Adaptive Neuro-fuzzy Inference System Training Research Articles

PREDICTION SYSTEM FOR THE AMOUNT OF SUGAR PRODUCTION USING ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM

Sugar is one of the staple foods most Indonesians use, so sugar production needs to be done optimally to meet people's needs. This research will design a prediction system for the amount of sugar production in PTPN XI PG Prajekan using the Adaptive Neuro-Fuzzy Inference System (ANFIS) method. ANFIS is a combined method of two systems, namely a fuzzy logic system and an artificial neural network system. This research consists of data collection, ANFIS system design, ANFIS training, ANFIS testing, accuracy calculation, and result analysis. The prediction system for the amount of sugar production is designed to predict the variable which is the amount of sugar production in the year using the input variables (sugarcane harvested area in year ), (amount of sugarcane in year ), (average of yield in year ), and (number of milling days in year ). The experiments in this research used variations of the type of membership function and the number of membership functions. The best model obtained in this research is a model with a difference between two sigmoidal membership functions and a product of two sigmoidal membership functions with a total of 2 membership functions for each input variable. Both models have the same Mean Absolute Percentage Error (MAPE) value, which is 1.79% in the training process and 4.82% in the testing process.

ANFIS-AMAL: Android Malware Threat Assessment Using Ensemble of ANFIS and GWO

Abstract The Android malware has various features and capabilities. Various malware has distinctive characteristics. Ransomware threatens financial loss and system lockdown. This paper proposes a threat-assessing approach using the Grey Wolf Optimizer (GWO) to train and tune the Adaptive Neuro-Fuzzy Inference System (ANFIS) to categorize Android malware accurately. GWO improves efficiency and efficacy in ANFIS training and learning for Android malware feature selection and classification. Our approach categorizes Android malware as a high, moderate, or low hazard. The proposed approach qualitatively assesses risk based on critical features and threats. Our threat-assessing mechanism’s scale categorizes Android malware. The proposed approach resolves the issue of overlapping features in different types of malware. Comparative results with other classifiers show that the ensemble of GWO is effective in the training and learning process of ANFIS and thus achieves 95% F-score, 94% specificity, and 94% accuracy. The ensemble makes fast learning possible and improves classification accuracy.

Feasibility of integrated numerical and Adaptive Neuro-Fuzzy Inference System models in predicting the thermophysical properties of unsteady coupled micropolar–Casson hybrid nanofluids

This article aims to probe postulated phenomena using a paired micropolar and Casson hybrid fluid over a rotating disk. The intension to design a numerical technique (Runge–Kutta fourth-order along with shooting technique) integrated Adaptive Neuro-Fuzzy Inference System (ANFIS) envisioned with a thermal and exponentially space-dependent heat source, nonlinear thermal radiation and entropy production is developed in this phase of work. To ensure that, a nonlinear partial differential equation set of equations has been transformed into an ordinary differential equation by using the proper self-similarity variables. The model's research results, with a few notable outliers, are mostly consistent with those from prior research that was merged into the dataset used to train the ANFIS model. With the impact of active factors, the results are esthetically exhibited for numerous profiles. This displays that with the rise in magnetic field and radiation, the velocity and temperature profiles increase sharply, resulting in a contradiction phenomenon with the decreasing electric field inputs. Also, tilting of vortex viscosity, spin gradient viscosity and microinertia density on the various microrotation components displays inclination. Moreover, ANFIS training was exploited to analyze the approximate solutions for specific scenarios, and the developed ANFIS was evaluated against a testing dataset to emphasize its performance. Due to their longer render, the nanoparticles exploited here are deemed suitable for use in bone implants, iodinated agents for blood imaging and red blood cell stimulation. Thus, the results of this study may be applied to therapeutic anemia therapies.

