**1. Introduction**

The constant advancement of ICT opens up grea<sup>t</sup> opportunities to improve systems' functionality, performance, and efficiency. Technologies such as IoT, Big Data, AI, WIFI 6, or 5G could come together to enhance the capabilities of equally emerging systems, oriented for use in the home, such as HEMS and Voice Assistants.

The residential sector is a key element in the context of both energy savings and people's well-being. The restructuring of the energy sector through Smart Grids and Microgrids [1], as well as the arrival of the 5G network [2–4] and WIFI 6, will allow an exponential development of connected devices and appliances, opening the doors of the network to the IoT both in the industrial sector and in the home [5].

Bringing all these elements together in the home environment can be challenging, but they form the fundamental structure of the Home Energy Management Expert Assistant (HERMES) system. However, it can offer us a revolution that can positively impact climate change, energy efficiency, or quality of life. Each of these elements separately already offers solutions to specific problems. The literature shows considerable evidence of this use and its benefits, as shown below, focusing on the most relevant, emphasizing the latest efforts and advances in applying the methodologies.

### *Home Energy Management Systems (HEMS)*

HEMS are hardware and software systems that enable advanced control of energyusing systems and devices in the home, continuously analyzing data to provide real-time information on the energy performance of the home [6], creating data streams (both external and internal such as weather, electricity price, or sensors) and making decisions

**Citation:** Linan-Reyes, M.; Garrido-Zafra, J.; Gil-de-Castro, A.; Moreno-Munoz, A. Energy Management Expert Assistant, a New Concept. *Sensors* **2021**, *21*, 5915. https://doi.org/10.3390/s21175915

Academic Editor: Geoff Merrett

Received: 27 July 2021 Accepted: 27 August 2021 Published: 2 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

for energy efficiency improvement [7], peak demand managemen<sup>t</sup> and demand response, so their specifications include the necessary integration (monitoring and control) and communication with all smart home devices, sensors, relays, and appliances regardless of their communication protocols [8]. A proper implementation would allow a reduction of about 35% of the total electricity bill, prioritizing load consumption based on the cost of energy [9]. Other studies, such as the ACEEE study not focused exclusively on energy cost, set this saving at a maximum of 17% [7].

Nevertheless, in practice, these systems have some limitations, mainly including interoperability between devices, lack of training of the users themselves, doubts about security or limitations of commitment to the customer, as well as the lack of studies showing the real possibilities of savings [5,7,10,11], as well as lacking true intelligence and the ability to manage demand peaks and demand response. To develop an efficient HEMS, it is necessary to know the characteristics and requirements of each of the technologies that will allow a complete communication and configuration of all the devices [12–15]. One solution to interoperability would be implementing a widely consolidated Building Management System (BMS) [16]. However, they are very closed systems, mostly proprietary solutions that can only be upgraded by the system manufacturer with relatively high costs, ranging from \$25 to \$70 per square meter of housing [17]. On the market, we can find a wide offer of both Open-Source and Proprietary HEMS [8,18]: Open-Source Home Assistant [19] and OpenHAB (based on Eclipse SmartHome ™) [20,21] that have a large number of protocols and configurable devices, BEMOSS [22] built-in Python on VOLTTRON [23], ioBroker [24], Open Energy Management (OGEMA) [25,26], and Open remote [27], among others.

In this paper, we will consider the following architecture of an advanced HEMS system (Figure 1):

We continue in the next section by detailing the main elements that will make up our HERMES system.

### **2. Materials and Methods**

In addition to the energy managemen<sup>t</sup> system described in the introduction, the HERMES system is developed integrating the following elements:

### *2.1. IoT: Smart Devices*

Depending on the context, there have been many definitions of IoT [28–31]. A good approximation to its current definition could be: "System of devices, machines, or everyday objects provided with unique identifiers (UID) with the ability to transfer data through a network without the need for interaction between people or between people and computers".

In recent years the number of Internet-connected devices has exploded, to such an extent that forecasts have become outdated [32]. There is currently no consensus on their number, offering figures that range from CISCO's 50 billion [33] to Intel's 200 billion [34]. In parallel to this growth, their price has been reduced by more than 90%, coining the concept of "democracy of devices" [35].

Not only devices or things but also household appliances are beginning to join the world of elements connected to the network, and its growth will be almost total in the coming years with the arrival of 5G and the IPv6 protocol without address exhaustion problems (2<sup>128</sup> addresses).

Therefore, IoT is a significant challenge for HEMS as it has to interact with a large number of "smart devices" with a wide variety of protocols, in addition to the large amount of data they will generate. In this regard, some emerging technologies can help address this challenge: big data, cloud computing, and AI, as also postulated in [17,36].
