Short-Term Energy Demand Forecast in Hotels Using Hybrid Intelligent Modeling

Abstract

The hotel industry is an important energy consumer that needs efficient energy management methods to guarantee its performance and sustainability. The new role of hotels as prosumers increases the difficulty in the design of these methods. Also, the scenery is more complex as renewable energy systems are present in the hotel energy mix. The performance of energy management systems greatly depends on the use of reliable predictions for energy load. This paper presents a new methodology to predict energy load in a hotel based on intelligent techniques. The model proposed is based on a hybrid intelligent topology implemented with a combination of clustering techniques and intelligent regression methods (Artificial Neural Network and Support Vector Regression). The model includes its own energy demand information, occupancy rate, and temperature as inputs. The validation was done using real hotel data and compared with time-series models. Forecasts obtained were satisfactory, showing a promising potential for its use in energy management systems in hotel resorts.

Computational Intelligence in the hospitality industry: A systematic literature review and a prospect of challenges

“…Evolutionary computation is also being used in the area (in a lesser way, though) to solve different optimization problems like resource allocation and forecasting, which are just a few of the main difficulties inside the hospitality panorama. A good example of this is energy consumption, an area where hospitality establishments frequently struggle to adapt because of their need of efficient energy management methods, necessary to guarantee their performance and sustainability [23].…”

A K-means Group Division and LSTM Based Method for Hotel Demand Forecasting

“…In the process of group division in this paper, K-means algorithm is used to cluster hotels with similar features to form multiple groups. K-means is a widely used unsupervised algorithm for creating data sets [28]. For a given sample set, K-means divides the samples into K clusters according to the distance between the samples, making the sample points in each cluster as close as possible, and the distance between each cluster as large as possible.…”

Hybrid Intelligent Modelling in Renewable Energy Sources-Based Microgrid. A Variable Estimation of the Hydrogen Subsystem Oriented to the Energy Management Strategy

“…Despite this, the problem could persist, depending on the nature of the nonlinearities. When this occurs, one of the possible solutions, that gives very satisfactory results, is the breakdown of the problem into areas with similar and/or linear behaviour [38,39].…”