Modbus multi-axis real-time synchronous control method for numerical control machine tool

Abstract

The application discloses a Modbus multi-axis real-time synchronous control method for a numerical control machine tool, and relates to the technical field of numerical control, which realizes microsecond-level synchronous control of multiple axes through the establishment of a global time reference, dynamic time slot allocation, synchronous execution of instruction pre-caching, feedforward-feedback dual-mode fault tolerance and other innovative mechanisms.The application breaks through the technical limitations of the traditional Modbus protocol, greatly reduces the cost while ensuring high performance, and the synchronous precision reaches ±0.8us, the roundness machining error is less than or equal to 9.5um, and the communication reliability reaches 99.998%.The application is particularly suitable for high-precision equipment such as five-axis linkage numerical control machine tools and high-speed machining centers.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of numerical control technology, in particular to a Modbus multi-axis real-time synchronous control method for a numerical control machine tool. BACKGROUND

[0002] Multi-axis synchronous control technology of numerical control machine tools is one of the core technologies in the field of modern precision manufacturing. It precisely coordinates the space-time relationship of multiple motion axes during processing to ensure that each motion axis moves accurately and synchronously according to the predetermined trajectory. In multi-axis numerical control machining, especially in high-end application scenarios such as five-axis linkage machining, high-speed finishing, and complex surface machining, the synchronization accuracy between the motion axes directly determines the contour accuracy, surface quality, machining efficiency, and product qualification rate of the machined parts.

[0003] The current multi-axis synchronous control technology route mainly used in the field of numerical control machine tools includes:

[0004] Special industrial bus technology: such as EtherCAT, PROFINET IRT, Powerlink, etc. These technologies can achieve microsecond-level or even nanosecond-level synchronization accuracy through hardware-level time synchronization mechanisms. However, these solutions have obvious cost problems. The single-axis cost of servo drives using special bus technology usually increases by 2000-5000 yuan. For multi-axis systems, the total cost increases significantly, which makes it difficult for small and medium-sized machine tool manufacturers to bear.

[0005] Traditional field bus technology: such as Modbus, Profibus, DeviceNet, etc. These technologies are based on traditional master-slave polling communication mechanisms. Although the cost is relatively low, there are inherent technical defects in multi-axis synchronous control, and the specific defects are as follows:

[0006] 1. Polling mechanism leads to time difference in command issuance: The traditional Modbus protocol-based communication method uses a master-slave polling mechanism, which leads to a significant time difference in the issuance of commands for each axis. In a typical polling cycle, the command interval between the first axis and the Nth axis reaches (N-1) x T_query, where T_query is the query response time of a single axis. This time difference directly leads to asynchronous motion of multiple axes;

[0007] 2. Communication jitter is uncontrollable: The communication delay based on a general operating system is uncontrollable, and the communication jitter usually reaches ±500 μs. This uncertainty seriously affects the synchronization accuracy;

[0008] 3. No synchronization execution mechanism: In the traditional scheme, the command is executed immediately after being received, and there is no unified synchronization trigger mechanism, so the space-time alignment of multi-axis motion cannot be achieved;

[0009] 4. Lack of effective fault-tolerant mechanism: in the case of communication anomalies, the system often needs to completely stop the processing process, affecting the continuity of production.

[0010] To solve the above technical problems, the existing improvement scheme is insufficient in that: the existing technical improvement scheme mainly focuses on communication path optimization, such as improving communication speed, optimizing network topology, etc., but fails to fundamentally solve the core contradiction of "instruction execution time inconsistency" in multi-axis synchronous control, especially lacking of "instruction pre-caching and synchronous triggering" innovative mechanism. The existing scheme is difficult to achieve a good balance between cost and performance, either the cost is too high to popularize, or the performance is insufficient to meet the high-precision machining demand.

