WO2014091419A1

WO2014091419A1 - A method of configuration for control application in a control system

Info

Publication number: WO2014091419A1
Application number: PCT/IB2013/060800
Authority: WO
Inventors: Abhilash G; Marius STANICA; Pietro Danelli
Original assignee: ABB Technology AG
Current assignee: ABB Technology AG
Priority date: 2012-12-13
Filing date: 2013-12-11
Publication date: 2014-06-19
Anticipated expiration: 2015-06-13

Abstract

A method of configuring a control application in a control system is provided. The method comprises of steps of identifying a computing device from a plurality of computing devices as a manage node device and configuring it for analyzing of configuration; having a configuration of a control logic for a computing device sent to the computing device; having the computing device detect a change in the configuration of the control logic; having the computing device send a notification on the control system network to the manage node device on detection of a change in the configuration of the control logic; having the manage node device analyze the change in the configuration of the control logic for optimized distribution of function blocks and communication configuration related to one or more computing devices connected in the control system network; having the one or more of the computing devices configured for optimized distribution of the function blocks and communication configuration based on the analysis carried out by the manage node device. The method is used for configuration of control system in a power plant.

Description

A METHOD OF CONFIGURATION FOR CONTROL APPLICATION IN A CONTROL SYSTEM

FIELD OF INVENTION

The invention generally relates to the field of communication networks and control systems and specifically relates to a method of configuration for control application in control systems.

BACK GROUND OF THE INVENTION

Power plants may be built with a starting capacity say 500MW and at a later point, the capacity may be increased to say 670MW and further to 1 GW capacity to suit a growing need or to have an attractive payback. With such an incremental growth of capacity in a power plant, the associated control system which controls the power plant also needs to be grown in conjunction with the plant. As well it could so happen that the automation plan may also have its priority items/areas of plant to be included in the first increment and then in the next increment phase and so on. Applications which can take advantage of this incremental strategy are called as incremental applications and these can minimize the costs of implementation and may use hardware that is already in place and thus reduce the implementation time and testing time with minimum effect on operation of the already commissioned part of the plant.

One of the problems with incremental application building and refurbishment is re-configuration. This could also be the case when refurbishing an existing plant with use of new systems. Refurbishment and reconfiguration are often accompanied with downtime that needs to be minimized for better system performance. Another aspect is in exploiting the capabilities of available equipment which include servers, controllers, Human Machine Interfaces (HMI), Input/ Output (I/O) devices and communication switches and so on. But the effort required to interconnect the newer parts of the application to the existing parts in the plant could turn out to be a daunting task.

Today this could be done with some effort at engineering level by interlinking the control logic with communication network for a control system network. For example making use of communication function blocks, but the Ethernet architectures pose several limitations of not being able to configure the Ethernet traffic below the application layer and to use the advantages of Ethernet based communication to the full. This mechanism could not be used to reconfigure the GOOSE (Generic object oriented substation event) control block subscriptions or any multicast based mechanisms using layer 2 of Ethernet communication.

In a general control system present in a plant, data from field devices is collected using field protocols and the data is provided through communication modules to a Process Control unit (PCU) consisting of controllers and switches (network switching devices). The PCU makes the data available on a data bus or plant bus which is available for Human Machine Interface (HMI). This process involves communications such as Controller to I/O Communication, Controller to Controller Communication, Process Control Unit to Process Control unit Communication, Controller to external device or system communication, Controller to HMI Communication, and HMI to HMI Communication.

The configuration for setting up the communications can be done using an engineering tool but involves many manual steps that need to be performed by engineers to install involved devices after the device hardware is made available. Some parameters related with communication settings could be pre-configured along with function blocks used for process engineering but this may be limited. Moreover such pre-configured concept may have dependencies i.e., they may work as long as all communication works via controllers. In general, setting up communication could involve setting the parameters such as IP Address, Logical MAC Address, Redundant Server Settings, Exception Reporting Publisher Subscriber settings or association between multiple devices to exchange information etc.

