CN102841582B - A kind of power distribution network self-healing control system and its implementation - Google Patents

Abstract

The invention discloses a distribution network self-healing control system, which is characterized in that the system includes an entity Agent, a function Agent, an interface Agent and a translation Agent; the entity Agent includes two types: element type and set type; The elemental entity Agent is embedded in the controlled device; the interface Agent is embedded in the control terminal; the collective entity Agent, function Agent and translation Agent are embedded in the server; the entity Agent is connected to other systems through the translation Agent, and the interface Agent Connect to the dispatcher, through the processing tasks of the functional Agent, to ensure the normal operation of the controlled equipment. The multi-agent system framework of the invention solves the difficult problem faced by the development of the distribution network self-healing control system.

Description

A distribution network self-healing control system and its implementation method

technical field

The invention relates to a multi-agent system framework and an implementation method in the field of power systems, in particular to a distribution network self-healing control system and an implementation method thereof.

Background technique

At present, the self-healing control of the distribution network is facing huge challenges: there are many software and hardware devices, and the distribution range is wide. There may be unexpected situations such as equipment failure, system crash, and misoperation during work. The overall analysis and processing data volume is large, and the calculation process is complicated. The computing speed is not ideal, and it is difficult to introduce new technologies, new equipment, and new systems through plug-and-play; the traditional centralized software modeling method is difficult to solve the above problems. The distribution, integration, fault tolerance, autonomy, flexibility and scalability of MAS can solve the difficulties faced by the distribution network self-healing control, but its development involves many knowledge fields and the engineering theory is not mature enough. From top to bottom, the idea of centralized control still prevails, which leads to the decline of the most important autonomy of Agent, and adversely affects the integration, fault tolerance, flexibility and scalability of the system.

Contents of the invention

Aiming at the deficiencies of the prior art, the present invention provides a distribution network self-healing control system and its implementation method, which not only can integrate numerous devices and functions, but also has strong flexibility, scalability and fault tolerance, and can also solve On-site problems, prevent the expansion of calculation objects, and improve the real-time response capability of the system.

A distribution network self-healing control system provided by the present invention is improved in that the system includes an entity Agent, a function Agent, an interface Agent, and a translation Agent; the entity Agent includes an element-type entity Agent and a set-type entity Agent ;

Embedding the element-type entity Agent into the controlled device; embedding the interface Agent into the control terminal; embedding the aggregate-type entity Agent, function Agent and translation Agent into the server;

The collective entity Agent connects to other systems through the translation Agent, connects to the dispatcher through the interface Agent, and ensures the normal operation of the controlled equipment through the processing tasks of the functional Agent.

Wherein, the element-type entity Agent corresponds to the controlled device, and the set-type entity Agent corresponds to an analysis unit.

Wherein, the element-type entity Agent obtains the operation data of the device under test, and performs emergency treatment when the data is abnormal.

Wherein, the collective entity Agent is used for coordinating the relationship among subordinate agents, monitoring state changes and performing coordinated control.

Wherein, the functional Agent is used to assist the entity Agent to complete the control task; one functional Agent corresponds to a function realization method; the same Agent with different realization methods of the function forms an alliance; the alliance is handled by an endorsement Agent External affairs and internal affairs are assigned to the remaining functional Agents according to the internal mechanism of the alliance. Task allocation methods are: master-slave control, negotiation, contract, auction or coordination.

Wherein, the interface Agent includes: a device-type interface Agent corresponding to standard equipment, a terminal-type interface Agent mapped by a standard set, and a management-type interface Agent for processing MAS events; the standard equipment of the present invention is a subset of measurable and controllable equipment. It is a standard, mass-produced, measurable and controllable device.

The device-type interface Agent is used to set and display the parameters and status of the device under test;

The terminal-type interface Agent is used to provide a dispatcher's control terminal;

The management interface Agent is used to monitor the events of the system and handle all abnormal problems of the Agent.

Wherein, the translation agent is used for conversion between external system information and the agent information.

Among them, each Agent has a relational database and an XML file;

The relational database is used to store voltage values, current values, active values and reactive values; the voltage values, current values, active values and reactive values include real-time values and historical values;

The XML file is used to store voltage limit value, current limit value, line loss limit value, power factor limit value, delay time and so on.

