Structure Of Fuzzy Neural Network Research Articles

Obstacle Avoidance Method for Wheeled Mobile Robots Using Interval Type-2 Fuzzy Neural Network

This paper proposes an obstacle avoidance method in the position stabilization of the wheeled mobile robots using interval type-2 fuzzy neural network (IT2FNN). Previously, we have proposed the unified strategies of obstacle avoidance and shooting method of the robot soccer system using type-1 fuzzy neural network (T1FNN). Even though the previous T1FNN method can achieve the required tasks, the performance of the previous T1FNN method is not satisfactory in the following sense. The previous T1FNN cannot reduce the influence of uncertainties effectively because it uses the crisp set as the membership values. In addition, it can result in the large oscillation behavior during the obstacle avoidance. Accordingly, we should design the IT2FNN method to improve the performance with smoother behavior as well as improved obstacle avoidance. The proposed IT2FNN method has the fuzzy neural network structure different from the T1FNN. Since the IT2FNN uses the fuzzy set instead of the crisp set as the membership values and it is robust against uncertainties, the performance of the robot behavior can be significantly improved especially in the presence of obstacles. Both simulation and experimental results using the actual wheeled mobile robot with the vision information are provided to show the validity and the advantages of the proposed method.

A Lyapunov stable type-2 fuzzy wavelet network controller design for a bilateral teleoperation system

A bilateral teleoperation controller should make the system stable and achieve optimal performance in the presence of time delays, plant disturbances, measurement noise, modeling errors, and unknown environments. This paper proposes a new method of controller design based on a type-2 fuzzy wavelet neural network structure. A new algorithm based on gradient descent is used. The controller has the advantage of fuzzy controllers and wavelet neural network controllers to allow design of a bilateral teleoperation controller. The salient characteristics of the T2FWNN controller are that the system is constructed on the basis of type-2 membership functions to handle uncertainties associated with information and data in the knowledge base; it can handle unknown environmental interactions; it will remove most noise added to data transmitted through a communication channel. The algorithm of the proposed controller is Lyapunov stable and tracking error is negligible. The proposed controller is compared with a type-1 fuzzy wavelet neural network controller and a conventional fuzzy controller and simulation results show the efficacy of the proposed method.

Performance Verification of a Fuzzy Wavelet Neural Network in the First Order Partial Derivative Approximation of Nonlinear Functions

Approximation of the first order partial derivative of a function modeled from a set of discrete data is the requirement of several applications. However, using a direct method for calculating the partial derivative from a set of discrete points is preferred rather differentiating the function which is obtained by modeling the discrete dataset. In this paper, the first order partial derivative of a fuzzy wavelet neural network structure is calculated to act as a direct differentiator. The structure of the network is described and its parameters are tuned by an adaptive gradient-based back propagation learning algorithm. It is shown that the proposed model outperforms the adaptive neuro-fuzzy inference-based and feed forward neural network-based differentiators in approximating the first order partial derivatives of multi-variable nonlinear functions.

Self-adaptive prediction of cloud resource demands using ensemble model and subtractive-fuzzy clustering based fuzzy neural network.

In IaaS (infrastructure as a service) cloud environment, users are provisioned with virtual machines (VMs). To allocate resources for users dynamically and effectively, accurate resource demands predicting is essential. For this purpose, this paper proposes a self-adaptive prediction method using ensemble model and subtractive-fuzzy clustering based fuzzy neural network (ESFCFNN). We analyze the characters of user preferences and demands. Then the architecture of the prediction model is constructed. We adopt some base predictors to compose the ensemble model. Then the structure and learning algorithm of fuzzy neural network is researched. To obtain the number of fuzzy rules and the initial value of the premise and consequent parameters, this paper proposes the fuzzy c-means combined with subtractive clustering algorithm, that is, the subtractive-fuzzy clustering. Finally, we adopt different criteria to evaluate the proposed method. The experiment results show that the method is accurate and effective in predicting the resource demands.

