Industrial controller fault diagnosis system

Abstract

The invention provides an industrial controller fault diagnosis system which comprises a monitoring system, an industrial control system and a redundant Ethernet link, the monitoring system comprises a data acquisition module, a data analysis module, a data display module and a data recording module, and the industrial control system comprises a controller module and an IO module. The industrial control system comprises two redundant controller modules. The monitoring system and the industrial control system are connected through a redundant Ethernet link, so that the communication reliability is improved. According to the industrial controller fault diagnosis system provided by the invention, the operation state information of the controller can be collected in real time and displayed through a picture, whether the operation information is abnormal or not is judged based on preset expert experience, if the operation information is abnormal, an alarm is given through the picture, and a state snapshot when the abnormality occurs is recorded; the speed of analyzing and positioning problem reasons is greatly improved, and the stability and reliability of a control system can be effectively improved.

Description

Technical Field

[0001] The invention relates to the field of device fault diagnosis systems, and in particular to an industrial controller fault diagnosis system. Background Art

[0002] As the scale of industrial systems becomes larger and larger, higher and higher requirements are placed on the functions and performance of controllers, which directly leads to the increasing complexity of the controller's software and hardware. While complex software and hardware systems bring rich functions and powerful performance, they also increase the probability of controller failure to a certain extent, and also bring great difficulties to troubleshooting. As the brain of the industrial control system, the industrial control system is responsible for the convergence calculation and output of the entire control system signal. Its stability and reliability are directly related to the safety of the entire industrial production process. Therefore, research on controller fault diagnosis is of great significance. At present, research on industrial fault diagnosis is mainly concentrated in the entire industrial production system, and the fault diagnosis and reliability of the controller itself mainly rely on the quality control of the research and development process. Summary of the invention

[0003] Purpose of the invention: The technical problem to be solved by the present invention is to provide an industrial controller fault diagnosis system for monitoring the controller status and quickly locating the controller problem in view of the shortcomings of the prior art.

[0004] The system includes a monitoring system and an industrial control system; the monitoring system and the industrial control system are connected via a redundant Ethernet link;

[0005] The industrial control system is used to push the controller operation status information to the monitoring system through the redundant Ethernet link;

[0006] The monitoring system analyzes, processes, displays and records the received controller operation status information, and issues an alarm for abnormal data.

[0007] The controller operation status information includes network status, network load, CPU load, memory load, backplane bus status and controller redundancy status. The network load is collected by reading the corresponding registers in the network card, the CPU and memory loads are obtained using the interface provided by the operating system, and the backplane bus status is detected by the Can Bus driver.

[0008] The industrial control system includes two redundant controller modules, an IO module and a backplane bus, and the redundant controller module and the IO module communicate through the backplane bus.

[0009] The two redundant controller modules include a master controller and a slave controller. If the master controller is abnormal, the slave controller will continue to collect and report the running status information of the controller, and report the abnormal status of the master controller to the monitoring system.

[0010] The monitoring system includes a data acquisition module, a data analysis module, a data display module and a data recording module;

[0011] The data acquisition module receives data pushed by the industrial control system based on a redundant Ethernet interface;

[0012] The data analysis module is responsible for analyzing and processing the data and identifying faults;

[0013] The data display module displays the analyzed and processed data on the computer screen and gives warning prompts for abnormal data;

[0014] The data recording module stores the data to be recorded in the disk of the local computer.

[0015] The redundant Ethernet link is a dual-ring redundant Ethernet link. The monitoring system is connected to the switch in the ring redundant link through two Category 6 shielded twisted-pair cables, and two mutually redundant controller modules in the controller system are connected to the switch in the ring redundant link through four Category 6 shielded twisted-pair cables. The ring redundant Ethernet link includes a dedicated switch and more than two ordinary switches. The dedicated switch includes a CPU and a switching core. The CPU is responsible for the ring network detection function, and the switching core is responsible for business data forwarding. During normal operation, one of the ports of the switching core is closed. The CPU has two network ports. The CPU periodically sends Ethernet messages with a unique CPU ID from one of the network ports and receives Ethernet messages from the other network port. If the message is not received or the ID in the message is inconsistent, it is determined that the network is faulty, and the switching core is immediately controlled to open another port to restore normal communication.

