Adaptive Neuro-fuzzy Interface System Research Articles

Prediction of colour strength in environmentally‐friendly dyeing of polyester fabric with madder using supercritical carbon dioxide

AbstractThe textile industry is one of the significant reasons for global water pollution, with dyeing processes being particularly environmentally detrimental. Researchers have explored alternative approaches to address this issue, such as using natural dyes, supercritical fluids and so forth. In addition to environment‐friendly approaches, reducing the number of experiments in studies, accurate production straightaway and using artificial intelligence (AI), one of the technologies of the present and the future that will provide significant support. Reaching clearer results with AI technology will not necessarily contribute to environment‐friendly technologies. However, AI techniques, including artificial neural networks (ANNs) and adaptive neuro fuzzy interface system (ANFIS) were employed to predict the colour strength (K/S) of the dyed fabric based on process parameters. A comprehensive experimental design involving pressure, temperature, and time variations was conducted, and the results were analysed using multi‐factor analysis of variance (MANOVA). The study demonstrates that supercritical carbon dioxide (scCO2) dyeing with madder on polyester fabric is a promising and environmentally friendly approach. Additionally, the optimised ANN and ANFIS models, aided by genetic algorithms (GAs), exhibit high predictive accuracy (less than 3%), providing insights into the impact of process parameters on colour strength. This research underscores the potential of AI‐driven automation in textile dyeing, offering solutions for dye formula prediction, colour matching, and defect detection, reducing the need for human intervention in these processes.

HETEROGENEOUS SERVER RETRIAL QUEUEING MODEL WITH FEEDBACK AND WORKING VACATION USING ARTIFICIAL BEE COLONY OPTIMIZATION ALGORITHM

This research delves into the dynamics of a retrial queueing system featuring heterogeneous servers with intermittent availability, incorporating feedback and working vacation mechanisms. Employing a matrix geometric approach, this study establishes the steady-state probability distribution for the queue size in this complex heterogeneous service model. Additionally, a range of system performance metrics is developed, alongside the formulation of a cost function to evaluate decision variable optimization within the service system. The Artificial Bee Colony (ABC) optimization algorithm is harnessed to determine service rates that minimize the overall cost. This work includes numerical examples and sensitivity analyses to validate the model's effectiveness. Also, a comparison between the numerical findings and the neuro-fuzzy results has been examined by the adaptive neuro fuzzy interface system (ANFIS).

Physics-based and machine-learning models for accurate scour depth prediction.

Scour phenomena remain a significant cause of instability in offshore structures. The present study estimates scour depths using physics-based numerical modelling and machine-learning (ML) algorithms. For the ML prediction, datasets were collected from previous studies, and the trained models checked against the statistical measures and reported outcomes. The numerical assessment of the scour depth has been also carried out for the current and coupled wave-current environment within a computational fluid dynamics framework with the aid of the open-source platform REEF3D. The outcomes are validated against the previously reported experimental studies. The results obtained from ML schemes demonstrated that the artificial neural network and adaptive neuro-fuzzy interface system models have an elevated level of effectiveness compared with the other models. Whereas the numerical analysis results show a good agreement against the reported values. For the current only conditions, the normalized scour depth (S/D) at the front and rear end of the pier is 0.65 and 0.81. For the wave-current conditions, the normalized scour depth (S/D) is 0.26. The study highlights the importance of machine learning and physics-based numerical modelling to assess scour depth within a reasonable time frame without compromising accuracy. This article is part of the theme issue 'Physics-informed machine learning and its structural integrity applications (Part 2)'.

