Abstract: This paper studies the key architecture and design of an intelligent flow measurement system based on CAN and virtual instrument (VI). The CAN-based communication mode ensures the high real-time performance and anti-jamming capability of the system, while the VI-based design significantly improves the system's data acquisition capability, openness, modularity, and intelligent characteristics. Keywords : CAN, VI, Flux measurement system 1. Introduction After years of development, the technological development trends of flow measurement systems can be summarized as follows: 1. Improving instrument intelligence. Instrument intelligence mainly refers to functional intelligence. Especially with the emergence and application of new microprocessors in recent years, instruments can fully utilize the functionality and flexibility of microcomputers by controlling and managing the entire measurement process through software. 2. Improved data acquisition capabilities and measurement performance. Measurement accuracy and stability are crucial technical indicators for general flow meters and are closely related to the data acquisition technology employed. By reducing interference and improving detection accuracy and instrument stability, the capabilities of the data acquisition card are greatly enhanced. 3. Enhanced CPU processing power and improved data processing functions. With advancements in microelectronics and computer technology, using high-performance integrated chips and microprocessors to improve signal amplification and processing accuracy, broaden the instrument's detection range, compensate for detection errors, and perform zero-point calibration has become one of the development directions for instruments today. 4. Increased system openness. Improving system openness includes enhancing the openness of hardware circuits, software structure, communication interfaces, and human-machine interfaces. 5. The trend towards system modularization. The data acquisition and signal processing sections of flow detection, including data sampling, filtering, amplification, and other functional modules, are modularized. In actual system assembly, only the relevant modules need to be selected and assembled according to the required functions. 2. Research on Intelligent Flow Detection System Based on CAN and Virtual Instrument (VI) 2.1 Introduction to Virtual Instrument (VI) A virtual instrument (VI) is a system that combines existing mainstream computer technologies with innovative, flexible, and easy-to-use software and high-performance modular hardware to establish a powerful and flexible computer-based test, measurement, and control system. Virtual instruments are a product of the deep integration of modern computer technology and instrument technology. They are an important technology in the field of computer-aided testing (CAT) and represent an effective combination of computer hardware resources, instrument and measurement control system hardware resources, and virtual instrument software resources. Virtual instruments (VI) are a new and dynamic type of instrument formed by the application of computer technology in the field of instrumentation. Virtual instruments provide users with a reusable source code library to build their own instruments, handling functions such as inter-module communication, timing, and triggering. It emphasizes transforming traditional manufacturer-defined instruments into user-defined, dedicated instruments composed of computer software and several modules, based on a general-purpose computer platform and through software and software panels. The emergence of virtual instruments has completely broken the traditional model where instruments are defined by manufacturers and cannot be changed by users. Because virtual instrument application software integrates all the functions of an instrument, such as acquisition, control, data analysis, result output, and user interface, it replaces some hardware and even the entire instrument with computer software. Therefore, in a sense, it can be said that "software is the instrument." Today, with the development of electrical measurement and network technologies, virtual instrument technology will inevitably become the main direction of instrumentation technology development. 2.2 Introduction to CAN Fieldbus CAN is short for Controller Area Network, a protocol first proposed by the German company BOSCH for data communication between internal automotive actuators. The CAN protocol is based on the Open Systems Interconnection model of the International Organization for Standardization, but its model structure has only three layers: physical layer, data link layer, and application layer. CAN data transmission uses a short frame format, with each frame containing 8 bytes of data, and a communication rate of up to 1 Mbit/s. CAN data transmission takes extremely short time, thus greatly reducing the probability of data interference on the bus; when a serious data transmission error occurs on a node, it also has the function of automatically shutting down the faulty node, thus possessing strong anti-interference capabilities. CAN supports a "multi-master" working mode, where any node on the CAN network can actively send information to other nodes at any time, facilitating "point-to-point," "one-to-many," and "global broadcast" communication. Because CAN uses non-destructive bus arbitration technology, when multiple nodes send data to the bus, lower-priority nodes will actively withdraw from transmission, while higher-priority nodes can continue transmitting data unaffected. 