Nature-Inspired Computing Paradigms in Systems: Reliability, Availability, Maintainability, Safety and Cost (RAMS+C) and Prognostics and Health Management (PHM)

✍ Scribed by Mohamed Arezki Mellal; Michael G. Pecht

• Publisher: Elsevier Science
• Year: 2021
• Tongue: English
• Leaves: 134
• Series: Intelligent Data-Centric Systems: Sensor Collected Intelligence
• Edition: 1
• Category: Library

✦ Synopsis

Nature-Inspired Computing Paradigms in Systems: Reliability, Availability, Maintainability, Safety and Cost (RAMS+C) and Prognostics and Health Management (PHM) covers several areas that include bioinspired techniques and optimization approaches for system dependability.

The book addresses the issue of integration and interaction of the bioinspired techniques in system dependability computing so that intelligent decisions, design, and architectures can be supported. It brings together these emerging areas under the umbrella of bio- and nature-inspired computational intelligence.

The primary audience of this book includes experts and developers who want to deepen their understanding of bioinspired computing in basic theory, algorithms, and applications. The book is also intended to be used as a textbook for masters and doctoral students who want to enhance their knowledge and understanding of the role of bioinspired techniques in system dependability.

• Provides the latest review
• Covers various nature-inspired techniques applied to RAMS+C and PHM problems
• Includes techniques applied to new applications

✦ Table of Contents

Front matter
Copyright
Contributors
Editor biographies
Preface
Acknowledgment
Reliability optimization of power plant safety system using grey wolf optimizer and shuffled frog-leaping algorith
Introduction
Literature review
Problem description
Grey wolf optimizer
Shuffled frog-leaping algorithm
Results and discussion
Conclusions
References
Design optimization of a car side safety system by particle swarm optimization and grey wolf optimizer
Introduction
Design optimization of a car side safety system
Particle swarm optimization
Grey wolf optimizer
Results and discussion
Conclusions
References
Genetic algorithms: Principles and application in RAMS
Introduction
GA construction
Genetic operators
Crossover operator
Mutation operation
Adaptive and hybrid approaches in the GA
The GA-PSO framework
Stop condition
GA applications
Reliability-based design optimization
Reliability allocation problems
Redundancy allocation problems
Redundancy allocation for a complex system
Multilevel redundancy allocation
Inspection and maintenance planning for one-shot systems
Joint optimization of spare parts inventory and maintenance policies
Industry 4.0 and optimization
Advantages and disadvantages of the GA
Conclusion
References
Evolutionary optimization for resilience-based planning for power distribution networks
Introduction
Problem description and formulation
Power distribution network
Preventive maintenance actions
Objective function
Constraints
Total number of replacements
Replacements per period
Subsequent replacements
Model
Solution methodology
Differential evolution
Binary differential evolution
Archiving-based adaptive tradeoff model (ArATM)
Results
Conclusions
References
Application of nature-inspired computing paradigms in optimal design of structural engineering problems&mdash
Introduction
Nature-inspired algorithms
Swarm intelligence algorithms
Bioinspired algorithms
Physics- and chemistry-based algorithms
Nature-inspired metaheuristics in optimal design of structural engineering problems
SI algorithms in optimal design of structural engineering problems
Bioinspired algorithms in optimal design of structural engineering problems
Physics- and chemistry-based algorithms in optimal design of structural engineering problems
Discussion
Conclusions
References
A data-driven model for fire safety strategies assessment using artificial neural networks and genetic algorithms
Introduction
Methodology
Development of ANN-based prediction model
Optimization using multiobjective-based genetic algorithms
Results and discussions
Investigation of fire safety predictors
Artificial neural network and genetic algorithm
Conclusions
Acknowledgments
References
Application of artificial neural networks in polymer electrolyte membrane fuel cell system prognostics
Introduction
Description of fuel cell test bench and experimental data
A hybrid approach for PEMFC prognosis
Effectiveness evaluation of control parameters with BPNN
Effectiveness evaluation of historical state with ANFIS
Proposed hybrid approach
Effectiveness of proposed hybrid approach in PEMFC predictions
Effectiveness of the proposed hybrid approach at static operating condition
Effectiveness of proposed hybrid approach at Quasistatic operating condition
Input parameter optimization using correlation-based analysis
Correlation-based analysis
Effectiveness of correlation-based analysis in PEMFC prognosis
Conclusion
References
Reliability redundancy allocation problems under fuzziness using genetic algorithm and dual-connection numbers
Introduction
Prerequisite mathematics
Problem formulation: Reliability redundancy allocation problem (RRAP)
Notations
Constraint satisfaction rule
Solution procedure: Genetic algorithm-based constrained handling approach
Numerical example
Concluding remarks
References
Index