A Forecasting Method of District Heat Load Based on Improved Wavelet Neural Network

2020 ◽  
Vol 142 (10) ◽  
Author(s):  
Zhongbin Zhang ◽  
Ye Liu ◽  
Lihua Cao ◽  
Heyong Si
Keyword(s):  
Neural Networks ◽  
Energy Conservation ◽  
Heat Load ◽  
Prediction Models ◽  
District Heating ◽  
Heating System ◽  
Load Prediction ◽  
Urban District ◽  
District Heating System ◽  
Forecasting Method

Abstract Energy conservation of urban district heating system is an important part of social energy conservation. In response to the situation that the setting of heat load in the system is unreasonable, the heat load forecasting method is adopted to optimize the allocation of resources. At present, the artificial neural networks (ANNs) are generally used to forecast district heat load. In order to solve the problem that networks convergence is slow or even not converged due to the random initial parameters in traditional wavelet neural networks (WNNs), the genetic algorithm with fast convergence ability is used to optimize the network structure and initial parameters of heat load prediction models. The results show that when the improved WNN is applied to forecast district heat load, the prediction error is as low as 2.93%, and the accuracy of prediction results is improved significantly. At the same time, the stability and generalization ability of the prediction model are improved.

scholarly journals Data-Driven Machine-Learning Model in District Heating System for Heat Load Prediction: A Comparison Study

2016 ◽  
Vol 2016 ◽  
pp. 1-11 ◽  
Author(s):  
Fisnik Dalipi ◽  
Sule Yildirim Yayilgan ◽  
Alemayehu Gebremedhin
Keyword(s):  
Machine Learning ◽  
Heat Load ◽  
District Heating ◽  
Heating System ◽  
Partial Least Square ◽  
Supervised Machine Learning ◽  
Data Driven ◽  
Support Vector ◽  
Load Prediction ◽  
District Heating System

We present our data-driven supervised machine-learning (ML) model to predict heat load for buildings in a district heating system (DHS). Even though ML has been used as an approach to heat load prediction in literature, it is hard to select an approach that will qualify as a solution for our case as existing solutions are quite problem specific. For that reason, we compared and evaluated three ML algorithms within a framework on operational data from a DH system in order to generate the required prediction model. The algorithms examined are Support Vector Regression (SVR), Partial Least Square (PLS), and random forest (RF). We use the data collected from buildings at several locations for a period of 29 weeks. Concerning the accuracy of predicting the heat load, we evaluate the performance of the proposed algorithms using mean absolute error (MAE), mean absolute percentage error (MAPE), and correlation coefficient. In order to determine which algorithm had the best accuracy, we conducted performance comparison among these ML algorithms. The comparison of the algorithms indicates that, for DH heat load prediction, SVR method presented in this paper is the most efficient one out of the three also compared to other methods found in the literature.

Applied machine learning: Forecasting heat load in district heating system

2016 ◽  
Vol 133 ◽  
pp. 478-488 ◽  
Author(s):  
Samuel Idowu ◽  
Saguna Saguna ◽  
Christer Åhlund ◽  
Olov Schelén
Keyword(s):  
Machine Learning ◽  
Heat Load ◽  
District Heating ◽  
Heating System ◽  
District Heating System ◽  
Applied Machine Learning

Physical-Rule Based Adaptive Neuro-Fuzzy Inferential Sensor vs. GA-BP Based Prediction Model in Indoor Temperature Predicting

2012 ◽  
Vol 594-597 ◽  
pp. 2179-2185
Author(s):  
Liang Huang ◽  
Zai Yi Liao
Keyword(s):  
Prediction Models ◽  
Back Propagation ◽  
District Heating ◽  
Heating System ◽  
Good Prediction ◽  
Indoor Temperature ◽  
District Heating System ◽  
Rule Based ◽  
Neuro Fuzzy ◽  
Indoor Comfort

The previous research on temperature prediction presented different approaches which are physical-rule based adaptive neuro-fuzzy inferential sensor (ANFIS) model and GA-BP (genetic algorithm back propagation) based model to estimate the average indoor temperature in the building environment. Their good prediction performances improved energy efficiency of district heating system and indoor comfort ratio. However, either of these two models has its drawback in a certain condition. In this paper, the two prediction models are reviewed and evaluated by three performance measures (RMSE, RMS, and R2). Their limitations are discussed and potential solution is proposed.

