Neural Network Logic Research Articles

P1380A SYSTEM FOR FORECASTING OF DRUG DOSAGE FOR ANEMIA CORRECTION IN PATIENTS ON CHRONIC DIALYSIS

Abstract Background and Aims Anemia is a most common complication of CKD in patients on chronic dialysis. Adequacy of anemia correction directly affects both patient’s life quality and patient’s long-term survival. The most important aspect of anemia correction is drug therapy. In this work, we present a system for forecasting of iron supplements and ESA (Epoetin alfa) dosage, that is based on logical rules and artificial intelligence (AI) models. Method As an input dataset, system uses patient’s anthropomorphic parameters, results of laboratory studies, and information on previous drug therapy. The output of the system are dosages of ESA an iron supplements that have to be administered in order to bring the patient’s hemoglobin as close as possible to target range of values (100-120 g/l). The system consists of two sequentially triggering forecasting models (for ESA and for iron supplements), where each model is a combination of logical rules and artificial neural network, that has been trained on a data, collected in more than 20 “Nefrosovet” hemodialysis clinics for the period of 3 years. The effect of system usage was examined for the group of 356 hemodialysis patients with median follow-up time of 4 month. The primary end-point was a number of patients in target range of hemoglobin values (100-120 g/l). Results During the study we determined that as a result of using the dose forecasting system, number of patients in target range of hemoglobin values significantly increased from 239 patients at the beginning of system employment to 266 patients at the end of the study (p=.0318). Furthermore, we observed that there was a concomitant effect of system usage – significant reduction of average monthly ESA dosage from 14300 IU at the beginning of system employment to 13900 IU at the end of the study (p=.0331). Conclusion Employment of automated drug dosage forecasting system based on logical rules and AI models, allows to improve the efficiency of anemia correction in hemodialysis patients and reduce the dosage of administered ESA, which, in turn, reduces the risk of possible complications and treatment cost.

A Case Study on Computer-Based Analysis of the Stochastic Stability of Mechanical Structures Driven by White and Colored Noise: Utilizing Artificial Intelligence Techniques to Design an Effective Active Suspension System

The goal of this research is to design an Artificial Intelligence controller for the active suspension system of vehicles. The Ring Probabilistic Logic Neural Network (RPLNN) architecture has been adopted to design the proposed controller, and the pavement condition has been modelled utilizing Gaussian white noise. The results show that the proposed RPLNN controller has an effective performance to reduce the unwanted stochastic effect of the road profile.

Analysis of vibration high-frequency dynamic characteristics of wheelless rail vehicle system based on fuzzy logic control

Track irregularity is the main source of vibration excitation of vehicle-track weighing system, which has an important impact on the safety, stability and comfort of train operation, and it is also the main factor limiting the speed of train operation. Higher requirements are put forward for track smoothness with the rapid development of China’s railways. Therefore, it has a great theoretical and practical significance to study the relationship between track irregularity and random vibration of vehicle-track cooker-in system and the evaluation method of track smoothness. In this paper, a model construction method is proposed based on machine learning fuzzy logic control and neural network algorithm to analyze the high frequency dynamic characteristics of ballastless track wheel-rail vehicle system. Firstly, a numerical analysis model of high frequency dynamic characteristics of ballastless track wheel-rail vehicle system is established by using linear discrete elastic-yellow damper element connection model; Secondly, the Rough Set Block Neural Network is introduced to optimize the dynamic characteristic analysis model, and the intelligent model of model optimization analysis is established. Finally, the validity of the proposed algorithm is verified by simulation experiments of practical examples.

Underwater Image Enhancement and Object Detection Using Edge Preserving and Multiscale Contextual Neural Network

The submerged perception circumstances cause incredible difficulties to the issue of article location from the low-goals submerged pictures. In this paper, we acquaint an effective strategy with improve the pictures caught submerged and corrupted in light of the medium dispersing and retention. It expands on the mixing of 2 pictures that are legitimately gotten from a shading redressed and white-adjusted adaptation of the first corrupted picture. In the wake of improving the submerged picture, plans to identify object that present in the submerged by utilizing novel edge saving and multiscale logical neural network. We concentrated for the most part on discovery of an item in the submerged that they are utilized to isolate them an article from the foundation by utilizing a mix of programmed difference extending pursued by picture number-crunching task, worldwide edge, and least channel. Our system could be a solitary picture approach that doesn't need particular equipment or information about the submerged conditions or scene structure. our upgraded pictures are described by better exposedness of the dull area, improved worldwide complexity and edge sharpness and our striking article recognition accomplishes both clear identification limit and multi-scale logical vigor at the same time in this manner accomplishes an enhanced presentation.

