1st Edition
Swarm Intelligence Algorithms Modifications and Applications
Nature-based algorithms play an important role among artificial intelligence algorithms. Among them are global optimization algorithms called swarm intelligence algorithms. These algorithms that use the behavior of simple agents and various ways of cooperation between them, are used to solve specific problems that are defined by the so-called objective function. Swarm intelligence algorithms are inspired by the social behavior of various animal species, e.g. ant colonies, bird flocks, bee swarms, schools of fish, etc. The family of these algorithms is very large and additionally includes various types of modifications to enable swarm intelligence algorithms to solve problems dealing with areas other than those for which they were originally developed.
This book presents 24 swarm algorithms together with their modifications and practical applications. Each chapter is devoted to one algorithm. It contains a short description along with a pseudo-code showing the various stages of its operation. In addition, each chapter contains a description of selected modifications of the algorithm and shows how it can be used to solve a selected practical problem.
This book should also be useful for undergraduate and postgraduate students studying nature-based optimization algorithms, and can be a helpful tool for learning these algorithms, along with their modifications and practical applications. In addition, it can be a useful source of knowledge for scientists working in the field of artificial intelligence, as well as for engineers interested in using this type of algorithms in their work.
If the reader wishes to expand his knowledge beyond the basics of swarm intelligence algorithms presented in this book and is interested in more detailed information, we recommend the book "Swarm Intelligence Algorithms: A Tutorial" (Edited by A. Slowik, CRC Press, 2020). It contains a detailed explanation of how each algorithm works, along with relevant program codes in Matlab and the C ++ programming language, as well as numerical examples illustrating step-by-step how individual algorithms work.
