Recurrent RBFN-based fuzzy neural network control for X-Y-theta motion control stage using linear ultrasonic motors

A recurrent radial basis function network (RBFN) based fuzzy neural network (FNN) control system is proposed to control the position of an X-Y-theta motion control stage using linear ultrasonic motors (LUSMs) to track various contours in this study. The proposed recurrent RBFN-based FNN combines the merits of self-constructing fuzzy neural network (SCFNN), recurrent neural network (RNN), and RBFN. Moreover, the structure and the parameter learning phases of the recurrent RBFN-based FNN are performed concurrently and on line. The structure learning is based on the partition of input space, and the parameter learning is based on the supervised gradient decent method using a delta adaptation law. The experimental results due to various contours show that the dynamic behaviors of the proposed recurrent RBFN-based FNN control system are robust with regard to uncertainties.     

Recurrent Radial Basis Function Network-Based Fuzzy Neural Network Control for Permanent-Magnet Linear Synchronous Motor Servo Drive

We propose a recurrent radial basis function network-based (RBFN-based) fuzzy neural network (FNN) to control the position of the mover of a field-oriented control permanent-magnet linear synchronous motor (PMLSM) to track periodic reference trajectories. The proposed recurrent RBFN-based FNN combines the merits of self-constructing fuzzy neural network (SCFNN), recurrent neural network (RNN), and RBFN. Moreover, it performs the structureand parameter-learning phases concurrently. The structure learning is based on the partition of input space, and the parameter learning is based on the supervised gradient descent method, using a delta adaptation law. Furthermore, all the control algorithms are implemented in a TMS320C32 DSP-based control computer. The simulated and experimental results due to periodic reference trajectories show that the dynamic behaviors of the proposed recurrent RBFN-based FNN control system are robust with regard to uncertainties     

Inexact rough-interval type-2 fuzzy stochastic optimization model supporting municipal solid waste management under uncertainty

In this study, an inexact rough-interval type-2 fuzzy stochastic linear programming (IRIT2FSLP) approach is developed for addressing uncertainties presented as rough-interval, type-2 fuzzy and random variables. The proposed method is applied to the case of a long-term municipal solid waste management system. The IRIT2FSLP approach is an extension of the inexact interval linear programming for handling nonlinear stochastic optimization problems where rough-interval and type-2 fuzzy parameters are integrated into a general framework. The results indicate that IRIT2FSLP normally leads to rough-interval solutions. Comparisons of the proposed model with scenarios without rough-interval and type-2 fuzzy parameters are also conducted. The results indicate the significant impact of dual-uncertain information on the system, which implies the reliability of IRIT2FSLP in handling waste flow allocation.     

一种改进的模糊神经网络控制器学习算法

提出了一种改进的模糊神经网络混合学习算法,运用遗传算法优化构成隶属函数的网络结构,运用最小二乘法进行解模糊,具有更高的学习精度和更快的收敛速度,解决了在多变量系统中采用模糊神经网络时学习收敛慢且易陷入局部极小点的问题。     

运用模糊神经网络的汽车座椅舒适性评价

汽车座椅舒适性受到消费者的高度关注。由于复杂性和主观性等因素的影响,使得汽车座椅舒适度评价成为座椅制造企业和整车企业的一个技术难点。提出一种基于模糊神经网络的座椅舒适度评价模型,应用8个座椅压力分布参数和2个人体参数等对座椅舒适性进行综合评价。实例分析显示:与普通BP网络模型对比,模糊神经网络模型具有更好的预测效果。     

大型氦低温系统多变量控制

大型氦低温系统广泛应用于各类大科学装置中,运行中往往会产生热脉冲,通过负载端传导给制冷系统,对制冷系统产生热冲击。为了研究和应对热冲击,建立了一种多变量控制策略并得到了相关仿真和实验结果。首先以真实系统为基础建立了氦低温系统的动态仿真模型,同时建立了一个基于模糊神经网络的多变量协同控制策略,并将其应用在仿真液化器模型和一个真实的氦透平制冷系统上,得到了低温系统降温过程和控制过程的仿真和实验数据。仿真和实验结果显示本策略的偏差积分量为0.016 5,下降时间为102 s,上升时间为112 s。普通PID的的偏差积分量为0.026 9,下降时间为154 s,上升时间为170 s。通过仿真和实验过程的比较,验证了本文建立的动态仿真模型具有可用的精度,证明了本策略具有较好的控制效果。     

