神经网络软测量在家用房间空调器制冷量测量中的研究与应用  

Research and Application of Soft Measurement by Neural Network in Cooling Capacity Measurement of Room Air Conditioner

作　　者：翁俊杰 

机构地区：[1]威凯检测技术有限公司

出　　处：《日用电器》2019年第8期152-158,共7页ELECTRICAL APPLIANCES

摘　　要：软测量是应用计算机技术对难以测量或无法直接测量的变量,选择与之相关的一些容易测量的变量,通过建模变量关系来推断或者估算。本文针对普通家用房间空调器系统的制冷量测量提出了一种基于BP神经网络的软测量方法。目前普通家用房间空调器多数为一拖一空调系统,通过测量采集空调系统中与制冷量变化相关的状态参数,采用BP神经网络预测系统制冷量,并在MATLAB中进行了仿真。同时采用平均影响值MIV对BP神经网络输入层节点参数的权重进行分析,从而达到变量的筛选,发现各变量在系统中作用的重要程度。结果表明,采用BP神经网络软测量的预测效果与实际测量结果偏差较小,可以用于家用房间空调系统制冷量的测量,且输入变量的权重分析结果与影响系统制冷量的组件结构特性较为一致。Soft measurement is the application of computer technology to select variables which are difficult to measure or can not be measured directly.According to selecting some variables which are easy to measured,we can infer or estimate target variables by modelling the relationship.In this paper,a soft measurement method based on BP neural network is proposed for measuring the cooling capacity of room air conditioner.At present,most of the air conditioners in common household used are one-to-one air conditioning system.By measuring and collecting the state parameters related to the change of cooling capacity,and used BP neural network to predict the cooling capacity of the system,and then simulate in MATLAB.At the same time,the average impact value MIV is used to analyze the weights of the input layer node parameters of BP neural network,so as to achieve the selection of variables and find the importance of each variable in the system.The results show that the predictive effect of soft measurement based on BP neural network is less deviated from the actual measurement result.It can be used to measure the cooling capacity of room air conditioning system.The importance analysis are shown thatthe MIV is consistent with the structural characteristics of the components that affect the cooling capacity of the system.

关 键 词：空调系统 制冷量 BP神经网络 软测量 MIV变量筛选 

分 类 号：TB6[一般工业技术—制冷工程]