CN106331037A - Computing node for distributed computing network - Google Patents
Computing node for distributed computing network Download PDFInfo
- Publication number
- CN106331037A CN106331037A CN201510378076.2A CN201510378076A CN106331037A CN 106331037 A CN106331037 A CN 106331037A CN 201510378076 A CN201510378076 A CN 201510378076A CN 106331037 A CN106331037 A CN 106331037A
- Authority
- CN
- China
- Prior art keywords
- cpn
- calculating node
- node
- calculating
- algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention discloses a computing node (CPN) for a distributed computing network. The CPN is a computer with information receiving, processing and sending functions. The CPU is equipped with a central processing unit, a memory and a communication interface. A plurality of CPNs form the distributed computing network. Each CPN carries out data interaction with adjacent CPNs in a topology where the CPN is located. Each CPN is equipped with a space attribute. The space attribute indicates an absolute space location where the CPN is located and/or a relative space location in a topology network where the CPN is located. An operating system is built in each computing node. An API interface is provided by each operating system. According to each API interface, various management/control demands and/or strategies are converted into standard computing sequences and the standard computing sequences are converted into instruction sequences capable of being identified by each operating system. The CPNs in the distributed computing network finish computing the sequences together in a distributed and self-organizing mode. The CPNs can be applied to a building automatic control system, and an open, flat and easy-programmable control platform is provided.
Description
Technical field
The present invention relates to calculate networking technology area, be specifically related to a kind of calculating node for distributed computing network.
Background technology
Since the eighties in 20th century, people start with information technology and realize the intelligent automatic control of building, such as towards spy
Determine the broadcast stereo system of service function, IC-card management system, hotel guest room management system, energy monitoring management system, with
And towards the air-conditioner control system of particular mechatronic, safety-protection system, fire-fighting system, cold station control system, Electrical Safety system
Deng, but problem is management control system of the prior art the most less " intelligent ", exceedes the building robot control system(RCS) of half
Can only realize remotely monitoring the operational factor of architectural environment and system equipment at central control room, and by middle control man machine interface hands
The start and stop of dynamic ground or regulation electromechanical equipment running status.The manual operation of operations staff is still depended critically upon on such system parenchyma,
It is not carried out automatization and intellectuality.The only building of only a few, it is possible to realize Automated condtrol and the management of building aspect, bag
Include the optimal control within subsystems, and the integrating control between subsystem.
The basic reason causing this situation is the centralized framework of robot control system(RCS), as it is shown in figure 1, of the prior art from
Ore-controlling Role uses centralized organizational structure, and all terminal measuring control points (sensor, executor, field controller) all pass through always
Line communication network connects, the information of the end of subsystems (illuminator, air conditioning system, fire-fighting system, safety-protection system)
Although measuring control point is largely distributed in same building subspace, but carries out longitudinally perpendicular according to different subsystems
Integrated, this centralized robot control system(RCS) has a following major defect:
1, the measuring control point of terminal needs carry out global communication name, definition physical attribute and define incidence relation each other,
When observing and controlling is counted more, it is very big that the work of this situ configuration and configuration becomes workload, very difficult；This job demand
Could carry out after architecture construction completes and electromechanical equipment is in place, it is possible to use construction period the shortest, thus time is pressing；?
Later stage architectural composition or function divide when changing, and robot control system(RCS) is difficult to change the most flexibly；
2, it is difficult to the real of information between each subsystem share, if wanting to realize cross-system information sharing it is necessary to existing
A new system is set up on the upper strata of dry systems, and it is high, no that this is accomplished by that the overall situation re-starts configuration and definition, difficulty and cost
Adapt to building control intellectuality, information-based, the demand of front end；
3, self-controlled platform closing, poor universality, it is often the most individually designed for individual building for controlling software, thus to being
During system is transformed and extended, new control strategy is difficult on existing self-controlled platform realize simply flexibly.
4, the integrated difficulty of cross-system, existing robot control system(RCS) and platform compatibility is poor, hardware environment is unfriendly, it is desirable to exploitation
Person has higher IT Professional knowledge, and the various control strategies run in robot control system(RCS) and control logic are (such as HVAC system
Control strategy, the control strategy of fire-fighting system, the control strategy of safety-protection system) (warm by the engineer of every field often again
Logical engineer, FSE teacher etc.) formulate, this engineer resulting in each field is difficult to the control made
Strategy and control logic and be converted into the control software of robot control system(RCS), thus results in the function that subsystems should realize and is difficult to integrated
In the middle of existing robot control system(RCS) and self-controlled platform.
Summary of the invention
It is contemplated that at least solve one of technical problem present in prior art.
To this end, it is an object of the present invention to propose a kind of calculating node (CPN) for distributed computing network, this use
Calculating node (CPN) in distributed computing network can be applied to build robot control system(RCS), it is provided that an opening, flat,
The control platform of easy programming.Although it is further to note that the calculating node for distributed computing network that the present invention provides
(CPN) result from and can apply to building automation field, but this is not intended that the restriction to protection scope of the present invention, by
The distributed computing network that the calculating node (CPN) for distributed computing network of the present invention is constituted is basic as one
Calculate network and can also be applied to other field.
To achieve these goals, embodiment of the invention discloses that a kind of calculating node for distributed computing network
(CPN), described calculating node (CPN) is for having information reception, process, the computer of sending function, and described calculating saves
Point (CPN) has central processing unit, memorizer, communication interface；Several described calculating nodes (CPN) constitute distributed
Calculate network；All adjacent with in its topology calculating node (CPN) of each calculating node (CPN) carries out data interaction；Institute
Stating calculating node (CPN) and have space attribute, described space attribute is presented as the absolute of described calculating node (CPN) place
Locus and/or the described calculating node (CPN) relative tertiary location in residing topological network；Described calculating node
Built In Operating System, described operating system provides api interface, and user can be by described api interface by various management/controls
Demand and/or strategy are converted into criterion calculation sequence；Calculating node (CPN) in described distributed computing network in a distributed manner,
Jointly complete the described sequence of calculation to self-organizing.
It addition, the calculating node (CPN) for distributed computing network according to the above embodiment of the present invention can also have as
Lower additional technical characteristic:
Further, described operating system supports the parallel computation of multiple criterion calculation sequences.
Further, the multiple function subnet of definable on described distributed computing network, each described calculating node (CPN)
Different function subnets can be under the jurisdiction of, be independent of each other between described function subnet.
Further, described calculating node (CPN) is associated to certain fundamental space unit or relevant with certain electromechanical equipment
Connection, the relevant information of described fundamental space unit or electromechanical equipment all describes with the form of normal data table, described normal data table
Form a set of standard information collection.
Further, described calculating node (CPN) based on the described automatic identification of standard information collection, it associated is specially certain
Fundamental space unit or certain electromechanical equipment, thus realize calculating the plug and play of node (CPN).
