Multi-Agent Recommendation System for Electrical Energy Optimization and Cost Saving in Smart Homes

Abstract

The European Union Establishes that for the next few years, a cleaner and more efficient energy system should be used. In order to achieve this, this work proposes an energy optimization method that facilitates the achievement of these objectives. Existing technologies allow us to create a system that optimizes the use of energy in homes and offers some type of benefit to its residents. Specifically, this study has developed a recommendation system based on a multiagent system that allows to obtain consumption data from electronic devices in a home, obtain information on electricity prices from the Internet, and provide recommendations based on consumption patterns of users and electricity prices. In this way, the system recommends new hours in which to use the appliances, offering the economic benefit that it would propose recommendations for the user. In this way, it is possible to distribute and optimize the use of energy in homes and reduce the peaks in electricity consumption. The system provides encouraging results in order to resolve the problem proposed by the European Union by optimizing the use of energy among different hours of the day and saving money for the customer.

The emergence of explainability of intelligent systems: Delivering explainable and personalized recommendations for energy efficiency

“…The individual consumption habits and preferences are evaluated against a set of rules and predefined decisions, which trigger—when met—the corresponding energy saving actions. Authors in Reference [10] implement a multiagent system, which enables to (i) collect power usage patterns from electrical appliances in domestic buildings; (ii) procure electricity price data from Internet; (iii) trigger appropriate recommendations for end‐users using consumption footprints electricity prices. To this end, the developed recommendation system furnishes information about the hours to use domestic devices, offering an economic benefit to end‐users.…”

Data fusion strategies for energy efficiency in buildings: Overview, challenges and novel orientations

“…Different sorts of recommender systems are proposed in the literature using the input data either to simply select actions of possible interests to the target consumer or to predict the consumer interest level for specific actions and then produce appropriate recommendations. Consequently, several recommendation engines are proposed such as collaborative filtering [103,6], context-aware recommendations [104], content-based recommendations [105,106] and multi-agent recommendations [107,108].…”

The emergence of Explainability of Intelligent Systems: Delivering Explainable and Personalised Recommendations for Energy Efficiency

“…The individual consumption habits and preferences are evaluated against a set of rules and predefined decisions, which trigger -when met -the corresponding energy saving actions. Authors in [10] implement a multi-agent system, which enables to (i) collect power usage patterns from electrical appliances in domestic buildings; (ii) procure electricity price data from Internet; (iii) trigger appropriate recommendations for end-users using consumption footprints electricity prices. To this end, the developed recommendation system furnishes information about the hours to use domestic devices, offering an economic benefit to end-users.…”