Modeling and development of an ANN-based smart pressure sensor in a dynamic environment

In many engineering applications, a capacitive pressure sensor (CPS) is placed in a dynamic environment in which the temperature variation is quite large. Since the response characteristics of a CPS are highly nonlinear and temperature dependent, in such situations, complex signal processing techniques are needed to obtain correct readout of the applied pressure. We have proposed an artificial neural network (ANN)-based smart capacitive pressure sensor, whose response characteristics can be estimated within an accuracy of ±1% error over a wide variation of temperature starting from −50°C to 150°C. This modeling scheme automatically takes care of all the nonidealities, such as, nonlinearity, offset, gain and temperature dependence, of the sensor. A novel idea of automatic collection of temperature information and its feeding into the ANN model is also proposed. In the practical implementation of this scheme, the hardware complexity poses a serious impairment. Since the tanh() functions are needed for implementation in the ANN-based model, to reduce the hardware requirement, we provide a simple scheme for computation of tanh(). Sensitivity analysis of the model with respect to the finite word-length constraint on the final stored weight values, and number of terms used in the implementation of tanh() function, have been carried out. A microcontroller-based implementation scheme for the ANN-based model is also suggested.     

人工神经网络与智能 CAD

通过分析目前ＣＡＤ专家系统中存在的问题,对人工神经网络在智能ＣＡＤ中应用的理论和方法进行了探讨。文中介绍了人工神经网络在模拟形象思维、知识获取、知识表示等方面的应用,并对存在的一些问题进行了详细的讨论。     

一种基于模糊神经网络确定电火花加工条件的新方法

在电火花加工中,影响因素较多,难以确定最优加工条件。应用模糊神经网络确定电火花加工条件是一种新的尝试,它可以充分发挥模糊逻辑和神经网络的长处。为方便操作者决定粗加工最优加工条件,并提高表面加工效果和粗加工速度,本文提出了一种基于模糊神经网络自动确定和优化电火花成形加工中加工参数的方法。经实践证明,用该方法所确定的最优加工条件,能保证较高的粗加工速度,方便了操作者对加工条件的确定。     

数控车削加工过程的刀具磨损动态监测

对车削加工刀具磨损的各阶段信号进行采集,通过动态时频域分析,找到车削加工过程中刀具磨损的重要参数变化,对其进行振铃记数和人工神经网络的模式识别,实现了车削加工刀具磨损的状态检测。     

人工神经网络技术在CBN砂轮磨削表面粗糙度研究中的应用

针对CBN砂轮磨削 ,采用人工神经网络方法建立由磨削用量确定表面粗糙度的预测模型。计算结果证明 ,所建立的人工神经网络模型可很好地描述砂轮速度、砂轮进给速度、工件转速对磨削表面粗糙度的影响。预测结果具有良好的精度并得到了验证试验的检验。通过本模型 ,利用有限的试验数据可得出整个工作范围内表面粗糙度的预测值 ,可大量减少试验费用     

Self-improving process control for molecular beam epitaxy

This paper addresses manufacturing research involving advances in material process control. The research objective has been to develop intelligent, self-directed and self-improving control systems which enablein situ (real-time) control path generation based on both product (material behaviour) and processing (control agent) feedback. A product-process control philosophy which emphasises product quality is described together with a generic architecture for representing product and process knowledge.Specific details are presented involving the development and application of a self-directed and self-improving material processing system for molecular beam epitaxy of gallium arsenide wafers. Special emphasis is given to the development of a neural model for self-improving control as well as future research directions.     

Hill非二次屈服准则次数m值识别及应变控制研究

正确确定本构模型中的物性参数是金属成形过程准确分析和模拟的基础。以Hill非二次屈服准则为基础,应用人工神经网络(ANN)技术建立了材料在变形过程中不同应力状态下物性参数m值的识别方法,并在MTS试验机上进行了薄壁管拉扭试验,通过试验中各阶段的实际应变增量值与m值识别前后计算所得理论应变增量值的比较,验证了识别所得m值以及根据识别所得m值进行应变控制的正确性。     

传感器动态补偿的神经网络逆系统方法

给出一种基于神经网络逆系统的传感器动态补偿策略。无需传感器具体模型和参数，即可实现传感器系统的近似单位线性化，达到动态补偿的目的。仿真实验和动态标定试验结果表明，应用这种新型的易于工程实现的动态补偿方法可显著地提高传感器的动态特性，有效改善传感器的动态品质。     