Further, described operating system provides the algorithms libraries at different levels from simple mathematical calculations to professional application algorithm, and concrete can
To be Operator Library and/or basic algorithm storehouse and/or advanced algorithm storehouse, user can call described when writing described criterion calculation sequence
Algorithm in algorithms library, described operating system automatically forms underlying programs code, thus realizes Agility programming.
It should be noted that described criterion calculation sequence is carried out according to standardized mode upon execution, specifically it is exactly every kind
The non-stop layer algorithm calculated, the definition of its input/output interface parameter determines that, is systematized.
Further, described Operator Library includes: add, and subtracts, and takes advantage of, and removes, weighted sum, long-pending, logical operations, maximizing,
Minima, set operation, spanning tree, Jacobi/ Gauss Saden that iteration, and other common basic mathematical computings；Described base
Plinth algorithms library includes: matrix calculus algorithm, steepest descent method, Newton method, genetic algorithm, neuron algorithm, and other are common
Underlying mathematical algorithms；Described advanced algorithm storehouse includes: sensor fault diagnosis algorithm, and algorithm, fire inverting are checked in number distribution
Algorithm, CFD algorithm based on region, other are for the advanced algorithm of various professional fields.
Further, described calculating node (CPN) particular by certain area control system and certain fundamental space unit or
Certain electromechanical equipment of person is associated, and described area control system is for collecting the relevant letter of described fundamental space unit or electromechanical equipment
Cease or for controlling the executor relevant to described fundamental space unit or electromechanical equipment.
Further, described calculating node (CPN) has several described communication interfaces, described communication interface be divided into A class and
B class；Described calculating node (CPN) is entered by the calculating node (CPN) that described A class communication interface is adjacent with in its topology
Row data interaction；Described calculating node (CPN) is entered with described area control system (DCS) by described B class communication interface
Row data interaction.
Further, described calculating node (CPN) uses when access network in local naming's mode, its title and its topology
Adjacent calculating node (CPN) is different, and in topology, non-conterminous calculating node (CPN) can have identical title.
Further, described api interface is specially the generic interface of a kind of api interface based on communication protocol or other forms.
Further, described criterion calculation sequence includes several computing units, and the definition of described criterion calculation sequence includes as follows
Content: the logical flow chart between several computing units described；Operator that each described computing unit relates to and/or algorithm,
Input variable, output variable, calculation process and/or step.
The calculating node (CPN) for distributed computing network that the present invention provides has a characteristic that
1, space-oriented: each calculating node (CPN) all with a fundamental space unit or the region of certain electromechanical equipment
Control system is interrelated, when calculating node (CPN) and being associated with described area control system, and fundamental space unit or electromechanics
Locational space information, relative position relation or the topological relation that equipment is had the most naturally is embodied in described calculating node
(CPN) on, thus the advantage with rapid deployment, it is possible to remove the magnanimity of original control system from and field wiring repeatedly,
Work adaptive, that debug, define, saves a large amount of manpower；
2, standardization: the relevant information of described fundamental space unit or electromechanical equipment is all retouched with the form of normal data table
State, calculate calculating node after node associates with area control system can automatically identify its associated be fundamental space unit or
Certain electromechanical equipment of person, it is thus possible to realize calculating node (CPN) plug and play, automatically identify；
3, non-stop layer calculates: the whole calculating network system being made up of the calculating node (CPN) for distributed computing network
It is flattening, non-stop layer, the complete equality in status of each node, distributed by the data interaction between each node
Complete the overall situation calculating, the various management and controlling tacticses run on the system are to be embodied by described distributed calculating
With complete；
4, the most friendly programmed environment: the system being made up of the calculating node (CPN) for distributed computing network provides
One opening, the programming platform of hommization, operator/algorithms library that user utilizes native system to provide can complete easily event/
The definition of task, system automatically compiles underlying programs code, thus realizes controlling the rapid software code of management strategy, has quick
The advantage of prompt exploitation；Described programming platform can be developed the application program of magnanimity, there is the most compatible and motility.
Accompanying drawing explanation
Above-mentioned and/or the additional aspect of the present invention and advantage the accompanying drawings below description to embodiment will be apparent from from combining and
Easy to understand, wherein:
Fig. 1 is the organisational chart of turn-key system of the prior art.
Fig. 2 is the structural representation of the distributed computing network system of embodiments of the invention.
Fig. 3 is the first implementation schematic diagram calculating node and the communication modes of DCS system of embodiments of the invention.
Fig. 4 is the another kind of implementation schematic diagram calculating node and the communication modes of DCS system of embodiments of the invention.
Fig. 5 is the structural representation of the calculating node (CPN) of embodiments of the invention.
Fig. 6 is the schematic diagram of the library module of the line module of the operating system of embodiments of the invention.
Fig. 7 is the schematic diagram of the user interactive module of the line module of the operating system of embodiments of the invention.
Fig. 8 is the schematic diagram of the kernel module of the operating system of embodiments of the invention.
Fig. 9 is topology and the naming rule schematic diagram of the distributed computing network system of embodiments of the invention.
Figure 10 is the structural representation of the communication frame between the line module of embodiments of the invention and kernel module.
Figure 11 is the principle schematic representing information from three dimensions.
Figure 12 is the composition schematic diagram of an event of embodiments of the invention.
Figure 13 is the schematic diagram of the example 1 of embodiments of the invention.
Figure 14 is another schematic diagram of the example 1 of embodiments of the invention.
Detailed description of the invention
Basic structure:
The distribution being made up of the calculating node (CPN) for distributed computing network of the present invention is shown in conjunction with Fig. 2, Fig. 2
Formula calculates the structure of network system, calculates network system and is calculated node (CPN, computing node) composition by several,
All calculating nodes have collectively constituted the acentric calculating network of a flattening, and the status of each calculating node is all equality
's.
Calculating node is to have information reception, process, the computer (such as minicomputer) of sending function, calculates the knot of node
Structure and composition are as it is shown in figure 5, each calculating node is respectively provided with processor 1, memorizer 2 and communication interface 3.
The calculating node that each calculating node is all adjacent with in its topology carries out data interaction, and data interaction is a jumping communication, passes through
Calculating node adjacent with in its topology again after the processed of information carries out data interaction, and described data interaction is also that a jumping is led to
Letter, completes calculating task to all of calculating Node distribution formula jointly；That is the distributed calculating network that the present invention provides
Task is disassembled for basic calculating typical, that can replicate, and basic calculating is to be saved from neighbours by each calculating node
Obtain input information at Dian, complete local computing, then send result of calculation to neighbor node completes, during whole network does not has
The heart or the concept of head, cooperate to the calculating of completion system with the coordination mechanism of a kind of self-organizing between each node.