Fuzzy Control Strategies Development for a 3-DoF Robotic Manipulator in Trajectory Tracking

This research delves into the development and evaluation of two distinct controllers for a 3-DoF robotic arm in the context of Industry 4.0. Two primary control strategies are presented in the study. The first is a Fuzzy Logic Controller that utilizes joint position error and its derivative as inputs, employing a set of 9 control knowledge rules. The second is an Adaptive Neuro-Fuzzy Inference System (ANFIS) Controller, trained to learn the inverse dynamic model of the robot through a structured dataset. The research emphasizes the importance of accurate parameter tuning and data acquisition to achieve optimal control system performance. Extensive experimentation was conducted to evaluate the controllers’ performance in trajectory tracking and their response against external disturbances, such as load variations. The controllers exhibited remarkable precision and proficiency in tracking reference trajectories, with minimal deviations, overshoots, or oscillations. A quantitative analysis using performance indices such as root mean square error (RMSE) and the integral of the absolute value of the time-weighted error (ITAE) further confirmed the controllers’ effectiveness. Notably, the ANFIS Controller consistently outperformed the Fuzzy Logic Controller, demonstrating superior precision in trajectory tracking. The study underscored the importance of selecting the right control method and obtaining high-quality training data. Challenges in parameter tuning for Fuzzy Logic Controllers and potential time constraints in training ANFIS were discussed. The findings have significant implications for advancing robotic control systems, particularly in the era of Industry 4.0.

RAINFALL PREDICTION IN JEMBER REGENCY WITH ADAPTIVE NEURO FUZZY INFERENCE SYSTEM BASED ON GSMaP SATELLITE DATA

Rainfall is very influential in daily life, including in agriculture. According to the Jember Regency Government, the majority of the economic activities of the Jember people come from the agricultural sector. Significant changes in rainfall conditions will adversely affect the agricultural sphere. The Water Resources Office of Jember Regency measures rainfall directly. Precipitation measurement can also be made indirectly using the Global Satellite Mapping of Precipitation (GSMaP), a project promoted by the Japan Aerospace Exploration Agency (JAXA) to produce rainfall accumulation globally. Rainfall predictions are urgently needed to address rainfall-related issues. The Adaptive Neuro-Fuzzy Inference System (ANFIS) method is an effective method for prediction because its working principle combines adaptive methods of artificial neural networks and fuzzy logic. The RMSE in the ANFIS training and testing process on daily rainfall was 12.7464 and 14.6268. Furthermore, RMSE in ANFIS training and testing on monthly rainfall was 7.6336 and 8.1456. The predicted daily rainfall in Jember Regency on January 1, 2023, is 3.1971 mm. Meanwhile, the predicted monthly rainfall in Jember Regency in January 2023 is 19.9114 mm.

Stability of classification performance on an adaptive neuro fuzzy inference system for disease complication prediction

It is crucial to detect disease complications caused by metabolic syndromes early. High cholesterol, high glucose, and high blood pressure are indicators of metabolic syndrome. The aim of this study is to use adaptive neuro fuzzy inference system (ANFIS) to predict potential complications and compare its performance to other classifiers, namely random forest (RF), C4.5, and naïve Bayesian classification (NBC) algorithms. Fuzzy subtractive clustering is used to construct membership functions and fuzzy rules throughout the clustering process. This study analyzed 148 different data sets. Cholesterol, random glucose, systolic, and diastolic blood pressure are all included in the data collection. This learning process was conducted using a hybrid algorithm. The consequent parameters are adjusted forward using the leastsquare approach, while the premise parameters are adjusted backward using the gradient-descent process. The performance of a system is determined by the following indicators: accuracy, sensitivity, specification, precision, area under the curve (AUC), and root mean squared error (RMSE). The results of the training prove that ANFIS is an "excellent classification" classifier. ANFIS has proven to have very good stability across the six performance parameters. The adaptive properties used in ANFIS training and the implementation of fuzzy subtractive clustering strongly support this stability.

