Neural network, ARX, and extreme learning machine models for the short-term prediction of temperature in buildings

Abstract

In this paper, the possibilities of developing machine learning based data-driven models for the short-term prediction of indoor temperature within prediction horizons ranging from 1 hour up to 12 hours are systematically investigated. The study was based on a TRNSYS emulation of a residential building heated by a heat pump, combined with measured weather data for a typical winter season in Ljubljana, Slovenia. Autoregressive models with exogenous inputs (ARX), neural network models (NN), and extreme learning machine models (ELM) are considered. The results confirm the finding that nonlinear models, particularly the NN model trained by regularization, consistently outperform linear models in both fitting and generalization performance, so they are the recommended choice as predictive models. The availability of future weather data considerably improved the predictive performance of all the tested models. Besides data about the future outdoor temperature, also data about future expected solar radiation significantly improve predictions of temperature in buildings. The linear models required embedding dimensions of 24 hours for accurate predictions, whereas the nonlinear models were not very sensitive to the use of past data. Nonlinear models required about three months of training data to reach good predictive performance, whereas the linear models converged to accurate predictions within six weeks. The RMSE prediction errors, averaged over all the data sets and all the prediction horizons, are within the range between 0.155°C for the linear ARX model (in the case of no future available weather data), and 0.065°C for the neural network model (in the case of available future weather data).

Long Term Solar Power Generation Prediction using Adaboost as a Hybrid of Linear and Non-linear Machine Learning Model

“…The main purpose of Adaboost is to ensemble machine learning models and improves the accuracy and precision. Adaboost has been extensively adopted in the recent years [36].In the mechanism of this model, weights are generated at each step of algorithm for better classification and prediction. In this paper, outputs taken from RNN, SVM and ARX are fed to Adaboost to form a strong classifier as shown in Fig.…”

Exploring the Predictability of Temperatures in a Scaled Model of a Smarthome

“…The proposed approaches are also evaluated on the SLM house dataset. Potočnik, Primož et al [33] are also predicting the short-term variations of temperature of buildings using neural networks, autoregressive models with exogenous inputs and extreme learning machines. The home considered in this case is a residential building heated by a heat pump in a typical winter in Slovenia.…”

A Survey of Research Progress and Hot Front of Natural Gas Load Forecasting From Technical Perspective

“…The main methods used are neural network model, support vector machine and linear regression model. In addition, the team also has some research achievements in building temperature prediction [14], heat energy prediction [15] and highway traffic flow prediction [16], which can be used for reference in natural gas load forecasting.…”