朴素贝叶斯下新型建筑室内空调瞬态能耗预测  

Transient Energy Consumption Prediction of Indoor Air Conditioning in New Buildings under Naive Bayes

作　　者：马博华[1] 周磊[1] MA Bo-hua;ZHOU Lei(Shenyang Ligong University,Shenyang Liaoning 110000,China)

机构地区：[1]沈阳理工大学,辽宁沈阳110000

出　　处：《计算机仿真》2024年第8期517-520,556,共5页Computer Simulation

基　　金：教育部协同育人项目(22087006115257)。

摘　　要：新型建筑通常具有较为复杂的结构,使得室内空调系统的能耗受到如建筑朝向、室内人员活动以及设备使用情况等因素影响,导致无法精准预测空调能耗。为此,提出新型建筑室内空调瞬态能耗朴素贝叶斯预测方法。采用K均值聚类(K-means clustering,K-Means)算法将新型建筑室内空调划分为除湿模式、制冷模式、通风模式、制热模式和自动模式,并分析影响空调瞬态能耗的因素,建立朴素贝叶斯预测模型,剔除空调瞬态能耗异常数据,输入至朴素贝叶斯预测模型中,实现新型建筑室内空调的瞬态能耗预测。仿真结果表明,所提方法可有效完成不同模式和运行时间下的瞬态能耗预测,与实际能耗之间的偏离程度较低,期望误差仅为2.31%,变异系数为4.76%,预测误差小。Generally,new buildings have complex structures,so the energy consumption of indoor air conditioning systems is affected by various factors such as building orientation,indoor human activities and equipment usage,resulting in inaccurate prediction of air conditioning energy consumption.Therefore,a prediction method for transient energy consumption of indoor air conditioning systems in new buildings was proposed based on naive Bayesian.Firstly,the K-means clustering(K-Means)algorithm was adopted to divide the indoor air conditioners in new buildings into several modes:dehumidification mode,refrigeration mode,ventilation mode,heating mode and automatic mode.Meanwhile,the factors affecting the transient energy consumption of air conditioners were analyzed.Moreover,a naive Bayesian prediction model was constructed.Furthermore,the abnormal data of transient energy consumption of air conditioners was eliminated and input into the naive Bayes prediction model.Finally,the transient energy consumption prediction was realized.Simulation results show that the proposed method can effectively predict the transient energy consumption under different modes and running times,and the deviation degree from the actual energy consumption is low.In addition,the expected error is only 2.31%,and the coefficient of variation is 4.76%.

关 键 词：新型建筑 空调瞬态能耗 模式划分 朴素贝叶斯预测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]