Computer Simulation Technology of Electric Power Safety Based on Fuzzy Neural Network

As the basis industry of the national economy electric power enterprise is shouldering significant social responsibility in the process of operation, also needs to face the operational risk generated by enterprise competition under the conditions of market economy. How to scientifically and efficiently manage the financial risks of power enterprises is one of the hot issues that are urgent needed to resolve in the current field. In this paper, on the basis of previous studies, firstly has combined with the structure characteristics of the fuzzy neural network model. Then it builds prediction analysis model of financial risks according to the fuzzy neural network structure. And it sets the selection of the number of neuron for the hidden layer based on the financial risks' characteristics of electric power enterprise. At last, it combines with 12 financial indicators data of electric power enterprise finance to make further computer simulation, so as to verify the scientificity of the model. And the results show that the model has strong reliability and a strong practical value.

Study on Parameters Optimization Method of Fuzzy Neural Network PID Controller

In allusion to the insufficient of the traditional parameters optimization of the traditional fuzzy neural network PID controller, the parallel search characteristics of the ant colony algorithm in the whole parameter space is used. A parameters optimization method of the PID controller based combining the ant colony algorithm and fuzzy theory and neural network is proposed in this paper. The method used the ant colony algorithm to comprehensively optimize the parameters and structure of fuzzy neural network, which to be used to train and determine the parameters of the PID controller in order to get the fuzzy neural network PID controller. This method is used in the practical application of nonlinear coupled system, the experimental results show that the optimized fuzzy neural network PID controller takes the faster approximation control objectives, the shorter response time, the smaller overshoot and higher control accuracy. Consequently, the research has the theoretical significance and practical application value.

Research of Fuzzy Neural Network in the Fault Diagnosis of Numerically-Controlled Machine Tool

At present, our rapid development of a processing industry requires the increasing number of numerically-controlled machine tool. At the same time, people grow the attention of diagnosing the fault in order to repair the machine one time. This article starts the research of fault diagnosis based on fuzzy neural network. At first, it evaluates the structure and learning method of fuzzy neural network. In the second place, this paper analyzes the numerically-controlled machine tool in details, and summarizes the fault and reason. At last, this article makes the comprehensive research of neural network and fault diagnosis and provides part of the code. It has positive guiding significance to the worker of numerically-controlled machine tool.

Predicting the risk of squamous dysplasia and esophageal squamous cell carcinoma using minimum classification error method

Early detection of squamous dysplasia and esophageal squamous cell carcinoma is of great importance. Adopting computer aided algorithms in predicting cancer risk using its risk factors can serve in limiting the clinical screenings to people with higher risks. In the present study, we show that the application of an advanced classification method, the Minimum Classification Error, could considerably enhance the classification performance in comparison to the logistic regression model and the variable structure fuzzy neural network, as the latest successful methods. The results yield the accuracy of 89.65% for esophageal squamous cell carcinoma, and 88.42% for squamous dysplasia risk prediction.

Fuzzy Neural Networks Learning by Variable-Dimensional Quantum-behaved Particle Swarm Optimization Algorithm

The evolutionary learning of fuzzy neural networks (FNN) consists of structure learning to determine the proper number of fuzzy rules and parameters learning to adjust the network parameters. Many optimization algorithms can be applied to evolve FNN. However the search space of most algorithms has fixed dimension, which can not suit to dynamic structure learning of FNN. We propose a novel technique, which is named the variable-dimensional quantum-behaved particle swarm optimization algorithm (VDQPSO), to address the problem. In the proposed algorithm, the optimum dimension, which is unknown at the beginning, is updated together with the position of swarm. The optimum dimension converged at the end of the optimization process corresponds to a unique FNN structure where the optimum parameters can be achieved. The results of the prediction of chaotic time series experiment show that the proposed technique is effective. It can evolve to optimum or near-optimum FNN structure and optimum parameters. DOI: http://dx.doi.org/10.11591/telkomnika.v11i10.2960

Design on Fuzzy Neural Network PID Control System of Diesel Engine

The diesel engA control system of diesel engine speed regulation based on the structure of fuzzy neural network is proposed in this paper, which is combined with PID controller in order to regulate PID parameters dynamically. The simulation result proves that the system has good control capability and robust stability, as well as good self-study and self-adaptive ability.

