Distributed Fault Diagnosis Method For Power System Based On Membrane Computing

Modern power system structure is becoming more and more complex, and the scale is expanding. When multiple complex faults occurs, a large amount of fault warning information swarm into dispatching center. There are many suspicious components, fault diagnosis decision table or rule knowledge representation is enormous, then if adopt the centralized fault diagnosis methods, diagnosis reasoning speed is difficult to meet the requirements. And short-term influx of a large number of alarm messages may cause the communication channel congestion, resulting in missing or error alarm information. Consequently, this paper adopts the distributed fault diagnosis theory. The power grid is divided into several sub networks, the fault diagnosis in the sub network, the diagnosis result to dispatch center as the form of fault briefing. Distributed diagnosis fault diagnosis can reduce computational complexity, so as to improve the diagnosis speed. At the same time reduce the pressure of data communication, to avoid the overcrowding on data in dispatching center. Fault diagnosis in the subnet, this paper adopts the method of membrane computing for power system fault sections identification. The main research contents are as follows:1. Optimization spiking neural P system is used to deal with power system fault alarm information. In this paper, the alarm information preprocessing optimization model is solved using optimization spiking neural P system, identify the missing and incorrect alarm information. And then eliminate the incorrect information and fill missing information, to improve the accuracy and reliability of fault alarm data. Finally, the detailed algorithm flow is given, through an example to verify the effectiveness of the method.2. Fault diagnosis of transmission network sections by using fuzzy reasoning spiking neural P system. This paper adopts the triangular fuzzy number in the diagnosis model, so the model can better deal with the problem of uncertainty in fault diagnosis. Then according to the fault production rules, respectively on line, bus and transformer construct fuzzy reasoning spiking neural P system diagnosis model. Through statistical data analysis and a large number of simulation experiments, the system parameter value and weight value is determined. Matrix description of the fault diagnosis model is given, and convenient for computer processing. Finally, through several of experimental results show that the fuzzy reasoning spiking neural P system fault diagnosis model is effective.3. Based on the idea of distributed fault diagnosis, a depth first search algorithm is used to divide the network into several sub networks. Fault diagnosis in the sub network, and the result of fault diagnosis is brief report. The simulation test was carried out on the IEEE-118 node system. The experimental results show that the distributed fault diagnosis method based on the membrane computing is feasible.