High-frequency real-time control method and device based on integrated bus architecture

Abstract

The embodiment of the invention provides a high-frequency real-time control method and device based on an integrated bus architecture, a system is divided into a resolving module and a control module, efficient data interaction is realized through a concurrent queue, an integrated architecture of an industrial Ethernet bus and a field bus is adopted, a load prediction model is trained based on historical data, and a high-frequency real-time control system is established. And dynamic adjustment of the task weight coefficient is realized. Adaptive initialization of different models of motors is realized by establishing a motor abstraction layer, and control parameters are dynamically generated based on a load prediction result. According to the method, the limitation of a traditional fixed scheduling strategy is broken through, and the real-time performance and reliability of multi-motor cooperative control are remarkably improved through adaptive task scheduling and load prediction compensation.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a high-frequency real-time control method and device based on an integrated bus architecture. Background Art

[0002] Traditional motor control systems usually use a single bus architecture, which has problems with real-time performance and reliability. Existing control methods often use fixed task scheduling strategies, which are difficult to dynamically optimize according to load changes. Especially in multi-motor collaborative control scenarios, it is difficult for the system to balance the resource usage and response time of different tasks.

[0003] At the same time, the existing system has obvious deficiencies in load prediction and parameter adaptation. Traditional methods lack effective use of historical data, cannot accurately predict load change trends, and fail to establish a dynamic association mechanism between task priority and load status. The system is also relatively mechanical in terms of motor abstract modeling and initialization configuration, making it difficult to adapt to the control requirements of different types of motors.

[0004] In addition, the existing technology also has problems in the coordination between the solution module and the control module. The lack of efficient data interaction mechanism and the failure to fully utilize concurrent processing capabilities have affected the overall performance of the system. Solving these problems is of great significance to improving the real-time performance and reliability of the motor control system. Summary of the invention

[0005] In response to the problems in the prior art, the present application provides a high-frequency real-time control method and device based on an integrated bus architecture, which can significantly improve the real-time and reliability of multi-motor collaborative control through adaptive task scheduling and load prediction compensation.

[0006] In order to solve at least one of the above problems, the present application provides the following technical solutions: In a first aspect, the present application provides a high-frequency real-time control method based on an integrated bus architecture, comprising: The control system is divided into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to perform data exchange, the control module includes a plurality of motor control units, each of which is connected to a corresponding motor through an integrated bus, the integrated bus includes an industrial Ethernet bus and a field bus, the solution module includes a plurality of solution tasks, each of which is set with a corresponding task weight coefficient, and a task history database is established, wherein the task history database records the response time and resource occupancy of the motor in executing each solution task under different load conditions; The real-time load data of the motor is collected through the integrated bus, wherein the real-time load data includes motor torque, motor speed and motor current, the real-time load data is input into a load prediction model, the load prediction model is obtained by training based on the task history database, the load prediction model outputs a load prediction result, the load prediction result includes a predicted load interval and a load change trend, the task weight coefficient is dynamically adjusted based on the load prediction result, a pre-adjustment weight coefficient is generated, the pre-adjustment weight coefficient changes nonlinearly with the change of the load prediction result, the solution tasks are prioritized based on the pre-adjustment weight coefficient, an adaptive task scheduling strategy is established according to the priority ranking, the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue; The control module receives the solution result, reads a motor configuration file, wherein the motor configuration file includes motor model parameters, motor initialization parameters and motor control parameters, abstractly models the motor according to the motor model parameters, generates a motor abstraction layer, adaptively initializes the motor by calling the motor initialization parameters based on the motor abstraction layer, matches the load prediction result with the motor control parameters, generates load adaptive control parameters, compensates and adjusts the solution result according to the load adaptive control parameters, and sends control instructions to the motor through the integrated bus. The control module receives encoding feedback information of the motor through the integrated bus, and stores the encoding feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model.

[0007] Furthermore, the control system is divided into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to exchange data, the control module includes a plurality of motor control units, each of which is connected to a corresponding motor through an integrated bus, and the integrated bus includes an industrial Ethernet bus and a field bus, including: A concurrent data queue structure is established based on a high-speed cache, a read-write controller and a data synchronizer are configured in the concurrent data queue structure, the read-write controller includes a mutex mechanism and a conditional variable, the data synchronizer is responsible for read-write synchronization and status monitoring of queue data, data transmission from a solution module to a control module is realized through the concurrent data queue structure, a data priority identification bit and a data check bit are set in the data queue, the data priority identification bit is used to mark the priority of data processing, and the data check bit is used to ensure the correctness of data transmission; A multi-channel motor control unit array is constructed in the control module, and an independent communication resource pool is allocated to each motor control unit. The communication resource pool includes an Ethernet communication interface and a field bus interface. A TCP / IP protocol stack and a real-time Ethernet protocol stack are configured in the Ethernet communication interface, and a CAN bus protocol is configured in the field bus interface. A corresponding communication protocol is selected according to the motor communication characteristics to establish a control link. The control link includes an uplink data channel and a downlink data channel. The sending of motor control instructions and the receiving of status feedback are realized through the control link.

[0008] Furthermore, the solution module includes a plurality of solution tasks, each of which sets a corresponding task weight coefficient, and establishes a task history database, which records the response time and resource occupancy of the motor executing each solution task under different load conditions, including: A task manager is established in the solution module. The task manager adopts a hierarchical structure design, including a task scheduling layer, a task execution layer and a resource management layer. A task priority queue is established in the task scheduling layer. A unique task identifier and an initial weight coefficient are assigned to each solution task. The task identifier includes a task type code and timestamp information. The initial weight coefficient is determined based on the real-time requirements and computational complexity of the task. The creation, destruction and status management of the solution task are realized through the task manager. A task history database of a distributed data storage architecture is constructed. The task history database includes a real-time data table and a historical data table. The execution information of each solution task under the current load state is recorded in the real-time data table. The execution information includes the task start time, completion time, CPU occupancy and memory usage. A load state index is established in the historical data table. The task execution information under different load states is stored in time series. The historical data is regularly archived and cleaned up through a data compression algorithm.

