Industrial control system of intelligent perforating machine

Abstract

The invention belongs to the field of industrial automation control, and particularly relates to an industrial control system of an intelligent puncher, which comprises a load sudden change sensing module, a bus state monitoring module, a dynamic synchronous compensation module and a cooperative control decision module, and is characterized in that the load sudden change sensing module is used for collecting current and torque signals of a motor driver in real time; and the instantaneous change characteristics of the load are identified through a preset sudden change detection algorithm. Through cooperative work of the load abrupt change sensing module and the bus state monitoring module, real-time and accurate sensing and quantification of load abrupt change and bus disturbance caused by the load abrupt change are realized; through a feed-forward compensation strategy introduced by the dynamic synchronous compensation module, time sequence jitter caused by sudden load change can be actively predicted and offset instead of passive response, so that the clock synchronization stability of the EtherCAT bus under severe dynamic load is remarkably improved, and the array precision of multi-axis cooperative perforation is effectively guaranteed.

Description

Technical Field

[0001] This invention belongs to the field of industrial automation control, specifically an industrial control system for an intelligent punching machine. Background Technology

[0002] Industrial automation control systems are a key technology area for achieving efficient and precise production in modern manufacturing. They integrate sensors, actuators, and controllers to form the nerve center of complex electromechanical equipment. In this field, industrial Ethernet bus technology serves as the backbone network connecting various industrial devices and facilitating real-time data exchange; its performance directly determines the response speed, synchronization accuracy, and operational stability of the entire control system.

[0003] Among them, motion control systems based on the EtherCAT bus are widely used in industrial scenarios involving multi-axis collaborative operations due to their high real-time performance and precise clock synchronization capabilities. This system employs a master-slave architecture, utilizing a mechanism where data frames are processed in real-time by slave devices during transmission, significantly reducing communication latency. Furthermore, it achieves nanosecond-level clock alignment across multiple devices through distributed clock technology, providing a foundation for precision motion control under high dynamic loads.

[0004] In existing technologies, the EtherCAT bus exhibits significant limitations in adaptability when faced with drastic load fluctuations, such as those caused by intelligent punching machines performing synchronous punching of multi-hole arrays. The instantaneous start and stop of motors lead to rapid changes in system load, resulting in severe fluctuations in bus current and voltage. This electrical disturbance directly affects the signal integrity of the EtherCAT communication link. More critically, sudden load changes can cause unexpected variations in message processing delays of slave devices, disrupting the synchronization stability of the distributed clock and triggering unpredictable timing jitter. In multi-axis collaborative drilling operations that require strict synchronization, this timing instability can directly lead to drilling position deviations, reduced array accuracy, and even motion conflicts between devices, severely restricting the realization of high-precision and high-efficiency drilling.

[0005] Therefore, the present invention provides an industrial control system for an intelligent punching machine. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0007] The technical solution adopted by this invention to solve its technical problem is as follows: An industrial control system for an intelligent punching machine, comprising a load change sensing module, a bus status monitoring module, a dynamic synchronization compensation module, and a collaborative control decision module. The load change sensing module is used to collect the current and torque signals of the motor driver in real time and identify the instantaneous change characteristics of the load through a preset change detection algorithm; The bus status monitoring module is used to continuously monitor the electrical parameters and message transmission status of the EtherCAT bus communication link. The electrical parameters include bus voltage fluctuations and signal noise levels, and the message transmission status includes the message processing delay and clock offset of each slave node. The dynamic synchronization compensation module is used to dynamically adjust the synchronization mechanism of the distributed clock based on the mutation characteristics output by the load mutation sensing module and the real-time status data output by the bus status monitoring module, so as to suppress timing jitter. The collaborative control decision module is used to integrate the load status and bus synchronization status to generate optimized control commands for multi-axis collaborative drilling operations, ensuring the accuracy of the drilling position and the consistency of the array.

[0008] Preferably, the load mutation sensing module implements the mutation detection algorithm as follows: First, the collected current and torque signals are subjected to sliding window mean filtering to eliminate high-frequency noise; then, the difference between the root mean square values ​​of the signals in adjacent sampling windows is calculated; when the difference exceeds the preset mutation threshold, it is determined that a load mutation event has occurred, and the time and intensity level of the mutation are recorded.

