A colossal amount of energy is wasted every year due to the inappropriate management of energy sources and household appliances within smart buildings (SBs) and smart homes (SHs), which constitute ever-evolving smart communities. In this direction, Wireless Sensor Networks (WSNs) are becoming irresponsive at handling the humongous and heterogeneous data being generated from a gigantic number of sensors connected via the Internet of Things (IoT). According to a report published by Security Today, 127 new IoT devices are added to the Internet every second, and the estimated number of these devices will rise to 75 billion devices by 2025. Thus, the penetration of more connected devices is inevitable shortly, and increasing their number implies increasing the possibilities of link failures and communications–which in turn exaggerate the energy consumption manifold. Thus, the energy demand of the connected IoT devices is projected to extensively increase in the coming years. To meet this growing concern, smart harvesting capabilities based on human motion, heat systems, and solar power is the need of the hour. One of the major sources of energy harvesting is solar radiation. However, their capabilities vary over the entire day and are dependent upon the component’s aging and weather conditions. Thus, these variations cause uncertainties in effectively managing such systems. Fortunately, one of the promising candidates in this regard is reinforcement learning (RL), an area of machine learning (ML). RL can accurately handle such variations and can also be used to schedule sensor nodes according to the amount of available energy from renewable energy sources.

The energy management strategies for smart communities based on ML techniques are already studied extensively. For instance, the modern smart grid architectures are now available with ML and intelligent support. However, the energy management of SHs/SBs/smart communities still needs conversion from traditional electrical systems to intelligent techniques. In this regard, many firms and telecommunication companies are working hard to design household appliances based on human-appliances interaction systems. For instance, Qualcomm has developed several SH solutions using AI techniques. The SHs appliances are now being controlled using smartphones over the Internet. Thus, continuous and uninterrupted connectivity requires continuous energy sources to fulfill the demands of SHs, SBs, and smart communities. This integration of IoT with every electrical appliance requires intelligent electrical networks to efficiently monitor and control the underlying devices. One of the aims of employing ML
techniques in such systems is to accurately model the energy consumption within SBs and SHs. Additionally, intelligent power-sharing techniques where the home appliances can intelligently share the spare energy with the rest of the home appliances needs powerful ML techniques.

The energy economy is now evolving its shape from electricity generation with new energy resources to smart data networks to efficiently monitored and planned energy systems. In this regard, a new concept of the “Internet of Energy (IoE)” has emerged, which intends to develop models that can efficiently utilize energy management on top of the smart girds. In a similar context, existing Home Energy Management Systems can control the energy consumption of household appliances but fail to offer intelligent services such as spare energy transfer between appliances. Thus, IoE needs further improvements and optimizations using ML-based techniques. Finally, we can say, instead of IoE, an intelligent IoE is needed for the next generation of smart communities.

For this special section, we invite the world’s renowned researchers, scientists, and data analysts to come up with new research findings for smart energy management at the consumer end. This special section will further elaborate on the use of the powerful ML and RL concepts in controlling the energy of the appliances within smart communities. Finally, we will be able to make smart home appliances more intelligent and enable them to form a communication network of sharing energy. Specific topics include, but are not limited to, the following:

Intelligent energy transfer within smart home appliances
RL-based smart appliances
Communication among appliances for energy sharing
Support of real-time decisions in SHs/SBs/smart communities
Machine- and deep-learning techniques for energy efficiency of SHs/SBs/smart communities
Home Edge Computing for energy management
Concepts related to human-appliances interactions
IoE for intelligent decision algorithms
Integrating ambient energy sources with smart home appliances
Data analytics of smart energy data
Integration of renewable energy sources with smart home appliances