Optimizing cleaner productions of sustainable energies: A co-design framework for complementary operations of offshore wind and pumped hydro-storages

There has been a constant evolution in developing clean energy technologies to reduce environmental concerns. In this context, wind energy technology is an ingenious option to make the energy sector more sustainable. However, the growing penetration of wind energy pose challenges to the power system reliability. An integrated system including offshore wind farms (OWFs) and pumped hydroelectric storages (PHSs) emerges as a promising solution to this issue. However, the current wind farm layout optimization (WFLO) always neglect the optimal designs of the PHSs, and also is not based on the optimal OWF controls. Therefore, this paper develops a co-design optimization framework to jointly optimize the wind farm layout and the PHSs’ configurations. The operation principle of the integrated system is first described and two optimization objectives are proposed. An evolutionary trained adaptive neuro-fuzzy inference system (ANFIS) is constructed to forecast OWF power output. The co-design optimization is then formulated as a constrained multi-objective optimization problem and evolutionary algorithms are adopted to find the optimal solutions. Two design cases considering the optimal/non-optimal OWF controls are proposed for the validations. The test results indicate that the trained ANFIS can efficiently identify the characteristics of OWF power generations. In the co-design optimization, the averaged power mismatch is lower than 5 MW, and there are several points that have zero power mismatch. The maximum OWF powers are around 30 MW in the optimal OWF control case, but are lower than 25 MW without using the optimal control. There also exist obvious differences between the optimal PHS parameters in the two cases due to the optimal OWF control.

ANFIS Based Real-Time Power Reference Generator for PV Applications

In this study, a real-time power estimation of a 250Wp solar panel is performed by using a commercial product SIS01-TC-T PV reference module and STM development board. Power estimation was carried out with Adaptive Neuro-Fuzzy Inference System (ANFIS). During the training process, the actual values of the Photo Voltaic (PV) panel with a 250Wp power were used. ANFIS training was accomplished with the hybrid learning algorithm. The power estimation process can be used as a reference power for various applications. The estimated power value can be used in real-time power monitoring for remote monitoring systems or optimum angle control applications for solar tracking mechanisms. It can also be used as a reference value for hybrid Maximum Power Point Tracking (MPPT) control applications or Proportional, Integral, and Derivative (PID) control. In addition, with this reference power value, the Pulse Width Modulation (PWM) signal required by various power electronics stages can be generated.

A Novel Approach for Estimation of Sediment Load in Dam Reservoir With Hybrid Intelligent Algorithms

Predicting the amount of sediment in water resource projects is one of the most important measures to be taken, while sediments have an unknown nature in their behavior. In this research, using the data recorded at the Mazrae station between 2002 and 2013, the amount of sediment in the catchment area of Maku Dam has been predicted using different models of intelligent algorithms. Recorded data including river flow (m3/s), sediment concentration (mg/L), and temperature (°C) were considered input data, and sediment load (ton/day) was considered output data. Initially, using the correlation test, the relationship between each input data with output data was considered. The results show high correlation of sediment concentration data and river flow with sediment load and low correlation of temperature data with these data. In order to find the best combination of data for prediction, the combination of single, binary, and triple data was considered in sensitivity analysis. In order to achieve the purpose of this study, first with the classical adaptive neuro-fuzzy inference system (ANFIS), the amount of sediment load was predicted, and then using evolutionary algorithms in ANFIS training, their performance was examined. The intelligent algorithms used in this study were ant colony optimization extended to continuous domain, particle swarm optimization, differential evolution, and genetic algorithm. The results showed that adaptive neuro-fuzzy inference system–ant colony optimization extended to continuous domain, adaptive neuro-fuzzy inference system–particle swarm optimization, adaptive neuro-fuzzy inference system–genetic algorithm, adaptive neuro-fuzzy inference system–differential evolution, and classical ANFIS had the best performance in predicting the amount of sediment load. In the meantime, it was observed that the coefficient of determination, root mean square error, and scatter index in the test mode for the adaptive neuro-fuzzy inference system–ant colony optimization extended to continuous domain algorithm with the best prediction dataset (sediment concentration + river flow) are equal to 0.991, 13.001, and (ton/day), 0.112, and those for the ANFIS with the weakest prediction (temperature + river flow) are equal to 0.490, 107.383 (ton/day), and 0.929, respectively. The present study showed that the use of intelligent algorithms in ANFIS training has been able to improve its performance in predicting the amount of sediment load in the catchment area of Maku Dam.

