Using Particle Swarm Optimization Algorithm as an Optimization Tool Within Developed Neural Networks

Developed neural networks as an output could have numerous potential outputs caused by numerous combinations of input values. When we are in position to find optimal combination of input values for achieving specific output value within neural network model it is not a trivial task. This request comes from profiling purposes if, for example, neural network gives information of specific profile regarding input or recommendation system realized by neural networks, etc. Utilizing evolutionary algorithms like particle swarm optimization algorithm, which will be illustrated in this chapter, can solve these problems.

Multi-Objective Binary Fish School Search

Fish school search (FSS) algorithm has inspired several adaptations for multi-objective problems or binary optimization. However, there is no particular proposition to solve both problems simultaneously. The proposed multi-objective approach binary fish school search (MOBFSS) aims to solve optimization problems with two or three conflicting objective functions with binary decision input variables. MOBFSS is based on the dominance concept used in the multi-objective fish school search (MOFSS) and the threshold technique deployed in the binary fish school search (BFSS). Additionally, the authors evaluate the proposal for feature selection for classification in well-known datasets. Moreover, the authors compare the performance of the proposal with a state-of-art algorithm called BMOPSO-CDR. MOBFSS presents better results than BMOPSO-CDR, especially for datasets with higher complexity.

Local Best Particle Swarm Optimization Using Crown Jewel Defense Strategy

Particle swarm optimization (PSO) is a swarm intelligence algorithm well known for its simplicity and high efficiency on various problems. Conventional PSO suffers from premature convergence due to the rapid convergence speed and lack of population diversity. It is easy to get trapped in local optima. For this reason, improvements are made to detect stagnation during the optimization and reactivate the swarm to search towards the global optimum. This chapter imposes the reflecting bound-handling scheme and von Neumann topology on PSO to increase the population diversity. A novel crown jewel defense (CJD) strategy is introduced to restart the swarm when it is trapped in a local optimum region. The resultant algorithm named LCJDPSO-rfl is tested on a group of unimodal and multimodal benchmark functions with rotation and shifting. Experimental results suggest that the LCJDPSO-rfl outperforms state-of-the-art PSO variants on most of the functions.

Swarm-Intelligence-Based Communication Protocols for Wireless Sensor Networks

Social insect communities are formed from simple, autonomous, and cooperative organisms that are interdependent for their survival. These communities are able to effectively coordinate themselves to achieve global objectives despite a lack of centralized planning, and the behaviour is referred to as swarm intelligence. This chapter presents a study of communication protocols for wireless sensor networks utilizing nature-inspired systems: social insect-based communities and natural creatures. Three types of insects are used for discussion: ants, termites, and bees. In addition, a study of the social foraging behavior of spider monkeys is presented. The performances of these swarm-intelligence-based algorithms were tested on common routing scenarios. The results were compared with other routing algorithms with varying network density and showed that swarm-intelligence-based routing techniques improved on network energy consumption with a control over best-effort service. The results were strengthened with a model of termite-hill routing algorithm for WSN.

Particle Swarm Optimization for Model Predictive Control in Reinforcement Learning Environments

This chapter introduces a model-based reinforcement learning (RL) approach for continuous state and action spaces. While most RL methods try to find closed-form policies, the approach taken here employs numerical online optimization of control action sequences following the strategy of nonlinear model predictive control. First, a general method for reformulating RL problems as optimization tasks is provided. Subsequently, particle swarm optimization (PSO) is applied to search for optimal solutions. This PSO policy (PSO-P) is effective for high dimensional state spaces and does not require a priori assumptions about adequate policy representations. Furthermore, by translating RL problems into optimization tasks, the rich collection of real-world-inspired RL benchmarks is made available for benchmarking numerical optimization techniques. The effectiveness of PSO-P is demonstrated on two standard benchmarks mountain car and cart-pole swing-up and a new industry-inspired benchmark, the so-called industrial benchmark.