Calculating node (CPN) and have space attribute, space attribute is presented as the absolute space position calculating node (CPN) place
Put and/or calculate the node (CPN) relative tertiary location in residing topological network.
Further in conjunction with Fig. 5, distributed computing network system inline operations system, operating system includes kernel module and line module,
It is identical that the kernel module of operating system is distributed in each kernel module calculated in node, and each calculating node.
It can thus be appreciated that the network being interconnected and form by calculating node (CPN) defines a big computer, calculating process is distributed in
In each calculating node (CPN)；Single calculating node oneself can also be one and have comprehensive information processing, operation calculating
The computer of ability, but the ability of the information of process and calculating is more weak, and the system formed after the interconnection of multiple calculating nodes possesses more
The ability that strong information stores, calculates and communicate.
Calculate in the memorizer of node (CPN) and there is several cache modules, for store tasks and/or the state of event,
Cache module is specifically as follows 1024 or any number of, and system can the simultaneously multiple task of parallel processing or event, it is possible to suitable
Answer control task demand complicated and changeable.
In conjunction with Fig. 9, each calculating node (CPN), when access network, all uses the naming method of localization, i.e. each
The name calculating node (CPN) has only to distinguish adjacent node, is not required to carry out global naming or through routing configuration,
As it is shown in figure 9, the entitled node of 1,4 or 5 the most repeatedly occurs, and this can't affect and calculate the normal of network
Running, each calculating node (CPN) directly and only with the calculating node of direct neighbor in its topology can carry out data interaction,
Thus adjacent nodename difference, it is thus possible to effectively evade the name of magnanimity scene and the work of system configuration.
Each calculating node (CPN) is respectively provided with several communication interfaces, and communication interface is divided into A class and B class two kinds, calculates
Node is interconnected by the calculating node that A class communication interface is adjacent with in its topology, by B class communication interface and Region control system
System (District Control System) interconnection.And described area control system and certain fundamental space unit (son such as building
Space: office, a corridor etc.) or certain electromechanical equipment (such as cold, water pump etc.) be associated, Region control system
System is for collecting described subspace or all observing and controlling information of electromechanical equipment or relevant to subspace or electromechanical equipment for controlling
Controller or executor, calculate and realize information alternately according to standard information collection between node (CPN) and corresponding DCS, from
And can obtain at DCS calculate required for relevant information, or calculated result is sent to corresponding DCS so that
It performs relevant control.
The communication modes calculated between node (CPN) and area control system (DCS) includes but not limited to following two:
The first communication modes combines shown in Fig. 3:
The zone controller (DCU) calculated in node and DCS directly carries out data interaction, DCU again with each controller,
Sensor or executor carry out data interaction, described DCU and each controller, sensor or executor and constitute master slave relation.
The second communication modes combines shown in Fig. 4:
Zone controller (DCU) in DCS and each controller, sensor, executor or lamp, BAIYE, FCU intelligence
Equipment, for being co-located at (can be a LAN) on a Lan, calculates node mutual with Lan
Connection, between described CPN, DCU and each controller, sensor, executor or lamp, BAIYE, FCU smart machine
Mutual, without master slave relation between above-mentioned each parts to carry out information.
Communication modes between CPN and DCS supports multiple kinds so that CPN can with in currently available technology
All kinds of DCU products compatible.
Additionally i.e. can increase a module being equivalent to DCU in CPN by the Function Integration Mechanism of DCU to CPN,
This module is corresponding with the subspace of building or electromechanical equipment, collects relevant information or performs relevant control.
Topological relation and function subnet:
The network that each CPN interconnection is formed belongs to the tangible network in a physical significance, can determine on this network
Justice goes out multiple virtual function subnet, such as indoor transportation network, air flow network, heat transfer network, distribution system, air-conditioning
Water system, running water system, hot water supply system, air conditioning duct system, smoke-discharging and ventilating system, VMC (Ventilation Mechanical Control System), fuel gas supply system,
Cold station, thermal substation etc., can calculate inside individual feature subnet, it is also possible to be associated multiple function subnets calculating,
Be thus able to the information that realizes in multiple subnets truly share.
Function subnet is determined by the content specifically applied, is the virtual net on the distributed calculating network that provides of the present invention, its
Topological structure is consistent with physical network.
Owing to can define multiple virtual subnet on the calculating network of the present invention, the most same calculating node (CPN) may
Multiple function subnet can be under the jurisdiction of；It is independent of each other between multiple function subnets, thus calculates node and can simultaneously, concurrently locate
Manage the calculating task of multiple function subnet.
Because of the difference of application function, the effect that each calculating node (CPN) is played in function subnet is the most variant, such as empty
In flow of air subnet, the CPN of all space cells is coequal；And in air-conditioning box wind system the CPN of air-conditioning box with
The CPN in subordinate space then constitutes master slave relation.
So, system is the most naturally completed by defined function subnet on the distributed calculating network that provides in the present invention
The work of configuration, is defined calculation procedure further and just can complete various merit at sub-online operation calculation procedure by function subnet
Management needed for energy subnet, control task.
Term is explained:
The term involved by calculating network system that the present invention provides has a following implication:
Criterion calculation sequence: the api interface in the operating system that user is provided by distributed computing network is by all controls
Demand and the task of management transform into criterion calculation sequence.
Criterion calculation sequence includes several computing units, and computing unit is specially a task or event, therefore criterion calculation sequence
The definition of row includes following content:
Logical flow chart between several computing units；
Operator that each computing unit relates to and/or algorithm, input variable, output variable, calculation process and/or step.
Introduce the definition of task and event in detail below.
Task: be one section of control management strategy cooperated by multiple calculating nodes (CPN), by several events according to one
Fixed Sequence composition.
Event: be to calculate non-stop layer that node (CPN) completes to calculate, by definition operator/algorithm, input variable,
Output variable, intermediate variable determine.
Variable: each physical message that fundamental space unit is relevant and/or relevant each physical message of electromechanical equipment and/or user
The temporary variable of definition；Such as: the region area in the essential information of region, the indoor temperature in environment and systematic survey parameter,
The feedback of the luminaire on off state in smart machine operational factor, the setting value of region inner blower coil pipe running status, indoor
Ambient temperature setting value, the fault alarm of fire-fighting equipment, past 10 minutes inner region grand mean electric power etc..
Algorithm: operator (belonging to the subordinate concept of algorithm), basic algorithm, advanced algorithm or the User Defined that system provides
Algorithm；Support that algorithm is downloaded to calculate node (CPN) by user, support that system uses the mechanism that similar transmission calculates, will
Algorithm passes to each relevant calculating node (CPN)；Can be deleted by the effective range of definition algorithm and/or effective time
Except user-defined algorithm.
In conjunction with concrete example, above-mentioned concept can be made further explanation and explanation further below.
Standard information collection: calculating node (CPN) why can identify automatically, and plug and play.
