ARTIFICIAL NEURAL NETWORK AND FUZZY LOGIC CONTROLLER FOR GTAW MODELING AND CONTROL

An artificial neural network(ANN) and a self-adjusting fuzzy logiccontroller(FLC) for modeling and control of gas tungsten arc welding(GTAW) process are presented.The discussion is mainly focused on the modeling and control of the weld pool depth with ANN and theintelligent control for weld seam tracking with FLC. The proposed neural network can produce highlycomplex nonlinear multi-variable model of the GTAW process that offers the accurate prediction ofwelding penetration depth. A self-adjusting fuzzy controller used for seam tracking adjusts thecontrol parameters on-line automatically according to the tracking errors so that the torch positioncan be controlled accurately.

Comparative Analysis between Conventional PI, Fuzzy Logic and Artificial Neural Network Based Speed Controllers of Induction Motor with Considering Core Loss and Stray Load Loss

Most of the controllers of IM （induction motor） for industrial applications have been designed based on PI controller without consideration of CL （core loss） and SLL （stray load loss）. To get the precise performances of torque as well as rotor speed and flux, the above mentioned losses should be considered. Conventional PI controller has overshoot effect at the transient period of the speed response curve. On the other hand, fuzzy logic and ANN （artificial neural network） based controllers can minimize the overshoot effect at the transient period because they have the abilities to deal with the nonlinear systems. In this paper, a comparative analysis is done between PI, fuzzy logic and ANN based speed controllers to find the suitable control strategy for IM with consideration of CL and SLL. The simulation analysis is done by using Matlab/Simulink software. The simulation results show that the fuzzy logic based speed controller gives better responses than ANN and conventional PI based speed controllers in terms of rotor speed, electromagnetic torque and rotor flux of IM.

Speed Control of Travelling Wave Type Ultrasonic Motors Using Artificial Neural Network

Ultrasonic motor (USM) is a newly developed motor, and it has some excellent performances and useful features, therefore, it has been expected to be of practical use. However, the driving principle of USM is different from that of other electromagnetic type motors, and the mathematical model is complex to apply to motor control. Furthermore, the speed characteristics of the motor have heavy nonlinearity and vary with driving conditions. Hence, the precise speed control of USM is generally difficult. This paper proposes a new speed control scheme for USM using an artificial neural network. An accurate tracking response can be obtained by random initialization of the weights of the network owing to the powerful on line learning capability. Two prototype ultrasonic motors of travelling wave type were fabricated, both having 100 mm outer diameters of stator and piezoelectric ceramic. The usefulness and validity of the proposed control scheme are examined in experiments.

Simultaneous Identification of Thermophysical Properties of Semitransparent Media Using a Hybrid Model Based on Artificial Neural Network and Evolutionary Algorithm

A hybrid identification model based on multilayer artificial neural networks(ANNs) and particle swarm optimization(PSO) algorithm is developed to improve the simultaneous identification efficiency of thermal conductivity and effective absorption coefficient of semitransparent materials.For the direct model,the spherical harmonic method and the finite volume method are used to solve the coupled conduction-radiation heat transfer problem in an absorbing,emitting,and non-scattering 2D axisymmetric gray medium in the background of laser flash method.For the identification part,firstly,the temperature field and the incident radiation field in different positions are chosen as observables.Then,a traditional identification model based on PSO algorithm is established.Finally,multilayer ANNs are built to fit and replace the direct model in the traditional identification model to speed up the identification process.The results show that compared with the traditional identification model,the time cost of the hybrid identification model is reduced by about 1 000 times.Besides,the hybrid identification model remains a high level of accuracy even with measurement errors.