[0009] Further, the real-time load data is input into a load prediction model, the load prediction model is obtained by training based on the task history database, the load prediction model outputs a load prediction result, the load prediction result includes a predicted load interval and a load change trend, the task weight coefficient is dynamically adjusted based on the load prediction result to generate a pre-adjusted weight coefficient, the pre-adjusted weight coefficient changes nonlinearly with the change of the load prediction result, and the solution tasks are prioritized based on the pre-adjusted weight coefficient, including: A load prediction model is constructed based on a long short-term memory neural network, wherein the load prediction model includes an input layer, a hidden layer, and an output layer, wherein a data normalization unit is set in the input layer to perform standardization processing on real-time load data, a forget gate and a memory gate are set in the hidden layer, invalid features are filtered through the forget gate, and important features are retained through the memory gate, an activation function is set in the output layer, the output of the neural network is mapped into a load prediction interval and a change trend, and the load prediction model is trained and parameter optimized using sample data in a task history database; A weight adjuster based on a fuzzy neural network is constructed. The weight adjuster receives the output result of the load prediction model, converts the load prediction result into a weight adjustment amount through a fuzzy rule base, and the fuzzy rule base includes a load interval rule set and a trend rule set. A pre-adjustment weight coefficient is calculated according to the weight adjustment amount, and the pre-adjustment weight coefficient is input into a task priority sorting module. The task priority sorting module uses an improved quick sorting algorithm to re-prioritize the solution tasks and generate a task execution sequence.

[0010] Further, the adaptive task scheduling strategy is established according to the priority sorting, the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue, including: Construct a multi-level feedback task scheduler, which includes a task allocation unit and a resource monitoring unit. Multiple priority queues are set in the task allocation unit. Each priority queue is configured with an independent time slice length and scheduling cycle. A dynamic time slice is set based on the response time in the task history database. A task switching threshold is set based on the resource occupancy rate. The resource monitoring unit collects the system resource utilization rate in real time. When the resource utilization rate exceeds the switching threshold, the task downgrade process is triggered. A parallel computing framework is established in the solution module, which includes a task decomposer and a result synthesizer. The task decomposer splits the solution task into multiple subtasks according to an adaptive task scheduling strategy, and allocates computing resources to process each subtask in parallel. The result synthesizer summarizes and verifies the calculation results of the subtasks, generates a final solution result, encapsulates the solution result into a data packet, and transmits the data packet to the control module through a concurrent queue mechanism.

[0011] Further, the motor is abstractly modeled according to the motor model parameters to generate a motor abstraction layer, the motor initialization parameters are called based on the motor abstraction layer to adaptively initialize the motor, the load prediction result is matched with the motor control parameter to generate a load adaptive control parameter, the solution result is compensated and adjusted according to the load adaptive control parameter, and a control instruction is sent to the motor through the integrated bus, including: Constructing a motor object abstraction layer, the motor object abstraction layer includes a parameter description module and a function interface module, establishing a motor characteristic model in the parameter description module, the motor characteristic model includes an electrical parameter model and a mechanical parameter model, configuring a basic parameter matrix according to motor model parameters, encapsulating a standardized control instruction set in the function interface module, the control instruction set includes motion control instructions and status query instructions, and realizing unified management of motors of different models through the motor object abstraction layer; A motor adaptive controller is established, wherein the adaptive controller includes a parameter matching unit and a compensation adjustment unit. The parameter matching unit receives a load prediction result, selects an optimal control parameter group from a motor control parameter library, and adaptively adjusts the control parameters according to the load state. The compensation adjustment unit corrects and compensates the solution result based on the load adaptive control parameter, optimizes the control performance by adopting a feedforward compensation strategy, and sends the corrected control instruction to the motor execution unit through an integrated bus.

[0012] Furthermore, the control module receives the encoder feedback information of the motor through the integrated bus, and stores the encoder feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model, including: Construct a feedback signal processor, the feedback signal processor includes a data acquisition unit and a signal analysis unit, the data acquisition unit periodically acquires position information, speed information and current information output by the motor encoder through an integrated bus, a signal filter and a data verifier are set in the signal analysis unit, the collected data is filtered for high-frequency noise through the signal filter, the data integrity is verified through the data verifier, and the processed feedback information is encoded and packaged according to a preset format; A data update manager is established, which includes a data classification module and a model optimization module. The data classification module performs time synchronization and feature extraction on the encoded feedback information and load data, and stores the processed data classifications in the corresponding data table of the task history database. The model optimization module periodically reads the newly added data samples, and uses an incremental learning algorithm to perform online fine-tuning on the network parameters of the load prediction model to achieve continuous optimization of the prediction model.

[0013] In a second aspect, the present application provides a high-frequency real-time control device based on an integrated bus architecture, comprising: A solution module is used to divide the control system into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to perform data exchange, the control module includes a plurality of motor control units, each of which is connected to a corresponding motor via an integrated bus, the integrated bus includes an industrial Ethernet bus and a field bus, the solution module includes a plurality of solution tasks, each of which is set with a corresponding task weight coefficient, and a task history database is established, wherein the task history database records the response time and resource occupancy of the motor in executing each solution task under different load conditions; A prediction module, used for collecting real-time load data of the motor through the integrated bus, wherein the real-time load data includes motor torque, motor speed and motor current, inputting the real-time load data into a load prediction model, the load prediction model is obtained by training based on the task history database, the load prediction model outputs a load prediction result, the load prediction result includes a predicted load interval and a load change trend, dynamically adjusting the task weight coefficient based on the load prediction result to generate a pre-adjusted weight coefficient, the pre-adjusted weight coefficient changes nonlinearly with the change of the load prediction result, prioritizing the solution tasks based on the pre-adjusted weight coefficient, establishing an adaptive task scheduling strategy based on the priority ranking, the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue; A control module is used for the control module to receive the solution result, read the motor configuration file, wherein the motor configuration file includes motor model parameters, motor initialization parameters and motor control parameters, abstractly model the motor according to the motor model parameters, generate a motor abstraction layer, adaptively initialize the motor by calling the motor initialization parameters based on the motor abstraction layer, match the load prediction result with the motor control parameters, generate load adaptive control parameters, compensate and adjust the solution result according to the load adaptive control parameters, and send control instructions to the motor through the integrated bus. The control module receives the encoding feedback information of the motor through the integrated bus, and stores the encoding feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model.

