摘要
为了研究煤矿节能减排规划期内每年度间的动态多目标关联性,构建了基于自适应控制、差分进化算法与混合聚类算法的智能优化算法,应用自适应控制实时调整变异尺度因子、交叉概率常数的更新策略,得到差分进化算法的pareto解,进而应用混合聚类算法获取投资优化方案,为管理者投资决策提供有效依据。实验结果表明:在较少的迭代步数内,煤炭生产总量、能源消耗量、污染物排放量可以根据目标要求获得协调发展,且提供的投资决策依据切实有效。
To study the dynamic multi-objective correlation each year during the planning period on the coal mine energy saving and emission reduction, an intelligent optimization algorithm is proposed based on adaptive control, differential evolution algorithm and hybrid clustering algorithm,which used adaptive control to real-time adjust the mutation scaling factor and crossover probability constant to obtain the Pareto solution of differential evolution algorithm.Then, the hybrid clustering algorithm is used to obtain the investment optimization scheme, which provides an effective basis for managers to invest. The experimental results show that the total amount of coal production, energy consumption and pollutant emissions can be coordinated and developed according to the requirements of the target in a few iterations and the investment decision-making is effective.
出处
《煤矿机械》
2017年第10期18-21,共4页
Coal Mine Machinery
关键词
差分进化
混合聚类
智能优化
自适应
多群体
决策
differential evolution
hybrid clustering
intelligent optimization
adaptive
multi group
decision making