[0009] Preferably, the message processing delay measurement process of the bus status monitoring module is as follows: the master station device embeds a precise timestamp when sending a data frame; after processing the data frame, each slave station device uses the difference between its own processing completion time and the received original timestamp as its local processing delay value, and returns it to the master station along with the response data; the master station summarizes the delay values ​​of all slave stations and, in conjunction with the reference clock of the distributed clock, calculates the relative clock offset of each slave station.

[0010] Preferably, the dynamic synchronization compensation module's synchronization mechanism dynamic adjustment process is as follows: When the load mutation sensing module detects a mutation event, the dynamic synchronization compensation module immediately starts the compensation program; the program predicts the timing jitter trend using a feedforward compensation strategy based on the mutation intensity level and the currently measured message processing delay and clock offset; by fine-tuning the offset compensation register value of the distributed clock, the local clock of the slave device is pre-corrected to offset the delay changes caused by load mutation.

[0011] Preferably, the optimized control command generation process of the collaborative control decision module is as follows: the module receives the synchronization status confirmation signal from the dynamic synchronization compensation module and the change intensity information from the load change sensing module; based on this information, the multi-axis motion trajectory is smoothed online using a pre-stored perforation process parameter library. Specifically, during periods of sudden load changes, the movement speed of non-critical axes is temporarily adjusted or a small delay is introduced to ensure that the positional accuracy of critical perforated axes is not affected by synchronization errors; once the bus synchronization stability is restored, all axes are instructed to return to their original coordinated motion trajectories.

[0012] Preferably, the system further includes an adaptive learning unit, which records historical load mutation events, corresponding compensation measures and their actual compensation effects. By analyzing historical data, the adaptive learning unit can gradually optimize the threshold setting of the mutation detection algorithm and the compensation parameters of the dynamic synchronous compensation module, enabling the system to adapt to different processing materials and piercing modes.

[0013] Preferably, the bus status monitoring module is also used to monitor the physical connection status of the bus network; when the signal attenuation exceeds the safety threshold or the bit error rate increases significantly, it sends a link quality warning to the collaborative control decision module; the collaborative control decision module can then decide whether to suspend the current processing operation or switch to a degraded operation mode.

[0014] Preferably, the system adopts a hierarchical control architecture, including an equipment layer, a control layer, and a management layer. The equipment layer consists of motors, sensors, and EtherCAT slave devices, and is responsible for performing specific perforation actions and data acquisition. The control layer integrates the load mutation sensing module, bus status monitoring module, dynamic synchronization compensation module, and collaborative control decision module, and is responsible for real-time control and compensation decisions. The management layer provides a human-machine interface and data storage functions for process parameter configuration, system status monitoring, and historical data analysis.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: This invention achieves real-time and accurate perception and quantification of load mutations and the bus disturbances they cause through the collaborative work of the load mutation sensing module and the bus status monitoring module, providing a precise data foundation for subsequent synchronization compensation and overcoming the shortcomings of existing technologies in responding to latent disturbances with lag.

[0016] The present invention introduces a feedforward compensation strategy through a dynamic synchronization compensation module, which can actively predict and offset timing jitter caused by load changes, rather than passively responding. This significantly improves the clock synchronization stability of the EtherCAT bus under severe dynamic loads and effectively ensures the array accuracy of multi-axis collaborative perforation.

[0017] This invention optimizes and smooths control commands online through a collaborative control decision module and integrates an adaptive learning unit into the system architecture. This enables the control system to not only cope with transient load shocks but also to have the self-optimization capability to adapt to different processing conditions over a long period of time, thus comprehensively improving the reliability, accuracy, and intelligence level of the intelligent drilling machine in complex industrial environments. Attached Figure Description

[0018] The invention will now be further described with reference to the accompanying drawings.