Each information point related to during operational management control strategy typically has the information of three dimensions, as shown in figure 11,
Where what one dimension solved be/what is, the most described information point where be positioned at (include residing space, position, near dry
Road or branch road etc. are with the information of space attribute), what is (being a room or a water pump or cold), second dimension
What degree solved is that the concrete meaning representated by data that described information point is transmitted (represents temperature or humidity or energy consumption etc. are concrete
Content information), the 3rd dimension is as the passage of time, the change etc. that data occur.
The distributed calculating network system provided by the present invention, is combined with DCS or communication calculating node (CPN)
That time, DCS self with space attribute (DCS or be a fundamental space unit, i.e. one room or corridor,
Being that an electromechanical equipment, fundamental space unit or electromechanical equipment are natural has position or space attribute) the most naturally
Being mapped in the middle of CPN, the network formed after CPN networking is also the formation of an Internet of Things with space attribute；Cause
And solve the characterization problems of first dimension.
And the sign of second dimension is realized by data normalization:
All information points involved by fundamental space unit and all kinds of electromechanical equipment are standardized processing, and i.e. form a set of standard
Information collection, standard information collection is made up of several normal data tables, and each fundamental space unit or electromechanical equipment are corresponding
One normal data table (as shown in table 1), this normal data table definition needs to gather which variable, name variable, variable
Reference format etc., all of fundamental space unit or electromechanical equipment are all retouched according to this unified standard information collection
Stating, such CPN just can obtain the information of corresponding region or equipment, such as, standard after carrying out data interaction with DCS
5 numbers that information is concentrated are " room air temperatures ", if the connected equipment of DCS has temperature point, then the data of DCS
In storehouse 5 numbers just to should the measured value of sensor, thus 5 numbers being mapped in CPN data base；And work as bottom
When not having measuring point, 5 numbers in DCS database are in default value state, are mapped to 5 numbers in CPN and are also at writing from memory
Recognize value.Even if measuring control point changes, during such as more exchange device, increase and decrease measuring control point, as long as normal data table is constant, would not be right
Calculating node (CPN) to have any impact, use this form, no matter bottom uses what equipment, what communications protocol,
CPN can at DCS acquisition information；Also it is in this way, after CPN is connected with on-the-spot DCS, passes through
The mutual CPN of information is just capable of identify that corresponding to this DCS be space or certain electromechanical equipment, such as cold or water pump
Deng such that it is able to realize the plug and play of CPN, automatically identify.
Table 1 is the example of described normal data table.The ending of description embodiment is shown in by complete normal data table.
Table 1
The sign of the 3rd dimension be by DCS or CPN record storage functional realiey, it is possible to by described information point according to
The time Change and Development situation of passage is recorded, and repeats no more.
It follows that the present invention provide distributed calculating network system in calculating node can plug and play, Intelligent Recognition,
There is extremely strong robustness and motility.
The description of line module:
Line module is specially a kind of API based on communication protocol, described API and includes library module and user interactive module, i.e. grasp
Making system and provide the algorithms libraries at different levels from simple mathematical calculations to professional application algorithm, user is adjustable when the standard of compiling sequence of calculation
With the algorithm in algorithms library, operating system automatically forms underlying programs code, thus realizes Agility programming；In conjunction with Fig. 6, storehouse
The built-in three grades of operator/algorithms libraries of module, respectively Operator Library, basic algorithm storehouse and senior level algorithms library.
Operator Library includes: add, and subtracts, and takes advantage of, and removes, weighted sum, long-pending, logical operations (ask with or, non-etc.), asks maximum
Value, minima, set operation (seek common ground, union etc.), spanning tree, your iteration of Jacobi/ Gauss Saden, and other are common
Basic mathematical computing.
Basic algorithm storehouse includes: matrix calculus algorithm, steepest descent method, Newton method, genetic algorithm, neuron algorithm, other
Common underlying mathematical algorithms.
Advanced algorithm storehouse includes: sensor fault diagnosis algorithm, and algorithm is checked in number distribution, and fire inversion algorithm, based on region
CFD algorithm, other are applied to the advanced algorithm of various professional field.
In conjunction with Fig. 7, user interactive module provides the user with following interface: task definition interface, event definition interface, algorithm are fixed
Justice interface, variable-definition interface；
Under task definition interfaces, user can define following every in one or more: task names, task execution condition,
Task step sum, the event that task includes and the execution step of event, task return the computation structure that result includes.
Under event definition interface, user can define following every in one or more: which event title, event be under the jurisdiction of
The operator that individual task, event execution condition, the input variable of event, output variable, the title of intermediate variable, event are called/
Algorithm.
Under algorithm definition interfaces, user can define following every in one or more: which task/thing algorithm acts on
Part, algorithm title, the particular content of algorithm；
Under variable-definition interface, user can define following every in one or more: name variable, variable membership are in which
Individual task and/or event, varying byte length, initial guess.
It is understandable that above-mentioned enumerating only schematically, does not has limit to be also impossible to exclusive list.
Line module can use calculating node to be carrier, it would however also be possible to employ area control system is carrier, it is also possible to use additionally
Independent software and hardware be carrier, set-up mode is flexible and changeable.
The description of kernel module:
In conjunction with Fig. 8: described kernel module farther includes:
Calculate processing module: for performing a basic calculating；
Parallel computation collaboration services module: be used for supporting and serving parallel computation；Make described system can carry out multiple standard meter
Calculate the parallel computation of sequence.
Communication module: be used for communication task.
Parallel computation collaboration services module farther includes following submodule: task management module, event manager module, operator/
Algorithm management module, variable management module.
Communication module farther includes following submodule: bottom communication interface driver module, communications protocol parsing module, communication frame
Editor module.
There is between module following conspiracy relation:
Bottom communication interface driver module receives information at communication interface and the information received is stored in reception caching；Receive
Caching is positioned in the internal memory calculating node (CPN).
Communication protocol parsing module is from receiving caching gathering information and being the relevant letter of task by information processing after dissection process
Information after described process is also pressed by breath, information concerning events, operator/algorithm relevant information, four subclasses of variable relevant information
Classification is stored respectively in task management space, incident management space, operator/algorithm management space, variable management space, described
Task management space, incident management space, operator/algorithm management space, variable management space are respectively positioned on calculating node (CPN)
Internal memory in.
Task relevant information in task management space is managed and safeguards by task management module.
Information concerning events in incident management space is managed and safeguards by event manager module.
Operator in operator/algorithm management space/algorithm relevant information is managed and safeguards by operator/algorithm management module.
Variable relevant information in variable management space is managed and safeguards by variable management module.
Calculate processing module to manage space from task management space, incident management space, operator/algorithm management space and variable
Call relevant information and calculate, then the result calculated being stored respectively in task management space, thing according still further to classification
Part management space, operator/algorithm management space, variable management space.
