Study On Optimal Control Of Internal Combustion Engine Waste Heat Utilization System Basde On Organic Rankine Cycle

The rapid development of human society and the progress of science and technology make the internal combustion engine as the highest thermal efficiency of the power machinery,widely used in various fields of production and life.With the rapid development of economy,a large amount of energy waste and environmental pollution are also caused.The recycling of the exhaust heat of internal combustion engine(ICE)by the method of organic Rankine cycle,not only can realize the reuse of energy,but also protect the environment.According to the Organic Rankine Cycle(ORC)based exhaust heat utilization system in internal combustion engine,the work and the main research results of this paper mainly include three parts.The first part is aimed at the operating conditions monitoring of waste heat utilization system of vehicle internal combustion engine.Based on support vector machine(SVM)method,the monitoring program of ICE-ORC system's running condition is designed,and several parameter optimization algorithms are used to optimize the parameters,and the remarkable optimization monitoring performance is obtained.The second part focuses on the control strategy of ORC system for automotive internal combustion engine.In order to ensure the heat can be used in the depth of the car running,the control problems need to be solved through the analysis of the control system.Based on the SVM preview controller,according to the prediction of automobile exhaust temperature and mass flow,the expansion speed,the speed of the pump,the throttle pressure and the degree of overheat of the system are controlled in a stable range,and the simulation results show that the control strategy can achieve satisfactory control results.The third part is mainly about the fault diagnosis in the operation process of ORC system.Using the extreme learning machine and the neighbor algorithm,two kinds of classification algorithms in machine learning domain,to classify the faults of ORC system.Simulation results show that the extreme learning machine shows better classification performance in the classification of fault data of small samples in ORC system.