[0014] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the high-frequency real-time control method based on an integrated bus architecture when executing the program.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the high-frequency real-time control method based on an integrated bus architecture.

[0016] In a fifth aspect, the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the high-frequency real-time control method based on an integrated bus architecture.

[0017] It can be seen from the above technical solution that the present application provides a high-frequency real-time control method and device based on an integrated bus architecture, which divides the system into a solution module and a control module, realizes efficient data interaction through concurrent queues, adopts an integrated architecture of industrial Ethernet bus and field bus, trains load prediction models based on historical data, and realizes dynamic adjustment of task weight coefficients. Adaptive initialization of motors of different models is achieved by establishing a motor abstraction layer, and control parameters are dynamically generated based on load prediction results. This method breaks through the limitations of traditional fixed scheduling strategies, and significantly improves the real-time and reliability of multi-motor collaborative control through adaptive task scheduling and load prediction compensation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is one of the flow charts of the high-frequency real-time control method based on the integrated bus architecture in the embodiment of the present application; Figure 2 This is a second flow chart of a high-frequency real-time control method based on an integrated bus architecture in an embodiment of the present application; Figure 3 This is a third flow chart of a high-frequency real-time control method based on an integrated bus architecture in an embodiment of the present application; Figure 4 This is a fourth flow chart of a high-frequency real-time control method based on an integrated bus architecture in an embodiment of the present application; Figure 5This is a fifth flow chart of a high-frequency real-time control method based on an integrated bus architecture in an embodiment of the present application; Figure 6 This is a sixth flow chart of a high-frequency real-time control method based on an integrated bus architecture in an embodiment of the present application; Figure 7 FIG7 is a flow chart of a high-frequency real-time control method based on an integrated bus architecture in an embodiment of the present application; Figure 8 It is a structural diagram of a high-frequency real-time control device based on an integrated bus architecture in an embodiment of the present application; Fig. 9 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application.

[0020] Reference numerals: Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0023] Taking into account the problems existing in the prior art, the present application provides a high-frequency real-time control method and device based on an integrated bus architecture, which divides the system into a solution module and a control module, realizes efficient data interaction through concurrent queues, adopts an integrated architecture of industrial Ethernet bus and field bus, trains load prediction models based on historical data, and realizes dynamic adjustment of task weight coefficients. Adaptive initialization of motors of different models is achieved by establishing a motor abstraction layer, and control parameters are dynamically generated based on load prediction results. This method breaks through the limitations of traditional fixed scheduling strategies, and significantly improves the real-time and reliability of multi-motor collaborative control through adaptive task scheduling and load prediction compensation.

[0024] In order to significantly improve the real-time performance and reliability of multi-motor coordinated control through adaptive task scheduling and load prediction compensation, the present application provides an embodiment of a high-frequency real-time control method based on an integrated bus architecture, see Figure 1 The high-frequency real-time control method based on the integrated bus architecture specifically includes the following contents: Step S101: dividing the control system into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to perform data exchange, the control module includes a plurality of motor control units, each of the motor control units is connected to a corresponding motor through an integrated bus, the integrated bus includes an industrial Ethernet bus and a field bus, the solution module includes a plurality of solution tasks, each of the solution tasks is set with a corresponding task weight coefficient, and a task history database is established, wherein the task history database records the response time and resource occupancy of the motor in executing each solution task under different load conditions; Optionally, the high-frequency real-time control method based on the integrated bus architecture provided in this embodiment first divides the control system into independent solution modules and control modules. This division is based on the demand for high-precision motion control in actual industrial scenarios. The solution module is responsible for complex trajectory planning, dynamic calculation and other computationally intensive tasks, while the control module focuses on real-time control and state monitoring of the motor. This functional separation can improve the real-time performance and reliability of the system.

[0025] This embodiment uses a concurrent queue to implement data interaction between the solution module and the control module. Specifically, the concurrent queue adopts a double-ended circular buffer structure, the write end of the queue is connected to the output interface of the solution module, and the read end is connected to the input interface of the control module. In order to ensure the real-time and reliability of data interaction, a semaphore-based mutual exclusive access mechanism is implemented in the concurrent queue. When the solution module completes a calculation task, the semaphore mechanism is used to wake up the control module to read new control data.

[0026] In the specific implementation of the control module, a multi-channel motor control unit array is constructed, and each control unit is independently responsible for the control of a motor. In order to meet the communication requirements of different types of motors, this embodiment designs an integrated bus architecture that integrates industrial Ethernet and fieldbus. Among them, the industrial Ethernet bus adopts the EtherCAT protocol, which supports high-speed data transmission and precise synchronous control; the fieldbus includes CAN bus and EtherCAT bus, which are used to connect traditional industrial equipment. This hybrid bus architecture not only ensures the real-time performance of the system, but also provides good compatibility.

[0027] In this embodiment, multiple solving tasks are set in the solving module, including trajectory planning tasks, kinematic solving tasks, dynamic calculation tasks, etc. Each solving task is assigned a task weight coefficient, which reflects the importance and real-time requirements of the task. For example, in the case of emergency braking of the robot, the weight coefficient of the trajectory planning task will be dynamically increased to ensure that the system can respond quickly to the braking command.

[0028] In order to realize dynamic scheduling and optimization of tasks, this embodiment establishes a task history database. The database adopts a distributed storage architecture and includes a real-time data area and a historical data area. In the real-time data area, the current load state of the motor (such as speed, torque, current, etc.) and the execution status of each solution task (such as start time, end time, CPU occupancy, memory usage, etc.) are recorded. The historical data area stores the long-term accumulated task execution data, which is classified and indexed according to the load state for subsequent task scheduling optimization and load prediction.