2.3 Intelligent Flow Detection Platform Based on CAN and Virtual Instruments 2.3.1 Requirements Analysis With the development of the national economy, valves are playing an increasingly important role in the industrial field, and their quality and performance requirements are becoming increasingly stringent. The flow coefficient of a valve is an indicator of its flow capacity; a higher flow coefficient value indicates less pressure loss when fluid flows through the valve. Valve manufacturers in industrialized countries often include flow coefficient values ​​for valves of different pressure ratings, types, and nominal diameters in their product catalogs for design departments and users to select from. The flow coefficient value varies with the size, type, and structure of the valve; different types and specifications of valves must be tested separately to determine the appropriate flow coefficient value. Currently, there are very few testing devices in my country capable of measuring valve flow coefficients, far from meeting actual needs. Furthermore, most of these devices rely on outdated methods such as traditional instrumentation and manual measurement, resulting in low accuracy. Therefore, this system employs advanced virtual instrument technology and industrial-grade real-time CAN communication to improve this situation. 2.3.2 CAN-Based Communication Scheme Design The detection terminal designed in this system, based on an ARM processor, is a digital integrated controller based on the CAN fieldbus. A typical integrated detection platform based on CAN and ARM intelligent detection terminals is shown in Figure 1. It is equipped with multiple CAN intelligent detection terminal nodes, display instruments, and an upper-level computer. The ARM9 intelligent detection terminal, as a control node at the CAN bus front end (slave node 1 in Figure 1), primarily undertakes the task of controlling field devices. Display instruments are optional and used to monitor system operation results. The upper-level computer system (master node in Figure 1) mainly consists of a master controller and a CAN bus communication interface adapter card connected to the master controller, responsible for managing the entire system, sending control commands, and transmitting data. The main control unit is a PIII or P4 level industrial computer, and a CAN bus communication interface adapter card with a PCI bus interface is also selected. [align=center] Figure 1. Integrated detection platform diagram based on CAN bus[/align] 2.3.3 Application of virtual instrument technology The LabVIEW data interface card is selected, and the working principle is shown in the figure: [align=center] Figure 2. Working principle diagram of LabVIEW-based data acquisition card[/align] A detection system consists of sensors, signal conditioning circuits, microprocessors/computers for data acquisition and processing, etc. When designing the detection system, the relationship between the system performance design indicators and the performance indicators of each component should be considered, and error allocation should be reasonably carried out to obtain the best static and dynamic performance indicators of the system with the least planning cost and the simplest implementation scheme. The system is used to dynamically test the valve flow coefficient, so the key is to accurately and quickly extract the physical quantities such as pressure, differential pressure, flow rate, and temperature that change dynamically over time in the test system, and the computer is the best tool for extracting and processing this information. Therefore, a computer-based virtual instrument testing system was constructed to obtain static and dynamic values ​​of parameters such as pressure, differential pressure, flow rate, and temperature with high accuracy, high sensitivity, and high efficiency. It can also filter and process the collected data to minimize random errors, systematic errors, and external interference, and perform functions such as automatic zero-point calibration, automatic display, data output, and test report printing. Constructing a flow detection system based on virtual instruments first requires a testing pipeline system, which is the foundation for the test. The pipeline design must meet the required pressure and flow rate of the test and maintain stability, while also considering its protective function for the sensors to extend the overall system lifespan. Secondly, sensors are needed to convert the physical signals to be measured into measurable electrical signals. The physical quantities to be measured by the system are pressure, differential pressure, flow rate, and temperature. Considering measurement accuracy and cost, ceramic pressure sensors, capacitive differential pressure sensors, turbine flow sensors, and PT100 resistance thermometers were selected to measure pressure, differential pressure, flow rate, and temperature, respectively. Because the turbine flow sensor outputs an irregular frequency signal, which is difficult for a computer to process, a secondary instrument was added after the flow sensor to convert the frequency into a standard current signal. Furthermore, building a flow detection system based on virtual instruments requires more than just a sensor; the sensor signal must undergo photoelectric isolation, amplification, and A/D conversion before being processed by the computer. Considering these factors, a data acquisition card with amplification, A/D conversion, and photoelectric isolation functions was added to the system. Additionally, auxiliary components such as a power supply, junction box, and signal cables are needed, with the power supply providing power to the sensor and power amplifier. Finally, since it is a