考虑控制电信号中间量的机床运动控制误差数据驱动建模方法

现有数据驱动的机床运动控制误差建模方法通常使用端到端的模型,即通过机器学习算法直接构建参考轨迹信息(速度、加速度等)与伺服误差之间的模型,以降低建模复杂度。然而,该方法忽视了控制电信号对运动控制系统非线性扰动的反映,而导致建立的模型精度受限。为解决此问题,提出了一种使用控制电信号作为中间量的数据驱动运动控制误差建模方法。该方法采集参考轨迹信息(速度、加速度、急动度等)、控制电信号、跟踪误差以及构造的换向特征,构建并训练基于参考轨迹信息的控制电信号预测网络,以及基于电信号和参考轨迹信息的运动控制误差预测网络,利用控制电信号这一中间量有效反应系统所受非线性扰动的特点,实现了高精度的运动控制误差数据驱动建模。在实际验证测试时,将参考轨迹信息输入电信号预测网络,而后将得到的预测控制电信号和参考轨迹信息输入跟踪误差预测网络,即可实现运动控制误差的预测。通过实验对所提出的建模方法进行了验证,所提出方法相对于传统的端到端建模方法,运动控制误差的预测精度在X轴和Y轴分别提升16.33%和20.42%,误差补偿后运动控制轮廓精度相较于未补偿提升85.59%,验证了所提出方法的可行性。     

小样本下改进MLE估计精度的二次估计方法

长寿命机电产品进行加速寿命试验常采用小样本定时截尾的方法,其统计分析方法常采用威布尔分布下的MLE估计.由于形状参数之间存在恒等约束,而分析过程始终没有考虑形状参数和特征寿命参数之间的相关性,这样就可能带来模型参数估计的误差.针对上述小样本恒加试验统计分析精度的问题,该文采用一种构造数据方法来解决分布参数之间的相关性.仿真结果表明:该方法可相应地提高小样本定时截尾恒加试验的MLE统计分析精度并可降低模型误差.     

基于空间绝对距离的机器人自标定研究

研究了一种工业机器人的自标定方法,对六自由度工业机器人的控制模型和误差模型进行了独立建模。提出的标定方法是基于空间中的绝对距离,对测量工具要求并不苛刻,测量过程简单,数据获取便捷可靠。采用解析法消除冗余误差参数,使得参数辨识的结果变得精确可靠。经实验证明,基于空间绝对距离的机器人自标定方法能将机器人的绝对精度提高十几倍。     

基于数据驱动建模的钣金装配过程误差分析

钣金装配过程的误差分析对于消除钣金装配质量故障具有重要意义。现有分析建模方法由于受钣金装配零件的材料、几何形状和装配工艺的限制,难于对钣金装配过程进行准确建模和误差分析。与分析建模方法不同,基于装配体关键产品特征的历史测量数据提出进行钣金装配过程误差分析的数据驱动建模方法。所提方法由工程经验和数学推导,建立钣金装配过程的多元一阶自回归模型和多元部分线性模型。基于极大似然估计方法和最小二乘核光滑估计方法给出所建立模型的参数和非参数估计。四元四工序典型汽车引擎盖的装配实例证明,所提方法在钣金装配误差分析过程中具有有效性。基于数据驱动的建模方法易于建模,分析结果准确可靠,可为钣金装配过程的误差分析提供新思路。     

基于状态空间模型的数控机床热误差建模

为了减小机床运行参数变化导致的机床热误差变化对模型预测精度的影响,提出了状态空间建模算法,该算法可根据机床运行参数的变化而自动调整模型,从而使模型对机床运行参数的变化具有良好的自适应性。通过实验比较了模型对机床处于不同条件下的热误差预测精度,并基于状态空间模型在Leaderway V-450型数控机床上进行了平面切削的热误差补偿实验。实验结果表明：与传统热误差建模算法相比较,所提算法的预测精度提高了58.12%,稳健性也得到了有效提升,且实际热误差补偿效果显著。     

基于高斯过程潜力模型的刀具磨损预测

刀具磨损预测对于提高加工精度和生产效率具有重要意义。刀具磨损预测模型主要包括基于物理的模型和基于数据驱动的模型。基于物理的模型一般使用经验公式或简化公式对刀具磨损过程进行建模，在切削参数变化的情况下其预测精度通常会变低。另一方面，数据驱动模型通过测量数据来估计刀具磨损，没有考虑刀具磨损机理，导致模型泛化性和结果可解释性较差。为了解决这些问题，提出了一种新的用于刀具磨损预测的高斯过程潜力模型。所提出的模型使用高斯过程对刀具磨损物理模型的未知参数进行建模，建立了一个物理信息机器学习模型。高斯过程潜力模型不仅避免了物理模型的参数识别，而且挖掘了来自物理域和数据域的隐藏信息。此外，通过将物理模型与高斯过程的协方差函数相结合，构建了一个物理信息协方差函数来约束模型的输出，提高了预测精度。多工况试验结果表明，所提方法的绝对平均误差和均方根误差分别为2.5945、3.740 8，比传统数据驱动模型的预测误差要更小，预测精度进一步提升。     

