Reinforcement Learning Based Neuro-fuzzy Controller for Coffee Roasting Process

Prophesying egg production based on energy consumption using multi-layered adaptive neural fuzzy inference system approach

This study investigated the accurate prediction of the calorific value of municipal solid waste (MSW) using soft computing systems, namely artificial neural networks (ANN), adaptive neural fuzzy inference system (ANFIS), support vector machine (SVM), and multi-layer perceptron (MLP). Calorific value of MSW is a crucial factor that exhibits the energy content of MSW, however, determining calorific value from conventional laboratory methods is quite expensive and difficult. The research focused on proximate analysis parameters obtained from the laboratory and utilized the measured calorific value to develop predictive models. All the models demonstrated a very good correlation between input and output, with consistently strong values of the coefficient of determination (R2). ANN SVM, MLP, and ANFIS models have respective R2 values of 0.9397, 0.8195, 0.7212, and 0.9979. ANFIS showed the best correlation with exceptional predictive power. Statistical parameters were determined to compare model accuracy, with ANFIS exhibiting the top performance, followed by ANN, and then MLP, which had the lowest values of mean square error (MSE), root mean square error (RMSE), mean absolute deviation (MAD), and mean absolute percentage error (MAPE) at 8.704E-07, 0.00019, 0.00016, and 1.295E-05 respectively. However, SVM has the least capability to predict calorific values accurately compared to other models. Soft computing Models, particularly ANFIS, demonstrated remarkable accuracy in predicting the calorific value.

This work focuses on the correlations among water quality parameters such as total hardness, total dissolved solids, electrical conductivity and total alkalinity with water parameters pH, temperature and the sum of mill equivalents of cations and anions in water reservoir in Kermanshah province (located in the middle of the western part of Iran). The data of water quality of monitoring sites were collected over year 2015. To predict and simulate water quality parameters, two data-driven models, i.e., adaptive neural fuzzy inference system and a hybrid adaptive neural fuzzy inference system structure trained by particle swarm optimization technique, were used. The main advantages of these methods are their high accuracy and very fast computational speed to predict unknown data. The results indicated that implementation of two models is highly satisfactory for predicting inorganic indicators of water quality. However, the flexibility of particle swarm optimization–adaptive neural fuzzy inference system method in modeling is better than adaptive neural fuzzy inference system approach. To prove this, the correlation coefficient, mean absolute error, root mean square error and t statistics were calculated as the error criterion. The overall (training and testing) mean relative error percentage, mean absolute error, root mean square error, correlation coefficient and t statistics obtained by the proposed particle swarm optimization–adaptive neural fuzzy inference system model are less than 3.50, 11.60, 18.90, 0.95 and 0.38%, respectively. The results provide a useful approach that uses water parameters to estimate water quality in water reservoir for water treatment and pollution management.

The flame retardancy of polymer materials is closely linked to the types and compositions of flame retardants used. However, due to the complex relationships involved, it is difficult to quantify and predict flame retardancy using linear equations and models. Alkyl phosphinate flame retardants have emerged as promising candidates for their environmentally friendly, low toxicity, and excellent flame retardant properties. In this study, we aimed to develop a novel, low‐cost alkyl phosphinate flame retardant by predicting the quantitative impact of flame retardant composition on flame retardancy. To establish prediction models for the limiting oxygen index (LOI) value of alkyl phosphinate flame retardants, we utilized artificial neural network (ANN) and adaptive neural fuzzy inference system (ANFIS) programs, namely ANN, ANFIS‐grid partition (GP), and ANFIS‐subtractive clustering (SC). The prediction models were trained and validated, and their reliability was assessed based on the root mean square error (RMSE), coefficient of determination (R2), and mean absolute percentage error (MAPE) criteria. Our results indicate that the ANFIS‐SC prediction model exhibits the highest reliability, with high accuracy and ideal prediction results.

