A reactive power optimization method based on day-ahead planning mode that adapts to grid load changes

Abstract

The invention discloses a reactive power optimization method in a day-ahead planning manner adaptive to changes in grid load, and belongs to the technical field of power systems and automation thereof. The invention solves the independent static reactive power optimization model of multiple planning modes through the optimal power flow, and obtains the optimal compensation amount of each capacitive reactor node in each mode; The planning method divides the time period, and solves the capacitive reactor node gear with the smallest total network loss for the different operation modes in the same segment based on linear programming, continuously reduces the number of segments and iterates until all the capacitive reactors meet the action requirements. The number of times is required; the segmental optimal gear position of the capacitive reactor node is fixed, and the continuous solutions such as generator reactive power of multiple day-ahead planning methods are further corrected through the optimal power flow, and an optimization method that finally satisfies various constraints is obtained. The present invention can provide a day-ahead optimized reactive power plan that is self-adaptive to the load variation of the power grid and meets the requirements for the number of operations of the capacitive reactor.

Description

technical field [0001] The invention belongs to the technical field of electric power system automation, and in particular the invention relates to a reactive power optimization method in a day-ahead planning mode that adapts to changes in grid load. Background technique [0002] Day-ahead reactive power optimization refers to formulating an optimized reactive power and voltage dispatching plan that satisfies the grid safety operation constraints and reactive power adjustment equipment action constraints of the next day based on the grid’s day-ahead load forecasting, active power generation, and maintenance plans, so as to achieve the total active power of the grid the next day. The goal of minimizing network loss is an important means to ensure the safe and economical operation of the power system. The traditional static reactive power optimization can only calculate the reactive power optimization results at a certain moment, and the actual load of the power system is ofte...

AI Technical Summary

Problems solved by technology

The traditional static reactive power optimization can only calculate the reactive power optimization results at a certain moment, and the actual load of the power system is often in constant change. Tracking the static reactive power optimization results will lead to frequent switching of discrete control equipment such as reactive power compensation devices. In order to improve the service life of the equipment and avoid the risk of misoperation in the actual power grid, the total number of adjustments within one day is strictly limited

Therefore, after comprehensive consideration of the above factors, the reactive power optimization problem of the multi-planning method is actually a nonlinear mixed integer dynamic optimization problem. Due to the large number of states, it is very difficult to solve directly

[0003] At present, there are two common indirect solution ideas: one is to obtain the static reactive power optimization results of multiple day-ahead planning methods first, and then solve the problem with the goal of minimizing the rounding deviation or the minimum system network loss increment. Considering the integer programming model constrained by the maximum number of switching times of reactive power compensation devices, but it ignores the interaction between capacitive reactors in the solution process, and cannot guarantee the continuity of the capacity curve of capacitive reactors, and the operating frequency may be high. In addition, When there are many planning methods (such as 96 points in general), it is also very difficult to solve the integer programming model containing the absolute value inequality constraints of the number of actions; The day-ahead planning method is divided into time periods, and the switching gears of typical load points in each time period are calculated through static reactive power optimization. are consistent), while ignoring the load characteristics of different sections in the large power grid, and the selection of typical load points in the same section lacks theoretical basis

Images