[0029] In practical applications, when the motor load changes significantly, such as a sudden increase in load caused by a robot holding a heavy object, this embodiment can quickly adjust the task weight coefficient and execution order based on the empirical data in the task history database. This adaptive scheduling mechanism based on historical data effectively improves the response speed and control accuracy of the system under different load conditions.

[0030] By modularizing the control system, using concurrent queues to achieve inter-module communication, integrating multiple bus technologies to build an integrated control network, and introducing dynamic task weight adjustment and historical data analysis mechanisms, this embodiment significantly improves the real-time control performance and adaptability of the system. In actual industrial scenarios, this solution can effectively cope with complex working conditions such as sudden load changes and emergency braking, and ensure the stability and reliability of the control system.

[0031] In addition, the task history database of this embodiment not only serves the current task scheduling, but also provides a data basis for the continuous optimization of the system. By analyzing the task execution mode and resource usage characteristics in the historical data, the task scheduling strategy can be continuously improved to enhance the overall efficiency of the system. For example, in periodic production tasks, the system can predictively adjust task weights and resource allocations to achieve a more efficient control process.

[0032] Step S102: collecting real-time load data of the motor through the integrated bus, wherein the real-time load data includes motor torque, motor speed and motor current, inputting the real-time load data into a load prediction model, the load prediction model is obtained by training based on the task history database, the load prediction model outputs a load prediction result, the load prediction result includes a predicted load interval and a load change trend, dynamically adjusting the task weight coefficient based on the load prediction result, generating a pre-adjustment weight coefficient, the pre-adjustment weight coefficient changes nonlinearly with the change of the load prediction result, prioritizing the solution tasks based on the pre-adjustment weight coefficient, establishing an adaptive task scheduling strategy based on the priority ranking, the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue; Optionally, this embodiment collects motor operation status data in real time through an integrated bus to build a complete load prediction and task scheduling optimization mechanism. Specifically, the industrial Ethernet bus and field bus are combined to synchronously collect the torque, speed and current data of the motor with a sampling period of 1 millisecond, and these data together reflect the actual load status of the motor.

[0033] Based on data collection, this embodiment designs a load prediction model based on deep learning. The model adopts a long short-term memory network (LSTM) structure, which can effectively capture the timing characteristics of the motor load. The input layer of the model contains three feature dimensions: motor torque, speed, and current. The hidden layer uses the attention mechanism to weightedly fuse the features of different time scales, and the output layer gives the predicted load range and load change trend. The predicted load range represents the possible value range of the motor load in the future, and the load change trend reflects the increase, decrease, or stable state of the load.

[0034] The load prediction model of this embodiment is trained based on the operating data accumulated in the task history database. The training data includes the motor load sequence under different working conditions and the corresponding solution task execution records. Through the analysis of historical data, the model can establish a mapping relationship between motor load characteristics and solution task performance. For example, in the case of a sudden increase in load, the response time of some computationally intensive solution tasks will be significantly extended. This correlation is accurately captured by the model and used for subsequent predictions.

[0035] Based on the load prediction results, this embodiment implements a dynamic adjustment mechanism for task weight coefficients. Specifically, a nonlinear mapping function is used to convert the predicted load interval and change trend into a weight adjustment amount. When it is predicted that the load is about to change dramatically, the weight coefficients of the solution tasks with high load correlation will be significantly increased to ensure that these critical tasks can respond to load changes in a timely manner. This nonlinear adjustment strategy can achieve rapid adaptation of task weights while ensuring system stability.

[0036] At the task scheduling level, this embodiment prioritizes the solving tasks according to the pre-adjusted weight coefficients. The sorting algorithm takes into account the weight coefficients, expected execution time, and resource requirements of the tasks to generate the optimal task execution sequence. For example, when the robot performs precision assembly tasks, trajectory planning and kinematic solving tasks will be given higher priority to ensure the accuracy of motion control.

[0037] The adaptive task scheduling strategy of this embodiment is based on an in-depth analysis of historical execution data. By analyzing the response time and resource occupancy recorded in the task history database, a correlation model between task execution characteristics and system performance is established. This optimization mechanism based on historical data can dynamically adjust the execution parameters of tasks, such as time slice allocation, resource constraints, etc., thereby achieving continuous optimization of task scheduling.

[0038] The solution module performs various solution tasks according to the adaptive task scheduling strategy and generates solution results including motor control parameters. These results are transmitted to the control module through a concurrent queue. The buffer mechanism of the queue ensures the reliability and real-time performance of data transmission. During the data transmission process, a timestamp-based synchronization mechanism is used to ensure that the solution results can be delivered to the control module in time and the corresponding control actions can be executed.

[0039] This embodiment significantly improves the control system's ability to adapt to load changes by building a complete load prediction and task scheduling framework. In practical applications, this solution can effectively predict and respond to various load changes, such as robot load mutations, multi-motor coordinated control, and other complex scenarios, ensuring the system's control accuracy and response speed.

[0040] In addition, the adaptive scheduling mechanism of this embodiment also has good scalability. By continuously accumulating and analyzing operating data, the system can continuously optimize the prediction model and scheduling strategy to adapt to more diverse application scenarios. At the same time, the nonlinear weight adjustment mechanism ensures that the system can still maintain stable operation under extreme conditions, improving the overall reliability.

[0041] Step S103: The control module receives the solution result, reads the motor configuration file, wherein the motor configuration file includes motor model parameters, motor initialization parameters and motor control parameters, abstractly models the motor according to the motor model parameters, generates a motor abstraction layer, adaptively initializes the motor by calling the motor initialization parameters based on the motor abstraction layer, matches the load prediction result with the motor control parameters, generates load adaptive control parameters, compensates and adjusts the solution result according to the load adaptive control parameters, and sends control instructions to the motor through the integrated bus. The control module receives the encoding feedback information of the motor through the integrated bus, and stores the encoding feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model.

[0042] Optionally, this embodiment implements an adaptive control mechanism based on the motor abstract model in the control module. The control module first receives the solution result from the solution module from the concurrent queue, and reads the motor configuration file stored locally, which contains the complete parameter information of the motor, providing basic data support for subsequent control strategy optimization.

