Adaptive Linear Neuron Network Research Articles

Performance Analysis of BLDC Motor Drive using Enhanced Neural Based Speed Controller for Electric Vehicle Applications

This paper deals with Brushless DC (BLDC) motor drive mathematical model with hysteresis current controlled based Voltage Source Inverter (VSI) which operates using Neural Network (NN). The developed simulation model is applied with the artificial neuro speed controller which is based on Least Mean Square (LMS) and Adaptive Linear Neuron (ADALINE) to improve the performance of BLDC motor drive. Under dynamic operating conditions, the NN speed controller is trained by the data obtained from closed loop speed controller of BLDC motor drive system. The developed conventional and proposed simulation models are simulated using MATLAB/Simulation software. The proposed neural based speed controller is important for tracking of motor speed as well as torque with their reference values with minimum transient time. In this paper ADALINE network model is developed for update of weights in NN. Here, a comparative study has been done for PI and NN controller for BLDC drive. The simulation results show that the NN based speed controller is more effective controller compared to classical PI based speed controller during most of the dynamic operating conditions.

DEA-ADALINE: an approach to improve the relative efficiency of 3PLs providers

PurposeBenchmarking is an approach to understand and evaluate the current position of different entities like third-party logistics (3PLs) service providers with respect to the best practices. The purpose of this paper is to develop an integrated approach to benchmark the 3PLs and minimize the transportation cost, operational cost, total carbon emissions and improve vehicle capacity utilization rate and demand management in an efficient manner.Design/methodology/approachConventional data envelopment analysis (DEA) is extended to develop an integrated approach involving DEA and adaptive linear neuron network (ADALINE). The proposed methodology is dichotomized to the benchmarking of DMUs using Charnes, Cooper, Rhodes ratio and, in the process, helps to suggest ways to reduce inefficiencies.FindingsThe approach also supports the decision makers to understand ways to increase the efficiency of decision-making units (DMUs) that are relatively inefficient for a set of 3PLs considered in this study. This is one of the potential researches as it studies how to improve the efficiency of inefficient DMUs.Research limitations/implicationsThe proposed approach would help the decision makers to better understand the complexities associated with the low-performing 3PLs. The application of the proposed DEA-ADALINE methodology is illustrated using a range of data set for 3PLs.Originality/valueThe paper examines and explains the value added to the conventional DEA model for obtaining the relative efficiencies with reference to the most efficient DMUs without any loss to the original characteristics of DEA and ADALINE.

A Novel Approach for an MPPT Controller Based on the ADALINE Network Trained with the RTRL Algorithm

The Real-Time Recurrent Learning Gradient (RTRL) algorithm is characterized by being an online learning method for training dynamic recurrent neural networks, which makes it ideal for working with non-linear control systems. For this reason, this paper presents the design of a novel Maximum Power Point Tracking (MPPT) controller with an artificial neural network type Adaptive Linear Neuron (ADALINE), with Finite Impulse Response (FIR) architecture, trained with the RTRL algorithm. With this same network architecture, the Least Mean Square (LMS) algorithm was developed to evaluate the results obtained with the RTRL controller and then make comparisons with the Perturb and Observe (P&O) algorithm. This control method receives as input signals the current and voltage of a photovoltaic module under sudden changes in operating conditions. Additionally, the efficiency of the controllers was appraised with a fuzzy controller and a Nonlinear Autoregressive Network with Exogenous Inputs (NARX) controller, which were developed in previous investigations. It was concluded that the RTRL controller with adaptive training has better results, a faster response, and fewer bifurcations due to sudden changes in the input signals, being the ideal control method for systems that require a real-time response.

Resolution-Enhanced Harmonic and Interharmonic Measurement for Power Quality Analysis in Cyber-Physical Energy System.

