The current work presents an optimized neural network for predicting depth of cut (DoC) in the abrasive water jet machining (AWJM) of Titanium Ti6Al4V Grade 5 material. Five process parameters were used for experimentation i.e., water pressure (Wp), traverse speed (Ts), nozzle to orifice diameter (N/Odia), abrasive mass flow rate (Amf), and abrasive orifice size (Aos). Experiments were carried out using the Taguchi based L27 Orthogonal array, resulting in the development of a regression equation describing the process behaviour. The DoC was modelled using two neural network architectures: a single hidden layer network (NN1), where neurons were varied from 1 to 10, and a deep neural network (NND), where both neurons and hidden layers were varied to determine the optimum network configuration. Four different activation functions, namely, Sigmoidal, Gaussian, Tanh, and Linear functions were employed to perform the optimization. The dataset for training and testing the neural network models was generated using the regression equation based on experimental data, rather than direct experimental measurements. The optimization of both the neural network architectures was evaluated using two performance metrics, i.e., root mean squared error (RMSE) and co-efficient of determination (<InlineEquation ID="IEq1"><EquationSource Format="TEX">\(\:{R}^{2}\)</EquationSource></InlineEquation>). The results concluded that both the neural network (NN) models were able to accurately predict DoC, however, the NND was found to be better in terms of accuracy with a 12.5% reduction in testing RMSE and 4.2% improvement in testing (<InlineEquation ID="IEq2"><EquationSource Format="TEX">\(\:{R}^{2}\)</EquationSource></InlineEquation>). Amongst the activation functions used, the Gaussian functions outperformed the other functions across both neural network architectures, as its symmetric response reflects the jet energy distribution peaking at the centre and decaying radially, thereby effectively capturing the non-linear penetration behavior. SEM analysis supported the computational results, showing micro-cutting with stable grooves for machined surfaces with higher DoC, whereas lower DoC experienced micro-ploughing and ridge formation. Contour plots further revealed that DoC in AWJM of Ti6Al4V is primarily governed by water pressure and traverse speed, thereby confirming the capability of the optimised NND architecture to capture the underlying physical mechanisms of AWJM.

Modelling and computational optimization of different neural network architectures for prediction of depth of cut in abrasive water jet machining of Ti6Al4V

期刊论文
出版模式：

开放获取
发表日期：

2026年6月30日

查看全文

Yakub Iqbal Mogul,
Jaimon Dennis Quadros,
Ganesan Subramanian,
Prashanth Thalambeti,
Ibtisam Mogul,
Ma Mohin

摘要

The current work presents an optimized neural network for predicting depth of cut (DoC) in the abrasive water jet machining (AWJM) of Titanium Ti6Al4V Grade 5 material. Five process parameters were used for experimentation i.e., water pressure (W_p), traverse speed (T_s), nozzle to orifice diameter (N/O_dia), abrasive mass flow rate (A_mf), and abrasive orifice size (A_os). Experiments were carried out using the Taguchi based L₂₇ Orthogonal array, resulting in the development of a regression equation describing the process behaviour. The DoC was modelled using two neural network architectures: a single hidden layer network (NN₁), where neurons were varied from 1 to 10, and a deep neural network (NN_D), where both neurons and hidden layers were varied to determine the optimum network configuration. Four different activation functions, namely, Sigmoidal, Gaussian, Tanh, and Linear functions were employed to perform the optimization. The dataset for training and testing the neural network models was generated using the regression equation based on experimental data, rather than direct experimental measurements. The optimization of both the neural network architectures was evaluated using two performance metrics, i.e., root mean squared error (RMSE) and co-efficient of determination (\(\:{R}^{2}\)). The results concluded that both the neural network (NN) models were able to accurately predict DoC, however, the NN_D was found to be better in terms of accuracy with a 12.5% reduction in testing RMSE and 4.2% improvement in testing (\(\:{R}^{2}\)). Amongst the activation functions used, the Gaussian functions outperformed the other functions across both neural network architectures, as its symmetric response reflects the jet energy distribution peaking at the centre and decaying radially, thereby effectively capturing the non-linear penetration behavior. SEM analysis supported the computational results, showing micro-cutting with stable grooves for machined surfaces with higher DoC, whereas lower DoC experienced micro-ploughing and ridge formation. Contour plots further revealed that DoC in AWJM of Ti6Al4V is primarily governed by water pressure and traverse speed, thereby confirming the capability of the optimised NN_D architecture to capture the underlying physical mechanisms of AWJM.