[0016] The network load includes unicast, multicast and broadcast; the network load consists of eight 64-bit data, each network port occupies two 64-bit data, representing the input and output traffic respectively, and the network load is divided into three states: normal working condition, debugging working condition, and abnormal working condition; the normal working condition refers to the normal operating condition after the controller is programmed and put into production, the debugging working condition refers to the engineers are configuring and programming the controller, and the abnormal working condition refers to the controller traffic obviously exceeding the normal range, and the specific range is determined by the specific controller model;

[0017] The network status consists of four bytes, which indicate the status of four network ports respectively. The highest bit of each byte indicates the current status of the network port, 0 indicates normal, 1 indicates abnormal, and the lower seven bits indicate the number of network port status switching times per minute. The rationality judgment basis of the number of network port status switching times per minute is as follows: 0 times is normal, 1 to 3 times is maintenance status, and more than 3 times is network abnormality. The data processing module processes the uploaded network port status data according to the rationality judgment basis and gives a status indication.

[0018] The CPU load supports independent monitoring of specific CPU cores and specific processes; the CPU load is determined by a variable-length data packet, the length of which depends on the number of CPU cores and the number of processes, each data packet is divided into more than two sub-data packets, each sub-data packet contains three fields: type, thread name, CPU core, and load, the packet type occupies 1 byte, the thread name occupies 10 bytes, the CPU core occupies 1 byte, and the load occupies 1 byte. The packet type is divided into CPU packets and thread packets, and the CPU packet does not contain a thread name field; during the data processing process, each field is first extracted according to the data packet format, the total CPU load and thread load of each CPU are parsed and calculated, and it is determined whether the accumulated sum of the loads of each thread is equal to the total CPU load, and then each thread load is determined one by one whether it is abnormal. Under common working conditions, the system single-core computing load is required to be less than 20%, and in some motion control systems with high data synchronization requirements, it may occupy more than 70% of the load. The specific abnormal threshold depends on the specific application scenario and can be set through the monitoring system interface.

[0019] The memory load supports independent monitoring of resource usage of a single process; the memory load is determined by a variable-length data packet, the length of which depends on the number of processes, each data packet includes a total packet header and more than two sub-packets, the packet header includes the total memory size and the total memory consumption, each sub-packet includes two fields, namely, the packet thread name and the memory load, the thread name occupies 10 bytes, and the load occupies 4 bytes; during the data processing process, each field is first extracted according to the data packet format, and the total memory load and thread load are parsed and calculated to determine whether the accumulated sum of the memory loads of each thread is equal to the total memory load, and then determine whether the memory load of each thread is abnormal one by one. Without integrating third-party components with large memory usage such as OPC UA, Profinet master station, DDS, etc., the memory occupied by conventional industrial controller applications is generally not higher than 50MB. The specific abnormal threshold depends on the specific application scenario and can be set through the monitoring system interface.

[0020] By adopting the above technical solution, the beneficial effects achieved by the present invention are:

[0021] Through redundant controllers and redundant Ethernet links, fault diagnosis information can be reliably collected and transmitted, greatly reducing the probability of diagnostic failure due to the loss of diagnostic information. The corresponding diagnostic information is selected based on the working scenario characteristics of the industrial control system, basically covering common fault points, making fault diagnosis more targeted, reducing the time of fault diagnosis, and improving the efficiency of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more clear.

[0023] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0024] Figure 2 It is a schematic diagram of the controller system structure of the present invention.

[0025] Figure 3 It is a schematic diagram of the dual-ring redundant Ethernet link structure of the present invention. DETAILED DESCRIPTION

[0026] The present invention provides an industrial controller fault diagnosis system, including a monitoring system, an industrial control system and a redundant Ethernet link.

