Preface

Optimization is a precise method that allows the planner to identify the best solution to a problem by using design restrictions and criteria. Optimization techniques have been used to solve a variety of practical problems in a variety of fields. These optimization methods have existed since the time of Newton, Lagrange and Cauchy. The contributions of Leibnitz and Newton to calculus are responsible for the growth of differential calculus approaches for optimization. Optimization techniques are useful tools for obtaining the required design parameters and operational conditions. They direct the experimental effort and lower the design and operations risk and cost. Finding the values of decision variables that correspond to and give the maximum or minimum of one or more specified objectives is referred to as optimization. An optimization algorithm is a process that compares numerous solutions iteratively until an optimum or satisfying solution is identified. Thus, optimization has become a feature of computer-aided design activities since the invention of computers. The formulation of goal functions and the optimization technique chosen determine the reliability of optimal solutions. A mathematical model that characterizes and predicts the process behavior is required for optimization. An optimization search could aid in the estimation of unknown parameters in complex nonlinear processes. In dynamic processes, robust optimization can be used to find uncertainty variables. Optimization can also be used to aid in the development of scale-up methodologies and the design of multiphase reactors and flow systems. Manufacturing and engineering activities currently being used will not be as efficient until design and operations are optimized. The purpose of design optimization might simply be to reduce production costs or to increase production efficiency. Optimization is particularly important in companies since it helps to cut costs, which can lead to increased earnings and success in a competitive environment.

There are two types of optimization approaches used: traditional and soft computing-based. Traditional optimization techniques can be used to identify the best solution or unconstrained maxima and minima of continuous and differentiable functions. These are mathematical methods that use differential calculus to get the best solution. Optimization can decrease readability and introduce code that is only needed to boost performance. This might make programs or systems more difficult to maintain and debug. As a result, performance tweaking or optimization is frequently done near the conclusion of the development stage. Customer experience optimization is critical since it boosts customer satisfaction and helps organizations improve their key performance indicators. More revenue and growth are frequently associated with satisfied customers and improved key performance indicators. Performance optimization is the practice of altering how a system functions to increase its efficiency and effectiveness. Debugging the optimization solution is more complex than debugging the rule-based simulation solution. It could be a mix of numerous limitations and input data from multiple places and time steps. Many fields have used optimization theory and methodologies to solve various practical challenges. Soft computing is the application of approximation computations to difficult computer problems to produce "mushy" but usable outcomes. The method produces results for issues that are either unsolvable or take too long to solve using traditional methods.

An optimization algorithm is a process that compares numerous solutions iteratively until an optimum or satisfying one is identified. It follows particular rules when moving from one solution to the next. Many fields of study employ optimization methods to find solutions that maximize or minimize certain study parameters, such as minimizing expenses in the manufacture of a thing or service, maximizing earnings, minimizing raw materials in the development of a good, or maximizing productivity.

The goal of optimizing an objective function is to identify a set of inputs that results in a maximum or minimal function evaluation. Many machine learning algorithms, from fitting logistic regression models to training artificial neural networks, are based on this concept. The purpose of the optimization process is to find choice variable values that result in an objective function's maximum or minimum. Optimization issues are characterized as linear, nonlinear, geometric, or quadratic programming problems based on the nature of the expressions for the objective function and constraints. An optimization model is a representation of the key features of the business problem being addressed. The objective function, decision variables, and business constraints are the three components of the model.

The purpose of optimization is to find the best design possible based on a set of prioritized criteria or constraints. These include, among other things, boosting productivity, strength, reliability, lifetime, efficiency, and utilization. Many fields have used optimization theory and methodologies to solve various practical challenges; and optimization techniques have become increasingly important and popular in various engineering applications as computing systems have advanced.

The aim of this book is to present some of the recent developments in the area of optimization theory, methods, and applications in engineering. It focuses on the metaphor of the inspired system and how to configure and apply the various algorithms. The book is organized into two parts: Part I - Soft Computing and Evolutionary-Based Optimization; and Part II - Decision Science and Simulation-Based Optimization, which contains application-based chapters. A brief description of each of the chapters is presented below:

Part I: Soft Computing and Evolutionary-Based Optimization

• Chapter 1 attempts to realistically represent the existing grey wolf optimizer in order to increase the algorithm's efficiency. Unlike grey wolf optimizers, the modeling of prey in this study is considered dynamic. To solve a dynamic economic dispatch problem with electric vehicle profiles, dynamism is added to the prey using the Levy flight distribution function.
• Chapter 2 presents an object detection approach devised for detecting plastics. To apply transfer learning, the algorithm is written in tensor flow. The two most commonly used object identification approaches, YOLO and Faster R-CNN, are compared.
• Chapter 3 strives to increase the use of real power in relation to reactive power in order to improve the power factor. The power factor is corrected closer to unity with the help of a smart power factor correction device. The power factor controller swaps the appropriate capacitor blocks in steps depending on the load and power factor of the network to raise the power factor to almost 0.95. MATLAB software is used to simulate the system.
• Chapter 4 focuses on the design and analysis of a solar-fed cascaded boost converter with a maximum power point tracking (MPPT) algorithm based on neural networks (NN), which is used for electric vehicle (EV) applications.
• Chapter 5 proposes modeling of a single and multi-junction solar cell with maximum power point tracking (MPPT) using an intelligent fuzzy logic algorithm (FLA) for maximum productivity.
• Chapter 6 discusses the utility of particle swarm optimization (PSO), as well as the weaknesses in the algorithm. The latest developments and improvements to the PSO parameters are also highlighted. Finally, it explores its hybridization with other notable algorithms and applications in a variety of disciplines and contexts during the last few decades.
• Chapter 7 delves into the uniformity subfields of sensor networks, as well as energy-efficient sensor networks, which actively investigate the use of genetic algorithms (GAs) as a fundamental component in developing deployment strategies and routing protocols. Since manual garbage collection is a time-consuming technique that cannot keep up with the ever-increasing demands of city rubbish, sensor networks can be used in garbage collection as the first step toward proper plastics treatment and recycling for a more environmentally friendly future.
• Chapter 8 presents several machine learning methods, such as random forests, linear regression, XGBoost, and support vector machine (SVM), that aid in the estimation of delamination factor based on known inputs such as feed rate, point angle, and spindle speed.
• Chapter 9 attempts using a differential evolutionary algorithm to perform sand elevation categorization of sand deposits acquired with a drone camera in a desert location. However, the results are unsatisfactory, prompting the development of an evolutionary algorithm-based contour identification approach for accurate elevation categorization.
• Chapter 10 mainly discusses promoting energy efficiency while maintaining a sustainable minimum spectral efficiency requirement. By obtaining the Pareto optimal solution set, the joint user clustering and multi-objective particle swarm optimization (MOPSO) method has been proposed for downlink non-orthogonal multiple access (NOMA). The simulation results reveal that the proposed MOPSO approach outperforms the existing systems in terms of parametric efficiency.
• Chapter 11 uses the Amazon fine food reviews dataset to automatically analyze product reviews and classify them as good or terrible. Four distinct types of classifiers were employed to identify the reviews...