Chapter five - IoT in renewable energy generation for conservation of energy using artificial intelligence

Abstract

Internet of things (IoT) consists of many devices interconnected virtually over remotely different technologies and devices over the Internet. The identification and collection of the data based on temporal and spatial information from the low-range communication devices in a massive environment is challenging. Data is collected from the low range communication devices and cooperatively forwarded to the base station (BS) with the help of low range and high range communication devices. A heterogeneous environment is challenging concerning their semantic context, the number of IoT devices increases in the spatial location and their nature of diverging concerning different IoT device type and technologies. Residual data or replication data concerning various types of devices and technologies need more energy and delay, directly affecting the network lifetime (NL) and conservation of energy (CE). Naming and addressing the low range communication device considerably influence the performance based on data collection, energy utilization, and latency. Connect standard range communication devices cooperatively to monitor the real-world phenomena and transfer the collected data through the higher range communication device to the BS. Increasing or decreasing the number of nodes in this layer significantly affects the frequency and bandwidth utilization of the networks that need IoT in renewable energy generation (REG). Discovering and transmitting the data independently over different communication devices will consume more energy. Various scenarios in sensing and the communication layer of an IoT in a REG environment are considered to overcome the above-discussed challenges. We work on IoT in REG for CE using artificial intelligence (AI). We proposed that AI-based reinforcement learning (RL) with long short-term memory (LSTM) construction is significant for improving IoT performance in REG and proposed fog computing-based architecture for IoT in REG for conservation using hybrid deep learning. Primarily AI-based Rewet LSTM is constructed based on a homogeneous network to analyze the coverage, delay, and energy utilization among IoT sensor devices in a large environment. All the devices in this topology are considered as same concerning the transmission range and energy.