[0011] Technical problem root analysis: traditional schemes mainly try to solve the synchronization problem by optimizing the communication path, but do not touch the essential contradiction of "instruction execution time inconsistency". Fundamentally speaking, the core problem of multi-axis synchronization is the uncertainty of instruction transmission time and execution time, and the lack of unified execution reference among axes. This essential contradiction leads to inherent limitations in multi-axis high-precision synchronous control, and innovative technical solutions are needed to break through this bottleneck. SUMMARY

[0012] To solve the above technical problems, the present application provides a Modbus multi-axis real-time synchronous control method for numerical control machine tools, aiming to solve the technical problems of poor real-time performance and low synchronization accuracy of economic Modbus communication technology in multi-axis control of numerical control machine tools, and provides a multi-axis real-time synchronous control scheme based on cycle alignment and synchronous broadcast.

[0013] To achieve the above technical purpose, the technical scheme adopted is: a Modbus multi-axis real-time synchronous control method for numerical control machine tools, comprising the following steps:

[0014] Establish a global time reference to achieve clock synchronization between the main controller and the group of servo drives;

[0015] A dynamic time slot allocation mechanism is adopted to adjust the execution time window of trajectory interpolation calculation, instruction packaging, Modbus synchronous broadcast transmission and state monitoring tasks in real time according to the complexity of the machining trajectory and the network load;

[0016] Through the instruction pre-caching and synchronous triggering mechanism, the instruction transmission time and execution time are decoupled, and the space-time alignment of multi-axis motion is realized;

[0017] Implement feedforward-feedback dual-mode fault-tolerant control to maintain the continuity of the processing process when communication anomalies are detected.

[0018] The clock synchronization between the main controller and the servo driver group is specifically implemented by unifying the interpolation period, the motion control period and the communication transmission period into a reference period T, and all devices work cooperatively by taking the period T as the minimum scheduling unit.

[0019] The dynamic time slot allocation mechanism adopts a machine learning algorithm based on historical operation data and real-time system state to predict and determine the optimal time slot allocation scheme.

[0020] The instruction pre-cache adopts a double buffering mechanism to divide a foreground buffer and a background buffer, the foreground buffer is used for the instruction execution of the current control period, and the background buffer is used for receiving the instruction data of the next control period.

[0021] A Modbus multi-axis real-time synchronous control method for a numerical control machine tool also comprises intelligent compression transmission of instruction data, adopts a data compression algorithm combining differential encoding and stroke encoding, and reduces the network transmission data amount.

[0022] The main controller is integrated with a FPGA timing processing unit capable of synchronous signal generation and accurate time slot scheduling function.

[0023] The prediction controller capable of predicting subsequent motion trajectories based on historical instruction sequences and making motion preparation in advance is built in the servo driver.

[0024] The Modbus broadcast transmission adopts a synchronous broadcast communication protocol, uses a Modbus broadcast address and an extended function code, and simultaneously sends a single data frame containing multi-axis motion instructions to all servo drivers; the protocol frame is extended with a synchronization information field and a high-precision timestamp field on the basis of the standard Modbus frame structure.

[0025] The specific implementation steps of the feedforward-feedback dual-mode fault-tolerant control are as follows: when the communication error rate is detected to be lower than a threshold value, feedforward control is executed, the feedforward control establishes a delay prediction model based on historical communication quality data, predicts network delay fluctuation according to the prediction model, actively adjusts the instruction sending time for compensation, keeps the system performance above 95%, and switches back to the normal mode until the abnormality is eliminated; when the communication error rate is detected to be greater than or equal to the threshold value, feedback control is executed, the feedback control first dynamically reallocates each task execution window, realizes gradual recovery by using a smooth transition strategy, and then completes rapid abnormal recovery within 50ms to eliminate the abnormality and switch back to the normal mode.

[0026] The present application has the following beneficial effects:

[0027] 1. The synchronization accuracy is greatly improved: the multi-axis synchronization error is reduced from ±85us of the traditional Modbus scheme to ±0.8us, which is improved by two orders of magnitude.

[0028] 2. The machining quality is obviously improved: the roundness machining error is reduced from 42μm to 9.5μm, and the workpiece surface machining quality is obviously improved.

[0029] 3. The system cost is obviously reduced: compared with the special bus solution, the hardware cost is reduced by more than 68%, and there is a significant economic advantage.

