Adaptive Neuro-fuzzy Inference Model Research Articles

Prediction of the Electricity Generation of a 60-kW Photovoltaic System with Intelligent Models ANFIS and Optimized ANFIS-PSO

The development and constant improvement of accurate predictive models of electricity generation from photovoltaic systems provide valuable planning tools for designers, producers, and self-consumers. In this research, an adaptive neuro-fuzzy inference model (ANFIS) was developed, which is an intelligent hybrid model that integrates the ability to learn by itself provided by neural networks and the function of language expression, how fuzzy logic infers, and an ANFIS model optimized by the particle swarm algorithm, both with a predictive capacity of about eight months. The models were developed using the Matlab® software and trained with four input variables (solar radiation, module temperature, ambient temperature, and wind speed) and the electrical power generated from a photovoltaic (PV) system as the output variable. The models’ predictions were compared with the experimental data of the system and evaluated with rigorous statistical metrics, obtaining results of RMSE = 1.79 kW, RMSPE = 3.075, MAE = 0.864 kW, and MAPE = 1.47% for ANFIS, and RMSE = 0.754 kW, RMSPE = 1.29, MAE = 0.325 kW, and MAPE = 0.556% for ANFIS-PSO, respectively. The evaluations indicate that both models have good predictive capacity. However, the PSO integration into the hybrid model allows for improving the predictive capability of the behavior of the photovoltaic system, which provides a better planning tool.

Performance improvement of the solid oxide fuel cell using optimal parameters identification through fuzzy logic based-modeling and different optimization algorithms

Fuel Cells are novel devices that have been proposed as new power generation systems. The advantages of solid oxide fuel cells are higher efficiency, higher stability, fuel flexibility, lower emissions, and generally lower cost. In the present study, the fuzzy model is employed to build the model of the solid oxide fuel cell considering various sputtering power, thickness of electrolyte, and temperatures of cell. The maximum iterations for the adaptive neuro-fuzzy inference model was considered 50 iterations. About 3500 samples were applied for the training process, and almost 900 samples were utilized for the testing. After modeling process, the genetic algorithm, particle swarm, simulated annealing, and hybrid firefly-particle swarm optimizers are applied to achieve the optimum value of current and power densities. The results showed that proposed fuzzy model could approximate the model the system with a good agreement with experimental data. Additionally, the obtained data confirm the accuracy, high convergence speed, and robustness of the proposed hybrid optimizer compared to three efficient optimization algorithms. Accordingly, the correlation factor for the proposed fuzzy model for the training and testing dataset was obtained to be 0.9298 and 0.9289, correspondingly.

An integrated approach of ML-metaheuristics for secure service placement in fog-cloud ecosystem

With the proliferation of the Internet of Things (IoT), a huge amount of data is being produced which requires processing. Usually, Cloud computing provides the resources for processing such requests. However, cloud has limitations in terms of latency and security. To overcome, Fog computing came into existence to complement cloud wherein the consolidation of Fog and Cloud computing substantially improves the Quality of Service (QoS). Due to inherent heterogeneity in the Cloud and Fog nodes, selecting an appropriate node becomes a complex task which requires utmost attention. Fog nodes have minimal and limited resources with a thin processing capacity. To utilize the fog nodes optimally, a secure two-step service placement framework has been proposed in this work. The framework consists of the classification of services to decide on the appropriate tier (Cloud/Fog) followed by the scheduling of the services at the Fog tier. For classification, a machine learning based improved adaptive neuro-fuzzy inference model (ANFIS) is implemented, which predicts the suitable computing tier. Further, a novel metaheuristic-based hybrid algorithm that integrates a chaotic-based grasshopper (CGOA) with genetic algorithm (GA) is applied for scheduling at the Fog tier. Basic GOA suffers from several shortcomings, such as falling into local minima, slow convergence, high time complexity, etc. The concept of chaos theory and opposition-based learning is incorporated into GOA to overcome this. The performance of the proposed model has been analysed on the Google trace data set, and its effectiveness has been studied on makespan, computational cost, and energy dissipation of the network where it yields respectively 9.2%, 4.25% and 2.75% average better results than state of art.

