Abstract: This paper mainly analyzes the current computer control technology of ceramic roller kilns, proposes a control system based on CAN fieldbus technology, and introduces its basic components, the network interface of the CAN bus, and the system software design flowchart. Keywords: CAN bus ; ceramic kiln; network node 0 Overview Ceramic roller kilns are the most important thermal equipment in the ceramic industry, mainly composed of kiln body, combustion system, flue gas system, cooling system, transmission system, and measurement and control system. Its control objective is to obtain good products that meet process requirements while minimizing overall consumption. However, due to the large inertia, pure time delay, strong nonlinearity, and coupling characteristics of multiple variables such as temperature and air pressure during product firing, effective control of the kiln is often difficult. Currently, the measurement and control systems of domestic ceramic kilns generally achieve the required temperature curve for the entire process by controlling the temperature of each combustion point. Each control point implements open-loop or closed-loop control through a PID controller, which generally achieves good results. However, there is no connection between adjacent controlled points, and the atmosphere in the kiln has no or only simple open-loop control. The fieldbus ceramic intelligent measurement and control system, while controlling each measurement and control point individually, considers the interrelationships between the points. It implements closed-loop control of the blower that affects the combustion air-oil ratio in the entire roller kiln, and also measures and controls the oxygen content in key parts of the firing zone to improve the quality of the fired product. 1 Control Method 1.1 PID Controller Individual control of each measurement and control point is generally achieved through a PID controller. It uses a microcontroller as the main body, combined with peripheral relays, proportional valves, controllers, and other detection and execution mechanisms to form a complete primary computer control system. Due to the limitations of the speed and capacity of the microcontroller, this control system can generally only use simple PID or fuzzy control algorithms to control the process according to the preset control parameters. However, this system is difficult to achieve functions such as adaptive parameter adjustment and online optimization. 1.2 CAN Bus-Based Control: Connecting each measurement and control point to the host computer via the CAN fieldbus enables more complex and advanced control and optimization, thereby adapting to product upgrades and significantly improving product quality. Predictive control further improves the temperature control accuracy of key components; by detecting the oxygen content in the kiln, ensuring that ceramic quality meets oxygen content requirements, energy saving of the control system is achieved through online air-oil ratio optimization software based on the highest combustion thermal efficiency; summarizing expert experience in improving product yield and quality, such as how to adjust the oxygen content in the kiln for atmosphere control, based on the temperature and pressure change curves of the kiln's spatial distribution, and the experience of the kiln operators, etc., and implementing this through software in the host computer, thereby improving product quality. 2. Basic Components of the Measurement and Control System The schematic diagram of the measurement and control system designed using CAN bus technology to meet the control requirements of ceramic kilns is shown in Figure 1. During system design, it was considered that the actuators on each kiln section and the corresponding thermocouples were controlled and signal acquired using a CAN bus control module, which was installed on-site to reduce potential interference during thermocouple transmission. Other thermocouples, oxygen content probes, and pressure sensors were controlled using a CAN bus A/D module; fan control was accomplished through a D/A module; flow meter data acquisition was performed through a counter module; and switch signals such as fan operating status were handled by a switch input module. [align=center] Figure 1 Schematic diagram of the measurement and control system[/align] The CAN bus network adapter card uses the PCL-841, which is compatible with IBM-PC and conforms to the CAN protocol. It is inserted into the expansion slot to complete CAN protocol communication. Analog signal acquisition and analog-to-digital and digital-to-analog conversion in each slave module are accomplished using ADAM series modules. 