基于区间二型模糊EDAS的多属性决策方法

针对多属性决策存在的复杂性和不确定性问题,提出基于区间二型模糊平均解距离法 (evaluation based on distance from average solution,EDAS) 的多属性决策方法。采用区间二型模糊集合 (interval type-2 fuzzy sets,IT2FS) 表达评价信息解决专家的偏好信息存在个体化差异问题,并纳入EDAS对备选方案进行排序。以区间二型模糊数表达评价信息构建决策矩阵,以计算得到的综合评价值的去模糊化结果作为最终的方案排序依据。针对EDAS中属性权重需要从外部获取的问题,采用区间二型模糊集合改进的最优最劣法 (best-worst method,BWM) 确定属性权重。最后,以某汽车制造企业选购新能源汽车云服务方案为例,验证所提方法的有效性。     

Generating Type 2 Trapezoidal Fuzzy Membership Function Using Genetic Tuning

Fuzzy inference system (FIS) is a process of fuzzy logic reasoning to produce the output based on fuzzified inputs. The system starts with identifying input from data, applying the fuzziness to input using membership functions (MF), generating fuzzy rules for the fuzzy sets and obtaining the output. There are several types of input MFs which can be introduced in FIS, commonly chosen based on the type of real data, sensitivity of certain rule implied and computational limits. This paper focuses on the construction of interval type 2 (IT2) trapezoidal shape MF from fuzzy C Means (FCM) that is used for fuzzification process of mamdani FIS. In the process, upper MF (UMF) and lower MF (LMF) of the MF need to be identified to get the range of the footprint of uncertainty (FOU). This paper proposes Genetic tuning process, which is a part of genetic algorithm (GA), to adjust parameters in order to improve the behavior of existing system, especially to enhance the accuracy of the system model. This novel process is a hybrid approach which produces Genetic Fuzzy System (GFS) that helps to enhance fuzzy classification problems and performance. The approach provides a new method for the construction and tuning process of the IT2 MF, based on the FCM outcomes. The result is compared to Gaussian shape IT2 MF and trapezoid IT2 MF generated by the classic GA method. It is shown that the proposed approach is able to outperform the mentioned benchmarked approaches. The work implies a wider range of IT2 MF types, constructed based on FCM outcomes, and an optimum generation of the FOU so that it can be implemented in practical applications such as prediction, analytics and rule-based solutions.     

An optimization model under interval and fuzzy uncertainties for a by-product gas system of an iron and steel plant

The effective utilization of by-product gas is essential for achieving the targets of energy conservation and emission reduction of iron and steel plants in China. The application of deterministic optimization methods may lead to oversimplification and inaccurate estimation of system parameters, and even to system failure. The major contributions made by this study are the development of a gas scheduling optimization model under fuzzy and interval uncertainties and it application to the gas scheduling system of the Baotai steel plant. The integration of type-1 and type-2 fuzzy sets and interval numbers was first used to describe specific model parameters, and the reduced fuzzy chance-constrained programming algorithm and interactive two-step interval algorithm were used for model solution. Compared with practical allocation patterns, it is shown that the proposed model could offer better solutions with more outstanding performance in rapid response to production fluctuations, as well as increases in system revenue.     

A hybrid interval-parameter fuzzy robust programming method and its application to filter management strategy in fluid power systems

An interval-parameter fuzzy robust programming (IFRP) method is developed for the assessment of filter allocation and replacement strategies in a fluid power system (FPS) under uncertainty. The developed IFRP can effectively handle the uncertainties expressed as fuzzy sets, interval values, and their combinations, which exist in contaminant ingression/generation of the system and contaminant-holding capacity of filter without making assumptions on their probabilistic distributions. The fuzzy decision space can be delimited into a more robust one with the uncertainties being specified through dimensional enlargement of the original fuzzy constraints, leading to enhanced robustness for the optimization process. Results indicate that the developed IFRP can not only help decision-maker to identify optimal filter allocation and replacement strategies to control the contamination level of FPS with a minimized system-cost and system-failure risk under multiple uncertainties, but also mitigate uncertainties through abating interval widths of the replacement periods and service life under different contamination ingression/generation rates.     