Power quality analysis issues, especially the measurement of harmonic and interharmonic in cyber-physical energy systems, are addressed in this paper. As new situations are introduced to the power system, the impact of electric vehicles, distributed generation and renewable energy has introduced extra demands to distributed sensors, waveform-level information and power quality data analytics. Harmonics and interharmonics, as the most significant disturbances, require carefully designed detection methods for an accurate measurement of electric loads whose information is crucial to subsequent analyzing and control. This paper gives a detailed description of the power quality analysis framework in networked environment and presents a fast and resolution-enhanced method for harmonic and interharmonic measurement. The proposed method first extracts harmonic and interharmonic components efficiently using the single-channel version of Robust Independent Component Analysis (RobustICA), then estimates the high-resolution frequency from three discrete Fourier transform (DFT) samples with little additional computation, and finally computes the amplitudes and phases with the adaptive linear neuron network. The experiments show that the proposed method is time-efficient and leads to a better accuracy of the simulated and experimental signals in the presence of noise and fundamental frequency deviation, thus providing a deeper insight into the (inter)harmonic sources or even the whole system.

An adaptive filter–based temperature compensation technique for structural health monitoring

Temperature variations have significant effects on guide wave propagation and therefore increase the detection uncertainty of the guided wave–based structural health monitoring system. A novel temperature compensation technique combining an adaptive filter and optimal baseline selection is developed to enhance the robustness and effectiveness of guided wave–based damage detection. The adaptive filter is the finite length unit impulse response digital filter based on adaptive linear neuron network. This article focuses on three main issues for practically implementing the proposed method: (a) establishment of temperature compensation standard, (b) parameter design of compensation filter, and (c) determination of temperature gradient to reduce the number of selected baselines. Experiments are conducted on two stiffened composite plates to verify the proposed method for effective and robust temperature compensation under a large temperature range from −40°C to 80°C. Results show that temperature interval for baselines of low-frequency signals, such as 50 kHz, can be up to 20°C to provide good temperature compensation with the proposed method, while temperature interval for baselines of high-frequency signals, such as 450 kHz, can be up to 12°C.

IMPROVING THE ACTIVE POWER FILTER PERFORMANCE WITH AN ADAPTIVE LINEAR NEURAL NETWORK BASED ON THE REFERENCE GENERATION

The use of artificial neural networks (ANNs) in power system problems is increasing attention due to their ability to learn and handle the nonlinearities and uncertainties of the systems. Since the last decade, different ANNs identification and filtering techniques have been applied in power systems and adaptive linear neuron networks (ADALINE) are actually widely used. More recently artificial neural networks have been introduced as a complement or an alternative to traditional control algorithms. The use of neural networks in control applications including process control, robotics, industrial manufacturing and aerospace applications, among others has recently experienced rapid growth. This paper proposes a new control method for shunt active power filter, which is based on ANN. An adaptive linear neuron (ADALINE) will be presented to estimate the reference compensation currents. The resultant compensation currents eliminate harmonics and reactive power compensation with a quick dynamic response. Theortical and expermental results in a good agreement

ADALINE approach for induction motor mechanical parameters identification

Two new methods to identify the mechanical parameters in induction motor based field oriented drives are presented in this paper. The identified parameters are: the moment of inertia and the viscous damping coefficient. The proposed methods are based on the adaptive linear neuron (ADALINE) networks. The two parameters are derived and optimized during the online training process. During the identification phase, the motor torque is controlled by the well-known field oriented control strategy. This torque is subjected to variations in order to obtain mechanical speed transients. The two proposed methods are simple to implement compared to the previous techniques. They require only the stator current and mechanical speed measurements. Finally, the effectiveness of the two methods and the accuracy of the derived parameters are proven experimentally by two direct starting tests. The originality of this work is the building of a model representation that it is suitable for implementation with ADALINE networks. This leads to a simple implementation and ease of mechanical parameters identification.