An appliance of adaptive Neuro-Fuzzy inference system for predicting the surface roughness of Al-4.5%Cu-TiC MMC in turning operation of CNC milling

In the recent decade, due to extensive application of computerized machine tools in the manufacturing industries, manufacturing processes are in need of highly reliable predictive model for the stable machining. Surface roughness(SR) essentially affects the strength of the developed parts, thus in order to attain high quality and productivity in CNC milling process prediction of surface roughness is very essential as it plays an imperative role in this sector. Hence in this present investigation, an effort has been given to predict the surface roughness of Al-4.5%Cu-TiC MMC, while turning in CNC milling by the use of Adaptive Neuro-Fuzzy Inference System (ANFIS). The experimentation is based on Taguchi’s L25 orthogonal array with the significant parameters such as cutting speed, depth of cut and feed rate. Using the experimental results, an ANFIS model is developed for prediction of surface roughness. The Gaussian fuzzy membership function is adopted in this present investigation for the analysis of ANFIS training as it gives better accuracy. Simulation results are validated through a number of trial cases and the results obtained by training of ANFIS are compared with the results obtained by regression analysis. Comparison result shows that the ANFIS results are better than the results obtained by regression model.

Development of adaptive neuro fuzzy inference system –Evolutionary algorithms hybrid models (ANFIS-EA) for prediction of optimal groundwater exploitation

The present study deals with the optimal extraction of groundwater resources. This approach has been developed for optimal integrated operation in an aquifer in Iran. The results of simulation / optimization models have been used to develop a predictive model based on machine learning. In the first stage, adaptive neuro-fuzzy inference system (ANFIS) was used to predict the Optimal Groundwater Exploitation (OGE) amount in each month having the amounts of several inputs, including the drop and amount of surface water at the end of the previous month and two months earlier, and the water demand of the current month. The results showed that the model’s performance in predicting the test data was undesirable. Therefore, to improve the prediction results, in the second stage several evolutionary optimization algorithms, i.e., particle swarm optimization (PSO), gray wolf optimization (GWO), and Harris hawk optimization (HHO), were used to train ANFIS model. The results indicated the appropriate performance of HHO in ANFIS training, which significantly improved the prediction accuracy of this model. The best scenario for the ANFIS-HHO model included all the input parameters, which resulted in RMSE = 1.45, MAE = 1.15, and R2 = 0.99 respectively, for the test data. In addition, the Taylor diagram (RMSD = 1.40, STD = 15.5 and CC = 0.99) showed ANFIS-HHO accuracy in estimating the OGE value. ANFIS-HHO was also able to improve the accuracy of anfis by RMSE = 4 and MAE = 2 MCM. In general, ANFIS-POS, ANFIS-GWO and ANFIS-HHO had good predictive accuracy compared to ANFIS. The results assure the authors to suggest the developed approach to experts for timely and cost-effective prediction of OGE in similar study areas.