Application of Random Search Methods in the Determination of Learning Rate for Training Container Dwell Time Data Using Artificial Neural Networks

Purpose: This study aimed to enhance the prediction of container dwell time, a crucial factor for optimizing port operations, resource allocation, and supply chain efficiency. Determining an optimal learning rate for training Artificial Neural Networks (ANNs) has remained a challenging task due to the diverse sizes, complexity, and types of data involved. Design/Method/Approach: This research used a RandomizedSearchCV algorithm, a random search approach, to bridge this knowledge gap. The algorithm was applied to container dwell time data from the TOS system of the Port of Tema, which included 307,594 container records from 2014 to 2022. Findings: The RandomizedSearchCV method outperformed standard training methods both in terms of reducing training time and improving prediction accuracy, highlighting the significant role of the constant learning rate as a hyperparameter. Research Limitations and Implications: Although the study provides promising outcomes, the results are limited to the data extracted from the Port of Tema and may differ in other contexts. Further research is needed to generalize these findings across various port systems. Originality/Value: This research underscores the potential of RandomizedSearchCV as a valuable tool for optimizing ANN training in container dwell time prediction. It also accentuates the significance of automated learning rate selection, offering novel insights into the optimization of container dwell time prediction, with implications for improving port efficiency and supply chain operations.

Study on optimization control method based on artificial neural network

In the goal optimization and control optimization process the problems with common artificial neural network algorithm are unsure convergence, insufficient post-training network precision, and slow training speed, in which partial minimum value question tends to occur. This paper conducted an in-depth study on the causes of the limi-tations of the algorithm, presented a rapid artificial neural network algorithm, which is characterized by integrating multiple algorithms and by using their complementary advan-tages. The salient feature of the method is self-organization, which can effectively prevent the optimized results from tending to be partial minimum values. Overall optimization can be achieved with this method, goal function can be searched for in overall scope. With op-timization control of coal mine ventilator as a practical application, the paper proves that by integrating multiple artificial neural network algorithms, best control optimization and goal optimized can be achieved.

A Review: Artificial Neural Networks as Tool for Control Food Industry Process

In the last year, interest in using Artificial Neural networks as a modeling tool in food technology is increasing because they have found extensive utilization in solving many complex real world problems. Due to this and as previous step at development of some project, this paper intends to introduce the reader inside neural networks: general characteristics of the ANN, their architectures, their rules of learning, types of networks and ANN’s create process. Also this paper presents a comprehensive review of food industrial applications of artificial neural networks in the last year. ANN industrial applications are grouped and tabulated by their main functions and what they actually performed on the referenced papers with except the applications in the olive oil industry that are described with special emphasis.

An Efficient and Robust Fall Detection System Using Wireless Gait Analysis Sensor with Artificial Neural Network (ANN) and Support Vector Machine (SVM) Algorithms

In this work, a total of 322 tests were taken on young volunteers by performing 10 different falls, 6 different Activities of Daily Living (ADL) and 7 Dynamic Gait Index (DGI) tests using a custom-designed Wireless Gait Analysis Sensor (WGAS). In order to perform automatic fall detection, we used Back Propagation Artificial Neural Network (BP-ANN) and Support Vector Machine (SVM) based on the 6 features extracted from the raw data. The WGAS, which includes a tri-axial accelerometer, 2 gyroscopes, and a MSP430 microcontroller, is worn by the subjects at either T4 (at back) or as a belt-clip in front of the waist during the various tests. The raw data is wirelessly transmitted from the WGAS to a near-by PC for real-time fall classification. The BP ANN is optimized by varying the training, testing and validation data sets and training the network with different learning schemes. SVM is optimized by using three different kernels and selecting the kernel for best classification rate. The overall accuracy of BP ANN is obtained as 98.20% with LM and RPROP training from the T4 data, while from the data taken at the belt, we achieved 98.70% with LM and SCG learning. The overall accuracy using SVM was 98.80% and 98.71% with RBF kernel from the T4 and belt position data, respectively.

