Pipeline damage identification based on an optimized back-propagation neural network improved by whale optimization algorithm

With the advantages of high economy and large transportation capacity, pipeline transportation is commonly used in industrial production. Pipeline damage induced by various factors will result in changes of physical properties, further leading to changes of dynamic parameters such as natural frequen...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Applied intelligence (Dordrecht, Netherlands) Ročník 53; číslo 10; s. 12937 - 12954
Hlavní autori: Wu, Lei, Mei, Jiangtao, Zhao, Shuo
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: New York Springer US 01.05.2023
Springer Nature B.V
Predmet:
Algorithms
Artificial Intelligence
Artificial neural networks
Back propagation
Back propagation networks
Computer Science
Damage assessment
Damage detection
Identification methods
Machines
Manufacturing
Mechanical Engineering
Neural networks
Optimization
Optimization algorithms
Physical properties
Processes
Resonant frequencies
Swarm intelligence
Transportation
Vibration mode
Back-propagation neural network
Whale optimization algorithm
Damage location and degree
Pipeline damage identification
ISSN:0924-669X, 1573-7497
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:With the advantages of high economy and large transportation capacity, pipeline transportation is commonly used in industrial production. Pipeline damage induced by various factors will result in changes of physical properties, further leading to changes of dynamic parameters such as natural frequency and vibration mode. Recently, as a new type of tool, artificial intelligence is widely used for pipeline damage identification. In this study, to promote the accuracy of pipeline damage identification, a novel method that employs the artificial neural network (ANN) and swarm intelligence algorithm is proposed. In detail, based on the original whale optimization algorithm (WOA), an improved WOA (IWOA) is presented in which an adaptive coefficient strategy and a stochastic optimal substitution strategy are introduced. Then, the IWOA and back-propagation neural network (BPNN) are hybridized into IWOA-BPNN. Subsequently, a damage location detector and a damage degree detector are established based on the proposed IWOA-BPNN. By taking a pipeline fixed at both ends and its curvature and displacement modes, the proposed damage identification method is verified to confirm its effectiveness and accuracy in different damage states. Experimental results demonstrate that the comprehensive performance of IWOA-BPNN is better than other compared models. The relative error of the predicted results obtained by IWOA-BPNN is less than 2.2% when evaluating the damage location and degree for 12 randomly selected test samples, indicating the superiority of the proposed method. The proposed method has broad application prospects in modern industries.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0924-669X
1573-7497
DOI:10.1007/s10489-022-04188-7