[0030] 4. The system reliability is enhanced: the communication transmission reliability reaches 99.998%, the communication exception recovery time is less than 50ms, and the system stability is greatly improved.

[0031] 5. The effect of "economic bus realizing high-end performance": the cost of the application is only 1 / 3 of the special bus solution, and the Modbus communication network of the application realizes the synchronous control performance close to 90% of the special bus, breaking the technical prejudice that "high performance must be high cost".

[0032] 6. The effect of "communication exception lossless machining": in the case of short communication link interruption (within 200ms), the system ensures that the machining process is not affected through instruction pre-caching, synchronous trigger mechanism and feedforward-feedback dual-mode fault-tolerant control, realizing the unique effect of "communication interruption and machining not stopping".

[0033] 7. The effect of "adaptive load balancing": the system can automatically adjust the control strategy according to the machining track complexity and network load state, and keep the performance stable within 95% of the load change range, effectively solving the technical problem of "high load collapse" of traditional systems. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 It is a system framework diagram of the application;

[0035] Figure 2 It is a dynamic time slot allocation mechanism schematic diagram of the application;

[0036] Figure 3 It is a four-layer synchronous signal architecture diagram;

[0037] Figure 4 It is a dual-mode fault-tolerant control strategy state diagram;

[0038] Figure 5 It is a performance comparison analysis diagram;

[0039] Figure 6 It is a flowchart of the application. DETAILED DESCRIPTION

[0040] The preferred embodiments of the application are described below in detail with reference to the accompanying drawings. Here, the application will be described in detail with reference to the accompanying drawings. It should be particularly noted that the preferred embodiments described herein are only used to illustrate and explain the application, and are not used to limit or define the application.

[0041] As shown in Figure 1 The method is implemented based on a numerical control system. The numerical control system architecture includes a main controller, a group of servo drives, a time-aware communication network, and a multi-layer synchronization signal generation architecture. The main controller is based on a micro-kernel real-time operating system, has a built-in intelligent time slot scheduler, and includes a precision clock source (temperature-compensated crystal oscillator, precision ±1ppm) and a motion pre-compensation module. The group of servo drives includes a plurality of servo drives supporting Modbus TCP / RTU communication protocol, and each drive has a built-in predictive controller. The time-aware communication network implements master-slave clock synchronization between the main controller and the drives based on IEEE 1588-2008 precision clock synchronization protocol, supports extended Modbus protocol, adds a synchronization information field and a high-precision timestamp field, supports priority marking of data transmission and dynamic monitoring of link quality. The multi-layer synchronization signal generation architecture includes hardware layer, network layer, software layer, and drive local layer synchronization signal generation and transmission.

[0042] The main controller uses an industrial-grade processor based on ARM Cortex-A15 core architecture, runs a seL4 micro-kernel real-time operating system that has undergone formal verification, and also integrates an FPGA timing processing unit to realize hardware-level synchronization signal generation and accurate time slot scheduling functions.

[0043] The group of servo drives is configured with 5 ASD-A2 series AC servo drives from Delta, supports Modbus TCP communication protocol, each servo drive has a double-ring instruction cache structure, a synchronization signal detection circuit, and a local precision timing function, and has a built-in predictive controller that can predict subsequent motion trajectories based on historical instruction sequences.

[0044] As shown in Figure 3 The multi-layer synchronization signal generation architecture includes four levels:

[0045] Hardware layer: FPGA precision generation, nanosecond-level precision;

[0046] Network layer: IEEE 1588 PTP, sub-microsecond synchronization;

[0047] Software layer: micro-kernel system call, deterministic scheduling;

[0048] Drive local layer: crystal oscillator compensation, temperature drift correction.

[0049] As shown in Figures 2-6As shown, a Modbus multi-axis real-time synchronization control method for a numerical control machine tool, comprising the following steps:

[0050] 1. Establish a global time reference based on a precision clock to achieve clock synchronization between the main controller and the servo driver group; The specific method for synchronization is:

[0051] Unify the interpolation period, motion control period and communication transmission period of the numerical control system into a reference period T;

[0052] Establish a globally unified time reference, and all devices in the numerical control system work cooperatively with the period T as the minimum scheduling unit.

