Novel Hybrid Optimization Techniques to Enhance Reliability from Reverse Osmosis Desalination Process

Abstract

Water is the most important resource of the Earth and is significantly utilized for agriculture, urbanization, industry, and population. This increases the demand for water; meanwhile, the climatic condition decreases the supply of it. A rise in temperature of 1 degree Celsius might dry up 20% of renewable water resources, and to circumvent the water scarcity, it is necessary to reuse, create, and consume less water without wasting it. Water desalination is the process used to reuse the used or saline water by promptly extracting the salt or unwanted minerals and producing fresh consumable water. Based on the International Desalination Association, around 300 million people rely on desalination and the people of the Middle East region rely the most upon it. Around 7% of desalination plants are located in countries such as Saudi Arabia, Bahrain, Kuwait, and the United Arab Emirates. Reverse osmosis (RO) is the relevant desalination process in this type of area however, the conventional methods include more complexities, and hence to address this issue we proposed a novel approach known as Hybrid Capuchin and Rat swarm algorithm (HCRS) for effective water desalination technology using conventional sources and renewable energy in the middle east region. Moreover, a hybrid reverse osmosis plant model is developed for identifying renewable sources such as wind and solar energy. The proposed optimization can be used to mitigate the life cycle cost and enhances the reliability of the hybrid schemes. The experiment is conducted in a MATLAB simulator and compared the results with state-of-art works over the metrics such as relative error, system cost, and reliability. Our proposed method outperforms all the other approaches.

Prediction of Water Quality Using SoftMax-ELM Optimized Using Adaptive Crow-Search Algorithm

“…The discussion in the previous sub-section reveals that the Extreme Learning Machine model is faster and more competent in providing solutions for several non-linear classification problems. At the same time, the characteristics of poor generalization and stability made ELM to yield poor classification performance [39]. The randomly selecting nature of ELM's input weight and hidden bias parameters is the primary reason behind its reduced performance.…”

Optimizing Energy Consumption in Smart Grids Using Demand Response Techniques

“…A spike in energy use at these times may put a strain on the grid, resulting in erratic voltage, power outages, and other problems. Less effective and more expensive power sources are typically used to accommodate the growing demand, driving up customer energy costs [20].…”