基于风险和成本的化工园区供电网络优化设计  

Optimized design for the power supply network of the chemical industry park based on the risk and cost evaluation

作　　者：刘储朝 钱新明[1] 段在鹏[1] 候云娟 

机构地区：[1]北京理工大学爆炸科学与技术国家重点实验室,北京100083

出　　处：《安全与环境学报》2016年第3期172-177,共6页Journal of Safety and Environment

基　　金："十二五"国家科技支撑计划项目(2012BAK13B01)

摘　　要：化工园区供电网络优化设计是国家综合管理部门、电网工程项目法人及融资方进行项目决策的重要依据。化工园区内企业在生产活动中极易释放易燃易爆气体,供电危险极大。主要目的是运用程序优化方法,找到一种既经济又安全的供电线路布局方案,最终得到基于风险和成本的安全决策。基于前人在计算机智能算法优化数学模型得到的经验,运用图论Floyd算法和GA遗传算法研究了图论算法和人工智能算法在基于风险和成本的化工园区供电网络优化中的设计和应用。设计了一个化工园区,总降变电站为源点,8个可选择的化工园区内具有火灾或爆炸高危险性生产企业位置为网络节点。建立了无浓度约束范围和有浓度约束范围两种模型,并在有浓度约束范围模型的优化中运用广度优先遍历和深度优先遍历相结合的思想。给出了3套方案并且得到了理想的优化结果。首次应用计算机智能算法优化的方法研究了化工园区供电网络安全性,研究表明,图论算法和人工智能算法在化工园区供电网络优化上的应用是可行的。对于网络节点较多的实际化工园区供电布局,GA遗传算法一般能得到比Floyd算法更快更好的全局最优解。In this paper we would like to present two optimized methods on the optimal design for the power supply network to be used for the chemical industry parks based on the risk and cost evaluation.For this purpose,we have developed two simulated chemical park models with a domain transformer substation as a power source,and eight high-risk fire or explosion producers as the network nodes.In proceeding with our research,we have prepared two simulated models,one of which is working without the gas concentration restriction whereas the other is working under the strict gas concentration restriction and limitation.With the above two models in hand,we have combined the breadth-first traversal algorithm and the depth-first traversal algorithm implementations with the safety decision made based on the Floyd algorithm and genetic algorithm.Thus,based on the formerly developed mathematical optimization models for artificial intelligent algorithm,the graph theory Floyd algorithm and the genetic algorithm,we have adopted the artificial intelligence algorithm,which was based on the risks and costs of the supply network optimization in the chemical industry parks,with all the algorithms being coded by MATLAB software.And,finally,we have achieved the ideal result from the three optimized methods.With the computer intelligent algorithm first adopted for searching measures for the power supply network safety,we have laid-out the length of the cable transversing an area with the density three times as much as that of an area without concentration.What is more,instead of the price of oxygen barrier retardant cable,we have adopted ordinary cable three times as much as usual,which helps to guarantee safety,economy,and reliability for the power supply system.The result of our trial application demonstrate that the graph theory algorithms and the artificial intelligence algorithms are suitable for the optimized simulated supply network in such parks while the genetic algorithm computation has been done faster with more accurate results th

关 键 词：安全工程 线路优化 FLOYD算法 遗传算法 

分 类 号：X394[环境科学与工程—环境工程]