[0027] The industrial control system is used to push the controller operation status information to the monitoring system through the redundant Ethernet link;

[0028] The industrial control system includes two redundant controller modules, a plurality of IO modules, and a backplane for connecting the IO modules and the controller modules. Figure 2 As shown, the industrial control system module periodically collects statistics on various status information of the entire industrial control system and pushes this information to the monitoring system. The monitoring system analyzes, processes, displays and records the received information and issues alarms for some abnormal data.

[0029] The monitoring system includes a data acquisition module, a data analysis module, a data display module, and a data recording module. The data acquisition module receives data pushed by the industrial control system based on a redundant Ethernet interface. The data analysis module analyzes and processes the data and identifies faults based on preset expert experience. The data display module displays the analyzed and processed data on a computer screen and gives an alarm prompt for abnormal data. The data recording module stores the data to be recorded in the local computer's disk for further analysis.

[0030] Preferably, the industrial control system and the monitoring system are connected via a dual-ring redundant Ethernet link, such as Figure 3As shown, the switch used in the ring redundant Ethernet link is an industrial switch, which supports automatic identification of ring link failures and network reconstruction, effectively improving the reliability of the fault diagnosis system at the communication level.

[0031] Preferably, the industrial control system adopts a pair of redundant controllers. Once the main controller is abnormal, the slave controller will continue to collect and report the status information of the controller system, and at the same time report the abnormal status of the main controller to the monitoring system, effectively improving the reliability of the fault diagnosis system at the diagnostic information collection level.

[0032] like Figure 1 As shown, in a specific embodiment of the present invention, an industrial controller fault diagnosis system is provided, which includes a monitoring system, an industrial control system, and a redundant Ethernet link. The monitoring system and the industrial control system are connected via a redundant Ethernet link.

[0033] (1) The industrial control system includes a controller module, an IO module, and a backplane bus. The backplane bus uses two CAN Buses to form a redundant backplane bus. The IO module and the controller module are connected to the backplane bus through two CAN Buses. A 120-ohm terminal resistor is connected to the tail of the backplane bus.

[0034] (2) The controller module includes two physically redundant Ethernet interfaces. When uploading controller status information, the two network interfaces report simultaneously.

[0035] (3) The controller status information includes network load, CPU load, memory load, backplane bus status, and controller redundancy status. The network load is collected by reading the corresponding registers in the network card. The CPU and memory loads are obtained using the interface provided by the operating system. The backplane bus status is detected by the Can Bus driver.

[0036] (4) The redundant Ethernet link is composed of two independent ring Ethernet links, each of which includes four switches and a certain number of Category 6 shielded twisted pair cables. Switch ports are selected from ring network A to connect to the two redundant controllers and the A network of the monitoring system respectively, and switch ports are selected from ring network B to connect to the two redundant controllers and the B network of the monitoring system respectively. Figure 1As shown. The ring redundant Ethernet link includes a dedicated switch and more than two ordinary switches. The dedicated switch contains a CPU and a switching core. The CPU is responsible for the ring network detection function, and the switching core is responsible for business data forwarding. During normal operation, one of the ports of the switching core is closed. The CPU has two network ports. The CPU periodically sends Ethernet messages with the CPU's unique ID from one of the network ports and receives the message from the other network port. If the message is not received or the ID in the message is inconsistent, it is considered that the network is faulty, and the switching core is immediately controlled to open another port to restore normal communication, as shown in FIG. Figure 3 shown.

[0037] (5) The monitoring system is implemented based on an industrial PC, and its hardware includes a computer host and a display. The core software modules include a data recording module, a data display module, a data analysis module, and a data acquisition module. The data acquisition module and the controller module communicate using the UDP protocol, and the data acquisition module further transmits the collected data to the data analysis module.

