为探究气象因素与光伏组件积灰之间的关系,提出一种基于改进秃鹰算法（IBES）优化的最小二乘支持向量机（LSSVM）的积灰预测模型。该模型以降雨量、风速等气象因素作为输入,对组件面积灰进行预测。通过引入高斯-柯西变异算子对种群最优个体进行变异,择优选取进入下一次迭代,改善原始秃鹰算法收敛速度慢、易陷入局部最优的缺点。将改进算法寻优得到的参数代入模型,仿真后与其他种类算法模型进行对比,结果表明IBES-LSSVM积灰预测模型预测误差更小,拟合效果更好。最后根据累计积灰计算发电损失,结合降雨情况对组件清洗进行指导。
Abstract
In order to explore the relationship between meteorological factors and photovoltaic panel ash deposition, an ash deposition prediction model based on least squares support vector machine （LSSVM） optimized by improved bald eagle algorithm （IBES） is proposed. The model uses meteorological factors such as rainfall and wind speed as input to predict the ash content of the plate area. By introducing the Gauss-Cauchy mutation operator to mutate the optimal individual of the population, the optimal selection is entered into the next iteration, which improves the shortcomings of the original vulture algorithm, such as slow convergence speed and easy to fall into local optimum. The parameters obtained by the improved algorithm are substituted into the model, and compared with other kinds of algorithm models after simulation. The results show that the prediction error of the IBES-LSSVM ash accumulation prediction model is smaller and the fitting effect is better. Finally, the power generation loss is calculated according to the cumulative ash accumulation, and the component cleaning is guided according to the rainfall conditions.
关键词
光伏组件 /
积灰 /
降雨 /
支持向量机 /
秃鹰算法
Key words
photovoltaic modules /
ash /
rainfall /
support vector machine /
bald eagle search
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 丁怡婷. 我国可再生能源发电装机容量超10亿千瓦[N]. 人民日报, 2021-11-29(001).
DING Y T. The installed capacity of renewable energy power generation in China exceeds 1 billion kW[N]. People’s daily, 2021-11-29(001).
[2] 陈金鑫, 潘国兵, 欧阳静, 等. 自然降雨下光伏组件积灰预测方法研究[J]. 太阳能学报, 2021, 42(2): 431-437.
CHEN J X, PAN G B, OUYANG J, et al.Study on prediction algorithm of dustfall on PV modules under natural rainfall[J]. Acta energiae solaris sinica, 2021, 42(2): 431-437.
[3] SAID S, WALWIL H M.Fundamental studies on dust fouling effects on PV module performance[J]. Solar energy, 2014, 107: 328-337.
[4] GHOLAMI A, KHAZAEE I, ESLAMI S, et al.Experimental investigation of dust deposition effects on photo-voltaic output performance[J]. Solar energy, 2018, 159: 346-352.
[5] CARON J R, LITTMANN B.Direct monitoring of energy lost due to soiling on first solar modules in California[J]. IEEE journal of photovoltaics, 2013, 3(1): 336-340.
[6] MICHELI L, MULLER M.An investigation of the key parameters for predicting PV soiling losses[J]. Progress photovoltaics research & opplications, 2017, 25: 291-307.
[7] JIANG Y, LU L,LU H.A novel model to estimate the cleaning frequency for dirty solar photovoltaic (PV) modules in desert environment[J]. Solar energy, 2016, 140: 236-240.
[8] GUO B, JAVED W, FIGGIS B.W, et al. Effect of dust and weather conditions on photovoltaic performance in Doha, Qatar[C]// 2015 First Workshop on Smart Grid and Renewable Energy (SGRE), Doha, Qatar, 2015.
[9] 张国宏, 谈建国, 郑有飞, 等. 上海市月降尘量与气象因子间关系研究[J]. 气象科学, 2006, 26(3): 3328-3333.
ZHANG G H, TAN J G, ZHENG Y F, et al.Relationships between monthly amount of dustfall and meteorological factors in Shanghai[J]. Scientia meteorologica sinica, 2006, 26(3): 3328-3333.
[10] JAVED W, GUO B, FIGGIS B W.Modeling of photovoltaic soiling loss as a function of environmental variables[J]. Solar energy, 2017, 157: 397-407.
[11] 韩坤. 基于ALSTM的光伏出力短期预测研究[J]. 仪器仪表用户, 2022, 29(2): 95-100.
HAN K.Research on short-term forecast of photovoltaic output based on ALSTM[J]. Instrumentation, 2022, 29(2): 95-100.
[12] 王粟, 隗磊锋, 曾亮. 基于PSO-RVM模型的光伏出力日前区间预测[J]. 湖北工业大学学报, 2021, 36(5): 37-41.
WANG S, WEI L F, ZENG L.Day ahead interval forecasting model of PV power based on PSO-RVM[J]. Journal of Hubei University of Technology, 2021, 36(5): 37-41.
[13] 郭枭, 田瑞, 邱云峰, 等. 清洁光伏组件温度预测及影响因素研究[J]. 太阳能学报, 2021, 42(11): 76-85.
GUO X, TIAN R, QIU Y F, et al.Study on temperature prediction and influence factors of clean photovoltaic modules[J]. Acta energiae solaris sinica, 2021, 42(11): 76-85.
[14] ALJARAH I, AL-ZOUBI A M, FARIS H, et al.Simultaneous feature selection and support vector machine optimization using the grasshopper optimization algorithm[J]. Cognitive computation, 2018, 10(2): 478-495.
[15] SUYKENS J A K, VANDEWALLE J. Least squares support vector machine classifiers[J]. Neural processing letters, 1999, 9(3): 293-300.
[16] 贾鹤鸣, 姜子超, 李瑶. 基于改进秃鹰搜索算法的同步优化特征选择[J]. 控制与决策, 2022, 37(2): 445-454.
JIA H M, JIANG Z C, LI Y.Simultaneous feature selection optimization based on improved bald eagle search algorithm[J]. Control and decision, 2022, 37(2): 445-454.
[17] ALSATTAR H A, ZAIDAN A A, ZAIDAN B B.Novel meta-heuristic bald eagle search optimisation algorithm[J]. Artificial intelligence review, 2020, 53(3): 2237-2264.
[18] 张燕, 陈慧丹, 周颖, 等. 基于改进的GWO-LSSVM磨矿粒度软测量模型[J]. 计算机仿真, 2020, 37(6): 298-304.
ZHANG Y, CHEN H D, ZHOU Y, et al.Soft-measuring model for grinding particle size based on the improved GWO-LSSVM model[J]. Computer simulation, 2020, 37(6): 298-304.
[19] 魏昕, 冯锋. 基于高斯-柯西变异的帝国竞争算法优化[J]. 计算机科学, 2021, 48(S2): 142-146.
WEI X, FENG F.Optimization of empire competition algorithm based on Gauss-Cauchy mutation[J]. Computer science, 2021, 48(S2): 142-146.
[20] HEGAZY Y G, SALAMA M, CHIKHANI A Y.Adequacy assessment of distributed generation systems using Monte Carlo simulation[J]. IEEE transactions on power systems, 2003, 18(1): 48-52.
基金
榆林市科学技术局产学研合作项目（CXY-2020-033）
PDF(1797 KB)