The limitations aforesaid can be overcome with a newer architecture of control system which allows I/O devices to be available on a Plant data bus, and Field Devices having connection capabilities to get connected using either wired or wireless protocols in a plant network. Thus, in a newer control system to control an electrical network, electrical relays and other control blocks can directly be a part of the plant network along with a controller. However, control systems need additional features for incremental configuration and to fully exploit the new architecture. BRIEF DESCRIPTION OF THE INVENTION

The invention provides a method of configuring a control application in a control system. The control system comprises of multiple computing devices, each computing device having associated control logic made using one or more function blocks, and a communication stack for communication in the control system network. The method comprises of:

a. identifying a computing device from the plurality of devices as a manage node device and configuring it for analyzing configuration;

b. having a configuration of a control logic for a computing device sent to the computing device;

c. having the computing device detect a change in the configuration of the control logic;

d. having the computing device send a notification on the control system to the device identified as manage node device on detecting of a change in the configuration of the control logic in the computing device ;

e. having the manage node device analyze the change in the configuration of the control logic for distribution of function blocks and for communication configuration of one or more computing devices connected in the control system network; and

f. having the one or more of the computing devices from the plurality of the computing devices configured as per the distribution of the function blocks and communication configuration received from the manage node device.

As an embodiment, the mange node device analyze the change in configuration based on performance of the devices to satisfy the requirements of the control application based on timing and criticality of control functions. In another embodiment, the analysis of change in configuration of control logic includes collecting information about connected computing devices on the control system network. Also, the analysis can be carried out by calculating turn-around time on the communication network based on the physical location of the computing devices on the control system network.

In yet another embodiment, the configuration of the control system network includes reconfiguring IP Address of one or more devices from the plurality of devices, or Identifying the Virtual LAN for arranging the devices in virtual rings, or registering to notifications like events and alarms.

In yet another embodiment, the configuring of the control system network includes reconfiguring of communication stack associated with the plurality of computing devices.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 : Figure showing an exemplary method of the invention;

Figure 2: Figure showing control system specific layers of communication stack running at two instances, one operating on layer 3 on TCP/IP layers and the other operating on Layer 2 which is able to accept data directly from Ethernet NIC cards;

Figure 3: Figure showing a power plant model with incremented capacity;

Figure 4: Figure showing a tripped Generator in the plant leading to load shedding by the control system for stability; and

Figure 5: Figure the showing the sequence of steps of the proposed method.

DETAILED DESCRIPTION

The invention discloses a method for configuring control application in a control system. The communication configuration assumes the following:

a. Existence of Ethernet based bus means communication need not always pass through controllers, all devices could participate in a two way communication directly. b. Configuration of logical organizations of VLANs or SubNetworks to control and manage the visibility of messages on the Plant data bus. c. Message Prioritization to ensure services match the capabilities of proprietary communication interfaces and connections existing in plants. d. Access Points and Configuration of Association between multiple access points is done as part of Message Exchange e. When reconfiguring the data or events are not lost and reliability and availability is as good as proprietary networks.

The user is authenticated into the system and has rights to perform the control logic configuration and communication configuration changes. The Devices (Control, I/O and HMI with computing capabilities) are chosen and control logics are built by the user by using standard device/control Libraries. The libraries includes the device type, function block type and communication block information.

The computing devices (PCU) need to support Ethernet or wireless communication or any other protocol that supports publisher subscriber mode and request response modes of information exchange.

Figure 1 illustrates an exemplary embodiment for the method of configuration for control application in a control system. The method involves configuration activity initiated from an Engineering Station (1 10) connected in a communication network that connect to numerous devices in a control system. The engineering station is used to perform engineering and configuring tasks, and also for distribution of engineered function blocks comprising a control logic to the device in a control system that is configured or engineered.

A control system is said to be consisting of several nodes and each node in a network (control system network) comes with a computing device. In the exemplary embodiment, several computing devices (130, 150, 1 70 in Figure 1 ). Each devices are computing devices in the communication network (control system network) having processing capabilities and resources. Each device also consists of communication stack comprising of physical interfaces (135) which could be network interface cards, layers 2 (137) and layer 3 (140) representing MAC/Data Link Layer and Application Protocol layers respectively. The nodes have communication stacks comprising of layer 2 and layer 3 and control system specific layers for control system communication, applications (145) , data storage/database (147) and agents (120) running on the nodes.

As mentioned earlier, one of the computing devices at device level or station level is made to act as a Manage Node Device (150) that runs a management agent and acts as aggregator for communication of all computing device. The Management Node Device can be a computing system, for example a server. The flow of data and events across the computing devices are bidirectional. The configuration of application in the control system using the method of invention (200) is carried out using the following steps (refer Figure 2):

Step1 (210): When an engineer using the engineering station (1 10) creates a control logic or makes any change to function blocks that constitute the control logic, and downloads the function block network to the associated computing device ( say controller device 1 , 130) in the control system network, it also sets a variable (object property of SNMP or a GOOSE dataset) called latest update timestamp.