A Novel Method for Identifying Parameters of Induction Motors at Standstill Using ADALINE

In this paper, we propose a new method for an online electric parameters identification of an induction motor (IM). This method is carried out in the IM standstill configuration and uses adaptive linear neuron (ADALINE) networks. In order to simplify the identification process, the IM model is approximated by two first-order subsystems: one is valid at low frequencies (LFs), called the slow system and the other is valid at high frequencies (HFs), called the fast system. By means of two ADALINE networks, the parameters of the slow system and the fast system are identified in LFs and HFs, respectively, and thus, the required IM parameters are derived. Finally, experimental results are presented in order to validate the proposed method and to check the accuracy of the obtained parameters. The originality of this paper is the building of a model representation that is suitable for implementation with ADALINE networks. This leads to a simple implementation and ease of parameters identification.

A self-synchronized ADALINE network for on-line tracking of power system harmonics

This paper presents a modified adaptive linear neuron (ADALINE) structure, called self-synchronized adaptive linear neuron (S-ADALINE) network for fast and accurate estimation of power system harmonics. The proposed network relies on the Levenberg gradient descent (LGD) method based parameter updating rule and is capable of dealing with both nominal and off-nominal frequency conditions, rather than the existing modified Widrow–Hoff delta rule based ADALINE network which provides good accuracy only at nominal frequency. Moreover, the S-ADALINE provides faster response and better noise immunity than the conventional approach. The only flaw of the proposed network is its high computational load. Based on simulation studies, performances of the proposed harmonic estimator at different operating conditions have been presented and its accuracy and response time have been compared with the conventional ADALINE structure.

Using complementary methods for improved flow forecasting

Forecasting river flow is important to water resources management and planning. Four artificial neural network (ANN) models were developed and compared in order to forecast river flow in the River Trent, UK. The first two are feed-forward networks and were trained with back-propagation and cascade-correlation algorithms; the third is an adaptive linear neuron network; and the fourth an Elman network. Using observations of river flow for 1996–1998, the ANN models were satisfactorily trained and verified. A method that allows the explicit description of the variations in the lag of the system was also used and, when applied together with a real-time updating method, resulted in improved model performance. The research shows that using a parallel data-driven error prediction model to complement the ANNs produces much better flow predictions in comparison to using the ANNs alone.

Forecasting Influent-Effluent Wastewater Treatment Plant Using Time Series Analysis and Artificial Neural Network Techniques

We consider the problem of predicting the future behavior of wastewater treatment plant quality indicators by creating prediction models using historical plant data. One of the main aims of this work is to be able to predict plant operational situations in advance so that corrective actions can be taken in time. Sets of historical plant data, such as BOD, COD and TSS were collected for a local wastewater treatment plant in Doha, the capital of the State of Qatar. These variables characterize the performance of any wastewater treatment plant and can be considered as quality indicators of the plant performance. Data were collected over a period of 4 years for the influent and effluent streams of the station. The plant influent and effluent predictions were performed using different techniques. These include time-series analysis, where the ARIMA (Autoregressive Integrated Moving Average) model was implemented in this case, and two Artificial Neural Networks (ANN) algorithms, namely Adaptive Linear Neuron networks (ADALINE) and Multi-layer Feedforward (ML-FF) neural networks. The predictions from the three techniques were presented and compared. The ML-FF model predictions proved to be more reliable than that of the equivalent ARIMA predictions followed by the ADALINE predictions, particularly for the finial effluent stream variables.

TIME DELAY COMPENSATION USING ADAPTIVE LINEAR NEURAL NETWORKS FOR NETWORKED CONTROL SYSTEMS

In real time systems, particularly control systems, delays or dropped packets may cause performance degradation and system destabilization. In order to consider the uncertainty of communication delays and packet losses, intelligent computational approaches such as fuzzy logic, neural networks, and genetic algorithm can be used. In this paper, The effect of time delay is compensated via building undelayed plant model based on delayed model data using the Adaptive Linear Neuron networks (ADALINE). In ADALINE the linear networks are adjusted at each time step based on new input and target vectors which can find weights and biases that minimize the network's sum-squared error for recent input and target vectors. The proposed works are applied on distributed control of a DC servo system. The network is built using the true time MATLAB toolbox. Several simulation examples are applied using CAN network to clarify the efficiency of the proposed methods.