Fault identification and classification algorithm for high voltage transmission lines based on a fuzzy-neuro-fuzzy approach

ABSTRACT Traditional techniques are used for fault location detection in high-voltage transmission lines that mostly depend on traveling waves and impedance-based techniques suffer from large errors owing to the intricacy of fault modeling for various types of faults. Although single-line to ground faults are dominant in high-voltage transmission lines, fault resistance as well as fault inception angle might distort the current fault detection techniques. In addition, other types of faults exist and that raises the need to develop an accurate fault detection technique to minimize the recovery time. The current paper introduces a fuzzy and neuro-fuzzy algorithm to detect, analyze, and locate different faults taking place in high-voltage transmission lines. A MATLAB Simulink Model is used for analyzing different fault cases; fault detection and classification are done by the Fuzzy Interface System (FIS), while fault location detection is done using the Adaptive Neuro-Fuzzy Interface System (ANFIS). The introduced algorithm is evaluated via the Mean Square Error (MSE) technique. The results showed full success in detecting and identifying different fault types, with a 0.0042 validity performance factor for fault location detection using ANFIS.

Power Quality Improvement using Dual Multilevel Converter for Micro Grid-Connected PV Energy Systems using ANFIS

This paper presents the implementation of dual voltage source inverter (DVSI) approach to improve the microgrid performance by enhancing the power quality. This paper also improves the power quality in photovoltaic (PV) generation interactive microgrids, respectively. The power generated from PV based distributive energy resources (DER) is perfectly applied to the microgrid through the two inverters, thus the nonlinear and unbalance load related problems are compensated. Thus, the power quality problems such as voltage sag, current drops, and power factor, active, and reactive powers are reduced by dual multilevel converter (DMLC). This converter also used for cooperative controlling to minimize the unbalances and voltage harmonics in microgrid by using adaptive neuro fuzzy interface system (ANFIS) based controller, which is applied in the shunt and series manner between the two VSIs. The total load current is mutually shared among the series and shunt VSI, if one VSI fails, then the other VSI continues its operation. Thus, the DMLC components failure rate is reduced, and the system lost energy is also reduced, which leads to improved reliability by maintain the reduction of down time price at environmental free conditions. The proposed control model is implemented in MATLAB/Simulink environment, and the obtained results shows that the superiority of total harmonic distortion (THD) reduction by the proposed ANFIS controller as compared to the conventional artificial neural network (ANN) and fuzzy logic controllers (FLC) by eliminating all the load current and grid voltage-based harmonics.

Frequency Stabilization of Multi Area Power System using Adaptive Neuro Fuzzy Interface System and Support Vector Machine Approach

Frequency Stabilization of Multi Area Power System using Adaptive Neuro Fuzzy Interface System and Support Vector Machine Approach

Application of a hybrid ANFIS with metaheuristic algorithms to estimate the aeration design parameters

Abstract Cavitation is a common and complex hydraulic phenomenon on the chute spillways and may cause damage to the structure. Aeration in the water flow is one of the best ways to prevent cavitation. To design an aerator, estimation of aeration coefficient (β), jet length (L/h0), and jet impact angle on chute (tanγ) are important in this study. The potential of a hybrid adaptive neuro-fuzzy interface system (ANFIS) with metaheuristic algorithms was investigated to estimate the required parameters to design an aerator. The ANFIS was combined with four metaheuristic algorithms, including differential evolution (DE), ant colony optimization (ACO), genetic algorithm (GA), and particle swarm optimization (PSO). Experimental data and dimensionless parameters were used to develop the proposed hybrid ANFIS models. Three statistical indicators, including root mean square error (RMSE), mean average error (MAE), and coefficient of determination (R2), were employed to compare the proposed methods with empirical relations. According to the statistical indicators, among the data-driven methods, the ANFIS–DE method had the best prediction in estimating β (RMSE = 0.018, R2 = 0.984, MAE = 0.013), L/h0 (RMSE = 1.293, R2 = 0.963, MAE = 1.082), and tanγ (RMSE = 0.009, R2 = 0.939, MAE = 0.007).

A methodology for performance prediction: Hydrodynamic investigation of spiral grooved thrust bearing

AbstractThis paper presents the design and numerical optimization of oil‐lubricated spiral grooved thrust bearing (SGTB) for its application at high speed and axial loading conditions. A numerical model is developed using nonlinear incompressible Reynold's equation and is solved using the finite volume method (FVM) to determine the static characteristics over the bearing surface. Further, the influence of groove parameters such as spiral angle, groove angle, film thickness ratio, number of grooves and speed on the static characteristics of the bearing has been investigated. The result shows that the designed oil‐lubricated SGTB can operate at high‐speed conditions and withstand high axial load. Further, the characteristic data sets acquired from the numerical analysis are trained using an artificial neural network (ANN), and their performance is evaluated through the computation of the regression coefficient. Then adaptive neuro‐fuzzy interface system (ANFIS) surface plot is obtained to determine the optimum bearing parameters.