[0038] (6) The data analysis module analyzes and processes the collected data according to the data processing strategy loaded when the monitoring system is started. The specific data analysis and processing process is as follows:

[0039] The network load consists of 8 64-bit data, each network port occupies two 64-bit data, representing the input and output traffic respectively. The network load is divided into three states: normal working condition, debugging working condition, and abnormal working condition. The normal working condition refers to the normal operating condition after the controller programming is completed and put into production. The debugging working condition refers to the engineer configuring and programming the controller. The abnormal working condition refers to the controller flow obviously exceeding the normal range. The specific range is determined by the specific controller model. The network state consists of 4 bytes, and the four bytes indicate the status of the four network ports respectively. The highest bit of each byte indicates the current status of the network port, 0 means normal, 1 means abnormal, and the lower seven bits indicate the number of network port status switches per minute. The rationality of the number of network port status switches per minute is determined as follows: 0 times is normal, 1 to 3 times is maintenance status, and more than 3 times is network abnormality. The data processing module processes the uploaded network port status data according to the above processing strategy and gives a status indication.

[0040] The CPU load is determined by a variable-length data packet. The length of the data packet depends on the number of CPU cores and processes. Each data packet is divided into multiple sub-data packets. Each sub-data packet contains three fields: type, thread name, CPU core, and load. The packet type occupies 1 byte, the thread name occupies 10 bytes, the CPU core occupies 1 byte, and the load occupies 1 byte. The packet type is divided into CPU packets and thread packets. The CPU packet does not contain the thread name field. During the data processing process, each field is first extracted according to the data packet format, and the total CPU load and thread load of each CPU are parsed and calculated to determine whether the accumulated sum of each thread load is equal to the total CPU load. Then, each thread load is judged one by one to see if it is abnormal. The specific abnormal threshold depends on the specific application scenario and can be set through the monitoring system interface.

[0041] The memory load is determined by a variable-length data packet. The length of the data packet depends on the number of processes. Each data packet contains a total packet header and multiple sub-packets. The packet header contains the total memory size and total memory consumption. Each sub-packet contains two fields: the packet thread name and the memory load. The thread name occupies 10 bytes and the load occupies 4 bytes. During the data processing process, each field is first extracted according to the data packet format, and the total memory load and thread load are parsed and calculated to determine whether the sum of the accumulated memory loads of each thread is equal to the total memory load. Then, the memory load of each thread is judged one by one to determine whether it is abnormal. The specific abnormal threshold depends on the specific application scenario and can be set through the monitoring system interface.

[0042] (7) The data display module displays the data processed by the data analysis module on a computer display for users to monitor and analyze.

[0043] (8) The data recording module records necessary data, and the recording period can be set. When a fault occurs, it can quickly record a status snapshot of the industrial control system for subsequent further analysis.

[0044] The present invention provides an industrial controller fault diagnosis system. There are many methods and ways to implement the technical solution. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the protection scope of the present invention. All components not specified in this embodiment can be implemented by existing technologies.

Claims

1. An industrial controller fault diagnosis system, characterized in that: It includes a monitoring system and an industrial control system; the monitoring system and the industrial control system are connected via a redundant Ethernet link; The industrial control system is used to push the controller operation status information to the monitoring system through the redundant Ethernet link; The monitoring system analyzes, processes, displays and records the received controller operation status information, and issues alarm prompts for abnormal data.

2. The system according to claim 1, characterized in that The controller operation status information includes network status, network load, CPU load, memory load, backplane bus status and controller redundancy status. The network load is collected by reading the corresponding registers in the network card, the CPU and memory loads are obtained using the interface provided by the operating system, and the backplane bus status is detected by the Can Bus driver.

3. The system according to claim 2, characterized in that The industrial control system includes two redundant controller modules, an IO module and a backplane bus, and the redundant controller module and the IO module communicate through the backplane bus.

