Developed Adaptive Neuro-fuzzy Inference System Model Research Articles

Advancement of an intelligent system based on ANFIS for predicting machining performance parameters of Inconel 690 – A perspective of metaheuristic approach

The convincing potential of sophisticated milling tools exploited for machining of Inconel alloys in CNC milling machine offers minimal surface roughness-cutting force-cutting temperature trade-off footprint as compared to the conventional machining operation, which afforded a strong motivation to accomplish an in-depth discovery of the parametric design of CNC milling operation. In this study, the synergistic potential of contemporary Cubic boron nitride coated tool is used for machining of Inconel 690 in CNC milling machine, which has been scrutinized in order to build up a correlation among the objective function and the control variables by a metamodel called Adaptive Neuro-fuzzy inference system (ANFIS). The developed ANFIS model was capable of predicting the performance parameters with commendable accuracy as observed from correlation coefficients within the range of 0.946542–0.988996, Mean absolute percentage error (MAPE) in the range of 3.879652–7.456275% along with noticeably low root mean square errors (RMSE). Moreover, the ANFIS acquired results were compared with an Artificial Neural Network (ANN) model, developed on the identical parametric ranges. The comparison of the obtained results indicated that the ANFIS overtakes the ANN model in predicting the preferred response variables, which suggests the modesty of the ANFIS model.

Real-time prediction of water level change using adaptive neuro-fuzzy inference system

ABSTRACTAccurate water levels modelling and prediction is essential for maritime applications. Water prediction is traditionally developed using the least-squares-based harmonic analysis method based on water level change (WLC) measurements. If long water level measurements are not obtained from the tide gauge, accurate water levels prediction cannot be estimated. To overcome the above limitations, the wavelet neural network (WNN) has recently been developed for the WLC prediction from short water level measurements. However, a new adaptive neuro-fuzzy inference system (ANFIS) model is proposed and developed in this paper. The ANFIS model is utilized to predict and select the WLC models of one month of hourly WLC for Yarmouth, Sain-John and Charlottetown stations in Canadian waters and compared with the current-state-of-the-art WNN model. The statistical analysis is applied to analyse the performance of the developed model in training and testing stages. The results showed an accurate modelling level using ANFIS technique for each station in training and testing stage. A comparison between the developed ANFIS method and the current-state-of-the-art WNN method shows that the accuracy of the developed ANFIS model is superior to the current-state-of-the-art model by 21.5% in average.

Development of artificial intelligence based systems for prediction of hydration characteristics of wheat

Hydration characteristics (moisture content, moisture ratio and hydration rate) of wheat kernel during soaking process were studied and simulated. Hydration procedure was performed at five different experimental water temperatures (30, 40, 50, 60 and 70 (°C)) with respect to hydration time changes. Hydration characteristics of samples were modeled using one of the most popular simulation frameworks of artificial intelligence, adaptive neuro-fuzzy inference system (ANFIS). A comparison was also made between results of the best developed ANFIS model and those of the another well-known artificial intelligence technique, artificial neural network (ANN). The hydration temperature and time were used as input parameters and moisture content, moisture ratio and hydration rate were taken as output parameters of the intelligent modeling frameworks. To attain the best model with the highest predictive ability, developed models were compared based on statistical parameters of coefficient of determination, root mean square error and mean relative deviation modulus. According to the results, although the best framework of both ANFIS and ANN were able to accurately predict hydration characteristics, the best ANN simulation framework with a simple structure (2–4–3) was easier to use than the ANFIS with three different structures. ANN surface plots also illustrated that increasing the hydration temperature and time increased moisture content and decreased moisture ratio. Moreover, ANN modeling results indicated that the hydration rate increased in the initial and decreased in the middle and final phase of hydration procedure.

On the exergetic optimization of continuous photobiological hydrogen production using hybrid ANFIS–NSGA-II (adaptive neuro-fuzzy inference system–non-dominated sorting genetic algorithm-II)

The aim of this work was to exergetically optimize the performance of a continuous photobioreactor for hydrogen production from syngas via water gas shift reaction by Rhodospirillum rubrum. To achieve this, a new multi-objective hybrid optimization technique was developed by coupling the elitist NSGA-II (non-dominated sorting genetic algorithm) with the ANFIS (adaptive neuro-fuzzy inference system) to optimize the operational conditions of the photobioreactor. The syngas flow rate and culture agitation speed were independent variables, while rational and process exergy efficiencies as well as normalized exergy destruction were dependent variables. The ANFIS was used to establish an objective function for each dependent variable individually based on the independent variables. The developed ANFIS model was then utilized by the NSGA-II approach to find the optimal operating conditions simultaneously leading to the highest rational and process exergy efficiencies and the lowest normalized exergy destruction. Consequently, the best operating conditions for the photobioreactor were extracted using a Pareto optimal front set consisting of seven optimum points. Accordingly, syngas flow rate of 13.34 mL/min and culture agitation speed of 383.33 rpm yielding process exergy efficiency of 21.66%, rational exergy efficiency of 85.64%, and normalized exergy destruction of 1.55 were found as the best operating conditions.