Prediction of Post-Operative Clinical Parameters in Posterior Scoliosis Surgery Through An Adaptive Neuro-Fuzzy Interface System

Background and Objective: Postoperative clinical indices should be estimated accurately in scoliosis correction surgeries, which have been analyzed in various studies such as experimental (in vitro or in vivo) trials through different modeling methods (finite element or multibody analysis). These costly and time-consuming methods can only be conducted on a large number of scoliotic patients. An adaptive neuro-fuzzy interface system (ANFIS) is used in this study to estimate the postoperative cobb and thoracic kyphosis angles in adolescent idiopathic scoliosis patients undergoing posterior scoliosis correction surgeries. Methods: Four groups of 55 patients with distinct preoperative clinical indices (thoracic cobb and pelvic incidence) were considered the ANFIS inputs, whereas postoperative thoracic cobb and kyphosis angles were used as the outputs. For robustness evaluation, the predicted values of postoperative angles were compared with measurements by calculating the root mean square errors and clinical correction deviation indices (the relative deviation of postoperative predicted angles from the real angles). Results: The least root mean square errors (3.0º and 6.3° for the main thoracic cobb and thoracic kyphosis estimations, respectively) were recorded in the group with the main thoracic cobb, pelvic incidence, thoracic kyphosis, and T1 spinopelvic inclination used as inputs. The clinical correction deviation indices were calculated 0.0086 and 0.0641 for cobb angles in two cases and 0.0534 and 0.2879 for thoracic kyphosis in two other cases. Conclusion: Greater differences between preoperative and postoperative cobb angles compared with those of thoracic kyphosis decreased the root-mean-square errors and clinical deviation indices but improved accuracy

Application of signal processing techniques and intelligent classifiers for high-impedance fault detection in ensuring the reliable operation of power distribution systems

High-impedance fault (HIF) is always a threat and the biggest challenge in the power transmission and distribution system (PTDS). For a PTDS to operate effectively, HIF diagnosis is essential. However, given the HIF’s nature and the involved complexity, detection, identification, and fault location are difficult. This will be even more complicated in conventional PTDSs as they are inefficient and highly vulnerable. Given the importance and urgent need for HIF diagnosis in PTDS, this study reviews state-of-the-art HIF phenomenon and detection techniques and proposes the use of “various signal processing techniques for fault feature extraction” and “ different classifiers for identifying HIF.” First, HIF current/voltage signals are analyzed using signal processing techniques, which include the discrete wavelet transform (DWT), pattern recognition, Kalman filtering, TT transform, mathematical morphology (MM), S transform (ST), fast Fourier transform (FFT), principal component analysis (PCA), linear discriminant analysis (LDA), and wavelet transforms, such as dual-tree, maximum overlap discrete wavelet transform (MODWT), and lifting wavelet transform (LWT). Second, the various HIF and non-HIF faults are classified using intelligent classifiers. The intelligent classifiers include artificial neural networks (ANNs), probabilistic neural networks (PNNs), genetic algorithms (GAs), fuzzy logic, adaptive neuro-fuzzy interface system, support vector machine (SVM), extreme learning machine (ELM), adaptive resonance theory, random forests (RFs), decision trees (DTs), and convolution neural networks (CNNs). In addition to the comparative discussion of various classifier techniques, their evaluation criterion and performance are prioritized. Third, this review also studied different test systems, such as radial distribution network, mesh distribution network, IEEE 4 node, IEEE 13 node feeder, IEEE 34 node feeder, IEEE 39 node feeder, IEEE 123 node feeder, Palash feeder, and test microgrid systems, to assess the pertinence of various HIF detection schemes and the behavior along with methods to locate the HIF. Overall, we believe this review would serve as a comprehensive compendium of advanced techniques for HIF diagnosis in different test systems.

