BJEST: A reverse algorithm for the real-time predictive maintenance system

An algorithm, to estimate the machine system parameters from the motion current signature, based upon non-linear time series techniques for use in the real-time predictive maintenance system is presented in this paper. Earlier work has introduced the use of a neural-network approach to learn non-linear mapping functions for condition monitoring systems. However, the performance of the neural-network largely depends upon the quality of the training data, and that of the quality and type of the pre-processing of the input data. A reverse algorithm called BJEST (Bansal–Jones Estimation), for estimating the machine input parameters using the motion current signature, has been designed and proven to be successful in estimating the macro-dynamics of the motion current signature. This motivated the enhancement of the predictive analysis to incorporate non-linear characteristic of the motion current signature. The results show considerable improvement in the estimation of the parameters using the enhanced BJEST algorithm due to estimation consistency, hence, improving the real-time predictive maintenance system.     

Application of a real-time predictive maintenance system to a production machine system

This paper develops earlier work that established the possibility of the classification of machine system parameters on the basis of motion current signature, using a neural network approach. A neural network requires a large amount of training data, which is impractical to generate using a production machine for real-time predictive maintenance system. Hence, a simulation model, which mapped the system parameters to the motion current signature, was developed. The accuracy of the system, to predict the changes in the value of the machine system parameter, is a direct function of the validity of the simulated data. Thus, the objective validation of the simulation model is important to ascertain that the simulation model is accurate with regards to its purpose.In this paper, the simulation model is validated against an on-line production machine. Various approaches to validate the simulation model are applied and a simulation model is developed.     

Deep learning-based human motion recognition for predictive context-aware human-robot collaboration

Timely context awareness is key to improving operation efficiency and safety in human-robot collaboration (HRC) for intelligent manufacturing. Visual observation of human workers’ motion provides informative clues about the specific tasks to be performed, thus can be explored for establishing accurate and reliable context awareness. Towards this goal, this paper investigates deep learning as a data driven technique for continuous human motion analysis and future HRC needs prediction, leading to improved robot planning and control in accomplishing a shared task. A case study in engine assembly is carried out to validate the feasibility of the proposed method.     

Vibration monitoring for defect diagnosis of rolling element bearings as a predictive maintenance tool: Comprehensive case studies

Vibration monitoring and analysis in rotating machineries offer very important information about anamolies formed internal structure of the machinery. The information gained by vibration analysis enable to plan a maintenance action. In this study, the vibration monitoring and analysis case studies were presented and examined in machineries that were running in real operating conditions. Failures formed on the machineries in the course of time were determined in its early stage by the spectral analysis. It was shown that the vibration analysis gets much advantage in factories as a predictive maintenance technique.     

基于HUNUC系统的数控机床维护技术

通过对基于HUNUC系统的数控机床的维护实例分析,介绍了如何对一台长期闲置的数控机床进行检修并恢复正常功能,重点介绍了技术资料不全的情况下,利用万用表等工具快速准确判别伺服运动轴接口与伺服驱动器接口定义,以达到正确通信的目的.     

A virtual machine concept for real-time simulation of machine tool dynamics

When designing CNC machine tools it is important to consider the dynamics of the control, the electrical components and the mechanical structure of the machine simultaneously. This paper describes the structure and implementation of a concept for real-time simulation of such machine tools using a water jet cutting machine as an application. The concept includes a real control system, simulation models of the dynamics of the machine and a virtual reality model for visualisation. The real-time capability of the concept, including the simulation of electrical and rather detailed mechanical component models is proofed. The validation process indicates good agreement between simulation and measurement, but suggests further studies on servo motor, connection and flexibility modelling. However, already from the initial simulation results presented in this paper it can be concluded that the influence of structural flexibility on manufacturing accuracy is of importance at desired feeding rates and accelerations. The fully automated implementation developed in this work is a promising base for dealing with this trade-off between productivity and accuracy of the manufacturing process through multidisciplinary optimisation.     