A type-2 fuzzy wavelet neural network for system identification and control

This paper proposes a novel, type-2 fuzzy wavelet neural network (type-2 FWNN) structure that combines the advantages of type-2 fuzzy systems and wavelet neural networks for identification and control of nonlinear uncertain systems. The proposed network is constructed on the base of a set of fuzzy rules that includes type-2 fuzzy sets in the antecedent part and wavelet functions in the consequent part. For structure identification, a fuzzy clustering algorithm is implemented to generate the rules automatically and for parameter identification the gradient learning algorithm is used. The effectiveness of the proposed system is evaluated for identification and control problems of time-invariant and time-varying systems. The results obtained are compared with those obtained by the use of type-1 FWNN based systems and other similar studies.

Adaptive variable structure fuzzy neural identification and control for a class of MIMO nonlinear system

This paper presents a novel adaptive variable structure (AVS) method to design a fuzzy neural network (FNN). This AVS-FNN is based on radial basis function (RBF) neurons, which have center and width vectors. The network performs sequential learning through sliding data window reflecting system dynamic changes, and dynamic growing-and-pruning structure of FNN. The salient characteristics of the AVS-FNN are as follows: (1) Structure-learning and parameters estimation are performed automatically and simultaneously without partitioning input space and selecting initial parameters a priori. The structure-learning approach relies on the contribution of the size of the output. (2) A set of fuzzy rules can be inserted or reduced during the learning process. (3) The connection weighting factors between the deduction layer and output layer generated quickly without resorting to iteration learning are updated by the least-squares algorithm. The proposed method effectively generates a fuzzy neural model with a highly accurate and compact structure. Simulation results demonstrate that the proposed AVS-FNN has a self-organizing ability, which can determine the structure and parameters of the FNN automatically. The application of this new approach has been applied successfully in the 3 DOF helicopter systems, showing the effectiveness and potential of the proposed design techniques.

Performance comparisons of intelligent load forecasting structures and its application to energy-saving load regulation

This study mainly focuses on the development of intelligent forecasting structures via a similar time method with historical load change rates for the hourly, daily and monthly load forecasting simultaneously based on the basic frameworks of fuzzy neural network (FNN) and particle swarm optimization (PSO). In the regulative aspect of network parameters, conventional back-propagation (BP) and PSO tuning algorithms are used, and varied learning rates are designed in the sense of discrete-time Lyapunov stability theory. The performance comparisons of different intelligent forecasting structures including neural network (NN) structure with BP tuning algorithm (NN-BP), FNN structure with BP tuning algorithm (FNN-BP), FNN structure with BP tuning algorithm and varied learning rates (FNN-BP-V), FNN structure with PSO tuning algorithm (FNN-PSO) and newly-designed adaptive PSO (APSO) structure are verified by numerical simulations. In order to verify the effectiveness of the superior APSO forecasting structure in practical energy-saving load regulation, the load forecasting during every 15 min is also given, and its result is used to manipulate the scheduled unloading control of a real case in Taiwan campus.

模糊神经网络在网络中心机房温度调节中的应用研究

网络中心机房的温度是大延时、大惯性的被控对象，针对此特点，以模糊控制和神经网络结合的思想，设计了一种基于模糊神经网络的控制器。建立了以T-S模糊模型为基础的5层模糊神经网络结构，并采用改进的BP算法和最小二乘法为模糊神经网络的学习算法。仿真结果表明，该方法下系统响应时间短、超调量小且稳态精度高，有较好的动态品质、稳定性和抗干扰性。 The temperature of network center room is large time delay, great inertia controlled object. In view of the characteristics, combined with fuzzy control and neural network, a kind of fuzzy neural network controller was de-signed. On the basis of T-S fuzzy model, five-layers fuzzy neural network structure was established. The improved BP algorithm and least square method for fuzzy neural network learning algorithm was used. The simulation results show that system response time is short and small overshoot and steady state of high accuracy, good dynamic quality and sta-bility and anti-jamming.

Designing a Soft Sensor with the Weighted Fuzzy Neural Network

Soft sensors are algorithms capable of estimate the process output that can not be measured directly in real time. A data-driven soft sensor is an inferential model developed from process observations. In this paper, the soft sensor modeling process based on the weighted fuzzy neural network was discussed. The proposed algorithm based on genetic algorithm and particle swarm optimization could obtain a near-optimal structure of fuzzy neural network, and the numerical experiments show that the soft sensor model has good performance.