威布尔分布参数估计新方法研究

由样本估计威布尔分布的三个参数是工程中的一个重要问题。应用改进的一阶灰色微分方程模型直接建模方法，可以导出不经迭代就估计出威布尔分布三个参数的灰色估计方法。用这种方法估计参数先得到形状参数，不用迭代，简单明了。实际数据计算和大量的蒙特卡洛模拟试验表明，对形状参数大于1的分布，拟合优度好，精度较高。     

基于多体系统理论的非球面磨削误差模型与补偿技术

为提高大中型非球面的磨削精度,从而提高非球面的加工效率,研究轴对称非球面磨削过程的误差模型,并对误差进行补偿.运用多体系统理论,基于一阶线性模型,建立非球面磨削成形的统一误差模型,并且推导各种误差对于最终面形误差的传递函数.基于传递函数特征相似误差集中补偿的方法,将所有趋势项误差转化为砂轮对刀误差以及砂轮形状误差进行补偿,并建立实用补偿模型,从而避免求解、校正各项具体误差.试验结果表明,建立的误差模型和辨识模型正确,可以使面形误差收敛到预期范围,从而解决了轴对称非球面磨削中的精度控制问题.     

无速度反馈的双臂空间机器人模糊滑模控制

针对存在不确定性且无速度反馈的自由漂浮双臂空间机器人关节轨迹跟踪控制问题,提出了一种基于状态观测器的模糊滑模控制方法.根据双臂空间机器人完全驱动动力学方程以及运动学方程,建立自由漂浮状态下系统的关节空间动力学方程.利用模糊系统的万能逼近特性对系统不确定部分进行逼近,并设计状态观测器在线估计系统关节运动的角速度信息.以关节角度和观测器获得的关节角速度作为系统状态反馈,在传统滑模控制方法基础上,进一步考虑系统惯性参数未知导致的建模误差,设计模糊滑模控制器,实现了双臂空间机器人系统关节角度的轨迹跟踪控制.数值仿真验证了所提控制方法的有效性.     

风速风向测量误差补偿算法的研究

为了提高风帆助力船舶中风速风向的测量精度,提出一种基于空间模型的风速风向测量误差补偿算法.首先建立了船舶运动状态下风速风向矢量的空间模型.在应用超声波风速风向测量技术的基础上,通过空间测量的方法确定测量误差与空间倾角的对应关系,并对测量误差进行单向拟合补偿计算和空间拟合补偿计算.最后,经过大量的实验和测量数据的对比,验证了基于空间模型误差补偿算法的有效性和可行性.从而解决了一类风帆助力船舶辅助推进控制系统的风速风向空间测量的误差补偿问题,使得风速风向测量精度满足风帆辅助推进控制系统的要求.     

Parameterized data-driven fuzzy model based optimal control of a semi-batch reactor

A parameterized data-driven fuzzy (PDDF) model structure is proposed for semi-batch processes, and its application for optimal control is illustrated. The orthonormally parameterized input trajectories, initial states and process parameters are the inputs to the model, which predicts the output trajectories in terms of Fourier coefficients. Fuzzy rules are formulated based on the signs of a linear data-driven model, while the defuzzification step incorporates a linear regression model to shift the domain from input to output domain. The fuzzy model is employed to formulate an optimal control problem for single rate as well as multi-rate systems. Simulation study on a multivariable semi-batch reactor system reveals that the proposed PDDF modeling approach is capable of capturing the nonlinear and time-varying behavior inherent in the semi-batch system fairly accurately, and the results of operating trajectory optimization using the proposed model are found to be comparable to the results obtained using the exact first principles model, and are also found to be comparable to or better than parameterized data-driven artificial neural network model based optimization results.     

Improved state space model predictive fault-tolerant control for injection molding batch processes with partial actuator faults using GA optimization

A novel model predictive fault-tolerant control (MPFTC) strategy adopting genetic algorithm (GA) is proposed for batch processes under the case of disturbances and partial actuator faults. Based on the extended state space model in which the tracking error is contained, there are more degrees of freedom provided for the controller design and better control performance is obtained. In order to enhance the control performance further, the GA is introduced to optimize the relevant weighting matrices in the cost function. The effectiveness of the proposed MPFTC approach is tested on the injection velocity regulation of the injection molding process.     