[0053] 2. Adopt a dynamic time slot allocation mechanism to adjust the execution time window of trajectory interpolation calculation, instruction packaging, Modbus synchronization broadcast transmission and state monitoring tasks in real time according to the complexity of the machining trajectory and the network load.

[0054] The dynamic time slot allocation mechanism adopts an intelligent time slot allocation scheme based on a machine learning algorithm to predict the optimal time slot allocation scheme based on historical operation data and real-time system state.

[0055] 3. Through the instruction pre-caching and synchronization triggering mechanism, the instruction transmission time and the execution time are decoupled to realize the space-time alignment of multi-axis motion; The specific implementation manner is:

[0056] The instruction pre-caching adopts a double buffering mechanism and is divided into a foreground buffer and a background buffer, the foreground buffer is used for instruction execution in the current control period, and the background buffer is used for receiving instruction data in the next control period; Based on a unified synchronization signal, new motion instructions are loaded and executed at the same time to realize multi-axis space-time alignment.

[0057] 4. Implement feedforward-feedback dual-mode fault-tolerant control to maintain the continuity of the machining process when communication abnormalities are detected.

[0058] 5. Intelligent data compression transmission, adopt a data compression algorithm combining differential encoding and run-length encoding to reduce network transmission data volume.

[0059] 6. Synchronization broadcast communication protocol, the specific implementation manner is:

[0060] Adopt an extended Modbus protocol, use a broadcast address (0x00) and a custom function code to realize that a single data frame simultaneously issues instructions to all connected servo drivers; The protocol frame is functionally extended on the standard Modbus structure, a synchronization information field is newly added to transmit control information such as period count and synchronization trigger flag; A high-precision timestamp field is integrated, based on IEEE1588 protocol, to mark the accurate transmission and scheduled execution time for each instruction frame, to provide a unified time reference for realizing space-time alignment.

[0061] 7. Hardware-level synchronization signal generation, by integrating FPGA timing units in the main controller, to realize hardware-level synchronization signal generation and precise time slot scheduling functions.

[0062] 8. Motion trajectory prediction control, by embedding a prediction controller in the servo driver, to predict the subsequent motion trajectory based on the instruction sequence, making the servo driver an enhanced servo driver.

[0063] In combination Figures 1-4 , the system workflow is as follows:

[0064] Step 1: System initialization process

[0065] Load the seL4 microkernel real-time operating system to create a task isolation domain;

[0066] Configure the intelligent time slot scheduler and set the time slot table for periodic task execution;

[0067] Establish a precise clock synchronization mechanism based on IEEE1588;

[0068] Initialize the extended ModbusTCP server and listen to the standard industrial Ethernet port 502.

[0069] Step 2: Dynamic time slot allocation combined with task scheduling Figure 2 The dynamic time slot allocation mechanism is shown in the figure, and the specific implementation of dynamic time slot allocation is as follows:

[0070] Real-time analysis of processing trajectory complexity and network load state using machine learning algorithms (based on LSTM neural network); trajectory complexity evaluation based on curvature change rate, acceleration continuity, inflection point density, etc. Multi-dimensional parameters; network load monitoring includes link utilization, error rate, transmission delay, etc. Key indicators; dynamically adjust the time slot allocation of each task according to the real-time evaluation results, optimize system resource utilization.

[0071] Periodic task scheduling execution, adjust the execution time window of trajectory interpolation calculation, instruction packaging, intelligent compression transmission, Modbus synchronization broadcast transmission and state monitoring tasks:

[0072] S1 time slot, trajectory interpolation calculation

[0073] Read the machining trajectory data generated by the G-code interpreter;

[0074] Execute linear / arc interpolation algorithms to calculate the target position of each servo axis;

[0075] Perform motion lookahead preprocessing and speed planning;

[0076] The motion pre-compensation module performs lookahead compensation on the instructions based on historical error data.

