Generating system reliability evaluation method based on sparse convolutional recursion

Abstract

The invention relates to a generating system reliability evaluation method based on sparse convolutional recursion. First of all, original data (load data and set technical data) is input, an original continuous curve is formed, N sets are sequenced in a default mode, generating capacity of a set i is calculated, an equivalent continuous load curve is corrected, and finally, reliability indexes LOLP and EENS are calculated, wherein a sparse convolutional recurrence method is employed according to discreteness of the equivalent continuous load curve. The method has the following advantages: the method is high in calculation precision, is not sensitive to the value of a reference unit, has obvious advantages over an equivalent electric quantity function method on an occasion with a quite high precision requirement for the LOLP index and can well satisfy application requirements of a power system for random production simulation.

Description

technical field [0001] The invention belongs to the field of power supply planning and power generation system reliability evaluation, in particular to a power generation system reliability evaluation method based on sparse convolution recursion. Background technique [0002] Power system stochastic production simulation is an algorithm that calculates system reliability indicators and system production costs by simulating the production conditions of each unit while considering the random fault characteristics and power constraints of the unit. It is used in power supply planning and power generation system reliability evaluation. Wide range of applications. [0003] The most basic algorithm for random production simulation is the convolution recursion method. The concept of this algorithm is clear, and the core is the convolution calculation formula for the probability distribution function of random variables in probability. In the calculation process, the equivalent loa...

AI Technical Summary

Problems solved by technology

Or use the semi-invariant based on random distribution of digital characteristics to describe the continuous load curve and the random outage function of the generator set. The semi-invariant method is only an approximate method, and the error is difficult to estimate and control, especially the calculation of reliability indicators often exists large error

Or the proposed equivalent power function method directly performs convolution and deconvolution operations based on power, which significantly reduces the amount of calculation. It is very suitable for stochastic production simulation of power systems containing multiple hydropower plants, and has been widely used. There may be large errors in the calculation of indicators

Images