Communication frame editor module from task management space, incident management space, operator/algorithm management space, variable management space
In recall relevant information and compile communication frame and be stored in transmission caching in, send caching be positioned at calculate node internal memory in.
The communication frame sent in caching is sent by bottom communication interface driver module by communication interface.
Kernel module and the communication of line module:
Carrying out information between kernel module and line module under certain communication protocol mutual, certain communication protocol can be existing skill
Various communications protocol present in art, it is also possible to be self-defining communications protocol, such as, combine shown in Figure 10, line module to
The communication frames that kernel module sends can use three-decker, and bottom is physical link layer protocol, including synchronous code, length, school
Testing part and data division, this physical link layer can use the communication skill of the various maturation such as Ethernet, Wifi, Zigbee
Art；The second layer is application bottom, the first and data division including frame, is used for processing the communications such as communication response, verification, sliced transmission
Problem；Third layer for application upper strata include type of message and message content part, message content part be used for processing task definition,
The problems such as the download of event definition, algorithm, variable assignments and system upgrade.The structure of application layer can be in the way of using other.
Line module utilizes in task that user defines by the form of the communication frame shown in Figure 10, event, algorithm, variable
One or more forms are sent to kernel module.
Example is explained:
The various operational management tasks and the strategy that run in the distributed computing network system that the present invention provides are all to pass through standard meter
Calculate sequence embody and complete, after network completes criterion calculation sequence, be just substantially finished task and the plan of operational management
Slightly, introducing calculating between node of native system below in conjunction with example is the Distributed Calculation that how to have cooperated.
Criterion calculation sequence is made up of some computing units, and computing unit refers to task or event, and if task be by
Secretary in charge of something's part is according to certain sequence composition, and event is to be calculated, by one, the basic calculating that node completes, and it is by input
Variable, output variable, intermediate variable (being likely not to have under some case) and operator/algorithm define, as shown in figure 12.
Example 1:
The automatic optimal of water pump pressure differential resetting value:
Problem describes: in building, the chilled water pump in air conditioning water system carries out runing adjustment according to end cooling situation, controls
Require on the premise of meeting each end refrigeration requirement, reduce the pressure differential resetting value of chilled water pump, to reduce pump energy consumption as far as possible.
Network structure: as shown in figure 13, the left-half of this figure shows that chilled water system controls network, figure grey water system end
Holding the network of the most multiple branch roads, simplify with a standpipe branch road and represent in figure, corresponding space nodes network is also multidimensional network knot
Structure, simplifies with chain structure in figure and represents, this chilled water system uses Primary pumps set-up mode, 2 corresponding 3 freezings of cold
Pump, chilled water coil pipe in water knockout drum flows to each room of end.The calculating of cold and water pump each corresponding device type
Node, end coil pipe is under the jurisdiction of the calculating node of a space type, and the refrigerating plant room at cold place arranges a space type
Calculating node, between described calculating node, interconnection forms a function subnet, the right half part institute of its topological structure such as Figure 13
Show.
Control logic: if there being the end-equipment water yield inadequate, then improve pressure differential resetting value according to the end quantity being unsatisfactory for requiring,
Increase pump rotary speed or increase water pump；In the case of all ends all meet requirement, if there being the end water yield excessive, then reduce
Pressure differential resetting value, thus reduce pump rotary speed or switch off the pump.
Whole problem can be decomposed into two relatively independent control loops, and one is to adjust chilled water pump according to end supply situation
Two ends pressure differential resetting value, another is that water pump number of units and frequency that water pump group is run according to the regulation of pressure differential resetting value make pump energy consumption
Minimum.Control the most how loop determines that the problem of pressure differential resetting value is to introduce the calculating that the present invention provides herein for first
The calculating process of network system.
Calculating process:
This task that is automatically adjusted of the pressure differential resetting value of water pump is completed by three composition of matter, as shown in table 2, and wherein event 1
With 2 can parallel computation, the result of event 1 and 2 is as the input of event 3, and whole flow process is unidirectional to be carried out, without circulation, often mistake
A period of time (such as 5 minutes) performs once.
Table 2
Detailed process comprises the steps:
1, the space nodes that each end coil pipe is subordinate to is according to the situation of change of temperature in region, it is judged that the cold of one's respective area
Whether demand enough, and basic logic is: the current measured temperature in this region in the range of the control accuracy of desired temperature, and
The valve opening (or dutycycle) of region inner coil pipe is not completely turned off, then it is assumed that current cold can meet demand；If
In current region, measured temperature is higher than the control accuracy scope of desired temperature, and the valve opening of region inner coil pipe or dutycycle
Standard-sized sheet, or in a threshold value the highest, then it is assumed that the cold in current region can not meet demand, sends " the hottest " signal；
If Current Temperatures measured value is less than the control accuracy scope of desired temperature, and the valve opening of region inner coil pipe or dutycycle
Fully closed, or in a threshold value the lowest, then it is assumed that the cold of current region has exceeded the refrigeration requirement of reality, sends " too
Cold " signal；Can define two variablees, variable 1 is " the most overheated ", is to be entered as 1, is otherwise entered as 0, variable 2
For " the most supercool ", it is to be entered as 1, is otherwise entered as 0.
2, the space nodes of end forms a spanning tree, calculates node from first and initiates, the value of variable 1 or 2 is transmitted
To neighbor node, the number that neighbours are transmitted by neighbor node and local data " local summation ", then this result of calculation is transmitted
To next node, the most repeatedly, the summation of all nodes is just completed when being delivered to endpoint node；During whole,
Each node has been involved in read group total, but neither one node knows there are how many nodes in whole system actually.Whether statistics
Overheated event 1 and the most supercool event 2 of statistics can be with parallel computations.
3, after event 1 and event 2 have calculated, the result of calculation of above-mentioned two event passes to any one and calculates node
(can be the calculating node at refrigerating water pump place), this calculating node calculates new pressure differential resetting value i.e. thing according to following algorithm
Part 3.
In conjunction with Figure 14.If " the hottest " ratio of stub area is higher than 5%, the pressure differential resetting value of the newest pressure differential resetting value=originally
+1；If " the hottest " ratio of stub area is less than or equal to 5% but " the coldest " ratio is higher than 5%, the newest pressure differential resetting value=former
The pressure differential resetting value-1 come；If " the coldest " ratio of stub area and " the hottest " ratio are below or are equal to 5%, then pressure differential resetting
It is worth constant.
In this example, " the coldest " or " the hottest " variable both can be by calculating the system variable that network system provides,
It can also be user's custom variable when compiler；Described " local summation " is to be provided by the library module in line module
Basic operator, can be built in each calculating node, and user can directly call, it is not necessary to goes specifically to compile realization again
Code；The algorithm of described " spanning tree " also by calculating network offer, can directly invoke；And above-mentioned event 1,2,3
Concrete input variable, output variable, sequencing etc. can be compiled by user.
