Deep reinforcement learning (RL) has attracted research interest in the manufacturing area in recent years, but real implemented applications are rarely found. This is because agents have to explore the given environments many times until they learn how to maximize the rewards for actions, which they provide to the environments. While training, random actions or exploration from agents may be disastrous in many real-world applications, and thus, people usually use computer generated simulation environments to train agents. In this paper, we present a RL experiment applied to temperature control of a chamber for ultra-precision machines. The RL agent was built in Python and PyTorch framework using a Deep Q-Network (DQN) algorithm and its action commands were sent to National Instruments (NI) hardware, which ran C codes with a sampling rate of 1 Hz. For communication between the agent and the NI data acquisition unit, a data pipeline was constructed from the subprocess module and Popen class. The agent was forced to learn temperature control while reducing the energy consumption through a reward function, which considers both temperature bounds and energy savings. Effectiveness of the RL approach to a multi-objective temperature control problem was demonstrated in this research.
This paper proposes a cycle time estimation algorithm of a CNC machine tool, using a block overlap based tool path generation algorithm. Velocity profile generation algorithm of CNC interpolator is proposed to compute the cycle time of the G-Code block. Because the CNC blends adjacent velocity profiles to reduce the cycle time and smooth the tool path, the cycle time is adjusted considering the block overlap. The in-position time of rapid traverse is compensated to improve the cycle time estimation accuracy. The simulation model was designed to estimate the cycle time of the CNC machine tool. A three-axis feed drive testbed was used to evaluate the cycle time estimation accuracy of the proposed algorithm.
This paper investigates the relationship between the preload level of a ball screw drive and the detected natural frequency of the system in an axial direction. A dynamic model to study the preload variation of the system is derived, and then a preload feature is proposed for extracting preload conditions based on the detected natural frequency of the system. A modified double-nut ball screw drive system with adjustable preload level is constructed. This is for the purpose of experimental verification. An accelerometer is attached to the ball screw nuts of the drive system to acquire vibration signals. The signals are analyzed to obtain the natural frequency of the ball screw drive system in an axial direction. By investigating the variation of the detected natural frequency, it is shown that the preload level can be diagnosed by the proposed preload feature. Both the experiment results and mathematical model show a direct correlation between the natural frequency and preload levels. Natural frequency increases when the preload level increases. This study provides a method to monitor the preload of a ball screw system which can be used as an indicator of the health status of the drive system.
Citations
Citations to this article as recorded by
Elastic contact analysis and fatigue life prediction of ball screws in automotive REPS systems Helong Wu, Xinrong Long, Xiaoyan Peng, Zheng Zhang, Minwei Wan, Wei Wu, Wangxin Jiang Mechanics Based Design of Structures and Machines.2025; : 1. CrossRef
A Novel Methodology for Incipient Ball Screw Backlash Measurement Using Capacitive Sensor Marcella Miller, Xu Han, Gregory William Vogl, Anita Penkova, Xiaodong Jia IEEE Transactions on Instrumentation and Measurement.2025; 74: 1. CrossRef
Robust feature design for early detection of ball screw preload loss Xu Han, Marcella Miller, Gregory W. Vogl, Guanyu Chen, Xiaodong Jia Manufacturing Letters.2024; 41: 1225. CrossRef
Detection of inception of preload loss and remaining life prediction for ball screw considering change in dynamics due to worktable position Pradeep Kundu, Marcella Miller, Prayag Gore, Xiaodong Jia, Jay Lee Mechanical Systems and Signal Processing.2023; 189: 110075. CrossRef
Ball Screw Health Monitoring With Inertial Sensors Vibhor Pandhare, Marcella Miller, Gregory William Vogl, Jay Lee IEEE Transactions on Industrial Informatics.2023; 19(6): 7323. CrossRef
A review of fault diagnosis, prognosis and health management for aircraft electromechanical actuators Zhengyang Yin, Niaoqing Hu, Jiageng Chen, Yi Yang, Guoji Shen IET Electric Power Applications.2022; 16(11): 1249. CrossRef
Experimental derivation of a condition monitoring test cycle for machine tool feed drives Maximilian Benker, Sebastian Junker, Johannes Ellinger, Thomas Semm, Michael F. Zaeh Production Engineering.2022; 16(1): 55. CrossRef
