Research Of Bus Protection Technology Based On Artificial Neural Network

Bus is one of the most important parts in power plants and transformer substations. Bus protection plays a key role in the reliable and secure operation of power system. The research on the bus protection of high-reliability and high-intelligence will be quite important.It is necessary to introduce new theory and technology to make better improvement in bus protection. Artificial intelegence is wildely used in the power system, and it was also researched in the relaying protection. As a mathematic tool with high intelligence, ANN (Artificial Neural Network) is suitable for the research on new protection.For a long while, the application of ANN to relay protection is based on classification ability. In this thesis, ANN model is trained by sample data, which can characterize fault. The various faults of power system can be distinguished and judged by the trained ANN. BP and RBF are both used to train the ANN model. Applying this model, bus protection failures were analyzed and compared in two different improved algorithms of neural networks, and two simulation results were acquired. The result shows that the trained ANN model could not only correspond to the normal bus operation but also to all types of faults, and satisfy the precision required by the bus protection.But, the lack of sample data restricts the abroad application of ANN protection to power system. This presis presents an ANN model which is used to replace the physical object of bus protection, and proposes a new method of bus protection which is based on function approximation ability of ANN. First, it is crucial to establish the functional relation between the input and the output of bus protection object. The inputs are synchronous currents on second side of each CT, and the output is current sum on primary side of CT. ANN model with different activation functions, like linear and Sigmoid, are built, and the training algorithm of the two models are discussed. The physical simulation in MATLAB of bus protection based on ANN model is conducted under different locations and types of faults, and the simulation results coincide with the estimated results.