Passivity ANFIS-Based Control for an Intelligent Compact Multilevel Converter

This article proposes a passivity method synthesized by an adaptive neuro fuzzy inference system (ANFIS) to control a grid-tied nine-level packed E-cell (PEC9) inverter. PEC9 is an optimized components count of single-phase, single-dc-source, compact multilevel converter with an appealing feature of dc capacitors horizontal extension. Indeed, PEC9 forms a single auxiliary dc-bus to actively balance its voltages so as grid-tied controller would not be charge of capacitors voltages balancing. Hence, the proposed intelligent-based passivity control (IPC) is designed using only the dynamical model of the PEC9 inverter current. The zero dynamics of state variable errors are obtained and a damping coefficient matrix is presented to theoretically analyze the convergence rate of PEC9 inverter currents under the proposed control functions. In addition, various operating margins of damping coefficients are inquired. So, using the margins, a trained ANFIS as an online estimator tracks the damping injections. It precisely regulates the damping coefficients in unstable conditions; thus, the excellent PEC9 inverter operation is guaranteed. Experiments and simulations validate high reliable performance of the proposed IPC technique under various grid-tied operational conditions.

An Optimal Approach of Adaptive Neuro -Fuzzy Inference System to Predict the Roof Ductility Demand of EBFs Subjected to Near-Fault Pulse-Like Ground Motions

Careful estimation of global ductility will certainly lead to greater accuracy in the design of structural members. In this paper, a new and optimal intelligent model is proposed to predict the roof ductility (μR) of EBF steel frames exposed to the near-fault pulse-like earthquakes, using the Adaptive Neuro-Fuzzy Inference System (ANFIS). To achieve this goal, a databank consisting of 12960 data is created. To establish different geometrical properties of models, 3-,6-, 9-, 12-, 15, 20-stories, steel EBF frames are considered with 3 different types of link beam, column stiffness, and brace slenderness. All models are analysed to reach 4 different performance levels using nonlinear time history under 20 near-fault earthquakes. About 6769 data are applied as ANFIS training data. Subtractive clustering and Fuzzy C-Mean clustering (FCM) methods are applied to generate the purposed model. The results show that FCM provides more accurate outcomes. Moreover, to validate the model, 2257 data are applied (as test data) in order to calculate the correlation coefficient (R) and mean squared error (MSE) between the predicted values of (μR) and the real values. The results of correlation analysis show the high accuracy of the proposed intelligent model.

Regional application of ANFIS in ionosphere time series prediction at severe solar activity period

The ionosphere and its activities play an important role in radio communications, space science and positioning. In this paper, we present efficiency of the adaptive neuro-fuzzy inference system (ANFIS) in ionosphere time series prediction at severe solar activity periods. The high accuracy prediction of the ionosphere total electron content (TEC) using ANFIS is the main purpose of the paper. Also, investigation of the daily, monthly and annually temporal behavior of the ionosphere is another goal of the paper. Three GPS stations (ANKR, ANTL and ZONG) from Turkish permanent GPS network (TPGN) are selected for analysis. These stations are at mid-latitudes. To predict and analyze ionospheric temporal behavior, observations of the 2014–2015 are selected. The index of solar activity in these 2 years is the maximum. Training of ANFIS is done by back-propagation (BP) algorithm. In order to analyze the accuracy of the proposed method, testing step has been performed in different days, months and seasons and compared with the global ionosphere map (GIM) TEC as a traditional ionosphere model, as well as, international reference ionosphere 2016 (IRI2016) and artificial neural network (ANN) models. To validate the results, root mean square error (RMSE), correlation coefficient, relative error, dTEC = |TECGPS-TECMODEL| and residual histogram are used. At testing step, the maximum correlation coefficient obtained 0.869 for the ANFIS. Also, the minimum RMSE of 3.131 TECU is computed for ANFIS at testing step. Using the ANFIS model, the daily, monthly and annual variations of the ionosphere time series are revealed with high accuracy. The results show that the ANFIS model is a reliable and high-precision alternative to the conventional ionosphere models in Turkish region.