[0077] S2 time slot, instruction encapsulation and intelligent compression transmission

[0078] Encapsulate each axis position instruction data into extended Modbus communication data frame;

[0079] Intelligent compression algorithm combining differential encoding and travel encoding:

[0080] Differential encoding: only transmit the difference value of adjacent period instructions, reduce data volume by 60-80%;

[0081] Travel encoding: compress and encode continuous same instructions;

[0082] Adaptive compression: dynamically select the optimal compression strategy according to the trajectory characteristics.

[0083] S3 time slot, Modbus synchronous broadcast transmission

[0084] Use extended synchronous broadcast communication protocol for data distribution:

[0085] Use broadcast address 0x00 and custom function code 0x18 (use Modbus reserved user function code area);

[0086] Data frame structure includes: standard Modbus header, synchronization information field (4 bytes, including period counter and synchronization flag), high-precision timestamp (6 bytes), multi-axis instruction data area, enhanced CRC check;

[0087] The main controller broadcasts the complete data frame to the network at a time in the fixed S3 communication window;

[0088] All servo drives receive the frame data at the same time and extract the instructions related to themselves and store them in the background buffer area.

[0089] S4 time slot, state monitoring task: system state monitoring and fault tolerance processing

[0090] Read the state register information of each servo drive;

[0091] Monitor communication link quality, calculate transmission error rate and response time;

[0092] Execute feedforward-feedback dual-mode fault tolerance control algorithm.

[0093] Step 3: Instruction pre-caching and synchronization trigger mechanism

[0094] Servo drive receives broadcast data frame and stores it in double-buffered instruction storage area:

[0095] Front buffer area: store current execution period instructions, read-only access;

[0096] Background buffer: Receives instructions for the next cycle and can be written to;

[0097] Perform integrity verification and decompression on the received data;

[0098] Waiting for the unified synchronization execution signal generated by the multi-layered synchronization signal generation architecture;

[0099] When the synchronization signal is triggered, all servo drives load and execute the new position command simultaneously;

[0100] The predictive controller predicts the subsequent motion trajectory based on the historical command sequence and adjusts the control parameters in advance.

[0101] Step 4: Feedforward-Feedback Dual-Mode Fault-Tolerant Control

[0102] The feedforward-feedback dual-mode fault-tolerant control is executed when a communication anomaly is detected, automatically switching between feedforward control mode and a combination of feedback control mode. Figure 4 The dual-mode fault-tolerant control strategy shown is as follows:

[0103] Feedforward control mode:

[0104] Real-time monitoring of system status, including communication error rate, response latency, etc.;

[0105] Automatically switch to feedforward control mode when a minor anomaly is detected (communication bit error rate < 5%);

[0106] Establish a delay prediction model based on historical communication quality data;

[0107] Predict network latency fluctuations in advance and proactively adjust the timing of command transmission to compensate for them;

[0108] Performance retention: Maintains over 95% of system performance under minor anomalies.

[0109] Feedback control mode:

[0110] Real-time monitoring of system status, including communication error rate, response latency, etc.;

[0111] Automatically switch to feedback control mode when a serious anomaly is detected (communication bit error rate ≥ 5%);

[0112] Time slot adjustment: Dynamically reallocates the execution windows for each task;

[0113] Gradual recovery: Employing a smooth transition strategy to avoid system shock;

[0114] Fast recovery: Completes the abnormal recovery process within 50ms;

[0115] Mode switching mechanism:

[0116] After the anomaly is eliminated, it automatically returns to normal mode from feedforward or feedback control mode;

[0117] Ensure the continuity of the processing during mode switching;

[0118] System performance verification:

[0119] Test environment configuration:

[0120] Testing equipment: Five-axis linkage high-speed machining center;

[0121] Test trajectory: Machining trajectory of complex 3D curved surfaces;

[0122] Measurement equipment: Precision measuring instruments such as laser interferometers and ballbars;

[0123] Performance test results are as follows Figure 5 As shown:

[0124] Multi-axis synchronous control accuracy: ±0.8μs;

[0125] Roundness machining error: 9.5μm;

[0126] Communication transmission reliability: 99.998%;

[0127] System cost: 68% lower compared to dedicated bus solutions.