神经网络在校准测量仪器中的应用

本文介绍了人工神经网络的基本原理及BP学习算法，提出了对测量仪器进行校准的一种新方法，并用万用表的校准实验证明了该方法的有效性。     

快走丝电火花线切割加工仿真系统

通过神经网络技术建立了快走丝电火花线切割加工工艺模型 ,利用穷举法建立了具有一定人工智能的工艺参数全局优化系统 ,开发了模具电火花加工过程仿真系统。该系统不仅可以精确预测加工效果 ,而且克服了工艺参数表的局限性 ,弥补了建立在工艺参数表基础上的参数自动选取系统的缺陷 ,实现了工艺参数全局最优化。测试结果及实际使用结果表明本文所建立的仿真系统反映了机床的加工工艺特性 ,预测误差基本控制在 8%内 ,系统的参数优化选取功能使机床的加工性能得以充分发挥。仿真系统具有广泛的通用性 ,可适用于不同类型的线切割加工机床。     

扩散硅压力传感器灵敏度温度漂移的调整

根据扩散硅压力传感器灵敏度温漂产生的原理,分析了灵敏度与温度的关系。并依据实验数据论述了灵敏度系数和扩散表面杂质浓度的关系,以及电阻温度系数与表面杂质浓度的关系。简单介绍通过调整和控制表面杂质浓度等参数,得到所需的灵敏度温度漂移曲线。     

Prediction of brake friction materials recovery performance using artificial neural networks

The brake friction materials in an automotive brake system are considered as one of the key components for overall braking performance of a vehicle. The sensitivity of friction material performance and accordingly brake performance, versus different operating regimes, has always been an important aspect of its functioning. In this paper, the influences not only on the brake operation conditions but also on the formulation and manufacturing conditions of friction materials have been investigated regarding friction materials recovery performance by means of artificial neural networks. A new neural network model of friction material recovery performance, trained by the Bayesian Regulation algorithm, has been developed.     

Accurate simplistic predictive modeling of nonlinear dynamic processes

In this article, a new simplistic way of predictive modeling of process variables in nonlinear dynamic processes is introduced. This approach, which is semi-empirical, is demonstrated on a simulated continuous stirred tank reactor. Model development uses a first-order-plus-dead-time structure and only two or three input changes for determining the coefficients. This approach is evaluated for a variety of situations which include measured output, unmeasured output, extrapolation beyond the input range, various levels of dead time, various levels of measurement error, large dynamics, and various levels of nonlinear behavior. In the situation of unmeasured output, the proposed approach is very accurate and in the other cases it is extremely accurate and far superior to linear regression and artificial neural twork models.     

人工神经网络在智能机械设计中的应用

介绍了智能ＣＡＤ的概念和发展,分析了人工神经网络（ＡＮＮ）的特点,针对目前机械设计专家系统存在的问题,提出将ＡＮＮ应用到专家系统的设计中,是进行智能ＣＡＤ的一条有效途径。介绍了ＡＮＮ在概念设计、设计过程中形象思维的模拟、知识的获取和表示、回溯问题的模拟等方面的应用。     

A smart force platform using artificial neural networks

The human body may interact with structures and these interactions are developed through the application of contact forces, for instance when walking. The aim of this paper is to propose a new methodology using Artificial Neural Network (ANN) for calibrating a force platform in order to reduce the uncertainties in the values of estimated vertical Ground Reaction Force and the positioning of the applied force in the human gait. Force platforms have been used to evaluate the pattern of human applied forces and to fit models for the interaction between pedestrians and structures. Linear relation assumptions between input and output are common in traditional Least Mean Square methods used in calibration. Some discrepancies due to nonlinearities in the experimental setup (looseness, wear, support settlements, electromagnetic noise, etc.) may harm the overall fitting. Literature has shown that nonlinear models, like ANN, can better handle this. During the calibration, the input data to the ANN were the reference voltages applied to the Wheatstone bridge, while the output data were the values of the standard weights applied in the force platform in defined sites. Supervised training based on k-fold cross validation was used to check the ANN generalization. The use of ANN shows significant improvements for the measured variables, leading to better results for predicted values with low uncertainty when compared to the results of a simple traditional calibration using Least Mean Squares.