GB2641079A - Systems and methods for controlling heat networks - Google Patents

Abstract

A heat network 100 comprises a plurality of components including a heating unit (e.g. heat pump 150 and/or electric heating elements 170) to heat water, pipework to circulate water and at least one flow control 140 (e.g. pumps, valves) to control the flow of water. One or more sensors 130 generate sensor data with respect to the plurality of components. A method of controlling the heat network by a control device 110 comprises: receiving sensor data from the sensor(s); determining at least one network performance metric based on the received data; and evaluating deterioration of the heat network based on the determined at least one network performance metric. In another aspect, a method of predictively maintaining the heat network comprises determining a trend of deterioration of the heat network based on the evaluated deterioration and projecting when maintenance of the heat network is required based on the trend. A system for controlling a heat network using a remote server that controls the heat network based on at least one network performance metric is also claimed.

Description

SYSTEMS AND METHODS FOR CONTROLLING HEAT NETWORKS

FIELD OF THE INVENTION

The present technology relates generally to systems and methods for controlling heat networks, in particular heat networks comprising a heat pump, for operating and maintaining heat networks, along with monitoring and detecting faults.

BACKGROUND

In general, operating a heat distribution network, or heat network, requires careful design of a system that comprises a plurality of electrical and mechanical components including, for example, circulation pumps, heat pumps, pipework, water and heat metering, controls for pipe flow, water pressure and temperature, thermal/heat storage, etc., and the system must be correctly commissioned according to design to ensure all components within the network operate synchronously. Herein, a heat distribution network or heat network refers to a network of components that comprises at least a heat provision unit to heat water, such as an electrical resistance heater, a gas boiler, a heat pump, etc., with suitable pipework to deliver or distribute heated (or cold) water around the network for the purpose of heating and/or provision of heated water.

Even a carefully designed and correctly commissioned heat network deteriorates over time and faults in one or more components occur from time to time. Conventionally, maintaining optimal functionality and accurately diagnosing faults and failures in such a complex system are immensely labour intensive, requiring skilled technicians or engineers to conduct diagnostics and tests on site, often through trial and error, in order for maintenance to be carried out.

Maintenance and operational health checks of heat networks require expertise and experience, and is generally conducted by trained engineers or technicians while physically on site. Assessment of various performance metrics of a heat network is performed by the qualified engineers or technicians who have access to certain performance data of the heat provision unit(s) of the heat network and/or various sensor data (e.g. pressure, temperature), heat/water meters, etc. In some cases, operation data may be obtained from the manufacturer of the heat provision unit (e.g. heat pump) that e.g. allows remote access when requested. However, maintenance and/or performance checks are often arbitrarily scheduled at best, or reactively scheduled when a fault or failure occurs.

At present, the standard of operations relies heavily on trained engineers, mechanics, electricians and/or technicians to attend a heat network in person and interact with the systems to perform diagnostics for maintenance. There is, therefore, scope for improving the control of heat networks for performance optimization during commissioning and during operation, fault detection and diagnosis, and maintenance scheduling.

SUMMARY

In view of the foregoing, an aspect of the present technology provides a control device for a heat network control system that controls a heat network, the heat network being formed of a plurality of components comprising a heat provision unit configured to heat water, pipework configured to circulate cold and/or heated water and at least one flow control configured to control the flow of the cold and/or heated water, and the heat network control system comprising one or more sensors configured to generate sensor data with respect to the plurality of components, the control device comprising: communication means configured to receive sensor data from the one or more sensors; and processing means configured to determine at least one network performance metric based on the received sensor data, and to evaluate performance/deterioration of the heat network based on the determined at least one network performance metric.

According to embodiments of the present technology, a control device is provided for controlling the operation of a heat network. Herein, a heat network or heat distribution network refers to a system formed of a plurality of components for delivering or distributing heat around a place, e.g. in the form of heated water.

Cold water may be heated by a heat provision unit, such as an electrical resistance heater, a gas boiler or a heat pump, and heated water may be distributed via suitably configured pipework e.g. facilitated by one or more flow controls such as circulation pumps, valves and actuators. The control device comprises communication means that communicates with one or more sensors provided to the heat network, which generate sensor data with respect to the components of the heat network. Thus, the communication means enables the control device to gather data on the components of the heat network. The control device may further communicate with the components of the heat network via the communication means. The communication means may be any suitable form of communication devices, modules and/or circuitry that receives and/or sends data via a physical (wired) connection and/or a wireless connection. The control device also comprises processing means that processes the received sensor data. The processing means may be any suitable form of processing circuitry, one or more processors/microprocessors, functional units, software or firmware. Using sensor data that represents the operation of various components of the heat network, the processing means determines at least one network performance metric (e.g., heat energy created and distributed versus electrical energy consumed) and uses the network performance metric(s) to evaluate the performance/deterioration of the heat network. The required data processing may be performed by the processing means of the control device or remotely by one or more other processing means communicatively connected to the control device and receiving data via the communication means, or it may be partially performed by the processing means of the control device and partially performed remotely. In some embodiments, the at least one network performance metric may be determined by applying a suitable algorithm using the sensor data as input, comparing the sensor data with a reference value or range, etc. In some embodiments, the deterioration of the heat network may be evaluated by applying one or more suitable algorithms using the network performance metric(s) as input, comparing the network performance metric(s) with a reference value or range, instantaneously or over time, reading a deterioration index off a predetermined table or database using the network performance metric(s) as a key, and/or applying a machine learning algorithm (MLA) trained to evaluate, forecast, project or extrapolate the deterioration of the heat network, etc. In doing so, it is possible not only to confirm (e.g. periodically) that the heat network is operating as expected or as designed, but to gauge the deterioration of the heat network as a whole and the components separately such that, for example, maintenance may be planned more effectively to prevent a fault in one or more components and/or a failure in the network.

Another aspect of the present technology provides a system for controlling a heat network, the heat network being formed of a plurality of components comprising a heat provision unit configured to heat water, pipework configured to circulate cold and/or heated water and at least one flow control configured to control the flow of the cold and/or heated water, the system comprising: one or more sensors configured to generate sensor data with respect to the plurality of components; and a control device comprising: communication means configured to receive sensor data from the one or more sensors; and processing means configured to determine at least one network performance metric based on the received sensor data, and to evaluate performance/deterioration of the heat network based on the determined at least one network performance metric.

A further aspect of the present technology provides a method of controlling a heat network by a control device, the heat network being formed of a plurality of components comprising a heat provision unit configured to heat water, pipework configured to circulate cold and/or heated water and at least one flow control configured to control the flow of the cold and/or heated water, the heat network being provided with one or more sensors configured to generate sensor data with respect to the plurality of components, the method comprising: receiving sensor data from the one or more sensors; determining at least one network performance metric based on the received sensor data; and evaluating performance/deterioration of the heat network based on the determined at least one network performance metric.

