Research On Improved Algorithm For Power System Reliability Considering Load Curve

With the growth of GDP and the scale of power system, power system reliability evaluation has been a very important assistant for power system's planning, design and operation, and valuable information for power system's planning and operation can be provided. However, along with the development of power system towards extra-high-voltage, long distance, great unit and huge capacity, the power system reliability evaluation becomes more and more complex. It is still the critical problems that how to reduce the computational complexity, how to obtain a balance point between computational precision and efficiency and how to deal with the uncertainty of power system reliability.Electric load forecasting is the forward work for power system planning, the accuracy of the forecasting results directly influence the correctness and rationality of the power system planning. But load forecasting are affected by many factors, there is no forecasting method with precise enough currently, so we can just choose an appropriate method based on the need of forecasting.The main contents in detail include the following:①The indices and basic theory of bulk power system adequacy evaluation are introduced.An arithmetic and the flow chart of program are analysised after an in-depth research on interralated calculations, including power flow calculation, load curtailment model etc.②classical methods, traditional methods and modern methods are compared respectively, it provides the theoretical basis for selecting reasonable load forecasting model and methods of reliability evaluation. Based on analysis of the characteristic indices of daily and yearly load curve forecasting, A load curve forecasting method is provided and appied to a actual engineering project.③Based on the comparability between Weibull distribution and load density curve, the load density curve is fitted, and then the probability density function of loads can be obtained, in other words the probabilities of each load in load intervals can be got. Compared with traditional multi-step model, this method is able to ensure the computed precision meanwhile reduce complexity of reliability computation.④The concept of critical load and its calculation method are discussed. The system probabilistic indices and frequency indices of a fault state can be calculated by comparing each load interval with the critical load and the energy indices can be calculated by evaluating the load interval upon the critical load. So the computed speed can be improved, by decreasing some power flow calculation.⑤By combining critical load with Weibull distribution, an improved algorithm for power system reliability considering load curve is proposed. The IEEE-RTS system and an actual engineering project are evaluated by the algorithm and the results are compared with other literatures, which prove that the algorithm can obtain a real reliability level of power system and process a fast computed speed.