Optimization of laser processing parameters through automated data acquisition and artificial neural networks

Objectives: To analyze the critical postures of the CNC milling machine operators by RULA (Rapid Upper Limb Assessment) scores and develop an ANN (Artificial Neural Network) prediction model. Methods: The methodology includes a postural analysis of 40 male CNC milling machine operators across Bangladesh, employing both manual (using manual RULA assessment worksheet) and digital (using CATIA V5R21 software) RULA methods complemented by an ANN prediction model. Finally, Digital RULA scores through DHM (Digital Human Modeling) and ANN predicted RULA scores would be compared. Findings: Digital RULA analysis reveals that lifting, carrying, and positioning are the most crucial ergonomic postures, and the most prominent high-risk category limbs are wrist and arm. The overall initial RULA score for lifting, carrying, and positioning are 7, 6, and 7, respectively, and reduced to 3, 3 and 4 respectively for ergonomically designed posture. The ANN model, structured with input, hidden, and output layers of 7, 10, and 1 nodes, significantly refines ergonomic risk prediction by aligning predicted scores closely with actual outcomes during the first stage, emphasized for training. It demonstrates a perfect correlation (R=1) in training, testing, validation, and overall performance for using manual RULA scores. The model's accuracy is further evidenced by minimal prediction offsets across all datasets for digital RULA score in the second stage, with correlation coefficients of 0.87003 (training), 0.93676 (validation), 0.89113 (testing), and (0.88395) for overall. This study contributes significant advancements in ergonomic risk assessment, highlighting the adoption of improved postures to reduce musculoskeletal disorders. Novelty: Employing both manual and DHM methods for RULA score calculation combined with ANN model, which can predict postural risk as floating number and fit a wider range of parameters. Keywords: ANN, CNC, Digital Human Modeling (DHM), Ergonomics, RULA

This study deals with artificial neural network (ANN) modelling of a fuel-staged gas turbine combustion rig to predict exhaust emissions and combustor outlet temperature. The data used for ANN training and testing was acquired from a natural gas burning test rig at various operating conditions. The ANN model uses a multi-layer feed-forward network architecture and was trained with experimental data using backpropagation. The ANN has 8 input neurons, 27 neurons in the hidden layer and 3 output neurons. The ANN model can predict the experimental results quite well with correlation coefficients in the range of 0.96 to 0.99. The ANN model was then used to create performance maps (response surfaces) and to predict two operating conditions that optimize the conflicting criteria of low emissions and high gas outlet temperature. The ANN predicted optimum operating conditions yielded NOx emissions below 20 ppm corrected to 15% O2 and a combustor outlet temperature of 1838 K. These optimum operating conditions were then experimental validated. The experimental validation showed that the first ANN predicted optimum operating condition was poorly predicted: the difference between the ANN predicted equivalence ratio and that of the experimental validation data for the optimum point was over 0.1. The second ANN predicted optimum operating condition was accurately predicted within the experimental uncertainty of the measurements. The difference in validity between the ANN predictions can be attributed to the sparsity of the data: for the first optimum operating condition there were 28 highly clustered training points, while the second optimum operating condition had 40 points that spanned a greater operating region.

Modeling of surface roughness in wire electrical discharge machining of Inconel 718 using artificial neural network

In this paper, we study a new model that represents the symmetric connection between capacitance–voltage and Schottky diode. This model has a symmetrical shape towards the horizontal direction. In recent times, works conducted on artificial neural network structure, which is one of the greatest actual artificial intelligence apparatuses used in various fields, stated that artificial neural networks are apparatuses that proposal very high forecast performance by the side of conventional structures. In the current investigation, an artificial neural network structure has been generated to guess the capacitance voltage productions of the Schottky diode with organic polymer edge, contingent on the frequency with a symmetrical shape. Of the dataset, 130 were grouped for training, 28 for validation, and 28 for testing. In order to evaluate the effect of the number of neurons on the prediction accuracy, three different models with different neuron numbers have been developed. This study, in which an artificial neural network model, although well-trained, could not predict the output values correctly, is a first in the literature. With this aspect, the study can be considered as a pioneering study that brings a novelty to the literature.

This paper proposes a technique to improve Artificial Neural Network (ANN) prediction accuracy using Particle Swarm Optimization (PSO) combiner. A hybrid system consists of two stages with the first stage containing two ANNs. The first ANN predictor is a multi-layer feed-forward network trained with error back-propagation and the second predictor is a functional link network. These two predictors are combined in the second stage using PSO combiner in a linear and non-linear fashion. The proposed method is applied to problem of predicting daily natural gas consumption. The performance of ANN predictors and combination methods are tested on real data from four different gas utilities. The experimental results show that the proposed particle swarm optimization combiners results in more accurate prediction compared to using single ANN predictor. Prediction accuracy improvement of the proposed PSO combiners have been shown using hypothesis testing.

Several researchers have investigated the relationships among different physical attributes of the fruits. For proper design and operation of grading systems, important relationships among the mass and other properties of fruits such as length, width, thickness, arithmetic mean diameter, geometric mean diameter, sphericity, surface area, volume, projected area, shape index, aspect ratio and elongations must be known. Recent researches have focused on artificial neural network (ANN) approaches to predict hard-to-find attributes of the fruits from easily-determined and readily available values. In this study, Modular Neural Network (MNN) and Radial Basis Neural Network (RBNN) structures of Artificial Neural Network (ANN) were employed to predict walnut mass from the physical attributes of the walnuts. Root mean square errors (RMSE) of MNN structure ranged from 0.60 to 0.89, while RMSE of RBNN structure were found to be very low (0.0002) in all of walnut varieties. These results showed that RBNN structures of Artificial Neural Network could potentially be used to estimate mass of walnuts and various physical attributes of walnuts were sufficient to predict the mass characteristics of a walnut.

