Power generation system reliability rapid assessment method based on state clustering

Abstract

The invention discloses a power generation system reliability rapid assessment method based on state clustering and belongs to the technical field of power system reliability assessment. Reliability assessment of a power generation system is carried out mainly through non-sequential Monte Carlo sampling and state clustering methods. Through the non-sequential Monte Carlo sampling method for sampling operation (failure) states of conventional power units of the system, and with the forced outage rate of the generator units being taken into consideration, the method not only replaces a conventional method for adding spare capacity to the system to ensure system reliability, but also establishes a system available generating capacity probability model with operation randomness and uncertainty of all units in the system being taken into full consideration; and by carrying out clustering on available generating capacity samples obtained through sampling to obtain a typical available generating capacity scene, workload of reliability calculation is reduced, reliability evaluation speed is allowed to be improved greatly, and the purpose of rapidly evaluating the reliability of the large-scale power system is achieved.

Description

technical field [0001] The invention relates to the technical field of electrical engineering, more specifically, it is a method for rapidly evaluating the reliability of a power generation system based on state clustering. Background technique [0002] Power system reliability assessment refers to the analysis and calculation of the adequacy or safety of the system under the consideration of various uncertain factors, including component outage model, system state selection, system state evaluation and system index calculation, etc. Content. [0003] With the increasing scale of the power system and the large-scale increase in the number of generating units in the power system, accurate evaluation of the reliability of the power generation system will inevitably increase the calculation workload and increase the calculation time. During the operation of the power system, the load level, unit start-up and shutdown, and power generation output all change in real time. When t...

AI Technical Summary

Problems solved by technology

Under the same accuracy requirements, the calculation time of the non-sequential Monte Carlo simulation method is much shorter than that of the sequential Monte Carlo method. In the case where the frequency and duration indicators do not need to be calculated, the non-sequential Monte Carlo method is preferred. However, due to the contradiction between the calculation accuracy and calculation time of the Monte Carlo method, it often takes a lot of calculation time to obtain a satisfactory calculation accuracy.

At present, variance reduction techniques are usually used, such as control variable method, equidistributed sampling method, etc., to speed up the convergence speed of Monte Carlo simulation and improve calculation efficiency, but these techniques only reduce the sampling times under the condition of ensuring a certain calculation accuracy. Rapid assessment of reliability cannot be achieved

Images