Research On The Control Method Of Low Temperature Waste Heat Power Generation System Power Conversion

The low-temperature waste heat power generation technology is based on OrganicRankine Cycle through the organic working fluid absorbing low-grade heat energy, thenswelling within the turbine to drive turbines rotate. The power to drive permanent magnetgenerators is from the utilization of industrial waste heat thermal energy. Because of thatthe power is limited, the utilization efficiency of energy is becoming more important.Therefore, the maximum power point tracking strategy of waste heat power generationsystem was studied, and the maximum efficiency of heat utilization was achieved throughadjusting the operating point.Firstly, with the basic theory of low-temperature waste heat power generation system,the Rankine cycle characteristics, the system features, application features and a variety ofworking fluid performance was analyzed, ultimately R600a was selected for circulatingrefrigerant; through the analysis of turbine thermodynamic properties, the factors ofcogeneration system maximum power point was researched; according turbinethermodynamic properties, as well as experimental data, the BP neural network wasestablished and the system mathematical model was fitted through the BP neural network.Secondly, according to the turbine characteristics and experimental data research, themaximum power point tracking (MPPT) control strategy of low-temperature waste heatpower generation system and the improvement of the program for hill-climbing searchmethod was proposed; velocity factor and the maximum step size of the hill-climbing wasdesigned; studied asynchronous motor vector control methods was studied, and combinedwith MPPT control to achieve maximum power tracking purposes; with MATLAB/Simulink simulation model as a platform, the system model was build and proved MPPTcontrol effectiveness.Finally, with the theory of waste heat power generation, through AutoCAD,experiment platform structure was designed through AutoCAD and experimental platformwas build; According to a large number of experimental data, the system output Efficiencywould be affected by working fluid pump frequency, the turbine speed and operating pressure; the experimental data was also recorded as BP neural network modeling trainingparameters.