Scour prediction in long contractions using ANFIS and SVM

Protection of the channel bed in waterways against scour phenomena in long contractions is a very significant issue in channels design. Several field and experimental investigations were carried out to produce a relationship between the scour depth due to the contracted channels width and the governing variables. However, existing empirical equations do not always provide accurate scour prediction due to the complexity of the scour process. This paper investigates local scour depth in long contractions of rectangular channels using the Adaptive Neuro-Fuzzy Inference System (ANFIS) and Support Vector Machines (SVM). For modeling of ANFIS and SVM, the input parameters that affect the scour phenomena are average flow velocity, critical threshold velocity of sediment movement, flow depth, median particle diameter, geometric standard deviation, un-contracted and contracted channel widths. Training and testing stages of the models are carried out using experimental data collected from different literature. The performances of the developed models are compared with those calculated using existing scour prediction equations. The results show that the developed ANFIS model can predict scour depth more accurately than SVM and the existing equations. A sensitivity analysis is also performed to determine the most important parameter in predicting the scour depth in long contractions.

Modeling of a HTPEM fuel cell using Adaptive Neuro-Fuzzy Inference Systems

In this work an Adaptive Neuro-Fuzzy Inference System (ANFIS) model of the voltage of a fuel cell is developed. The inputs of this model are the fuel cell temperature, current density and the carbon monoxide concentration of the anode supply gas.First an identification experiment which spans the expected operating range of the fuel cell is performed in a test station. The data from this experiment is then used to train ANFIS models with 2, 3, 4 and 5 membership functions.The performance of these models is then compared and it is found that using 3 membership functions provides the best compromise between performance and fast model evaluation. This model has a mean absolute error of 0.70%. It is concluded that the developed ANFIS model is suitable for optimization of fuel cell systems and as the steady state component in larger dynamic system models.

Adaptive neuro-fuzzy interference system modelling of carbon nanotube-based electrical discharge machining process

This study deals with modelling of surface roughness with carbon nanotube (CNT)-based electrical discharge machining (EDM) of AISI D2 tool steel material by means of adaptive neuro-fuzzy inference system (ANFIS) approach. The full factorial design of experimental techniques was adapted to conduct the experimental works. The CNT mixed dielectric nanofluids were prepared and used in the EDM process to analyze the surface roughness. The first-order sugeno type fuzzy interference modeling was used to predict the output parameters and compared with experimental values. The ANFIS model has been developed in terms of machining parameters for the prediction of surface roughness using trained data. The ANFIS predictions for the surface roughness with CNT the testing error was 0.20276 and correlation coefficient was 0.997 with the experimental data and for without CNT the testing error was 0.26529 and correlation coefficient was 0.889. The developed ANFIS model were compared in terms of their performances and shows that high residual R2 value indicate that the predicted model very well fits with the experimental data for using CNT on EDM process. The proposed model can also be used for estimating surface roughness on-line.

Prediction of coke quality using adaptive neurofuzzy inference system

The production rate, quality and cost of hot metal production through the blast furnace route depend mainly on the quality of the metallurgical coke. Coke reactivity index and coke strength after reaction (CSR) are the most important parameters used for the assessment of the high temperature properties of coke. Many coke plants and blast furnaces around the world use CSR as a specification just as important as cold strength, size and chemistry. The present work aims to fulfil the need for a model that will predict the coke CSR from coal blend characteristics. In this work, the functional relationship between the coal blend properties (ash, volatile matter, average vitrinite reflectance, crucible swelling number, total reactives, vitrinite distribution V9–V13 and basicity index) and the corresponding coke CSR has been mapped using an adaptive neurofuzzy inference system (ANFIS). The ANFIS model is formulated with different sets of coal blend properties as input variable, and the singular value decomposition and QR factorisation based techniques have been employed for model reduction. It has been found that the developed ANFIS model predicts the CSR with reasonable accuracy.

Non-linear modelling and control of a conveyor-belt grain dryer utilizing neuro-fuzzy systems

The grain drying process is characterized by its complex and non-linear nature. As a result, conventional control system design cannot handle this process appropriately. This work presents an intelligent control system for the grain drying process, utilizing the capabilities of the adaptive neuro-fuzzy inference system (ANFIS) to model and control this process. In this context, a laboratory-scale conveyor-belt grain dryer was specifically designed and constructed for this study. Utilizing this dryer, a real-time experiment was conducted to dry paddy (rough rice) grains. Then, the input–output data collected from this experiment were presented to an ANFIS network to develop a control-oriented dryer model. As the main controller, a simplified proportional–integral–derivative (PID)-like ANFIS controller is utilized to control the drying process. A real-coded genetic algorithm (GA) is used to train this controller and to find its scaling factors. From the robustness tests and a comparative study with a genetically tuned conventional PID controller, the simplified ANFIS controller has proved its remarkable ability in controlling the grain drying process represented by the developed ANFIS model.