摘要
针对直膨式太阳能热泵(DX-SAHP)系统运行工况复杂、系统热力性能预测较为困难的问题,提出基于多元线性回归算法的系统性能预测模型。在对环境参数和运行参数与系统性能系数(COP)之间相关性分析的基础上,利用Python语言编写了具有自学习能力的训练模型程序,分别将春秋季、夏季和冬季工况下的实验数据作为模型样本训练数据,实现模型的自我完善。获得不同季节工况下的系统性能预测模型,对模型的准确性进行F检验和T检验。结果表明,全年工况下DX-SAHP系统COP预测值和实测值之间的平均相对误差为6.4%,验证了所构建系统性能预测模型的准确性。
To solve the problem that the complex operating conditions make it difficult to predict the thermal performance of a directexpansion solar-assisted heat pump(DX-SAHP)system,a performance prediction model of the DX-SAHP system by using a multiple linear regression algorithm is proposed.On the basis of analyzing the correlation between environmental and operating parameters and system coefficient of performance(COP),a training model program with self-learning ability is written in Python language,and the experimental data in spring,autumn,summer and winter conditions are taken as the training data of the model samples respectively to realize its self-improvement.The system performance prediction models under different seasonal working conditions are obtained,and the accuracy of the model is also validated by F test and T test.The results show that under the whole-year working condition,the mean relative error between the predicted and measured COP values of the DX-SAHP system is 6.4%,which verifies the accuracy of the proposed model.
作者
孔祥强
刘晓东
尚燕平
李瑛
李见波
Kong Xiangqiang;Liu Xiaodong;Shang Yanping;Li Ying;Li Jianbo(College of Mechanical and Electronic Engineering,Shandong University of Science and Technology,Qingdao 266590.China)
出处
《太阳能学报》
EI
CAS
CSCD
北大核心
2022年第1期443-449,共7页
Acta Energiae Solaris Sinica
基金
国家自然科学基金面上项目(51776115)
山东省研究生导师指导能力提升项目(SDYY17037)
山东科技大学研究生导师指导能力提升计划(KDYC17009)。
关键词
太阳能
热泵系统
热水器
性能系数
多元线性回归
预测模型
solar energy
heat pump systems
water heaters
coefficient of performance
multiple linear regression
prediction model