Design optimization of proton exchange membrane fuel cell bipolar plate

The bipolar plate geometry design is one of the fuel cell's key features that determines the cell's power. It equally has a direct correlation to the thermal and water management of the cell as it tends to regulate the amount of by-product water that can be expunged from the fuel cell. This study, therefore, explored the development of novel bipolar plate geometry designs, namely the square baffled channel, the rectangular baffled channel, the parallel channel design, and the double serpentine geometry design. This was further compared with the traditional serpentine design to ascertain the design with the optimum fuel cell performance. With the squared baffled channel presenting the best results, varying operating conditions that will influence the performance of the novel fuel cell channel design were also evaluated. It was observed that the hydrogen mass fraction increased by 22.6% for the square baffled channel design compared with the other geometry designs considered in the present study. The square baffle channel showed 12.11% increase in power density and 14.54% increase in current density compared to the rectangular baffle channel. In terms of the parallel channel design, the square baffle showed 18.941% increase in power density and 22.278% increase in current density. The least performing channel geometry design was the double serpentine design. Comparing the double serpentine channel geometry design to the square baffle channel geometry design, there was an increase in current density by 67.72% and 77.88% in terms of power density in favour of the square baffle channel geometry design. The square baffle channel also showed 50% increase in current density and 58.23% increase in power density compared to conventional serpentine channel flow plate geometry design. An adaptive neuro fuzzy inference system (ANFIS) was also adopted to predict the output power of the cell. This was then compared with Feed Forward Back Propagation Neural Network to determine the model with the most accurate results. The adaptive neuro-fuzzy inference model accurately predicted the non-linearities associated with fuel cell performance, hence recommended as ideal for Proton Exchange membrane fuel cell prediction. The main contribution for the study is the development of optimal flow plate geometry design that will ensure maximum fuel cell performance. The current study is aimed at providing technical information to policy makers and the fuel cell industry on how optimization of the flow plate design via the introduction of baffles could increase the cell performance hence accelerates its commercialization and widen their applications in various sectors beyond the automotive industry.

Classification of Sentiment Analysis based on Adaptive Neuro Fuzzy Inference Model (ANFIS)

With the growing demand of Social media and its various networks and with the emergence of virtual nature of Online Shopping users are not easily able to determine whether the quality of the product or service is good or worse. For this work there is no such specific model existed that can find similar or dissimilar products from the people posting tweets in the form of post, messages or hashtags using similar sort of textual content. Also the companies are giving their Product reviews, which contain the direct / Indirect feedback from the customers. The companies are also getting feedback from the social media in the form of tweets / post / blogs from the twitter in the form of text for the same product. So, to find the sentiments of an individual based on the product recommendations the work had been proposed. The reviews from various intelligent algorithm for sentimental classification based on Adaptive Neuro-Fuzzy Inference System were studied. The research work is based on the comparison of the other traditional based Sentiment Analysis methodology with the ANFIS. Therefore, we are able to identify the best method to be used for the reviews identification in the best possible method.

Machine Learning Models for Acoustic Data Prediction During Drilling Composite Lithology Formations

Abstract The rock acoustic data that provide important information about the formation petrophysics and geomechanics are highly needed to design the wells drilling programs, in addition to, reservoir stimulation and field development plans. The acoustic data acquisition through the conventional methods in the petroleum industry either by logging or lab measurements has many drawbacks as the cost of well log operations and the job time. Determining the acoustic data through the common correlations did not provide high accuracy, and there are limitations for using these correlations. The new trend in the petroleum industry with the fourth industrial revolution is to employ machine learning for such problems to provide effective solutions. Therefore, this study utilized the machine learning for developing prediction models for acoustic compressional and shear slowness using adaptive neuro-fuzzy inference system and support vector machines tools. The study presents novel contributions for predicting acoustic slowness from only the surface drilling data while drilling different formations of composite lithology (limestone, sandstone, shale, and carbonate). The study utilized real field data (2800 data points) to build and test the two models through deep sensitivity analysis, in addition, further testing for the models by another 2800 data points from the same field for the validation phase. The obtained results ensured the capability of machine learning for predicting the acoustic slowness with high accuracy as adaptive neuro-fuzzy inference system (ANFIS) achieved a correlation coefficient (R) higher than 0.98 and error less than 1.43% as average absolute percentage error (AAPE) between the actual and predicted acoustic values. Adaptive neuro-fuzzy inference model showed the highest accuracy during the model training as R was 1.0; in addition, R for testing results showed 0.98, and AAPE ranged from 0.25% to 0.92%. The validation phase ensured the high performance for the acoustic prediction of the developed models as R is higher than 0.98 and AAPE lower than 1.46%. The machine learning applications through the developed models for the acoustic data will provide cost and time savings for acoustic data acquisition for the field applications.