1 Ant Colony Optimization, Modi□cations, and Application
Pushpendra Singh, Nand K. Meena, and Jin Yang
1.1 Introduction
1.2 Standard Ant System
1.2.1 Brief of Ant Colony Optimization
1.2.2 How arti□cial ant selects the edge to travel?
1.2.3 Pseudo-code of standard ACO algorithm
1.3 Modi□ed Variants of Ant Colony Optimization
1.3.1 Elitist ant systems
1.3.2 Ant colony system
1.3.3 Max-min ant system
1.3.4 Rank based ant systems
1.3.5 Continuous orthogonal ant systems
1.4 Application of ACO to Solve Real-life Engineering Optimization
Problem
1.4.1 Problem description
1.4.2 Problem formulation
1.4.3 How ACO can help to solve this optimization problem?
1.4.4 Simulation results
1.5 Conclusion
2 Arti□cial Bee Colony □ Modi□cations and An Application to Software Requirements Selection
Bahriye Akay
2.1 Introduction
2.2 The Original ABC algorithm in brief
2.3 Modi□cations of the ABC algorithm
2.3.1 ABC with Modi□ed Local Search
2.3.2 Combinatorial version of ABC
2.3.3 Constraint Handling ABC
2.3.4 Multi-objective ABC
2.4 Application of ABC algorithm for Software Requirement Selection
2.4.1 Problem description
2.4.2 How can the ABC algorithm be used for this problem?
2.4.2.1 Objective Function and Constraints
2.4.2.2 Representation
2.4.2.3 Local Search
2.4.2.4 Constraint Handling and Selection Operator
2.4.3 Description of the Experiments
2.4.4 Results obtained
2.5 Conclusions
References
3 Modi□ed Bacterial Forging Optimization and Application
Neeraj Kanwar, Nand K. Meena, Jin Yang, and Sonam Parashar
3.1 Introduction
3.2 Original BFO algorithm in brief
3.2.1 Chemotaxis
3.2.2 Swarming
3.2.3 Reproduction
3.2.4 Elimination and dispersal
3.2.5 Pseudo-codes of the original BFO algorithm
3.3 Modi□cations in Bacterial Foraging Optimization
3.3.1 Non-uniform elimination-dispersal probability distribution
3.3.2 Adaptive chemotaxis step
3.3.3 Varying population
3.4 Application of BFO for Optimal DER Allocation in Distribution Systems
3.4.1 Problem description
3.4.2 Individual bacteria structure for this problem
3.4.3 How can the BFO algorithm be used for this problem?
3.4.4 Description of experiments
3.4.5 Results obtained
3.5 Conclusions
4 Bat Algorithm □ Modi□cations and Application
Neeraj Kanwar, Nand K. Meena, and Jin Yang
4.1 Introduction
4.2 Original Bat Algorithm in Brief
4.2.1 Random □y
4.2.2 Local random walk
4.3 Modi□cations of the Bat algorithm
4.3.1 Improved bat algorithm
4.3.2 Bat algorithm with centroid strategy
4.3.3 Self-adaptive bat algorithm (SABA)
4.3.4 Chaotic mapping based BA
4.3.5 Self-adaptive BA with step-control and mutation mechanisms
4.3.6 Adaptive position update
4.3.7 Smart bat algorithm
4.3.8 Adaptive weighting function and velocity
4.4 Application of BA for optimal DNR problem of distribution system
4.4.1 Problem description
4.4.2 How can the BA algorithm be used for this problem?
4.4.3 Description of experiments
4.4.4 Results
4.5 Conclusion
5 Cat Swarm Optimization - Modi□cations and Application
Dorin Moldovan, Adam Slowik, Viorica Chifu, and Ioan Salomie
5.1 Introduction
5.2 Original CSO algorithm in brief
5.2.1 Description of the original CSO algorithm
5.3 Modi□cations of the CSO algorithm
5.3.1 Velocity clamping
5.3.2 Inertia weight
5.3.3 Mutation operators
5.3.4 Acceleration coe□cient c1
5.3.5 Adaptation of CSO for diets recommendation
5.4 Application of CSO algorithm for recommendation of diets
5.4.1 Problem description
5.4.2 How can the CSO algorithm be used for this problem?
5.4.3 Description of experiments
5.4.4 Results obtained
5.4.4.1 Diabetic diet experimental results
5.4.4.2 Mediterranean diet experimental results
5.5 Conclusions
References
6 Chicken Swarm Optimization - Modi□cations and Application
Dorin Moldovan and Adam Slowik
6.1 Introduction
6.2 Original CSO algorithm in brief
6.2.1 Description of the original CSO algorithm
6.3 Modi□cations of the CSO algorithm
