Life Prediction Of Photovoltaic Inverter Based On Mission Profile

With the gradual depletion of traditional fossil energy reserves,a large number of new energy sources are connected to the power system.As the key link of photovoltaic grid-connected networks,photovoltaic inverters’ reliability issues have received more and more extensive attention.After the photovoltaic system is put into use,there are great limitations in improving the reliability of the photovoltaic inverter by improving the topology and control strategy,as they have been identified.Therefore,establishing the life prediction model of photovoltaic inverter to improve its reliability has become an urgent problem to be solved.Firstly,the concept of reliability and the evaluation method of reliability have been introduced.Two types of devices that are most vulnerable to failure in photovoltaic inverters:power switching devices IGBTs and electrolytic capacitors’failure mechanisms and life prediction models have been introduced and analyzed.It is pointed out that temperature cycling caused by power fluctuations is the main cause of fatigue failure of IGBTs.The internal temperature rise caused by ESR of electrolytic capacitors is the main reason for the aging of electrolytic capacitors.Both reveal that temperature is one of the main factors leading to device failure.Then,in order to obtain the IGBT junction temperature variation curve when the load changes abruptly,the traditional Cauer thermal resistance network model is improved,and a thermal resistance network model that considers the influence of temperature on the thermal resistance of the IGBT layer materials is established.The IGBT dynamic junction temperature curves under traditional model and improved model were compared and analyzed.Based on the improved thermal resistance network model,the IGBT junction temperature variation curve under different load changes is obtained,and the three-dimensional safe operation domain of the IGBT is obtained by combining the thermal breakdown critical value.The established IGBT power loss model and thermal resistance network model laid the foundation for subsequent IGBT life prediction.Then,the life prediction process of the device based on the mission profile is given,and the actual environment of photovoltaic inverter in actual operation within one year is taken as the input,and the conversion of mission profile to the output power of photovoltaic inverter,current profile,power loss profile,temperature profile,and predicted lifetime of the device are completed.In order to obtain the current profile more simply and quickly,the current spectrums of IGBT and electrolytic capacitor are solved based on the double Fourier series,and compared with the simulation results,the correctness and accuracy of the method are verified.The rain flow counting method and fatigue cumulative damage theory are implemented in the program.Taking a 10 kw H bridge resistive load inverter as an example,the junction temperature distribution diagram of the IGBT is obtained by the rain flow counting method and the Miner cumulative damage criterion is used to obtain the device’s Predicted lifetime value.Finally,based on the model Carlo simulation method,the lifetime distribution of the 10,000 samples of the device was obtained.The unreliability model of the device was established through the lifetime distribution.The system-level unreliability model was established based on the reliability block diagram method,and the curve of unreliability of PV inverter with running time was obtained.The prediction results show that,the PV inverter system unreliability becomes 1%and 10%when the operating time reaches 3.1 years and 4.1 years respectively.The confidence level value can be selected according to the reliability requirements of different photovoltaic systems to predict the life of the photovoltaic inverter.This method is valuable in engineering for the reliability evaluation of photovoltaic inverters and the selection of inspection cycles.