新型磁流变脂阻尼器对铁路连续梁桥地震响应模糊神经网络控制

为解决传统磁流变液阻尼器(Magneto Rheological Fluid Damper,MRFD)磁场利用率低及磁流变液沉降导致控制特性劣化难题,提出新型阻尼器—磁流变脂阻尼器(Magneto Rheological Grease Damper,MRGD)。采用神经网络(Neural Network,NN)对足尺MRGD动力特性进行辨识,通过将改进的限幅最优(Modified Clipped-Optimal,MCO)算法整合到模糊神经网络(Fuzzy Neral Network,FNN)理论来设计适合MRGD的FNN/MCO半主动控制策略,并构建SIMULINK仿真分析平台。以典型三跨铁路连续梁桥为工程背景,分别对未控制、FNN/MCO半主动控制及线性二次型高斯(Linear Quadratic Gaussian,LQG)主动控制下桥梁各项评价指标进行分析。结果表明,所提FNN/MCO半主动控制策略对桥梁地震响应控制效果明显优于LQG主动控制策略;FNN/MCO策略较LQG策略更利于控制装置性能发挥;FNN/MCO策略稳定性、鲁棒性均明显优于LQG策略。     

Recurrent Functional-Link-Based Fuzzy Neural Network Controller With Improved Particle Swarm Optimization for a Linear Synchronous Motor Drive

A recurrent functional link (FL)-based fuzzy neural network (FNN) controller is proposed in this study to control the mover of a permanent-magnet linear synchronous motor (PMLSM) servo drive to track periodic reference trajectories. First, the dynamic model of the PMLSM drive system is derived. Next, a recurrent FL-based FNN controller is proposed in this study to control the PMLSM. Moreover, the online learning algorithms of the connective weights, means, and standard deviations of the recurrent FL-based FNN are derived using the back-propagation (BP) method. However, divergence or degenerated responses will result from the inappropriate selection of large or small learning rates. Therefore, an improved particle swarm optimization (IPSO) is adopted to adapt the learning rates of the recurrent FL-based FNN online. Finally, the control performance of the proposed recurrent FL-based FNN controller with IPSO is verified by some simulated and experimental results.      

基于模糊RBF神经网络的水下机械臂控制研究

针对水下复杂工作环境下机械臂控制性能易受影响,而传统控制方法效果不佳的问题,提出了一种基于模糊RBF（radial basis function,径向基函数）神经网络的智能控制器,用于精确、稳定地控制水下机械臂。考虑到在水扰动环境下,机械臂通常受到附加质量力、水阻力和浮力的影响,运用拉格朗日法和Morison方程,建立包含水动力项的二杆机械臂动力学模型,通过模糊RBF神经网络对水下机械臂动力学方程中的水动力不确定项进行总体识别并拟合,利用模糊系统启发式搜索和RBF神经网络推理速度较快的优点,使水下机械臂系统具有较高的控制精度和较强的自适应性。考虑到水动力项,采用Lyapunov稳定性理论验证了水下机械臂系统的稳定性。最后利用MATLAB对二杆机械臂进行轨迹跟踪控制仿真实验,并对比模糊RBF神经网络与常规RBF神经网络识别方法和传统模糊控制方法的控制效果。仿真结果表明：与常规RBF神经网络识别方法相比,模糊RBF神经网络控制下二杆机械臂关节1的响应时间缩短了91%,相对误差减小了88%,关节2的响应时间缩短了92%,相对误差降低了77%;与传统模糊控制方法相比,关节1的相对误差减小了65%,关节2的相对误差减小了10%。研究结果表明模糊RBF神经网络的控制效果优于常规RBF神经网络识别方法和传统模糊控制方法,可为水下机械臂的控制提供一种精度较高、较有效的方法。     