In some embodiments, the processing means may be further configured to compare the determined at least one network performance metric against a corresponding reference performance metric, and to determine a failure when the at least one network performance metric deviates from the corresponding reference performance metric. The determined network performance metric may be used in various ways to evaluate the performance of the heat network and its components. Comparing the determined network performance metric with a reference value or range provides a quick and simple evaluation, such that if the determined network performance metric is outside of (deviates from) the expected value or range, it may reasonably be assumed that there is a fault or failure within the system. The reference value or range may e.g. be based on system design, manufacturing parameters, learned experience, tolerance ranges and/or recorded commissioning values. For example, some components may not have independent metrics that can immediately indicate a fault in the component. In such cases, it may be possible to monitor changes in one or more metrics over time to observe patterns in these metrics when those components experience faults. In addition or alternatively, the performance metric(s) of, or in relation to, a component may be recorded when it is commissioned, and this recorded metric(s) then becomes the reference value of this component.

In some embodiments, the processing means may be further configured to monitor a change in the deterioration of the heat network to determine a trend of deterioration of the heat network. Through monitoring the change in deterioration, it is possible to estimate a trend, which may for example be extrapolated to e.g. project or forecast when the performance of (one or more components of) the heat network will fall below an acceptable level (e.g. due to a component failure through wear and tear).

Thus, in some embodiments, the processing means may be further configured to predictively project when maintenance of the heat network is required based on the trend of deterioration of the heat network and, optionally, to schedule the maintenance to prevent a network failure. In doing so, maintenance may be performed before the heat network fails to avoid disruption of service.

Throughout a day or a week, demands on a heat network, e.g. for heating an indoor space and/or for hot water, can vary. However, the variations in demands may follow a repetitive pattern over a 24-hour, weekly or seasonal period. For example, there may be a higher demand during the early morning and evening, but a lower demand during the day and late at night in a domestic setting. Thus, in some embodiments, the processing means may be further configured to analyse the at least one performance metric to project a network performance requirement over a predetermined time period so as to prepare the heat network in advance to meet the projected network performance requirement over the predetermined time period. For example, the analysis and/or the projection may be performed by a previously trained MLA. By projecting a network performance requirement over a predetermined time period (e.g. 24 hours) to prepare the heat network in advance, it is possible to ensure that demands on the heat network can be met. This is particularly relevant when a heat pump is used as a primary heat source, as a heat pump may require a period of time to reach operation temperature. Moreover, through the ability to project a network performance requirement that allows the preparation of the heat network in advance to meet the projected network performance requirement, it is possible to generate heat (by any available means including heat pump and electrical resistance heater, etc.) during off-peak periods, when demands on electricity nationally is lower (thus lower-cost electricity is available), and storing the heat for later use. This may be particularly advantageous when electricity is generated by certain renewable sources with fluctuating availability. Further, demand forecasting (as described herein) enables the heat network to generate a sufficient 15 amount of heat to meet demands without generating an excess amount of heat above demands that leads to wasted energy.

In some embodiments, the processing means may be further configured to analyse the at least one network performance metric to detect a fault in the heat network, and to identify the detected fault within the heat network based on the sensor data. The at least one network performance metric is determined based on sensor data that is representative of the plurality of components of the heat network. By strategically placing sensors around the heat network, it is possible, based on received sensor data, to isolate a portion of the heat network in which a fault is detected. Then, it is possible to identify the fault with a particular component or a group of components that causes the fault. In doing so, it may be possible to avoid performing lengthy diagnostics and/or trial and error to identify the cause of a fault.

In some embodiments, the processing means may be further configured, during a test phase, to generate an operation signal for the plurality of components to perform a test run, and to analyse the sensor signal generated from the test run to determine whether the heat network is correctly commissioned. When a heat network is installed, the control device may be configured to autonomously or semi-autonomously determine if the heat network has been correctly commissioned with or without a trained engineer on site. For example, the control device may send a run signal, e.g. through the communication means, to a pump and then check the appropriate sensor(s) that a flow can be detected downstream of the pump. Where an actuator valve is used, the control device may send a signal to open the actuator valve and then check whether water is running downstream of the actuator valve. Where a heat pump or a thermal store is used, the control device may signal the heat pump or thermal store to turn on or off and check for thermal responses (e.g. changes in temperature) in the components to and from the heat pump or thermal store. Moreover, the control device may be configured to confirm that the heat network has been correctly commissioned by monitoring the heat network over a period of time, and determine whether components of the heat network activate and deactivate as programmed or designed. Then, deviation or variation from expected behaviour may indicate that there is an error in the commissioning of the heat network and e.g. trigger a maintenance call.

In some embodiments, the at least one network performance metrics may comprise a calibrated agreement between two or more sensors of a same type, a flow rate against power consumed, a water chemistry measure, and wherein, optionally, a reference performance metric may represent a performance metric derived from manufacture setting, system design setting, optimal performance setting, or a combination thereof. A calibrated agreement between two or more sensors of the same type may include e.g. two (or more) temperature sensors, two (or more) flow sensors, two (or more) pressure sensors, etc. For example, when two temperature sensors were first commissioned, the two sensors may read a 0.5 degrees difference when measuring the same fluid. Over time, the agreement between the two sensors begin to deviate, which indicates that a recalibration is required or one or both of the sensors may need replacing. Measuring a flow rate against power consumed e.g. by a circulation pump enables detection of an increase in power consumption over time to achieve the same flow rate, which indicates that the circulation pump may require servicing or replacing.

In some embodiments, the one or more sensors may comprise one or more water temperature sensors, one or more ambient temperature sensors, one or more water flow sensors, one or more water pressure sensors, one or more heat meters, one or more water chemistry sensors, one or more sensors for measuring electrical frequency or wavelength from components, one or more acoustic sensors, one or more vibration sensors, one or more weather data sensors, or any combination thereof.

In some embodiments, the sensor data may comprise water temperature, ambient temperature, water flow rate, water pressure, heat provision unit operation data, water chemistry data, component electrical frequency or wavelength, acoustic data, vibration data, weather data, or any combination thereof.

A yet further aspect of the present technology provides a method of predictively maintaining a heat network by a control device, the heat network being formed of a plurality of components comprising a heat provision unit configured to heat water, pipework configured to circulate cold and/or heated water and at least one flow control configured to control the flow of the cold and/or heated water, the heat network being provided with one or more sensors configured to generate sensor data with respect to the plurality of components, the method comprising: receiving sensor data from the one or more sensors; determining at least one network performance metric based on the received sensor data; evaluating deterioration of the heat network based on the determined at least one network performance metric; determining a trend of deterioration of the heat network based on the evaluated deterioration; and projecting when maintenance of the heat network is required based on the trend of deterioration of the heat network.