Thermal behavior of a machine tool equipped with linear motors

Development of a feed drive system with high speed and accuracy has been a major issue in the machine tool industry. Linear motors can be used as efficient tool to achieve the high speed and accuracy. However, a high speed feed drive system with linear motors, in turn, can generate heat problems. Also, frictional heat is produced at the ball or roller bearing of LM block when driven at high speed. It can affect the thermal deformation of the linear scale as well as that of the machine tool structure.In this paper, important heat sources and resulting thermal errors in a machine tool equipped with linear motors were investigated when it was operated at high speed. The thermal deformation characteristics were identified through measuring the thermal error caused from thermal deformation of the linear scale and the machine tool structure. The dominant thermal error components were identified from the thermal error analysis using finite element method. It was shown that the proposed analysis scheme is efficient in identifying the dominant thermal error components and its magnitudes such as the thermal expansion and movement of the linear scale, thermal deformation of the machine tool slide.     

神经网络在机床主轴热位移监控系统中的应用

本文在介绍机床主轴热位移监控系统的基础上,重点分析检测热位移的电涡流传感器的工作原理,及其神经网络在在线定标中的应用。     

A generalized modelling methodology for optimized real-time compensation of thermal deformation of machine tools and CMM structures

Thermal deformation of the machine tool structure plays a critical role in controlling the level of machining accuracy. Since thermally-induced errors cannot completely be eliminated at the design stage, the use of indirect compensation systems emerges as the most economical and inevitable course of action. These systems are based on two different, but yet complimentary, approaches; the use of empirical compensation function and the on-line execution of numerical simulation models. Recognizing the existence of a mathematical similarity between the real process and a simplified dynamic model, the generalized modelling approach is proposed to resolve the main problems associated with existing indirect compensation methods. It is demonstrated that the proposed methodology can be used to identify the true global optimum position and minimum number of the temperature sensors with the highest contribution to the compensation function. These issues have been the focus of a recent research activity world-wide and remain the main obstacles limiting the acceptance of the empirical compensation approach by industry. The use of the generalized modelling method to improve the thermal deformation predictions, through filtering out random temperature measurement errors, is also demonstrated. The application of the generalized modelling method to compensation systems based on numerical simulation is discussed. It is shown that the method can reproduce the accuracy of the finite-element solution, but two orders of magnitude faster, for on-line prediction of the transient temperature field in the machine tool structure. The use of the generalized modelling method to test and to improve the accuracy of finite-element models is also discussed.     

Switching and memory devices based on a polythiophene derivative for data-storage applications

In this article, we report electrical characteristics of devices based on oriented and unoriented films of a polymer, namely poly[3-(6-methoxyhexyl)thiophene]. The current–voltage characteristics of sandwiched devices, based on unoriented polymer, showed hysteresis behavior, while oriented versions exhibited switching characteristics, i.e. presence of two conducting states depending on sweep direction of voltage scans. The ratio between the device current of two conducting states has been as high as 10⁵. This is comparable, if not better, than the results reported so far with complicated device architecture or doped polymeric materials. We have also demonstrated that the switching devices have an associated memory effect for data-storage applications.     

A novel approach for modelling of water jet peening

In this paper, a novel approach, proposed for predicting residual stresses induced on materials treated with high pressure water jets, i.e. water jet peening, is presented. This approach considers the impact pressure distribution due to high velocity droplets impinging on the material surface instead of stationary pressure distribution considered in Trans ASME J Eng Mat Technol 121 (1999) 336 for prediction of residual stresses on water jet peened surfaces. It makes use of Reichardt’s theory for predicting the velocity distribution of droplets and liquid impact theory for predicting the impact pressure and duration of impact of high velocity droplets. For predicting residual stresses on the surface and sub surface of material subjected to water jet peening, finite element modelling approach was adopted by considering the transient, dynamic nature of droplets for analysis. The effectiveness of the proposed approach was demonstrated by comparing the predicted residual stresses with those predicted employing the approach proposed in Trans ASME J Eng Mat Technol 121 (1999) 336. Finally, the practical relevance of the proposed approach was shown by comparing the predicted results with the experimental results obtained by water peening of 6063-T6 aluminium alloy.     

A model for predicting austenite to bainite phase transformation in producing dual phase steels

Based on the theory of free energy change between austenite and bainite phases during the cooling period, a material model is developed to predict the incubation time of the transformation from austenite to bainite for 50CrNiMo deformed at different strain levels and austenite states. Further constitutive equations are developed to model the growth of the nucleated bainite phase. These enable the bainite volume fraction to be predicted for different material cooling conditions after hot forming. To determine the material constants within the phase transformation model, an objective function is formulated to quantify the relative errors between experimental and computed data. The phase transformation model and the objective functions are implemented into an Evolutionary Programming (EP) based software package through a user defined subroutine for the optimisation. The experimental data of 50CrNiMo is used to evaluate the capability of the model in prediction of TTT diagrams and bainite phase growth for pre-plastic deformations of 0, 25 and 50%.     