Magnetic and structural characterization of Sn doped cobalt ferrites; A visible light-driven photocatalysts for degradation of rhodamine-B and modeling the process by artificial intelligence tools

Tin (Sn) substituted cobalt ferrites Co1−xSnxFe2O4 with composition (x = 0.0, 0.1, 0.2, 0.3, 0.4 and 0.5) were synthesized utilizing sol-gel auto combustion method. All synthesized powders were characterized for their structural, magnetic and optical characterization after calcination. The single phase spinel with good crystallisation and a decreasing crystallite size with Sn substitution is confirmed by the powder X-ray diffraction (XRD) pattern. The magnetic measurements were carried out at room temperature where saturation magnetization values show sufficient magnetic nature of calcinated ferrites. The bandgap values for pure cobalt and Sn-substituted cobalt ferrites were closer to the experimental value. The visible light photocatalytic degradation of Rhodamine B was carried out in presence of Sn substituted cobalt ferrites that shows good catalytic activity up to 87% degradation for Co0.5Sn0.5Fe2O4 catalyst. Additionally, this study used Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Interface System (ANFIS) models to simulate degradation efficiency with inputs including time, pH, and catalyst dosage. The results revealed that the 3–2–2–1 structure ANN4 model outperformed the other ANN and ANFIS models in the testing phase (RMSE = 6.90% and NSE = 0.92). According to the findings of this study, artificial intelligence models are capable of precisely predicting Rhodamine B degradation.

Comparative Study of Soft Computing and Metaheuristic Models in Developing Reduced Exhaust Emission Characteristics for Diesel Engine Fueled with Various Blends of Biodiesel and Metallic Nanoadditive Mixtures: An ANFIS-GA-HSA Approach.

Since the discovery of petrol-based products, a surge in energy-requiring equipment has been established across the world. Recent depletion of the existing crude oil resources has motivated researchers to opt for and analyze potential fuels that could potentially provide a cost-effective and sustainable solution. The current study selects a waste plant known as Eichhornia crassipes through which biodiesel is generated, and its blends are tested in diesel engines for feasibility. Different models using soft computing and metaheuristic techniques are employed for the accurate prediction of performance and exhaust characteristics. The blends are further mixed with nanoadditives, thereby exploring and comparing the changes in performance characteristics. The input attributes considered in the study comprise engine load, blend percentage, nanoparticle concentration, and injection pressure, while the outcomes are brake thermal efficiency, brake specific energy consumption, carbon monoxide, unburnt hydrocarbon, and oxides of nitrogen. Models were further ranked and chosen based on their set of attributes using the ranking technique. The ranking criteria for models were based on cost, accuracy, and skill requirement. The ANFIS harmony search algorithm (HSA) reported a lower error rate, while the ANFIS model reported the lowest cost. The optimal combination achieved was 20.80 kW, 2.48047, 150.501 ppm, 4.05025 ppm, and 0.018326% for brake thermal efficiency (BTE), brake specific energy consumption (BSEC), oxides of nitrogen (NOx), unburnt hydrocarbons (UBHC), and carbon monoxide (CO), respectively, thereby furnishing better results than the adaptive neuro-fuzzy interface system (ANFIS) and the ANFIS-genetic algorithm model. Henceforth, integrating the results of ANFIS with an optimization technique with the harmony search algorithm (HSA) yields accurate results but at a comparatively higher cost.

Intelligent cascaded adaptive neuro fuzzy interface system controller fed KY converter for hybrid energy based microgrid applications