Implementation of Radial Basis Function Artificial Neural Network into an Adaptive Equivalent Consumption Minimization Strategy for Optimized Control of a Hybrid Electric Vehicle

Continued increases in the emission of greenhouse gases by passenger vehicles have accelerated the production of hybrid electric vehicles. With this increase in production, there has been a parallel demand for continuously improving strategies of hybrid electric vehicle control. The goal of an ideal control strategy is to maximize fuel economy while minimizing emissions. Methods exist by which the globally optimal control strategy may be found. However, these methods are not applicable in real-world driving applications since these methods require a priori knowledge of the upcoming drive cycle. Real-time control strategies use the global optimal as a benchmark against which performance can be evaluated. The goal of this work is to use a previously defined strategy that has been shown to closely approximate the global optimal and implement a radial basis function (RBF) artificial neural network (ANN) that dynamically adapts the strategy based on past driving conditions. The strategy used is the Equivalent Consumption Minimization Strategy (ECMS), which uses an equivalence factor to define the control strategy and the power train component torque split. An equivalence factor that is optimal for a single drive cycle can be found offline with a priori knowledge of the drive cycle. The RBF-ANN is used to dynamically update the equivalence factor by examining a past time window of driving characteristics. A total of 30 sets of training data (drive cycles) are used to train the RBF-ANN. For the majority of drive cycles examined, the RBF-ANN implementation is shown to produce fuel economy values that are within ±2.5% of the fuel economy obtained with the optimal equivalence factor. The advantage of the RBF-ANN is that it does not require a priori drive cycle knowledge and is able to be implemented in real-time while meeting or exceeding the performance of the optimal ECMS. Recommendations are made on how the RBF-ANN could be improved to produce better results across a greater array of driving conditions.

NONLINEAR MODELING AND CONTROLLING OF ARTIFICIAL MUSCLE SYSTEM USING NEURAL NETWORKS

The pneumatic artificial muscles are widely used in the fields of medicalrobots, etc. Neural networks are applied to modeling and controlling of artificial muscle system. Asingle-joint artificial muscle test system is designed. The recursive prediction error (RPE)algorithm which yields faster convergence than back propagation (BP) algorithm is applied to trainthe neural networks. The realization of RPE algorithm is given. The difference of modeling ofartificial muscles using neural networks with different input nodes and different hidden layer nodesis discussed. On this basis the nonlinear control scheme using neural networks for artificialmuscle system has been introduced. The experimental results show that the nonlinear control schemeyields faster response and higher control accuracy than the traditional linear control scheme.

Optimization of the Conceptual Model of Green-Ampt Using Artificial Neural Network Model (ANN) and WMS to Estimate Infiltration Rate of Soil (Case Study: Kakasharaf Watershed, Khorram Abad, Iran)

Determination of the infiltration rate in a watershed is not easy and in empirical and theoretical point of view, it is important to access average value of infiltration. Infiltration models has main role in managing water sources. Therefore different types of models with various degrees of complexity were developed to reach this aim. Most of the estimating methods of soil infiltration are expensive and time consuming and these methods estimate infiltration with hypothesis of zero slope. One of the conceptual and physical models for estimating soil infiltration is Green-Ampt model which is similar to Richard model. This model uses slope factor in estimating infiltration and this is the power point of Green-Ampt model. In this research the empirical model of Green-Ampt was optimized with integrating artificial neural network model (ANN) and a model of geographical information system WMS to estimate the infiltration in Kakasharaf watershed. Results of the comparison between the output of this method and real value of infiltration in region (through multiple cylinders) showed that this method can estimate the infiltration rate of Kakasharaf watershed with low error and acceptable accuracy (Nash-Sutcliff performance coefficient 0.821, square error 0.216, correlation coefficient 0.905 and model error 0.024).

Predicting pollutant removal in constructed wetlands using artificial neural networks(ANNs)

Growth in urban population,urbanisation,and economic development has increased the demand for water,especially in water-scarce regions.Therefore,sustainable approaches to water management are needed to cope with the effects of the urbanisation on the water environment.This study aimed to design novel configurations of tidal-flow vertical subsurface flow constructed wetlands(VFCWs)for treating urban stormwater.A series of laboratory experiments were conducted with semi-synthetic influent stormwater to examine the effects of the design and operation variables on the performance of the VFCWs and to identify optimal design and operational strategies,as well as maintenance requirements.The results show that the VFCWs can significantly reduce pollutants in urban stormwater,and that pollutant removal was related to specific VFCW designs.Models based on the artificial neural network(ANN)method were built using inputs derived from data exploratory techniques,such as analysis of variance(ANOVA)and principal component analysis(PCA).It was found that PCA reduced the dimensionality of input variables obtained from different experimental design conditions.The results show a satisfactory generalisation for predicting nitrogen and phosphorus removal with fewer variable inputs,indicating that monitoring costs and time can be reduced.