4. The system according to claim 3, characterized in that The two redundant controller modules include a master controller and a slave controller. If the master controller is abnormal, the slave controller will continue to collect and report the running status information of the controller, and report the abnormal status of the master controller to the monitoring system.

5. The system according to claim 4, characterized in that The monitoring system includes a data acquisition module, a data analysis module, a data display module and a data recording module; The data acquisition module receives data pushed by the industrial control system based on a redundant Ethernet interface; The data analysis module is responsible for analyzing and processing the data and identifying faults; The data display module displays the analyzed and processed data on the computer screen and gives warning prompts for abnormal data; The data recording module stores the data to be recorded in the disk of the local computer.

6. The system according to claim 5, characterized in that The redundant Ethernet link is a dual-ring redundant Ethernet link. The monitoring system is connected to the switch in the ring redundant link through two Category 6 shielded twisted-pair cables, and two mutually redundant controller modules in the controller system are connected to the switch in the ring redundant link through four Category 6 shielded twisted-pair cables. The ring redundant Ethernet link includes a dedicated switch and more than two ordinary switches. The dedicated switch includes a CPU and a switching core. The CPU is responsible for the ring network detection function, and the switching core is responsible for business data forwarding. During normal operation, one of the ports of the switching core is closed. The CPU has two network ports. The CPU periodically sends Ethernet messages with a unique CPU ID from one of the network ports and receives Ethernet messages from the other network port. If the message is not received or the ID in the message is inconsistent, it is determined that the network is faulty, and the switching core is immediately controlled to open another port to restore normal communication.

7. The system according to claim 2, characterized in that The network load includes unicast, multicast and broadcast; the network load consists of eight 64-bit data, each network port occupies two 64-bit data, representing input and output traffic respectively, and the network load is divided into three states: normal working condition, debugging working condition and abnormal working condition; The normal operating condition refers to the normal operating condition after the controller is programmed and put into production. The debugging condition refers to the condition when engineers are configuring and programming the controller. The abnormal condition refers to the condition when the controller flow rate obviously exceeds the normal range. The specific range is determined by the specific controller model.

8. The system according to claim 2, characterized in that The network status consists of four bytes, and the four bytes indicate the status of four network ports respectively. The highest bit of each byte indicates the current status of the network port, 0 indicates normal, 1 indicates abnormal, and the lower seven bits indicate the number of network port status switching times per minute. The rationality of the number of network port status switching times per minute is determined based on the following: 0 times is normal, 1 to 3 times is maintenance status, and more than 3 times is network abnormality; the data processing module processes the uploaded network port status data based on the rationality determination basis and gives a status indication.

9. The system according to claim 2, characterized in that The CPU load supports independent monitoring of specific CPU cores and specific processes; the CPU load is determined by a variable-length data packet, the length of which depends on the number of CPU cores and the number of processes, each data packet is divided into more than two sub-data packets, each sub-data packet contains three fields: type, thread name, CPU core, and load, the packet type occupies 1 byte, the thread name occupies 10 bytes, the CPU core occupies 1 byte, and the load occupies 1 byte. The packet type is divided into CPU packets and thread packets, and the CPU packet does not contain a thread name field; in the data processing process, each field is first extracted according to the data packet format, the total CPU load and thread load are parsed and calculated, and it is determined whether the accumulated sum of the thread loads is equal to the total CPU load, and then each thread load is determined one by one whether it is abnormal.

10. The system according to claim 2, characterized in that The memory load supports independent monitoring of resource usage of a single process; the memory load is determined by a variable-length data packet, the length of which depends on the number of processes, each data packet contains a total packet header and more than two sub-packets, the packet header contains the total memory size and total memory consumption, and each sub-packet contains two fields, the thread name and the memory load, the thread name occupies 10 bytes and the load occupies 4 bytes; in the data processing process, each field is first extracted according to the data packet format, and each memory The total load and thread load are calculated to determine whether the sum of the memory loads of each thread is equal to the total memory load. Then determine whether the memory load of each thread is abnormal.

Images