The agent (120) in the computing Device (130) detects a change in the timestamp value and thereby detects that a change has occurred in the control logic.

Step2 (220): The agent (120) in device 130 now sends a notification on communication subsystem which has a computing device identified as Manage Node device (1 50), the Management Node Device can be a computing system(for example a server). The Manage Node Device receives this notification over the communication network as an event message (a multicast or asynchronous message). An example is a SNMP trap message or a GOOSE multicast message notifying a change in system configuration.

In case of IEC 61850 protocol is supported in the communication network, the multicast capability of GOOSE messages to communicate across the network may be used. GOOSE Message in such a case comes with a Dataset containing the 'NotifyOnConfigChange' configured from GOOSE Control Block and going to all the subscribers configured for the block.

Step3 (230): The Manage Node Device (management agent (160)) identifies the impact of the change. The impact of the change can be analyzed by analyzing the configuration data prior to making the update. The analysis is done by analyzing the function blocks and based on CPU/resources of the computing devices, it checks if there is a need to distribute/redistribute any function blocks or functions across a wider system network. On identifying the need for such distribution, the Manage Node Device then analyzes the entire network of resources for an optimal distribution.

The Manage Node Device performs the analysis based on the following changes:

1 . System Level Logic changes

a. Time synchronization changes

b. Interfaces would include external and internal interfaces for example CPU = processor interface, Ethernet card communication interface, support for a specific protocol (say SNTP) is seen as full interface support for the protocol. These are available based on device type information

c. Event and alarm flow related changes

2. Device Level Logic changes

a. Sequence of Events Changes

b. Protection function changes

c. Supervision function changes

3. The reorganized function network could mean

a. Time Synchronization Blocks are placed in a different device

b. Function Blocks for I/O Handling based on interface type is distributed across devices

c. Function Blocks responsible for HMI interaction

d. Function Blocks responsible for decision or main functions

4. The system starts by defining a configuration based on existing configuration and adding the communication elements towards the end of each device end point and analyzes the performance to satisfy the requirements of the application with respect to timing and criticality.

a. System detects the start time and from the input to the output of Time Synchronization roundtrip, the system analyses the execution sequence and checks for the time that it takes for the execution.

b. Considering that the configuration will go into multiple devices seeing virtual device boxes and considering a factor of communication overhead of 20%, the execution times are compared with the total reaction times of the fast load shedding requirements <100ms

c. The analysis continues iteratively until the conditions of the solution criteria are satisfied.

Step4 (240): The manage node device arrives at an optimal distribution of function blocks, for example the analysis can be based on turn-around time/round trip times on the network based on the physical location of the computing devices in the control system network. The Manage Node Device collects this information from the Computing devices and performs a detailed analysis as mentioned in Step 3, for having an optimal configuration of communication network. The configuration of communication could include modifying the IP Addresses, or Identifying the VLAN's (VLAN IDs) for organizing the devices in virtual rings, or re-register to notifications like events and alarms. This distribution of function blocks according to logical changes leads to an optimal communication path.

For example if the incremental applications needs another processing resources, it can still be added without really affecting the existing configuration.

Step5 (250): The agent on Manage Node Device (150) now initiates the configuration of communication (540) and also distribution of the function blocks by setting the values of the computing devices for the respective changes into the network object (SNMP object or GOOSE dataset). In addition a variable <netreconf> gets set on the private sections of SNMP structure or GOOSE dataset to reset the stacks.

The method includes using a shared memory to hold data which is passed to Control System Specific Communication Stack. The Control System Specific Communication layers can be run at two instances. The following process shows reconfiguration of the stack on the computing devices.

i) The agent in devices in turn initiates an application which performs the stack reconfiguration.

ii) With reference to Figure 3, the application initializes the Layer 2 (320), lower layer based communication component which will continue to support event and data flow when the layer 3 (330) based communication stack is restarted. Once the layer 2 component is ready using a flag the initialization completed status is indicated to the application. iii) The application now initiates a shutdown call on the layer 3 (330) or application layer communication using the commands. iv) In order to perform reconfiguration with minimal downtime MMRP (Multiple

MAC Registration Protocol) and GMRP (GARP Multicast Registration Protocol) can be used which allows the reconfiguration dynamically without restarting stacks. These would need VLAN aware switches already used in control systems.

v) In case of IEC 61850, the messages would lead to the SCD file getting updated with GOOSE notifications and thereby the stacks to be aware of these changes and can send the multicast group,

vi) Once the Layer 3 components are successfully initialized, the flag is set to application and the application now sets the flag which leads to redirection from layer 2 component to stop and layer 3 component takes over.