6.3.1 Improved Chicken Swarm Optimization (ICSO)
6.3.2 Mutation Chicken Swarm Optimization (MCSO)
6.3.3 Quantum Chicken Swarm Optimization (QCSO)
6.3.4 Binary Chicken Swarm Optimization (BCSO)
6.3.5 Chaotic Chicken Swarm Optimization (CCSO)
6.3.6 Improved Chicken Swarm Optimization - Rooster Hen Chick (ICSO-RHC)
6.4 Application of CSO for Detection of Falls in Daily Living Activities
6.4.1 Problem description
6.4.2 How can the CSO algorithm be used for this problem?
6.4.3 Description of experiments
6.4.4 Results obtained
6.4.5 Comparison with other classi□cation approaches
6.5 Conclusions
References
7 Cockroach Swarm Optimization □ Modi□cations and Application
Joanna Kwiecien
7.1 Introduction
7.2 Original CSO algorithm in brief
7.2.1 Pseudo-code of CSO algorithm
7.2.2 Description of the original CSO algorithm
7.3 Modi□cations of the CSO algorithm
7.3.1 Inertia weight
7.3.2 Stochastic constriction coe□cient
7.3.3 Hunger component
7.3.4 Global and local neighborhoods
7.4 Application of CSO algorithm for traveling salesman problem
7.4.1 Problem description
7.4.2 How can the CSO algorithm be used for this problem?
7.4.3 Description of experiments
7.4.4 Results obtained
7.5 Conclusions
References
8 Crow Search Algorithm - Modi□cations and Application
Adam Slowik and Dorin Moldovan
8.1 Introduction
8.2 Original CSA in brief
8.3 Modi□cations of CSA
8.3.1 Chaotic Crow Search Algorithm (CCSA)
8.3.2 Modi□ed Crow Search Algorithm (MCSA)
8.3.3 Binary Crow Search Algorithm (BCSA)
8.4 Application of CSA for Jobs Status Prediction
8.4.1 Problem description
8.4.2 How can CSA be used for this problem?
8.4.3 Experiments description
8.4.4 Results
8.5 Conclusions
References
9 Cuckoo Search Optimisation □ Modi□cations and Application
Dhanraj Chitara, Nand K. Meena, and Jin Yang
9.1 Introduction
9.2 Original CSO Algorithm in Brief
9.2.1 Breeding behavior of cuckoo
9.2.2 Levy Flights
9.2.3 Cuckoo search optimization algorithm
9.3 Modi□ed CSO Algorithms
9.3.1 Gradient free cuckoo search
9.3.2 Improved cuckoo search for reliability optimization problems
9.4 Application of CSO Algorithm for Designing Power System Stabilizer
9.4.1 Problem description
9.4.2 Objective function and problem formulation
9.4.3 Case study on two-area four machine power system
9.4.4 Eigenvalue analysis of TAFM power system without and with PSSs
9.4.5 Time-domain simulation of TAFM power system
9.4.6 Performance indices results and discussion of TAFM power system
9.5 Conclusion
10 Improved Dynamic Virtual Bats Algorithm for Identifying a Suspension System Parameters
Ali Osman Topal
10.1 Introduction
10.2 Original Dynamic Virtual Bats Algorithm (DVBA)
10.3 Improved Dynamic Virtual Bats Algorithm (IDVBA)
10.3.1 The weakness of DVBA
10.3.2 Improved Dynamic Virtual Bats Algorithm (IDVBA)
10.4 Application of IDVBA for identifying a suspension system
10.5 Conclusions
11 Dispersive Flies Optimisation: Modi□cations and Application
Mohammad Majid al-Rifaie, Hooman Oroojeni M. J., and Mihalis Nicolaou
11.1 Introduction
11.2 Dispersive Flies Optimisation
11.3 Modi□cations in DFO
11.3.1 Update Equation
11.3.2 Disturbance Threshold,
11.4 Application: Detecting false alarms in ICU
11.4.1 Problem Description
11.4.2 Using Dispersive Flies Optimisation
11.4.3 Experiment Setup
11.4.3.1 Model Con□guration
11.4.3.2 DFO Con□guration
11.4.4 Results
11.5 Conclusions
References
12 Improved Elephant Herding Optimization and Application
Nand K. Meena and Jin Yang
12.1 Introduction
12.2 Original Elephant Herding Optimization
12.2.1 Clan updating operator
12.2.2 Separating operator
12.3 Improvements in Elephant Herding Optimization
12.3.1 Position of leader elephant
12.3.2 Separation of male elephant
12.3.3 Chaotic maps
12.3.4 Pseudo-code of improved EHO algorithm
12.4 Application of IEHO for Optimal Economic Dispatch of Microgrids
12.4.1 Problem Statement
12.4.2 Application of EHO to solve this problem
