基于BP神经网络方法的热管中冷器热工性能分析

摘要

基于BP神经网络建立了热管中冷器的传热性能和阻力性能预测模型,采用基于Levenberg Marquardt的trainlm训练算法,对热管中冷器的传热性能和阻力性能进行了预测,经试验数据的验证,预测值与试验值吻合较好,传热性能网络预测最大相对误差为8.0%,平均相对误差为3.5%,阻力性能网络预测最大相对误差为13.1%,平均相对误差为5.1%,说明该预测模型能较精确地预测热管中冷器的热工性能,用于指导工程设计.最后利用该预测模型对热管中冷器的结构参数进行优化,得到最佳设计参数,为热管中冷器的开发研究与应用提供了依据.%Models for heat transfer performance and thermal resistance of heat pipe intercooler were established based on BP neural network. The heat transfer performance and thermal resistance were predicted by means of the Levenberg Marquardt training algorithm. The prediction results were in good agreement with the experimental results. For the network model of heat transfer, the maximum relative error is 8.0％ and the average relative error is 3.5％. For the network model of thermal resistance, the maximum relative error is 13.1％ and the average relative error is 5. 1％. Thus the prediction model of the heat pipe intercooler can be used for engineering design. The structural parameters were optimized with the BP neural network model and the optimal design parameters were obtained. The study will be the foundation for the development and application of heat pipe intercooler.