| With the gradual acceleration of industrial development,the consumption of fossil fuels and the deterioration of the global environment have become urgent problems that are receiving attention from countries around the world.To solve the problem of energy shortages and improve the ecological environment,the renewable energy generation industry has emerged.Developing and utilizing solar energy for power generation,and photovoltaic(PV)power generation,which is clean and pollution-free,has become a trend.Currently,the installed capacity and power generation of PV are increasing year by year,and the reliability of PV inverters cannot be underestimated.PV inverters are the only flexible and controllable devices that connect PV power generation systems and the grid,and their safety is crucial to the smooth operation of the entire PV system.Among them,power electronic switching devices are the most fragile components of the entire inverter system and are prone to fatigue failure due to fluctuations in junction temperature.To solve the problem of thermal fatigue failure,active and passive thermal control can be implemented.Passive thermal control methods include increasing the heat sink area and fan flow rate,but their effectiveness is limited.Active thermal control(essentially active power control)involves dynamically adjusting the operation and control parameters of the photovoltaic inverter during operation,to reduce the power device power fluctuation caused by fluctuations in the photovoltaic output power,and thereby reduce device junction temperature fluctuation and improve thermal reliability.The most commonly used method in active power control is dynamic adjustment of pulse width modulation(PWM)frequency control,which does not require additional hardware and is highly flexible in the control process.However,the control loop of the traditional frequency-based active power control method has the following issues:1.The filter parameters used are not thoroughly discussed;2.The setting principle of the control frequency uses empirical values that cannot be obtained in advance;3.The focus of the control is on power device loss fluctuation control only,which is used to map junction temperature fluctuation.Regarding the first point,this paper adds a real-time filter parameter setting process based on traditional active power dissipation control.This process utilizes real-time rainflow counting method to select the filter time constant that maximizes the real-time calculated value of the power device’s lifetime within each control cycle.This improvement enhances the control effectiveness,and the proposed method’s effectiveness is verified through simulation.Regarding the second and third points,this paper proposes a generalized active power dissipation control method with the direct objective of achieving smooth junction temperature.The method redefines the principles for setting the switching frequency:changing the switching frequency to maximize the tracking of the junction temperature reference value.Depending on the selection principles for the junction temperature reference value,the method can be divided into three approaches:junction temperature tracking,equal-weighted averaging,and linear weighting.The junction temperature tracking approach adjusts the switching frequency to make the current junction temperature equal to the junction temperature at the previous sampling instant.The equal-weighted averaging approach sets the current junction temperature to the average value of all historical junction temperatures with equal weighting coefficients.The linear weighting approach modifies the weighting coefficients of historical junction temperature data in the equal-weighted averaging approach,giving higher weights to junction temperatures closer to the current sampling instant,thereby calculating the junction temperature reference value.The effectiveness of the proposed methods is verified through program simulation in the article. |