LSTM Model for Various Types of Load Forecasting in Energy System Integration

Increasing energy efficiency and reducing pollution to the environment, a comprehensive energy system is of great significance for building an energy internet and promoting the transformation of the energy structure. Energy demand forecasting plays an important role in optimizing comprehensive energy system planning, and its accuracy is of great significance for maintaining stability and economic operation of the system. A load forecasting method based on Long Short-Term Memory Network Model (LSTM) is proposed in this paper. Firstly, calculating the mutual information load value of the past 1 hour time to the past 168 hour times, and the time to be predicted, and then using the maximum correlation and minimum redundancy to filter the input variables. Finally, on the basis of selecting the optimal input variable set, the multi-element load forecasting model of integrated energy system based on long-short-term memory neural network (LSTM) is established to realize the load data forecasting of the regional integrated energy system.

[1]  Demis Hassabis,et al.  Mastering the game of Go with deep neural networks and tree search , 2016, Nature.

[2]  Tao Hong,et al.  A Naïve multiple linear regression benchmark for short term load forecasting , 2011, 2011 IEEE Power and Energy Society General Meeting.

[3]  Atsushi Nakamura,et al.  Efficient WFST-Based One-Pass Decoding With On-The-Fly Hypothesis Rescoring in Extremely Large Vocabulary Continuous Speech Recognition , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[4]  Chia-Nan Ko,et al.  Short-term load forecasting using SVR (support vector regression)-based radial basis function neural network with dual extended Kalman filter , 2013 .

[5]  C.S. Ozveren,et al.  Short term load forecasting using Multiple Linear Regression , 2007, 2007 42nd International Universities Power Engineering Conference.

[6]  Guoqiang Sun,et al.  Application of functional deep belief network for estimating daily global solar radiation: A case study in China , 2020 .

[7]  Priyanka Singh,et al.  Integration of new evolutionary approach with artificial neural network for solving short term load forecast problem , 2018 .

[8]  Christian Szegedy,et al.  DeepPose: Human Pose Estimation via Deep Neural Networks , 2013, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Dong Yu,et al.  Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition , 2012, IEEE Transactions on Audio, Speech, and Language Processing.

[10]  Abbas Khosravi,et al.  Short-Term Load and Wind Power Forecasting Using Neural Network-Based Prediction Intervals , 2014, IEEE Transactions on Neural Networks and Learning Systems.

[11]  Tao Ding,et al.  Estimation and validation of daily global solar radiation by day of the year-based models for different climates in China , 2019, Renewable Energy.

[12]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[13]  Qia Ding,et al.  Long-Term Load Forecast Using Decision Tree Method , 2006, 2006 IEEE PES Power Systems Conference and Exposition.

[14]  Tao Ding,et al.  Day-ahead photovoltaic power forecasting approach based on deep convolutional neural networks and meta learning , 2020, International Journal of Electrical Power & Energy Systems.

[15]  Zhang Jianhua,et al.  A Short-Term Load Forecasting Method for Micro-Grid Based on EMD-KELM-EKF and Parameter Optimization , 2014 .