Modern Optimization Algorithms and Particle Swarm Variations.pdf (original) (raw)

2013, 4th Inverse Problems, Design and Optimization Symposium (IPDO-2013)

Many designs and inverse problems can be formulated as optimization problems. Due to the cost of evaluating real-world objective functions, optimization algorithms must be both fast and robust. While Particle Swarm (PS) is one of the fastest and most robust optimization algorithms ever developed, it cannot solve all problems and is poorly suited in many cases. This paper proposes a modification to PS and investigates five algorithms similar to PS: Quantum Particle Swarm (QPS), Modified Quantum Particle Swarm (MQP), Firefly Algorithm (FA), Cuckoo Search (CKO), and Bat-Inspired Metaheuristic Algorithm (BAT). Their performance is compared to standard PS using a subset of the Schittkowski & Hock's test cases. The modified PS is observed to outperform standard PS in twenty-seven percent of the test cases. QPS and MQP similarly outperform PS in roughly twenty-nine percent of test cases, while BAT and FA outperform PS in twenty-six percent of test cases.

Sign up for access to the world's latest research.

checkGet notified about relevant papers

checkSave papers to use in your research

checkJoin the discussion with peers

checkTrack your impact