Alarming Processing Of Electric Power System Based On Artificial Intelligence

A tabu search (TS) based approach is proposed for alarming processing in power systems. First, several existing evaluation criteria describing the alarm processing problem are briefly discussed, and a new criterion is proposed. Secondly, a novel method is developed to solve this problem using a TS based method. Finally, the example is used to demonstrate the feasibility and efficiency of the developed method. The paper also presents a comparison between the developed TS based and the more established genetic algorithm (GA) based approaches to the alarm processing problem. Many simulation results show that the TS-based approach is more efficient than the GA based approach. Key features of this proposed method are that it has solid mathematical foundation and can find multiple global optimal solutions directly and efficiently in a single run. This is very suitable for complex alarm processing problems especially for situations with missing or false alarms, because different combinations of events can produce the same set of alarms under these circumstances. The test results suggest that the developed TS based method is promising.Artificial intelligence (AI) is a method simulating human being's intelligence to resolve a real problem and Expert system is a primary branch of AI. An expert system may be a computer program, which employs intensive knowledge and expertise, to solve a complex problem which can be settled only by full knowledge and experience.The principle of artificial intelligence and expert system are discussed especially the application of expert system in power system. An expert system developed in VC++ prolog to identify faulted sections and interpret protective apparatus operation in large interconnected power systems is presented. The expert system is capable of identifying bus fault, line fault and fault sections in the common area between a specific bus and a line. Also, the expert system can identify the misinformation from relays or breakers. Finally the expert system is expanded to classify the fault type of the faulted section by taking real-time measurements of current and voltage signals.Computer simulation results showed that the expert system for fault section identification and fault type classification is correct and valid.