Using Cuckoo Search Algorithm for Hybrid Flow Shop Scheduling Problems Under Makespan Criterion

In this chapter, cuckoo search algorithm (CSA) is used to solve the multistage hybrid flow shop (HFS) scheduling problems with parallel machines. The objective is the minimization of makespan. The HFS scheduling problems are proved to be strongly non-deterministic polynomial time-hard (NP-hard). Proposed CSA algorithm has been tested on benchmark problems addressed in the literature against other well-known algorithms. The results are presented in terms of percentage deviation (PD) of the solution from the lower bound. The results indicate that the proposed CSA algorithm is quite effective in reducing makespan because average PD is observed as 1.531, whereas the next best algorithm has result of average PD of 2.295, which is, in general, nearly 50% worse, and other algorithms start from 2.645.

Optimized Base Station Sleeping and Smart Grid Energy Procurement Scheme to Improve Energy Efficiency

In this chapter, the authors propose a joint BS sleeping strategy, resource allocation, and energy procurement scheme to maximize the profit of the network operators and minimize the carbon emission. Then, a joint optimization problem is formulated, which is a mixed-integer programming problem. To solve it, they adopt the bi-velocity discrete particle swarm optimization (BVDPSO) algorithm to optimize the BS sleeping strategy. When the BS sleeping strategy is fixed, the authors propose an optimal algorithm based on Lagrange dual domain method to optimize the power allocation, subcarrier assignment, and energy procurement. Numerical results illustrate the effectiveness of the proposed scheme and algorithm.

Image Reconstruction of Electrical Impedance Tomography Using Fish School Search and Differential Evolution

Electrical impedance tomography (EIT) is a noninvasive imaging technique that does not use ionizing radiation with application both in environmental sciences and in health. Image reconstruction is performed by solving an inverse problem and ill-posed. Evolutionary and bioinspired computation have become a source of methods for solving inverse problems. In this chapter, the authors investigate the performance of fish school search (FSS) and differential evolution (DE) using non-blind search (NBS) considering meshes of 415, 3190, and 9990 finite elements. The methods were evaluated using numerical phantoms consisting of electrical conductivity images with objects in the center, between the center and the edge, and on the edge of a circular section. Twenty simulations were performed for each configuration. Results showed that both FSS and DE are able to perform EIT image reconstruction with large meshes and converge faster by using non-blind search.

Squeeze Casting Parameter Optimization Using Swarm Intelligence and Evolutionary Algorithms

This chapter is focused to locate the optimum squeeze casting conditions using evolutionary swarm intelligence and teaching learning-based algorithms. The evolutionary and swarm intelligent algorithms are used to determine the best set of process variables for the conflicting requirements in multiple objective functions. Four cases are considered with different sets of weight fractions to the objective function based on user requirements. Fitness values are determined for all different cases to evaluate the performance of evolutionary and swarm intelligent methods. Teaching learning-based optimization and multiple-objective particle swarm optimization based on crowing distance have yielded similar results. Experiments have been conducted to test the results obtained. The performance of swarm intelligence is found to be comparable with that of evolutionary genetic algorithm in locating the optimal set of process variables. However, TLBO outperformed GA, PSO, and MOPSO-CD with regard to computation time.

A Computational Comparison of Swarm Optimization Techniques for Optimal Load Shedding Under the Presence of FACTS Devices to Avoid Voltage Instability

Voltage instability has become a serious threat to the operation of modern power systems. Load shedding is one of the effective countermeasures for avoiding instability. Improper load shedding may result in huge technical and economic losses. So, an optimal load shedding is to be carried out for supplying more demand. This chapter implements bat and firefly algorithms for solving the optimal load shedding problem to identify the optimal amount of load to be shed. This is applied for a multi-objective function which contains minimization of amount of load to be shed, active power loss minimization, and voltage profile improvement. The presence of with and without static VAR compensator (SVC), thyristor-controlled series capacitor (TCSC), and unified power flow controller (UPFC) on load shedding for IEEE 57 bus system has been presented and analyzed. The results obtained with bat and firefly algorithms were compared with genetic algorithm (GA) and also the impact of flexible AC transmission system (FACTS) devices on load shedding problem has been analyzed.