Introduction
A certain plastics company's main product is disposable PVC gloves, producing 3.6 billion gloves annually, making it one of the world's leading manufacturers of medical-grade PVC gloves. For a company striving to become a modern manufacturing enterprise, the data acquisition and automation control system of its entire production line is currently not yet fully developed. At present, the plastics plant's production data is mainly generated by a combination of multi-channel inspection instruments and manual observation and recording. A significant portion of the data remains unobserved and unrecorded, and its real-time performance and accuracy cannot meet the ever-increasing production demands. Furthermore, the plant's production operations are primarily performed manually, and some essential processes cannot yet be automated.
Boiler Rooms: There are currently five boiler rooms in the factory area, but only one is equipped with an automatic control system; the others are operated manually. The main data to be collected in the boiler rooms include: inlet and outlet oil temperatures, grate speed, thermal oil flow rate, furnace temperature, blower and induced draft fan current, furnace negative pressure, steam drum water level, water pump frequency, power supply voltage, oil circulation pump current, inlet and outlet pressure, and steam pressure.
Batching Room: Batching includes primary and secondary batching. Currently, there is no data collection for either batching stage, and production records are entirely manual. Batching control relies on workers' years of experience. Based on the actual site conditions, the quantities that need to be collected and controlled mainly include: the amount of powder and liquid materials used, the vacuum level in the vacuum tank, the speed of the vacuum pump, etc.
Dip Coating Workshop: The dip coating workshop requires the collection of a large amount of data. On-site monitoring primarily relies on multi-channel inspection instruments to monitor the temperature and speed of the hand mold conveyor belt throughout the production line. However, displaying data through these instruments has two drawbacks: firstly, the data has a significant delay, leading to inaccurate readings; secondly, if the data exceeds normal production limits, it cannot promptly trigger alarms to notify operators. Additionally, monitoring also includes the liquid level and viscosity in the dip coating tank, the temperature of the hand molds in the drying oven, and the flow rate of the heat transfer oil, among other things.
Therefore, designing a comprehensive data acquisition and automation control system is of vital importance for improving productivity, reducing production costs, improving product quality, and enhancing the competitiveness of enterprises.
Two-system design
A modern enterprise management system is mainly divided into the following levels (as shown in Figure 1).
Figure 1 Enterprise Management Levels
For this plastics factory, its construction was also carried out according to these three levels. However, based on the company's current actual needs, the current focus is on building the PCS (Process Control System) layer. The main task of the PCS layer is to control relevant parameters in the production process (temperature, pressure, flow rate, level, composition, humidity, pH value, and physical properties, etc.) to keep them constant or change according to a certain pattern, ensuring the continuous production process proceeds automatically while guaranteeing product quality and production safety. Process control uses various detection and control instruments and computer automation technologies to automatically detect and control the entire production process. Therefore, the entire process control system consists of a data acquisition and detection section and a control section.
2.1 System Overall Design
The data acquisition and monitoring system (as shown in Figure 2) should include a power equipment monitoring subsystem, a boiler and water supply and drainage monitoring subsystem, a power supply and distribution management system and a power energy management center, a flue gas recovery control room and a production line monitoring subsystem.
The power equipment monitoring subsystem includes vacuum, compressed air, and refrigeration systems;
The boiler and water supply and drainage system are located in the boiler room monitoring room, which includes boiler monitoring and water supply and drainage monitoring.
The power distribution management system is mainly used to monitor the high-voltage power supply used for electrostatic deposition.
The flue gas recovery control room system includes the temperature and flow rate of the flue gas.
The production line monitoring subsystem (four branch plants) mainly collects and monitors various data related to production, such as temperature, pressure, liquid level, flow rate, speed, etc.
Considering the dispersed nature and relatively independent functions of these devices, the data acquisition and monitoring system comprehensively utilizes modern communication, computer IT, and automatic control technologies to establish a distributed computer network monitoring and management system with centralized management and decentralized control. This distributed system architecture allows the entire data acquisition and monitoring system to achieve centralized monitoring and management of each subsystem from the monitoring center, ensuring unified coordination. Furthermore, decentralized control facilitates operation, debugging, and maintenance, thereby improving system reliability.
Figure 2 System Structure Diagram
2.2 System Hardware Design
PLC is the core of realizing field automatic control. Each PLC branch control station is mainly responsible for collecting signals from various sensors in the field. In addition to converting the signals and transmitting them to the host computer for display and control, it also needs to make corresponding automatic controls on the field control cabinets and actuators according to the control requirements. Therefore, each branch PLC not only has the role of central control, but also acts as a data relay station.
Based on the control requirements of the control system and the basic principles of hardware selection, and after a thorough investigation and understanding of the current industrial automation market, we conducted detailed demonstrations and investigations of the solutions and quotations of several well-known brands such as SIEMENS, Modicon, AB, and OMRON. We also fully considered the situation of other control equipment in the workshop. After strict screening, we finally decided to adopt the PLC control system of Siemens (as shown in Figure 3).
For a complete PCS system, which includes data acquisition and automated control of the production process, the following configuration is adopted:
Boiler Room (2 units) 300 PLC Station: 1 power supply module, 1 300 CPU, 1 16-channel digital input module, 1 16-channel digital output module, 4 8-channel analog input modules, 1 DC 24V power supply, etc.
Each factory production line (4 units) has a 300 PLC station: 1 power supply module, 1 300 CPU, 1 16-channel digital input module, 1 16-channel digital output module, 4 8-channel analog input modules, 1 DC 24V power supply, etc.