Investigation on the ball screws no-load drag torque in presence of lubrication through MBD simulations Antonio Carlo Bertolino, Giovanni Jacazio, Stefano Mauro, Massimo Sorli Mechanism and Machine Theory.2021; 161: 104328. CrossRef
Lumped parameters modelling of the EMAs’ ball screw drive with special consideration to ball/grooves interactions to support model-based health monitoring Antonio Carlo Bertolino, Massimo Sorli, Giovanni Jacazio, Stefano Mauro Mechanism and Machine Theory.2019; 137: 188. CrossRef
Study of ball screw system preload monitoring during operation based on the motor current and screw-nut vibration Tung Lam Nguyen, Seung-Kook Ro, Jong-Kweon Park Mechanical Systems and Signal Processing.2019; 131: 18. CrossRef
This paper introduces a system to machine micro-sized patterns effectively on surface based on micro-milling process using tools with simultaneous rotation and oscillation, oscillation micro milling. To review the effectiveness of proposed concept, we integrated a micro-spindle supported by active magnetic bearings with a precision 3-axis air bearing stage using double-wedge mechanism, and tested this oscillation milling. Two types of oscillation milling were tested, which are linear oscillation milling with a flat end mill and elliptical oscillation milling with a ball end mill with 0.3 mm of diameter. The spindle was rotating 110 krpm and workpiece was moving constant speed of 2~8 mm/sec during the oscillation milling. As the results, multiple oval shape dimples were generated in regular spacing, and the variation of elliptical motion made different shapes of patterns. The results showed that proposed oscillation milling can be successfully used for machining repeated micro-patterns.
In this paper, we report a new manufacturing method for friction reduction using micro-AAJ (abrasive air-jet) machining. AAJ machining employs compressed air to accelerate a jet of highspeed particles to mechanically machine features, including micro-channels and micro-holes, into glass, metal, or polymer substrates for use in microfluidics, MEMS (micro electromechanical systems). And we introduce the micro-AAJ machining system, which consists of a micro-AAJ nozzle and a five-axis positioning system. Various micro-AAJ nozzles can be used, depending on the required surface structure, and three-dimensional machining is possible. We machined samples under six different conditions and describe machining results obtained while using it. We also measured the coefficient of friction of micro-textured surfaces. We report the coefficient of friction of micro-textured surfaces patterned using micro-AAJ machining for engine piston ring.
In this paper, we proposed an air bearing stage with active magnetic preloads in vertical directions compensating motion errors and attenuating vibrations to improve dynamic characteristics. This preloaded design gives simpler configuration of the stage, and active control of preload can be used for compensating motion errors by feedforward method. To improve dynamic characteristics, vibration of the table is monitored by an accelerometer, and controlled by a DSP based digital controller with integrator and band pass filters for suppressing roll and pitch vibration modes. The modes were evaluated by measuring frequency response functions, and compared with compensated responses. This showed effective results for suppressing poorly damped regenerative vibration of air bearings.
Control systems in machinery equipment provide correction signals to motion units in order to reduce or cancel out the mismatches between sensor feedback signals and command or desired values. In this paper, we introduce a simulator for control characteristics of machinery equipment. The purpose of the simulator development is to provide mechanical system designers with the ability to estimate how much dynamic performance can be achieved from their design parameters and selected devices at the designing phase. The simulator has a database for commercial parts, so that the designers can choose appropriate components for servo controllers, motors, motor drives, and guide ways, etc. and then tune governing parameters such as controller gains and friction coefficients. The simulator simulates the closed-loop control system which is built and parameter-tuned by the designer and shows dynamic responses of the control system. The simulator treats the moving table as a 6 degrees-of-freedom rigid body and considers the motion guide blocks stiffness, damping and their locations as well as sensor locations. The simulator has been under development for one and a half years and has a few years to go before the public release. The primary achievements and features will be presented in this paper.