Purpose. This article proposes a new control strategy for KY (DC-DC voltage step up) converter. The proposed hybrid energy system fed KY converter is utilized along with adaptive neuro fuzzy interface system controller. Renewable energy sources have recently acquired immense significance as a result of rising demand for electricity, rapid fossil fuel exhaustion and the threat of global warming. However, due to their inherent intermittency, these sources offer low system reliability. So, a hybrid energy system that encompasses wind/photovoltaic/battery is implemented in order to obtain a stable and reliable microgrid. Both solar and wind energy is easily accessible with huge untapped potential and together they account for more than 60 % of yearly net new electricity generation capacity additions around the world. Novelty. A KY converter is adopted here for enhancing the output of the photovoltaic system and its operation is controlled with the help of a cascaded an adaptive neuro fuzzy interface system controller. Originality. Increase of the overall system stability and reliability using hybrid energy system fed KY converter is utilized along with adaptive neuro fuzzy interface system controller. Practical value. A proportional integral controller is used in the doubly fed induction generator based wind energy conversion system for controlling the operation of the pulse width modulation rectifier in order to deliver a controlled DC output voltage. A battery energy storage system, which uses a battery converter to be connected to the DC link, stores the excess power generated from the renewable energy sources. Based on the battery’s state of charge, its charging and discharging operation is controlled using a proportional integral controller. The controlled DC link voltage is fed to the three phase voltage source inverter for effective DC to AC voltage conversion. The inverter is connected to the three phase grid via an LC filter for effective harmonics mitigation. A proportional integral controller is used for achieving effective grid voltage synchronization. Results. The proposed model is simulated using MATLAB/Simulink, and from the obtained outcomes, it is noted that the cascaded adaptive neuro fuzzy interface system controller assisted KY converter is capable of maintaining the stable operation of the microgrid with an excellent efficiency of 93 %.

Prediction of Post-operative Clinical Indices in Scoliosis Correction Surgery Using an Adaptive Neuro-fuzzy Interface System.

Accurate estimation of post-operative clinical parameters in scoliosis correction surgery is crucial. Different studies have been carried out to investigate scoliosis surgery results, which were costly, time-consuming, and with limited application. This study aims to estimate post-operative main thoracic cobb and thoracic kyphosis angles in adolescent idiopathic scoliosis patients using an adaptive neuro-fuzzy interface system. Distinct pre-operative clinical indices of fifty-five patients (e.g., thoracic cobb, kyphosis, lordosis, and pelvic incidence) were taken as the inputs of the adaptive neuro-fuzzy interface system in four categorized groups, and post-operative thoracic cobb and kyphosis angles were taken as the outputs. To evaluate the robustness of this adaptive system, the predicted values of post-operative angles were compared with the measured indices after the surgery by calculating the root mean square errors and clinical corrective deviation indices, including the relative deviation of post-operative angle prediction from the actual angle after the surgery. The group with inputs for main thoracic cobb, pelvic incidence, thoracic kyphosis, and T1 spinopelvic inclination angles had the lowest root mean square error among the four groups. The error values were 3.0° and 6.3° for the post-operative cobb and thoracic kyphosis angles, respectively. Moreover, the values of clinical corrective deviation indices were calculated for four sample cases, including 0.0086 and 0.0641 for the cobb angles of two cases and 0.0534 and 0.2879 for thoracic kyphosis of the other two cases. In all scoliotic cases, the post-operative cobb angles were lesser than the pre-operative ones; however, the post-operative thoracic kyphosis might be lesser or higher than the pre-operative ones. Therefore, the cobb angle correction is in a more regular pattern and is more straightforward to predict cobb angles. Consequently, their root-mean-squared errors become lesser values than thoracic kyphosis.

Modelling and prediction of WEDM parameters for sustainable machining of titanium grade-2 alloy

PurposeThe purpose of the study is to establish a predictive model for sustainable wire electrical discharge machining (WEDM) by using adaptive neuro fuzzy interface system (ANFIS). Machining was done on Titanium grade 2 alloy, which is also nicknamed as workhorse of commercially pure titanium industry. ANFIS, being a state-of-the-art technology, is a highly sophisticated and reliable technique used for the prediction and decision-making.Design/methodology/approachKeeping in the mind the complex nature of WEDM along with the goal of sustainable manufacturing process, ANFIS was chosen to construct predictive models for the material removal rate (MRR) and power consumption (Pc), which reflect environmental and economic aspects. The machining parameters chosen for the machining process are pulse on-time, wire feed, wire tension, servo voltage, servo feed and peak current.FindingsThe ANFIS predicted values were verified experimentally, which gave a root mean squared error (RMSE) of 0.329 for MRR and 0.805 for Pc. The significantly low RMSE verifies the accuracy of the process.Originality/valueANFIS has been there for quite a time, but it has not been used yet for its possible application in the field of sustainable WEDM of titanium grade-2 alloy with emphasis on MRR and Pc. The novelty of the work is that a predictive model for sustainable machining of titanium grade-2 alloy has been successfully developed using ANFIS, thereby showing the reliability of this technique for the development of predictive models and decision-making for sustainable manufacturing.