Figure 3 System PLC Distribution
2.3 System Software Design
Configure the monitoring and management system software according to the requirements of the data monitoring and management system:
It can provide a complete set of software, including system software, application software, and application programming packages, which meet the system operation functions, secondary development requirements, are easy to maintain, and conform to the development system standards.
The system platform can allocate and manage resources in real-time, multi-user, and multi-process scenarios. The system will feature event-driven sequencing and optimized architecture to handle real-time situations and urgent tasks simultaneously. Furthermore, the platform includes network management, standard network protocols (including TCP/IP, IPX/SPX, etc.), remote communication management, and complies with the requirements of future computer technology development. The software is designed in a modular fashion to facilitate program expansion and modification.
The system ensures that the controller continues to function independently even if the energy and power monitoring and management center malfunctions. In the event of a problem with the controller's power supply, relevant status data will also be transmitted to the energy and power management center. Upon restoration of power after a power outage, all affected equipment and controllers should automatically reset without requiring resetting. All application software uses the same high-level language and a unified database management system. The user interface is fully localized in Chinese, features multi-window functionality, dynamic graphics, and intuitive, user-friendly operation. The software is practical, comprehensively improving enterprise productivity and efficiency, enhancing energy management quality, and reducing costs and raw material consumption.
The system software employs a unique distributed client/server architecture, allowing users to easily access real-time process information at all levels within the enterprise, enabling interactive operations whether in the control room or conference room. This facilitates faster and more effective decision-making. The system features self-diagnostic capabilities and rapid fault diagnosis response. Subsystems can be switched between each other.
Three key process control designs
3.1 Vehicle Speed Control Scheme
The speed of the hand mold conveyor belt directly affects the drying quality of the hand molds in the oven and is a crucial aspect of the entire process. Therefore, controlling the speed is essential for product quality. The control principle of the oven section speed is shown in the diagram below. The Siemens MM440 frequency converter's setpoint is based on the overall line speed, provided by the host computer and transmitted via PROFIBUS bus to the frequency converter controlling the hand mold conveyor belt. The actual motor speed is measured by a rotary encoder and fed back to the frequency converter. The frequency converter uses a built-in PID control algorithm to control the frequency of the output circuit, ensuring control of the motor's actual speed and thus achieving the goal of controlling the oven section speed.
Figure 4 Schematic diagram of vehicle speed control principle
3.2 Counting control of PVC gloves
Due to the inherent characteristics of PVC gloves, namely their semi-transparent state, general measuring instruments cannot determine whether a hand mold contains PVC gloves. After comprehensively comparing domestic and international testing solutions for similar situations, machine vision can be used to measure the gloves.
Machine vision is a method of using computers to analyze images acquired from cameras. The analysis results are used to report information (detection results), control processes, or move objects. Compared to general image processing systems such as multimedia systems, machine vision emphasizes accuracy, speed, and reliability in industrial environments. A CCD camera converts the captured target into an image signal, which is then transmitted to a dedicated image processing system. Based on pixel distribution and information such as brightness and color, this signal is converted into a digital signal. The image system performs various operations on these signals to extract target features, such as area, length, quantity, and position. Finally, based on preset tolerances and other conditions, it outputs results such as size, angle, offset, number, pass/fail, presence/absence, etc. Machine vision is characterized by automation, objectivity, non-contact operation, and high precision. Compared to general image processing systems, machine vision emphasizes accuracy, speed, and reliability in industrial environments. Machine vision is ideally suited for measurement, inspection, and identification in mass production processes, such as: part assembly integrity, assembly dimensional accuracy, part machining accuracy, position/angle measurement, part identification, and feature/character recognition. Machine vision inspection technology boasts significant advantages such as high speed, high precision, and high automation, effectively meeting the needs of modern manufacturing and demonstrating broad application prospects in practice. The advantages of machine vision technology over traditional inspection techniques in the field of inspection are as follows:
(1) Non-contact
(2) Fast detection speed
(3), high precision
(4) It has strong real-time performance and can achieve fully automatic detection.
(5) Strong on-site anti-interference capability
In this system, because the gloves will curl on the hand mold after drying, a grayscale change will occur between the glove and hand mold images when the camera is used to take a picture. This allows the system to identify whether the hand mold contains PVC gloves. The resulting image is shown in Figure 5.
Figure 5-1 Hand mold with PVC gloves Figure 5-2 Hand mold without PVC gloves
The specific image processing flow is as follows:
Smoothing filtering process to eliminate image noise;
Rapidly locate and determine the detection edge; this step is crucial and directly affects the speed of image processing.
Grayscale transformation: The purpose is to highlight the contrast between the grayscale values of the foreign object and the background;
Image segmentation: Binarize the image based on an adaptive feature threshold;
Image identification: Edge detection is performed on the circular image. Because the grayscale range of the image is relatively small, based on the characteristics of image processing, it can be concluded that the hand mold with a larger grayscale value has PVC gloves, and the hand mold without PVC gloves has a smaller grayscale value.
After the image is processed, the signal is sent to the host computer, which can then measure the number of PVC gloves.
IV. Conclusion
This automated data acquisition, monitoring, and management system is based on computer network and database technologies. It is a highly automated centralized control system tailored to the characteristics of each subsystem in a plastics factory. It achieves comprehensive and effective monitoring and management of public facilities, ensuring equipment operates safely, efficiently, energy-savingly, and at its optimal state. This management information system also features process control, significantly improving the work efficiency of equipment and personnel.
About the Author
He Xiaohu (1986-) is a male postgraduate student whose main research interests are computer control and intelligent systems.
