新型矿用管芯式散热器热工性能的预测方法  

Prediction of Thermal Performance for a New Type of Mining Tube Core Radiator

作　　者：班淑珍[1] 席明智[1] 

机构地区：[1]内蒙古工业大学能源与动力工程学院

出　　处：《工程机械》2009年第6期43-46,共4页Construction Machinery and Equipment

基　　金：内蒙古自治区自然科学基金(200308020209)

摘　　要：新型管芯式散热器既有良好的散热性和抗振性,又维修方便,在工作条件恶劣的矿用汽车上应用优势明显。通过分析影响散热器传热性能的主要因素,利用改进BP神经网络方法建立新型矿用管芯式散热器传热性能预测模型,并利用建立的阻力性能计算模型编制预测阻力性能的程序,对具有2、3、4排管芯式散热器的热工性能作出预测,同时,搭建风筒试验台架对管芯式散热器进行热工性能试验。通过比较预测结果和试验结果发现,两者相对误差在±5%左右,可用于工程设计。A new type of tube core radiator is of good heat dissipating and anti-vibration capabilities and features unique serviceability so it presents an obvious superiority when applied on mining trucks under rough work conditions. Through an analysis on main factors influencing heat transfer of the radiator, a prediction model for heat transfer behavior of the new type mining tube core radiator is set up with improved BP nerval network method and a program to predict resistance performance is compiled with the established calculation model of resistance performance. Its thermal characteristics test is conducted on constructed windbarrel test bench. BP nerval network prediction program is compiled with MATLAB nerval network toolbox. Applying nerval network method and the program model of resistance performance, predictions for thermal performances of the second, the third and the forth of tube rows are fulfilled. Comparing the predicted results with the test results, it is found out that the relative error is around 5%, which can be accepted when applied to engineering design.

关 键 词：改进BP神经网络 管芯式散热器 热工性能 预测模型 

分 类 号：U464.13[机械工程—车辆工程] U469.6[交通运输工程—载运工具运用工程]