3 Characteristics of CAN Bus CAN is an open fieldbus. For real-time performance and cost reduction considerations, the CAN bus only uses the two most critical layers of the Open System Interconnection (OSI) 7-layer reference model, namely the physical layer and the data link layer. According to the CAN technology specification version 2.0 formulated by Philips, its communication distance can reach 10Km (5kbps) and the highest speed can reach 1Mbps (40m). The physical layer of the CAN bus mainly specifies the mechanical, electrical, functional and procedural characteristics of the communication medium. The main function of the data link layer is to package the data to be sent, that is, to add data error check bits, control information of the data link protocol, header and footer identifiers and other additional information to form a data frame, and send it out from the physical channel. After receiving the data frame, the additional information is removed to obtain the communication data. The main characteristics of the CAN bus are: (1) Real-time performance: The CAN bus data segment length is up to 8 bytes, which can meet the requirements of control commands or data acquisition for a working point in the industrial field. (2) No master: Any node on the network can actively send data to other nodes, which can easily form a multi-machine backup system. (3) Reliability: Each frame of data has cyclic redundancy check (CRC) and other verification measures, resulting in a low data error rate. Moreover, when a node on the network encounters a serious error, it can automatically shut down the bus and exit network communication, ensuring that other operations on the bus are not affected. 4 CAN Bus Network Interface and System Design 4.1 CAN Bus Network Interface The CAN bus network interface consists of Philips SJA1000 and PCA82C250 chips. The SJA1000 is an independent CAN controller that can operate in both BasicCAN and PeliCAN modes and supports the CAN2.0B protocol. It has functions such as extended interface buffer, enhanced error handling capability, and enhanced acceptance filtering, and can directly interconnect CAN buses. To enhance the driving capability, the 82C250 bus transceiver is used in conjunction with it, and its connection circuit diagram is shown in Figure 2. To improve the topology capability of the network nodes, 120Ω (R22) anti-reflection terminating resistors are required at both ends of the CAN bus, which play a very important role in matching the bus impedance. Ignoring this resistor will significantly reduce the anti-interference and reliability of digital communication, and may even prevent communication altogether. [align=center] Figure 2 CAN bus interface circuit diagram[/align] 4.2 System Software Design The control program for multiple field measurement and control devices includes nonlinear compensation for thermocouple temperature measurement, execution functions for control quantities such as pressure and flow, and communication with the host computer via CAN bus. Multiple measurement and control units complete a certain part of the system's functions. When a unit fails, its impact is only local; when the host fails, each smart card can still independently complete the measurement and control tasks. Each measurement and control device collects and processes field signals locally, and then sends them to the host computer via CAN bus for calculation, analysis, control, and storage. The host computer is also responsible for human-computer interaction and data management. The main tasks of the host computer in the roller kiln control system are: to communicate with each node via the CAN bus interface, receive data from the lower end and send instructions to each node in the CAN network; process data uploaded by nodes and calculate control quantities. The main computing task of the host computer is to accurately calculate the control quantities of the field monitoring and control points based on the temperature, oxygen content, and flow rate transmitted by the nodes, using control algorithms (such as fuzzy neural networks, expert systems, etc.) set in the program. The key to the host computer program design is the design of the communication program, which consists of three parts: the initialization program of the CAN controller, the sending program, and the receiving program. The difficulty in the design is the initialization of SJA1000. If the initialization is not correct, the system will not be able to send and receive data correctly. The initialization of SJA1000 includes the initialization of its clock divider register (CDR), acceptance code register (ACR), acceptance mask register (AMR), two bus timing registers (BTR0 and BTR1), and output control register. Its initialization flowchart is shown in Figure 3. [align=center] Figure 3 Initialization program flowchart[/align] 5 Conclusion and the author's innovation points Applying advanced fieldbus technology (CAN BUS) to the intelligent monitoring and control system of ceramic kilns greatly improves the reliability of the system, the yield of ceramic products, and the energy-saving effect is obvious. The CAN bus network node design implemented using chips such as the 89C52 microcontroller, SJA1000, and 82C250 is stable, highly scalable, and can be directly ported to other systems, providing a certain reference for researchers. References [1] Liu Lei, Gao Jun. Application of fieldbus technology in equipment monitoring system [J]. Microcomputer Information, 2006, 7-1: P73-74 [2] Yuan Shujuan, Chen Renwen. Implementation and application of CAN bus network node [M]. Journal of Jiangnan University (Natural Science Edition), 2005, 4 (3): 1671-7147. [3] Qiu Yin'an, Yang Aimin. Design and implementation of a new type of PID decision fuzzy controller. Microelectronics and Computer. 2005, Vol. 22, No. 2 [4] Liu Meijun. Design of control system for ceramic roller kiln based on CAN bus [J]. Low Voltage Electrical Appliances, 2006, 3, 1001-5531 [5] Yu Yonglai, Gu Qin, Yu Guohua. Application of centralized computer control technology in ceramic shuttle kiln [J]. China Ceramics. 2006, 6, 1001-9642 [6] Jiang Shubo, Zhang Huanchun, Implementing Embedded Communication Interfaces Using Software Thread Integration Technology. Microelectronics & Computer, 2002, Vol. 19, No. 7.