Study on Genetic Fuzzy Wavelet Neural Network Controller of Robotic Manipulator

This paper deals with the tracking controller design of robotic manipulator using genetic algorithm (GA). A genetic fuzzy wavelet neural network (GFWNN) controller is designed and implemented based on MATLAB in this paper, whose parameters are optimized by GA. The structure and algorithm of fuzzy wavelet neural network (FWNN) are described at first. Then the key content of GA used in this paper and the steps for using GA to optimize FWNN are demonstrated. Finally, a numerical simulation of tracking control for 2-link robotic manipulator is given to verify the effectiveness of the proposed method.

A variable structure fuzzy neural network model of squamous dysplasia and esophageal squamous cell carcinoma based on a global chaotic optimization algorithm

Identification of squamous dysplasia and esophageal squamous cell carcinoma (ESCC) is of great importance in prevention of cancer incidence. Computer aided algorithms can be very useful for identification of people with higher risks of squamous dysplasia, and ESCC. Such method can limit the clinical screenings to people with higher risks.Different regression methods have been used to predict ESCC and dysplasia. In this paper, a Fuzzy Neural Network (FNN) model is selected for ESCC and dysplasia prediction. The inputs to the classifier are the risk factors. Since the relation between risk factors in the tumor system has a complex nonlinear behavior, in comparison to most of ordinary data, the cost function of its model can have more local optimums. Thus the need for global optimization methods is more highlighted. The proposed method in this paper is a Chaotic Optimization Algorithm (COA) proceeding by the common Error Back Propagation (EBP) local method. Since the model has many parameters, we use a strategy to reduce the dependency among parameters caused by the chaotic series generator. This dependency was not considered in the previous COA methods. The algorithm is compared with logistic regression model as the latest successful methods of ESCC and dysplasia prediction. The results represent a more precise prediction with less mean and variance of error.

Application of genetic algorithms in fuzzy wavelet neural network for fetal electrocardiogram extraction

In this paper, we propose a novel fuzzy wavelet neural networks method to extract fetal electrocardiogaram from cutaneous potential abdominal and noise contaminations electrocardiogram recordings. As the fuzzy wavelet neural networks is adaptive to the non-linear and time-varying features of electrocardiogram signal therefore, the fuzzy wavelet neural networks has been used to extract the fetal electrocardiogaram signal. Using this fuzzy wavelet neural networks approach, the maternal electrocardiogram has been suppressed from the abdominal electrocardiogram by correlation detraction, so that the output can be considered as only fetal electrocardiogaram. The structure of fuzzy wavelet neural networks is based on the basis of fuzzy rules including wavelet functions in the consequent parts of rules. In order to improve the extraction fetal electrocardiogaram accuracy and general capability of the fuzzy wavelet neural networks system, an efficient genetic algorithm approach is used to adjust the parameters of dilation, translation, weights, and membership functions.

Modeling of Distillation Tower Temperature Based on D-FNN

Distillation temperature control system is characteristics of nonlinear time-varying and we use dynamic fuzzy neural network to model the temperature of distillation. Firstly, we introduce the structure and algorithm of dynamic fuzzy neural network; Second, after data preprocessing of distillation process, we use dynamic Fuzzy neural network modeling the temperature of distillation. Dynamic fuzzy neural network adopt dynamic learning algorithm, and characteristic of approximation. The simulation results show the effect and accuracy of Dynamic fuzzy neural network model ing method.

The Research and Application of the Fuzzy Neural Network Control Based on Genetic Algorithm

How fuzzy technology and neural networks and genetic algorithm combine with each other has become the focus of research. A fuzzy neural network controller was proposed based on defuzzification and optimization around the fuzzy neural network structure. Genetic algorithm of fuzzy neural network was brought forward based on optimal control theory. Optimal structure and parameters of fuzzy neural network controller were Offline searched by way of controller performance indicators of genetic algorithm. Fuzzy neural network controller through genetic algorithm was accessed in fuzzy neural network intelligent control system.