According to embodiments of the present technology, instead of scheduling maintenance work arbitrarily or in reaction to a fault or failure in the system, by projecting or forecasting when maintenance may be required based on the trend of deterioration, it is possible to intelligently schedule maintenance before a fault or failure occurs while avoiding unnecessary maintenance. The method may, for example, be performed by a suitably trained MLA executing on the control device. The control device, or an artificial intelligent executing on the control device, may be configured to autonomously schedule maintenance work based on the projection.

A yet further aspect of the present technology provides a system for controlling a heat network, the heat network being formed of a plurality of components comprising a heat provision unit configured to heat water, pipework configured to circulate cold and/or heated water and at least one flow control configured to control the flow of the cold and/or heated water, the system comprising: one or more sensors configured to generate sensor data with respect to the plurality of components; a control device comprising: communication means configured to receive sensor data from the one or more sensors, to transmit the received sensor data, and to receive control instructions; and processing means configured to control the communication means to transmit the sensor data; and a remote server configured to determine at least one network performance metric based on the sensor data received from the control device, to evaluate performance of the heat network based on the determined at least one network performance metric, and to control operation of the heat network based on the at least one network performance metric by transmitting one or more control instructions.

In some embodiments, the remote server may be configured to receive data from a plurality of control devices respectively controlling a plurality of heat networks, and the remote server has executing thereon a machine learning algorithm trained using data collected from the plurality of control devices to determine operation of a given heat network.

Implementations of the present technology each have at least one of the above-mentioned objects and/or aspects, but do not necessarily have all of them. It should be understood that some aspects of the present technology that have resulted from attempting to attain the above-mentioned object may not satisfy this object and/or may satisfy other objects not specifically recited herein.

Additional and/or alternative features, aspects and advantages of implementations of the present technology will become apparent from the following description, the accompanying drawings and the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments will now be described, with reference to the accompanying drawings, in which: FIG. 1 shows schematically an exemplary heat distribution system controlled by a control device; FIG. 2 shows schematically examples of inputs and functions of the control device of FIG. 1; and FIG. 3 shows schematically examples of processing and actions of the control device of FIG. 1.

DETAILED DESCRIPTION

Operation of a heat network is generally maintained in reaction to the occurrence of a fault, and relies heavily on trained engineers, mechanics, electricians and/or technicians to attend on site and interact with the system to perform diagnostics and maintenance. It is therefore to improve the control of heat networks in terms of performance optimization, fault detection and diagnosis, and maintenance scheduling.

Thus, the present technology provides a control device for controlling the operation of a heat network. Herein, a heat network refers to a system of multiple components for delivering or distributing heat, e.g. in the form of heated water. Cold water is heated by a heat provision unit and heated water is distributed via suitably configured pipework, facilitated by one or more flow controls such as circulation pumps, valves and actuators. Other systems have been contemplated including heat networks/systems in which a heat provision unit heats cold or ambient fluid, generally in a central location, and then the heated fluid is distributed to the dwellings, and heat networks/systems in which cold or ambient fluid is distributed to the dwellings and inside each dwelling is provided a small-scale heat provision unit (or more than one units) that locally heats the fluid for use. The control device communicates, wirelessly or via a physical connection, with one or more sensors provided to the heat network to obtain sensor data representative of the operation of various components of the heat network. The control device may also communicate with the components of the heat network to provide control signals. The control device then processes the received sensor data. Using the received sensor data, the processing means determines at least one network performance metric (e.g. a calibrated agreement between two or more sensors of a same type, a flow rate against power consumed, a water chemistry measure) and uses the network performance metric to evaluate the deterioration of the heat network. The determination of one or more performance metrics and/or the evaluation of deterioration may be performed, for example, by applying a suitable algorithm and/or by applying a previously trained machine learning algorithm (MLA), etc. In doing so, it is possible not only to confirm (e.g. periodically) that the heat network is operating as expected or as designed (e.g. optimally), but to monitor the deterioration of the heat network as a whole and the components separately such that, for example, maintenance may be planned effectively to avoid a failure in the heat network.

A control system according to the present technology generally comprises electronic telemetry and hardware installed in a heat network, e.g. a heat pump-based heat network, which transmits performance data (e.g. time, heat pump metrics, heat meter metrics, water temperature, water flow, water pressure, water chemistry, vibration, electricity usage, etc.) to a control device, which may be situated remotely. The control device analyses the performance data through computation of performance metrics and monitoring and observation of the performance metrics over time. One or more analysis methods performed e.g. by artificial intelligence (AI) e.g. using one or more MLA5 may be incorporated to analyse the performance metrics and/or performance data.

Through the use of a system of electronic hardware, e.g. including multi-sensor telemetry and remote data transmission ability, together with processing and analysis provided by the control device, a control system according to the present technology is capable of: * Remotely assessing the operational health of the heat network through e.g. checking the heat network performance and metrics against design and operational settings. Based on the assessment, settings of various components of the heat network may be remotely controlled or adjusted. The remote assessment functionality allows detection when the heat network is down, i.e. when the heat network fails, not operational, or is outputting reduced or no heat.

* Implementing preventative and predictive maintenance through monitoring performance metrics over time and evaluate the deterioration of heat network equipment. In doing so, it is possible to flag a component or section of the heat network for maintenance well before a fault occurs. Moreover, parts and components may be replaced only when necessary instead of according to arbitrarily set timelines.

* Conducting remote fault diagnostics and fault analysis through routinely analysing the network's performance metrics, for example using a combination of advanced analytics and AI (machine learning and deep learning), to identify faults and determine the causes. In doing so, engineers may be informed of the fault before attending a maintenance call and thus mitigates the need for engineers to diagnose a complex heat network. It is thus possible to improve the efficiency of maintenance visits as the engineers may avoid performing time-consuming diagnostics and may arrive on site with the correct materials and replacement components.

* Remotely controlling the heat network, including remote rebooting, resetting, and/or reconfiguring the system. Through the ability of the control device to communicate with the heat network, it is able to control the heat network to adjust performance, heat production, timings, etc. to e.g. optimise performance. It further allows operation of the heat network to be adaptive and evolve with the changing heat demands on the network.

* Predicting or forecasting heat demand and electricity requirements on the heat network through the analytics ability of the control device. The control device, e.g. through one or more MLA5, can learn the performance requirements and/or heat demands of the heat network from sensor data and analytic data collected over time, and predict the heat network's performance requirements for a predetermined period of time in advance (e.g. 24 hours). In doing so, the heat network may be prepared in advance to ensure heat supply meets expected demands, for example by activating a heat pump to prepare heat storage before an expected rise in heat demand, while minimising heat waste, and taking into account performance optimisation against electricity tariffs.

