Research Of Fault Diagnosis Method For Digital Substation

With the improvement of social modernization, power plays an increasingly important role in the national economy and people’s lives, depending on electricity supply increased. We need quality, reliable and stable electricity, while power failure would cause significant economic losses to the user. The scale, capacity and coverage of modern power grid are increasing and operational complexity is growing. Power grid runs closer to the limit, the losses is huger than past when power failed. There are many problems to resolve for preventing large area power failer. It involved in many aspects of power system, for example, the design of power system, operationing, relaying protection, automatic control and so on. Finding the fault location as soon as possible, correctly handle the accident and early resumption of normal power supply to reduce power losses is an effective means.First of all, this thesis studies the structure, technical features and unconventional transformer working principle and structure of digital substation in smart grid. On this basis, proposed a fault diagnosis scheme of unconventional transformer based on wavelet analysis. For the problem of signal interference caused by electrical equipment in the digital substation, design a new wavelet thresholding function. The function has a good performance proved by simulation. According to the analysis of secondary signals produced by unconventional transformer, combine with the characteristics of mutations of voltage and current under fault, presented a diagnostic criteria of unconventional transformer, it proved correctly by simulation.Sencondly, the fault and fault diagnosis method of the main equipment for the digital substation are analyzied, as the fault of digital substation is complicate and distributed, the multi-agent technology is applied to it’s the fault diagnosis method. The decision method of fault diagnosis based on multi-agent is designed and the transformer fault case is analyzed to varify the the superiority of this method.At the end of this thesis, considering the characteristics of substation, apply multi-agent technology to removing the fail distribution line and the restoration of power supply. There is a faster and more reliable result than the traditional fault diagnosis method. Come up with a method for adjusting the load current. Examples proved the adjusting is necessary and the method is feasible.