virtual instrument, a computer is essential; considering the stability of the entire testing system, an industrial control computer from Huabei Industrial Control was selected. In summary, the hardware design of the valve flow detection system based on virtual instruments can be divided into: detection pipeline design, sensor location planning, and the design of the computer testing subsystem. The following diagram illustrates the design of the detection pipeline: [align=center]Figure 3, Detection Pipeline Design[/align] In the diagram, 1 is the controllable water source, 2 is the check valve, 3 is the flow regulating valve, 4 is the filter, 5 is the temperature sensor, 6 is the flow sensor, 7 is the flow digital accumulator, 8 is the pressure sensor, 9 is the differential pressure transmitter, and 10 is the valve being measured. The controllable water source of the system consists of a high-power water pump; the check valve and flow regulating valve mainly prevent backflow and maintain the stability of the system flow rate; the filter's function is to filter impurities in the water and protect the turbine of the turbine flow sensor from damage. The overall hardware platform design block diagram of the detection system is shown in Figure 4. [align=center]Figure 4, Hardware Platform of the Detection System[/align] The various parameters being measured (pressure, differential pressure, flow rate, temperature) are converted into electrical signals that are easy to process by the sensors. If the sensor output signal is too weak or the signal quality is too poor, it should be amplified and filtered by the front-end preprocessing circuit. The data is then converted into digital values ​​by the data acquisition subsystem and fed into the microcomputer system via the data bus. The computer processes the data and verifies whether it meets the test requirements. If the conditions are not met (e.g., test temperature is too low, flow rate is unstable, pressure is too high), an alarm device is triggered, and an adjustment or test prohibition command is issued to protect the test equipment from damage. If the test conditions are met, the computer begins data calculation and displays the results, records them in the database, and plots them as curves and graphs. Then, the I/O subsystem completes the human-computer interaction for basic parameters such as valve specifications, model, and applicable conditions, and finally prints out the test report. 3. Application Case The success of a test system is determined by verifying the validity of the test results. Since the system was built at the university's Valve Flow Detection Research Center, it has undergone multiple tests. The tests show that the interface is easy to operate, and all functions meet the predetermined design requirements and can satisfy the requirements of experimental research. Taking the SZ45X series valves as an example, the measured data was tested under turbulent flow conditions in the pipeline, and the Reynolds coefficient of the fluid met the requirement of being between 4 x 10⁵ and 1 x 10⁶. Comparing the system test data and the instrument display data, the absolute value of the relative error is less than 1.4%, meeting the requirement of ±2% for the effective test error range specified in JB/T 5296-91. This proves that the system design meets the requirements and is in a leading position domestically. Its computer-acquired, processed, and displayed data meets the system accuracy requirements and can be used for testing valve flow coefficients. Simultaneously, the system can conveniently provide various valve parameters and test results, including: test date, basic valve information, original test data and data curves, valve flow coefficient test results, etc. 4. Conclusion and Outlook Based on expert evaluation and analysis of actual test results, the intelligent flow detection system based on CAN and virtual instruments can be used to determine the flow resistance coefficient of gate valves, globe valves, throttle valves, ball valves, butterfly valves, diaphragm valves, plug valves, check valves, foot valves, and pressure reducing valves, as well as the flow coefficient of gate valves, throttle valves, ball valves, and butterfly valves, thereby achieving the purpose of flow measurement. Its CAN bus-based communication method ensures the real-time nature of the measured parameters and the system's high anti-interference characteristics. The use of virtual instrument technology makes it easier to add detection system functions in software, giving the system openness, intelligence and modularity, which is in line with the current development trend of flow detection systems. Its key technologies can be used in most industrial control systems, including real-time detection, monitoring and control and process control. Innovations: 1. Based on the combination of CAN and virtual instrument technology for flow detection system 2. Provides an advanced solution that can integrate real-time measurement and control and process measurement and control References: [1] David J. Kland. Nioll M. Adams Pattern Detection and Discovery[M]. Springer, 2001 [2] Li Shiping. PC Computer Detection Technology and Application[M]. Xi'an University of Electronic Science and Technology Press, 2003 [3] Sang Qiang, Zhang Hongjian. Design and application of computer control system based on data acquisition card and output card[J]. Mechanical and Electrical Engineering, 2003 (4) [4] US National Instruments Inc. Measurement and automation catalogue [M/CD]. USA, 2003 [5] Liang Xingyan, He Weixing. LabVIEW realizes remote data acquisition and transmission. Microcomputer Information, 2004.9, 44-45