Hybridization of metaheuristic algorithm in training radial basis function with dynamic decay adjustment for condition monitoring / Chong Hue Yee

Chong , Hue Yee (2023) Hybridization of metaheuristic algorithm in training radial basis function with dynamic decay adjustment for condition monitoring / Chong Hue Yee. PhD thesis, Universiti Malaya.

	PDF (The Candidate's Agreement) Restricted to Repository staff only Download (175Kb)
	PDF (Thesis PhD) Download (2787Kb)

Abstract

Condition monitoring is a major component in predictive maintenance which monitors the condition and identifies significant changes of the machinery parameter to perform early detection and prevent equipment defects that could cause unplanned downtime or incur unnecessary expenditures. An effective condition monitoring model is helpful to reduce the frequency of unexpected breakdown incidents and thus, facilitate in maintenance. Artificial neural network (ANN) has shown effective in various condition monitoring and fault detection applications. ANN is popular due to its capability of identifying the complex nonlinear relationships among features in a large dataset and hence, it can perform with an accurate prediction. However, a drawback is that the performance of ANN is sensitive to the parameters (i.e., number of hidden neurons and the initial values of connection weights) in its architecture where the settings of these parameters are subject to tuning on a trial-and-error basis. Hence, a wide range of studies has been focused on determining the optimal weight values of ANN models and the number of hidden neurons. In this research work, the motivation is to develop an autonomous learning model based on the hybridization of an adaptive ANN and a metaheuristic algorithm for optimizing ANN parameters so that the network could perform learning and adaptation in a more flexible way and handle condition classification tasks more accurately in industries, such as in power systems. Different types of metaheuristic optimization tools have their unique features which vary from others and hence lead to different suitability in the particular application. Two metaheuristic algorithms i.e., Harmony Search (HS) and Gravitational Search Algorithm (GSA), are selected to integrate separately with a Radial Basis Function Network with Dynamic Decay Adjustment (RBFN-DDA) to perform condition monitoring in industrial processes. RBFN-DDA performs incremental learning wherein its structure expands by adding new hidden units to include new information. As such, its training can reach stability in a shorter time compared to the gradient-descent based methods. To achieve optimal RBFN-DDA performance, HS (or GSA) is proposed to optimize the center and the width of each hidden unit in a trained RBFN. By integrating with the HS (or GSA) algorithm, the proposed metaheuristic neural networks (i.e., RBFN-DDA-HS and RBFN-DDA-GSA) can optimize the RBFN-DDA parameters and improve classification performances from the original RBFN-DDA up to 28.69% in two benchmarks datasets, which are numerical records from a bearing and steel plate system and a condition-monitoring system in a power plant (i.e., the circulating water (CW) system). The results also show that the proposed RBFN-DDA-HS and RBFN-DDA-GSA are compatible, if not better than, the classification performances of other state-of-art machine learning methods.

Actions (For repository staff only : Login required)