12.4.3 Application in Matlab and Source-code
12.5 Conclusions
Acknowledgement
References
13 Fire□y Algorithm: Variants and Applications
Xin-She Yang
13.1 Introduction
13.2 Fire□y Algorithm
13.2.1 Standard FA
13.2.2 Special Cases of FA
13.3 Variants of Fire□y Algorithm
13.3.1 Discrete FA
13.3.2 Chaos-Based FA
13.3.3 Randomly Attracted FA with Varying Steps
13.3.4 FA via Lévy Flights
13.3.5 FA with Quaternion Representation
13.3.6 Multi-objective FA
13.3.7 Other Variants of FA
13.4 Applications of FA and its Variants
13.5 Conclusion
References
14 Glowworm Swarm Optimization - Modi□cations and Applications
Krishnanand Kaipa and Debasish Ghose
14.1 Introduction
14.2 Brief Description of GSO
14.3 Modi□cations to GSO Formulation
14.3.1 Behavior Switching Modi□cation
14.3.2 Local Optima Mapping Modi□cation
14.3.3 Coverage Maximization Modi□cation
14.3.4 Physical Robot Modi□cation
14.4 Engineering Applications of GSO
14.4.1 Application of Behavior Switching to Multiple Source Localization and Boundary Mapping
14.4.2 Application of Local Optima Mapping Modi□cation to Clustering
14.4.3 Application of Coverage Maximization Modi□cation to Wireless Networks
14.4.4 Application of Physical Robot Modi□cation to Signal Source Localization
14.5 Conclusions
References
15 Grasshopper Optimization Algorithm - Modi□cations and Applications
Szymon ukasik
15.1 Introduction
15.2 Description of the Original Grasshopper Optimization Algorithm
15.3 Modi□cations of the GOA technique
15.3.1 Adaptation to Other Optimization Domains
15.3.2 Structural Modi□cations
15.3.3 Hybrid algorithms
15.4 Application Example: GOA-based Clustering
15.4.1 Clustering and Optimization
15.4.2 Experimental Setting and Results
15.5 Conclusion
References
16 Grey wolf optimizer □ Modi□cations and Applications
Ahmed F. Ali and Mohamed A. Tawhid
16.1 Introduction
16.2 Original GWO algorithm in brief
16.2.1 Description of the original GWO algorithm
16.3 Modi□cations of the GWO algorithm
16.3.1 Chaotic maps
16.3.2 Chaotic grey wolf operator
16.4 Application of GWO algorithm for Engineering optimization problems
16.4.1 Engineering optimization problems problem
16.4.1.1 Welded beam design problem
16.4.1.2 Pressure vessel design problem
16.4.1.3 Speed reducer design problem
16.4.1.4 Three-bar truss design problem
16.4.1.5 Tension compression spring problem
16.4.2 Description of experiments
16.4.3 Convergence curve of CGWO with engineering optimization problems
16.4.4 Comparison between CGWO and GWO with engineering optimization problems
16.5 Conclusions
References
17 Hunting Search Optimization Modi□cation and Application
Ferhat Erdal, Osman Tunca, and Erkan Dogan
17.1 Introduction
17.2 Original HuS Algorithm in Brief
17.2.1 Description of the original hunting search algorithm
17.2.1.1 Description of the global version of the HuS algorithm
17.3 Improvements in the hunting search algorithm
17.4 Applications of the algorithm to theWelded Beam Design Problem
17.4.1 Problem description
17.4.2 How can the hunting search algorithm be used for this problem?
17.4.3 Description of experiments
17.4.4 Result obtained
17.5 Conclusions
References
18 Krill Herd Algorithm □ Modi□cations and Applications
Ali R. Kashani, Charles V. Camp, Hamed Tohidi, and Adam Slowik
18.1 Introduction
18.2 Original KH algorithm in brief
18.3 Modi□cations of the KH algorithm
18.3.1 Chaotic KH
18.3.2 Levy-□ight KH
18.3.3 Multi-stage KH
18.3.4 Stud KH
18.3.5 KH with linear decreasing step
18.3.6 Biography-based krill herd
18.4 Application of KH algorithm for optimum design of retaining walls
18.4.1 Problem description
18.4.2 How can KH algorithm be used for this problem?
18.4.3 Description of experiments
18.4.4 Results obtained
18.5 Conclusions
References
19 Modi□ed Monarch Butter□y Optimization and Real-life Applications
Pushpendra Singh, Nand K. Meena, and Jin Yang
19.1 Introduction
19.2 Monarch butter□y optimization