Fuzzy logic for scaling finite element solutions of electromagneticfields

Artificial intelligence (AI) has been used to determine the quasi-stationary two-dimensional electromagnetic fields within rectangular boundaries. Amplitude and phase of magnetic vector potential have been calculated in an iron slot with an embedded current carrying conductor. A suitable fuzzy neural network (FNN) for scaling finite elements electromagnetic field calculations has been developed. FNN has been trained, using finite elements calculations within rectangular boundaries. Then, FNN has been used to calculate the field in a new geometry differing significantly from the geometries used for training. It was concluded that FNN may be used to scale results from one geometry to another with negligible errors     

模糊神经技术在刀具故障检测中的应用

提出了一种刀具故障检测方法，该方法把模糊逻辑和神经网络结合起来，并用神经网络分解技术，建立了一个刀具状态识别网络。该网络适用于多传感器对刀具复杂状态进行识别和分类，具有训练时间短，执行速度快，可靠性高，抗噪能力强等特点。     

基于高斯基模糊神经网络的漂浮基柔性空间机械臂自学习控制

讨论了载体位置不受控、姿态受控情况下,自由漂浮柔性空间机械臂的高斯基模糊神经网络自学习控制问题。利用拉格朗日方程和模态综合法可建立柔性空间机械臂的动力学模型,但由于此类空间机器人系统严格遵守动量守恒,其动力学方程表现出强烈的非线性性质。结合神经网络和模糊控制,即利用神经网络来实现模糊推理可使模糊控制具有自学习能力,在此基础上,设计了柔性空间机械臂关节空间的高斯基模糊神经网络自学习控制方案。由于将动量守恒定理耦合到系统动力学方程的推导过程中,所提出的控制方案具有不需要测量、反馈载体位置、移动速度和移动加速度的显著优点。系统的数值仿真,证实了方法的有效性。     

Neuro-fuzzy network for flavor recognition and classification

This paper presents the neuro-fuzzy Takagi-Sugeno-Kang (TSK) network for the recognition and classification of flavor. The important role in this process fulfills the self-organizing process used for the creation of the inference rules. The self-organizing neurons perform the role of clustering data into fuzzy groups with different membership values (the preprocessing stage). Applying the automatic control of clusters, we have the optimal size of the TSK network. The developed measuring system has been applied for the recognition of flavor of different brands of beer. The fuzzy neural network is used for processing signals obtained from the semiconductor sensor array. The results of numerical experiments have confirmed the excellent performance of such solutions.     

Prediction of the mechanical properties of forged Ti–10V–2Fe–3Al titanium alloy using FNN

In this paper, a fuzzy neural network (FNN) prediction model has been employed to establish the relationship between processing parameters and mechanical properties of Ti–10V–2Fe–3Al titanium alloy. In establishing these relationships, deformation temperature, degree of deformation, solution temperature and aging temperature are entered as input variables while the ultimate tensile strength, yield strength, elongation and area reduction are used as outputs, respectively. After the training process of the network, the accuracy of fuzzy model was tested by the test samples and compared with regression method. The obtained results with fuzzy neural network show that the predicted results are much better agreement with the experimental results than regression method and the maximum relative error is less than 7%. And the optimum matching processing parameters can be quickly selected to achieve the desired mechanical property based on the fuzzy model. It proved that the model has a good precision and excellent ability of predicting.     

基于混合智能的刀具状态在线识别

提出了一个将小波包分析方法,模糊理论及人工神经网络技术相结合的智能刀具状态在线监测系统,系统利用小波包方法将声发射信号分解为不同频带的时间序列,从中抽取出与刀具切削状态紧密相关的序列信号的构方根值（RMS）作为信号特征值,为了表示刀具状态与特征值之间的关系提出了一个模糊神经网络模型,采用了自组织竞争学习与BP算法相结合的混合学习算法,可迅速,可靠地对刀具状态进行识别。     

基于模糊神经网络的设备故障预测研究

对模糊神经网络技术进行了研究，提出了预测分析的模糊神经网络模型；建立了故障指标评定方法，利用预测算法运用参数历史故障指标对参数指标进行趋势预测，预测得到的参数指标可以根据专家诊断系统判据进行诊断，对未来设备的健康状况进行可信度较高的评估。经仿真结果验证，该算法预测精度较高，预测结果可信.