自适应神经网络模糊小波语音消噪算法

针对有色噪声,采用自适应神经网络模糊系统模糊(Auto Neural Fuzzy Inference System,ANFIS)逼近有色噪声,利用自适应神经模糊推理系统ANFIS对噪声的非线性动态特性进行建模,提出了语音自适应神经网络模糊小波消噪算法,建立并训练了消噪系统。对被有色噪声污染的测量信号经模糊消噪后,根据信号和噪声的小波系数在不同分解尺度上的传递性,进行中值滤波和小波重构,得到了干净的语音。对算法进行了仿真实验,结果表明,消噪效果明显。     

基于自适应神经模糊推理的形变数据仿真计算

应用自适应神经模糊推理系统的原理,建立构筑物变形数据的预测计算模型,并用此模型完成大坝变形数据预报。它克服了以往人工神经网络变形数据计算中当解空间稍大时,便难以收敛到所需精度的缺陷,在较大的解空间内模型收敛速度快、变形预报精度高,是一种良好的变形数据预报方法。     

Prediction of ground motion parameters using randomized ANFIS (RANFIS)

In this paper, a novel neuro-fuzzy learning machine called randomized adaptive neuro-fuzzy inference system (RANFIS) is proposed for predicting the parameters of ground motion associated with seismic signals. This advanced learning machine integrates the explicit knowledge of the fuzzy systems with the learning capabilities of neural networks, as in the case of conventional adaptive neuro-fuzzy inference system (ANFIS). In RANFIS, to accelerate the learning speed without compromising the generalization capability, the fuzzy layer parameters are not tuned. The three time domain ground motion parameters which are predicted by the model are peak ground acceleration (PGA), peak ground velocity (PGV) and peak ground displacement (PGD). The model is developed using the database released by PEER (Pacific Earthquake Engineering Research Center). Each ground motion parameter is related to mainly to four seismic parameters, namely earthquake magnitude, faulting mechanism, source to site distance and average soil shear wave velocity. The experimental results validate the improved performance of the machine, with lesser computation time compared to prior studies.     

带并行机调度问题中基于ANFIS的自适应算法

针对用规则的线性组合方法解决实际的复杂调度中,如何确定哪些规则需要组合和确定权系数一直缺乏有效的方法,提出了根据多个不同的生产调度目标采用ANFIS将不同调度目标的规则进行自适应的综合,进行学习后产生模糊规则的算法,并将其用于带并行机Jobshop调度问题中。数值计算结果表明,该算法能够较好地对一些规则进行综合,并且在此类调度问题中有比这些规则更好的效果。今后的研究方向是调整ANFIS的结构使其更适合调度问题。     

基于ANFIS的非线性系统辨识研究

系统辨识是控制系统设计的基础,对非线性系统进行辨识是当前的难点;文献[1]提出了用模糊建模方法,文献[2]提出了用神经网络方法,在总结上述方法不足的基础上,该文提出了用自适应神经模糊推理系统(ANFIS)对非线性系统进行辨识的方法,仿真结果表明,ANFIS进行非线性系统辨识是可行的,其辨识精度很高。     

基于自适应神经模糊推理控制的无刷直流电机转子位置控制研究

为提高无刷直流电机(Brushless DC Motor, BLDCM)在瞬态和稳态条件下的控制性能,提出一种基于自适应神经模糊推理系统(ANFIS)的转子位置控制器。推导出无刷直流电机位置控制数学模型;基于Takagi-Sugeno模型设计ANFIS控制器,分别利用反向传输算法和最小二乘估计法对输入和输出参数进行了优化;对正弦信号和斜坡信号下的转子位置控制效果进行了仿真和实验研究。结果表明:该控制器可实现无刷直流电机快速、准确的动态响应,具有良好的稳态性能。通过仿真和试验,验证了ANFIS控制策略的有效性。     

Reactive power compensation of an isolated hybrid power system with load interaction using ANFIS tuned STATCOM

This paper presents an adaptive neuro fuzzy interference system (ANFIS) based approach to tune the parameters of the static synchronous compensator (STATCOM) with frequent disturbances in load model and power input of a wind-diesel based isolated hybrid power system (IHPS). In literature, proportional integral (PI) based controller constants are optimized for voltage stability in hybrid systems due to the interaction of load disturbances and input power disturbances. These conventional controlling techniques use the integral square error (ISE) criterion with an open loop load model. An ANFIS tuned constants of a STATCOM controller for controlling the reactive power requirement to stabilize the voltage variation is proposed in the paper. Moreover, the interaction between the load and the isolated power system is developed in terms of closed loop load interaction with the system. Furthermore, a comparison of transient responses of IHPS is also presented when the system has only the STATCOM and the static compensation requirement of the induction generator is fulfilled by the fixed capacitor, dynamic compensation requirement, meanwhile, is ful-filled by STATCOM. The model is tested for a 1% step increase in reactive power load demand at t = 0 s and then a sudden change of 3% from the 1% at t = 0.01 s for a 1% step increase in power input at variable wind speed model.     

音频信号用于球磨机工况识别的应用研究

为了达到制粉系统的智能自动控制,必需要对球磨机内的工作状态进行识别,利用音频信号的特征频率是一种适用的方法.而在现场测得的声音信号一定会包含有多种噪声信号,特别是当特征信号与噪声信号存在严重的频谱交叠,采取频域滤波方法无法获得较好的效果时,利用基于自适应神经模糊系统(ANFIS)的消噪技术,可以得到良好的分离效果.提高测得数据的精确度,进而有效地实现球磨机的自动控制.     

基于ANFIS的电力系统谐波分析算法

选用电力系统的谐波测量数据,给出了该算法谐波分析的仿真结果,并与傅立叶算法和改进的神经网络算法的计算结果进行了比较,还选用电力系统的高次谐波信号数据,并给出了高次谐波分析仿真结果,表明该算法在处理高次谐波信号方面具有较好的应用前景。     

基于ANFIS的电力客户信用判别模型研究

电费收缴工作的好坏,将直接影响电力公司的经营状况.为了约束电力客户的用电行为,保证电费的及时收缴,事先准确地识别客户的信用状况,是最为关键的内容.考虑到电力公司对客户信息的掌握情况以及信用管理的要求,设计了一套符合实际管理需要的,能够反映电力客户实际信用水平的指标体系,并在此指标体系的基础上利用自适应神经模糊推理系统(ANFIS)建立信用评级模型.该模型经测试样本检验证明是精确可靠的,适用于实际配电管理中对电力客户信用类别的判别.从而为电力公司预知信用风险,并采取不同的信用管理对策,提供了有力的参考依据.