| 摘要: |
| 基于节能减排背景和双碳目标，既有
建筑绿色改造是一项迫切需求。屋面增设光伏
阵列是既有建筑节能改造的一项有利途径，但
目前常用的基于地域最佳倾角的固定式光伏板
铺设方法存在不美观、经济效益低等问题。如
何借助数字化技术优化和确定固定式光伏板的
倾角，从而收获良好的经济和环境效益，成为
人们关注的热点问题。选择四川省成都市一处
既有建筑屋面为研究对象，借助Grasshopper工
具搭建参数化模拟优化平台，以总日照辐射量
和单位面积日照辐射量作为性能评价指标，对
坡屋面增设光伏阵列形态设计进行多目标优化
研究，得到屋面坡度、面积和朝向确定条件下光
伏阵列的一系列几何形态优化设计方案，经筛
选后优化方案较对照方案有较大提升。并对其
进行数据分析，发现以优化性能目标为方向的
几何形态设计建议：阵列倾角取场地条件范围
下最大值；同时光伏阵列宽距比满足倾角的正
弦函数关系，以此方案设计的光伏阵列的单位
面积日照辐射量和总日照辐射量相对符合经济效益需求。明确了该多目标优化方法流程对既有建筑屋面光伏阵列设计的积极作用。 |
| 关键词: 既有建筑 建筑附加光伏 参数化工具 遗传算法 多目标优化 |
| DOI：10.13791/j.cnki.hsfwest.20240120 |
| 分类号: |
| 基金项目:中建股份科技研发子课题（CSCEC-2022-Z-1-8） |
|
| Multi-objective optimization study on photovoltaic arrays on roofs of existing buildings |
|
LIN Geng,LIU Yucheng,LIU Li,CHEN Rui,XIE Rong
|
| Abstract: |
| Based on the background of energy saving, emission reduction and the dual-carbon goal,
green retrofit of existing buildings is an urgent need. Adding photovoltaic arrays on the roof is a
favorable way for energy-saving retrofit of existing buildings, but laying fixed photovoltaic panels
based on the geographically optimal inclination angle has problems, such as unattractive appearance
and low economic benefits. How to optimize and determine the tilt angle of fixed photovoltaic panels
with the help of digital technology, so as to reap good economic and environmental benefits, has
become a hot issue of concern.
An existing building’s roof in Chengdu, Sichuan Province, is selected as the research target, and
the PV array is laid along the north slope direction of the roof, with the overall lateral width equal to
the length of the roof. The length of the PV panels in this system is selected from 700 to 1 260 mm in
combination with the commonly used specifications in the project. The minimum value of the PV panel
spacing is taken as the value of the smallest specification of PV panels laying densely along the plane of 0.7
m, and the maximum value is taken as the width of the roof of 14 m. Due to the need to take into account
the effect of the wind loads on the PV arrays, the maximum angle of the PV panels with the roof surface is
taken as 45 degrees, also the thickness of the PV panels is taken as the commonly used specification of 35
mm. Taking the south edge of each PV panel as the axis and the roof surface as the starting plane, the PV
array can be rotated along the axis according to the tilt parameter, and the distance between each other can
be changed according to the spacing parameter. The semiconductor cell side of the PV panels is used as
the receiving solar radiation object, and the roof and the PV panel blocks are used as the relevant shading
objects. With the help of Grasshopper tool, a parametric simulation optimization platform is established.
SPEA-II genetic algorithm is selected as the optimization algorithm. Optimization objectives include the
total solar radiation and unit area solar radiation. Genetic algorithm calculations obtain 14 pareto frontier
on the scheme, which means 14 programs as the optimal design of the single PV panel vertical width, PV
array spacing and PV array tilt angle of PV arrays under the design conditions. There are two programs
comprehensively better than the other programs. And then the study considers the expected cost of the
optimization of the most in line with the project’s actual needs of the program filtered out. Taking the two
conventional design forms of direct laying along the roof and intensive laying according to the optimal
inclination angle in Chengdu as the comparison group, the screened multi-objective optimization scheme
has a greater improvement in the radiation per unit area and the total radiation than the comparison group.
Through data analysis, some design recommendations regarding comparable design in this district was
found: array tilt angle to take the maximum value of the range of site conditions; at the same time, the
PV array aspect ratio to meet the sinusoidal function of the tilt angle of the relationship between the solar
radiation per unit area of the PV array designed by this program and the total solar radiation is relatively
in line with the economic benefits of the needs of the PV array.
It is shown that this multi-objective optimization method process is oriented to the green
performance of the building, using genetic algorithms to generate the results of the scheme in a selforganized manner, compared with the other two scheme-simulation-optimization of the traditional
parametric design techniques in reducing the workload at the same time to obtain a more excellent
performance of the energy-saving retrofit of the building and resource conservation play a positiveeffect, and the results of this method can be used to improve the performance of the PV array. It confirms the feasibility of optimizing the design of fixed
PV arrays on the roofs of existing buildings through genetic algorithms. The optimization design idea, framework and non-dominated solution selection
proposed in the study can be used as a useful supplement to the design method of additional PV arrays on the roofs of existing buildings. |
| Key words: existing building building attached photovoltaic parametric tools genetic algorithm multi-objective optimization |
