Research On Key Technology Of Multi-objective Particle Swarm Optimization And Its Application In Air Conditioning System

The air conditioning system is often designed according to the standard operating conditions,but the actual operating conditions of air conditioning are real-time changes,so the air conditioning system running at fixed frequency is difficult to meet people's increasing needs in comfort and energy saving.Therefore,more and more scholars are engaged in the research of frequency conversion air conditioning technology.Inherent characteristics such as time delay,nonlinearity and coupling of air conditioning system,as well as the phase change characteristics of refrigerant inside the system,make the frequency conversion control of air conditioning system extremely complicated.Based on the comparative study of four commonly used multi-objective optimization algorithms,this thesis selects MOPSO algorithm which is suitable for the optimization control of air conditioning system to be improved,and applies it to the optimization control of compression air conditioning system,and puts forward a set of frequency conversion optimization control strategy of air conditioning system based on multi-objective optimization.The main research contents of this thesis include the following aspects:1.By summarizing the research background and progress of the optimization control of compression refrigeration system and the multi-objective particle swarm optimization algorithm,the advantages and disadvantages of the currently applied multi-objective optimization algorithm and the variable frequency control strategy of compression refrigeration system are summarized,and the research content of this thesis is proposed.2.The working principle of compression refrigeration system is discussed,and the experimental platform of compression refrigeration unit is set up.The experimental platform adopts the control board independently developed by the research group to realize the acquisition of data signals such as temperature,pressure and flow,as well as the transmission of control signals of compressor,condenser,fan,water pump and other frequency conversion equipment and electronic expansion valve.At the same time,the upper computer monitoring software is developed on Lab VIEW,and the real-time monitoring and storage of experimental data between the upper computer and the control board is realized by using RS485 communication mode,which provides data support for the modeling of compression refrigeration system components.3.The modeling method of each component of the compression refrigeration system is discussed.The model parameters are identified by using experimental data,and the equation constraint relation is prepared for the multi-objective optimization of the whole system.4.By analyzing and comparing the performance of four heuristic multi-objective optimization algorithms(NSGA-II,MOEA/D,RVEA and MOPSO)which are widely used in recent years,MOPSO algorithm is selected to improve.Based on the framework of multi-objective particle swarm optimization algorithm,the dynamic inertia weight adjustment strategy of crowded distance value is proposed to improve the virtual reference plane and reference vector construction method proposed by predecessors.It is results show that the proposed dynamic inertia weight adjustment strategy can maintain good development ability in the later iteration of the algorithm.5.Using the improved MOPSO algorithm,the frequency conversion optimization control strategy of the compression refrigeration system is proposed.The strategy to system for optimization target of minimizing the total energy consumption and energy consumption of the compressor system components model as the equality constraint,upper and lower boundaries of each variable as the inequality constraints,using the improved multi-objective particle swarm optimization algorithm according to the different working condition to optimization control system,reduces the power consumption of refrigeration units.