带时滞的ADALINE神经网络在时变线性系统自适应建模中的应用研究

本文提出了带时滞的ADALINE神经网络对时变复杂线性系统的自适应建模方法,通过对性能函数是权系数二次型的神经网络的算法改进,以梯度变化的共轭向量为搜索方向,以解析式确定最优步长,理论分析和仿真实验证明了此算法的快速性和高精度逼近能力,也证明了基于这种快速算法的带时滞的ADALINE网络实现复杂时变线性系统高精度动态自适应建模的有效性。     

Two degree of freedom PID based inferential control of continuous bioreactor for ethanol production

This article presents the development of inferential control scheme based on Adaptive linear neural network (ADALINE) soft sensor for the control of fermentation process. The ethanol concentration of bioreactor is estimated from temperature profile of the process using soft sensor. The prediction accuracy of ADALINE is enhanced by retraining it with immediate past measurements. The ADALINE and retrained ADALINE are used along with PID and 2-DOF-PID leading to APID, A2PID, RAPID and RA2PID inferential controllers. Further the parameters of 2-DOF-PID are optimized using Non-dominated sorted genetic algorithm-II and used with retrained ADALINE soft sensor which leads to RAN2PID inferential controller. Simulation results demonstrate that performance of proposed RAN2PID controller is better in comparison to other designed controllers in terms of qualitative and quantitative performance indices.     

自适应智能化谐波电流检测方法

有源电力滤波器检测谐波电流的实时性、精确性对电能质量的提高至关重要,文中提出了基于非线性最小二乘法与自适应人工神经网络结合的检测方法。非线性最小二乘法用于检测基波电压的频率,自适应人工神经网络用于检测基波电压的初始相位和基波电流的幅值,由基波电压的频率和初始相位获得单位幅值的基波电流。文中方法在0.02 s内可准确检测出基波和谐波电流,检测精度较传统方法有显著提高,通过仿真验证了该方法的有效性和优越性。     

Local stereovision matching through the ADALINE neural network

This paper presents an approach to the local stereovision matching problem using edge segments as features with four attributes. Based on these attributes we compute a matching probability between pairs of features of the stereo images. A correspondence is said to be true when this probability is maximum. The probability value is a weighted sum of the attributes. We use two combined ADALINE neural networks to compute the weight for each attribute. A comparative analysis among other recent matching methods is illustrated.     

Learning method of the ADALINE using the fuzzy logic system

We propose a learning method for the ADALINE. The proposed method exploits fuzzy logic system for automatic tuning of the weights of the ADALINE. The inputs of the fuzzy logic system are error and change of error, and the output is the weight variation. We used same membership functions and different scaling factor for each weights. In order to verify the effectiveness of the proposed method, we performed the simulation and experimentation for the cases of the noise cancellation and the inverted pendulum control. The results show that the proposed method does not need the learning rate and the derivative, and improves the performance compared to the Widrow–Hoff delta rule for ADALINE.     

A novel approach to adaptive single phase auto-reclosing scheme for EHV transmission lines

In modern interconnected power systems, nearly 80% of faults in high voltage transmission lines are intrinsically transient. The necessity of rapid fault clearing has resulted in fast development of protection equipments. Moreover, need for reliable supply of loads has led to improvements in single phase auto-reclosing equipments. In this paper, a novel and efficient method is proposed that leading to improved performance and efficiency of single phase auto-reclosing. In the case of occurrence of permanent faults, the proposed yardstick is accurate and authentic to diagnose fault type (transient or permanent). To validate accuracy and authenticity of the proposed method, a 400 kV transmission system is simulated using EMTP software and results are presented.     

基于ADALINE的有源电力滤波器预测电流控制策略

为有效补偿有源电力滤波器(APF)的控制延时,提出一种基于自适应线性神经元(ADALINE)的APF预测电流控制策略。该策略通过ADALINE预测下一控制周期的参考电流,根据两相静止坐标系下APF系统模型确定参考电压,利用空间矢量脉宽调制(SVPWM)方法,使APF输出电流快速、准确地跟踪参考电流。仿真与APF样机上的实际运行结果表明,和采用传统无预测算法的SVPWM方法相比,采用本文提出的预测电流控制策略,APF的稳态补偿精度和动态跟踪性能均明显提高。     

基于MRAS的异步电机无速度传感器应用研究

针对异步电机基于转子磁链模型参考自适应(MRAS)速度观测器中的积分直流偏置和PI自适应率参数不能在线修正的问题,将转子磁链的电压模型进行改进,并提出一种自适应线性神经元(ADALINE)速度观测器。通过一阶惯性环节代替了电压模型中的纯积分环节,消除了积分直流偏置和误差积累,利用ADALINE算法对MRAS观测器中的自适应率进行改进,实现了在中高速情况下对电机转速的准确观测。仿真结果表明,改进后的MRAS速度观测器对速度的辨识度较高,动态性能良好,验证了该控制策略的正确性与可行性。     

A quadratic polynomial signal model and fuzzy adaptive filter for frequency and parameter estimation of nonstationary power signals

Accurate estimation of amplitude, phase and frequency of a sinusoid in the presence of harmonics/inter harmonics and noise plays an important role in a wide variety of power system applications, like protection, control and state monitoring. With this objective, the paper presents a novel hybrid approach for the accurate estimation of dynamic power system frequency, phasor and in addition to suppressing the effect of harmonics/interharmonics and noise in the voltage and current signals. The algorithm assumes that the current during a fault occurring on a power system consists of a decaying dc component, and time variant fundamental and harmonic phasors. For accurate estimation of fundamental frequency, phasor, decaying dc and ac components in the fault current or voltage signal, the algorithm uses a quadratic polynomial signal model and a fuzzy adaptive ADALINE filter with a modified Gauss–Newton algorithm. Extensive study has been carried out to demonstrate the performance analysis and fast convergence characteristic of the proposed algorithm. The proposed method can also be implemented for accurate estimation of dynamic variations in the amplitude and phase angles of the harmonics and inter harmonics mixed with high noise conditions.     

基于ADALINE神经网络的电能质量监测终端设计

谐波频谱检测是电能质量监测仪器的核心功能,其检测频谱是进行各种电能质量特征值运算的前提。传统以STFT为检测算法的电能质量终端由于时间窗固定,不具有暂态情况下的谐波分析能力。而诸如小波变换、S变换等算法则由于运算量巨大不利于谐波实时在线监测。针对这种情况,设计了一种以DSP为处理器,基于自适应线性神经网络(ADALINE)的电能质量监测终端。详细介绍信号接口电路、调理电路、PLL倍频电路、AD转换电路的硬件设计,给出了自适应线性神经算法推导和DSP数据处理框图。实验表明,所构建系统在运算量不大的情况下增强了暂态谐波测量能力,同时利用ADALINE误差信号可对电压暂降进行精确时间点定位。