This method leads to an intelligent configuration of the Control system network. The method eliminates the concept of Communication Function Blocks and lets Communication Stand independent but automatically taken care based on impact of control network changes. This method provides a mechanism to ensure availability of the Control system function with zero loss of data and events while the configuration takes place.

As illustrative example of configuration for control application in a control system,

Considering an example of a power plant with increased capacity and configuring an incremental application into the power plant control system.

Figure 4 shows a power plant substation as an example. In this example the power plant 400 comprising of generators for power production and motors as load with circuit breakers to control feeding of power from generators, is started with 500 MW capacity and then additional capacities are added to the plant increasing the capacity to 670MW by adding generators 410. To have the plant controlled with latest available control system features, the automation system deployment needs to be optimally planned to minimize new cost implications or squeeze functions. For example, an upgrade strategy can be to have more PCU added and have the existing HMI extended to include visualization of events and alarms.

In this example of capacity addition, the upgrade strategy would mean new field equipment (410, 420) which when added has to be operated for control and protection functions by the deployed automation system. Adding capacities also mean new auxiliaries appearing in the control system and giving rise to new monitoring and protection scenarios. Moreover there is energy efficiency logic which has to be extended to comprise inputs from newly added PCUs of the automation system. The upgrade will result in a new feeder comprising of a generator, Circuit breakers getting added as primary equipment and also new Relays or Intelligent Electronic Devices (IED) or Controllers getting added as secondary equipment.

One of the essential control functions in a substation of a power plant is to handle the load. It is done by managing (reducing or shedding) the load on the system depending upon the available generation. In a power plant when a generator trips, based on its capacity, a certain load is shed by isolating a part of the system by the control system available on the plant, thereby avoiding system failure. The load shedding or power balance management is controlled by a power management controller in the electrical network system. The power management controller may be an independent controller device or a process controller in a power plant that is managing processes of the power plant.

To support load shedding in the upgraded plant provided with new generation facility 410, additional switchgears have to be installed. It is required to define the loads based on its priorities to identify critical loads and accordingly carry out load shedding management. For fast load shedding application, it is important that the data of the switch gear is available in the action point within say 100ms. The exchange of data about the newly added switchgear including the quality of the signals have to be managed through the Power management controller. This requires configuration of communication to facilitate this. The change in configuration of communication includes configuration of GOOSE Control Block in the controller to be active and the new additional switch gear equipment like the lEDs needs to be registered as subscribers in the GOOSE Control Block and the dataset needs to be defined which would contain the information that needs to be exchanged for having fast load shedding.

The configuration for control application in control system of the power plant using the method of the invention can be done in the following manner:

The moment control logic is configured for load shedding in the upgraded substation, the controller has a change in the control logic. In an electrical integration case like load shedding there involves Distributed Control System as well as Electrical systems/auxiliaries. New capacity in power plant means new feeders getting added. The system wide logic changes include

a) a new feeder which comprises of Generators, Circuit Breakers, lEDs and Controllers.

b) a new Time synchronization source

c) a new HMI for visualization

The device level logic changes include protection function change, or supervision function change or changing the parameters of the functions. Say a protection function we need to set the HMI related functions to be modified, the Sequence of Events (SOE) Slave, functions governing the motor. Function blocks examples includes Time Synch Blocks, Function Blocks related to I/O, Pulse Input/Period (PIPER) function Blocks, Analog Input Slave (AIS/FBS - Field Bus Slave), HMI Related Function Blocks, Analog Exception Report Function Blocks (AO/L), Sequence of Events Slave, Basic Function Blocks could include the AND Blocks, Remote Control Memory Blocks, Function Blocks related to Decision Making and Execution, Multi State Device Driver, and PID Error which is capable of providing proportional, integral and derivative actions on error signal from process variables and set point inputs.