Example 2:
Statistician is distributed:
On calculating network system provided by the present invention, it is also possible to the personnel's distribution in statistics regional.It is connected in space
Region doorway at installation infrared detector, when there being people to leave from a-quadrant, enter B region time, this Infrared Detectors meeting
The signal of one number-1 of transmission is to the calculating node of a-quadrant, and calculating of a-quadrant can be by one's respective area after node receives signal
Number-1, can calculate node B to neighbours simultaneously and initiate a calculating, make the number+1 in B region, thus in described system
Upper emulation completes personnel in corresponding real process and enters the physical process in B region from a-quadrant.
All Infrared Detectorss communicating junction, region that are arranged on, can detect the redundant labor transfer situation in building in real time, and will
Signal is sent to the calculating node in the region of correspondence, triggers the real-time calculating of whole system, such that it is able to obtain the people in all regions
Member's number distribution.This information can based on other functional subsystems of information realization (such as air conditioning system, illuminator)
Optimal control.
Example 3:
Collect roomed power consumption values:
Each electromechanical equipment of system all can record the energy consumption of oneself by DCS or CPN, or by DCS or CPN
The operation time according to electromechanical equipment, calculate the power consumption of this equipment, thus the energy of all devices in each fundamental space unit
Consumption value can by described system statistics out.In the angle of property management personnel, have and the power consumption values in all regions is collected
Get up to compare the demand of analysis.
From the perspective of mathematical operation, collect roomed power consumption values be substantially the process that an overall situation seeks union.In reality
During the calculating on border, property management personnel the calculating node accessed initiates this energy consumption statistic task, and the overall situation asks union, CPN
The total energy consumption value of the involved equipment that variable is each fundamental space unit of operation.After task is initiated, by the meter of the task of initiation
Operator node spreads to the overall situation, and whole network system is initially formed a spanning tree, then each branch endpoint node by oneself
Power consumption values returns to its neighbor node；After the most each calculating node receives the power consumption values that neighbours pass over, corresponding to oneself
The power consumption values in region is also added, and repacks, and issues other neighbours (i.e. other neighbour outside input consumption information neighbor node
Occupy node)；In described calculating network system, all of calculating node all follows this principle, and the consumption information of final all nodes is all
Can transmit at the calculating node of task promoter, i.e. complete the overall situation and seek union, collect the energy of all fundamental space unit
Consumption is worth this task.
The calculating node (CPN) for distributed computing network that the present invention provides has a characteristic that
1, space-oriented: each calculating node (CPN) all with a fundamental space unit or the region of certain electromechanical equipment
Control system is interrelated, when calculating node (CPN) and being associated with described area control system, and fundamental space unit or electromechanics
Locational space information, relative position relation or the topological relation that equipment is had the most naturally is embodied in described calculating node
(CPN) on, thus the advantage with rapid deployment, it is possible to remove the magnanimity of original control system from and field wiring repeatedly,
Work adaptive, that debug, define, saves a large amount of manpower；
2, standardization: the relevant information of described fundamental space unit or electromechanical equipment is all retouched with the form of normal data table
State, calculate calculating node after node associates with area control system can automatically identify its associated be fundamental space unit or
Certain electromechanical equipment of person, it is thus possible to realize calculating node (CPN) plug and play, automatically identify；
3, non-stop layer calculates: the whole calculating network system being made up of the calculating node (CPN) for distributed computing network
It is flattening, non-stop layer, the complete equality in status of each node, distributed by the data interaction between each node
Complete the overall situation calculating, the various management and controlling tacticses run on the system are to be embodied by described distributed calculating
With complete；
4, the most friendly programmed environment: the system being made up of the calculating node (CPN) for distributed computing network provides
One opening, the programming platform of hommization, operator/algorithms library that user utilizes native system to provide can complete easily event/
The definition of task, system automatically compiles underlying programs code, thus realizes controlling the rapid software code of management strategy, has quick
The advantage of prompt exploitation；Described programming platform can be developed the application program of magnanimity, there is the most compatible and motility.
Complete normal data table is as follows, it is to be appreciated that the description that described normal data table is merely exemplary, not structure
The restriction of protection scope of the present invention in pairs:
A.1 space cell
Subordinate list A.1 space information collection describes
A.2 distributor
Subordinate list A.2 distributor information point describes
A.3 generating equipment
Subordinate list A.3 generating equipment information point describes
A.4 cold
Subordinate list A.4 Cooling and Heat Source information point describes
A.5 boiler
Subordinate list A.5 boiler information point describes
A.6 water pump
Subordinate list A.6 water pump information point describes
A.7 cooling tower
Subordinate list A.7 cooling tower information point describes
A.8 air-conditioning box
Subordinate list A.8 air-conditioning box information point describes
A.9 Fresh air handing device
Subordinate list A.9 Fresh air handing device information point describes
A.10 new wind heat reclaimer
Subordinate list A.10 new wind heat reclaimer information point describes
A.11 energy storage pool
Subordinate list A.11 energy storage pool information point describes
A.12 bypass valve
Subordinate list A.12 bypass valve information point describes
A.13 ventilation blower
Subordinate list A.13 ventilation blower information point describes
A.14 heat exchanger
Subordinate list A.14 heat exchanger information point describes
A.15 elevator
Subordinate list A.15 elevator information point describes
Claims (12)
1. the calculating node (CPN) for distributed computing network, it is characterised in that described calculating node (CPN)
For having information reception, process, the computer of sending function, described calculating node (CPN) has central processing unit, storage
Device, communication interface；
Several described calculating nodes (CPN) constitute distributed computing network；Each calculating node (CPN) is all topological with it
Upper adjacent calculating node (CPN) carries out data interaction；
Described calculating node (CPN) has space attribute, and described space attribute is presented as described calculating node (CPN) place
Absolute spatial position and/or the described calculating node (CPN) relative tertiary location in residing topological network；
Described calculating node Built In Operating System, described operating system provides api interface, and user can pass through described api interface
Demand and/or the strategy of various management/controls are converted into criterion calculation sequence；
Calculating node (CPN) in described distributed computing network is in a distributed manner, complete the described sequence of calculation to self-organizing jointly.
Calculating node (CPN) the most according to claim 1, it is characterised in that described operating system supports multiple standards
The parallel computation of the sequence of calculation.
Calculating node (CPN) the most according to claim 1 and 2, it is characterised in that at described distributed computing network
On the multiple function subnet of definable, each described calculating node (CPN) can be under the jurisdiction of different function subnets, described function
It is independent of each other between subnet.
