Power Load Forecasting Based on the Combined Model of LSTM and XGBoost

Accurate power load forecasting can provide effective and reliable guidance for power construction and grid operation, and plays a very important role in the power grid system. In order to improve the accuracy of power load forecasting, this paper proposes a combined forecast model based on LSTM and XGBoost. The LSTM forecast model and the XGBoost forecast model are firstly established and the power load is predicted by using the two models respectively. Then the combined model predicts the power load by using the error reciprocal method to combine the results from the two single models. Through the experimental verification of the power load data of The Electrician Mathematical Contest in Modeling, the forecast error of the combined model we got is 0.57%, which is significantly lower than the single forecast model.

[1]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[2]  Jing Chang,et al.  Comparison of Random Forest and SVM for electrical short-term load forecast with different data sources , 2016, 2016 7th IEEE International Conference on Software Engineering and Service Science (ICSESS).

[3]  Bingyue Pan,et al.  Application of XGBoost algorithm in hourly PM2.5 concentration prediction , 2018 .

[4]  Wei-Yin Loh,et al.  Classification and regression trees , 2011, WIREs Data Mining Knowl. Discov..

[5]  Donghui Shi.  Application of outlier mining in power load forecasting , 2010, 2010 International Conference on Computer Application and System Modeling (ICCASM 2010).

[6]  Francisco Herrera,et al.  A Study on the Noise Label Influence in Boosting Algorithms: AdaBoost, GBM and XGBoost , 2017, HAIS.

[7]  R. Olshen,et al.  Points of Significance: Classification and regression trees , 2017, Nature Methods.

[8]  Hong Li,et al.  Short Term Load Forecasting by Adaptive Neural Network , 2018, IOP Conference Series: Materials Science and Engineering.

[9]  Juan Chen,et al.  A Study on Exponential Smoothing Model for Load Forecasting , 2012 .

[10]  J. Ben Hadj Slama,et al.  Day-ahead load forecast using random forest and expert input selection , 2015 .

[11]  Ponnuthurai N. Suganthan,et al.  Ensemble incremental learning Random Vector Functional Link network for short-term electric load forecasting , 2018, Knowl. Based Syst..

[12]  Junzo Watada,et al.  Short-Term Load Forecasting Using Optimized LSTM Networks Based on EMD , 2018, 2018 10th International Conference on Communications, Circuits and Systems (ICCCAS).

[13]  Marcin Luckner,et al.  Application of XGBoost Algorithm in Fingerprinting Localisation Task , 2017, CISIM.

[14]  Alex Graves,et al.  Supervised Sequence Labelling with Recurrent Neural Networks , 2012, Studies in Computational Intelligence.

[15]  Min Shen,et al.  Based on the ARIMA model with grey theory for short term load forecasting model , 2012, 2012 International Conference on Systems and Informatics (ICSAI2012).

[16]  Jie Gao,et al.  Research on Hot Micro-blog Forecast Based on XGBOOST and Random Forest , 2018, KSEM.

[17]  N. Kamel,et al.  Autoregressive method in short term load forecast , 2008, 2008 IEEE 2nd International Power and Energy Conference.

[18]  Ponnuthurai N. Suganthan,et al.  Ensemble Incremental Random Vector Functional Link Network for Short-term Crude Oil Price Forecasting , 2018, 2018 IEEE Symposium Series on Computational Intelligence (SSCI).

[19]  Sokratis Papadopoulos,et al.  Short-term electricity load forecasting using time series and ensemble learning methods , 2015, 2015 IEEE Power and Energy Conference at Illinois (PECI).

[20]  Wang Xiao.  Study on power system load prediction method for medium and long term based on comprehensive model , 2013 .

[21]  Guan Shi-lei.  Theoretical Study of Combination Model for Medium and Long Term Load Forecasting , 2010 .

[22]  Yuan Zhang,et al.  Short-Term Residential Load Forecasting Based on LSTM Recurrent Neural Network , 2019, IEEE Transactions on Smart Grid.

[23]  Jian Sun,et al.  HyperAdam: A Learnable Task-Adaptive Adam for Network Training , 2018, AAAI.

[24]  Wenbin Ma.  Power System Short-Term Load Forecasting Based on Improved Support Vector Machines , 2008, 2008 International Symposium on Knowledge Acquisition and Modeling.

[25]  Ke Wang,et al.  Study on the Supporting System for Smart Grid Control Center , 2011 .

[26]  Wenyin Liu,et al.  A Hybrid Model Combining Convolutional Neural Network with XGBoost for Predicting Social Media Popularity , 2017, ACM Multimedia.

[27]  Mianxiong Dong,et al.  Everything is Image: CNN-based Short-Term Electrical Load Forecasting for Smart Grid , 2017, 2017 14th International Symposium on Pervasive Systems, Algorithms and Networks & 2017 11th International Conference on Frontier of Computer Science and Technology & 2017 Third International Symposium of Creative Computing (ISPAN-FCST-ISCC).

[28]  Yasuhiro Hayashi,et al.  Long Term Load Forecasting using Improved Recurrent Neural Network , 1993 .

[29]  Donghui-Shi.  Application of outlier mining in power load forecasting , 2010 .