Wherein, the element-type entity Agent includes a switch Agent, a capacitor Agent, a reactance Agent, a transformer Agent, a voltage regulator Agent, and a DGA Agent;

The switch Agent is a mapping model of a switch;

The capacitance Agent is a mapping model of a capacitor;

The reactance Agent is a mapping model of a reactor;

The transformer Agent is a mapping model of a transformer;

The voltage regulator Agent is a mapping model of the voltage regulator;

The DGAgent is a mapping model of distributed power generation.

Wherein, the collective entity Agent includes a feeder Agent; the feeder Agent is a mapping model composed of a set of measurable and controllable objects on a feeder.

Wherein, the functional Agent includes risk identification and control Agent, optimization identification and control Agent, failure analysis and processing Agent, power flow calculation Agent, network topology Agent and endorsement Agent;

The risk identification and control Agent is used for risk identification and risk control;

The optimal identification and control Agent is used for optimal identification and formulating an optimal control scheme;

The fault analysis and processing Agent is used to analyze faults, locate faults and formulate fault recovery schemes;

The power flow calculation Agent is used for power flow calculation;

The network topology Agent is used for topology analysis;

The endorsement Agent handles the external affairs of the alliance and assigns internal tasks of the alliance;

Wherein, the interface Agent includes a device Agent, a terminal Agent and a management Agent;

The device Agent is used to display the parameters of the standard element type entity Agent;

The terminal Agent is used to display the parameters of the standard collective entity Agent;

The management agent is used for automatically repairing the abnormally exiting agent, checking the running status of all agents, and manually controlling the life cycle of all agents. The life cycle includes creating, sleeping, waking up, killing, etc.

The present invention provides an implementation method based on the above-mentioned system structure based on another purpose, and its improvement is that the method includes the following steps:

(1) The entity Agent monitors the operating status data of the controlled equipment in real time and judges the operating status; interacts with other systems through the translation Agent; other systems can be distribution automation systems, etc., and its interaction is to obtain existing data;

(2) The entity Agent sends the feedback information and the request for assistance to the interface Agent;

(3) Each functional Agent communicates with each other, responds to the service request of the entity Agent and returns the result;

(3) The interface Agent obtains the feedback information and the request for assistance, and transmits the dispatcher's control instructions and the operating parameters set by the dispatcher;

(4) The entity Agent receives the dispatcher's instruction and the operating parameters to be adjusted, and adjusts the parameters of the controlled equipment.

Wherein, the controlled equipment includes switches, capacitors, reactors, distribution transformers, voltage regulators and distributed power sources.

Wherein, the data monitored by the entity Agent in step (1) in real time includes data such as active power, reactive power, voltage and current.

Among them, the entity Agent seeks assistance from the functional Agent to obtain the required feedback information;

The information that the entity Agent needs to feed back is the information for the entity Agent to judge the fault type and handle the fault;

The failure types include real-time failures and potential risks.

Among them, the steps for the entity Agent to handle real-time faults include:

(i) collecting the real-time active value data, real-time reactive value data, real-time voltage data and real-time current value data of the controlled equipment;

(ii) judging the state of the controlled device, if it is in an abnormal state, proceed to step (iii), otherwise end;

(iii) Execute emergency processing, that is, use the state obtained in ii as the search condition, search the rule base to obtain the processing plan;

(iv) Step (iii) If the plan is obtained, then request the relevant Agent to perform the corresponding operation, and the processing process ends after the operation is completed; if the plan is not obtained or the plan is rejected by other Agents, then proceed to step (v);

(v) In-depth processing of the organization: the entity Agent divides the task of handling this abnormal situation into subtasks and distributes it to other Agents for completion, and is responsible for controlling the process and merging results by itself; organization means that the Agent has the behavior to handle this situation Logic, it seeks help from other Agents according to its own needs; depth is relative to emergency/quick processing, which means that it can use all available resources to get the most correct results, and the implementation process has been reflected in the behavior logic , to get a solution.

Among them, the steps for entity Agent to deal with potential risks include:

① Obtain the active power, reactive power, voltage and current data of the controlled equipment; the active power, reactive power, voltage and current data include historical data, real-time data and forecast data;

②Judging whether there is a potential risk based on the data, if so, proceed to step ③, otherwise end;

③ In-depth processing of the organization: the entity Agent divides the task of handling this abnormal situation into subtasks and assigns them to other Agents to complete, and is responsible for controlling the process and merging results by itself; organization means that the Agent has the behavioral logic to handle this situation, It seeks help from other Agents according to its own needs; depth is relative to emergency/quick processing, which means that it can use all available resources to get the most correct results. The implementation process has been reflected in the behavior logic, and finally obtained solution.