Thanks to their convex formulation, kernel regressions have shown an improved accuracy with respect to artificial neural network (ANN) structures in regression problems where a reduced set of training samples are available. However, despite the above interesting features, kernel regressions are inherently less flexible than ANN structures since their implementations are usually limited to scalar-output regression problems. This article presents a vector-valued (multioutput) formulation of the kernel ridge regression (KRR) aimed at bridging the gap between multioutput ANN structures and scalar kernel-based approaches. The proposed vector-valued KRR relies on a generalized definition of the reproducing kernel Hilbert space (RKHS) and on a new multioutput kernel structure. The mathematical background of the proposed vector-valued formulation is extensively discussed together with different matrix kernel functions and training schemes. Moreover, a compression strategy based on the Nystrom approximation is presented to reduce the computational complexity of the model training. The effectiveness and the performance of the proposed vector-valued KRR are discussed on an illustrative example consisting of a high-speed link and on the optimization of a Doherty amplifier.

Application of neural networks in multiphase flow through porous media: Predicting capillary pressure and relative permeability curves

An artificial neural network (ANN) is an information processing modelling of the human brain inspired by the way biological nervous systems behave. There are about 100 billion neurons in the human brain. Each neuron has a connection point between 1,000 and 100,000. The key element of this paradigm is the novel structure of the information processing system. In the human brain, information is stored in such a way as to be distributed, and we can extract more than one piece of this information when necessary from our memory in parallel. We are not mistaken when we say that a human brain is made up of thousands of very powerful parallel processors. It is composed of a large number of highly interconnected processing elements (neurons) working in union to solve specific problems. ANN, like people, learns by example. The chapter includes characteristics of artificial neural networks, structure of ANN, elements of artificial neural networks, pros and cons of ANN.

Soil Coefficient of Consolidation (Cv) is a crucial mechanical parameter and used to characterize whether the soil undergoes consolidation or compaction when subjected to pressure. In order to define such a parameter, the experimental approaches are costly, time-consuming, and required appropriate equipment to perform the tests. In this study, the development of an alternative manner to estimate the Cv, based on Artificial Neural Network (ANN), was conducted. A database containing 188 tests was used to develop the ANN model. Two structures of ANN were considered, and the accuracy of each model was assessed using common statistical measurements such as the coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). In performing 600 simulations in each case, the ANN structure containing 14 neurons was statistically superior to the other one. Finally, a typical ANN result was presented to prove that it can be an excellent predictor of the problem, with a satisfying accuracy performance that yielded of RMSE = 0.0614, MAE = 0.0415, and R2 = 0.99727. This study might help in quick and accurate prediction of the Cv used in civil engineering problems.

In this study, sisal and coir fiber-based epoxy composites were fabricated using compression molding route. Optimization technique using Taguchi and grey relational analysis (GRA) was employed to find the best condition for tensile, flexural and impact properties. Sisal (0–20 wt%), coir fibers (0–20 wt%), NaOH treatment (0–8 wt%), compression pressure (10–16 MPa) and temperature (100–120 °C) are the different factors with various levels. Multi-response optimization using GRA showed significance of both the fibers in enhancing mechanical properties. The response table of GRA concluded that combination of sisal 20%/coir 15%/NaOH treatment 5%/compression pressure 10 MPa and temperature 120 °C has the best mechanical results. SEM images indicated the advancement in adhesion of reinforcement and epoxy resin by hybridization of fibers. The artificial neural network (ANN) prediction states predicted and experimental values for various trials are almost same, and regression graph showed a straight line with R-value 97.321%. These fabricated composites can be used in aerospace and automobile industries for their interior parts.

Influence of random topology in artificial neural networks: A survey

Wind speed forecasting by subspace and nuclear norm optimization based algorithms

An artificial neural network (ANN) is a network designed with adaptation to a computer system. The developed computer system will perform functions oriented to the way the brain works (neuron concept). This study is an extension to the study of the suitability of ANN to be applied in numbers of applications, especially in the field of medical engineering. ANN has been widely being used in medicine, ANN is widely applied in education, research, and even decision making. In this study, ANN will be trained for pre-testing to predict the cardiac abnormalities symptom based on selected reference parameters. This reference parameter is better known as the input parameter to the ANN to detect cardiac abnormalities, among which are the of the height of peak/wave (amplitude) and time occurrence of peak/wave (duration of time) extracted from the electrocardiogram (ECG) signal. A complete ECG complex contains a P peak, a QRS wave, and a T peak. For each P peak, QRS wave, and T peak, amplitude height and duration will be measured to serve as input parameters. This makes six parameters defined as inputs to the ANN. This study has used a Multilayer Perceptron (MLP) network as ANN structure by being trained using three different training algorithms namely Backpropagation (BP), Lavenberg Marquardt (LM) and Bayesian Regularization (BR). At the end of the study, it showed the MLP network which by BR training algorithm gave the highest accuracy prediction (94.04%), followed by LM (92.95%) and BP (88.77%). In this study all MLP networks were activated using the Tansig activation function.

Revealing prediction of perched cum off-centered wick solar still performance using network based on optimizer algorithm