acquisition is a basic and vitally important procedure to prepare and collect all needed process and metrology data for developing intelligent applications. In general, process data consists of sensing signals and manufacturing parameters. More details are introduced in Section 2.2.
Data Preprocessing
Without reliable data acquired from sensors and/or controllers, an intelligent application will not be feasible. To correctly interpret the acquired process and metrology data for deriving useful information, data preprocessing must be performed in advance.
Data preprocessing focuses on improving data quality through techniques such as de‐noising, synchronization, modification, and signal compression to enhance the transmission speed and storage efficiency as well as emphasize the major components and key features of signals. More commonly adopted techniques are introduced in Section 2.3 in three steps of data preprocessing: segmentation, cleaning, and feature extraction.
Intelligent Application Development
Based on the extracted features, final decisions and actions for the current situation can be carried out through learned functions by artificial intelligence (AI) approaches, including machine learning and/or deep learning techniques [5]. The ultimate purpose of the intelligent application is to extract useful knowledge and explanation from the AI models, so that correct decisions and actions can be made.
This chapter presents the existing techniques for data acquisition and data preprocessing in general; while the adoption of selected AI models for solving the problems in different industries, such as Thin‐Film‐Transistor Liquid‐Crystal Display (TFT‐LCD), solar cell, semiconductor, automotive, aerospace, chemical, and bottle industries, will be illustrated in Chapter 11 respectively.
2.2 Data Acquisition
Figure 2.2 illustrates the connection of an equipment and an external data acquisition system with a basic hardware architecture meeting the aforementioned requirements for implementing intelligent applications. The analog‐to‐digital converter (ADC) connects to various sensors installed on the equipment side to convert analog sensing signals into digital signals via its analog input/output (AIO) ports. The external data acquisition system is connected to the controller of the equipment through an Ethernet card for retrieving the manufacturing parameters. The corresponding metrology data by the measurement tool for training and tuning the AI models can also be acquired through the Ethernet card, which interconnects network devices based on the Transmission Control Protocol/Internet Protocol (TCP/IP) protocol. In this way, required process and metrology data are collected and sent to an industrial personal computer (PC) for further processing.
Figure 2.2 An external data acquisition system for acquiring process and metrology data from the equipment and measurement tool.
2.2.1 Process Data Acquisition
By acquiring the process data, including sensing signals and manufacturing parameters, the machining stability can then be evaluated and the tool health status can be monitored. Details are introduced as below.
2.2.1.1 Sensing Signals Acquisition
A sensor is a device that detects and measures a physical quantity from the real‐world environment and converts it into signals. Almost an infinite number of parameters can be acquired, such as light, temperature, location, displacement, movement, sound, pressure, moisture, voltage, current, and a great number of other environmental phenomena. Sensors are the key enabling devices for improving manufacturing capability and productivity.
With the emergence of the Fourth Industrial Revolution (Industry 4.0), cyber‐physical systems (CPSs), and industrial Internet of Things (IIoT), the number of sensors on equipment is increasing rapidly. Currently, various sensors can be seamlessly connected to equipment through generic platforms for integrating numerous devices, various tools, and shop floor information into a smart factory. In ubiquitous sensing, all possible sensors are connected to provide essential data for both IoT and big data in a given environment. These pervasive sensors enhance awareness of time‐varying information for physical manufacturing processes, thus they bridge the cyber and physical worlds. The output of a sensor is generally a converted digital format that is human‐readable at the sensor location or electronically transmitted to an external system to serve as the input for further processing and application. In the following, various sensing techniques and the issues of sensor selection and installation are presented.
Sensing Techniques
To measure process accuracy or production quality of a process tool, direct and indirect techniques may be applied. For direct techniques, the process accuracy or production quality can be measured in the machine by various sensors such as touch sensor, charge‐coupled device (CCD), laser detector, and ultrasonic sensors. However, direct techniques are limited in practice due to extreme environment of machine workspace (such as affected by cutting fluid and chips). Furthermore, valuable production cycle time is reduced along with the measurement of device accuracy or workpiece quality. Also, these sensors are usually very expensive and difficult to apply to the in‐line production due to the increased cycle time.
Relatively, indirect techniques for measuring process accuracy or production quality are less‐accurate. However, using the sensors to sense indirect factors (such as force, vibration, temperature, and power consumption) is more economical and feasible for achieving the purpose of on‐line and real‐time diagnosis and prognosis.
Sensor Selection and Installation
When it comes to selecting appropriate sensors for monitoring a specific machining process, the cutting force is regarded as the best indicator to describe the performance of cutting processes and determine product quality. Five commonly used sensors for capturing cutting‐force information are listed in Table 2.1.
Table 2.1 Sensor comparison.
| Physical quantity | Measuring type | Detecting principle | Typical device | Cost | Intrusive nature |
|---|---|---|---|---|---|
| Force | Direct | Deformation | Strain gauge | High | High |
| Current | Indirect | Hall effect |
