Fault is an undesired and unexpected event that changes the system behaviour resulting in performance degradation or even instability, so how to detect and diagnose fault become a great deal in engineering community. ...
详细信息
Fault is an undesired and unexpected event that changes the system behaviour resulting in performance degradation or even instability, so how to detect and diagnose fault become a great deal in engineering community. In this study, an adaptive fuzzy wavelet network-based fault detection and diagnosis (AFWN-FDD) scheme is proposed for non-linear systems subject to unstructured uncertainty. The proposed scheme is composed of a diagnostic estimator and an adaptive fuzzy wavelet network (AFWN). Diagnostic estimator is designed for residual generation and fault detection and AFWN based on multi-resolution analysis of wavelet transform and fuzzy concept is proposed to approximate the model of fault. learning algorithm of the proposed AFWN-FDD scheme is derived in the Lyapunov stability sense. The proposed scheme can simultaneously detect and estimate multiple incipient and abrupt faults in the presence of uncertainty. Stability analysis for the presented fault detection and diagnosis (FDD) scheme is provided. Furthermore, an extension of the proposed scheme for a class of non-linear systems with unmeasured states is presented. The efficiency and performance of the proposed scheme is evaluated through simulations that are performed for two well-known case studies. Comparison results highlight the superiority and capability of the proposed scheme.
The Kushilevitz-Mansour (KM) algorithm is an algorithm that finds all the "large" Fourier coefficients of a Boolean function. It is the main tool for learning decision trees and DNF expressions in the PAC mo...
详细信息
The Kushilevitz-Mansour (KM) algorithm is an algorithm that finds all the "large" Fourier coefficients of a Boolean function. It is the main tool for learning decision trees and DNF expressions in the PAC model with respect to the uniform distribution. The algorithm requires access to the membership query (MQ) oracle. The access is often unavailable in learning applications and thus the KM algorithm cannot be used. We significantly weaken this requirement by producing an analogue of the KM algorithm that uses extended statistical queries (SQ) (SQs in which the expectation is taken with respect to a distribution given by a learning algorithm). We restrict a set of distributions that a learning algorithm may use for its statistical queries to be a set of product distributions with each bit being 1 with probability rho, 1/2 or 1 - rho for a constant 1/2 > rho > 0 (we denote the resulting model by SQ-D-rho). Our analogue finds all the "large" Fourier coefficients of degree lower than clog n (we call it the Bounded Sieve (BS)). We use BS to learn decision trees and by adapting Freund's boosting technique we give an algorithm that learns DNF in SQ-D-rho. An important property of the model is that its algorithms can be simulated by MQs with persistent noise. With some modifications BS can also be simulated by MQs with product attribute noise (i.e., for a query x oracle changes every bit of x with some constant probability and calculates the value of the target function at the resulting point) and classification noise. This implies learnability of decision trees and weak learnability of DNF with this non-trivial noise. In the second part of this paper we develop a characterization for learnability with these extended statistical queries. We show that our characterization when applied to SQ-Dp is tight in terms of learning parity functions. We extend the result given by Blum et al. by proving that there is a class learnable in the PAC model with random classification noise a
This paper describes the applicability of artificial neural networks (ANNs) to predict performance of a horizontal ground-coupled heat pump (GCHP) system. Performance forecasting is the precondition for the optimal co...
详细信息
This paper describes the applicability of artificial neural networks (ANNs) to predict performance of a horizontal ground-coupled heat pump (GCHP) system. Performance forecasting is the precondition for the optimal control and energy saving operation of heat pump systems. ANNs have been used in varied applications and they have been shown to be particularly useful in system modelling and system identification. In order to train the ANN, limited experimental measurements were used as training data and test data. In this study, in input layer, there are air temperature entering condenser unit and air temperature leaving condenser unit, and ground temperatures (I and 2 in);coefficient of performance of system (COPS) is in output layer. The back propagation learning algorithm with three different variants, namely Levenberg-Marguardt (LM), Pola-Ribiere conjugate gradient (CGP), and scaled conjugate gradient (SCG), and tangent sigmoid transfer function were used in the network so that the best approach can find. The most suitable algorithm and neuron number in the hidden layer are found as LM with seven neurons. For this number level, after the training, it is found that Root-mean squared (RMS) value is 1%, and absolute fraction of variance (R-2) value is 99.999% and coefficient of variation in percent (COV) value is 28.62%. It is concluded that, ANNs can be used for prediction of COPS as an accurate method in the systems. (C) 2007 Elsevier Ltd. All rights reserved.
The results of an investigation of a fuzzy logic model for short term load forecasting are presented. The proposed methodology uses fuzzy rules to incorporate historical weather and load data. These fuzzy rules are ob...
详细信息
The results of an investigation of a fuzzy logic model for short term load forecasting are presented. The proposed methodology uses fuzzy rules to incorporate historical weather and load data. These fuzzy rules are obtained from the historical data using a learning-type algorithm. Test results from daily peak and total load forecasts for one year of data from a large scale power system indicate that the fuzzy rule bases can produce results similar in accuracy to more complicated statistical and back-propagation neural network methods. Copyright (C) 1996 Elsevier Science Ltd.
