An Optimization Of Heat Pump And Refrigeration System By The Method Of Genetic Simulated Annealing Algorithm

The match of parts in heat pump and refrigeration system is a key for systemic performance. Reasonable matching could improve systemic performance and reduce running expenditure. The traditional optimization algorithms, including grads-Newton method, penalty function method, single shape method, polyhedron method and so on, have some disadvantages such as easy divergence,high dependability on initial value and long consumed time in the optimization of intricate system. The genetic algorithm and the simulated algorithm are two important methods in artificial intelligence and often used to the study of the optimization of intricate combinated system for their high parallel and quick convergence. The genetic simulated annealing algorithm, as an improved genetic algorithm by means of the simulated annealing algorithm, has high parallel and fast optimization rate in systemic optimization.For the one-rank steam compressed system, based on the steady mathematic models of heat pump and refrigeration system, the genetic simulated annealing algorithm was applied to optimize the heat pump and refrigeration system. With the value of systemic COP as the index, this paper maily discussed the match of the output capacity of compressor, the area of condensor and evaporator at the heat load of 36.6kW and the cold load of 36.47kW, and the change tendency of the output capacity of compressor, the area of condensor and evaporator in some range of systematic capacity.Results indicate that the genetic simulated annealing algorithm is better than Genetic algorithm in stability, thus it can be used to investigate into the problem of optimization and matching in heat pump and refrigeration system.