Compared with the prior art, the beneficial effects of the present invention are:

The self-healing control system of the present invention can integrate a large number of distribution network equipment through the layered entity Agent structure, and the equipment can handle its own abnormal conditions, preventing the expansion of the calculation scale, thereby improving the real-time response capability, and ensuring the separation of entities and functions The scalable performance is achieved, the management interface Agent realizes the fault tolerance function, the operation matching between the interface and the entity improves the reusability and flexibility, and the translation Agent integrates other external software systems; thus, the multi-agent system framework solves the problem of configuration Difficulties faced in the development of power grid self-healing control system.

Description of drawings

Fig. 1 is an Agent type model of the distribution network self-healing control system provided by the present invention.

Fig. 2 is a working principle diagram of the entity Agent of the distribution network self-healing control system provided by the present invention.

Fig. 3 is a structural model of the distribution network self-healing control MAS provided by the present invention.

Fig. 4 is an agent interaction model of the distribution network self-healing control system provided by the present invention.

Fig. 5 is a structural model of the distribution network self-healing control MAS provided by the present invention.

Fig. 6 is a distribution network self-healing control MAS deployment model provided by the present invention.

Detailed ways

The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings.

Distribution network self-healing control MAS is composed of four types of members: Entity Agent, Functional Agent, Interface Agent and Translation Agent. Function Agent's processing tasks to ensure the normal operation of the controlled equipment. The entity Agent mainly analyzes and controls the distribution network equipment, the functional Agent performs system analysis and calculation, the interface Agent is the interface of the dispatcher's operating system, and the translation Agent is the bridge to communicate with other distribution systems.

Entity Agent is divided into element type and collection type. The measurable and controllable equipment in the distribution network corresponds to the element type, and the commonly used analysis units correspond to the collection type; an analysis unit can include many measurable and controllable equipment, and can also include some sub-analysis units, so the collection Types can also be elements of other collections. The element-type entity Agent directly obtains equipment operation data, has emergency processing functions, and has analysis and processing logic; it constantly evaluates its own status and mines possible risks; if any abnormal situation is found, it will immediately carry out emergency treatment to protect the equipment from damage; If the result of the emergency treatment is unsatisfactory or if a potential risk is discovered, the analysis and processing logic is used to organize other system members to solve the problem in depth and maintain the long-term normal operation of the equipment. The collective entity Agent coordinates the relationship between subordinate agents, does not participate in the internal data calculation of subordinates, but only cares about its affairs and state changes, and coordinates and controls on this basis, thereby avoiding the expansion of the calculation scale.

The functional Agent provides various services and assists the entity Agent to complete various analysis and control tasks. If there are multiple realization methods for a function, one method corresponds to one Agent, and the Agents of these methods form a coalition; in the coalition, a representative Agent centrally handles external affairs, and internal affairs are assigned to members according to alliance rules. When a certain function needs the support of other functions, its Agent will make a service request, and its working method is similar to that of an entity Agent requesting service.

There are three types of interface Agents: the equipment type corresponding to standard equipment, which mainly displays and sets equipment parameters and status; the terminal type mapped by the standard set, as the control terminal of the dispatcher, is the operation platform for many functions and control objects; processing MAS events The management type monitors system events, handles Agent exceptions, and can also be directly operated by the dispatcher. When the system is working, an interface Agent needs to match an application object, but there is not only one object that can be matched by an interface, on the contrary, an object can also correspond to multiple interfaces.

An external system corresponds to a translation agent, and the translation agent converts its communication language and agent communication language (Agent Communication Language, ACL), so that the two systems can transmit data and information to each other.

The implementation method of the distribution network self-healing control system of this embodiment includes the following steps:

(1) The entity Agent monitors the operating status data of the controlled equipment in real time and judges the operating status; interacts with other systems through the translation Agent; other systems can be distribution automation systems, etc., and its interaction is to obtain existing data;

(2) The entity Agent sends the feedback information and the request for assistance to the interface Agent;

(3) Each functional Agent communicates with each other, responds to the service request of the entity Agent and returns the result;

(3) The interface Agent obtains the feedback information and the request for assistance, and transmits the dispatcher's control instructions and the operating parameters set by the dispatcher;

(4) The entity Agent receives the dispatcher's instruction and the operating parameters to be adjusted, and adjusts the parameters of the controlled equipment.

