Study On Operational Reliability Evaluation Models For Components And Power Systems

With the rapid growth of power demands in China and the implementation of the"Transmitting Electricity from the West to the East, and Interconnecting a Nationwide Electric Power Grid"strategy, the scale of power grid is expanding increasingly and the degree of power network interconnected is raising continuously. In this case, the operational reliability evaluation of power systems has become increasingly an important task in utilities. Therefore, evaluating and forecasting the operational reliability is becoming a hot topic of research on power system reliability fields.Load forecasting has become an important task for the designers, planners and operators of power systems. The accuracy of load forecasting has a significant impact on the system reliability. There are many factors affecting the accuracy of load forecasting models, then none of them can be used to forecast the load with a precision of 100%. Therefore, it must be choice appropriate models based on the characteristics of the load. Based on stationary and non-stationary weights, combination forecasting models for short-term load were proposed in this thesis. In order to improve the accuracy of combination forecasting models, a grey relationship analyzing technique and redundancy verifying technique are used to judge the forecasting performance of each model. The case studies on a practical power system show that the proposed model has a higher precision in forecasting load, and it presents a novel idea for load forecasting.Combining with the characteristics of operational reliability, neural network model of components'failure rate which suiTab for operational reliability evaluation is established in this thesis based on traditional reliability evaluation theory and method. Artificial neural network is a large-scale nonlinear adaptive model, it can establish a operational failure rate model based on environmental factor. In this paper ,a three tier structure of improved BP neural network is established ,and forecast the failure rate on various relative factor.To improve the accuracy of operational reliability evaluation, this paper proposes an algorithm of operational reliability evaluation which considers the components'operational failure rate and load curve. In this algorithm making fault tree analyze is involving components'failure rate in that situation. Monte Carlo system state simulation method is utilized to sample system state. Making a contingency analyze is involving DC power flow calculation, Depth-first Search for recognize of system islanding, and load curtailment calculation. Indices and algorithm framework of operational reliability evaluation are also given in this paper.