Research On Solar Cell Optimization Model Design And Solution Algorith

As one of new and potential energy sources,solar energy,which possesses the advantages of nonpollution and renewable,has attracted much attention in the field of new energy.As the crucial component of the photovoltaic power generation system,photovoltaic cells can convert solar energies into electrical ones.Nevertheless,it has been being an important issue to guarantee the normal operation of such a system in the field of new energy,find the cause of operation failure in time,and accurately grasp the change of battery parameters.Accordingly,based on the operation principles of photovoltaic cells and arrays,studies on photovoltaic voltage parameter identification and related intelligent optimization approaches are helpful for not only understanding the changeable characteristics of the photovoltaic cells’ parameters and guaranteeing the maximum of electric energy production and power,but also timely catching the operation status of the system and tracking the maximum power point.These are of important theoretical and practical significance for promoting the development of photovoltaic power generation industry.To this end,several identification models and related improved intelligent optimization algorithms are developed to solve the problem of parameter identification of photovoltaic cell and array in static and noisy environments,based on the measured and calculated values of output voltage and current.The acquired achievements help to not only timely know the influence of solar cell parameters on electrical energy,but also promote the development of intelligent optimization in the field of new energy research and application.The main work and achievements can be summarized below:A.To overcome the influence of leakage current and drift current in that the double-diode photovoltaic model cannot completely characterize the status of the cell operation,a four-diode parameter identification optimization model with multiple parameters is constructed based on photovoltaic cells,after extending the double-diode structure of the cells to a four-diode structure.Herein,the whole difference between the measured and calculated output currents of photovoltaic cells is taken as the performance index.Further,an improved capuchin search algorithm is developed to solve the acquired identification model,after the strategies of gray wolf predation and Cauchy variation are introduced into the capuchin search algorithm.Comparative experiments show that not only the identification model is rational and effective,but also the acquired algorithm has significant advantages over the compared approaches with the aspects of solution search,efficiency,parameter identification effect,and the potential to solve complex optimization problems.B.The existing photovoltaic arrays usually require that the characteristics of all photovoltaic cells are the same,which necessarily causes the difficulty of fault detection in the operation of Photovoltaic array.Thus,the parameter identification optimization models of the single-and fourdiode photovoltaic array models are constructed based on the designed structural models of a singlediode photovoltaic array and a four-diode one,in which for a given photovoltaic array,the whole difference between the measured and calculated values of photovoltaic array output voltage is taken as the performance index.Subsequently,based on the above improved capuchin search algorithm,another improved approach is developed to seek the optimal identification schemes of the above identification models,after the strategies of population partitioning and sine-cosine update are introduced into such an improved algorithm.Comparative experiments show that the constructed optimization models for single-and four-diode photovoltaic arrays are rational and effective,and the acquired algorithm can not only solve the benchmark test problems,but also execute the photovoltaic array’s parameter identification effectively.C.The cell performances of photovoltaic arrays are seriously influenced by noisy environments,which results that it is an important topic in the designs of such arrays to effectively inhibit the influence of noise on output voltage and timely know the changeable characteristics of cell parameters.Therefore,the parameter identification problems of the above static single-and fourdiode photovoltaic array models are transformed into expected value programming models.Thereafter,an improved vulture search algorithm suitable for noisy optimization is proposed to solve such identification models,after introducing adaptive control factors,t-distribution mutation,and dynamic inertia weights into the classical vulture search algorithm.Comparative experiments show that,not only the designs of the acquired identification models are rational,but also the improved algorithm can not only effectively inhibit noise interference to solution search effect,but also obtain the satisfactory parameter identification schemes of the photovoltaic arrays in noisy environments.