The fluctuations in solar irradiance and temperature throughout the year require an accurate methodology for forecasting the generated current of a PV system based on its specifications. The optimal technique must effectively manage rapid weather fluctuations while maintaining high accuracy in forecasting the performance of a PV panel. This work presents a comparative examination of supervised learning algorithms optimized with particle swarm optimization for estimating photovoltaic output current. The empirical formula’s measured currents are compared with outputs from various neural networks techniques, including feedforward neural networks (FFNNs), the general regression network known as GRNN, cascade forward neural networks also known as CFNNs, and adaptive fuzzy inference systems known as ANFISs, all optimized for enhanced accuracy using the particle swarm optimization (PSO) method. The ground data utilized for these models comprises hourly irradiations and temperatures from 2023, sourced from several places in South Africa. The accuracy levels indicated by statistical error margins from the root mean square error (RMSE), mean bias error (MBE), and mean absolute percentage error (MAPE) imply a universal enhancement in the algorithms’ precision upon optimization.

Within the domain of cybersecurity, Intrusion Detection Systems (IDSs) are crucial to protecting network infrastructures from hostile actions. This paper presents a novel approach that leverages Particle Swarm Optimization (PSO) for feature selection to enhance the performance of an Adaptive Neuro-Fuzzy Inference System (ANFIS)-based IDS. The PSO algorithm identifies the most significant features of a dataset, reducing dimensionality and computational complexity. Subsequently, the selected features are utilized by the ANFIS model to detect and classify intrusions with greater accuracy and efficiency. The proposed PSO-ANFIS framework outperforms traditional methods in terms of detection accuracy and false positive rate, as evidenced by experimental results. The combination of PSO for feature selection with ANFIS for intrusion detection offers a solution to cybersecurity issues, especially in dynamic and intricate systems such as the Internet of Battlefield Things (IoBT).

Three-dimensional realizations of flood flow in large-scale rivers using the neural fuzzy-based machine-learning algorithms

Modeling output energy based on fossil fuels and electricity energy consumption on dairy farms of Iran: Application of adaptive neural-fuzzy inference system technique

With the wide application of batteries, the battery state of health (SOH) is increasingly concerned by people. The battery SOH reflects its ability to store charge and plays a key role in electric power systems. This paper proposes an adaptive neural fuzzy inference system (ANFIS) based method to estimate the battery SOH. The main steps of the proposed method include model creating, model training and SOH estimation. The constant current charging time, the voltage drop at the beginning of discharge and the released energy within a certain depth of discharge are used as the inputs for the model training. Since the basic ANFIS algorithm has large estimation errors and the model trains slowly, A Fletcher-Reeves based method is proposed to improve the basic ANFIS. The trained model is then used for the estimation of the SOH. Experimental results show that the Fletcher-Reeves based ANFIS method is effective and efficient.

This paper presents a Neuro-Fuzzy adaptive controller for speed control of a three phase direct torque controlled induction motor drive. The Direct Torque Control (DTC) scheme is one of the most advanced methods for controlling the flux and electromagnetic torque of machines. Control of electromagnetic torque/speed in these drives for high performance applications requires a highly robust and adaptive controller. Adaptive Neural-Fuzzy Inference System (ANFIS) is a hybrid between Artificial Neural Networks (ANN) and Fuzzy Logic Control (FLC) that enhances the execution of direct torque controlled drives and overcomes the difficulties in the physical implementation of high performance drives. MATLAB/SIMULINK implementation of 15 hp, 50 Hz, 4 pole squirrel cage induction motor controlled with the DTC scheme is presented in this paper. The PI controller used for speed control in conventional DTC drives is substituted by the ANFIS based controller. Simulation results show the use of ANFIS decreases the response time along with reduction in torque ripples.