In the distribution network self-healing control system entity Agent works in two parts, as shown in Figure 2. The flow chart on the left is a process of identifying and handling real-time faults, and the flow chart on the right is a process of identifying and handling potential risks;

The steps for the entity Agent to handle real-time faults include:

(i) collecting the real-time active value data, real-time reactive value data, real-time voltage data and real-time current value data of the controlled equipment;

(ii) judging the state of the controlled device, if it is in an abnormal state, proceed to step (iii), otherwise end;

(iii) Execute emergency processing, that is, use the state obtained in ii as the search condition, search the rule base to obtain the processing plan;

(iv) Step (iii) If the plan is obtained, then request the relevant Agent to perform the corresponding operation, and the processing process ends after the operation is completed; if the plan is not obtained or the plan is rejected by other Agents, then proceed to step (v);

(v) Organize in-depth processing; the entity Agent divides the task of handling this abnormal situation into subtasks and assigns it to other Agents to complete, and is responsible for controlling the process and merging results by itself; organization means that the Agent has the behavior to handle this situation Logic, it seeks help from other Agents according to its own needs; depth is relative to emergency/quick processing, which means that it can use all available resources to get the most correct results, and the implementation process has been reflected in the behavioral logic. Finally got the solution.

The steps for entity Agent to deal with potential risks include:

① Obtain data such as active power, reactive power, voltage, and current of the controlled equipment, which includes historical, real-time, and forecast data;

②Judging whether there is a potential risk based on the data, if so, proceed to step ③, otherwise end;

③ In-depth processing of the organization: the entity Agent divides the task of handling this abnormal situation into subtasks and assigns them to other Agents to complete, and is responsible for controlling the process and merging results by itself; organization means that the Agent has the behavioral logic to handle this situation, It seeks help from other Agents according to its own needs; depth is relative to emergency/quick processing, which means that it can use all available resources to get the most correct results. The implementation process has been reflected in the behavior logic, and finally obtained solution.

The entity Agent executes two processes cyclically respectively, the time interval of the fault handling process is relatively short, and the time interval of the risk control process is relatively long. In order to complete this process, the tasks undertaken by the entity Agent include: protecting the corresponding equipment from being damaged, maintaining the normal working state of the equipment, sending the equipment status to the dispatcher, who can set various parameters, and assist other Agents to complete the task if possible; function The task of the Agent is: to allow the dispatcher to set parameters and processing objects, provide its own functional services, and be able to find other functional support; the interface Agent displays data and information, provides a platform for the dispatcher to edit instructions, select plans, and formulate plans, and issue control instructions; The translation agent completes the mutual conversion between general communication language and ACL.

The distribution network self-healing control MAS structural model is shown in Figure 3. Agents of the entity type form a tree structure, and the measurable and controllable equipment corresponds to its leaf nodes. The leaf nodes extend out of the middle nodes in the manner shown in the figure. Many such nodes constitute There are two types of members in the functional group: the functional Agent and the functional alliance. The alliance is a black box to the outside world, and the endorsement Agent is the only channel for its internal communication; the terminal Agent is the center of the interface group, through which other interface agents can be opened , but there is no direct connection between the two after opening; the translation Agent corresponds to other external systems one by one.

The agent interaction model of distribution network self-healing control system is shown in Figure 4, ① is the interaction between entity Agent and interface Agent, the content mainly includes equipment status and control instructions; ② is the interaction between entity Agent and functional Agent, Mainly service requests and processing results; ③ is the interaction between the entity Agent and the translation agent, including request information and return results; ④ is the interaction between the entity Agents, the content includes data, status and scheme, etc.; ⑤ is the interaction between the functional Agent The content of the interaction between them is the same as ②; the dotted line is the management interaction of the interface Agent, and the content includes Agent affairs and processing measures. The interaction tasks in the system include: the entity Agent sends the feedback information and request for assistance to the interface Agent, obtains the desired information through the translation Agent, obtains the required service through the function Agent, requests assistance from other entity Agents, accepts or Reject assistance requests from other entity Agents and perform corresponding actions; functional Agents respond to entity Agent service requests and return results, respond to other functional Agent service requests and return results, and seek service support from other related functional Agents; interface Agents obtain other Agents The transmitted feedback information is transmitted to other Agents to transmit the dispatcher's control instructions, operating parameters, etc., to obtain the abnormal operation information of other Agents and handle them accordingly; the translation Agent responds to the translation request and returns the result.