The controller (device) which is affected by the change will update the latest update timestamp variable. The agent service running in the computing device identifies a change in timestamp variable and it identifies the need for communication change. For example the list of communication changes at system level includes publishing and subscription of GOOSE Control Block and the signals to be included in the dataset. As well as the time synchronization source to be added, the flow of all the time synch information across the network devices will have to be configured including the path including parameters of IP Addresses or MAC addresses, adding HMI means for reporting or exception reports to update HMI, the path needs to be added and events and data reach the new HMI. For device level changes, this could be triggering system level changes as well. Also supervision function changes like signals would translate to the reporting data and exception reports and their path of IP Address or MAC address. More signals to be connected to Sequence of Events would mean the resultant events and their paths need to be defined in the control system network.

The agent service running in the affected computing device then sends a notification on communication sub system which has one of the computing devices or operations/HMI devices identified as Manage Node Device. This notification is sent using multicast or broadcast approaches. For the devices like switches which can support SNMP, notification can be sent using an SNMP trap message containing <controlconfigchanged=true>. For the computing devices that support IEC 61850 GOOSE, it can be using IEC 61850 GOOSE updated with dataset <controlconfigchanged=true> to reflect that configuration has changed.

The management agent on the Manage Node Device runs analysis to identify the impact of the change. The Impact of the change is analyzed by comparing the previous configuration with the new configuration with respect to control logic and checks if there are additions of control logic and functions which would need communication changes. The Manage Node Device analyses the change in the control block functions for the fast load shedding and performs a detailed analysis for distribution/redistribution of the function blocks.

The Manage Node Device collects the information from the rest of the computing devices in the network and performs a detailed analysis for having and optimal configuration of the network. Aligned with the distribution an optimal communication configuration is prepared which involves modifying IP Addresses, subscriptions added for computing devices. This would include configuring the existing computing devices and newly added computing devices in the plant.

The Management Node Device initiates the reconfiguration by assigning the Change Parameters List as a dataset or set of parameters for each computing device and triggers the changes to be deployed on the automation control system network devices which are affected. Further to the ChangeParameters List, a variable called <netreconf> is set on every computing device which triggers the update of the parameters.

The Control System Specific Communication layers will be running at two instances by making use of two different layer. They would get data from the shared memory and this allows redirection. An example is 2 instances of IEC 61850 stack running, first one running over Layer 3, and another directly working from Layer 2. Layer 2 being below Layer 3, there is all possibility to handle data without missing any data or events. Thus the reconfigurations may be carried out efficiently.

While only certain features of the invention have been illustrated and described herein, many modifications and changes will occur to those skilled in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.

Claims

Claims:

1 . A method of configuring a control application in a control system comprising of plurality of computing devices, each computing device from the plurality of computing devices having associated control logic comprised from one or more function blocks, a communication stack for communication in the control system network, the method comprising of:

a. identifying a computing device from the plurality of computing devices as a manage node device and configuring it for analyzing configuration;

b. having a configuration of a control logic for a computing device sent to the computing device ;

c. having the computing device receiving the configuration of a control logic detect a change in the configuration of the control logic;

d. having the computing device send a notification on the control system network to the manage node device on detecting the change in the configuration of the control logic in the computing device;

e. having the manage node device analyze the change in the configuration of the control logic for distribution of function blocks and for communication configuration of one or more computing devices connected on the control system network; and

f. having one or more of the computing devices from the plurality of the computing devices configured as per the analysis carried out in the manage node device for distribution of the function blocks and communication configuration.

2. The method of claim 1 , wherein having the computing device to send a notification on the control system network as a multicast or as a broadcast.

3. The method of claim 1 , wherein having the manage node device analyze the change includes comparing one or more configurations corresponding to the change in the configuration of the control logic and checking the resources of the devices for distribution of the function blocks.

4. The method of claim 1 , wherein the mange node device analyze the change in a configuration based on performance of the computing devices to satisfy the requirements of a control application based on timing and criticality of function.

5. The method of claim 1 , wherein the manage node device analyze the change in the configuration of the control logic for distribution of function blocks comprises of collecting information of the computing devices connected on the control system network.

6. The method of claim 1 , wherein the manage node device analyze the communication configuration by calculating turn-around time on the control system network based on the physical location of the plurality of computing devices on the control system network.

7. The method of claim 1 , wherein configuring communication includes reconfiguring IP Address of one or more devices from the plurality of devices, or Identifying the Virtual LAN for arranging the devices in virtual rings, or registering to notifications like events and alarms.

8. The method of claim 1 , wherein configuring communication includes reconfiguring of communication stack associated with the computing devices.

9. The method of claim 1 , wherein the method is used in control system configuration in a power plant.