Neuro-fuzzy control of sit-to-stand motion using head position tracking

Based on the clinical evidence that head position measured by the multisensory system contributes to motion control, this study suggests a biomechanical human-central nervous system modeling and control framework for sit-to-stand motion synthesis. Motivated by the evidence for a task-oriented encoding of motion by the central nervous system, we propose a framework to synthesize and control sit-to-stand motion using only head position trajectory in the high-level-task-control environment. First, we design a generalized analytical framework comprising a human biomechanical model and an adaptive neuro-fuzzy inference system to emulate central nervous system. We introduce task-space training algorithm for adaptive neuro-fuzzy inference system training. The adaptive neuro-fuzzy inference system controller is optimized in the number of membership functions and training cycles to avoid over-fitting. Next, we develop custom human models based on anthropometric data of real subjects. Using the weighting coefficient method, we estimate body segment parameter. The subject-specific body segment parameter values are used (1) to scale human model for real subjects and (2) in task-space training to train custom adaptive neuro-fuzzy inference system controllers. To validate our modeling and control scheme, we perform extensive motion capture experiments of sit-to-stand transfer by real subjects. We compare the synthesized and experimental motions using kinematic analyses. Our analytical modeling-control scheme proves to be scalable to real subjects’ body segment parameter and the task-space training algorithm provides a means to customize adaptive neuro-fuzzy inference system efficiently. The customized adaptive neuro-fuzzy inference system gives 68%–98% improvement over general adaptive neuro-fuzzy inference system. This study has a broader scope in the fields of rehabilitation, humanoid robotics, and virtual characters’ motion planning based on high-level-task-control scheme.

Training of ANFIS with simulated annealing algorithm on flexural buckling load prediction of aluminium alloy columns

In this study, we propose a simulated annealing algorithm (SA) to train an adaptive neurofuzzy inference system (ANFIS). We performed different types of optimization algorithms such as genetic algorithm (GA), SA and artificial bee colony algorithm on two different problem types. Then, we measured the performance of these algorithms. First, we applied optimization algorithms on eight numerical benchmark functions which are sphere, axis parallel hyper-ellipsoid, Rosenbrock, Rastrigin, Schwefel, Griewank, sum of different powers and Ackley functions. After that, the training of ANFIS is carried out by mentioned optimization algorithms to predict the strength of heat-treated fine-drawn aluminium composite columns defeated by flexural bending. In summary, the accuracy of the proposed soft computing model was compared with the accuracy of the results of existing methods in the literature. It is seen that the training of ANFIS with the SA has more accuracy.
 
 Keywords: Soft computing, ANFIS, simulated annealing, flexural buckling, aluminium alloy columns.

Optimizing a Neuro-Fuzzy System Based on Nature-Inspired Emperor Penguins Colony Optimization Algorithm

A neuro-fuzzy system is a learning machine that finds the parameters of a fuzzy system using approximate techniques of neural networks. Both neural network and fuzzy system have common features. These can solve problems that have no mathematical models. Adaptive neuro-fuzzy inference system (ANFIS) is an adaptive network that uses supervised learning on learning algorithm. To achieve effective results with ANFIS, selecting the optimization method in training is very important. Heuristics and metaheuristics algorithms attempt to find the best solution out of all possible solutions to an optimization problem. ANFIS training can be based on nonderivative algorithms. Heuristics and metaheuristics are nonderivative algorithms that can lead to better performance in ANFIS training. Most heuristic and metaheuristic algorithms are taken from the behavior of biological systems or physical systems in nature. The newly released emperor penguins colony (EPC) algorithm is a population-based and nature-inspired metaheuristic algorithm. This algorithm has much potential for solving various problems. In this article, an optimized ANFIS based on the new EPC algorithm is proposed. The optimized ANFIS is compared with other nonderivative algorithms on benchmark data sets. Eventually, the proposed algorithm is used to solve the classical inverted pendulum problem. The results show that the proposed ANFIS based on the EPC algorithm has less error and better performance than other state-of-the-art algorithms in both training and testing phase.