[0019] Figure 1 This is a schematic diagram of the industrial control system for the intelligent punching machine proposed in this invention; Figure 2 This is a schematic diagram of the core principle framework of the collaborative operation of load change sensing and bus status monitoring in this invention; Figure 3 This is a logical flow diagram of the dynamic synchronization compensation module in this invention; Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow of the collaborative control decision module in this invention; Figure 5 This is a schematic diagram comparing the technical effects and principles of the adaptive learning unit in this invention. Detailed Implementation

[0020] Example 1 This embodiment details the specific implementation structure and workflow of an industrial control system for an intelligent punching machine. Please refer to the appendix. Figure 1 The system adopts a hierarchical control architecture, including an equipment layer, a control layer, and a management layer. The equipment layer consists of servo motors, high-precision encoders, current sensors, torque sensors, and several EtherCAT slave devices, responsible for executing the physical actions of the punching machine and collecting real-time operating data. The control layer integrates a load mutation sensing module, a bus status monitoring module, a dynamic synchronization compensation module, and a collaborative control decision-making module, achieving high-speed data processing and decision generation through an embedded industrial computer and a field-programmable gate array (FPGA). The management layer is deployed on an industrial host computer, providing a graphical human-machine interface and a distributed database for process parameter configuration, system status monitoring, and historical data traceability.

[0021] It should be noted that the specific implementation of the load change sensing module is as follows: This module synchronously acquires the three-phase current signal and output shaft torque signal of the motor driver at a sampling frequency of 10,000 times per second through current and torque sensors at the device layer. The acquired raw signals are first processed by a hardware anti-aliasing filter to suppress high-frequency noise above 5000 Hz; Continuing from the above, please refer to the appendix. Figure 2The filtered signal is input to the sliding window processing unit of the load change sensing module. This unit uses a rectangular window with a length of 256 sampling points and continuously calculates the root mean square (RMS) value of the signal within the window by overlapping every 128 points. The difference in RMS values ​​between adjacent windows is calculated in real time by the arithmetic logic unit, and the calculation formula is as follows: in, Indicates the current window number. This represents the root mean square value of the signal within the window. This difference is fed into a comparator and compared with a preset mutation threshold. The mutation threshold is dynamically set based on 5% of the motor's rated torque. When the difference exceeds the threshold for 2 milliseconds, a load mutation event is determined to have occurred. At this time, the timestamp generation unit records the precise moment the event occurred, and the intensity assessment unit classifies the mutation into three levels based on the magnitude of the difference: Level 1 corresponds to a difference between the threshold and 150% of the threshold; Level 2 corresponds to a difference between 150% and 250% of the threshold; and Level 3 corresponds to a difference above 250% of the threshold. According to the above-described solution provided by the embodiments of the present invention, the type, time, and level data of the mutation event are encapsulated into a structure and transmitted to the bus status monitoring module and the dynamic synchronization compensation module through the direct memory access channel.

[0022] It should be noted that the specific implementation of the bus status monitoring module is as follows: This module connects to the bus network via an EtherCAT master chip and a physical layer transceiver. At the beginning of each communication cycle, the master device triggers a data frame transmission action by a hardware clock, embedding a 64-bit nanosecond-precision timestamp in the frame header. After receiving the data frame, each slave device parses the timestamp using its local microprocessor and records the processing completion time after completing the instruction processing for its node. The slave defines the processing delay value as the difference between its local completion time and the received timestamp, filling it into a specific field of the response data frame. The master station extracts the delay data of all slaves by parsing the return frame and, in conjunction with the distributed clock reference value, calculates the relative clock offset of each slave. Meanwhile, the bus status monitoring module integrates an electrical parameter acquisition submodule. This submodule samples the bus power supply voltage and differential signal amplitude in real time via an analog-to-digital converter, calculating voltage fluctuation rate and signal-to-noise ratio. The message transmission status analysis submodule statistically analyzes the bit error rate and packet loss rate within a given period. When the bit error rate exceeds one in a million or the voltage fluctuation rate is greater than two percent, a link quality warning flag is generated. All monitoring data is updated to the shared memory area every millisecond for use by the dynamic synchronization compensation module.

[0023] It should be noted that the specific implementation method of the dynamic synchronization compensation module is as follows: Please refer to the appendix. Figure 3This module uses feedforward compensation as its core mechanism, and its inputs include the mutation event data from the load mutation sensing module and the real-time delay and offset data from the bus status monitoring module. When a mutation event is triggered, the compensation logic control unit immediately initiates a priority interrupt service routine. The routine first queries a preset compensation parameter mapping table based on the mutation level to obtain the basic compensation coefficient. Subsequently, the timing jitter prediction unit calculates the expected timing jitter value within the next 5 milliseconds using a linear extrapolation algorithm, based on the gradient of the current message processing delay change and the historical trend of the clock offset. The compensation calculation unit dynamically adjusts the value of the distributed clock offset compensation register according to the expected value, using the following formula: Where α is the compensation coefficient related to the mutation level, and ΔRMS is the root mean square difference output by the load mutation sensing module. This represents the time derivative of the message processing delay. The register value is sent to the target slave station via the EtherCAT mailbox communication protocol, and the slave station corrects the phase of its local clock accordingly. To ensure stability, the single adjustment range of the compensation amount is limited to ±10% of the reference value, and the continuous adjustment interval is no less than 0.5 milliseconds. After compensation is completed, the synchronization status verification unit verifies the compensation effect by comparing the master-slave clock difference. If the difference still exceeds the limit, a secondary compensation process is triggered.