An affective learning-based system for diagnosis and personalized management of diabetes mellitus

Diabetes Mellitus is a major health problem with high global morbidity and mortality rates. While, conventional diagnosis methods are based on monitoring blood glucose levels, variables such as body mass index and diastolic blood pressure have been reported as having stable correlations with the incidence and prevalence of diabetes. Recently, machine learning approaches are developed for diagnosis of diabetes, but the existing models are mostly trained and validated on single dataset while the apt strategies needed for proper management of diabetes mellitus are omitted. In this study, we develop an affective learning-based system for diagnosis and therapy of diabetes mellitus. The integrated system consists of a multimodal adaptive neuro-fuzzy inference model designed for diagnosis of diabetes, and a knowledge-based diets recommender model designed for personalized management of diabetes. The diagnosis model was trained and validated with 87.5% and 13.5% of Pima Indians diabetes dataset, respectively; and re-validated with Schorling diabetes dataset; both of which are publicly available. Then, the model was applied retrospectively on a private dataset consisting of 14 female patients’ records obtained at Obafemi Awolowo University Teaching Hospital Complex, Ile-Ife, Nigeria. Also, the recommender model combines users’ diagnoses results with their eating formulae to determine users’ food-per-day consumption and generate weekly personalized food-plan from an expert-designed template. Evaluation results from the studies show that both models performed well. Specifically, the multimodal model attained training and validation accuracies of 83.8% and 79.2%, respectively for Pima dataset, and prediction accuracies of 72.9% and 94.3% for the Schorling and private dataset cases, respectively. In addition, the model shows the best performance when compared with individual baseline models, and ten existing machine learning methods used in related studies. Similarly, the proposed recommender model received the highest average score when compared with several existing diet recommender systems used for chronic diseases therapy. With these promising features, the proposed affective learning-based system could effectively reduce the morbidity and mortality rates of diabetes mellitus in the world.

A novel hybrid learning based Ada Boost (HLBAB) classifier for channel state estimation in cognitive networks

Spectrum sensing, through efficient channel estimation methods utilizing cognitive radio networks, has become an increasingly researched area in recent times. This has become more pronounced in recent times, especially with increasing scarcity in availability of radio frequency spectrum. Wireless state of the art communication standards demand high bandwidth to provide seamless connectivity and high degree of mobility, which requires more of radio frequency band for functioning. Hence, intelligent methods of spectrum allocation have been an increasing challenge in recent times. This paper proposes a hybrid learning based-Ada Boost classifier model for efficient spectrum allocation through channel state estimation through a learning and double classification approach. The proposed algorithm has been experimented in a high bandwidth characterized 5G communication simulation settings and observed for its performance measures namely collision rate analysis, throughput, probability of detection, false alarm detection and bit error rate. The proposed technique has been compared against benchmark techniques such as conventional fast Fourier transform based energy detector, fuzzy cognitive engine and adaptive neuro fuzzy inference model without Ada Boost and found to exhibit superior performance in all performance measures. The proposed technique exhibits a spectral efficiency of nearly 90% and considered to be a suitable spectrum sensing scheme for high bandwidth and narrowband utilities.

Improving precision in the prediction of laser texturing and surface interference of 316L assessed by neural network and adaptive neuro-fuzzy inference models

Laser-based surface texturing provides highly controlled interference fit between two parts. In this work, artificial intelligence-based models were used to predict the surface properties of laser processed stainless steel 316 samples. Artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) were used to predict the characteristics of laser surface texturing. The models based on feedforward neural network (FFNN) were developed to examine the effect of the laser process parameters for surface texturing on 316L cylindrical pins. The accuracy of the models was measured by calculating the root mean square error and mean absolute error. The reliability of the ANFIS and FFNN models for the output prediction of the laser surface texturing (LST) system were investigated by using the data measured from experiments based on a 3^3 factorial design, with main processing parameters set as laser power, pulse repetition frequency, and percentage of laser spot overlap. The relative assessment of the models was performed by comparing percentage error prediction. Finally, the impact of input data was examined using predicted response surface plots. Results showed that ANFIS prediction was 48% more accurate compared with that provided by the FFNN model.

An Adaptive Neuro-Fuzzy Inference Model to Predict Punching Shear Strength of Flat Concrete Slabs

An adaptive neuro-fuzzy inference system (ANFIS)-based model was developed to predict the punching shear strength of flat concrete slabs without shear reinforcement. The model was developed using a database collected from 207 experiments available in the existing literature. Five key input parameters were used to build the model, which were slab effective depth, concrete strength, reinforcement ratio, yield tensile strength of reinforcement, and width of square loaded area. The output parameter of the model was punching shear strength. The results from the adaptive neural fuzzy inference model were compared to those from the simplified punching shear equations of ACI, BS-8110, Model Code 2010, Euro-Code 2, and also experimental results. The root mean square error (RMSE) and the correlation coefficient (R) were used as evaluation criteria. Parametric studies were presented using ANFIS to assess the effect of each input parameter on the punching shear strength and to compare ANFIS results to those from the equations proposed in commonly used codes. The results showed that the ANFIS model is simple and provided the most accurate predictions of the punching shear strength of two-way flat concrete slabs without shear reinforcement.

