A hybrid fault diagnosis approach using neural networks

A hybrid fault diagnosis method is proposed in this paper which is based on the parity equations and neural networks. Analytical redundancy is employed by using parity equations. Neural networks then are used to maximise the signal- to- noise ratio of the residual and to isolate different faults. Effectiveness of the method is demonstrated by applying it to fault detection and isolation for a hydraulic test rig. Real data simulation shows that the sensitivity of the residual to the faults is maximised, whilst that to the unknown input is minimised. The simulated faults are successfully isolated by a bank of neural nets.     

Feedback linearizing control of switched reluctance motors

Motivated by technological advances in power electronics and signal processing, and by the interest in using direct drives for robot manipulators, we investigate the control problem of high-performance drives for switched reluctance motors (SRM's). SRM's are quite simple, low cost, and reliable motors as compared to the widely used dc motors. However, the SRM presents a coupled nonlinear multivariable control structure which calls for complex nonlinear control design in order to achieve high dynamic performances. We first develop a detailed nonlinear model which matches experimental data and establish an electronic commutation strategy. Then, on the basis of recent nonlinear control techniques, we design a state feedback control algorithm which compensates for all the nonlinearities and decouples the effect of stator phase currents in the torque production. The position dependent logic of the electronic commutator assigns control authority to one phase, which controls the motion, while the remaining phase currents are forced to decay to zero. Simulations for a direct drive, single link manipulator with the SRM are reported, which show the control performance of the algorithm we propose in nominal conditions and test its robustness versus the most critical parameter uncertainties of payload mass and stator resistance.     

Feedback linearizing control of vehicle longitudinal acceleration

The application of recently developed feedback linearization techniques to controlling the powertrain of an automobile is discussed. The objective of powertrain control is to regulate the output torque of the engine and transmission system to achieve a desired longitudinal acceleration of the vehicle. With higher model order and increasing system complexity, the design process for feedback linearization becomes harder and more tedious. However, the design process can be simplified for a class of nonlinear systems including the automobile powertrain. Simplification of the design process is accomplished by incorporating the single perturbation technique into the input-output linearization technique. Furthermore, the controller obtained by the simplified linearization technique is computationally simpler than the one obtained by the conventional linearization technique. Nonetheless, it provides good dynamic performance for the powertrain system     

A hybrid approach to vehicle routing using neural networks and genetic algorithms

A competitive neural network model and a genetic algorithm are used to improve the initialization and construction phase of a parallel insertion heuristic for the vehicle routing problem with time windows. The neural network identifies seed customers that are distributed over the entire geographic area during the initialization phase, while the genetic algorithm finds good parameter settings in the route construction phase that follows. Computational results on a standard set of problems are also reported.     

Feedback linearisation control of an induction machine augmented by single-hidden layer neural networks

We consider adaptive output feedback control methodology of highly uncertain nonlinear systems with both parametric uncertainties and unmodelled dynamics. The approach is also applicable to systems of unknown, but bounded dimension. However, the relative degree of the regulated output is assumed to be known. This new control strategy is proposed to address the tracking problem of an induction motor based on a modified field-oriented control method. The obtained controller is then augmented by an online neural network that serves as an approximator for the neglected dynamics and modelling errors. The network weight adaptation rule is derived from the Lyapunov stability analysis, that guarantees boundedness of all the error signals of the closed-loop system. Computer simulations of an output feedback controlled induction machine, augmented via single-hidden-layer neural networks, demonstrate the practical potential of the proposed control algorithm.     

Training recurrent neural networks using a hybrid algorithm

This paper proposes a new hybrid approach for recurrent neural networks (RNN). The basic idea of this approach is to train an input layer by unsupervised learning and an output layer by supervised learning. In this method, the Kohonen algorithm is used for unsupervised learning, and dynamic gradient descent method is used for supervised learning. The performances of the proposed algorithm are compared with backpropagation through time (BPTT) on three benchmark problems. Simulation results show that the performances of the new proposed algorithm exceed the standard backpropagation through time in the reduction of the total number of iterations and in the learning time required in the training process.     

Enhanced avatar control using neural networks

This paper presents realistic avatar movements using a limited number of sensors. An inverse kinematics algorithm, SHAKF, is used to configure an articulated skeletal model, and a neural network is employed to predict the movement of joints not bearing sensors. The results show that the neural network is able to give a very close approximation to the actual rotation of the joints. This allows a substantial reduction in the number of sensors to configure an articulated human skeletal model.     

采用资格迹的神经网络学习控制算法

强化学习是解决自适应问题的重要方法,被广泛地应用于连续状态下的学习控制,然而存在效率不高和收敛速度较慢的问题.在运用反向传播(back propagation,BP)神经网络基础上,结合资格迹方法提出一种算法,实现了强化学习过程的多步更新.解决了输出层的局部梯度向隐层节点的反向传播问题,从而实现了神经网络隐层权值的快速更新,并提供一个算法描述.提出了一种改进的残差法,在神经网络的训练过程中将各层权值进行线性优化加权,既获得了梯度下降法的学习速度又获得了残差梯度法的收敛性能,将其应用于神经网络隐层的权值更新,改善了值函数的收敛性能.通过一个倒立摆平衡系统仿真实验,对算法进行了验证和分析.结果显示,经过较短时间的学习,本方法能成功地控制倒立摆,显著提高了学习效率.     