[0024] It should be noted that the specific implementation method of the collaborative control decision-making module is as follows: Please refer to the appendix. Figure 4 This module receives synchronization confirmation signals from the dynamic synchronization compensation module and mutation intensity information from the load mutation sensing module, while simultaneously accessing the perforation process parameter library pre-stored by the management layer. The parameter library contains ideal motion trajectory data corresponding to over 200 process combinations, including material type, aperture specifications, and array layout. When the online trajectory optimization unit detects a load mutation or synchronization warning, it activates a real-time smoothing algorithm. This algorithm first identifies the critical and non-critical perforation axes in the current operation. The critical axis is defined as the main axis that directly affects the aperture position accuracy, while non-critical axes include auxiliary positioning axes and feeding axes. For non-critical axes, the optimization unit temporarily reduces their movement speed to 80% of its original value or inserts a variable delay of 1 to 5 milliseconds to offset the interference of synchronization errors on the positioning of the critical axis. Meanwhile, the motion trajectory of the critical axis is locally replanned using cubic spline interpolation to ensure that the target position accuracy is maintained while maintaining continuous velocity. All adjustment commands are sent to the device layer driver via EtherCAT periodic synchronization position mode. When the bus status monitoring module confirms that the link quality has stabilized and the dynamic synchronization compensation module reports a synchronization error of less than 50 nanoseconds, the cooperative control decision module instructs all axes to asymptotically return to the original trajectory within 10 milliseconds.

[0025] It should be noted that the specific implementation method of the system-integrated adaptive learning unit is as follows: Please refer to the appendix. Figure 5 This unit is deployed on the industrial server in the management layer and uses a circular buffer to store characteristic data of the most recent 10,000 load surge events, applied compensation parameters, and the compensated synchronization error values. The characteristic data includes the surge level, duration, associated slave station number, and bus voltage fluctuation amplitude. The compensation effectiveness evaluation submodule quantifies the effectiveness of each compensation by calculating the percentage decrease in synchronization error before and after compensation. The parameter optimization engine, based on stochastic gradient descent, aims to minimize the average synchronization error by iteratively adjusting the threshold of the load mutation sensing module and the compensation coefficient of the dynamic synchronization compensation module. The learning cycle is set to perform full-data training every 24 hours, and the training results are updated online to each module of the control layer after being confirmed by engineers. In addition, the adaptive learning unit also constructs a mapping relationship library between the processed materials and the optimal compensation parameters. When the system switches to processing new materials, it can prioritize loading the historical optimal parameter combination to reduce the adaptation time.

[0026] The physical connection status monitoring function of the bus status monitoring module is implemented through the link quality diagnosis submodule. This submodule periodically sends test data frames to all slave stations and statistically analyzes the signal strength and bit error rate of the response frames. When signal attenuation exceeds -3 dB or the bit error rate exceeds 5 parts per million for three consecutive cycles, a link quality early warning message is generated and notified to the collaborative control decision module via real-time interruption. The collaborative control decision module executes a tiered response based on the warning level: For Level 1 alerts, the system continues to operate but records event logs. For Level 2 alerts, the system suspends non-emergency tasks and activates redundant communication paths. For a Level 3 warning, the system immediately interrupts the processing operation and switches to safe mode, while simultaneously issuing an alarm through the human-machine interface.

[0027] In the system's layered control architecture, communication between each layer is guaranteed by standard industrial protocols. The device layer and control layer exchange data at the microsecond level via EtherCAT. Internal modules within the control layer use shared memory and semaphore mechanisms for data synchronization. The control layer and management layer communicate asynchronously via the OPC UA protocol, ensuring decoupling between real-time control and higher-level management. All critical data is timestamped and includes a checksum, and cyclic redundancy checks ensure transmission integrity. Upon system startup, a self-test process is executed to verify the firmware version and hardware configuration of each module, and the effectiveness of the compensation logic is verified through simulated load surge scenarios.

