A Two-Stage Continuous Discrete Var Heuristic Optimization Method with Embedded Benders Decomposition

Abstract

The invention discloses a two-stage continuous discrete reactive power heuristic optimization method embedded with Benders decomposition, which decomposes the problem into continuous sub-problems and discrete main problems through Benders decomposition, and converts the secondary objective function with the smallest voltage deviation into a linear objective function, so that the discrete main problem is transformed into an easy-to-solve linear integer programming problem, and the heuristic algorithm is used to solve the main problem to provide the value of the discrete variable to the sub-problem, and then by solving the continuous sub-problem, return the feasible cut or optimal cut to the discrete The master problem, which ultimately determines the sequence of actions of discrete devices through iteration. The decision result of the discrete equipment is returned to the original secondary voltage control, and the pure continuous problem is solved to obtain the output of the continuous equipment. The invention effectively reduces the difficulty of solving the mixed secondary voltage control problem, not only avoids the problem of infeasible solution caused by continuous equipment adjustment steps after the discrete equipment optimization action, but also ensures that the normal node voltage reactive power is qualified.

Description

technical field [0001] The invention relates to a two-stage continuous discrete reactive power heuristic optimization method embedded with Benders decomposition, and belongs to the technical field of automatic operation and control of power systems. Background technique [0002] The secondary voltage control realizes the coordinated calculation of various reactive power sources and control targets in the region, which is the core function of automatic voltage control. At present, there are a large number of discrete and continuous devices in the power grid at all levels. These devices make the voltage and reactive power of the whole network qualified. Mathematically, this is a complex mixed integer programming problem. The traditional method adopts "discrete devices act first, continuous devices finely The idea of ​​"adjustment" separates discrete and continuous problems, but this method has the following two problems: [0003] 1. The priority action of discrete equipment m...

AI Technical Summary

Problems solved by technology

[0003] 1. The priority action of discrete equipment mainly refers to the priority action of discrete equipment in this station. However, in the secondary voltage control, for substations that exceed the limit on the high voltage side, the priority action of discrete equipment may be caused. Substations that exceed the limit exceed the limit, which in turn requires more output from continuous equipment to correct the limit;

[0004] 2. Discrete equipment operates first. Although it can ensure that the voltage or reactive power of the station is qualified, it does not consider whether the continuous equipment can reach the feasible range of ensuring the voltage and reactive power of the entire network within the adjustment step after the action;

[0005] On the one hand, the principle of the priority action of discrete devices is to consider the local balance of reactive power to reduce network loss, but if the reactive power is over-compensated, local switching may lead to increased network loss. This heuristic strategy is between economy and safety. The balance between them is not good; the priority action of discrete devices essentially reflects the corrective effect of discrete variables on constraints, and does not consider the contribution to the objective function

Images