基于内点法和遗传算法相结合的交直流系统无功优化

Reactive Power Optimization of Hybrid AC/HVDC Power System Combining Interior Point Algorithm and Genetic Algorithm

为简化传统交直流混合系统潮流计算方法,提出了一种改进算法。该算法采用Gauss-Seidel迭代法求解直流系统,通过调整换流变压器的变比以使换流器的触发角运行在合适的值域,再与交流系统交替迭代求解。针对遗传算法适合处理离散变量但容易陷入局部最优而内点法具有方向性但不适合处理离散变量的不足,提出一种内点法和遗传算法相结合的混合算法对交直流系统无功优化模型求解,交流系统中少量的离散控制变量及直流系统控制变量采用遗传算法求解而交流系统中大量的连续控制变量则采用内点法求解。混合算法结合了两种算法的优点,可以方便地处理离散变量且具有明显的方向性。通过实例仿真验证了该算法具有收敛性好、运算速度快等优点。

In order to simplify the traditional method for calculating AC and DC power flow, we proposed an improved calculation method which uses the Gauss-Seidel iterative method to figure out power flow of the DC system by adjusting the transformer turns ratio to make the firing angle of the inverter run at the right range to solve the DC system, and then alternately iterating with AC power system. As genetic algorithm（GA） is suitable for handling discrete variables but easy to fall into local optimum and interior point algorithm is directional but not suitable for handing discrete variables, we proposed a hybrid algorithm to deal with the reactive power optimization model in the AC-DC system, which uses GA to solve a small number of discrete control variables in AC system and DC system control variables and uses interior point method to solve a large number of continuous control variables in AC system. The hybrid algorithm combines the advantages of two algorithms that can easily handle discrete variables and has a clear direction. Example simulation demonstrates that the algorithm has good convergence and fast operation speed.