The MAS structure model of distribution network self-healing control is shown in Fig. 5, which shows the process of how the Agent finds the Agent that can provide the functions it needs. A1, A2, and A3 publish services in advance. When A4, A5, and A6 need a certain service, they first search for the service to obtain a list of facilitators, and then directly interact with the service provider to obtain the required service.

The distribution network self-healing control MAS deployment model is shown in Figure 6. The element entity Agent is deployed on the device, the interface Agent is deployed on the control terminal, and the function Agent, translation Agent and collective entity Agent are deployed on the server. Each Agent has its own relational database and XML file. The relational database stores real-time data such as voltage, current, active power, and reactive power. The XML file stores static data such as voltage limit, current limit, and delay time.

In an embodiment, each Agent is refined into a specific role according to the type of equipment in the power distribution network. The element type entity Agent includes Switch Agent, Capacitor Agent, Reactance Agent, Transformer Agent, Voltage Regulator Agent and DGA Agent. The element type entity Agent is the software model of the switch. Including the main transformer and the distribution transformer; the voltage regulator Agent is the software model of the voltage regulator; DGAgent is the software model of various distributed power sources. The element-type entity Agent is used as the software model of the controlled equipment; it obtains the measurement data of the equipment according to the power distribution demand, uses the analysis and calculation method of the distribution network to confirm the environmental state of the equipment, and controls the action of the equipment to adapt to the environment. Variety. The collective entity Agent is mainly the feeder Agent; the feeder Agent is a software model composed of all the measurable and controllable objects on a feeder, including various switches, capacitors, reactors, distribution transformers, voltage regulators, distributed power supplies, etc. Functional Agents include risk identification and control Agent, optimization identification and control Agent, fault analysis and processing Agent, power flow calculation Agent, network topology Agent, and endorsement Agent. Risk identification and control Agent conducts risk identification and risk control, optimization identification and control Agent conducts optimization identification and formulates optimization control plan, fault analysis and processing Agent analyzes faults, locates faults, formulates fault recovery plan, power flow calculation Agent is responsible for power flow calculation, network Topology Agent conducts topology analysis, acts on behalf of Agent to handle alliance external affairs and assign alliance internal tasks. The interface Agent includes equipment agent, terminal agent and management agent; the equipment agent is the interface of each standard element type entity agent, the terminal agent is the interface of each standard collection type entity agent, and the management agent is the system control terminal interface with MAS management function. The translation agent is in one-to-one correspondence with other external systems.

In the embodiment, the system interaction model still has type characteristics, so it is modeled in units of types, and the interaction protocol standard used is FIPA (The Foundation for intelligent Physical Agents). The interaction performed by the entity Agent includes: first, publish the service, and interact with the DF (DirectoryFacilitator) as the initiator at the time of initialization. Second, send a notification, and act as the initiator to interact with other associated entities Agent when its own state changes. Third, issue instructions/negotiations/requests, when formulating a plan that requires assistance from other Agents, encountering problems that cannot be solved by itself, needing functional support, or needing to know the latest status of related Agents, as the initiator and upper/middle/lower-level entities Agent, functional Agent or interface Agent interaction. Fourth, search for services, and interact with DF as an initiator when a certain service is needed. Fifth, respond to instructions/negotiations/requests, and respond appropriately as responders. Sixth, send out the status, when it is necessary to feed back the status to the dispatcher, including fixed time intervals and special events, etc., interact with the interface Agent as the initiator. The interaction of functional Agents includes: first, assigning tasks, and interacting with internal functional Agents of the alliance as an initiator when obtaining tasks. Second, return the result and react appropriately as a responder. The interaction of the agent on the interface includes: first, when sending instructions, the dispatcher acts as the initiator to interact with the specified Agent when modifying parameters and formulating a plan. The second, the accepting state, displays data information as a responder. The translation agent, as the responder, returns the translation result to the requester.

