An Enhanced Energy Optimization Model for Industrial Wireless Sensor Networks Using Machine Learning

Abstract

Industrial Wireless Sensor Networks (WSNs) are becoming increasingly popular due to their enhanced scalability and low cost of deployment. However, they also present new challenges, such as energy optimization and network maintenance, which industrial users must address. In order to meet the challenges, Machine Learning techniques have been used to create an enhanced energy optimization model for Industrial WSNs. This model utilizes knowledge-based learning to identify and optimize the energy consumption of the nodes, allowing Industrial WSNs to consume the least amount of energy for the given tasks. In addition, the model also evaluates the effectiveness of feedback control schemes and predicts the best possible outcomes for its application in Industrial WSNs to ensure higher efficiency and longer network lifetime. The model also enables the exploration of potential trade-offs between power consumption and communication performance to ensure a better energy-efficient solution. The proposed EEOM obtained 64.72% transmission energy consumption, 35.28% transmission energy saving, 67.27% received energy consumption, 32.73% received energy storage, 52.16% idle-mode energy consumption, 47.84% idle-mode energy storage, 66.31% sleep-mode energy consumption, and 33.69% sleep-mode energy storage. It also obtained 90.44% prevalence threshold, 90.33% critical success index, 93.93% Delta-P, 90.06% MCC and 92.17% FMI rates. It also provides the ability to identify the best selection of nodes and paths for data transmission to reduce network traffic. When applied in conjunction with manual intervention, these automated knowledge-based techniques will make Industrial WSNs more reliable, efficient, and energy-cost effective.

Keywords

• Industrial
• wireless
• sensor
• networks
• WSN
• scalability