Application of Artificial Neural Networks Based Monte Carlo Simulation in the Expert System Design and Control of Crude Oil Distillation Column of a Nigerian Refinery

This research work investigated comparative studies of expert system design and control of crude oil distillation column (CODC) using artificial neural networks based Monte Carlo (ANNBMC) simulation of random processes and artificial neural networks (ANN) model which were validated using experimental data obtained from functioning crude oil distillation column of Port-Harcourt Refinery, Nigeria by MATLAB computer program. Ninety percent (90%) of the experimental data sets were used for training while ten percent (10%) were used for testing the networks. The maximum relative errors between the experimental and calculated data obtained from the output variables of the neural network for CODC design were 1.98 error % and 0.57 error % when ANN only and ANNBMC were used respectively while their respective values for the maximum relative error were 0.346 error % and 0.124 error % when they were used for the controller prediction. Larger number of iteration steps of below 2500 and 5000 were required to achieve convergence of less than 10-7?for the training error using ANNBMC for both the design of the CODC and controller respectively while less than 400 and 700 iteration steps were needed to achieve convergence of 10-4?using ANN only. The linear regression analysis performed revealed the minimum and maximum prediction accuracies to be 80.65% and 98.79%;and 98.38% and 99.98% when ANN and ANNBMC were used for the CODC design respectively. Also, the minimum and maximum prediction accuracies were 92.83% and 99.34%;and 98.89% and 99.71% when ANN and ANNBMC were used for the CODC controller respectively as both methodologies have excellent predictions. Hence, artificial neural networks based Monte Carlo simulation is an effective and better tool for the design and control of crude oil distillation column.

Artificial Neural Networks for Controlling the Temperature of Internally Cooled Turning Tools

By eliminating the need for externally applied coolant, internally cooled turning tools offer potential health, safety and cost benefits in many types of machining operation. As coolant flow is completely controlled, tool temperature measurement becomes a practical proposition and can be used to find and maintain the optimum machining conditions. This also requires an intelligent control system in the sense that it must be adaptable to different tool designs, work piece materials and machining conditions. In this paper, artificial neural networks (ANN) are assessed for their suitability to perform such a control function. Experimental data for both conventional tools used for dry machining and internally cooled tools is obtained and used to optimise the design of an ANN. A key finding is that both experimental scatter characteristic of turning and the range of machining conditions for which ANN control is required have a large effect on the optimum ANN design and the amount of data needed for its training. In this investigation, predictions of tool temperature with an optimised ANN were found to be within 5°C of measured values for operating temperatures of up to 258°C. It is therefore concluded that ANN’s are a viable option for in-process control of turning processes using internally controlled tools.

Predictive active control of building structures using LQR and artificial intelligence

This study presents a neural network-based model for predicting linear quadratic regulator(LQR)weighting matrices for achieving a target response reduction.Based on the expected weighting matrices,the LQR algorithm is used to determine the various responses of the structure.The responses are determined by numerically analyzing the governing equation of motion using the state-space approach.For training a neural network,four input parameters are considered:the time history of the ground motion,the percentage reduction in lateral displacement,lateral velocity,and lateral acceleration,Output parameters are LQR weighting matrices.To study the effectiveness of an LQR-based neural network(LQRNN),the actual percentage reduction in the responses obtained from using LQRNN is compared with the target percentage reductions.Furthermore,to investigate the efficacy of an active control system using LQRNN,the controlled responses of a system are compared to the corresponding uncontrolled responses.The trained neural network effectively predicts weighting parameters that can provide a percentage reduction in displacement,velocity,and acceleration close to the target percentage reduction.Based on the simulation study,it can be concluded that significant response reductions are observed in the active-controlled system using LQRNN.Moreover,the LQRNN algorithm can replace conventional LQR control with the use of an active control system.