4. according to the calculating node (CPN) described in any one of claims 1 to 3, it is characterised in that described calculating node (CPN)
It is associated with certain fundamental space unit or is associated with certain electromechanical equipment, described fundamental space unit or the phase of electromechanical equipment
Pass information all describes with the form of normal data table, and described normal data table forms a set of standard information collection.
Calculating node (CPN) the most according to claim 4, it is characterised in that described calculating node (CPN) based on
The described automatic identification of standard information collection its associated specially certain fundamental space unit or certain electromechanical equipment, thus realize meter
The plug and play of operator node (CPN).
6. according to the calculating node (CPN) described in any one of claim 1 to 5, it is characterised in that described operating system carries
For from the algorithms libraries at different levels of simple mathematical calculations to professional application algorithm, user can call when writing described criterion calculation sequence
Algorithm in described algorithms library, described operating system automatically forms underlying programs code, thus realizes Agility programming.
Calculating node (CPN) the most according to claim 6, it is characterised in that
Described Operator Library includes: add, and subtracts, and takes advantage of, and removes, weighted sum, long-pending, logical operations, maximizing, minima,
Set operation, spanning tree, Jacobi/ Gauss Saden that iteration, and other common basic mathematical computings；
Described basic algorithm storehouse includes: matrix calculus algorithm, steepest descent method, Newton method, genetic algorithm, neuron algorithm,
Other common underlying mathematical algorithms；
Described advanced algorithm storehouse includes: sensor fault diagnosis algorithm, number distribution check algorithm, fire inversion algorithm, based on
The CFD algorithm in region, other are for the advanced algorithm of various professional fields.
8. according to the calculating node (CPN) described in any one of claim 1 to 7, it is characterised in that described calculating node (CPN)
It is associated with certain fundamental space unit or certain electromechanical equipment particular by certain area control system, described Region control
System for collect described fundamental space unit or the relevant information of electromechanical equipment or for control with described fundamental space unit or
The executor that electromechanical equipment is relevant.
9. according to the calculating node (CPN) described in any one of claim 1 to 8, it is characterised in that described calculating node (CPN)
Having several described communication interfaces, described communication interface is divided into A class and B class；
Described calculating node (CPN) is carried out by the calculating node (CPN) that described A class communication interface is adjacent with in its topology
Data interaction；
Described calculating node (CPN) carries out data friendship by described B class communication interface and described area control system (DCS)
Mutually.
10. according to the calculating node (CPN) described in any one of claim 1 to 9, it is characterised in that described calculating node
(CPN) using local naming's mode when access network, the calculating node (CPN) that its title is adjacent from its topology is different,
In topology, non-conterminous calculating node (CPN) can have identical title.
11. according to the calculating node (CPN) described in any one of claim 1 to 10, it is characterised in that described api interface
It is specially a kind of api interface based on communication protocol or the generic interface of other forms.
12. calculating nodes (CPN) according to claim 1, it is characterised in that
Described criterion calculation sequence includes that several computing units, the definition of described criterion calculation sequence include following content:
Logical flow chart between several computing units described；
Operator that each described computing unit relates to and/or algorithm, input variable, output variable, calculation process and/or step.
Priority Applications (5)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510378076.2A CN106331037B (en) | 2015-06-30 | 2015-06-30 | Computing node for distributed computing network |
| JP2017568390A JP6742353B2 (en) | 2015-06-30 | 2016-06-01 | Distributed computing network system and computing nodes used in the system |
| US15/740,146 US10732588B2 (en) | 2015-06-30 | 2016-06-01 | Decentralized computing network system and computing processing node used for the same |
| PCT/CN2016/084332 WO2017000738A1 (en) | 2015-06-30 | 2016-06-01 | Distributed computing network system and computing node used therefor |
| EP16817102.3A EP3318938A4 (en) | 2015-06-30 | 2016-06-01 | DISTRIBUTED COMPUTER NETWORK SYSTEM AND COMPUTER NODE USED THEREFOR |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510378076.2A CN106331037B (en) | 2015-06-30 | 2015-06-30 | Computing node for distributed computing network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106331037A true CN106331037A (en) | 2017-01-11 |
| CN106331037B CN106331037B (en) | 2023-09-19 |
Family
ID=57726134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510378076.2A Active CN106331037B (en) | 2015-06-30 | 2015-06-30 | Computing node for distributed computing network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106331037B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106325229A (en) * | 2015-06-30 | 2017-01-11 | 邻元科技（北京）有限公司 | Distributed computing network system |
| CN107450434A (en) * | 2017-08-16 | 2017-12-08 | 山东建筑大学 | A kind of intelligent node, automatic building control system and method based on two grade network |
| CN110083072A (en) * | 2019-05-20 | 2019-08-02 | 中国建筑科学研究院有限公司 | A kind of fluid network intelligent control method, intelligent node and system |
| CN111343227A (en) * | 2018-12-18 | 2020-06-26 | 富士通株式会社 | Information processing apparatus, information processing method, and information processing program |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS62212763A (en) * | 1986-03-13 | 1987-09-18 | Fujitsu Ltd | Job execution method in computer network |
| EP1087567A2 (en) * | 1999-09-24 | 2001-03-28 | Xerox Corporation | Decentralized network system |
| US20020156932A1 (en) * | 2001-04-20 | 2002-10-24 | Marc Schneiderman | Method and apparatus for providing parallel execution of computing tasks in heterogeneous computing environments using autonomous mobile agents |
| US7031945B1 (en) * | 2000-07-24 | 2006-04-18 | Donner Irah H | System and method for reallocating and/or upgrading and/or rewarding tickets, other event admittance means, goods and/or services |
| US20080075020A1 (en) * | 2004-04-30 | 2008-03-27 | Daimlerchrysler Ag | Data Communications Network with a Decentralized Communications Management |
| US20080147234A1 (en) * | 2006-12-14 | 2008-06-19 | Palo Alto Research Center Incorporated | Module identification method and system for path connectivity in modular systems |
| KR20090124438A (en) * | 2008-05-30 | 2009-12-03 | 버츄얼빌더스 주식회사 | Intelligent Space Awareness Method and System |
| US7656822B1 (en) * | 2003-12-22 | 2010-02-02 | Sun Microsystems, Inc. | Method and apparatus for decentralized device and service description and discovery |