Continuous process control using neural networks

We present some adaptive control strategies based on neural networks that can be used for designing controllers for continuous process control problems. Specifically, a learning algorithm has been formulated based on reinforcement learning, a weakly supervised learning technique, to solve set-point control and control scheduling for continuous processes where the process cannot be modeled easily. It is shown how reinforcement learning can be used to learn the control strategy adaptively based on exploration of the control space without making assumptions about the process model. A new learning scheme, handicapped learning, was developed to learn a control schedule that specifies a schedule of set points. Applications studied include the control of a nonisothermal continuously stirred tank reactor at its unstable state and the learning of the daily time-temperature schedule for an environment controller. Experimental results demonstrate good learning performance, indicating that the learning algorithm can be used for solving transient startup and boundary value control problems.     

A hybrid recommender system using artificial neural networks

In the context of recommendation systems, metadata information from reviews written for businesses has rarely been considered in traditional systems developed using content-based and collaborative filtering approaches. Collaborative filtering and content-based filtering are popular memory-based methods for recommending new products to the users but suffer from some limitations and fail to provide effective recommendations in many situations. In this paper, we present a deep learning neural network framework that utilizes reviews in addition to content-based features to generate model based predictions for the business-user combinations. We show that a set of content and collaborative features allows for the development of a neural network model with the goal of minimizing logloss and rating misclassification error using stochastic gradient descent optimization algorithm. We empirically show that the hybrid approach is a very promising solution when compared to standalone memory-based collaborative filtering method.     

Feedback linearizing and sliding mode control of a variable reluctance motor

Direct-drive connection of electric motors represents a suitable solution to friction and backlash problems introduced by mechanical reduction gears. Variable reluctance (VR) are a special type of switched reluctance motors whose construction is well suited for direct-drive connection. Although these motors are traditionally conceived as stepper motors, continuous motion can be obtained by implementing suitable closed-loop control in the drive. The authors' main aim is to design a high-performance robust controller for a VR motor intended for velocity trajectory tracking in robotics applications where continuous motion is required. A cascade controller structure (velocity-torque) is considered. In the design of the torque controller, both feedback linearizing and sliding mode techniques are considered. Feedback linearization performs slightly better in the ideal case, but under more realistic operating conditions the sliding mode controller demonstrates comparable or even better performance. A very simple but extremely robust velocity controller is designed using a dynamic sliding mode approach, ensuring robustness to large variations of load torque and inertia, typical of direct-drive robotics applications. A simulation experiment of the overall controller with the motor connected to a single link robotic arm shows very good tracking properties as well as insensitivity to large variations of load torque and inertia.     

Customer churn prediction by hybrid neural networks

As churn management is a major task for companies to retain valuable customers, the ability to predict customer churn is necessary. In literature, neural networks have shown their applicability to churn prediction. On the other hand, hybrid data mining techniques by combining two or more techniques have been proved to provide better performances than many single techniques over a number of different domain problems. This paper considers two hybrid models by combining two different neural network techniques for churn prediction, which are back-propagation artificial neural networks (ANN) and self-organizing maps (SOM). The hybrid models are ANN combined with ANN (ANN + ANN) and SOM combined with ANN (SOM + ANN). In particular, the first technique of the two hybrid models performs the data reduction task by filtering out unrepresentative training data. Then, the outputs as representative data are used to create the prediction model based on the second technique. To evaluate the performance of these models, three different kinds of testing sets are considered. They are the general testing set and two fuzzy testing sets based on the filtered out data by the first technique of the two hybrid models, i.e. ANN and SOM, respectively. The experimental results show that the two hybrid models outperform the single neural network baseline model in terms of prediction accuracy and Types I and II errors over the three kinds of testing sets. In addition, the ANN + ANN hybrid model significantly performs better than the SOM + ANN hybrid model and the ANN baseline model.     

Sound quality analysis of cars using hybrid neural networks

In this paper, a procedure of testing and evaluation on the sound quality of cars are proposed and sound quality is analysed through the cars’ road running test on the providing ground, which was carried out with varying running speed. In addition to this experimental analysis, a neural network predictor is also designed to model the system for possible experimental applications. The proposed neural network is a recurrent type network, which consists of two types of neuron function in the hidden layer. As basic factors for sound quality, only objective factors are considered such as loudness, sharpness, speech intelligibility, and sound pressure level. The correlation between sound pressure level and another factor are discussed from a point of view of running speed dependency. Results of both computer simulations and experiments show that the neural predictor algorithm gives good results at accommodating different cases and provides superior prediction on two cars’ sound analysis.     

