基于人工智能机器学习的热网管道保温优化设计  

Thermal Insulation Optimization Design of Heat Pipe Network Based on Artificial Intelligence and Machine Learning

作　　者：俞李斌 林俊光 蒋月红 马聪 YU Libin;LIN Junguang;JIANG Yuehong;MA Cong(ZheJiang Energy Research&Development Institute,Hangzhou 311121,China;The International Joint Lab on Low-grade Energy Utilization of Zhejiang Province(The International Joint Lab),Hangzhou 311121,China;Zhejiang Zheneng Jiahua Power Generation Co.,Ltd.,Jiaxing 314201,Zhejiang,China)

机构地区：[1]浙江浙能技术研究院有限公司,杭州311121 [2]浙江-北欧(浙能)低品位能源利用国际联合实验室,杭州311121 [3]浙江浙能嘉华发电有限公司,浙江嘉兴314201

出　　处：《能源研究与管理》2020年第4期116-120,共5页Energy Research and Management

摘　　要：由于保温管径和保温层厚度对管道经济性的影响并不是独立的,故传统“两步法”分步设计管径和保温层经济厚度具有较大的误差,经济性相对不高。提出了基于保温散热损失特性的热网热损模型,同时采用人工智能机器学习,对多重不确定性因素影响下的热网保温管道设计进行优化,弥补了传统设计“两步法”存在的缺点,达到保温层厚度和管径两者耦合影响的最优设计,对提高热力管网的经济性具有决定性意义。Because the insulation pipe diameter and the insulation layer thickness are independently affected by the economics of the pipeline,the traditional"two-step method"design of the pipe diameter and the insulation layer has a large error and low economic efficiency.This paper proposes a heat loss model of the heat pipe network based on the characteristics of thermal insulation and heat dissipation loss.At the same time,it uses artificial intelligence machine learning to optimize the design of the heat pipe network insulation under the influence of multiple uncertain factors,which makes up for the shortcomings of the traditional design"two-step method".The optimal design to achieve the coupling effect of insulation thickness and pipe diameter.It is of decisive significance to improve the economic efficiency of the heat pipe network.

关 键 词：人工智能 机器学习 热网管道 保温设计 

分 类 号：TK018[动力工程及工程热物理]