Hourly Heat Load Prediction Model Based on Temporal Convolutional Neural Network

Jiancai Song,Yunpeng Ma,Xuhua Pan,Han Li,Guixiang Xue

doi:10.1109/access.2020.2968536

Abstract

Smart district heating system (SDHS) is an important way to realize green energy saving and comfortable heating in the future, which is conducive to improving energy utilization efficiency and reducing pollution emissions. The accurate prediction algorithm of heating load plays an important role in on-demand heat supply, however, the heating load prediction is a complicated nonlinear optimization problem, and the prediction accuracy is limited due to the poor nonlinear expression ability of the traditional prediction algorithms. This paper proposes a heating load prediction model based on temporal convolutional neural network (TCN), which implements the rapid extraction of complex data features due to the integration of both the parallel feature processing of convolution neural network (CNN) and the time-domain modeling capability of recurrent neural network (RNN). The engineering data of four heat exchange stations located in Anyang, China in the 2018 heating season is used to evaluate and verify the performance of proposed prediction algorithm based on TCN, and the comprehensive comparisons with state-of-the-art algorithms, such as RFR, ETR, GBR, SVR, NuSVR, SGD, Bagging, Boosting, MLP, RNN, LSTM, etc., were analyzed carefully. The experimental results shown that the proposed heat load prediction algorithm based on TCN has performance superiority.

Highlights

Over the past two decades, China’s energy consumption has grown very rapidly, which brings a heavy burden on environmental protection [1]
The deep learning algorithm is introduced and applied into district heating systems (DHS), and a heating load prediction algorithm based on temporal convolutional neural network (TCN) is proposed
(2) The proposed heat load prediction algorithm based on TCN considers the operating parameters of the heat exchange station and meteorological parameters as the features, which is helpful to improving the complex features expressiveness by means of convolution network

Summary

INTRODUCTION

Over the past two decades, China’s energy consumption has grown very rapidly, which brings a heavy burden on environmental protection [1]. Three different prediction models based on machine learning are evaluated for forecasting the heat load of the large DHS [26]. Lecun et al [31] proposed the deep neural network structure, which can be layered to extract complex features of different depths, so that the deep neural network can make more accurate prediction It can solve the problem of insufficient nonlinear expression ability of shallow network, and deep neural network represented by deep learning algorithm has made great progress in image processing [32], [33], speech [34], [35] and natural language process [36]. The deep learning algorithm is introduced and applied into DHS, and a heating load prediction algorithm based on TCN is proposed. (2) The proposed heat load prediction algorithm based on TCN considers the operating parameters of the heat exchange station and meteorological parameters as the features, which is helpful to improving the complex features expressiveness by means of convolution network.

FEATURE REPRESENTATION

DATA PREPROCESSING

EVALUATION CRITERIA

Findings

CONCLUSION