EFFECT OF COLD WORKING ON THE AGING PROPERTIES OF Cu-Cr-Zr-Mg ALLOY BY ARTIFICIAL NEURAL NETWORK

A developmental research has been carried out to deal with the high performance of Cu-Cr-Zr-Mg lead frame alloy by artificial neural network (ANN). Using the cold working to assist in the aging hardening can improve the the hardness and electrical conductivity properties of Cu-Cr-Zr-Mg lead frame alloy. This paper studies the effect of different extent of cold working on the aging properties by a supervised ANN to model the non-linear relationship between processing parameters and the properties. The back-propagation (BP) training algorithm is improved by Levenberg-Marquardt algorithm. A basic repository on the domain knowledge of cold worked aging processes is established via sufficient data mining by the network. The predicted values of the ANN coincide well with the tested data. So an important foundation has been laid for prediction and optimum controlling the rolling and aging properties of Cu-Cr-Zr-Mg alloy.

Prediction of roadheaders' performance using artificial neural network approaches (MLP and KOSFM)

A pplication o f m echanical excavators is one o f th e m o st com m only used excavation m eth o d s because itcan bring th e p ro ject m ore productivity, accuracy and safety. A m ong th e m echanical excavators, roadhead ers are m echanical m iners w h ich have b een extensively u se d in tu n n elin g , m ining an d civil indu stries. Perform ance pred ictio n is an im p o rta n t issue for successful ro a d h e a d e r application andgenerally deals w ith m achine selection, p ro d u ctio n rate an d b it consu m p tio n . The m ain aim o f thisresearch is to investigate th e c u ttin g p erfo rm an ce （in stan tan eo u s c u ttin g rates （ICRs）） o f m ed iu m -d u tyro ad h ead ers by using artificial neural n etw o rk （ANN） approach. T here are d ifferent categories forANNs, b u t based o n train in g alg o rith m th e re are tw o m ain k in d s： supervised and u n su p erv ised . Them u lti-lay er p ercep tro n （MLP） an d K ohonen self-organizing feature m ap （KSOFM） are th e m o st w idelyused neu ral netw o rk s for supervised an d u n su p erv ised ones, respectively. For gaining this goal, ad atab ase w as prim arily provided from ro ad h e a d e rs＇ p erfo rm an ce an d geom echanical characteristics o frock form ations in tu n n els and d rift galleries in Tabas coal m ine, th e larg est an d th e only fullymech an ized coal m ine in Iran. T hen th e datab ase w as analyzed in o rd e r to yield th e m ost im p o rtan tfactor for ICR by using relatively im p o rta n t factor in w hich G arson eq u atio n w as utilized. The MLPn etw o rk w as train ed by 3 in p u t p ara m e te rs including rock m ass pro p erties, rock quality d esignation（RQD）, in tact rock p ro p erties such as uniaxial com pressive stre n g th （UCS） an d Brazilian ten sile stren g th（BTS）, and o n e o u tp u t p a ra m e te r （ICR）. In o rd e r to have m ore v alidation o n MLP o u tp u ts, KSOFM visualizationw as applied. The m ean square e rro r （MSE） an d regression coefficient （R ） o f MLP w e re found tobe 5.49 an d 0.97, respectively. M oreover, KSOFM n etw o rk has a m ap size o f 8 x 5 and final qu an tizatio nan d topographic erro rs w e re 0.383 an d 0.032, respectively. The results show th a t MLP neural n etw orkshave a strong capability to p red ict an d ev alu ate th e perfo rm an ce o f m ed iu m -d u ty ro ad h ead ers in coalm easu re rocks. Furtherm ore, it is concluded th a t KSOFM neural n etw o rk is an efficient w ay for u n d e rstand in g system beh av io r an d know ledge extraction. Finally, it is indicated th a t UCS has m ore influenceo n ICR b y applying th e b e st train ed MLP n etw o rk w eig h ts in G arson eq u atio n w h ich is also confirm ed byKSOFM.