* Sharing knowledge through each individual control device connected to different heat networks being connected to a corresponding virtual machine in a network, enabling the gathering and sharing of collective knowledge of multiple heat network systems. In doing so, a newly installed heat network may be controlled by a control device with access to the collective knowledge that can be used for detecting and diagnosing malfunctions or faults and for optimising operation performance.

The use of a control device in a control system according to the present technology substantially reduces the need for engineers on site, improving the system's performance while reducing operating costs, reducing waste e.g. in heat, electricity, spare parts, increasing heat supply reliability, and ultimately facilitating the provision of low-carbon, low-cost heat and hot water. A control device according to the present technology controls a heat network centrally and remotely, thus allowing a control system incorporating the control device to be scalable to different heat networks through strategically providing suitable sensors to various components and/or sections of a heat network.

FIG. 1 schematically illustrates a simplified heat network 100 controlled by a control device 110 as part of a control system according to an embodiment. The heat network 100 comprises a plurality of components including flow control 140 for controlling water flow such as flow rate and pressure, and may include e.g. one or more circulation pumps, valves and/or actuators. The heat network further comprises one or more heat provision units for heating water, such as a heat pump 150 which deposits heat or thermal energy into a thermal storage 160 (e.g. water storage tank) and electric heating elements 170 (e.g. electrical resistance heater). It should be noted that the thermal storage 160 is provided to the present embodiment as example only; in other heat networks or systems, a thermal storage may not be necessary. The heat pump 150 and the electric heating elements 170 may be operated at the same time or separately, as needed, for example the electric heating elements 170 may be operated only as a backup for the heat pump 150. The heat network 100 further comprises suitably configured pipework for circulating and distributing cold and heated water around the network. A plurality of sensors 130-1, 130-2, 130-3 are provided as part of the control system at various points of the heat network 100 to generate sensor data representative of the operation of various components or sections of the heat network 100. For example, sensors may be provided at the upstream end and the downstream end of flow control 140 to provide sensor data for flow before and after the flow control 140, sensors may be provided at the upstream end and downstream end of the heat pump 150 and thermal storage 160 to provide sensor data for temperature changes, and similarly sensors may be provided at the upstream end and downstream end of the electric heating elements 170 to provide sensor data for temperature changes. Weather data sensors may also be provided as desired.

The control device comprises communication means 111 configured to communicate with the heat network 100 to receive sensor data e.g. from the sensors 130-1, 130-2, 130-3 and to transmit control signals to the various components of the heat network 100. The communication means 111 may further receive operation data from various components of the heat network such as the heat pump 150 and the electric heating elements 170, amongst others. The control device further comprises processing means 112 configured to process sensor data and/or operation data received via the communication means 111 e.g. to determine one or more performance metrics that can be analysed to evaluate the state of the heat network 100, for example the level of deterioration of the heat network 100, whether the heat network 100 is operating optimally, whether there is a fault, etc. The control device 110 is further configured to monitor the performance metrics of the heat network 100 over time such that a trend of deterioration may be determined or projected, e.g. by means of one or more MLAs executing on the processing means 112. The trend of deterioration may be used for predicting or forecasting maintenance needs, such as anticipating when a component may fail, and maintenance may be scheduled well before the failure occurs to avoid disruption of service.

The control device, or the one or more MLAs executing thereon, may learn the operation requirements pattern or demands pattern of the heat network 100, and forecast the operation requirements or demands over e.g. a 24-hour period in advance to prepare the heat network 100 so as to meet its operation requirements.

The control device 110 may also be configured to transmit control signals to various components of the heat network 100 to operate the heat network, e.g. for testing purposes. For example, when the heat network 100 is first installed, the control device 110 may operate various components of the heat network 100 and check the sensor data that is generated as a result to confirm whether the components are operating as expected (according to system design). The control device 110 may further monitor the operation of the heat network 100 following the initial testing to confirm whether the heat network 100 continues to operate as expected. Thus, the control device may be used to confirm whether the heat network 100 has been commissioned correctly.

In some embodiments, the control device 110 may be a single device installed to a heat network for controlling the heat network. In other embodiments, the control device 110 may instead be distributed such that a physical device at the heat network performs the function of receiving and transmitting data and control instructions, while the data processing functions are performed remotely e.g. by a remote server and/or as a virtual machine running on a remote server. Both implementations are contemplated. In the latter implementation, a remote server or virtual machine perform the data processing to compute the required performance metrics, and determines the appropriate response and/or operation control for the heat network based on the performance metrics. Current and past performance metrics may also be used to generate performance insights on the heat network for reference and/or for machine learning purposes. Such remote/distributed arrangement also enables important or critical control to be performed centrally, such as shutting down or reconfiguring certain components.

In the embodiments, for each heat network, there may be several sources of raw and processed data. Data received from the heat network may include sensor data from simple discrete sensors such as for temperature, water flow and pressure, data from heat meters (which logs the combination of water flow and temperature), and operating data from one or more heat pumps (e.g. on/off, run signals to pumps/compressors/valves, internal pressures, temperatures, etc.). The various sources of raw operating data can be combined to provide a perspective on the performance of the heat network. The data may be graphed and queried over different time intervals; however, such raw data is generally incomprehensible even to a trained user. Thus, according to the present technology, the control device is configured to apply one or more algorithms to calculate and report one or more performance metrics, for example, heat produced per day, compressor runtime per day, etc. In the embodiments, the control device determines the heat network is down, non-operational or experiencing a failure when it detects a fault signal from one or more component, or when analysis of the operational data indicates that a part or section of the heat network is not operating as expected, e.g. one or more performance metrics are outside an acceptable range defined by a reference setting. The reference setting is the setting or operation performance according to which the heat network is designed to operate. For example, a heat pump of the heat network may be set to reach a predetermined operation temperature e.g. to ensure legionella disinfection, or the heat pump may be set to operate at a predetermine time and predetermined flowrate. The (processing means of the) control device may be configured to execute one or more suitable algorithms to determine if the heat network and its components are operating at expected times and in the expected manners. In particular, the control device may determine that the heat network is down when it experiences a critical fault, for example, and typically, when water flow and/or heat flow stops. The control device may be configured such that detection of a critical fault triggers an alert (e.g. a notification such as an automated email to a system operator or an engineer) to prompt a maintenance visit to the site. In addition, the control device may be configured such that a critical fault further triggers an activation of one or more backup heat provision units (e.g. electrical immersion heaters) and adjust other operation settings to temporarily manage heat demands. For non-critical faults, the control device may be configured to, upon detection of such a fault, alert or notify engineers or operators to schedule maintenance visits in order to avoid the heat network trending toward a critical fault.

