基于改进DEMO算法的高速公路分布式光伏设施布局优化  

Layout Optimization of Distributed Photovoltaic Facilities on Expressway Based on Improved Differential Evolution for Multi-objective Optimization Algorithm

作　　者：王宇[1,2] 耿庆桥[2] 张路凯 孙东冶 王登科 WANG Yu;GENG Qing-qiao;ZHANG Lu-kai;SUN Dong-ye;WANG Deng-ke(School of Transportation Science and Engineering,Harbin Institute of Technology,Harbin 100059,Heilongjiang,China;Transport Planning and Research Institute,Ministry of Transport,Beijing 100028,China;China Transport Telecommunications&Information Center,Beijing 100011,China;School of Civil Engineering,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]哈尔滨工业大学交通科学与工程学院,黑龙江哈尔滨100059 [2]交通运输部规划研究院,北京100028 [3]中国交通通信信息中心,北京100011 [4]北京交通大学土木建筑工程学院,北京100044

出　　处：《中国公路学报》2024年第7期264-279,共16页China Journal of Highway and Transport

基　　金：国家自然科学基金项目(71340020);中央高校基本科研业务费专项资金项目(2021YJS080)。

摘　　要：合理的光伏能源利用及设施布局优化对于高速公路的低碳化建设和“净零”排放目标的趋近具有重要意义。从能源自洽角度,以经济成本最低为布局优化目标,构建考虑选型建设成本、位置距离成本、运营维护成本及额外能源增益的光伏设施布局优化模型(Photovoltaic Layout Optimization Model,PLOM),对光伏设施的布局面积、密度设置、布局数量及倾角方位等约束进行条件限制。为避免解空间分布不均匀及漏解现象,引入控制参数调整策略、精英选择双变异策略与动态拥挤距离排序策略,提出改进多目标差分进化算法(Improved Differential Evolution for Multi-objective Optimization,IDEMO)进行布局方案求解。依托G40沪陕(上海—陕西)高速(陕西段)及其周围区域的地理信息及路网数据,探讨服务区屋面区域及道路边坡区域光伏设施的布局方案及发电效果。研究结果表明:建筑物屋面面积、太阳辐射强度、边坡位置及设施倾角方位是高速公路光伏设施选址布局的关键要素,当倾角设置为18°~21°、边坡朝南时能量转换和发电效果最佳;IDEMO算法与标准DEMO算法、非支配解排序遗传算法(Non-dominated Sorting Genetic Algorithm,NSGA-II)、粒子群算法(Particle Swarm Optimization,PSO)、混沌猫群算法(Chaos Cat Swarm Optimization,CCSO)及禁忌搜索算法(Tabu Search,TS)的性能对比分析结果显示,IDEMO算法在各基准函数下具有更好的搜索能力和更高的收敛精度,更容易获得全局最优解,其算法寻优效率和寻优可信性较高,整体具有更好的寻优性能。所提出的研究方法可为高速公路的低碳化建设和零碳目标的趋近提供理论基础及思路参考。Reasonable photovoltaic-energy utilization and facility-layout optimization are crucial for the low-carbon construction of expressways and the approach of“net-zero”emission targets.From the perspective of energy self-consistency and based on the lowest economic cost as the objective of layout optimization,a photovoltaic layout optimization model was constructed by considering the selection and construction costs,location-distance cost,operation and maintenance costs,and additional energy gain.The layout area,density setting,layout number,and inclination azimuth of photovoltaic facilities were restricted.To avoid uneven spatial distribution and missing solutions,a control-parameter adjustment strategy,an elite-selection double-variation strategy,and a dynamic crowding-distance ranking strategy were introduced,and improved differential evolution for multi-objective optimization(IDEMO)was proposed to solve the layout scheme.Based on the geographic information and road network data of the G40 Shanghai-Shaanxi Expressway(Shaanxi section)and its surrounding areas,the layout scheme and power-generation effect of photovoltaic facilities on the roof of the service area and road slope area are discussed.The results show that roof area,solar-radiation intensity,slope location,and facility-inclination orientation are key factors that determine the location layout of expressway photovoltaic facilities.When the inclination was set to 18°-21°and the slope was oriented south,the energy conversion and power-generation effects were the best.Moreover,the performance of the IDEMO algorithm was compared with that of the standard differential evolution for multi-objective optimization algorithm,non-dominated sorting genetic algorithm,particle swarm optimization algorithm,chaos cat swarm optimization algorithm,and tabu search algorithm.Results of comparative analysis show that the IDEMO algorithm offers better searching ability and convergence accuracy under each benchmark function,as well as obtains the global optimal solution m

关 键 词：交通工程 光伏设施 IDEMO算法 布局优化 PLOM模型 

分 类 号：U491.8[交通运输工程—交通运输规划与管理]