[0043] In the motor model construction phase, this embodiment performs abstract modeling based on motor model parameters. Specifically, different types of motors (such as servo motors, stepper motors, etc.) are abstracted into a unified mathematical model, including core characteristics such as electrical equations and mechanical equations. This abstract modeling method makes the control strategy have good versatility and portability. As an intermediate interface, the motor abstraction layer shields the hardware differences between different motor models and provides a unified calling interface for the upper-level control algorithm.

[0044] Based on the motor abstraction layer, this embodiment implements an adaptive initialization process of motor parameters. The initialization process first reads the motor initialization parameters, including basic parameters such as rated voltage, rated current, and rated speed, and then dynamically adjusts these parameters according to the current load state. For example, when it is detected that the motor load is heavy, the system will increase the current limit value accordingly to ensure that the motor can provide sufficient output torque.

[0045] This embodiment innovatively matches and maps the load prediction results with the motor control parameters to generate load adaptive control parameters. This process involves the coordinated optimization of multiple control parameters, such as PID parameters, feedforward compensation coefficients, etc. Specifically, when the load prediction results indicate that the load is about to increase, the system will increase the proportional coefficient and integral time constant in advance to improve the anti-interference ability of the control system; when it is predicted that the load will change sharply, the differential coefficient will be adjusted accordingly to enhance the dynamic response characteristics of the system.

[0046] Compensation adjustment of the solution result is another innovation of this embodiment. Based on the load adaptive control parameters, the system performs real-time compensation for the control instructions generated by the solution module. For example, when the robot performs a position control task, if it is predicted that the load increase will lead to an increase in the position error, the system will automatically calculate the compensation amount and superimpose it on the original control instruction to ensure control accuracy.

[0047] During the execution of the control instructions, this embodiment collects the encoder feedback information of the motor in real time through the integrated bus. This feedback information includes state quantities such as actual position, speed and acceleration, and is recorded in the task history database together with the current load data. This closed-loop feedback mechanism is not only used for real-time control, but also provides continuously updated training data for the optimization of the load prediction model.

[0048] The online optimization and update mechanism of this embodiment is the key to ensure the long-term stable operation of the system. By analyzing the newly added historical data, the system can promptly discover the prediction deviation of the load prediction model and fine-tune the model using incremental learning. This adaptive learning mechanism enables the prediction model to continuously adapt to the aging characteristics of the equipment and changes in operating conditions, maintaining a high prediction accuracy.

[0049] In practical applications, this embodiment significantly improves the adaptability and reliability of the motor control system. For example, in the flexible manufacturing scenario of industrial robots, when the robot needs to frequently switch workpieces of different weights, this solution can accurately predict load changes and adjust the control strategy in advance, effectively avoiding overshoot and oscillation problems in traditional control solutions.

[0050] In addition, the motor abstraction layer design of this embodiment facilitates subsequent functional expansion. By adding new interface definitions in the abstraction layer, the system can easily support the access of new motors while maintaining the stability of the original control strategy. This modular design concept greatly improves the maintainability and scalability of the system.

[0051] By establishing a complete feedback loop, this embodiment not only achieves high precision and high reliability of motor control, but also builds a continuously optimized learning system that can continuously improve control performance and prediction accuracy, providing an effective solution for the intelligent upgrade of the industrial automation field.

[0052] From the above description, it can be seen that the high-frequency real-time control method based on the integrated bus architecture provided by the embodiment of the present application can divide the system into a solution module and a control module, realize efficient data interaction through concurrent queues, adopt the integrated architecture of industrial Ethernet bus and field bus, train the load prediction model based on historical data, and realize dynamic adjustment of task weight coefficients. Adaptive initialization of motors of different models is achieved by establishing a motor abstraction layer, and control parameters are dynamically generated based on load prediction results. This method breaks through the limitations of traditional fixed scheduling strategies, and significantly improves the real-time and reliability of multi-motor collaborative control through adaptive task scheduling and load prediction compensation.

[0053] In one embodiment of the high-frequency real-time control method based on the integrated bus architecture of the present application, see Figure 2 , and can also include the following: Step S201: Establish a concurrent data queue structure based on the cache, configure a read-write controller and a data synchronizer in the concurrent data queue structure, the read-write controller includes a mutex mechanism and conditional variables, the data synchronizer is responsible for read-write synchronization and status monitoring of queue data, and data transmission from the solution module to the control module is realized through the concurrent data queue structure, and a data priority identification bit and a data check bit are set in the data queue, the data priority identification bit is used to mark the priority of data processing, and the data check bit is used to ensure the correctness of data transmission; Step S202: construct a multi-channel motor control unit array in the control module, and allocate an independent communication resource pool to each motor control unit, wherein the communication resource pool includes an Ethernet communication interface and a field bus interface, a TCP / IP protocol stack and a real-time Ethernet protocol stack are configured in the Ethernet communication interface, and a CAN bus protocol is configured in the field bus interface. The field bus interface can adopt a CAN physical interface, and a corresponding communication protocol is selected according to the motor communication characteristics to establish a control link. The control link includes an uplink data channel and a downlink data channel, and the sending of motor control instructions and the receiving of status feedback are realized through the control link.

[0054] Optionally, this embodiment implements an adaptive control mechanism based on the motor abstract model in the control module. The control module first receives the solution result from the solution module from the concurrent queue, and reads the motor configuration file stored locally, which contains the complete parameter information of the motor, providing basic data support for subsequent control strategy optimization.

[0055] In the motor model construction phase, this embodiment performs abstract modeling based on motor model parameters. Specifically, different types of motors (such as servo motors, stepper motors, etc.) are abstracted into a unified mathematical model, including core characteristics such as electrical equations and mechanical equations. This abstract modeling method makes the control strategy have good versatility and portability. As an intermediate interface, the motor abstraction layer shields the hardware differences between different motor models and provides a unified calling interface for the upper-level control algorithm.

[0056] Based on the motor abstraction layer, this embodiment implements an adaptive initialization process of motor parameters. The initialization process first reads the motor initialization parameters, including basic parameters such as rated voltage, rated current, and rated speed, and then dynamically adjusts these parameters according to the current load state. For example, when it is detected that the motor load is heavy, the system will increase the current limit value accordingly to ensure that the motor can provide sufficient output torque.

