Optimal design of convection heating surface of boiler based on genetic algorithm

钟崴,吴燕玲,童水光,葛俊旭,周懿,谢金芳

DOI: https://doi.org/10.3785/j.issn.1008-973X.2010.12.010

2010-01-01

Abstract:An optimal design model of boiler convection heating surface which defined key structure parameters as decision variables was established for optimizing the structure and then reducing the manufacturing cost of boiler. An optimization method based on genetic algorithm was proposed to minimize the heat transfer area while accomplish a specified heat transfer task. The structure parameters of convection heating surface were defined by integer-code and controlled by introducing parameter constraints, dominant function constraints and recessive function constraints to satisfy the performance requirements. Supported by boiler design case library, the initial population of design scheme was generated via case-based method combined with random generation method which included design experience and search ability of genetic algorithm. Uniform crossover and arithmetic crossover, as well as uniform mutation were used simultaneously to generate new individuals of design scheme. Proportional selection and elitist preservation were adopted to guide the genetic algorithm process oriented to the global optimization. The results indicate that the model and method is suitable for the structure optimal design of boiler convection heating surface and can improve the design quality effectively.