Feasibility of Harris Hawks Optimization in Combination with Fuzzy Inference System Predicting Heating Load Energy Inside Buildings

Heating and cooling systems account for a considerable portion of the energy consumed for domestic reasons in Europe. Burning fossil fuels is the main way to produce this energy, which has a detrimental effect on the environment. It is essential to consider a building’s characteristics when determining how much heating and cooling is necessary. As a result, a study of the related buildings’ characteristics, such as the type of cooling and heating systems required for maintaining appropriate indoor air conditions, can help in the design and construction of energy-efficient buildings. Numerous studies have used machine learning to predict cooling and heating systems based on variables that include relative compactness, orientation, overall height, roof area, wall area, surface area, glazing area, and glazing area distribution. Fuzzy logic, however, is not used in any of these methods. In this article, we study a fuzzy logic approach, i.e., HHO−ANFIS (combination of Harris hawks optimization and adaptive neuro-fuzzy interface system), to predict the heating load in residential buildings and investigate the feasibility of this technique in predicting the heating load. Fuzzy techniques obtain perfect results. The analysis results show that the HHO−ANFIS with a population size of 400, the highest value of R2 (0.98709 and 0.98794), and the lowest value of RMSE (0.08769 and 0.08281) in the training and testing dataset, respectively, can predict the heating load with high accuracy. According to the high value of R2 (98%) and low value of RMSE, HHO−ANFIS can be used in predicting the heating load of residential buildings.

Teaching–Learning–Based Optimization (TLBO) in Hybridized with Fuzzy Inference System Estimating Heating Loads

Nowadays, since large amounts of energy are consumed for a variety of applications, more and more emphasis is placed on the conservation of energy. Recent investigations have experienced the significant advantages of using metaheuristic algorithms. Given the importance of the thermal loads’ analysis in energy-efficiency buildings, a new optimizer method, i.e., the teaching–learning based optimization (TLBO) approach, has been developed and compared with alternative techniques in the present paper to predict the heating loads (HLs). This model is applied to the adaptive neuro–fuzzy interface system (ANFIS) in order to overcome its computational deficiencies. A literature-based dataset acquired for residential buildings is used to feed these models. According to the results, all the applied models can appropriately predict and analyze the heating load pattern. Based on the value of R2 calculated for both testing and training (0.98933, 0.98931), teaching–learning-based optimization can help the adaptive neuro–fuzzy interface system to enhance the results’ correlation. Also, the high R2 value means that the model has high accuracy in the HL prediction. In addition, according to the estimated RMSE, the training error of TLBO–ANFIS in the testing and training stages was 0.07794 and 0.07984, respectively. The low value of root–mean–square error (RMSE) indicates that the TLBO–ANFIS method acts favorably in the estimation of the heating load for residential buildings.

The Applicability of Biogeography-Based Optimization and Earthworm Optimization Algorithm Hybridized with ANFIS as Reliable Solutions in Estimation of Cooling Load in Buildings