Adaptive dimensional control based on in-cycle geometry monitoring and programming for CNC turning center

This paper presents a method to compensate the dimensional deviation, irrespective of the sources for its components, and to integrate the dimensional verification and dimensional control processes. Nowadays, approach in compensation of dimensional deviation is based on decomposing the deviation. The decomposing criterion is the error source such as positioning errors, thermal deformation, mechanical loads, tool wear, kinematical errors, dynamic force, and motion control. Then, one or even more components are modeled and compensated. On contrary, the proposed method is based on the decomposing of the tool path and consideration of the batch samples. The decomposition criteria ignores the error sources as: (1) speed of variation in space of the total deviation value for the tool path decomposition and (2) the speed of variation in time of the deviation model parameters values for batch samples decomposition. The data from the geometry holistic monitoring are used for both modeling and compensation of systematic component of the total error, also for checking the compliance with technical requirements. Two algorithms for processing of the data provided by geometry monitoring, namely the adaptive?Cpredictive algorithm and adaptive?Coptimal algorithm, are presented. Nine experimental batches were machined to verify the efficiency of the proposed method using various model structures and processing algorithms. The results of method application have shown a reduction of deterministic and even nondeterministic part of the total error in what concern accuracy and precision. For the entire batch, the level of remanent error is less than 5% for deterministic part, and less than 75% for nondeterministic part. These results are clearly better than the other results reported; moreover, they refer to the whole processing error and entire batch.     

Improved design of constrained model predictive tracking control for batch processes against unknown uncertainties

In this paper, an improved constrained tracking control design is proposed for batch processes under uncertainties. A new process model that facilitates process state and tracking error augmentation with further additional tuning is first proposed. Then a subsequent controller design is formulated using robust stable constrained MPC optimization. Unlike conventional robust model predictive control (MPC), the proposed method enables the controller design to bear more degrees of tuning so that improved tracking control can be acquired, which is very important since uncertainties exist inevitably in practice and cause model/plant mismatches. An injection molding process is introduced to illustrate the effectiveness of the proposed MPC approach in comparison with conventional robust MPC.     

Hybrid modeling and dynamic simulation of automated batch plants

A proposal is presented for the hybrid modeling and dynamic simulation of automated batch processing plants with dominant discrete-event behavior. The proposal encompasses current techniques for modeling continuous-time/discrete-event processes, synthesizing discrete-event controllers as well as the use of industrial standards for batch control. The result is a hierarchical-modular model of a plant in which process and control tasks are clearly differentiated. Implementation rules are established for a specific dynamic simulator capable of handling hybrid systems. The approach is demonstrated by building and verifying a complete model for an automated milk pasteurization plant. Results show that handling problems of realistic complexity is feasible using state-of-the-art technology.     

基于分段线性MPCA的充液氨过程建模及监控

为深入分析晶变改性装备中反应釜充液氨工艺过程特性,同时也为了进行更精确、有效的过程监控,提出了一种分段线性多向主元分析(MPCA)方法,实现该间歇过程的建模及故障诊断。首先将三维建模数据降维展开并进行主元分析(PCA),得到采样点数据特征矩阵;然后基于间歇过程的多操作时段特性,根据特征矩阵相似度最大隶属原则,通过设定窗口长度可调的滑动聚类器,在系统运行轨迹上分别从数据变异的方向和幅值对采样点进行2次聚类划分,将间歇过程操作周期分割为一系列运行特性相同的子时段,最后利用各个子时段的线性MPCA模型逼近系统非线性运行过程,实现间歇生产过程建模及在线监测。通过反应釜充液氨工艺过程的实际应用,结果表明提出的建模方法能够对该间歇生产进行合理的、易于从过程特性角度解释的时段划分,所建模型具有可靠的监控性能,能及时准确检测出系统运行过程中的异常情况。     

Sliding mode control with self-tuning law for uncertain nonlinear systems

A robust sliding mode control that follows a self-tuning law for nonlinear systems possessing uncertain parameters is proposed. The adjustable control gain and a bipolar sigmoid function are on-line tuned to force the tracking error to approach zero. Control system stability is ensured using the Lyapunov method. Both simulation and experimental application of a planetary gear type inverted pendulum control system verify the effectiveness of the developed approach.     

基于并行计算的平面可调七杆机构动力学解析模型研究

基于运动学分析、凯恩动力学方程及数字-符号法,提出采用并行计算建立平面可调七杆机构动力学解析模型的方法,并研究了构件杆长及惯性参数变化对驱动力/力矩的影响.利用封闭矢量法对平面可调七杆机构进行运动学分析,得到各构件的速度和加速度表达式;将独立广义坐标、杆长及惯性参数作为符号量,其余参数处理为数值量,导出动力学解析模型的数字-符号表达式,并构造了解析模型的并行算法.由于动力学模型及实时代码优化是离线建立的,并且采用并行计算结构,减少了在线计算时间,从而为实时控制打下了基础.给出的仿真实例证明了此方法的有效性.     

Fault detection and classification in chemical processes based on neural networks with feature extraction

Feed forward neural networks are investigated here for fault diagnosis in chemical processes, especially batch processes. The use of the neural model prediction error as the residual for fault diagnosis of sensor and component is analyzed. To reduce the training time required for the neural process model, an input feature extraction process for the neural model is implemented. An additional radial basis function neural classifier is developed to isolate faults from the residual generated, and results are presented to demonstrate the satisfactory detection and isolation of faults using this approach.