| CN101741696A (en) * | 2008-11-07 | 2010-06-16 | 中国测绘科学研究院 | A Multi-user Real-time Collaboration System in Distributed Geographical Information Environment |
| US7765307B1 (en) * | 2006-02-28 | 2010-07-27 | Symantec Operating Corporation | Bulk network transmissions using multiple connections primed to optimize transfer parameters |
| US20120143386A1 (en) * | 2010-12-02 | 2012-06-07 | Sony Corporation | Method and apparatus for control |
| US20130209583A1 (en) * | 2010-10-27 | 2013-08-15 | Nestec S.A. | Methods and compositions suitable for promoting healthy skin |
| US20130325525A1 (en) * | 2012-05-21 | 2013-12-05 | Boost3, Llc | Systems and methods for an integrated online portal and marketplace for event-related items |
| US20140351010A1 (en) * | 2008-11-14 | 2014-11-27 | Thinkeco Power Inc. | System and method of democratizing power to create a meta-exchange |
| CN104374060A (en) * | 2014-11-04 | 2015-02-25 | 清华大学 | Water pump system control method based on centerless network |
| CN104620183A (en) * | 2012-05-21 | 2015-05-13 | 泰塔制造有限责任公司 | Automation and motion control system using a distributed control model |
-
2015
- 2015-06-30 CN CN201510378076.2A patent/CN106331037B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS62212763A (en) * | 1986-03-13 | 1987-09-18 | Fujitsu Ltd | Job execution method in computer network |
| EP1087567A2 (en) * | 1999-09-24 | 2001-03-28 | Xerox Corporation | Decentralized network system |
| US6671737B1 (en) * | 1999-09-24 | 2003-12-30 | Xerox Corporation | Decentralized network system |
| US7031945B1 (en) * | 2000-07-24 | 2006-04-18 | Donner Irah H | System and method for reallocating and/or upgrading and/or rewarding tickets, other event admittance means, goods and/or services |
| US20020156932A1 (en) * | 2001-04-20 | 2002-10-24 | Marc Schneiderman | Method and apparatus for providing parallel execution of computing tasks in heterogeneous computing environments using autonomous mobile agents |
| US7656822B1 (en) * | 2003-12-22 | 2010-02-02 | Sun Microsystems, Inc. | Method and apparatus for decentralized device and service description and discovery |
| US20080075020A1 (en) * | 2004-04-30 | 2008-03-27 | Daimlerchrysler Ag | Data Communications Network with a Decentralized Communications Management |
| US7765307B1 (en) * | 2006-02-28 | 2010-07-27 | Symantec Operating Corporation | Bulk network transmissions using multiple connections primed to optimize transfer parameters |
| US20080147234A1 (en) * | 2006-12-14 | 2008-06-19 | Palo Alto Research Center Incorporated | Module identification method and system for path connectivity in modular systems |
| KR20090124438A (en) * | 2008-05-30 | 2009-12-03 | 버츄얼빌더스 주식회사 | Intelligent Space Awareness Method and System |
| CN101741696A (en) * | 2008-11-07 | 2010-06-16 | 中国测绘科学研究院 | A Multi-user Real-time Collaboration System in Distributed Geographical Information Environment |
| US20140351010A1 (en) * | 2008-11-14 | 2014-11-27 | Thinkeco Power Inc. | System and method of democratizing power to create a meta-exchange |
| US20130209583A1 (en) * | 2010-10-27 | 2013-08-15 | Nestec S.A. | Methods and compositions suitable for promoting healthy skin |
| US20120143386A1 (en) * | 2010-12-02 | 2012-06-07 | Sony Corporation | Method and apparatus for control |
| US20130325525A1 (en) * | 2012-05-21 | 2013-12-05 | Boost3, Llc | Systems and methods for an integrated online portal and marketplace for event-related items |
| CN104620183A (en) * | 2012-05-21 | 2015-05-13 | 泰塔制造有限责任公司 | Automation and motion control system using a distributed control model |
| CN104374060A (en) * | 2014-11-04 | 2015-02-25 | 清华大学 | Water pump system control method based on centerless network |
Non-Patent Citations (1)
| Title |
|---|
| 张发平;陈爱清;董长欣;: "基于时间序列的数控设备状态智能评判方法", 北京理工大学学报, no. 09 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106325229A (en) * | 2015-06-30 | 2017-01-11 | 邻元科技（北京）有限公司 | Distributed computing network system |
| CN106325229B (en) * | 2015-06-30 | 2020-03-17 | 邻元科技（北京）有限公司 | Distributed computing network system |
| CN107450434A (en) * | 2017-08-16 | 2017-12-08 | 山东建筑大学 | A kind of intelligent node, automatic building control system and method based on two grade network |
| CN107450434B (en) * | 2017-08-16 | 2021-02-19 | 山东建筑大学 | Building automatic control system and method based on secondary network |
| CN111343227A (en) * | 2018-12-18 | 2020-06-26 | 富士通株式会社 | Information processing apparatus, information processing method, and information processing program |
| CN111343227B (en) * | 2018-12-18 | 2022-12-13 | 富士通株式会社 | Information processing apparatus, information processing method, and information processing computer-readable storage medium |
| CN110083072A (en) * | 2019-05-20 | 2019-08-02 | 中国建筑科学研究院有限公司 | A kind of fluid network intelligent control method, intelligent node and system |
| CN110083072B (en) * | 2019-05-20 | 2021-12-03 | 中国建筑科学研究院有限公司 | A fluid network intelligent control method, intelligent node and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106331037B (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106325229A (en) | Distributed computing network system | |
| US10732588B2 (en) | Decentralized computing network system and computing processing node used for the same | |
| Bottaccioli et al. | Building energy modelling and monitoring by integration of IoT devices and building information models | |
| Yan et al. | When artificial intelligence meets building energy efficiency, a review focusing on zero energy building | |
| CN104238510B (en) | Intelligent building district management system based on Internet of things | |
| US11236917B2 (en) | Building control system with zone grouping based on predictive models | |
| Castaldo et al. | Uses of dynamic simulation to predict thermal‐energy performance of buildings and districts: a review | |
| JP6186409B2 (en) | Wireless communication network system and management method for managing multiple areas | |
| CN106331037A (en) | Computing node for distributed computing network | |
| Yang et al. | Building energy management systems | |
| CN108319172A (en) | A kind of green building energy management method based on network without center | |
| CN103809541B (en) | For building the control system with groups of building and method | |
| CN205864470U (en) | A kind of calculating node for distributed computing network and distributed computing network | |
| Wang et al. | A decentralized swarm intelligence algorithm for global optimization of HVAC system | |
| Noye et al. | A wireless sensor network prototype for post-occupancy troubleshooting of building systems | |
| Doellner et al. | Towards concepts for climate and energy-oriented digital twins for buildings | |
| Zhang et al. | A local area network-based insect intelligent building platform | |
| Gonzalez-Gil et al. | Interoperable and intelligent architecture for smart buildings | |
| CN119047103A (en) | Pipeline path determining method and device, storage medium and electronic equipment | |
| Goyal et al. | An agent-based test bed for building controls | |
| Arnesano et al. | An iot solution for energy management at building and district level | |
| CN206055876U (en) | Central air-conditioning system and control system thereof | |
| Dymora et al. | Sensor network infrastructure for intelligent building monitoring and management system | |
| Pattarello et al. | Model predictive control of HVAC systems: design and implementation on a real case study | |
| Building load control and optimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |