Research Of Intelligent Control Method And System Integration For Photovoltaic Grid-connected Inverter

With the development of science and technology in promoting development of human society,but also consumes a lot of energy,causing pollution on the environment.Therefore,the use of renewable energy,reduce environmental pollution is currently one of the important tasks in scientific research.Renewable energy resources include hydroenergy,wind energy,solar energy,biomass energy and so on.Photovoltaic power generation is an important part of renewable energy application and photovoltaic power generation system can be divided into off-grid and grid-connected.Because grid-connected PV system is of low cost and flexible for installation,it is currently in possession of a large market share in the photovoltaic system.Grid-connected PV system has now entered the practical stage after years of research.And with the development of photovoltaic technology,further relevant research has still to be carried out in terms of security,reliability and efficiency of grid-connected solar system.At present,most of the control methods applied in inverter are traditional methods,which,based on an accurate model of the controlled object and lacking flexibility and adaptability,are applicable to such relatively simple control problems as linear or time-invariant problems.However the PV inverter is a nonlinear,time-varying system.So traditional control methods can not satisfy the control needs of the photovoltaic inverter.Intelligence control is an interdisciplinary approach which combines control theory and artificial intelligence techniques in order to adapt itself to the complexity and uncertainty of a controlled object.This paper chooses domestic 3KW single phase grid-connected inverter as the study object and in-depth research is carried out concerning the extraction method of the parameters of photovoltaic module,maximum power point tracking method and controlling method for grid-connected inverter.In this paper,such modern intelligent control technologies as fuzzy control,particle swarm optimization algorithm,sliding mode control and so on are applied to controlling photovoltaic grid-connected inverter so as improve dynamic and steady properties of the system,so that the quality of output power meets the grid requirements.This paper includes the following parts:Chapter One gives a brief introduction to the background,significance,current situation and the main contents of the research.In Chapter Two,parameter extract methods of photovoltaic cells are studied.It is have great significance to extract the parameters of solar cells according to measurements of solar cell I-V curve for improve the performance of solar cells and modeling photovoltaic power generation system.This paper presents a method to extract the parameters of solar cell based on combined chaotic particle swarm algorithm,it initialized particle swarm by the chaotic algorithm,then the optimal location of populations by chaotic optimization method.In order to verify the validity of the proposed method,extract the parameters of solar cell in different search range by chaotic particle swarm optimization algorithm,the traditional particle swarm optimization and genetic algorithms respectively,and conduct a comparative analysis.The results show that the method proposed in this paper with advantages of high accuracy,faster search speed.Chapter Three focuses on the study of Maximum power point tracking strategy of photovoltaic power generation system.Currently,conversion efficiency of photovoltaic cells is also relatively low.To improve the efficiency of energy conversion of photovoltaic power generation system,the maximum power point tracking(maximum power point tracking MPPT)technology should be applied.Currently,frequently-used maximum power point tracking methods include constant voltage method,perturb observation and incremental conductance method.Although these methods improve the efficiency of the system to some extent,there still exist some shortcomings.In dealing with the problem that the output power of photovoltaic power generation system with perturb observation method oscillates around the maximum power,this paper designs a fuzzy controller with PSO algorithm(particle swarm optimization PSO),and applies it to PV maximum power point tracking system(maximum power point tracking MPPT).The controller,which adopts particle swarm algorithm to optimize the membership function of fuzzy controller,is easy to implement.It can adjust the real-time track step to ensure that system possesses fast dynamic response speed and higher steady-state accuracy when light intensity or temperature changes.In this paper,perturb observation method,fuzzy control method and fuzzy controllers optimized with particle swarm optimization algorithm are also applied to the simulation in the same situation and the test results show that the method is effective and robust.Chapter Four is involved in the study of control strategies of grid-connect photovoltaic power generation system.In this part,the operation principle of the single-phase PV grid-connected inverter is analyzed and its mathematical model is established.PV system is a nonlinear control object,and sliding mode control strategy is a nonlinear control strategy with fast response,strong robustness,and insensitivity to the disturbance of outside interference and the parameter variation.However,the traditional response of sliding mode control includes such two stages as approaching motion and sliding motion.Sliding mode control system in a sliding motion stage is robust,but the system in an approaching motion stage does not have robustness because parameter perturbations and external disturbances have a significant impact on the system.To deal with the shortcomings of the traditional sliding mode control,this paper,based on the global sliding mode control theory,designs a dynamic non-linear sliding surface equation to ensure that the system is robust throughout the whole response process.To cope with the chattering phenomenon of sliding mode control,in this paper,a combination control strategy of the fuzzy control and sliding mode control is used in single-phase grid-connected inverter control,fuzzy rules are used to estimate the switching gain of the sliding mode controller to reduce the chattering phenomenon of the sliding mode controller,and Moreover feedforward control is applied to compensating for the voltage so that the system has better robustness and adaptability.To validate the effectiveness of proposed control strategy,this paper simulates the proposed method in Matlab/Simulink and compares the grid-connected inverter controlled by the proposed method.The results show that output current of the inverter with the global sliding mode control can better track the grid voltage,with higher power factors and low THD(total harmonic distort).When grid power supply is stopped,a self-powered island formed with the distributed power generation system and neighboring load.In an islanding state,the grid frequency and voltage are not under a power company’s control.So in such a state,harm to man and equipment is inevitable.Therefore,grid-connected PV system must have a function of islanding detection.This paper,based on the analysis of the current program of islanding detection,adopts an improved frequency-shift method to achieve the active islanding detection.Through simulation,the method turns out to be feasible.In Chapter Five,according to the control method proposed in previous sections,a test platform for single-phase photovoltaic inverter is constructed,the hardware and software debugging is completed and the relevant tests are conducted and the test results prove the effectiveness and feasibility of the control strategy proposed in this paper.Chapter Six summarizes the innovative points and deficiencies of this paper,and points out the prospect of the future research.