基于家族优生学进化算法的电力系统动态无功优化

针对大规模电网无功优化收敛速度慢的问题，提出了基于家族优生学的无功优化算法，该方法将正交设计技术引入家庭子代培植过程中，以加强个体行为改善，避免早熟，加快了进化后期的收敛速度。采用改进的动态无功优化数学模型和相适应的编码方法，并根据控制变量的可调档位数动态分配码串长度，由控制变量确定基因组，以基因组中控制变量的可调档位数为约束条件，修正不可行解，提高了计算效率。算例表明将家族优生学进化算法应用于无功优化，取得了良好的效果。

Dynamic Reactive Power Optimization on the Basis of Family Eugenics Based Evolution Algorithms

To improve the convergence speed of reactive power optimization in large-scale power grid, a reactive power optimization algorithm on the basis of family eugenics based evolution is proposed. In this method the orthogonal design technique is led into the course of offspring’s breeding inside a family, thus the individual behavior amelioration can be improved, the prematurity can be avoided and the convergence speed at the late period of evolution can be quickened. In addition, an improved mathematical model for dynamic reactive power optimization and corresponding encoding method are adopted; according to the binary codes corresponding to the numbers of adjustable taps which belong to both transformers and reactive power compensating devices, the length of code string is dynamically assigned; the genome is decided by control variables; taking the binary codes of adjustable taps in control variables of genome as constraint condition, the infeasible solution is modified, thereby the computational efficiency is improved. Results of calculation example show that good effect is achieved while the family eugenics based evolution is applied to reactive power optimization.