Prediction of tool wear based on GA-BP neural network

Abstract

The anisotropy and nonuniformity of wood-plastic composites (WPCs) affect the milling tool, which rapidly wears during high-speed milling of WPCs. Thus, the evolution mechanism of tool failure becomes complicated, and the prediction of tool wear cannot be precisely described mathematically. A neural network based on tool wear test was proposed to predict the tool wear condition during high-speed milling of WPCs. The traditional backpropagation (BP) neural network easily falls into the local optimal solution. A genetic algorithm (GA-BP) neural network prediction model was established by using the GA to optimise its initial weight and threshold. The BP model and the GA-BP model were evaluated in terms of mean square error and training times, and the generalisation verification was applied to the prediction model. After analysing and comparing the results of the two models, the GA-BP neural network model has better training speed and accuracy under the test conditions. The relative error between the predicted value and the actual value is controlled within 5%.

Distributed deep learning enabled prediction on cutting tool wear and remaining useful life

“…An approach for tool wear prediction was designed based on a genetic algorithm and a backpropagation (BP) neural network [8]. An approach integrating a CNN model and a stacked bi-directional/ unidirectional LSTM model was designed to predict the wear condition and RUL of a cutting tool [9]. In this approach, the CNN model was designed to extract features from monitoring signals for dimensionality reduction, and the LSTM model was applied to train these features to achieve tool wear and RUL prediction.…”

“…8 An approach integrating a CNN model and a stacked bi-directional/ unidirectional LSTM model was designed to predict the wear condition and RUL of a cutting tool. 9 In this approach, the CNN model was designed to extract features from monitoring signals for dimensionality reduction, and the LSTM model was applied to train these features to achieve tool wear and RUL prediction. A data-based approach to predict cutting tool wear and RUL for micro-milling was developed.…”

Development of tool wear condition on-line monitoring method for impeller milling based on new data processing approach and DAE-BP-ANN-integrated modeling

“…Nouioua and Bouhalais 24 decomposed the vibration signal collected in TNMG carbide insert turning process into 21 intrinsic mode functions by using complete ensemble empirical mode decomposition with adaptive noise, and corresponding root mean square values and spectral centroid indicator values were used as inputs to an artificial neural network to predict tool wear condition. To reduce the effect of traditional backpropagation (BP) neural network that tend to fall into local optimal solution, Wei et al 25 proposed a genetic algorithm (GA-BP) neural network model to predict tool wear condition with the mean square error (MSE) and the training times being used to evaluate the model. Results showed that the model had better training speed and lower error which was controlled within 5%.…”

The multi-sensor-based measurement of machining signals and data fusion to develop predictive tool wear models for TiAlN-PVD coated carbide inserts during end milling of Inconel 617

“…Zhang et al 22 proposed a generic wear model with adjustable coefficients by dividing the tool life into three wear zones and using a genetic algorithm for real-time monitoring of tool life and wear. Wei et al 23 developed genetic algorithm backpropagation (GA-BP) neural network and compared with the traditional backpropagation (BP) found that GA-BP has better training speed and accuracy based on mean square error and training times. Wu et al 24 proposed deep learning based tool wear prediction during machining.…”