Research On Decentralized Autonomous Self-healing Mode For Distribution Network With Widely Accessed Distributed Generation

With increasingly serious energy shortage and environment pollution,the concept of sustainable development has become a global consensus.Distributed generation that uses renewable clean energy has accessed to power system more and more widely,especially in distribution network.The current operation mode of single-terminal power source system has been changed as DGs are widely assessed.Different types of DG will make significant impact in the aspect of detecting shortcut current,identifying fault location and self-healing.The centralized control mode,relying on dispatch center,has exposed problem and has been proved it is no longer applicable to the situation where DGs are widely assessed.Therefore,it is necessary to carry out the research for distribution network self-healing control mode which is adjust to widely assessed DGs.This paper firstly introduces the research status of fault location and fault restoration method for distribution network,then proposes decentralized autonomous self-healing mode in order to realize fault location,isolation and restoration as soon as possible.The main work of this paper includes the following aspects:1.The output characteristics of different types of DG has been studied.Among them,the output characteristics of asynchronous motor type generation and converter type generation are performed on software DIgSILENT.The influence on control and protection of distribution network,made by DGs,is analyzed.2.This paper proposes a two-layer decentralized autonomous self-healing structure based on intelligent multi-agent technology.When it comes to the decision problem involve multiple feeders which requires global information to optimize,it is still remain to the tradition centralized decision making mode of upper layer.When the problem expects higher instantaneity,speed and flexibility,including fault protect and control response on feeders,self-healing of distribution network,it will be processed by decentralized autonomous self-healing mode.3.This paper establishes the model of control and protection intelligent agent which makes up the intelligent multi-agent system on feeders and processes task analysis for the modules and elements needed.By researching the characteristic of different types of faults and different positions upstream/downstream the fault location,combining operation laws under open loop and closed loop,the protection scheme based on directional current protection is proposed.Taking the lack of signal and abnormal status of CPIA,this paper raise a fault-tolerant scheme,achieving cooperation between protection and backup fault-tolerance.The effectiveness of the proposed method is verified by simulation results of DIgSILENT examples.4.The recovery strategy for power loss buses after fault isolation is proposed.According to the recovery mode of “decomposition first-aggregation second”,the critical load is given priority to ensure their power supply supported by DG,and the neighboring buses with the aggregation capability are searched.Through the communication of distributed control and protection agent,the healing progress based on the real-time load conditions of the grid is realized.The effectiveness and flexibility of the proposed method is verified by simulation results of DIgSILENT examples.