In the embodiments, the control device is configured to evaluate the deterioration of a heat network by observing the heat network and its components over time. For example, a specific set of algorithms may be executed by the control device to determine the performance of each component when it performs a function, e.g. when a circulation pump causes water flow or when a heat pump heats the water in a storage tank, based on sensor data corresponding to the component. The set of algorithms corresponding to a component may generate a score or an indicator on the level of performance of that component (e.g. a performance metric), based on sensor data, and then monitor the changes in the score or indicator over time. For example, a circulation pump may initially have a flowrate of 2L/s when first installed but overtime the flowrate may reduce to 1.6L/s. A heat pump may heat the water in a storage tank to a pre-set temperature at an optimal rate when first installed but may after a period of time (e.g. a few years) require twice the amount of time to achieve the same temperature. These measurements or sensor data can serve as metrics of deterioration, which can be used to assess appropriate times to service or replace components e.g. based on cost-benefit analyses. For example, continuous monitoring and analysis may indicate that, at a certain point in time, it becomes more cost-effective to replace an immersion heater compared to the consumption of additional electricity for the immersion heater to heat an amount of water to the required temperature. Moreover, a further set of algorithms may be applied to extrapolate the performance and/or deterioration of a component or a part of the heat network over time (trend analysis and forecasting). In doing so, it is possible to schedule maintenance before the need arises having predicted or forecasted when a component may fail or reach the end of its useful life.

It may be possible to monitor the performance of every single component of a heat network, to detect when a fault occurs in the component and to predict or forecast its deterioration, depending on the extent to which each component can be constrained. For example, if a sensor is provided on either side of a component with no other variables between the two sensors, then when sensor data is received it is possible to deduce any changes are caused by that component alone. In practice, at least some parts of a heat network have multiple components between a pair of sensors, which, depending on the changes seen in the sensor data, can either allow the responsible component to be identified that's deteriorating or inference of one or more components that are responsible.

Typically, major components such as heat pumps, thermal storages, circulation pumps, immersion heaters are provided with sensors and monitored, as well as general pipework and water conditions.

The present technology is particularly relevant to heat pump-based heat networks. Heat pumps typically log and record tens of metrics internally, including their own fault signals. However, while conventional heat pumps can send a fault signal when a fault occurs, they generally lack the ability to self-diagnose and cannot always identify the cause of the fault. Through the analysis, monitoring and forecasting by the control device, embodiments of the present technology is able to use heat pump data and/or sensor data to identify faults and determine possible causes. For example, a heat pump may indicate/signal for a pressure fault, and (algorithm running on) the control device may determine the likely cause based on past and present heat pump data points. Examples of such causes include e.g. a refrigerant leaks or a faulty compressor.

According to embodiments, for circulation pumps, flow rates and run signals (whether a pump is running) are monitored to determine if and when a circulation pump is pumping, if it is pumping at the correct or expected flowrate, and if the flowrate changes overtime, etc. Circulation pumps is typically operated when there is a pressure drop or in response to run signals, so the control device may deduce a potential leak in the pipework based on one or more circulation pumps operating unexpectedly. To analyze circulation pump performances, run signals or operation signals are sent to a circulation pump (e.g. in the form of an electrical log of l's and 0's) and a flow meter corresponding to the circulation pump is monitored. It is possible to identify both electrical and mechanical faults in this manner. For example, upon sending a run signal to the circulation pump, no flow is detected by the corresponding flowmeter, then this may be due to a broken pump or an electrical fault. If a circulation pump is operating at a progressively lower flowrate, this may indicate deterioration over time. In response, an engineer may be sent to service or replace the deteriorated circulation pump and thus restoring the flow rate. For pipework, detection of a reduction in flowrate while water pressure remains constant may infer a build-up of scale. If water pressure changes, equipment damage such as a leak may be inferred.

In some embodiments, the control device may be configured to remotely reboot, reset and/or reconfigure a heat pump. Operations of heat pumps can be complex and each heat pump typically has its own internal programmed settings and behaviors. For example, some heat pumps may experience faults when they encounter certain operation settings and they are therefore programmed to avoid such operation settings. In some circumstances, it may be appropriate to detect these settings and to override them, for example when the real cause of a fault is known and is unrelated to such settings. Remotely resetting and configuring also enables testing of certain operating conditions and diagnose faults without the physical presence of an engineer on site. Moreover, this functionality may improve the security of heat networks, in that (algorithms running on) the control device may sweep the network settings of a heat network and confirm that nothing has been tampered with (and reconfiguring them if necessary).

Embodiments of the control device may be configured to determine if a heat network has been commissioned successfully/correctly. Firstly, the control device is configured such that it is able to distinguish the source of various sensor data, e.g. circulation pumps, heat pumps, thermal storages, heat meters, actuator valves, etc. The control device may determine whether a component is commissioned correctly by confirming whether the various components respond to use. For example, the control device may send a run (operation) signal to a circulation pump and then check whether a flow can be detected downstream of the circulation pump. Similarly, the control device may control an actuator valve and then check whether water is running or not running downstream of the actuator valve. For heat pumps and thermal storages, the control device may turn a heat pump on or off and check for thermal responses in parts leading to and leaving these components. Moreover, the control device may monitor the heat network over a period of time and determine whether all the components are activating and deactivating as programmed and designed. Deviation from expected behavior may indicate that the heat network has not been commissioned properly according to the designed operating method.

FIG. 2 schematically illustrates examples of input data used by a control device according to the embodiments and possible functions that may be performed by the control device. It should be noted that, in the present embodiment, a control device may refer to a physical device installed to a heat network, or in a distributed or remote system in which the functions are performed remotely by a remote server and/or a virtual machine running on the remote server. In the example, the control device, such as the control device 110, can receive one or more of heat pump data 201, heat meter data 202, thermal storage data 203, temperature data from one or more temperature sensors 204, pressure data from one or more pressure sensors 205, flow data 206 at various points of a heat network, valve signal data 207 from one or more actuator valves, circulation pump data 208 from one or more circulation pumps and other resource data 209 such as weather data, electrical supply grid data and/or user supplied data from the heat network, and use the received data as input for data analysis 210, for example, by applying one or more algorithms and/or executing one or more MLAs, in order to output one or more functions. Based on the analysis and monitoring, the control device may be configured to perform functions such as generating a network health report 221, determine when maintenance is required (e.g. based on deterioration trends) and predictively schedule the required maintenance 222 in advance of a failure, perform fault diagnostics 223 when a fault signal is detected, provide forecasting or projection of heat and electricity demands 224 (e.g. over a 24-hour period) for example to prepare the heat network in advance of expected high demands, and/or perform commission checks 225 when the heat network is first installed or when changes have been made to the heat network.