[0128] This technical solution has been validated in an industrial production environment and demonstrates the following advantages in practical applications:

[0129] The significant reduction in hardware costs makes it possible to popularize high-precision multi-axis synchronous control technology in small and medium-sized machine tools; the significant reduction in system complexity improves system reliability and maintainability; the greatly improved ease of maintenance reduces maintenance costs throughout the equipment's life cycle; it is particularly suitable for upgrading and retrofitting existing CNC machine tools and has broad market application prospects.

[0130] The above are merely preferred embodiments of the present invention and are not intended to limit or restrict the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection declared by the present invention.

Claims

1. A Modbus multi-axis real-time synchronous control method for CNC machine tools, characterized in that, Includes the following steps: Establish a global time base to achieve clock synchronization between the main controller and the servo drive group; A dynamic time slot allocation mechanism is adopted to adjust the execution time window of trajectory interpolation calculation, instruction encapsulation, Modbus synchronous broadcast transmission and status monitoring tasks in real time according to the complexity of the processing trajectory and network load; By using instruction pre-caching and synchronous triggering mechanisms, instruction transmission time and execution time are decoupled, achieving spatiotemporal alignment of multi-axis motion. The instruction pre-caching adopts a dual buffering mechanism, dividing the system into a front buffer and a back buffer. The front buffer is used for instruction execution in the current control cycle, and the back buffer is used to receive instruction data for the next control cycle. Implement feedforward-feedback dual-mode fault-tolerant control to maintain the continuity of the processing when a communication anomaly is detected.

2. The Modbus multi-axis real-time synchronous control method for CNC machine tools as described in claim 1, characterized in that: The specific implementation of clock synchronization between the main controller and the servo drive group is to unify the interpolation period, motion control period and communication transmission period into a reference period T, and all devices work together with period T as the smallest scheduling unit.

3. The Modbus multi-axis real-time synchronous control method for CNC machine tools according to claim 1, characterized in that, The dynamic time slot allocation mechanism uses a machine learning algorithm based on historical operating data and real-time system status to predict and determine the optimal time slot allocation scheme.

4. The Modbus multi-axis real-time synchronous control method for CNC machine tools according to claim 1, characterized in that: It also includes intelligent compressed transmission of instruction data, using a data compression algorithm that combines differential coding and run-length coding to reduce the amount of data transmitted over the network.

5. The Modbus multi-axis real-time synchronous control method for CNC machine tools according to claim 1, characterized in that: The main controller integrates an FPGA timing processing unit capable of synchronous signal generation and precise time slot scheduling.

6. The Modbus multi-axis real-time synchronous control method for CNC machine tools according to claim 1, characterized in that: The servo driver has a built-in predictive controller that can predict subsequent motion trajectories based on historical command sequences and prepare for motion in advance.

7. The Modbus multi-axis real-time synchronous control method for CNC machine tools according to claim 1, characterized in that: Modbus broadcast transmission uses a synchronous broadcast communication protocol, employing Modbus broadcast addresses and extended function codes to simultaneously send a single data frame containing multi-axis motion commands to all servo drives. The protocol frame extends the standard Modbus frame structure with synchronization information fields and high-precision timestamp fields.

8. The Modbus multi-axis real-time synchronous control method for CNC machine tools according to claim 1, characterized in that: The specific implementation steps of the feedforward-feedback dual-mode fault-tolerant control are as follows: When the communication bit error rate is detected to be lower than the threshold, feedforward control is executed. The feedforward control establishes a delay prediction model based on historical communication quality data, predicts network delay fluctuations according to the prediction model, and actively adjusts the timing of command transmission to compensate for the network delay fluctuations, keeping the system performance above 95% until the anomaly is eliminated and the system switches back to normal mode. When the communication bit error rate is detected to be greater than or equal to the threshold, feedback control is executed. The feedback control first dynamically reallocates the execution windows of each task, adopts a smooth transition strategy to achieve gradual recovery, and then completes rapid anomaly recovery within 50ms, eliminating the anomaly and switching back to normal mode.

Images