基于内点法和改进遗传算法的无功优化组合策略

提出了一种求解无功优化问题的组合策略，该策略将无功优化问题分解为连续优化和离散优化2个子问题，分别用预测–校正内点法和改进遗传算法进行求解。考虑到实际电网在进行无功优化控制时，发电机是主要的调节手段，先不考虑离散变量的约束，采用预测–校正内点法优化连续变量；然后保持连续变量不变，用改进遗传算法优化离散变量；再返回到连续优化阶段，如此交替求解。当出现相邻的连续优化阶段和离散优化阶段网损变化的差值小于设定值时，停止优化。IEEE14、30、57、118节点系统的仿真结果表明，该策略比其它组合算法在收敛性和计算效率上更具优越性。

A Combination Strategy for Reactive Power Optimization Based on Predictor-Corrector Interior Point Method and Improved Genetic Algorithm

Based on predictor-corrector interior point method (PCIPM) and improved genetic algorithm (IGA), a combination strategy for reactive power optimization is proposed, which divides reactive power optimization problem into two sub-problems, i.e., continuous optimization and discrete optimization, then PCIPM and IGA are used to solve them respectively.Because the regulation of generator voltage is the main reactive power optimization measure in actual power network, so at first the constraints of discrete variables are not taken into account and the continuous variables are optimized by PCIPM;then keeping continuous variables invariant, the discrete variables are optimized by IGA;and then the alternately solving processes are repeatedly performed until the difference of network loss variation between the adjacent continuous optimization stage and discrete optimization stage is less than the set value.The simulation results of IEEE 14-bus system, IEEE 30-bus system, IEEE 57-bus system and IEEE 118-bus system show that the proposed combination strategy is superior to other combination strategies in convergence and calculation efficiency.