ENERGY CONSUMPTION PREDICTION BY USING MACHINE LEARNING FOR SMART BUILDING
Abstract : Building Energy Management System (BEMS) has been a substantial topic nowadays due to its importance in reducing energy wastage. However, the performance of one of BEMS applications which is energy consumption prediction has been stagnant due to problems such as low prediction accuracy. Thus, this research aims to address the problems by developing a predictive model for energy consumption in Microsoft Azure cloud-based machine learning platform. Three methodologies which are Support Vector Machine, Artificial Neural Network, and kNearest Neighbour are proposed for the algorithm of the predictive model. Focusing on real-life application in Malaysia, two tenants from a commercial building are taken as a case study. The data collected is analysed and pre-processed before it is used for model training and testing. The performance of each of the methods is compared based on RMSE, NRMSE, and MAPE metrics. The experimentation shows that each tenant’s energy consumption has different distribution characteristics.
• The prediction performance of the stacking model in contrast with existing well-known prediction models including Random Forest, Gradient Boosted Decision Tree, Extreme Gradient Boosting, Support Vector Machine, and KNearest Neighbor.
• Most of the existing research is based on single or integration models, which lack improvements in the essence of the algorithm.
• Some existing studies have confirmed the effectiveness of using machine learning algorithms to build meta-models of building performance in early building performance evaluation.
• Significant opportunities exist to take advantage of external data sources including real-time occupancy sensor networks, changing space schedules, weather forecasts, grid carbon intensity, and other environmental conditions that could help us better predict space set-points and schedules 24x7.
• Although the term “smart building” (SB) may bring a thought of a fictional smart space from science-fiction movies, but the reality is that SBs exist today, and their number is getting increased.
• It describes the different recently created strategies for the simplification of this problem, which deal with the engineering, statistical and artificial intelligence techniques.
• The ANN-based model proved to be more productive than the regression model, and thereby consumed fewer electrical units, as it deals with nonlinear problems in a better way.
• Many building structures’ energy utilization undertakings can be perceived as multi objective optimization problems by reasonably recognizing and confirming the multiple objectives.
• The measure of energy utilized in various regions is impacted by various factors, e.g., water resources, windand temperature.
• A deep learning predictive methodology with a long term transient memory intelligence model for the regulation of energy usage in urban buildings to decrease the information amount and coordinate the inter-building impact with the information-driven energy model.
• By integrating the machine learning-based approach with model optimization, we propose one architecture for better energy efficiency analysis and consumption in smart city public sector buildings.
• By integrating the data collection, data pre-processing, data analysis and generation with a deep neural network model, we propose an architecture for better energy efficiency and consumption in urban buildings of smart cities.
• Energy consumption dataset of 300 commercial buildings was used to assess the effectiveness of the proposed algorithm.
• In this paper we propose a new algorithm for selecting and fine - tuning the hyper parameters of the energy consumption model and tested it on the example of conglomerate of the commercial buildings.
• In this paper, we propose a method for estimating the consumption of electric power by large commercial centers and business buildings, based on the GBM gradient boosting algorithm with adaptive tuning of hyper parameters using the ??-fold cross-validation and randomization procedures.
• Some of these methods are also used along with predictors in the optimization of the energy consumption performance.
• The performance of an energy consumption predictive intelligence model is dependent on the sample dataset size that is used for the prediction, and the proper selection ofthe deployed model.
• A sensitivity analysis is required to quantify the association between the data size and the prediction model’s performance.
• The role of computational intelligent techniques in developing an intelligent framework for overseeing energy efficiency.
• It is hard to allocate a sufficient number of data points to accurately assess the predictive performance of the models without affecting the quality of the assessment.
• Intelligent energy saving and energy efficiency technologies are the modern large-scale global trend in the energy systems development.
• This led to the use of modern and efficient machine learning methods that provide promising opportunities to obtain more accurate forecasts of energy consumption of the building, and thus increase energy efficiency.
|