scholarly journals Stochastic Generation of District Heat Load

Energies ◽  
2021 ◽  
Vol 14 (17) ◽  
pp. 5344
Author(s):  
Andrea Menapace ◽  
Simone Santopietro ◽  
Rudy Gargano ◽  
Maurizio Righetti
Keyword(s):  
Heat Load ◽  
Time Series Data ◽  
Empirical Relationship ◽  
District Heating ◽  
Heating System ◽  
Series Data ◽  
New Paradigm ◽  
Flexible Tool ◽  
District Heating System ◽  
Demand Variation

Modelling heat load is a crucial challenge for the proper management of heat production and distribution. Several studies have tackled this issue at building and urban levels, however, the current scale of interest is shifting to the district level due to the new paradigm of the smart system. This study presents a stochastic procedure to model district heat load with a different number of buildings aggregation. The proposed method is based on a superimposition approach by analysing the seasonal component using a linear regression model on the outdoor temperature and the intra-daily component through a bi-parametric distribution of different times of the day. Moreover, an empirical relationship, that estimates the demand variation given the average demand together with a user aggregation coefficient, is proposed. To assess the effectiveness of the proposed methodology, the study of a group of residential users connected to the district heating system of Bozen-Bolzano is carried out. In addition, an application on a three-day prevision shows the suitability of this approach. The final purpose is to provide a flexible tool for district heat load characterisation and prevision based on a sample of time series data and summary information about the buildings belonging to the analysed district.

scholarly journals Heat load prediction of small district heating system using artificial neural networks

2016 ◽  
Vol 20 (suppl. 5) ◽  
pp. 1355-1365 ◽  
Author(s):  
Milos Simonovic ◽  
Vlastimir Nikolic ◽  
Emina Petrovic ◽  
Ivan Ciric
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Artificial Neural Network Model ◽  
District Heating ◽  
Heating System ◽  
Load Prediction ◽  
District Heating System ◽  
Artificial Neural

Accurate models for heat load prediction are essential to the operation and planning of a utility company. Load prediction helps a heat utility to make important and advanced decisions in district heating systems. As a popular data driven method, artificial neural networks are often used for prediction. The main idea is to achieve quality prediction for a short period in order to reduce the consumption of heat energy production and increased coefficient of exploitation of equipment. To improve the short term prediction accuracy, this paper presents a kind of improved artificial neural network model for 1 to 7 days ahead prediction of heat consumption of energy produced in small district heating system. Historical data set of one small district heating system from city of Nis, Serbia, was used. Particle swarm optimization is applied to adjust artificial neural network weights and threshold values. In this paper, application of feed forward artificial neural network for short-term prediction for period of 1, 3, and 7 days, of small district heating system, is presented. Two test data sets were considered with different interruption non-stationary performances. Comparison of prediction accuracy between regular and improved artificial neural network model was done. The comparison results reveal that improved artificial neural network model have better accuracy than that of artificial neural network ones.

scholarly journals Heat Load Numerical Prediction for District Heating System Operational Control

2021 ◽  
Vol 58 (3) ◽  
pp. 121-136
Author(s):  
D. Rusovs ◽  
L. Jakovleva ◽  
V. Zentins ◽  
K. Baltputnis
Keyword(s):  
Ambient Temperature ◽  
Heat Load ◽  
District Heating ◽  
Heating System ◽  
Training Data ◽  
Mean Bias Error ◽  
Percentage Error ◽  
Data Set ◽  
District Heating System ◽  
Heating Systems

Abstract To develop an advanced control of thermal energy supply for domestic heating, a number of new challenges need to be solved, such as the emerging need to plan operation in accordance with an energy market-based environment. However, to move towards this goal, it is necessary to develop forecasting tools for short- and long-term planning, taking into account data about the operation of existing heating systems. The paper considers the real operational parameters of five different heating networks in Latvia over a period of five years. The application of regression analysis for heating load dependency on ambient temperature results in the formulation of normalized slope for the regression curves of the studied systems. The value of this parameter, the normalized slope, allows describing the performance of particular heating systems. Moreover, a heat load forecasting approach is presented by an application of multiple regression methods. This short-term (day-ahead) forecasting tool is tested on data from a relatively small district heating system with an average load of 20 MW at ambient temperature of 0 °C. The deviations of the actual heat load demand from the one forecasted with various training data set sizes and polynomial orders are evaluated for two testing periods in January of 2018. Forecast accuracy is assessed by two parameters – mean absolute percentage error and normalized mean bias error.

Sign in / Sign up

Export Citation Format

Share Document