A fully autonomous kernel-based online learning neural network model and its application to building cooling load prediction

An improved error transfer BP neural network model is use to predict the dynamic heating load in a house or a dwelling unit with the character of hour heating load. Compared with the conventionally physical model, the computation consumption is reduced greatly for the less number of the parameters by improving the error transfer ways. The numerical simulation and experimental measure in a low energy consumption building of Dalian city are performed and the BP neural network model was based entirely on the field survey data. The results show that the simulated results are well agreed with the experimental data and the averaged relative error is less than 5%. Furthermore, this improved model can predict accurately hour heating load during the course of next 24 hours and it is favorable for predicting the short time heating load problems.

Download Full-text

Short-Term Load Prediction Based on Ant Colony Clustering-Elman Neural Network Model

2009 Second International Workshop on Computer Science and Engineering ◽

10.1109/wcse.2009.695 ◽

2009 ◽

Cited By ~ 2

Author(s):

Dong-xing Duan

Keyword(s):

Neural Network ◽

Network Model ◽

Neural Network Model ◽

Ant Colony ◽

Load Prediction ◽

Short Term ◽

Elman Neural Network ◽

Ant Colony Clustering

Download Full-text

A novel clustering-enhanced adaptive artificial neural network model for predicting day-ahead building cooling demand

Journal of Building Engineering ◽

10.1016/j.jobe.2020.101504 ◽

2020 ◽

Vol 32 ◽

pp. 101504 ◽

Cited By ~ 4

Author(s):

X.J. Luo

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Network Model ◽

Neural Network Model ◽

Artificial Neural Network Model ◽

Artificial Neural ◽

Building Cooling

Download Full-text

Cooling Load Forecasting via Predictive Optimization of a Nonlinear Autoregressive Exogenous (NARX) Neural Network Model

Sustainability ◽

10.3390/su11236535 ◽

2019 ◽

Vol 11 (23) ◽

pp. 6535 ◽

Cited By ~ 1

Author(s):

Kim ◽

Seong ◽

Choi

Keyword(s):

Neural Network ◽

Network Model ◽

Neural Network Model ◽

Missing Values ◽

Training Data ◽

Model Parameters ◽

Cooling Load ◽

Cooling Loads ◽

Nonlinear Autoregressive ◽

Narx Neural Network

Accurate calculations and predictions of heating and cooling loads in buildings play an important role in the development and implementation of building energy management plans. This study aims to improve the forecasting accuracy of cooling load predictions using an optimized nonlinear autoregressive exogenous (NARX) neural network model. The preprocessing of training data and optimization of parameters were investigated for model optimization. In predictive models of cooling loads, the removal of missing values and the adjustment of structural parameters have been shown to help improve the predictive performance of a neural network model. In this study, preprocessing the training data eliminated missing values for times when the heating, ventilation, and air-conditioning system is not running. Also, the structural and learning parameters were adjusted to optimize the model parameters.

Download Full-text

Attention-based interpretable neural network for building cooling load prediction

Applied Energy ◽

10.1016/j.apenergy.2021.117238 ◽

2021 ◽

Vol 299 ◽

pp. 117238

Author(s):

Ao Li ◽

Fu Xiao ◽

Chong Zhang ◽

Cheng Fan

Keyword(s):

Neural Network ◽

Cooling Load ◽

Load Prediction ◽

Building Cooling

Download Full-text

Predicting the determinants of online learning adoption during the COVID-19 outbreak: a two-staged hybrid SEM-neural network approach

Interactive Technology and Smart Education ◽

10.1108/itse-08-2020-0165 ◽

2021 ◽

Vol ahead-of-print (ahead-of-print) ◽

Author(s):

Nattaporn Thongsri ◽

Chalothon Chootong ◽

Orawan Tripak ◽

Piyaporn Piyawanitsatian ◽

Rungtip Saengae

Keyword(s):

Neural Network ◽

Higher Education ◽

Online Learning ◽

Data Analysis ◽

Empirical Study ◽

Network Model ◽

Neural Network Model ◽

Learning Context ◽

Content Type ◽

Study Results

Purpose This study aims to study the adoption of online learning in higher education through the perspective of the readiness of the following factors: self-directed learning (SDL), motivation for learning (ML), online communication self-efficacy (OCE) and learner control (LC). This was an empirical study in the context of developing countries, specifically Thailand. Design/methodology/approach This research applied a quantitative study method by collecting data from 605 higher education students in autonomous government institutions. The data analysis applied a structural equation model (SEM) to identify the significant determinants that affected the adoption of online learning. Moreover, this study applied a neural network model to examine the findings from the SEM. Findings From the data analysis using the SEM and neural network model, the results matched each other. The results of the empirical study were firm and supported that the readiness factors of students had statistical significance in the following order: SDL, OCE, LC and ML. Practical implications The study results showed an operational perspective to be prepared for online teaching, both for the related department of the Ministry of Education to support the infrastructure for online learning and for universities and instructors to create learning conditions and design teaching processes consistently with the online learning context. Originality/value Since the learning management in the 21st century is focused on student-centred learning, the empirical results obtained from this study presented the view of learners’ readiness that would influence the acceptance of online learning. In addition, this research presented the challenges and opportunities of online instruction during the COVID-19 pandemic.

Download Full-text

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

Thermal Science ◽

10.2298/tsci16s5355s ◽

2016 ◽

Vol 20 (suppl. 5) ◽

pp. 1355-1365 ◽

Cited By ~ 5

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.

Download Full-text

A Comparison of NARX and BP Neural Network in Short-Term Building Cooling Load Prediction

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.513-517.1545 ◽

2014 ◽

Vol 513-517 ◽

pp. 1545-1548 ◽

Cited By ~ 3

Author(s):

Yan Li Xu ◽

Hong Xun Chen ◽

Wang Guo ◽

Qiu Yu Zhu

Keyword(s):

Neural Network ◽

Bp Neural Network ◽

Prediction Models ◽

Back Propagation ◽

Prediction Errors ◽

Cooling Load ◽

Climate Data ◽

Load Prediction ◽

Short Term ◽

Building Cooling

A comparison of nonlinear autoregression with exogenous inputs (NARX) neural network and back-propagation (BP) neural network in short-term prediction of building cooling load is presented in this dissertation. Both predictive models have been applied in a group of commercial buildings and analysis of prediction errors has been highlighted. Training and testing data for both prediction models have been generated from DeST (Designers Simulation Toolkits) with climate data of Shanghai. The simulation results indicate that NARX method can achieve better accuracy and generalization ability than traditional method of BP neural network. This work provides a key support in smooth and optimizing control in air-conditioning system.

Download Full-text