A neural network-based shape control system for cold rolling operations

In cold steel rolling, strip shape is crucial to product quality. For modern rolling mills, there are a number of different ways to control the strip shape, including adjusting the side depression, bending the work rolls and axial shifting the middle roll. However, these controls are not independent and hence, must be used with great care. This paper introduces a new method for strip shape control. It takes two steps: the first step is to use an Artificial Neural Network (ANN) to recognize the strip shape pattern. The second step is to apply one or a combination of several controls accordingly. This process may take several iterative steps. The new method is validated on an 8000 KN HC mill. The results demonstrated the new method could reduce the strip shape error step by step.     

A six-degree-of-freedom measurement system for the motion accuracy of linear stages

This paper presents a new six-degree-of-freedom measurement system (6DMS) which has the capability to measure simultaneously all six motion errors of a linear stage. This system employs three parallel laser beams to detect three relative linear distances of a moving body by Doppler effect so that the positioning, yaw and pitch errors can be separated. With an additional beam splitter and two quadrant photodetectors to detect the lateral shifts of the returned beams, the remaining two straightness errors and roll error can be obtained at the same time. In comparison with the HP5528A system, the accuracy of the positioning error is about 0.01 μm to the range of 10 m, the straightness error is about 1 μm within the measuring range of ± 0.1 mm, and angular errors are all about 1 arcsec within the range of ± 50 arcsecs. This system is simple in principle and can be easily equipped to any moving stages, such as linear stages, X-Y tables, CMMs, and machine tools.     

A theoretical method for analyzing the glass-crystal transformation observed by DSC in the continuous heating regime. Application to the Sb0.16As0.36Se0.48 semiconducting alloy

A theoretical procedure has been developed for deriving the evolution equation with time of the actual volume fraction transformed, for integrating the above mentioned equation under non-isothermal regime, for deducing the kinetic parameters and for analyzing the glass-crystal transformation mechanisms in solid systems involving formation and growth of nuclei. By defining an extended volume of transformed material and assuming spatially random transformed regions, a general expression of the extended volume fraction has been obtained as a function of the temperature. Considering the mutual interference of regions growing from separate nuclei (impingement effect) and from the above mentioned expression the actual volume fraction transformed has been deduced. The kinetic parameters have been obtained, assuming that the reaction rate constant is a time function through its Arrhenian temperature dependence. Besides, it has been shown that the different models, used in the literature for analyzing the glass-crystal transformation, are particular cases of the general expression deduced for the actual volume fraction transformed. The theoretical method described has been applied to the crystallization kinetics of the Sb_0.16As_0.36Se_0.48 glassy alloy, thus obtaining values for the kinetic parameters that agree satisfactorily with the calculated results by the Austin–Rickett kinetic equation, under non-isothermal regime. This fact shows the reliability of the theoretical method developed.     

Comparison of optimal motion planning algorithms for intelligent control of robotic part micro-assembly task

Two intelligent motion (or path) planning algorithms, one based on a neural network and the other based on a fuzzy coordinator, to mate a part with an assembly hole or a receptacle (target) without a jamming related to a robotic quasi-static part micro-assembly task are introduced. These algorithms are then compared by experiment results and several criteria. In the first algorithm, a neural network control strategy with a fuzzy entropy measure for avoiding jamming during the part micro-assembly is presented. An entropy function, which is a useful measure of the variability and the information in terms of uncertainty, is introduced to measure its overall performance of a task execution related to the part micro-assembly task. In the second algorithm, a learning control strategy with a fuzzy coordinator to minimize entropy and eliminate unneeded events in the plan related to avoiding jamming is described. Fuzzy set theory is introduced to address the uncertainty associated with the part micro-assembly procedure. The degree of uncertainty associated with the part micro-assembly is used as an optimality criterion, e.g. minimum fuzzy entropy, for a specific task execution. It is shown that the machine organizer using a sensor system can intelligently determine an optimal control value, based on explicit performance criteria. The algorithms utilize knowledge processing functions such as machine reasoning, planning, inferencing, learning, and decision-making. The results show the effectiveness of the proposed approaches. The proposed techniques are applicable to a wide range of robotic tasks including pick and place operations, motion planning, and part mating with various shaped parts.