Adaptive observer backstepping control using neural networks

This paper extends the application of neurocontrol approaches to a new class of nonlinear systems diffeomorphic to output feedback nonlinear systems with unmeasured states. A neural-based adaptive observer is introduced for state estimation as well as system identification using only output measurements during online operation. System identification is achieved via the online approximation of a priori unknown functions. The controller is designed using the backstepping control design procedure. Leakage terms in the adaptive laws and nonlinear damping terms in the backstepping controller are introduced to prevent instability from arising due to the inherent approximation error. A primary benefit of the online function approximation is the reduction of approximation errors, which allows reduction of both the observer and controller gains. A semi-global stability analysis for the proposed approach is provided and the feasibility is investigated by an illustrative simulation example.     

Adaptive manipulator trajectory control using neural networks

A unified study of adaptive control and neural network based control schemes for the trajectory tracking problem of robot manipulators is presented. Efficacy of parametrized adaptive algorithms in compensating the structured uncertainties in robot dynamics is verified through extensive simulation. The ability of neural networks to provide a robust adaptive framework in the presence of both structured and unstructured uncertainties is investigated. A case study is carried out in support of a parametrized adaptive scheme using neural networks. Simulation results clearly indicate that the neural network based adaptive controller achieves better tracking in the presence of parametric uncertainties as well as unmodelled effects compared to the simple direct adaptive scheme.     

Exchange rate prediction using hybrid neural networks and trading indicators

This paper describes a hybrid model formed by a mixture of various regressive neural network models, such as temporal self-organising maps and support vector regressions, for modelling and prediction of foreign exchange rate time series. A selected set of influential trading indicators, including the moving average convergence/divergence and relative strength index, are also utilised in the proposed method. A genetic algorithm is applied to fuse all the information from the mixture regression models and the economical indicators. Experimental results and comparisons show that the proposed method outperforms the global modelling techniques such as generalised autoregressive conditional heteroscedasticity in terms of profit returns. A virtual trading system is built to examine the performance of the methods under study.     

An intelligent hybrid approach for industrial quality control combining neural networks, fuzzy logic and fractal theory

The application of type-2 fuzzy logic to the problem of automated quality control in sound speaker manufacturing is presented in this paper. Traditional quality control has been done by manually checking the quality of sound after production. This manual checking of the speakers is time consuming and occasionally was the cause of error in quality evaluation. For this reason, by applying type-2 fuzzy logic, an intelligent system for automated quality control in sound speaker manufacturing is developed. The intelligent system has a type-2 fuzzy rule base containing the knowledge of human experts in quality control. The parameters of the fuzzy system are tuned by applying neural networks using, as training data, a real time series of measured sounds produced by good sound speakers. The fractal dimension is used as a measure of the complexity of the sound signal.     

Adaptive control using neural networks and approximate models

The NARMA model is an exact representation of the input-output behavior of finite-dimensional nonlinear discrete-time dynamical systems in a neighborhood of the equilibrium state. However, it is not convenient for purposes of adaptive control using neural networks due to its nonlinear dependence on the control input. Hence, quite often, approximate methods are used for realizing the neural controllers to overcome computational complexity. In this paper, we introduce two classes of models which are approximations to the NARMA model, and which are linear in the control input. The latter fact substantially simplifies both the theoretical analysis as well as the practical implementation of the controller. Extensive simulation studies have shown that the neural controllers designed using the proposed approximate models perform very well, and in many cases even better than an approximate controller designed using the exact NARMA model. In view of their mathematical tractability as well as their success in simulation studies, a case is made in this paper that such approximate input-output models warrant a detailed study in their own right.     

Optimal control of terminal processes using neural networks

Feedforward neural networks are capable of approximating continuous multivariate functions and, as such, can implement nonlinear state-feedback controllers. Training methods such as backpropagation-through-time (BPTT), however, do not deal with terminal control problems in which the specified cost function includes the elapsed trajectory-time. In this paper, an extension to BPTT is proposed which addresses this limitation. The controller design is reformulated as a constrained optimization problem defined over the entire field of extremals and in which the set of trajectory times is incorporated into the cost function. Necessary first-order stationary conditions are derived which correspond to standard BPTT with the addition of certain transversality conditions. The new gradient algorithm based on these conditions, called time-optimal backpropagation through time, is tested on two benchmark minimum-time control problems.     

Coverage control in unknown environments using neural networks

This paper proposes a distributed adaptive control algorithm for coverage control in unknown environments with networked mobile sensors. An online neural network weight tuning algorithm is used in order for the robots to estimate the sensory function of the environment, and the control law is derived according to the feedforward neural network estimation of the distribution density function of the environment. It is distributed in that it only takes advantage of local information of each robot. A Lyapunov function is introduced in order to show that the proposed control law causes the network to converge to a near-optimal sensing configuration. Due to neural network nonlinear approximation property, a major advantage of the proposed method is that in contrary to previous well known approaches for coverage, it is not restricted to a linear regression form. Finally the controller is demonstrated in numerical simulations. Simulation results have been shown that the proposed controller outperforms the previous adaptive approaches in the sense of performance and convergence rate.