[0028] Example 2 This embodiment provides an alternative implementation scheme for the industrial control system of an intelligent punching machine. Its core feature is that the dynamic synchronization compensation module adopts a fuzzy adaptive control strategy to replace the feedforward compensation mechanism in Embodiment 1. Please refer to the attached document. Figure 3 The input variables for this strategy include the intensity of load mutation, the rate of change of message processing delay, and the absolute value of clock offset. The fuzzification interface converts precise quantities into three linguistic variables: low, medium, and high, with their membership functions using a triangular distribution. The fuzzy rule base contains 27 empirical rules. For example, if the mutation strength is high and the delay change rate is high, the compensation amount is positive. The inference engine uses the Mamdani min-max method for fuzzy implication operations, and the centroid method is used in the defuzzification stage to output the precise compensation register adjustment amount. This scheme is suitable for complex load scenarios with significant nonlinear characteristics, reducing the system's dependence on precise mathematical models through the robustness of fuzzy logic.

[0029] In this embodiment, the collaborative control decision module introduces a dynamic priority scheduling mechanism. This mechanism decomposes the drilling task into three sub-tasks: positioning, drilling, and tool retraction, and assigns a real-time priority to each sub-task. When the bus status monitoring module detects an increase in synchronization error, the scheduler dynamically increases the priority of the positioning task to ensure that the positional accuracy is not affected by subsequent actions. Meanwhile, the task queue manager implements time-slice round-robin scheduling for non-critical tasks to avoid a single task blocking system resources for a long time.

[0030] In this embodiment, the adaptive learning unit integrates a reinforcement learning algorithm. This algorithm uses the reciprocal of the synchronization error as the reward function and iteratively optimizes the weights of the fuzzy rule base through Q-learning. The learning process is pre-trained 1000 times in a simulation environment before being deployed to the actual system. During the online learning phase, an ε-greedy strategy is used to balance exploration and exploitation. The ε value is initially set to 0.1 and decays to 0.01 over time.

[0031] In this embodiment, the bus status monitoring module incorporates a spectrum analysis function. This function analyzes the frequency domain characteristics of the bus signal using a Fast Fourier Transform (FFT). When interference at a specific frequency is detected, it automatically activates a digital filter bank to suppress noise. The filter parameters are dynamically configured based on the interference spectrum to ensure signal integrity.

[0032] In this embodiment, the system management layer deploys a digital twin model. This model receives control layer data in real time, synchronously maps the physical system state in virtual space, and predicts load trends and synchronization risks within the next 5 seconds. The prediction results are published to the collaborative control decision module via OPC UA to assist it in making forward-looking control decisions.

[0033] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An industrial control system for an intelligent punching machine, characterized in that, include: The load change sensing module is used to collect the current and torque signals of the motor driver in real time, and identify the instantaneous change characteristics of the load through a preset change detection algorithm, so as to output the change event type, occurrence time and intensity level. The bus status monitoring module is used to continuously monitor the electrical parameters and message transmission status of the EtherCAT bus communication link. The electrical parameters include bus voltage fluctuations and signal noise levels, and the message transmission status includes the message processing delay and clock offset of each slave node. The dynamic synchronization compensation module is used to dynamically adjust the synchronization mechanism of the distributed clock based on the mutation characteristics output by the load mutation sensing module and the real-time status data output by the bus status monitoring module, so as to suppress timing jitter. The collaborative control decision module is used to generate optimized control commands for multi-axis collaborative drilling operations by comprehensively considering the load status and bus synchronization status. The load mutation sensing module's mutation detection algorithm implementation process includes: performing sliding window mean filtering on the collected current and torque signals to eliminate high-frequency noise; calculating the difference between the root mean square values ​​of the signals in adjacent sampling windows; and determining that a load mutation event has occurred when the difference exceeds a preset mutation threshold. The message processing delay measurement process of the bus status monitoring module includes: the master station device embeds a precise timestamp when sending a data frame; after processing the data frame, each slave station device uses the difference between its own processing completion time and the received original timestamp as the local processing delay value, and returns it to the master station along with the response data; the master station summarizes the delay values ​​of all slave stations, and calculates the relative clock offset of each slave station by combining the reference clock of the distributed clock. The dynamic adjustment process of the synchronization mechanism of the dynamic synchronization compensation module includes: when the load mutation sensing module detects a mutation event, it immediately starts the compensation program; the program predicts the timing jitter trend by using a feedforward compensation strategy based on the mutation intensity level and the currently measured message processing delay and clock offset; and pre-corrects the local clock of the slave device by fine-tuning the offset compensation register value of the distributed clock. The optimized control command generation process of the collaborative control decision module includes: receiving the synchronization status confirmation signal from the dynamic synchronization compensation module and the change intensity information from the load change sensing module; based on this information, using the pre-stored perforation process parameter library, performing online smoothing processing on the motion trajectory of the multi-axis.