In an embodiment, the agent behavior still has type characteristics, so it is modeled in units of types. The behaviors of entity Agent include: first, initialization behavior, a behavior of initializing itself or loading new objects, mainly loading environment (power grid results, parameters) information, rule base, knowledge base and publishing services. Second, notification behavior, a behavior when the state changes, mainly updates the corresponding parameters and sends the new state to the relevant Agent. Third, state identification and control behaviors, cyclic behaviors at a certain time interval, are responsible for reading real-time operating data of the power grid, quickly determining the state of the power grid, and quickly processing dangerous states. Fourth, the hidden danger elimination behavior, the cyclical behavior at a certain time interval, uses its own hidden danger handling logic, organizes relevant Agents to work together, and jointly discovers, analyzes and eliminates hidden dangers. Fifth, obtain status behavior, which is a behavior when you need to know the relevant status, and request the target Agent to notify the current latest status. Sixth, the cooperative operation behavior is an action when executing the plan, contacting and coordinating all the Agents in the plan to cooperate to complete the operation. Seventh, respond to the request behavior, the cycle behavior of message activation, get the message and give a reply according to its own state. Eighth, send status behavior, a behavior when a special event occurs, and send the status to the interface Agent, so that the dispatcher can understand the status of the device. The behaviors of the functional Agent include: first, the behavior of responding to requests, the cycle behavior of message activation, obtaining task requests, and assigning them to internal members of the alliance. Second, the reply behavior is a behavior after being requested and getting the result, sending the calculation result to the requesting Agent. The behavior of the interface Agent is: first, the behavior of receiving status, the circular behavior of message activation, obtaining the device status, and displaying it on the interface. Second, the action of issuing instructions, an action triggered by the dispatcher, sends control instructions, operating parameters, etc. to the target Agent. The translation behavior of the translation agent, the circular behavior activated by the message, accepts the translation task, and sends the translation result back to the requester.

The functions of Agent in the present invention can be realized by corresponding programs.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: the present invention can still be Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention shall be covered by the scope of the claims of the present invention.

Claims (17)

1. a power distribution network self-healing control system, is characterized in that, described system comprises entity agent, Functions Agent, interface agent and translation Agent; Described entity agent comprises element type entity agent and assembly type entity agent;

Described element type entity agent is embedded in controlled device; Described interface agent embeds in control terminal; Described assembly type entity agent, Functions Agent and translation Agent embed in server;

Described assembly type entity agent connects other system by translation Agent, connects yardman by interface agent, by the Processing tasks of Functions Agent, ensures the normal operation of controlled device;

Described Functions Agent completes control task for assisting described entity agent; A corresponding a kind of function realizing method of described Functions Agent; The Agent of the different implementation methods of same described function forms an alliance; Represent Agent process external affairs by one in alliance, internal affairs then distribute to remaining Functions Agent process by alliance's internal mechanism.

2. power distribution network self-healing control system as claimed in claim 1, is characterized in that, the corresponding described controlled device of described element type entity agent, described assembly type entity agent correspondence analysis unit.

3. power distribution network self-healing control system as claimed in claim 2, is characterized in that, described element type entity agent obtains the service data of equipment under test, when data occur abnormal, then carries out emergency processing.

4. power distribution network self-healing control system as claimed in claim 2, is characterized in that, described assembly type entity agent for coordinating relation between subordinate's intelligent body, monitored state changes and carries out cooperation control.

5. power distribution network self-healing control system as claimed in claim 1, is characterized in that,

Described interface agent comprises: the management type interface agent of the terminal type interface agent that the equipment type interface agent that standard device is corresponding, standard set map and process MAS event;

Described equipment type interface agent is for arranging and showing parameter and the state of equipment under test;

Described terminal type interface agent is for providing the control terminal of yardman;

Described management type interface agent, for monitoring the event of described system, processes all Agent abnormal problems.

6. power distribution network self-healing control system as claimed in claim 1, is characterized in that, described translation Agent is used for the conversion between external system information and intelligent body information.

7. power distribution network self-healing control system as claimed in claim 1, it is characterized in that, each Agent is provided with relational database and XML file;

Described relational database is used for storage voltage value, current value, has work value and without work value; Described magnitude of voltage, current value, there is work value and comprise instantaneous value and history value without work value;

Described XML file is used for storage voltage limit value, current limit, line loss limit value, power factor limit value and time delay.

