There is a term that appeared around the same time as the 4th Industrial Revolution. Industry 4.0, which originated in Germany, refers to a flexible and efficient production system based on a cyber-physical system (CPS).


Industry 4.0, defined by the German government to revitalize German manufacturing, pursues mass production of a wide variety of products, unlike the heavy and rigid Industry 3.0, through modular processes, variable and flexible facilities, autonomous and distributed control, wireless communication, and real-time location tracking.

On the 10th, during the 50th Korea Electronics Show, the '4th Industrial Revolution Advanced Sensor Technology Seminar 2019' hosted by the Korea Semiconductor Industry Association and My Forum was held as a side event.

On this day, Dae-seong Lee, head of the Smart Sensor Research Center at the Electronics and Telecommunications Research Institute (KETI), gave a seminar on the topic of ‘Smart Sensor Technology Applicable to Smart Factories.’


Classification of sensors used in smart factories
The sensors mounted on each piece of equipment in smart factories are evolving from board-type sensors to micro-electro-mechanical systems (MEMS) that integrate sensor SoCs with sensing elements and intelligent algorithms to achieve Industry 4.0.

Pressure sensor, acceleration sensor, 9-axis motion sensor, microphone, infrared imager, size not exceeding mm Sensors, magnetic sensors, etc. are now commercialized and are actively used in various fields such as smart factories and smartphones.

The areas where sensors are utilized in smart factories can be broadly divided into three categories. First, ▲equipment control, which monitors equipment in real time and remotely controls equipment. It also reports and responds to delays in manufacturing schedules and abnormal signs.

Next is ▲Energy and environmental efficiency, which helps managers conduct remote metering by identifying energy consumption, environmental pollution levels, etc. It can evaluate the impact of the factory environment on the manufacturing process, analyze the cause of environmental abnormalities, and also aggregate energy usage.

Lastly, ▲logistics efficiency. It automatically transmits inventory quantity information collected based on robots or RFID, tracks inventory progress by process, analyzes appropriate inventory quantity, and provides optimal movement route guidance based on 3D digital maps.

Factory automation products are diverse. There are products that include sensors such as proximity sensors, capacitive sensors, inclination and acceleration sensors, ultrasonic sensors, photo sensors, automatic door sensors, vision sensors, and safety sensors, as well as rotary encoders, RFID and barcode image solutions, and WCS/PCV position control systems.


Optical sensing technology that 3D images objects
Optical sensing technology, which is widely used for object recognition in smart factories, can be divided into active and passive methods.

The active method detects objects by shooting light and the like and calculating the time difference, angle difference, etc. for the return. The passive method does not shoot light, but uses two or more sensors to detect objects using trigonometry, etc.

Representative 3D image sensors include stereo cameras, ToF cameras, and structured light cameras. As mentioned above, stereo cameras use two cameras to detect objects and use triangulation to create 3D images.

ToF cameras calculate the distance data by calculating the time it takes for the shot light to return and convert it into a 3D image. Structured light cameras utilize a camera and a projector. Search for matching points of structured light codes in camera images and projector images and obtain the 3D coordinates of those points.

In addition, LiDAR, which emits a pulse laser to a target and measures the time and intensity it takes for the light to return to detect distance, direction, speed, temperature, material distribution, and concentration characteristics, has been used in autonomous vehicles.

However, it is also being used for collision avoidance in unmanned transport robots within smart factories, and as prices drop, accuracy is improving from 8 channels to 16 channels and 32 channels.


Black holes in the electromagnetic spectrum, cameras utilizing THz waves
Non-destructive testing is a method of testing without modifying the product's integrity or surface condition, and recently, there has been a move to utilize terahertz (THz) waves. THz waves are an electromagnetic spectrum between microwaves and infrared waves, with a band of 0.1 to 10 Hz and a wavelength of 3 mm (millimeters) to 30 μm (micrometers). Director Lee Dae-seong explained that it is a spectrum like a black hole that has been less studied than other electromagnetic spectra.

However, cameras capable of imaging these THz waves have recently been released. Companies such as NEC in Japan, Agiltron in the US, and LETI in France have released products or are developing prototypes. In Korea, Somo Energy & Technology is developing an antenna-coupled nano-bolometer camera.

THz cameras have proven useful in detecting foreign substances in food. In 2014, the Korea Food Research Institute and the Korea Electrotechnology Research Institute demonstrated that they are more effective than photography and X-rays in detecting foreign insects in flour powder and in testing the quality of red ginseng.


Sensors used in facility diagnostics
Sensors play a major role in the prognostics and health management (PHM) function that is linked to the lifespan of a smart factory. Various IoT sensors mounted on the equipment detect changes in equipment vibration, speed, motor current, temperature, and pressure.

The establishment of a smart factory's equipment diagnosis function is divided into three stages. The first stage is equipment monitoring, which visualizes equipment data. The second stage is equipment diagnosis monitoring, which visualizes the processed and analyzed equipment data. The third stage is equipment predictive diagnosis monitoring, which collects and integrates data, processes and analyzes it, and learns and visualizes it through machine learning.

The equipment diagnostic solution from SnK in Germany records normal sensor strengths over time during the product manufacturing process. It then sets an acceptable range. When the product is repeatedly produced, any deviation from this range is detected as an equipment abnormality. Each facility has a detection function, and data from each facility is linked to the Manufacturing Execution System (MES) to comprehensively manage the status of all facilities within the factory.

How will the sensors be utilized when applying the above equipment diagnostic solution to the forging field? Pressure, shock, and temperature sensors are installed in the press contact part of the press machine, the rotating shaft, and other physical quantity detection parts. The sensors can collect data such as pressure, shock vibration and sound, temperature change, and torque, and transmit them wirelessly to the MES.

Center Director Lee Dae-seong concluded the seminar by saying, “Sensors entering smart factories are becoming smart and IoT-ized,” and “In order to create intelligent sensors that communicate wirelessly, we will need to actively research semiconductor and MEMS technologies.”