Fuzzy Logic and Back-Propagation Neural Networks for Optimal Performance

Thus far, the Taguchi technique is found only efficient in obtaining the combination of optimal factor settings when a single product/process response is considered. In today’s dynamic environment, customers are interested in multiple quality responses. This research, therefore, utilizes fuzzy logic and backward-propagation neural networks (BPNNs) to optimize process performance for products of multiple quality responses. In this research, quality characteristics are transformed to signal to noise (S/N) ratios, which are then used as inputs to a fuzzy model to obtain a single common output measure (COM). Next, BPNNs are employed to obtain full-factorial experimental data. Finally, the combination of factor levels that maximizes the average COM value is chosen as the optimal combination. Three case studies are provided for illustration; in all of which the proposed approach provided the largest total anticipated improvement. This indicates that the proposed approach is more efficient than Taguchi-fuzzy, grey-Taguchi, and Taguchi-utility methods. In conclusion, the fuzzy-BPNN approach may greatly assist process/product engineers in optimizing performance with multiple responses in a wide range of business applications.

Adaptive Fault Estimation and Fault-Tolerant Control for Nonlinear System With Unknown Nonlinear Dynamic

This paper investigates the problems of fault estimation and fault-tolerant control for nonlinear system. The nonlinear part of the system is assumed to be unknown. Based on the adaptive approximation technique, which can be performed by fuzzy logic systems or neural networks, an adaptive fault estimation observer is designed, and the fault and the system state can be estimated simultaneously. Based on the estimation information, the observer-based fault-tolerant controller is designed. In this paper, the parameter matrices of the adaptive law and the observer and the controller can be calculated by slowing LMIs. To verify the proposed scheme, a simulation example is provided at the end of this paper.

The control to aggregates of pumping stations using a regulator based on a neural network with fuzzy logic

A pumping station control system is considered using a controller based on a fuzzy logic neural network. The simulation of the classical and fuzzy regulators. The possibility of the implementation of the controller in the form of an adaptive multilayer neural network is shown. The use of the theory of fuzzy sets in combination with the theory of neural networks to create a fuzzy-neural regulator to control pumping units provides a promising approach. Simulation modeling and real operation have shown that fuzzy-logic regulators have a number of advantages over classical regulators, which allow the use of form and limitations. Using the neural network model allows you to add the properties of adaptability and learning. The fuzzy-neural controller for controlling pumping units is promising in terms of efficiency and safety by controlling pumping stations.

Performance Evaluation of Fuzzy Logic and Back Propagation Neural Network for Hand Written Character Recognition

Performance Evaluation of Fuzzy Logic and Back Propagation Neural Network for Hand Written Character Recognition

퍼지 로직과 인공신경망을 이용한 자동차용 연료전지 스택의 지능형 진단방법 개발

퍼지 로직과 인공신경망을 이용한 자동차용 연료전지 스택의 지능형 진단방법 개발

Prediction of Flashover Voltage of High-Voltage Polluted Insulator Using Artificial Intelligence

The objective of this study is to predict the flashover voltage of a high-voltage insulator artificially contaminated using artificial intelligence (AI). First, practical tests were performed on a high-voltage insulator to collect a database used in the implementation of the artificial intelligence concept. These tests were realized for different levels of artificial pollution (saline distilled water). Each pollution level presented an amount of artificial pollution, in milliliters, in each petticoat (zone) of the insulator. The collecting database gives flashover voltage values corresponding to different amounts of artificial pollution in each insulator zone and its conductivity. Second, we have introduced fuzzy logic (FL) and artificial neural networks (ANN) as two AI techniques to predict the flashover voltage of the high-voltage insulator and to estimate the insulating state of artificial pollution. The proposed prediction concepts based on FL and ANN are implemented using MATLAB’s graphical user interface. Finally, a comparison was made between the results obtained by AI and practical ones. The database used in this comparison is different from that used in concepts based on FL and ANN implementation. The obtained results show a high efficacy of FL and ANN techniques in predicting the flashover voltage of high-voltage insulators compared with those obtained by practical tests.

Human Knowledge in Constructing AI Systems — Neural Logic Networks Approach towards an Explainable AI

To build an easy-to-use AI and ML, it is crucial to gain user’s trust. Trust comes from understanding the reasoning behind an AI system’s conclusions and results. The recent research efforts on Explainable AI (XAI) reflect the importance of explainability in responding to the criticism of “black box” type of AI. Neural Logic Networks (NLN) is a research to embed logic reasoning (being binary or fuzzy) to connectionist models having humans’ domain knowledge taken into consideration. The reasoning carried out on such network structures allows possible interpretation beyond binary logic. This article intends to discuss the potential contribution of NLN approach in making reasoning more explainable.