[0057] This embodiment innovatively matches and maps the load prediction results with the motor control parameters to generate load adaptive control parameters. This process involves the coordinated optimization of multiple control parameters, such as PID parameters, feedforward compensation coefficients, etc. Specifically, when the load prediction results indicate that the load is about to increase, the system will increase the proportional coefficient and integral time constant in advance to improve the anti-interference ability of the control system; when it is predicted that the load will change sharply, the differential coefficient will be adjusted accordingly to enhance the dynamic response characteristics of the system.

[0058] Compensation adjustment of the solution result is another innovation of this embodiment. Based on the load adaptive control parameters, the system performs real-time compensation for the control instructions generated by the solution module. For example, when the robot performs a position control task, if it is predicted that the load increase will lead to an increase in the position error, the system will automatically calculate the compensation amount and superimpose it on the original control instruction to ensure control accuracy.

[0059] During the execution of the control instructions, this embodiment collects the encoder feedback information of the motor in real time through the integrated bus. This feedback information includes state quantities such as actual position, speed and acceleration, and is recorded in the task history database together with the current load data. This closed-loop feedback mechanism is not only used for real-time control, but also provides continuously updated training data for the optimization of the load prediction model.

[0060] The online optimization and update mechanism of this embodiment is the key to ensure the long-term stable operation of the system. By analyzing the newly added historical data, the system can promptly discover the prediction deviation of the load prediction model and fine-tune the model using incremental learning. This adaptive learning mechanism enables the prediction model to continuously adapt to the aging characteristics of the equipment and changes in operating conditions, maintaining a high prediction accuracy.

[0061] In practical applications, this embodiment significantly improves the adaptability and reliability of the motor control system. For example, in the flexible manufacturing scenario of industrial robots, when the robot needs to frequently switch workpieces of different weights, this solution can accurately predict load changes and adjust the control strategy in advance, effectively avoiding overshoot and oscillation problems in traditional control solutions.

[0062] In addition, the motor abstraction layer design of this embodiment facilitates subsequent functional expansion. By adding new interface definitions in the abstraction layer, the system can easily support the access of new motors while maintaining the stability of the original control strategy. This modular design concept greatly improves the maintainability and scalability of the system.

[0063] By establishing a complete feedback loop, this embodiment not only achieves high precision and high reliability of motor control, but also builds a continuously optimized learning system that can continuously improve control performance and prediction accuracy, providing an effective solution for the intelligent upgrade of the industrial automation field.

[0064] In one embodiment of the high-frequency real-time control method based on the integrated bus architecture of the present application, see Figure 3 , and can also include the following: Step S301: Establish a task manager in the solution module. The task manager adopts a hierarchical structure design, including a task scheduling layer, a task execution layer and a resource management layer. A task priority queue is established in the task scheduling layer. A unique task identifier and an initial weight coefficient are assigned to each solution task. The task identifier includes a task type code and timestamp information. The initial weight coefficient is determined based on the real-time requirements and computational complexity of the task. The creation, destruction and status management of the solution task are realized through the task manager. Step S302: Construct a task history database of a distributed data storage architecture, wherein the task history database includes a real-time data table and a historical data table. The execution information of each solution task under the current load state is recorded in the real-time data table. The execution information includes the task start time, completion time, CPU occupancy and memory usage. A load state index is established in the historical data table, and the task execution information under different load states is stored in time series. The historical data is regularly archived and cleaned up through a data compression algorithm.

[0065] Optionally, this embodiment first implements an efficient hierarchical task management architecture in the field of task management. Specifically, the task manager adopts a three-layer structure design, including a task scheduling layer, a task execution layer, and a resource management layer. Each layer communicates through a standardized interface, thereby realizing modularization and decoupling of task management.

[0066] At the task scheduling layer, this embodiment innovatively introduces a priority-based task queue management mechanism. Each solution task is assigned a unique task identifier, which consists of a task type code and timestamp information. The task type code is used to distinguish different types of solution tasks, such as trajectory planning, dynamics calculation, etc., while the timestamp information is used to track the life cycle of the task. At the same time, based on the real-time requirements and computational complexity of the task, an initial weight coefficient is assigned to each task. For example, trajectory planning tasks that require real-time response will be assigned a higher initial weight, while non-real-time tasks such as background data analysis will be assigned a lower weight.

[0067] This embodiment implements complete task lifecycle management in the task manager. Starting from the creation of a task, the task manager will allocate the necessary computing resources for it, monitor the execution status of the task, and recycle resources after the task is completed. This refined task management mechanism ensures the efficient use of system resources and avoids resource waste and task accumulation problems.

[0068] In terms of data storage, this embodiment builds a distributed task history database, using a storage structure that combines real-time data tables and historical data tables. The real-time data table mainly records the task execution information under the current load state, including key indicators such as the task start time, completion time, CPU occupancy rate, and memory usage. These real-time data provide an important basis for the system's dynamic load balancing and resource scheduling.

[0069] This embodiment innovatively establishes a load state index in the historical data table, and stores the task execution information under different load states in time series. This index mechanism greatly improves the retrieval efficiency of historical data and provides reliable data support for subsequent load prediction and performance optimization. At the same time, historical data is regularly archived and cleaned up through a data compression algorithm, effectively solving the storage space problem during long-term operation.

[0070] In order to ensure the real-time performance and reliability of the system, this embodiment establishes a concurrent data queue structure based on the cache. The read-write controller configured in this structure implements secure access control to shared resources through a mutex mechanism and conditional variables. The data synchronizer is responsible for read-write synchronization and status monitoring of queue data, ensuring reliable data transmission between the solution module and the control module.

[0071] In this embodiment, a data priority flag and a data check bit are set in the data queue. The priority flag is used to mark the priority of data processing so that the system can give priority to important control data, while the check bit ensures the correctness of data transmission through algorithms such as cyclic redundancy check.

