Study of bio-inspired neural networks for the prediction of liquid flow in a process control system
Küçük Resim Yok
Tarih
2021
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Elsevier
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
This research investigates the use of a bio-inspired algorithm (BIA) in preparing an artificial neural network in the field of a nonlinear liquid flow model. To improve the expanding multifaceted nature and operational productivity of fluid stream measures, plans that can improve model recognizable proof of exceptionally nonlinear frameworks are required. As a contextual investigation, the WFT-20-I measure control arrangement for flow rate measurement and control issue is discussed. The key issue is to locate a specific design for an artificial neural network (ANN) that best models the fluid stream measure. The authors propose the utilization of BIAs with particle swarm improvement using particle swarm optimization and firefly calculation using a firefly algorithm as systems to upgrade the synaptic weights of ANN. All these algorithms were used to improvise the weight of an ANN so that it poduces local optimum of objective function of a liquid flow rate. The proposed model was formulated with broad tests & factual investingation to show the appropriateness of the whole demonstration. The outcomes were investigated utilizing the root mean square error exactness, and other measures to evaluate the level of identification performance of the liquid flow contextual analysis model. The trustworthiness of the present models was compared with earlier models for similar subsystems utilizing competitive intelligent methodologies. The outcomes of the present model show that the proposed approach gives satisfactory output and outperforms its rivals. © 2022 Elsevier Inc. All rights reserved.
Açıklama
Anahtar Kelimeler
Firefly Algorithm; Liquid Flow Process; Modeling; Neural Network; Particle Swarm Optimization
Kaynak
Cognitive Big Data Intelligence with a Metaheuristic Approach
WoS Q Değeri
Scopus Q Değeri
N/A