Research on a new teaching quality evaluation method based on improved fuzzy neural network for college English

For the heavy workload and complicated statistics of college English teaching work, the progress and limitations of neural network, and the existing characteristics of fuzzy information, fuzzy logic and RBF neural network are introduced to integrate the advantages of learning, association, identification, adaptation and fuzzy information processing to propose an improved fuzzy RBF neural network model based on back-propagation learning. Then the teaching quality evaluation method of college English based on improved fuzzy neural network is proposed to obtain the more objective and reasonable evaluation result. To test the effectiveness of the teaching quality evaluation method, the college English teaching in Henan University of Chinese Medicine is selected as study case. The results show that the proposed teaching quality evaluation method can effectively overcome the subjectivity and randomness of the traditional teaching quality evaluation methods, and make the evaluation results more in line with the actual situation.

SUPPLY CHAIN RISK MANAGEMENT BY MONTE CARLO METHOD

In this paper, the conceptual model of risk-based cost estimation for completing tasks within supply chain is presented. This model is a hybrid. Its main unit is based on Monte Carlo Simulation (MCS). Due to the fact that the important and difficult to evaluate input information is vector of risk-occur probabilities the use of artificial intelligence method was proposed. The model assumes the use of fuzzy logic or artificial neural networks – depending on the availability of historical data. The presented model could provide support to managers in making valuation decisions regarding various tasks in supply chain management.

ANFIS CONTROLLER FOR AN ACTIVE MAGNETIC BEARINGS TO SUSPEND A FLYWHEEL ENERGY STORAGE SYSTEM

<p><span style="font-family: Times New Roman;">A flywheel energy storage system (FESS) is an effective energy-saving device. It works by accelerating a rotor flywheel disc at a very high speed and maintaining the energy in the system as rotational energy. Active magnetic bearings (AMBs) are ideally suited for use at high-speed and are so used in FESSs. This work develops a mathematical model of the electromagnet force and rotor dynamic of a flywheel. The systems for controlling the position and velocity of the flywheel are designed based on the emerging approaches of an adaptive neuro-fuzzy inference system (ANFIS). Fuzzy logic has occurred as a mathematical tool to deal with the uncertainties in human perception. It also provides a framework for applying approximate human reasoning capabilities to knowledge-based systems. Additionally, ANFIS has emerged as an intelligent controller with learning and adaptive capabilities. ANFIS is combined the fuzzy logic controller (FLC) and neural networks (NNs). In the method that is developed herein, the control model uses Takagi-Sugeno fuzzy logic, in which the back-propagation algorithm processes information from neural networks to adjust suitably the parameters of the fuzzy controller, and the output control signal tracks the input signal. This method can be applied to improve the control performance of nonlinear systems. The output signal responses transient performance of systems use an ANFIS that must be trained through a learning process to yield suitable membership functions and weightings. The results of the FESS indicated that the system responds with satisfactory control performance to reduce overshoot, a zero-error steady-state, and short rise time. The proposed controller can be feasibly applied to FESS with various external disturbances, and the effectiveness of the ANFIS with self-learning and self-improving capacities is proven.</span></p>

ANFIS CONTROLLER FOR AN ACTIVE MAGNETIC BEARINGS TO SUSPEND A FLYWHEEL ENERGY STORAGE SYSTEM

<p><span style="font-family: Times New Roman;">A flywheel energy storage system (FESS) is an effective energy-saving device. It works by accelerating a rotor flywheel disc at a very high speed and maintaining the energy in the system as rotational energy. Active magnetic bearings (AMBs) are ideally suited for use at high-speed and are so used in FESSs. This work develops a mathematical model of the electromagnet force and rotor dynamic of a flywheel. The systems for controlling the position and velocity of the flywheel are designed based on the emerging approaches of an adaptive neuro-fuzzy inference system (ANFIS). Fuzzy logic has occurred as a mathematical tool to deal with the uncertainties in human perception. It also provides a framework for applying approximate human reasoning capabilities to knowledge-based systems. Additionally, ANFIS has emerged as an intelligent controller with learning and adaptive capabilities. ANFIS is combined the fuzzy logic controller (FLC) and neural networks (NNs). In the method that is developed herein, the control model uses Takagi-Sugeno fuzzy logic, in which the back-propagation algorithm processes information from neural networks to adjust suitably the parameters of the fuzzy controller, and the output control signal tracks the input signal. This method can be applied to improve the control performance of nonlinear systems. The output signal responses transient performance of systems use an ANFIS that must be trained through a learning process to yield suitable membership functions and weightings. The results of the FESS indicated that the system responds with satisfactory control performance to reduce overshoot, a zero-error steady-state, and short rise time. The proposed controller can be feasibly applied to FESS with various external disturbances, and the effectiveness of the ANFIS with self-learning and self-improving capacities is proven.</span></p>