FIG. 3 schematically illustrates an example of data analysis that may be performed by the control device of FIG. 2. According to embodiments, the control device uses input data 300, such as sensor data 301 and resource data 302, to perform various analyses 310. In the present embodiment, the control device receives input data 300 and compile the received data 311. The control device then generates one or more performance metrics 312 using the received and compiled data. The performance metrics are then analysed 313 as well as being written to a metric database 315. The results of the metric analysis 313 are used to generate an AI database 314, and data from the AI database 314 is used for training one or more MLA5 318. The metric database 315 may be queried periodically to perform time-dependent metric analyses 316, such that changes in performance metrics may be monitored over time to determine e.g. performance deteriorations. The results of both the current/instant metric analysis 313 and the time-dependent metric analysis are used for insight generation 317, such as deterioration trending, maintenance/faults forecasting, demands/requirements predictions, etc. The generated insights are then used as inputs for machine learning 318, together with data stored in the AI database 314. The control device then performs one or more functions 320 as described above, based on the results of the data analysis 310, and in doing so, the control device is able to autonomously perform one or more actions 330 including remotely controlling 331 (e.g. resetting, reconfiguring, rebooting, adjusting settings of various components, etc.) the heat network and its various components, generating various performance reports 332 as needed, generating various metering and/or billing reports 333, scheduling maintenance 334 such as site visits, parts ordering, etc. before the heat network experience a failure, and/or sharing insight learning amongst a plurality of connected heat networks 335.

The following gives a brief overview of a number of different types of machine learning algorithms for embodiment(s) in which one or more MLAs are used.

However, it should be noted that the use of an MLA in these embodiment(s) is a non-limiting example of implementing the present technology, and the use of an MLA is not essential.

Overview of MLAs There are many different types of MLAs known in the art. Broadly speaking, 10 there are three types of MLAs: supervised learning-based MLAs, unsupervised learning-based MLAs, and reinforcement learning-based MLAs.

Supervised learning MLA process is based on a target -outcome variable (or dependent variable), which is to be predicted from a given set of predictors (independent variables). Using this set of variables, the MLA generates a function using training data that maps inputs to desired outputs during training. The training process continues until the MLA achieves a desired level of accuracy on validation data. Examples of supervised learning-based MLAs include: Regression, Decision Tree, Random Forest, Logistic Regression, etc. Unsupervised learning MLA does not involve predicting a target or outcome variable but learns patterns from untagged data. Such MLAs are capable of self-organization to capture patterns as probability densities, and are used e.g. for clustering a population of values into different groups. Clustering is used in many fields including pattern recognition, image analysis, bioinformatics, data compression, computer graphics, etc. Examples of unsupervised learning MLAs include: apriori algorithm and k-means algorithm.

Reinforcement learning MLA is trained to take actions or make decisions that maximize cumulative reward (e.g. a user-provided score). During training, the MLA is exposed to a training environment where it learns through trial and error to develop an optimal or near-optimal policy that maximizes reward. In doing so, the MLA learns from past experience and attempts to capture the best possible knowledge to make desirable decisions. An example of reinforcement learning MLA is a Markov Decision Process.

It should be understood that different types of MLAs having different structures or topologies may be used for various tasks. One particular type of MLAs includes artificial neural networks (ANN), also known as neural networks (NN).

Neural Networks (NN) Generally speaking, a given NN consists of an interconnected group of artificial "neurons", which process information using a connectionist approach to computation. NNs are used to model complex relationships between inputs and outputs (without actually knowing the relationships) or to find patterns in data.

NNs are first conditioned in a training phase in which they are provided with a known set of "inputs" and information for adapting the NN to generate appropriate outputs (for a given situation that is being attempted to be modelled). During this training phase, the given NN adapts to the situation being learned and changes its structure such that the given NN will be able to provide reasonable predicted outputs for given inputs in a new situation (based on what was learned). Thus, rather than attempting to determine a complex statistical arrangements or mathematical algorithms for a given situation, the given NN aims to provide an "intuitive" answer based on a "feeling" for a situation. The given NN is thus regarded as a trained "black box", which can be used to determine a reasonable answer to a given set of inputs in a situation giving little importance to what happens inside the "box".

NNs are commonly used in many such situations where an appropriate output based on a given input is important, but exactly how that output is derived is of lesser importance or is unimportant. For example, NNs are commonly used to optimize the distribution of web-traffic between servers and in data processing, including filtering, clustering, signal separation, compression, vector generation and the like.

Deep Neural Networks In some non-limiting embodiments of the present technology, the NN can be 30 implemented as a deep neural network. It should be understood that NNs can be classified into various classes of NNs. Below are a few non-limiting example classes of NNs.

Recurrent Neural Networks (RNNs) RNNs are adapted to use their "internal states" (stored memory) to process sequences of inputs. This makes RNNs well-suited for tasks such as unsegmented handwriting recognition and speech recognition, for example. These internal states of the RNNs can be controlled and are referred to as "gated" states or "gated" memories.

It should also be noted that RNNs themselves can also be classified into various sub-classes of RNNs. For example, RNNs comprise Long Short-Term Memory (LSTM) networks, Gated Recurrent Units (GRUB), Bidirectional RNNs (BRNN5), and the like.

LSTM networks are deep learning systems that can learn tasks that require, in a sense, "memories" of events that happened during very short and discrete time steps earlier. Topologies of LSTM networks can vary based on specific tasks that they "learn" to perform. For example, LSTM networks may learn to perform tasks where relatively long delays occur between events or where events occur together at low and at high frequencies. RNNs having particular gated mechanisms are referred to as GRUB. Unlike LSTM networks, GRUB lack "output gates" and, therefore, have fewer parameters than LSTM networks. BRN Ns may have "hidden layers" of neurons that are connected in opposite directions which may allow using information from past as well as future states.

Residual Neural Network (ResNet) Another example of the NN that can be used to implement non-limiting embodiments of the present technology is a residual neural network (ResNet).

Deep networks naturally integrate low/mid/high-level features and classifiers in an end-to-end multilayer fashion, and the "levels" of features can be enriched by the number of stacked layers (depth).

Convolutional Neural Network (CNN) CNNs are also known as shift invariant or space invariant artificial neural networks (SIANN), based on the shared-weight architecture of the convolution kernels or filters that slide along input features and provide translation equivariant responses known as feature maps. They are most commonly applied to analyze visual imagery and have applications in image and video recognition, recommender systems, image classification, image segmentation, medical image analysis, natural language processing, brain-computer interfaces, and financial time series.