19.2.1 Migration Operator
19.2.2 Butter□y adjusting operator
19.3 Modi□ed Monarch Butter□y Optimization Method
19.3.1 Modi□ed migration operator
19.3.2 Modi□ed butter□y adjustment operator
19.4 Algorithm of Modi□ed MBO
19.5 Matlab Source-code of GCMBO
19.6 Application of GCMBO for Optimal Allocation of Distributed Generations
19.6.1 Problem Statement
19.6.2 Optimization framework for optimal DG allocation
19.7 Conclusion
20 Particle Swarm Optimization □ Modi□cations and Application
Adam Slowik
20.1 Introduction
20.2 Original PSO algorithm in brief
20.2.1 Description of the original PSO algorithm
20.3 Modi□cations of the PSO algorithm
20.3.1 Velocity clamping
20.3.2 Inertia weight
20.3.3 Constriction coe□cient
20.3.4 Acceleration coe□cients c1 and c2
20.4 Application of PSO algorithm for IIR digital □lter design
20.4.1 Problem description
20.4.2 How can the PSO algorithm be used for this problem?
20.4.3 Description of experiments
20.4.4 Results obtained
20.5 Conclusions
References
21 Salp Swarm Algorithm: Modi□cation and Application
Essam H. Houssein, Ibrahim E. Mohamed , and Aboul Ella Hassanien
21.1 Introduction
21.2 Salp Swarm Algorithm (SSA) in brief
21.2.1 Inspiration Analysis
21.2.2 Mathematical Model for salp Chains
21.3 Modi□cations of SSA Algorithm
21.3.1 Fuzzy Logic
21.3.2 Robust
21.3.3 Simplex
21.3.4 Weight Factor and Adaptive Mutation
21.3.5 Levy Flight
21.3.6 Binary
21.3.7 Chaotic
21.3.8 Multi-Objective Problems (MOPS)
21.4 Application of SSA for welded beam design problem
21.4.1 Problem description
21.4.2 How to use SSA to optimize this problem?
21.4.3 Result obtained
21.5 Conclusion
References
22 Social Spider Optimization □ Modi□cations and Applications
Ahmed F. Ali and Mohamed A. Tawhid
22.1 Introduction
22.2 Original SSO algorithm in brief
22.2.1 Description of the original SSO algorithm
22.3 Modi□cations of the SSO algorithm
22.3.1 Chaotic maps
22.3.2 Chaotic Female cooperative operator
22.3.3 Chaotic Male cooperative operator
22.4 Application of SSO algorithm for economic load dispatch problem
22.4.1 Economic load dispatch problem
22.4.2 Problem Constraints
22.4.3 Penalty Function
22.4.4 How can the SSO algorithm be used for economic load dispatch problem?
22.4.5 Description of experiments
22.4.6 Results obtained
22.5 Conclusions
References
23 Stochastic Di□usion Search: Modi□cations and Application
Mohammad Majid al-Rifaie and J. Mark Bishop
23.1 Introduction
23.2 SDS algorithm
23.3 Further modi□cations and adjustments
23.3.1 Recruitment Strategies
23.3.1.1 Passive Recruitment Mode
23.3.1.2 Active Recruitment Mode
23.3.1.3 Dual Recruitment Mode
23.3.1.4 Context Sensitive Mechanism
23.3.1.5 Context Free Mechanism
23.3.2 Initialisation and Termination
23.3.3 Partial Function Evaluation
23.4 Application: Identifying metastasis in bone scans
23.4.1 Experiment setup
23.4.2 Results
23.4.3 Concluding remarks
23.5 Conclusion
References
24 Whale Optimization Algorithm □ Modi□cations and Applications
Ali R. Kashani, Charles V. Camp, Moein Armanfar, and Adam Slowik
24.1 Introduction
24.2 Original WOA algorithm in brief
24.3 Modi□cations of WOA algorithm
24.3.1 Chaotic WOA
24.3.2 Levy-□ight WOA
24.3.3 Binary WOA
24.3.4 Improved WOA
24.4 Application of WOA algorithm for optimum design of shallow foundation
24.4.1 Problem description
24.4.2 How can WOA algorithm be used for this problem?
24.4.3 Description of experiments
24.4.4 Results obtained
24.5 Conclusions
References
Biography
Adam Slowik (IEEE Member 2007; IEEE Senior Member 2012) is an Associate Professor in the Department of Electronics and Computer Science, Koszalin University of Technology. His research interests include soft computing, computational intelligence, and, particularly, bio-inspired optimization algorithms and their engineering applications. He was a recipient of one Best Paper Award (IEEE Conference on Human System Interaction - HSI 2008).