An intelligent system based on optimized ANFIS and association rules for power transformer fault diagnosis

This research work put forward an intelligent method for diagnosis and classification of power transformers faults based on the instructive Dissolved Gas Analysis Method (DGAM) attributes and machine learning algorithms. In the proposed method, 14 attributes obtained through DGAM are utilized as the initial and unprocessed inputs of Adaptive Neuro-Fuzzy Inference System (ANFIS). In this method, attribute selection and improved learning algorithm are utilized to enhance fault detection and recognition precision. In the propounded fault detection and classification method, the most instructive attributes obtained by DGAM are selected by association rules learning technique (ARLT). Using efficient enlightening attributes and eliminating tautological attributes lead to higher accuracy and superior operation. Furthermore, appropriate training of ANFIS has significant effect on its precision and robustness. Therefore, Black Widow Optimization Algorithm (BWOA) is applied to train the ANFIS. Having excellent exploration and extraction capability, fast convergence speed and simplicity is the main reason for choosing the BWOA as the learning algorithm. Two industrial datasets are utilized to test and evaluate the performance of the put forward method. The results show that the propounded diagnosis system has high accuracy, robust performance and short run time. Selecting the most educative attributes of DGAM, training ANFIS optimally, improving the robustness of ANFIS and increasing the classification accuracy are the main contribution of this paper in the field of power transformer fault detection and classification.

Comparative evaluation of intelligent algorithms to improve adaptive neuro-fuzzy inference system performance in precipitation modelling

Existent algorithms to train adaptive neuro-fuzzy inference system (ANFIS) such as gradient descent and least-square methods are useful, but they have problems such as trapping in local optimum as well as high-volume of computations. This paper examines the application of genetic algorithm (GA), ant colony optimization for the continuous domain (ACOR), particle swarm optimization (PSO) and differential evolution (DE) for improving the performance of ANFIS models in simulating monthly rainfall magnitudes. Monthly rainfall records of nine foothills to arid weather stations of Isfahan province, central Iran were used for evaluating the proposed methodology. The correlation coefficient and non-linear sensitivity analysis were performed on the available rainfall records. Monthly rainfall modelling was carried out based on three input groups and used separately in five applied models (e.g. ANFIS, ANFIS-PSO, ANFIS-ACOR, ANFIS-DE, and ANFIS-GA). The obtained results showed that proposed hybrid models had better accuracy than the simple ANFIS model in escaping local optima. The best model was ANFIS-ACOR (with R2 = 0.92, RMSE = 2.73 mm and SI = 0.26), while sole ANFIS model (with R2 = 0.33, RMSE = 9.98 mm and SI = 0.90) presented the worst output. Moreover, the obtained results showed that: (1) Coupling the optimization algorithms with ANFIS can improve its performance in modeling the monthly rainfall time series; (2) the correlation coefficient and sensitivity analysis of input data provided a valid input selection that could improve the model accuracy with neglecting the redundant input parameters.

SOFT-COMPUTING ESTIMATION OF PRE-TREATED MANGO (MANGIFERA INDICA) KERNEL MOISTURE CONTENT IN A TRAY DRYER

This study presents Adaptive Neuro-fuzzy Inference System (ANFIS) technique for the estimation of pre-treated mango kernel moisture content in a tray dryer. The blanching and drying experiments were conducted at different drying air temperatures (45 -750C), drying time (0 - 420 minutes), blanching temperatures (40 - 1000C) and blanching time (3 - 9 minutes). The experimental input and output (moisture ratio) data were used to architect different ANFIS structures at different epoch numbers, input and output Membership Functions (MFs). The best structure for the estimation was achieved using trap input MF, constant output MF and 1500 epoch number. The Root Mean Squared Error (RMSE) and correlation coefficient R2 showed 0.0097 and 0.958 respectively for the ANFIS structure; furthermore, K value that compares ratio of the best ANFIS checking and training error is 0.87. The results of this investigation show the capability of ANFIS for the estimation of pre-treated mango kernel moisture content in a tray dryer.