One of the traditional optical methods to monitor a tool is a CCD sensor-based vision system which captures an aspect of the tool in real time. In the case using the CCD sensor, specific lensmodules are necessary to monitor the tool with higher resolution than its pixel size, and a microprocessor is required to attain desired data from captured images. Thus theses additional devices make the entire measurement system complex. Another method is to use a pair of an optical source and a detector per measuring axis. Since the method is based on the intensity modulation, the structure of the measurement system is simper than the CCD sensor-based vision system. However, in the case measuring the three dimensional position of the tool, it is difficult to apply to micro machine-tools because there may not be space to integrate three pairs of an optical source and a detector. In this paper, in order to develop a tool-origin measurement system which is employed in micro machine-tools, the improved method to measure a tool origin in x, y and z axes is introduced. The method is based on the intensity modulation and employs one pair of an optical source radiating divergent beams and a quadrant photodiode to detect a three dimensional position of the tool. This paper presents the measurement models of the proposed tool-origin sensor. The models were verified experimentally. The verification results show that the proposed method is possible and the induced models are available for design.
This paper presents a linear air bearing stage with compensated motion errors by active control of preloads generated by magnetic actuators with combination of permanent and electro-magnets. A 1-axis linear stage motorized with a linear motor with 240 mm of travel range is built for verifying this design concept and tested its performances. The three motions of the table are controlled with four magnetic actuators driven by current amplifiers and a DSP based digital controller. Three motion errors were measured combined method with laser interferometer and two-probe method with 0.085 um of repeatability for straightness error. The measured motion errors were modeled as functions of the stage position, and compensation were carried out with feedforward control because the characteristics of the motion control with magnetic actuators are linear and independent for each degree-of-freedoms. As the results, the errors were reduced from 1.09 um to 0.11 um for the vertical motion, from 9.42 sec to 0.18 sec for the pitch motion and from 2.42 sec to 0.18 sec for roll motion.
In this paper, we propose a precise linear motion stage supported by magnetically preloaded air bearings. The eight aero static bearings with rectangular carbon porous pads were located only one side of vertical direction under the platen where four bearings are in both sides of horizontal direction as wrap-around-design, and this gives simpler configuration than which constrained by air bearings for all direction. Each of the magnetic actuators has a permanent magnet generating static magnetic flux for required preload and a coil to perturb the magnetic force resulting adjustment of air￢bearing clearance. The characteristics of porous aero static bearing are analyzed by numerical analysis, and analytic magnetic circuit model is driven for magnetic actuator to calculate preload and variation of force due to current. A l-axis linear stage motorized with a core less linear motor and a linear encoder was designed and built to verify this design concept. The load capacity, stiffness and preload force were examined and compared with analysis. With the active magnetic preloading actuators controlled with DSP board and PWM power amplifiers, the active on-line adjusting tests about the vertical, pitching and rolling motion were performed. It was shown that motion control for three DOF motions were linear and independent after calibration of the control gains.
In this paper, a new air bearing stage with magnetic preload and a linear motor has been developed for the small precision machine systems. The new air bearing stage is unique in the sense that permanent magnets attached bottom of the iron core of table are used not only for preloading air bearings in vertical direction but also for generating thrust force by current of the coil at base. The characteristics of air bearings using porous pads were analyzed with numerical method, and the magnetic circuit model was derived for linear motor for calculating required preload force and thrust force. A prototype of single axis miniature stage with size of 120(W)×120(L)×50(H) ㎣ was designed and fabricated and examined its performances, vertical stiffness, load capacity, thrust force and positioning resolution.
First
Prev