Recent psychophysics experiment has showed that the noise strength could affect the perceived image quality. This work gives an adaptive process for achieving the optimal perceived image quality in a simple image perc...
详细信息
Recent psychophysics experiment has showed that the noise strength could affect the perceived image quality. This work gives an adaptive process for achieving the optimal perceived image quality in a simple image perception array, which is a simple model of an image sensor. A reference image from memory is used for constructing a cost function and defining the optimal noise strength where the cost function gets its minimum point. The reference image is a binary image, which is used to define the background and the object. Finally, an adaptive algorithm is proposed for searching the optimal noise strength. Computer experimental results show that if the reference image is a thresholded version of the sub-threshold input image then the output of the sensor array gives an optimal output, in which the background and the object have the biggest contrast. If the reference image is different from a thresholded version of the sub-threshold input image then the output usually gives a sub-optimal contrast between the object and the background. (C) 1998 Elsevier Science B.V.
This paper presents a method for solving linear Fredholm integro-differential equations using a feedforward neural network based on Legendre polynomials. Firstly, the Legendre polynomials are used to approximate the u...
详细信息
This paper presents a method for solving linear Fredholm integro-differential equations using a feedforward neural network based on Legendre polynomials. Firstly, the Legendre polynomials are used to approximate the unknown function and the kernel function in the equations. Secondly, The roots of Legendre polynomials are obtained by using Newton iteration method to obtain Gaussian integration points, and the obtained Gaussian integration points are used as the input nodes of the neural network, and the corresponding weight is learned by the gradient descent method to obtain an approximate solution. Finally, through the numerical value Case analysis to verify the effectiveness of the method.
This letter presents a novel fuzzy neural network, which is composed of an antecedent network and a consequent network. The antecedent network matches the premises of the fuzzy rules and the consequent network impleme...
详细信息
This letter presents a novel fuzzy neural network, which is composed of an antecedent network and a consequent network. The antecedent network matches the premises of the fuzzy rules and the consequent network implements the consequences of the rules. In the network learning and training phase, a concise and effective algorithm based on the fuzzy hierarchy error approach (FHEA) is proposed to update the parameters of the network. This algorithm is simple to implement and it does not require as many Calculations as some other classic neural network learning algorithms, A model reference adaptive control structure incorporating the proposed fuzzy neural network is studied. Simulation results Of a cart-pole balancing system demonstrate the effectiveness of proposed method.
This paper develops a non-singleton type-2 fuzzy neural network (NT2FNN) with type-2 3-dimensional membership functions (MFs) and adaptive secondary membership. A new approach based on the squareroot cubature quadratu...
详细信息
This paper develops a non-singleton type-2 fuzzy neural network (NT2FNN) with type-2 3-dimensional membership functions (MFs) and adaptive secondary membership. A new approach based on the squareroot cubature quadrature Kalman filter (SR-CQKF) is proposed for the training the level of the secondary membership and the centers of membership functions. The consequent parameters are learned by using rule-ordered extended Kalman filter (EKF). To show the applicability and effectiveness of proposed NT2FNN in high dimensional problems, four real-world datasets with 4, 7, 13 and 32 input variables are considered. Additionally, the performance of NT2FNN with the proposed learning algorithm is compared with other well-known neural networks and learning algorithms. The simulations demonstrate that the developed method results in high performance in contrast to the other methods. (C) 2019 Elsevier B.V. All rights reserved.
In this study a new machine learning technique is presented to solve singular multi-pantograph differential equations (SMDEs). A new optimized type-3 fuzzy logic system (T3-FLS) by unscented Kalman filter (UKF) is pro...
详细信息
In this study a new machine learning technique is presented to solve singular multi-pantograph differential equations (SMDEs). A new optimized type-3 fuzzy logic system (T3-FLS) by unscented Kalman filter (UKF) is proposed for solution estimation. The convergence and stability of presented algorithm are ensured by the suggested Lyapunov analysis. By two SMDEs the effectiveness and applicability of the suggested method is demonstrated. The statistical analysis show that the suggested method results in accurate and robust performance and the estimated solution is well converged to the exact solution. The proposed algorithm is simple and can be applied on various SMDEs with variable coefficients.
In order to provide more comprehensive network services, a concept of integrated heterogeneous network system was introduced. Until now, lots of researchers have focused on how to efficiently integrate different types...
详细信息
In order to provide more comprehensive network services, a concept of integrated heterogeneous network system was introduced. Until now, lots of researchers have focused on how to efficiently integrate different types of wireless and mobile networks. To exploit the heterogeneous network system operation, an important issue is how to properly manage the network bandwidth. In this study, a new bandwidth management scheme has been proposed by employing the principal-agent game model. Among heterogeneous networks, we have analyzed the asymmetric information situation and developed an effective bandwidth allocation algorithm. Under diverse network condition changes, our principal-agent game approach is essential to provide a suitable tradeoff between conflicting requirements. Simulation results demonstrate that the proposed scheme can obtain a better network performance and bandwidth efficiency than other existing schemes.
暂无评论