CNNs are regularized fully connected networks, that is, each neuron in one layer is connected to all neurons in the next layer. CNNs use relatively little preprocessing compared to other image classification algorithms and learn to optimize the filters (or kernels) through automated learning.

To summarize, the implementation of at least a portion of the one or more MLAs in the context of the present technology can be broadly categorized into two phases -a training phase and an in-use or deployed phase. First, the given MLA is trained in the training phase using one or more appropriate training data sets. Then, once the given MLA learned what data to expect as inputs and what data to provide as outputs, the given MLA is executed using in-use data in the in-use or deployed phase. Further, while deployed, the given MLA may continue to learn from the in-use data based for example on user feedback.

The various MLAs described above may refer to the same or different MLA. If multiple MLAs are implemented, one or some or all of the MLAs may be executed on the device, and one or some or all of the MLAs may be executed on a server (e.g. a cloud server) in communication with the device via a suitable communication channel. It will be understood by those skilled in the art that the embodiments above may be implemented in any combinations, in parallel or as alternative strategies as desired.

As will be appreciated by one skilled in the art, the present techniques may be embodied as a system, method or computer program product. Accordingly, the present techniques may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware.

Furthermore, the present techniques may take the form of a computer 30 program product embodied in a computer readable medium having computer readable program code embodied thereon. The computer readable medium may be a computer readable signal medium or a computer readable storage medium.

A computer readable medium may be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing.

Computer program code for carrying out operations of the present techniques 5 may be written in any combination of one or more programming languages, including object-oriented programming languages and conventional procedural programming languages.

For example, program code for carrying out operations of the present techniques may comprise source, object or executable code in a conventional programming language (interpreted or compiled) such as C, or assembly code, code for setting up or controlling an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array), or code for a hardware description language such as VerilogTM or VHDL (Very high-speed integrated circuit Hardware Description Language).

The program code may execute entirely on the user's computer, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network. Code components may be embodied as procedures, methods or the like, and may comprise sub-components which may take the form of instructions or sequences of instructions at any of the levels of abstraction, from the direct machine instructions of a native instruction set to high-level compiled or interpreted language constructs.

It will also be clear to one of skill in the art that all or part of a logical method according to the preferred embodiments of the present techniques may suitably be embodied in a logic apparatus comprising logic elements to perform the steps of the method, and that such logic elements may comprise components such as logic gates in, for example a programmable logic array or application-specific integrated circuit. Such a logic arrangement may further be embodied in enabling elements for temporarily or permanently establishing logic structures in such an array or circuit using, for example, a virtual hardware descriptor language, which may be stored and transmitted using fixed or transmittable carrier media.

The examples and conditional language recited herein are intended to aid the reader in understanding the principles of the present technology and not to limit its scope to such specifically recited examples and conditions. It will be appreciated that those skilled in the art may devise various arrangements which, although not explicitly described or shown herein, nonetheless embody the principles of the present technology and are included within its scope as defined by the appended claims.

Furthermore, as an aid to understanding, the above description may describe relatively simplified implementations of the present technology. As persons skilled in the art would understand, various implementations of the present technology 10 may be of a greater complexity.

In some cases, what are believed to be helpful examples of modifications to the present technology may also be set forth. This is done merely as an aid to understanding, and, again, not to limit the scope or set forth the bounds of the present technology. These modifications are not an exhaustive list, and a person skilled in the art may make other modifications while nonetheless remaining within the scope of the present technology. Further, where no examples of modifications have been set forth, it should not be interpreted that no modifications are possible and/or that what is described is the sole manner of implementing that element of the present technology.

Moreover, all statements herein reciting principles, aspects, and implementations of the technology, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof, whether they are currently known or developed in the future. Thus, for example, it will be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative circuitry embodying the principles of the present technology. Similarly, it will be appreciated that any flowcharts, flow diagrams, state transition diagrams, pseudo-code, and the like represent various processes which may be substantially represented in computer-readable media and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.

The functions of the various elements shown in the figures, including any functional block labeled as a "processor", may be provided through the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functions may be provided by a single dedicated processor, by a single shared processor, or by a plurality of individual processors, some of which may be shared. Moreover, explicit use of the term "processor" or "controller" should not be construed to refer exclusively to hardware capable of executing software, and may implicitly include, without limitation, digital signal processor (DSP) hardware, network processor, application specific integrated circuit (ASIC), field programmable gate array (FPGA), read-only memory (ROM) for storing software, random access memory (RAM), and non-volatile storage. Other hardware, conventional and/or custom, may also be included.

Software modules, or simply modules which are implied to be software, may be represented herein as any combination of flowchart elements or other elements indicating performance of process steps and/or textual description. Such modules may be executed by hardware that is expressly or implicitly shown.

It will be clear to one skilled in the art that many improvements and modifications can be made to the foregoing exemplary embodiments without departing from the scope of the present techniques.

Claims (30)