An adaptive neuro–fuzzy inference model for prediction of coal char conversion in CO2 gasification

ABSTRACTToday, in industry, the gasification of coal as an important energy source is an interesting perspective. In this study, the application of adaptive neuro–fuzzy inference system (ANFIS) technique for estimation of char conversion in CO2 gasification is investigated. The main variables affecting char conversion are particle size, reaction time, and reaction temperature, which are chosen as input variables of the proposed model. Experimental data which are gathered from the literature are applied for training, testing, and validation of developed ANFIS model. The results reveal the exact estimation of char conversions with the corresponding experimental values with the regression coefficients (R2) greater than 0.99.

Preliminary estimation of rock mass strength using diamond bit drilling operational parameters

In many mining engineering applications such as prospecting, development, production and grouting, diamond bit drilling is widely used due to high penetration rate, core recovery and its ability to drill with less deviation. It has been well known that the operational parameters of diamond bit drilling are closely related with rock mass strength properties. One of the most widely discussed subjects in drilling is the possibility of using diamond drill bit operational parameters for preliminary estimation of rock mass strength and deformability properties used in many mining engineering design projects. Once such rock properties are estimated, it will be possible to make tactical planning decisions as mining progresses. In this study two different techniques, multiple regression and adaptive neuro-fuzzy inference system (ANFIS) were used to develop the models for preliminary estimation of rock mass strength. The variables used in the models are widely known and recorded operational parameters of diamond bit drilling such as bit load, bit rotation and penetration rate. To develop the models, a database covering the rock properties and the machine operational parameters collected from seven different drill holes in Turkey was constructed. Results indicate that both regression and ANFIS-based models can successfully be used to predict the rock mass strength. Adaptive neuro-fuzzy inference model exhibits better performance according to statistical performance indicators. By means of the developed models it is possible to estimate the strength of rock mass during drilling operation, especially in weak and highly fractured rock masses. The estimated strength parameters can be related to further mining engineering applications such as the assessment of excavatability, blast design and even mine design studies.

Penetration rate prediction for diamond bit drilling by adaptive neuro-fuzzy inference system and multiple regressions

In many mining, civil, and petroleum engineering applications diamond bit drilling is widely used due to high penetration rate, core recovery and its ability to drill with less deviation. Recently, many research have been conducted to estimate the penetration rate of diamond drilling which can be considered as one of the most important parameters in project planning and cost estimation of the operation.A database covering the rock properties and the machine operational parameters collected from seven different drilling sites in Turkey is constructed. Construction of an adaptive neuro-fuzzy inference system and the multiple regression models for predicting the penetration rate of diamond drilling is described. In the models, rock properties such as the uniaxial compressive strength, the rock quality designation, and the equipment operational parameters like bit load and bit rotation are considered. Although the prediction performance of multiple regression models is high, the adaptive neuro-fuzzy inference model exhibits better performance based on the comparison of performance indicators. By using the models, penetration rate of diamond bit drilling can be predicted effectively.

Adaptive neuro-fuzzy inference models for speech and video quality prediction in real-world mobile communication networks

This article presents a unified Quality of Service (QoS) prediction methodology based on neuro-fuzzy inference systems that can be used in contemporary rollout mobile communication networks. Our work is concentrated on radio key performance indicators of mobile radio access networks that affect speech and video quality of wireless multimedia communications, and on how we can estimate Quality of end-user Experience (QoE) using fuzzy-based techniques. We propose a methodology that is based on modern experimental drive-test equipment with which a measurement campaign is configured and conducted in various environments. Afterwards, ANFIS models are developed based on real network measurements and numerical results are presented.

Prediction of the performance of impact hammer by adaptive neuro-fuzzy inference system modelling

Impact type excavators are widely used for excavations, performed in weak-laminated-foliated-anisotropic rocks. Therefore the prediction of the performance of impact hammer is very important in many mining and civil engineering projects. This paper describes the construction of adaptive neuro-fuzzy inference system model for predicting the performance of impact hammer type excavator by considering rock and excavating machine properties such as block punch strength index, geological strength index system and impact hammer power. Extensive field and laboratory studies were conducted in the tunnel construction route of the second stage of Izmir Metro Project, which excavated in laminated-foliated flysch rocks. The results of the constructed adaptive neuro-fuzzy inference system and traditional multiple regression models were compared. Although the prediction performance of traditional multiple regression model is high, it is seen that adaptive neuro-fuzzy inference model exhibits better prediction performance according to statistical performance indicators. By means of the developed model, the performance of impact type excavators can be predicted in terms of net excavation based on the selected rock and machine properties.

Adaptive neurofuzzy ANFIS modeling of laser surface treatments

This paper introduces a new ANFIS adaptive neurofuzzy inference model for laser surface heat treatments based on the Green’s function. Due to its high versatility, efficiency and low simulation time, this model is suitable not only for the analysis and design of control systems, but also for the development of an expert real time supervision system that would allow detecting and preventing any failure during the treatment.