Precision Machinery Manufacturing Systems
The primary objective of the project will be to develop green precision mechanisms and advanced metallic materials with energy-efficient properties. Energy conservation studies show that only 25% of the energy consumption is used in actual machining. With 75% of the energy wasted in non-machining motion, machine idling, friction, and heat dissipation, there is plenty of opportunity for energy conservation. As a result, two research directions are planned in this subproject, namely, (1) improving efficiency of motion with reduced non-machining motion and idling time and (2) innovative mechanisms and materials with green characteristics such as light weight, structural strength and better heat dissipation. An example of a green material is porous metal foam composites. The structural/mechanical research will focus on an energy efficient intelligent feed drive system and design of machine tools, for example, a minimal-chatter feed drive system and “thermally friendly” structure. Meanwhile, investigation into the physical principles underlying energy consumption patterns and health status (like tool breakdown) of the whole manufacturing system will be conducted in support of the other two subprojects, the “intelligent control” and “smart sensing”.
Real-time multi-core intelligent control
The main objective of the project will be to develop the controllers required to realize intelligent multi-axis motion systems by taking advantage of the multi-processor technique. It is anticipated that these controllers will utilize the ICT (information and communication technology) to process the signals received from embedded wireless sensors within the manufacturing system and will then provide the system with various intelligent functions for machining and health monitoring. In implementing such controllers, the subproject will devise effective schemes which enable the controller to maintain a real-time capability despite the need to process huge volumes of information received from all the sensors in the various motion axes simultaneously. Specifically, the research issues include source-level analysis at the upper level, software support at the middle level, and operating systems at the lower level, including synchronization and communication, computation migration, code scheduling and parallelism exploration, middleware design, library support, and real-time operating systems.
The main objective of the project will be to develop advanced embedded wireless sensors like a high resolution accelerometer, concentricity sensor for axial alignment, temperature sensor, etc., for in-line intelligent status monitoring, diagnosis, and control. One major challenge of this subproject is how to develop sensors with robust wireless communication in the harsh machining environment of high temperature and vibration. In practice, high-speed motor rotation and metal cutting will generate enormous low frequency (< 30 MHz) EMI noise, seriously interfering with wireless communication. Moreover, in the confined operating conditions of machine tools, slow signal fading due to reflection by chamber walls and fast signal fading due to serial reflection and deflection by fast-rotating parts pose other serious wireless communication issues. Furthermore, machine vibration will cause an internal sensing noise (so-called microphonic noise) in the electronic phase-locked loop, which will in turn create unwanted variation in the carrier frequency. This will give rise to a higher bit transmission error rate that will require resending of the signal; excessive increase of transmission latency due to frequent signal re-transmission will eventually violate the real-time transmission requirement of the sensor information. Thus, one key issue in embedded wireless sensors for machine tools is robust real-time wireless communication. In this subproject, both broad spectrum (including direct sequence broad spectrum and rapid carrier frequency hopping methods) and error coding techniques will be exploited to achieve real-time wireless communication in the harsh machining environment.
Intelligent Industrial Applications and Management
The theme project will be led by Fan-Tien Cheng (an Endowed Professor of National Cheng-Kung University and director of eMRC). The research plan of this project includes the following four directions: 1) Cloud Platform: Considering that enterprises may be concerned for the security issue of the public cloud, we will develop a new-generation AMC fully based on private cloud for promoting the AMC to the industry. 2) Prediction Capability: We will adopt the VM-based Baseline Predictive Maintenance (BPM) technology to develop the capabilities of fault diagnosis and predictive maintenance of the machine tool’s key components. 3) Knowledge Inference: We will create cutting tool evaluation functions, as well as develop the machine tool recommendation service and cutting tool evaluation service in the private cloud. 4) Virtual Machine Tool: We will increase the efficiency of collision detection, improve the precision of the cutting module, and provide estimation of machining time