[0072] In the implementation of the control module, this embodiment constructs a multi-channel motor control unit array and allocates an independent communication resource pool to each control unit. The communication resource pool includes an Ethernet communication interface and a fieldbus interface, and is configured with a TCP / IP protocol stack, a real-time Ethernet protocol stack, and a CAN bus protocol to meet the communication requirements of different types of motors.

[0073] This embodiment dynamically selects the most suitable communication protocol to establish a control link based on the communication characteristics of the motor. The control link includes an uplink data channel for sending control instructions and a downlink data channel for receiving status feedback, forming a complete two-way communication mechanism. This flexible communication architecture enables the system to support multiple types of motor control at the same time, significantly improving the compatibility and scalability of the system.

[0074] In practical applications, the task management and data storage architecture of this embodiment significantly improves the operating efficiency and reliability of the system. For example, in the collaborative control scenario of a multi-joint robot, the system can reasonably allocate computing resources and communication bandwidth according to the control requirements of different joints, ensure the real-time performance of key control tasks, and provide rich historical data to support the adjustment of optimized control strategies.

[0075] In one embodiment of the high-frequency real-time control method based on the integrated bus architecture of the present application, see Figure 4 , and can also include the following: Step S401: constructing a load prediction model based on a long short-term memory neural network, the load prediction model includes an input layer, a hidden layer and an output layer, a data normalization unit is set in the input layer to perform standardization processing on the real-time load data, a forget gate and a memory gate are set in the hidden layer, invalid features are filtered through the forget gate, important features are retained through the memory gate, an activation function is set in the output layer, the output of the neural network is mapped into a load prediction interval and a change trend, and the sample data in the task history database are used to train and optimize the parameters of the load prediction model; Step S402: construct a weight adjuster based on a fuzzy neural network, the weight adjuster receives the output result of the load prediction model, converts the load prediction result into a weight adjustment amount through a fuzzy rule base, the fuzzy rule base includes a load interval rule set and a trend rule set, calculates a pre-adjustment weight coefficient according to the weight adjustment amount, inputs the pre-adjustment weight coefficient into a task priority sorting module, and the task priority sorting module uses an improved quick sorting algorithm to re-prioritize the solution tasks and generate a task execution sequence.

[0076] Optionally, this embodiment builds an intelligent load prediction model based on the long short-term memory neural network (LSTM) to accurately predict the load status and change trend of the system. At the input layer, this embodiment designs a data normalization unit to standardize the load data collected in real time, and maps the original data such as CPU occupancy and memory usage to the [-1,1] interval, eliminating the impact of different dimensions on the prediction accuracy.

[0077] In the design of the hidden layer, this embodiment innovatively optimizes the gating mechanism of LSTM. The forget gate screens the input features through the sigmoid function, automatically filtering out features with low contribution to load prediction, such as instantaneous fluctuating load data. The memory gate retains feature information with predictive value, such as the periodic change pattern and long-term development trend of the load. This dual-gate structure significantly improves the model's ability to perceive load changes.

[0078] This embodiment uses a combined activation function design in the output layer to map the output of the neural network into two dimensions: load prediction interval and change trend. The load prediction interval reflects the possible range of system load in the future, while the change trend indicates the direction of load development. This multi-dimensional prediction result provides a more comprehensive decision-making basis for subsequent task scheduling.

[0079] During the model training process, this embodiment makes full use of the sample data in the task history database. By extracting task execution information under different load states, a training set containing multiple load scenarios is constructed. The training process uses the back propagation algorithm to optimize the network parameters by minimizing the prediction error and improve the generalization ability of the model.

[0080] In order to achieve dynamic adjustment of task priority, this embodiment constructs a weight adjuster based on fuzzy neural network. The adjuster receives the output result of the load prediction model and converts the prediction result into a specific weight adjustment amount through a predefined fuzzy rule base. The fuzzy rule base includes a load interval rule set and a trend rule set, which correspond to the two input dimensions of load level and change trend respectively.

[0081] This embodiment fully considers the needs of actual application scenarios in the design of fuzzy rules. For example, when the load is predicted to be on an upward trend, the rule base will appropriately reduce the weight of computationally intensive tasks and increase the priority of lightweight tasks to prevent system overload. On the contrary, when the load is at a low level and the predicted trend is stable, the rule base will tend to increase the weight of heavy tasks to improve resource utilization efficiency.

[0082] The weight adjuster calculates the pre-adjusted weight coefficient based on the fuzzy reasoning results, which takes into account the original weight of the current task and the load forecast information. The pre-adjusted weight coefficient is then input into the task priority sorting module, which uses an improved quick sorting algorithm to re-prioritize the solution tasks. The improved sorting algorithm improves the processing efficiency of approximately ordered sequences by optimizing the partitioning strategy.

[0083] The innovation of this embodiment is to achieve a deep integration of load prediction and task scheduling. The LSTM network captures the time series characteristics of load changes, and the fuzzy neural network is combined to dynamically adjust the weights, forming a closed-loop task optimization mechanism. This mechanism can predictably adjust task priorities and avoid the delay problem caused by the traditional passive response method.

[0084] In practical applications, this embodiment significantly improves the system's load balancing capability and task processing efficiency. For example, in a multi-robot collaborative operation scenario, the system can adaptively adjust the execution priority of each robot control task according to the predicted load state, ensuring timely response to key control instructions, while smoothly processing non-critical tasks to maintain stable operation of the system.

[0085] This embodiment not only solves the blindness and lag problems in traditional task scheduling, but also achieves optimal allocation of system resources through intelligent prediction and fuzzy control. This data-driven intelligent scheduling solution provides a new solution for task management of complex control systems.

[0086] In one embodiment of the high-frequency real-time control method based on the integrated bus architecture of the present application, see Figure 5 , and can also include the following: Step S501: construct a multi-level feedback task scheduler, which includes a task allocation unit and a resource monitoring unit. A plurality of priority queues are set in the task allocation unit. Each priority queue is configured with an independent time slice length and a scheduling cycle. A dynamic time slice is set based on the response time in the task history database. A task switching threshold is set based on the resource occupancy rate. The resource monitoring unit collects the system resource utilization rate in real time. When the resource utilization rate exceeds the switching threshold, the task downgrade process is triggered. Step S502: Establish a parallel computing framework in the solution module, the parallel computing framework includes a task decomposer and a result synthesizer, the task decomposer splits the solution task into multiple subtasks according to the adaptive task scheduling strategy, and allocates computing resources to process each subtask in parallel, the result synthesizer summarizes and verifies the calculation results of the subtasks, generates a final solution result, encapsulates the solution result into a data packet, and transmits the data packet to the control module through a concurrent queue mechanism.