2. The industrial control system for an intelligent punching machine according to claim 1, characterized in that, The system also includes an adaptive learning unit, which records historical load mutation events, corresponding compensation measures, and their actual compensation effects. By analyzing historical data, the adaptive learning unit can gradually optimize the threshold setting of the mutation detection algorithm and the compensation parameters of the dynamic synchronization compensation module.

3. The industrial control system for an intelligent punching machine according to claim 2, characterized in that, The adaptive learning unit is deployed on the industrial server of the management layer and uses a circular buffer to store the feature data of the most recent 10,000 load mutation events, the applied compensation parameters, and the compensated synchronization error value. The feature data includes mutation level, duration, associated slave station number, and bus voltage fluctuation amplitude.

4. The industrial control system for an intelligent punching machine according to claim 3, characterized in that, The adaptive learning unit also includes a compensation effect evaluation submodule and a parameter optimization engine; The compensation effect evaluation submodule quantifies the effectiveness of each compensation by calculating the percentage decrease in synchronization error before and after compensation. The parameter optimization engine is based on stochastic gradient descent and aims to minimize the average synchronization error by iteratively adjusting the threshold of the load mutation sensing module and the compensation parameters of the dynamic synchronization compensation module.

5. The industrial control system for an intelligent punching machine according to claim 1, characterized in that, The bus status monitoring module is also used to monitor the physical connection status of the bus network; When signal attenuation exceeds the safety threshold or the bit error rate increases significantly, a link quality warning is sent to the collaborative control decision module.

6. The industrial control system for an intelligent punching machine according to claim 5, characterized in that, The collaborative control decision module responds to link quality warnings by executing tiered response control: For Level 1 alerts, the system continues to operate but records event logs. For Level 2 alerts, the system suspends non-emergency tasks and activates redundant communication paths; For a Level 3 warning, the system immediately interrupts the processing operation and switches to safe mode.

7. The industrial control system for an intelligent punching machine according to claim 1, characterized in that, The system adopts a layered control architecture, including a device layer, a control layer, and a management layer; The equipment layer consists of motors, sensors, and EtherCAT slave devices; The control layer integrates the load mutation perception module, bus status monitoring module, dynamic synchronization compensation module, and collaborative control decision module; the management layer provides a human-machine interface and data storage functions.

8. The industrial control system for an intelligent punching machine according to claim 7, characterized in that, The device layer and the control layer exchange data at the microsecond level via EtherCAT; the internal modules of the control layer use shared memory and semaphore mechanisms for data synchronization; and the control layer and the management layer communicate asynchronously via the OPC UA protocol.

9. The industrial control system for an intelligent punching machine according to claim 1, characterized in that, The compensation calculation process adopted by the dynamic synchronization compensation module includes: querying the preset compensation parameter mapping table according to the mutation level to obtain the basic compensation coefficient; Based on the gradient of current message processing latency changes and the historical trend of clock offset, a linear extrapolation algorithm is used to calculate the expected value of timing jitter within the next 5 milliseconds; the value of the offset compensation register of the distributed clock is dynamically adjusted according to the expected value.

10. The industrial control system for an intelligent punching machine according to claim 1, characterized in that, The online smoothing process of the collaborative control decision module includes: Identify the critical and non-critical perforated shafts in the current operation; For non-critical axes, temporarily reduce their motion speed to 80% of the original value, or insert a variable delay of 1 to 5 milliseconds; For the critical axis, a cubic spline interpolation method is used for local trajectory replanning; once the bus synchronization stability is restored, all axes are instructed to asymptotically return to the original trajectory within 10 milliseconds.

Images