摘要
为了满足电网调度对火电厂负荷优化分配的实时性要求,提出了负荷优化分配总时间的概念并给出了具体计算公式。对煤耗、污染物排放、负荷调整时间的多目标数学模型进行了改进,将全厂煤耗量、污染物排放量两个目标统一为经济性成本目标,将负荷调整时间目标处理为机组变负荷范围的约束条件。通过引入遗传算法中的自然选择机制,提出了自然选择粒子群算法,并对算法的惯性权重及约束条件的处理进行了改进。实例计算结果表明:该方法与现有多目标优化方法相比可有效地缩短优化计算时间,在满足电网调度对负荷分配总时间的硬性要求的同时,更能满足电厂降低成本需求。
The total time of optimal load dispatch and the concrete calculation equation were proposed in order to meet the real-time demand of optimal load dispatch in power plant for power network dispatching.The multi-objective mathematical model of coal consumption,pollutant emission and load adjustment time objective was improved.The coal consumption of the entire plant and the pollutant emission were unified as an economic cost objective.The load adjustment time objective was treated as constraints of variable load range for units.By introducing the natural selection mechanism of genetic algorithm,natural selection particle swarm optimization（NSPSO） was given out,besides,the processing method of inertia weight and constraint condition in the algorithm was improved.The example calculation results show that the method can effectively shorten the optimal calculation time in comparison with the multi-objective optimization method.It can fulfill the mandatory demand of total time of optimal load dispatch for power network dispatching.Meanwhile,it can more satisfy the request of reducing economic costs for power plant.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2011年第32期122-128,共7页
Proceedings of the CSEE
基金
吉林省科技发展计划项目(20080523)~~
关键词
负荷优化分配
负荷分配总时间
实时调度
多目标
自然选择粒子群算法
optimal load dispatch
total time of optimal load dispatch
real-time dispatching
multi-objective
natural selection particle swarm optimization
