自适应加权最小二乘支持向量机的空调负荷预测

被引：20

作者：

赵超

戴坤成

机构：

[1] 福州大学石油化工学院

来源：

重庆大学学报 | 2016年 / 39卷 / 01期

关键词：

空调负荷; 预测; 自适应加权; 最小二乘; 支持向量机; 粒子群优化;

D O I：

暂无

中图分类号：

TU831.1 [空气特性]; TP18 [人工智能理论];

学科分类号：

081404 [供热、供燃气、通风及空调工程]; 140502 [人工智能];

摘要：

为了提高建筑空调负荷的预测精度,在分析空调负荷主要影响因素的基础上提出了一种基于自适应加权最小二乘支持向量机(AWLS-SVM)的建筑空调负荷预测方法。该方法根据预测误差的统计特性,采用基于改进正态分布加权规则,自适应地赋予每个建模样本不同的权值,以克服异常样本点对模型性能的影响。建模过程中采用粒子群优化(PSO)算法对模型参数进行优化,以进一步提高模型预测精度。基于DeST模拟数据将AWLS-SVM方法应用于南方地区某办公建筑的逐时空调负荷预测中,并与径向基神经网络(RBFNN)模型、LS-SVM模型及WLS-SVM模型作比较,其平均预测绝对误差分别降低了51.84%、13.95%和3.24%,并进一步基于实际空调负荷数据将该方法应用于另一办公建筑的逐日空调负荷预测中。预测结果表明:AWLS-SVM预测的累积负荷误差为4.56MW,亦优于其他3类模型,证明了AWLS-SVM具有较高的预测精度和较好的泛化能力,是建筑空调负荷预测的一种有效方法。

引用

页码：55 / 64

页数：10

共 22 条

[1]

Forecasting short-term electricity consumption using the adaptive grey-based approach—An Asian case.[J].Der-Chiang Li;Che-Jung Chang;Chien-Chih Chen;Wen-Chih Chen.Omega.2011, 6

[2]

Modeling and prediction of Turkey’s electricity consumption using Support Vector Regression.[J].Kadir Kavaklioglu.Applied Energy.2010, 1

[3]

Parallel Support Vector Machines Applied to the Prediction of Multiple Buildings Energy Consumption [J].

Zhao, Hai Xiang ;

Magoules, Frederic .

JOURNAL OF ALGORITHMS & COMPUTATIONAL TECHNOLOGY, 2010, 4 (02) :231-249

[4]

Particle Swarm Optimization-based LS-SVM for Building Cooling Load Prediction [J].

Li Xuemei ;

Shao Ming ;

Ding Lixing ;

Xu Gang ;

Li Jibin .

JOURNAL OF COMPUTERS, 2010, 5 (04) :614-621

[5]

Weighted least squares support vector machine local region method for nonlinear time series prediction.[J].Tingwei Quan;Xiaomao Liu;Qian Liu.Applied Soft Computing Journal.2009, 2

[6]

Greek long-term energy consumption prediction using artificial neural networks.[J].L. Ekonomou.Energy.2009, 2

[7]

Artificial neural networks for energy analysis of office buildings with daylighting.[J].S.L. Wong;Kevin K.W. Wan;Tony N.T. Lam.Applied Energy.2009, 2

[8]

Adaptive weighted least square support vector machine regression integrated with outlier detection and its application in QSAR [J].

Cui, Wentong ;

Yan, Xuefeng .

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2009, 98 (02) :130-135

[9]

Applying support vector machine to predict hourly cooling load in the building.[J].Qiong Li;Qinglin Meng;Jiejin Cai;Hiroshi Yoshino;Akashi Mochida.Applied Energy.2008, 10

[10]

Predicting hourly cooling load in the building: A comparison of support vector machine and different artificial neural networks.[J].Qiong Li;Qinglin Meng;Jiejin Cai;Hiroshi Yoshino;Akashi Mochida.Energy Conversion and Management.2008, 1

← 1 2 3 →