With the construction of smart grid,lots of renewable energy resources such as wind and solar are deployed in power system.It might make the power system load varied complex than before which will bring difficulties i...With the construction of smart grid,lots of renewable energy resources such as wind and solar are deployed in power system.It might make the power system load varied complex than before which will bring difficulties in short-term load forecasting area.To overcome this issue,this paper proposes a new short-term load forecasting framework based on big data technologies.First,a cluster analysis is performed to classify daily load patterns for individual loads using smart meter data.Next,an association analysis is used to determine critical influential factors.This is followed by the application of a decision tree to establish classification rules.Then,appropriate forecasting models are chosen for different load patterns.Finally,the forecasted total system load is obtained through an aggregation of an individual load’s forecasting results.Case studies using real load data show that the proposed new framework can guarantee the accuracy of short-term load forecasting within required limits.展开更多
Although the recent load information is critical to very short-term load forecasting(VSTLF), power companies often have difficulties in collecting the most recent load values accurately and timely for VSTLF applicatio...Although the recent load information is critical to very short-term load forecasting(VSTLF), power companies often have difficulties in collecting the most recent load values accurately and timely for VSTLF applications.This paper tackles the problem of real-time anomaly detection in most recent load information used by VSTLF.This paper proposes a model-based anomaly detection method that consists of two components, a dynamic regression model and an adaptive anomaly threshold. The case study is developed using the data from ISO New England. This paper demonstrates that the proposed method significantly outperforms three other anomaly detection methods including two methods commonly used in the field and one state-of-the-art method used by a winning team of the Global Energy Forecasting Competition 2014. Finally, a general anomaly detection framework is proposed for the future research.展开更多
提出了一种基于贝叶斯证据框架下加权最小二乘支持向量机(Weighted Least Squares Support Vector Machine,WLS-SVM)的短期负荷预测模型和算法。在对历史负荷数据进行完预处理基础上,分析影响负荷变化的重要因素,然后选择最佳的输入数...提出了一种基于贝叶斯证据框架下加权最小二乘支持向量机(Weighted Least Squares Support Vector Machine,WLS-SVM)的短期负荷预测模型和算法。在对历史负荷数据进行完预处理基础上,分析影响负荷变化的重要因素,然后选择最佳的输入数据作为LS-SVM训练模型的输入向量。通过贝叶斯证据三层推断寻找到模型的最佳参数:第一层推断确定LS-SVM的权向量w和偏置值b,第二层推断确定模型的超参数γ,第三层推断确定核函数的超参数σ。为了提高模型的鲁棒性,赋予了每个样本误差不同的权系数,建立了具有良好泛化性能的WLS-SVM回归模型,从而进一步提高了模型预测的精度。采用上述方法对黑龙江电网短期负荷进行了预测,结果证明了该方法具有良好的预测效果。展开更多
本文提出了一种基于贝叶斯证据框架下加权最小二乘支持向量机(Weighted Least Squares Support Vector Machine,WLS-SVM)的短期负荷预测模型和算法。在对历史负荷数据进行完预处理基础上,分析影响负荷变化的重要因素,然后选择最佳的输...本文提出了一种基于贝叶斯证据框架下加权最小二乘支持向量机(Weighted Least Squares Support Vector Machine,WLS-SVM)的短期负荷预测模型和算法。在对历史负荷数据进行完预处理基础上,分析影响负荷变化的重要因素,然后选择最佳的输入数据作为LS-SVM训练模型的输入向量。通过贝叶斯证据三层推断寻找到模型的最佳参数:第一层推断确定LS-SVM的权向量w和偏置值b,第二层推断确定模型的超参数γ,第三层推断确定核函数的超参数σ。为了提高模型的鲁棒性,赋予了每个样本误差不同的权系数,建立了具有良好泛化性能的WLS-SVM回归模型,从而进一步提高了模型预测的精度。采用上述方法对一固定预测区电网中期负荷进行了预测,结果证明了该方法具有良好的预测效果。展开更多
文摘With the construction of smart grid,lots of renewable energy resources such as wind and solar are deployed in power system.It might make the power system load varied complex than before which will bring difficulties in short-term load forecasting area.To overcome this issue,this paper proposes a new short-term load forecasting framework based on big data technologies.First,a cluster analysis is performed to classify daily load patterns for individual loads using smart meter data.Next,an association analysis is used to determine critical influential factors.This is followed by the application of a decision tree to establish classification rules.Then,appropriate forecasting models are chosen for different load patterns.Finally,the forecasted total system load is obtained through an aggregation of an individual load’s forecasting results.Case studies using real load data show that the proposed new framework can guarantee the accuracy of short-term load forecasting within required limits.
基金supported in part by the National Natural Science Foundation of China(No.71701035)the US Department of Energy,Cybersecurity for Energy Delivery Systems(CEDS)Program(No.M616000124)
文摘Although the recent load information is critical to very short-term load forecasting(VSTLF), power companies often have difficulties in collecting the most recent load values accurately and timely for VSTLF applications.This paper tackles the problem of real-time anomaly detection in most recent load information used by VSTLF.This paper proposes a model-based anomaly detection method that consists of two components, a dynamic regression model and an adaptive anomaly threshold. The case study is developed using the data from ISO New England. This paper demonstrates that the proposed method significantly outperforms three other anomaly detection methods including two methods commonly used in the field and one state-of-the-art method used by a winning team of the Global Energy Forecasting Competition 2014. Finally, a general anomaly detection framework is proposed for the future research.
文摘提出了一种基于贝叶斯证据框架下加权最小二乘支持向量机(Weighted Least Squares Support Vector Machine,WLS-SVM)的短期负荷预测模型和算法。在对历史负荷数据进行完预处理基础上,分析影响负荷变化的重要因素,然后选择最佳的输入数据作为LS-SVM训练模型的输入向量。通过贝叶斯证据三层推断寻找到模型的最佳参数:第一层推断确定LS-SVM的权向量w和偏置值b,第二层推断确定模型的超参数γ,第三层推断确定核函数的超参数σ。为了提高模型的鲁棒性,赋予了每个样本误差不同的权系数,建立了具有良好泛化性能的WLS-SVM回归模型,从而进一步提高了模型预测的精度。采用上述方法对黑龙江电网短期负荷进行了预测,结果证明了该方法具有良好的预测效果。
文摘本文提出了一种基于贝叶斯证据框架下加权最小二乘支持向量机(Weighted Least Squares Support Vector Machine,WLS-SVM)的短期负荷预测模型和算法。在对历史负荷数据进行完预处理基础上,分析影响负荷变化的重要因素,然后选择最佳的输入数据作为LS-SVM训练模型的输入向量。通过贝叶斯证据三层推断寻找到模型的最佳参数:第一层推断确定LS-SVM的权向量w和偏置值b,第二层推断确定模型的超参数γ,第三层推断确定核函数的超参数σ。为了提高模型的鲁棒性,赋予了每个样本误差不同的权系数,建立了具有良好泛化性能的WLS-SVM回归模型,从而进一步提高了模型预测的精度。采用上述方法对一固定预测区电网中期负荷进行了预测,结果证明了该方法具有良好的预测效果。