Forecasting of financial data: a novel fuzzy logic neural network based on error-correction concept and statistics

First, this paper investigates the effect of good and bad news on volatility in the BUX return time series using asymmetric ARCH models. Then, the accuracy of forecasting models based on statistical (stochastic), machine learning methods, and soft/granular RBF network is investigated. To forecast the high-frequency financial data, we apply statistical ARMA and asymmetric GARCH-class models. A novel RBF network architecture is proposed based on incorporation of an error-correction mechanism, which improves forecasting ability of feed-forward neural networks. These proposed modelling approaches and SVM models are applied to predict the high-frequency time series of the BUX stock index. We found that it is possible to enhance forecast accuracy and achieve significant risk reduction in managerial decision making by applying intelligent forecasting models based on latest information technologies. On the other hand, we showed that statistical GARCH-class models can identify the presence of leverage effects, and react to the good and bad news.

Optimization of Material Removal Rate in Wire-EDM Using Fuzzy Logic and Artifical Neural Network

Present work aims to develop a model and optimize the material removal rate (MRR) in complex wire electric discharge machining (EDM) process. Initially various percentages of aluminium alloy hybrid Nanocomposites were fabricated by novel ultrasonication method. The sample specimens were cut and machined using wire EDM. Experiments were carried out using Taguchi’s L18 orthogonal array under different cutting parameters like Pulse-on, Pulse- off, current and servo voltage. Fuzzy-based Taguchi method and artificial neural network (ANN) with back propagation algorithm were used to optimize the material removal rate. Both ANN and fuzzy logic based models were developed in MATLAB software and the models were trained for estimating the MRR and improving the machining parameter.

A Comparative Study for Position Regulation and Anti-Swing Control of Highly Non-Linear Double Inverted Pendulum (DIP) System Using Different Soft Com

This paper presents a comparative analysis for stabilization and control of highly non-linear, complex and multi-variable Double Inverted Pendulum on cart. A Matlab-Simulink model of DIP has been built using governing mathematical equations. The objective is to control both the pendulums at vertical position while cart is free to move in horizontal direction. The control of DIP was achieved using three different soft-computing techniques namely Fuzzy logic reasoning, Neural networks (NN's) and Adaptive neuro fuzzy inference system (ANFIS). The results show that the ANFIS controller is more effective as compared to other two controllers in terms of settling time (sec), maximum overshoot (degree) and steady state error. The regression (R) and mean square error (MSE) values obtained after training of Neural network were adequate and the training error obtained in ANFIS was also optimum. All the three controllers were able to stabilize the DIP system but ANFIS control provides better results as illustrated with the help of graphs and tables.

Introducing a Novel Hybrid Artificial Intelligence Algorithm to Optimize Network of Industrial Applications in Modern Manufacturing

Recent advances in modern manufacturing industries have created a great need to track and identify objects and parts by obtaining real-time information. One of the main technologies which has been utilized for this need is the Radio Frequency Identification (RFID) system. As a result of adopting this technology to the manufacturing industry environment, RFID Network Planning (RNP) has become a challenge. Mainly RNP deals with calculating the number and position of antennas which should be deployed in the RFID network to achieve full coverage of the tags that need to be read. The ultimate goal of this paper is to present and evaluate a way of modelling and optimizing nonlinear RNP problems utilizing artificial intelligence (AI) techniques. This effort has led the author to propose a novel AI algorithm, which has been named “hybrid AI optimization technique,” to perform optimization of RNP as a hard learning problem. The proposed algorithm is composed of two different optimization algorithms: Redundant Antenna Elimination (RAE) and Ring Probabilistic Logic Neural Networks (RPLNN). The proposed hybrid paradigm has been explored using a flexible manufacturing system (FMS), and results have been compared with Genetic Algorithm (GA) that demonstrates the feasibility of the proposed architecture successfully.

Modelling the effects of petroleum product contaminated sand on the compressive strength of concretes using fuzzy logic and artificial neural networks: A case study of diesel

In this study an artificial neural network (ANN) and fuzzy logic (FL) were used to predict the compressive strength of concrete produced with diesel contaminated sand. Concrete was produced using sand contaminated with diesel at 0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 5 and 10 percent, and each batch cured for 7, 14, 28, 58, 90 and 118 days. The compressive strength of the concretes was measured for each percentage contamination and curing time. Subsequently, an ANN and FL models were developed used to predict the compressive strength of the concrete. The ANN model predictions with a correlation coefficient (R) of 0.99316 predicted better than two FL models which predicted with correlation coefficient (R) values of 0.9086 and 0.8038 respectively. The results show that ANN and FL models could be used to predict the compressive strength of concretes produced with diesel contaminated sand.