In recent times, gold has been one of the prioritized commodities in terms of long term as well as short term investments since the investors consider gold as a hedgerow against the unforeseen events leading to chaos in the market. Consequentially, the price of gold in the market plays an important role. In this research work, time-series gold price prediction models have been developed using the support vector regression and anfis models for the prediction of daily gold prices. The support vector model was designed using epsilon support vector regression method while the adaptive neural fuzzy inference systems have been developed using grid partition and subtractive clustering methods. The gold prices obtained for the training and testing were obtained from Perth Mint of Australia. The evaluation criteria for the comparison of the models are Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Nash-Sutcliffe model efficiency coefficient (E) and Mean absolute percentage error (MAPE). It was observed that the models obtained using support vector regression outperformed the ANFIS models. In the ANFIS models, it was observed that ANFIS-GP performed slightly better than the ANFIS-SC model.

Landfill leachate is one of the sources of surface water pollution in Selangor State (SS), Malaysia. Leachate volume prediction is essential for sustainable waste management and leachate treatment processes. The accurate estimation of leachate generation rates is often considered a challenge, especially in developing countries, due to the lack of reliable data and high measurement costs. Leachate generation is related to several variable factors, including meteorological data, waste generation rates, and landfill design conditions. Large variations in these factors lead to complicated leachate modeling processes. The aims of this study are to determine the key elements contributing to leachate production and then develop an adaptive neural fuzzy inference system (ANFIS) model to predict leachate generation rates. Accuracy of the final model performance was tested and evaluated using the root mean square error (RMSE), the mean absolute error (MAE), and the correlation coefficient (R). The study results defined dumped waste quantity, rainfall level, and emanated gases as the most significant contributing factors in leachate generation. The best model structure consisted of two triangular fuzzy membership functions and a hybrid training algorithm with eight fuzzy rules. The proposed ANFIS model showed a good performance with an overall correlation coefficient of 0.952.

Prediction and simulation of monthly groundwater levels by genetic programming

Cement is an indispensable raw material of civil construction works although manufacturing methodology with massive energy and heat transfer impacts to environment and human health adversely. Tracking of flue gas emissions continuously and investing online supervisory control systems is mandatory to comply with instructions of regulations and standards imposing to monitor, control, and limit air pollutants, and sustainability. In this study, sulfur dioxide content in effluent gas of Turkish cement factory, located in Kocaeli, is modeled using 9 input parameters and 3 alternative forecasting systems are investigated to solve complex environmental engineering problem. Demanding neuro-fuzzy-based soft computing, ANFIS (Adaptive neural fuzzy inference system), artificial intelligence tool, ANN (Artificial neural networks), and statistical method, MLR (Multiple linear regression), are applied and output data is compared using RMSE (Root mean squared error) term and R2 (Coefficient of determination). Prediction outputs reveal that proposed ANFIS air pollution forecasting system is found highly successful and promising with accurate R2 and RMSE rates, 0.95, and 1.603, respectively. Further, offline-online simulator or expert modeler using the ANFIS forecasting software is recommended to be combined with existing plant control systems to measure efficiency as future study.

This paper presents Artificial Neural Networks (ANN) and Adaptive Neural-Fuzzy Inference System (ANFIS) for reduction of torque and flux ripples in transient and steady state response of Direct Torque Control (DTC) for Induction Motor drive. The Flux and Electromagnetic torque can be controlled by using efficient Direct Torque Control (DTC) scheme This proposed technique is to improve the torque, speed and flux response with the Artificial Neural Network (ANN) and then with the Adaptive Neuro-Fuzzy Inference (ANFIS). This paper shows implementation of DTC system using ANN and ANFIS on three phase induction motor to optimize the flux and to improve the performance of fast stator flux response in transient state. To improve the performance of DTC with the modern technique using ANN and ANFIS approach is implemented and performance of ANN DTC compared with CDTC and ANN DTC with ANFIS is done, conclusion is about the ANN approach shows the better performance than CDTC and ANFIS shows superior performance than ANN. The performance has been tested by using MATLAB/SIMULINK and NEURAL NETWORK toolbox.