A robust behavior of Feed Forward Back propagation algorithm of Artificial Neural Networks in the application of vertical electrical sounding data inversion

The applications of intelligent techniques have increased exponentially in recent days to study most of the non-linear parameters. In particular, the behavior of earth resembles the non- linearity applications. An efficient tool is needed for the interpretation of geophysical parameters to study the subsurface of the earth. Artificial Neural Networks （ANN） perform certain tasks if the structure of the network is modified accordingly for the purpose it has been used. The three most robust networks were taken and comparatively analyzed for their performance to choose the appropriate network. The single- layer feed-forward neural network with the back propagation algorithm is chosen as one of the well- suited networks after comparing the results. Initially, certain synthetic data sets of all three-layer curves have been taken tk^r training the network, and the network is validated by the field datasets collected from Tuticorin Coastal Region （78°7＇30＂E and 8°48＇45＂N）, Tamil Nadu, India. The interpretation has been done successfully using the corresponding learning algorithm in the present study. With proper training of back propagation networks, it tends to give the resistivity and thickness of the subsurface layer model of the field resistivity data concerning the synthetic data trained earlier in the appropriate network. The network is trained with more Vertical Electrical Sounding （VES） data, and this trained network is demon- strated by the field data. Groundwater table depth also has been modeled.

Prediction of TBM jamming risk in squeezing grounds using Bayesian and artificial neural networks

This study presents an application of artificial neural network(ANN)and Bayesian network(BN)for evaluation of jamming risk of the shielded tunnel boring machines(TBMs)in adverse ground conditions such as squeezing grounds.The analysis is based on database of tunneling cases by numerical modeling to evaluate the ground convergence and possibility of machine entrapment.The results of initial numerical analysis were verified in comparison with some case studies.A dataset was established by performing additional numerical modeling of various scenarios based on variation of the most critical parameters affecting shield jamming.This includes compressive strength and deformation modulus of rock mass,tunnel radius,shield length,shield thickness,in situ stresses,depth of over-excavation,and skin friction between shield and rock.Using the dataset,an ANN was trained to predict the contact pressures from a series of ground properties and machine parameters.Furthermore,the continuous and discretized BNs were used to analyze the risk of shield jamming.The results of these two different BN methods are compared to the field observations and summarized in this paper.The developed risk models can estimate the required thrust force in both cases.The BN models can also be used in the cases with incomplete geological and geomechanical properties.

Recovery of saturated signal waveform acquired from high-energy particles with artificial neural networks

Artificial neural networks(ANNs)are a core component of artificial intelligence and are frequently used in machine learning.In this report,we investigate the use of ANNs to recover the saturated signals acquired in highenergy particle and nuclear physics experiments.The inherent properties of the detector and hardware imply that particles with relatively high energies probably often generate saturated signals.Usually,these saturated signals are discarded during data processing,and therefore,some useful information is lost.Thus,it is worth restoring the saturated signals to their normal form.The mapping from a saturated signal waveform to a normal signal waveform constitutes a regression problem.Given that the scintillator and collection usually do not form a linear system,typical regression methods such as multi-parameter fitting are not immediately applicable.One important advantage of ANNs is their capability to process nonlinear regression problems.To recover the saturated signal,three typical ANNs were tested including backpropagation(BP),simple recurrent(Elman),and generalized radial basis function(GRBF)neural networks(NNs).They represent a basic network structure,a network structure with feedback,and a network structure with a kernel function,respectively.The saturated waveforms were produced mainly by the environmental gamma in a liquid scintillation detector for the China Dark Matter Detection Experiment(CDEX).The training and test data sets consisted of 6000 and 3000 recordings of background radiation,respectively,in which saturation was simulated by truncating each waveform at 40%of the maximum signal.The results show that the GBRF-NN performed best as measured using a Chi-squared test to compare the original and reconstructed signals in the region in which saturation was simulated.A comparison of the original and reconstructed signals in this region shows that the GBRF neural network produced the best performance.This ANN demonstrates a powerful efficacy in terms of solving the saturation recovery problem.The proposed method outlines new ideas and possibilities for the recovery of saturated signals in high-energy particle and nuclear physics experiments.This study also illustrates an innovative application of machine learning in the analysis of experimental data in particle physics.