Heating load prediction based on attention long short term memory: A case study of Xingtai

Abstract An accurate heating load prediction algorithm can play an important role in smart district heating systems (SDHS), which is helpful for realizing on-demand heating and fine control. However, most of the traditional heating load prediction algorithms neglect the indoor temperature feedback from the household and cannot form closed-loop control. This paper designs an intelligent sensor based on the Narrow band Internet of Thing (NB-IoT) to collect the indoor temperature of a typical household and proposes an algorithm based on attention long short term memory (ALSTM) to predict the heating load for an integrated "heat exchange station - heat user". The attention mechanism is designed to obtain more accurate nonlinear prediction models between the heating load and influencing factors, such as indoor temperature, outdoor temperature, and historical heat consumption. A performance comparison with other state-of-the-art algorithms shows that the proposed ALSTM algorithm has the best performance, achieving an accuracy of 97.9%. Besides, a Kalman filter is introduced to identify and remove outliers while reducing the random error of the measurement.

[1]  Shahaboddin Shamshirband,et al.  Forecasting of consumers heat load in district heating systems using the support vector machine with a discrete wavelet transform algorithm , 2015 .

[2]  Michael C. Georgiadis,et al.  An integrated stochastic multi-regional long-term energy planning model incorporating autonomous power systems and demand response , 2015 .

[3]  Long Chen,et al.  Short-Term Load Forecasting Using EMD-LSTM Neural Networks with a Xgboost Algorithm for Feature Importance Evaluation , 2017 .

[4]  Yeong-Koo Yeo,et al.  Optimization of district heating systems based on the demand forecast in the capital region , 2009 .

[5]  Peder Bacher,et al.  Online short-term forecast of greenhouse heat load using a weather forecast service , 2017 .

[6]  Sam Kwong,et al.  Deep sequential fusion LSTM network for image description , 2018, Neurocomputing.

[7]  Gabriele Comodi,et al.  The role of data sample size and dimensionality in neural network based forecasting of building heating related variables , 2016 .

[8]  Abdul Hanan Abdullah,et al.  Heat load prediction in district heating systems with adaptive neuro-fuzzy method , 2015 .

[9]  Yinhua Liu,et al.  Deep self-taught learning for facial beauty prediction , 2014, Neurocomputing.

[10]  Na Deng,et al.  Multi-criteria comprehensive study on predictive algorithm of heating energy consumption of district heating station based on timeseries processing , 2020 .

[11]  Shahaboddin Shamshirband,et al.  Appraisal of the support vector machine to forecast residential heating demand for the District Heating System based on the monthly overall natural gas consumption , 2015 .

[12]  Jean-Luc Gauvain,et al.  Optimization of RNN-Based Speech Activity Detection , 2018, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[13]  P. Brum,et al.  Lack of α2C-Adrenoceptor Results in Contrasting Phenotypes of Long Bones and Vertebra and Prevents the Thyrotoxicosis-Induced Osteopenia , 2016, PloS one.

[14]  Fu Xiao,et al.  A short-term building cooling load prediction method using deep learning algorithms , 2017 .

[15]  Edvard Govekar,et al.  Linear and Neural Network-based Models for Short-Term Heat Load Forecasting , 2015 .

[16]  Qinghua Hu,et al.  Transfer learning for short-term wind speed prediction with deep neural networks , 2016 .

[17]  Shahaboddin Shamshirband,et al.  Intelligent forecasting of residential heating demand for the District Heating System based on the monthly overall natural gas consumption , 2015 .

[18]  H. Madsen,et al.  Short-term heat load forecasting for single family houses , 2013 .

[19]  Hwataik Han,et al.  Simplified dynamic neural network model to predict heating load of a building using Taguchi method , 2016 .

[20]  T Ivan Ciric,et al.  Heat load prediction of small district heating system using artificial neural networks , 2016 .

[21]  Risto Lahdelma,et al.  Evaluation of a multiple linear regression model and SARIMA model in forecasting heat demand for district heating system , 2016 .

[22]  Robert S. McLeod,et al.  Forecasting indoor temperatures during heatwaves using time series models , 2018, Building and Environment.

[23]  Yeong-Koo Yeo,et al.  Heat consumption forecasting using partial least squares, artificial neural network and support vector regression techniques in district heating systems , 2010 .

[24]  Hermann Ney,et al.  From Feedforward to Recurrent LSTM Neural Networks for Language Modeling , 2015, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[25]  Magnus Dahl,et al.  Using ensemble weather predictions in district heating operation and load forecasting , 2017 .

[26]  Weiqiang Zhang,et al.  RNN language model with word clustering and class-based output layer , 2013, EURASIP J. Audio Speech Music. Process..

[27]  Shouxiang Wang,et al.  Multi-energy load forecasting for regional integrated energy systems considering temporal dynamic and coupling characteristics , 2020 .

[28]  Myungjong Kim,et al.  Speaker-Independent Silent Speech Recognition From Flesh-Point Articulatory Movements Using an LSTM Neural Network , 2017, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[29]  Shahaboddin Shamshirband,et al.  Extreme learning machine for prediction of heat load in district heating systems , 2016 .

[30]  Frank K. Soong,et al.  Effective Spectral and Excitation Modeling Techniques for LSTM-RNN-Based Speech Synthesis Systems , 2017, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[31]  M. P. van den Heuvel,et al.  Deep learning predictions of survival based on MRI in amyotrophic lateral sclerosis , 2016, NeuroImage: Clinical.

[32]  Joyce Kim,et al.  Personal comfort models: Predicting individuals' thermal preference using occupant heating and cooling behavior and machine learning , 2018 .

[33]  J. Gonzalez-Dominguez,et al.  Language Identification in Short Utterances Using Long Short-Term Memory (LSTM) Recurrent Neural Networks , 2016, PloS one.

[34]  Giuseppina Ciulla,et al.  Annual heating energy requirements of office buildings in a European climate , 2016 .

[35]  Shahaboddin Shamshirband,et al.  Prediction of heat load in district heating systems by Support Vector Machine with Firefly searching algorithm , 2016 .

[36]  Saguna Saguna,et al.  Applied machine learning: Forecasting heat load in district heating system , 2016 .

[37]  Mingming Gao,et al.  Dynamic modeling for NOx emission sequence prediction of SCR system outlet based on sequence to sequence long short-term memory network , 2020 .

[38]  M. Greenstone,et al.  Evidence on the impact of sustained exposure to air pollution on life expectancy from China’s Huai River policy , 2013, Proceedings of the National Academy of Sciences.

[39]  Kody M. Powell,et al.  Heating, cooling, and electrical load forecasting for a large-scale district energy system , 2014 .

[40]  Trond Andresen,et al.  Dynamic optimization of control setpoints for an integrated heating and cooling system with thermal energy storages , 2020 .