8. the power distribution network self-healing control system as described in as arbitrary in Claims 2 or 3, it is characterized in that, described element type entity agent comprises switch Agent, electric capacity Agent, reactance Agent, transformer Agent, pressure regulator Agent and DGAgent;

Described switch Agent is the mapping model of switch;

Described electric capacity Agent is the mapping model of capacitor;

Described reactance Agent is the mapping model of reactor;

Described transformer Agent is the mapping model of transformer;

Described pressure regulator Agent is the mapping model of pressure regulator;

Described DGAgent is the mapping model of distributed power source.

9. power distribution network self-healing control system as claimed in claim 4, it is characterized in that, described assembly type entity agent comprises feeder line Agent; A feeder line can form the mapping model gathered by observing and controlling object to described feeder line Agent.

10. power distribution network self-healing control system as claimed in claim 1, it is characterized in that, described Functions Agent comprises Risk Identification and control agents, Optimal Identification and control agents, fault analysis and process Agent, Load flow calculation Agent, network topology Agent and represents Agent;

Described Risk Identification and control agents are used for Risk Identification and risk control;

Described Optimal Identification and control agents are used for Optimal Identification and formulate optimization control scheme;

Described fault analysis is used for analysis of failure, localizing faults and formulation fault recovery scheme with process Agent;

Described Load flow calculation Agent is used for Load flow calculation;

Described network topology Agent is used for topological analysis;

The described Agent of representing process alliance's external transactions also distributes alliance's internal task.

11. as arbitrary in claim 1 or 5 as described in power distribution network self-healing control system, it is characterized in that, described interface agent comprises equipment type interface agent, terminal type interface agent and management type interface agent;

Described equipment type interface agent is for showing the parameter of leading element type entity agent;

Described terminal type interface agent is for showing the parameter of regular set mould assembly entity agent;

Described management type interface agent is used for the life cycle of automatically repairing the Agent extremely exited, the running status of checking all Agent, all Agent of Non-follow control.

12. 1 kinds of implementation methods based on the arbitrary described system of claim 1-11, it is characterized in that, described method comprises the steps:

(1) controlled device described in entity agent Real-Time Monitoring running state data and judge running status; Undertaken alternately by translation Agent and other system;

(2) request needing feedack and seek to assist is sent to interface agent by entity agent;

(3) communicate with one another between each Functions Agent, respond the services request of entity agent and return results;

(4) interface agent obtains described feedack and seeks the request of assistance, and transmits the steering order of yardman and the operational factor of yardman's setting;

(5) entity agent is received yardman's instruction and the operational factor that need adjust and is carried out parameter adjustment to described controlled device.

13. implementation methods as claimed in claim 12, it is characterized in that, described controlled device comprises switch, capacitor, reactor, distribution transforming, pressure regulator and distributed power source.

14. implementation methods as claimed in claim 12, is characterized in that, the data of step (1) described entity agent Real-Time Monitoring comprise meritorious, idle, voltage and current.

15. implementation methods as claimed in claim 12, is characterized in that, entity agent is sought to assist to obtain the feedback information needed to Functions Agent;

Entity agent needs feedack to be the information of entity agent failure judgement type and handling failure;

Described fault type comprises real time fail and potential risk.

16. implementation methods as claimed in claim 15, it is characterized in that, the step of entity agent process real time fail comprises:

I () gathers described controlled device and to gain merit instantaneous value data, idle instantaneous value data, voltage instantaneous value data and electric current instantaneous value data;

(ii) judge described controlled device state, if abnormality then carries out step (iii), otherwise terminate;

(iii) emergency treatment is performed;

(iv) step (iii) is if the scheme of obtaining, then the relevant Agent of request performs respective operations, and after having operated, this processing procedure terminates; If the scheme of not obtaining or scheme are refused by other Agent, then carry out step (v);

V the process of () tissue depth, achieve a solution scheme.

17. implementation methods as claimed in claim 15, it is characterized in that, the step of entity agent process potential risk comprises:

1. meritorious, idle, the voltage and current data of described controlled device are obtained; Described meritorious, idle, voltage and current data comprise historical data, real time data and predicted data;

2. determine whether potential risk according to data, if having, carry out step 3., otherwise terminate;

3. tissue depth process, achieve a solution scheme.