1. CLAIMS1. A control device for a heat network control system that controls a heat network, the heat network being formed of a plurality of components comprising a heat provision unit configured to heat water, pipework configured to circulate cold and/or heated water and at least one flow control configured to control the flow of the cold and/or heated water, and the heat network control system comprising one or more sensors configured to generate sensor data with respect to the plurality of components, the control device comprising: communication means configured to receive sensor data from the one or more sensors; and processing means configured to determine at least one network performance metric based on the received sensor data, and to evaluate deterioration of the heat network based on the determined at least one network performance metric.
2. 2. The control device of claim 1, wherein the processing means is further configured to compare the determined at least one network performance metric against a corresponding reference performance metric, and to determine a failure when the at least one network performance metric deviates from the corresponding reference performance metric.
3. 3. The control device of claim 1 or 2, wherein the processing means is further configured to monitor a change in the deterioration of the heat network to determine a trend of deterioration of the heat network.
4. 4. The control device of claim 3, wherein the processing means is further configured to predictively project when maintenance of the heat network is required based on the trend of deterioration of the heat network and, optionally, 30 to schedule the maintenance to prevent a network failure.
5. 5. The control device of any preceding claim, wherein the processing means is further configured to analyse the at least one performance metric to project a network performance requirement over a predetermined time period so as to 35 prepare the heat network in advance to meet the projected network performance requirement over the predetermined time period.
6. 6. The control device of any preceding claim, wherein the processing means is further configured to analyse the at least one performance metric to detect a fault in the heat network, and to identify the detected fault within the heat network based on the sensor data.
7. 7. The control device of any preceding claim, wherein the processing means is further configured, during a test phase, to generate an operation signal for the plurality of components to perform a test run, and to analyse the sensor signal generated from the test run to determine whether the heat network is correctly commissioned.
8. 8. The control device of any preceding claim, wherein the at least one network performance metrics comprise a calibrated agreement between two or more sensors of a same type, a flow rate against power consumed, a water chemistry measure, and wherein, optionally, a reference performance metric represents a performance metric derived from manufacture setting, system design setting, optimal performance setting, or a combination thereof.
9. 9. A system for controlling a heat network, the heat network being formed of a plurality of components comprising a heat provision unit configured to heat water, pipework configured to circulate cold and/or heated water and at least one flow control configured to control the flow of the cold and/or heated water, the system comprising: one or more sensors configured to generate sensor data with respect to the plurality of components; and a control device comprising: communication means configured to receive sensor data from the one or 30 more sensors; and processing means configured to determine at least one network performance metric based on the received sensor data, and to evaluate deterioration of the heat network based on the determined at least one network performance metric.
10. 10. The system of claim 9, wherein the one or more sensors comprise one or more water temperature sensors, one or more ambient temperature sensors, one or more water flow sensors, one or more water pressure sensors, one or more heat meters, one or more water chemistry sensors, one or more sensors for measuring electrical frequency or wavelength from components, one or more acoustic sensors, one or more vibration sensors, one or more weather data sensors, or any combination thereof.
11. 11. The system of claim 9 or 10, wherein the sensor data comprises water temperature, ambient temperature, water flow rate, water pressure, heat provision unit operation data, water chemistry data, component electrical frequency or wavelength, acoustic data, vibration data, weather data, or any combination thereof.
12. 12. The system of any of claims 9 to 11, wherein the processing means is further configured to compare the determined at least one network performance metric against a corresponding reference performance metric, and to determine a failure when the at least one network performance metric deviates from the corresponding reference performance metric.
13. 13. The system of any of claims 9 to 12, wherein the processing means is further configured to monitor a change in the deterioration of the heat network to determine a trend of deterioration of the heat network
14. 14. The system of claim 13, wherein the processing means is further configured predictively project when maintenance of the heat network is required based on the trend of deterioration of the heat network and, optionally, to schedule the maintenance to prevent a network failure.
15. 15. The system of any of claims 9 to 14, wherein the processing means is further configured to analyse the at least one performance metric to project a network performance requirement over a predetermined time period so as to prepare the heat network in advance to meet the projected network performance requirement over the predetermined time period.
16. 16. The system of any of claims 9 to 15, wherein the processing means is further configured to analyse the at least one performance metric to detect a fault in the heat network, and to identify the detected fault within the heat network based on the sensor data.
17. 17. The system of any of claims 9 to 16, wherein the processing means is further configured, during a test phase, to generate an operation signal for the plurality of components to perform a test run, and to analyse the sensor signal generated from the test run to determine whether the heat network is correctly commissioned.
18. 18. The system of any of claims 9 to 17, wherein the at least one network performance metrics comprise a calibrated agreement between two or more sensors of a same type, a flow rate against power consumed, a water chemistry measure, and wherein, optionally, a reference performance metric represents a performance metric derived from manufacture setting, system design setting, optimal performance setting, or a combination thereof.
19. 19. A method of controlling a heat network by a control device, the heat network being formed of a plurality of components comprising a heat provision unit configured to heat water, pipework configured to circulate cold and/or heated water and at least one flow control configured to control the flow of the cold and/or heated water, the heat network being provided with one or more sensors configured to generate sensor data with respect to the plurality of components, the method comprising: receiving sensor data from the one or more sensors; determining at least one network performance metric based on the received sensor data; and evaluating deterioration of the heat network based on the determined at 30 least one network performance metric.
20. 20. The method of claim 19, wherein the sensor data comprises water temperature, ambient temperature, water flow rate, water pressure, heat provision unit operation data, water chemistry data, component electrical frequency or wavelength, acoustic data, vibration data, weather data, or any combination thereof.
21. 21. The method of claim 19 or 20, further comprising comparing the determined at least one network performance metric against a corresponding reference performance metric, and determining a failure when the at least one network performance metric deviates from the corresponding reference performance metric.
22. 22. The method of any of claims 19 to 21, further comprising monitoring a change in the deterioration of the heat network to determine a trend of deterioration of the heat network.
23. 23. The method of claim 22, further comprising predictively projecting when maintenance of the heat network is required based on the trend of deterioration of the heat network and, optionally, scheduling the maintenance to prevent a network failure.
24. 24. The method of any of claims 19 to 23, further comprising analysing the at least one performance metric to project a network performance requirement over a predetermined time period, and preparing the heat network in advance to meet the projected network performance requirement over the predetermined time period.
25. 25. The method of any of claims 19 to 24, further comprising analysing the at least one performance metric to detect a fault in the heat network, and identifying the detected fault within the heat network based on the sensor data.
26. 26. The method of any of claims 19 to 25, further comprising, during a test phase, generating an operation signal for the plurality of components to perform a test run, and analysing the sensor signal generated from the test run to determine whether the heat network is correctly commissioned.
27. 27. The method of any of claims 19 to 26, wherein the at least one network performance metrics comprise a calibrated agreement between two or more 35 sensors of a same type, a flow rate against power consumed, a water chemistry measure, and wherein, optionally, a reference performance metric represents a performance metric derived from manufacture setting, system design setting, optimal performance setting, or a combination thereof.
28. 28. A method of predictively maintaining a heat network by a control device, the heat network being formed of a plurality of components comprising a heat provision unit configured to heat water, pipework configured to circulate cold and/or heated water and at least one flow control configured to control the flow of the cold and/or heated water, the heat network being provided with one or more sensors configured to generate sensor data with respect to the plurality of components, the method comprising: receiving sensor data from the one or more sensors; determining at least one network performance metric based on the received sensor data; evaluating deterioration of the heat network based on the determined at least one network performance metric; determining a trend of deterioration of the heat network based on the evaluated deterioration; and projecting when maintenance of the heat network is required based on the 20 trend of deterioration of the heat network.
29. 29. A system for controlling a heat network, the heat network being formed of a plurality of components comprising a heat provision unit configured to heat water, pipework configured to circulate cold and/or heated water and at least one flow control configured to control the flow of the cold and/or heated water, the system comprising: one or more sensors configured to generate sensor data with respect to the plurality of components; a control device comprising: communication means configured to receive sensor data from the one or more sensors, to transmit the received sensor data, and to receive control instructions; and processing means configured to control the communication means to transmit the sensor data; and a remote server configured to determine at least one network performance metric based on the sensor data received from the control device, to evaluate performance of the heat network based on the determined at least one network performance metric, and to control operation of the heat network based on the at least one network performance metric by transmitting one or more control instructions.
30. 30. The system of claim 29, wherein the remote server is configured to receive data from a plurality of control devices respectively controlling a plurality of heat networks, the remote server has executing thereon a machine learning algorithm trained using data collected from the plurality of control devices to determine operation of a given heat network.