The foundation of energy-efficient architectural design is modeling heating and cooling loads (HLs and CLs), which defines the heating and cooling apparatus constraints necessary to maintain a suitable interior air environment. It is possible that analytical models for energy-efficient buildings might offer an accurate evaluation of the influence that various building designs would have. The implementation of these instruments, however, might be a process that requires a significant amount of manual labor, a significant amount of time, and is reliant on user experiences. In light of this, the authors of this paper present two unique methods for estimating the CL of residential structures in the form of complex mathematical concepts. These methodologies include an evolutionary web algorithm (EWA), biogeography-based optimization (BBO), and a hybridization of an adaptive neuro-fuzzy interface system (ANFIS), namely BBO-ANFIS and EWA-ANFIS. The findings initiated from each of the suggested models are evaluated with the help of various performance metrics. Moreover, it is possible to determine which model is the most effective by comparing their coefficient of determination (R2 ) and its root mean square error (RMSE) to each other. In mapping non-linear connections between input and output variables, the observed findings showed that the models used have a great capability. In addition, the results showed that BBO-ANFIS was the superior forecasting model out of the two provided models, with the lowest value of RMSE and the greatest value of R2 (RMSE = 0.10731 and 0.11282 and R2 = 0.97776 and 0.97552 for training and testing phases, respectively). The EWA-ANFIS also demonstrated RMSE and R2 values of 0.18682 and 0.17681 and 0.93096 and 0.93874 for the training and testing phases, respectively. Finally, this study has proven that ANN is a powerful tool and will be useful for predicting the CL in residential buildings.

Effects of flexible fins on melting process in a phase change material filled circular cavity

In this study, melting process in a phase change material (PCM) filled annular closed space by using flexible fins is analyzed. Cases with different number of flexible fins and without fin are compared in terms of phase change dynamics while two-way fluid structure interaction analysis is used. Temperature differences of 30 °C and 60 °C are considered between the walls of the circular enclosure while flexible fin numbers are varied between 0 and 10. It is observed that the melting is significantly accelerated by using the elastic fins. Due to the deflection of the elastic fins, melt process dynamics is influenced by varying fin number and its elastic modulus. Further reduction in the full phase transition occurs when the temperature difference is increased. As compared to un-finned case at temperature difference of 30 °C by using 10 flexible fins at ∆T = 60 °C, there is 87 % reduction in the complete phase transition time. The ANFIS (Adaptive Neuro-Fuzzy Interface System) based modeling approach is used for phase transition dynamics in circular enclosure equipped with flexible fins.

Predicting crop yields using a new robust Bayesian averaging model based on multiple hybrid ANFIS and MLP models

Predicting crop yield is an important issue for farmers. Food security is important for decision-makers. The agriculture industry can more accurately supply human demand for food if the crop yield is predicted accurately. Tomato is one of the most important crops so that 160 million tonnes of tomatoes are produced annually around the world. In this study, tomato yield based on data of 40 cities of Iran country including annual average temperature (T), relative humidity (RH), effective rainfall (R), wind speed (WS), and Evapotranspiration (EV) for the period of 1968–2018 was predicted using a new Bayesian model averaging (BMA). The paper's main innovation is the use of the new BMA so that it allows the modellers to quantify the uncertainty of model parameters and inputs simultaneously. For this aim, first, the multiple Adaptive neuro-fuzzy interface system (ANFIS) and multi-layer perceptron (MLP) were used for predicting crop yield. To train the ANFIS and MLP model, a new algorithm, namely, multi verse optimization algorithm (MOA) was used. Also, the ability of MOA was benchmarked against the particle swarm optimization (PSO), and firefly algorithm (FFA). In the next level, the new BMA used the outputs of the ANFIS-MOA, MLP-MOA, ANFIS, FFA, MLP-FFA, ANFIS-PSO, MLP-PSO, ANFIS, and MLP for predicting tomato yield in an ensemble framework. The five- input combination of RH, T, and R, WS, and EV gave the best result. The mean absolute error (MAE) of the BMA in the testing level was 20.12 (Ton/ha) while it was 24.12, 24.45, 24.67, 25.12, 29.12, 30.12, 31.12, and 33.45 for the ANFIS-MOA, MLP-MOA, ANFIS-FFA, MLP-FFA, ANFIS-PSO, MLP-PSO, ANFIS, and MLP models. Regarding the results of uncertainty analysis, the uncertainty of BMA was lower than those of the ANFIS-MOA, MLP-MOA, ANFIS-FFA, MLP-FFA, ANFIS-PSO, MLP-PSO, ANFIS, and MLP models while the MLP model provided the highest uncertainty. The results of this study indicated that BMA using multiple MLP and ANFIS model was useful for predicting tomato yield.