Research On Power System Complex Fault Diagnosis Method

Power grid fault diagnosis is one of the basic subject of making intelligent decision for power grid dispatching, and it is also the inevitable trend of future grid dispatching operation and control. With the continuous enhancements of the interconnectedness of power grid, the dynamic behavior of power system are more complex. The complex faults with the malfuctions of protective relay and circuit breaker and developmental failure and cascading failure caused by failing to timely remove fault equipments are more and more. Especially under the background of the smart grid construction, the great access of intermittent energy and the interactive requirements with users ask dispatching center to have higher handling efficiency, and put forward more urgent practical requirements for intelligent decision-making for handling complex fault as well. This dissertation will get down to research from the effectiveness under the condition of error fault information and the uncertainty of time the spatial dimension, whose restrict the online operation of fault diagnosis system. The main work are as follow:To study the influence factors and the difficulty and analyze the basic concept of power grid complex fault diagnosis by comparing the complex fault diagnosis methods at home and abroad. Combining the basic theory of grid operating, the related factors of involved in power grid complex fault diagnosis were revealed with complex data environment, and their comprehensive effect mechanism is analyzed. Finally, it points out that the uncertainty of time the spatial dimension and the objective function measured the rationality of fault hypothesis are the problems for the complex fault diagnosis that has not been solved.Combining with characteristics of actual faults in power grid, a power grid fault diagnosis method based on genetic algorithm with variable length chromosome was presented, which can reliable to diagnose fault under severe environment, such as no obvious fault feature, data error or no credible information, and even analyze and recreate the complex fault process of power grid. According to logic constraint relationship between different events, the fault hypothesis was constructed; and remediation strategies were proposed considering complicated conditions such as data error, loss or time-sequence distortion. Combining the two, the complete fault hypothesis set was set up. To measure the credibility of fault hypothesis, the objective function with operation consistency of protective relays taken into account was proposed. Whose optimal solution was calculated by adopting genetic algorithm with variable length chromosome, to meet the requirements of dynamic change of events length for fault hypothesis. Finally, based on the basic data of dispatching automation platform, the online fault diagnosis system is designed, and the the effectiveness and practicability of the system is verified by the offline tests and online running.