[0087] Optionally, this embodiment constructs an efficient multi-level feedback task scheduler, which adopts dynamic time slice allocation and resource adaptive adjustment mechanism to realize intelligent scheduling of solution tasks and optimal resource allocation. Multiple priority queues are set in the task allocation unit, and each queue independently configures the time slice length and scheduling cycle, forming a hierarchical task processing structure.

[0088] This embodiment innovatively introduces a dynamic time slice allocation mechanism based on historical data. By analyzing the task response time recorded in the task history database, the average execution time and fluctuation range of different types of tasks are calculated, and the time slice length of each priority queue is dynamically adjusted. For example, for computationally intensive trajectory planning tasks, longer time slices are allocated to reduce task switching overhead; while for lightweight status monitoring tasks, shorter time slices are configured to increase response speed.

[0089] In terms of resource monitoring, this embodiment designs a refined resource monitoring unit. This unit collects system resource indicators such as CPU usage, memory occupancy, I / O load, etc. in real time, and dynamically calculates resource utilization based on these indicators. When the usage of a certain type of resource is monitored to exceed the preset switching threshold, the task downgrade processing mechanism is triggered to release system resources by reducing task priority or suspending non-critical tasks.

[0090] This embodiment adopts an innovative task decomposition strategy in the design of the parallel computing framework. The task decomposer adaptively selects the optimal task splitting scheme based on the computing characteristics of the task and the current system resource status. For parallelizable solving tasks, such as the kinematic calculation of a multi-degree-of-freedom robot, it is decomposed into multiple independent subtasks for parallel processing, which significantly improves the computing efficiency.

[0091] In the process of task decomposition, this embodiment takes into account data dependencies and computational load balancing. By analyzing the data flow graph of the task, the computational units that can be executed in parallel are identified, and the computational resources are reasonably allocated according to the complexity of each computational unit. For example, in the robot joint trajectory planning, the kinematic calculations of different joints are allocated to different processing cores, while ensuring that the coordination constraints between the joints are met.

[0092] The result synthesizer adopts a design that combines a double buffer mechanism and a verification mechanism. After the subtask calculation is completed, the result is first stored in a temporary buffer for data integrity and correctness verification. After verification, the result is transferred to the main buffer, and the status flag is updated through atomic operations to ensure data consistency. Finally, the verified calculation result is encapsulated into a standard data packet and transmitted to the control module through the concurrent queue mechanism.

[0093] This embodiment adopts a lock-free design in the implementation of concurrent queues, and ensures the safe transmission of data through atomic operations and memory barriers. Read-write pointers and status flags are set in the queue to support multi-producer and multi-consumer mode, avoiding the performance overhead caused by traditional lock mechanisms.

[0094] In practical applications, this embodiment significantly improves the computing efficiency and resource utilization of the system. For example, in a multi-robot collaborative operation scenario, the system can dynamically adjust the computing resource allocation according to the control requirements of each robot to ensure the real-time performance of key control instructions, while fully utilizing the computing power of multi-core processors through parallel computing.

[0095] When the system load changes suddenly, the multi-level feedback mechanism of this embodiment can respond quickly and maintain the stable operation of the system through task downgrading and resource reallocation. This adaptive scheduling strategy is particularly suitable for the real-time processing requirements of complex control systems and effectively solves the adaptability problem of traditional fixed scheduling schemes under dynamic loads.

[0096] This embodiment realizes intelligent scheduling and efficient utilization of computing resources through the organic combination of multi-level feedback scheduling and parallel computing framework. This adaptive scheduling solution driven by historical data provides a reliable technical solution for real-time computing of complex control systems.

[0097] In one embodiment of the high-frequency real-time control method based on the integrated bus architecture of the present application, see Figure 6 , and can also include the following: Step S601: constructing a motor object abstraction layer, the motor object abstraction layer includes a parameter description module and a function interface module, establishing a motor characteristic model in the parameter description module, the motor characteristic model includes an electrical parameter model and a mechanical parameter model, configuring a basic parameter matrix according to motor model parameters, encapsulating a standardized control instruction set in the function interface module, the control instruction set includes motion control instructions and status query instructions, and realizing unified management of motors of different models through the motor object abstraction layer; Step S602: Establish a motor adaptive controller, the adaptive controller includes a parameter matching unit and a compensation adjustment unit, the parameter matching unit receives the load prediction result, selects the optimal control parameter group from the motor control parameter library, and adaptively adjusts the control parameters according to the load state, the compensation adjustment unit corrects and compensates the solution result based on the load adaptive control parameter, adopts a feedforward compensation strategy to optimize the control performance, and sends the corrected control instruction to the motor execution unit through the integrated bus.

[0098] Optionally, this embodiment constructs a motor object abstraction layer, and realizes unified management and control of motors of different models through parameterized modeling and standardized interface design. The motor characteristic model established in the parameter description module includes an electrical parameter model and a mechanical parameter model, which fully describes the dynamic characteristics of the motor.

[0099] In the electrical parameter model, this embodiment takes into account the key parameters of the motor, such as the rated voltage, rated current, winding resistance, inductance, etc., and establishes the correlation between these parameters. The mechanical parameter model includes mechanical characteristic parameters such as rotor inertia, friction coefficient, maximum speed, and rated torque. These parameters are uniformly described by the basic parameter matrix, so that the characteristics of motors of different models can be expressed and processed under the same framework.

[0100] This embodiment designs a hierarchical control instruction set in the functional interface module. The motion control instructions include basic instructions such as position control, speed control, torque control, and compound motion control instructions. The status query instructions cover functions such as motor operation status, fault diagnosis, and parameter reading. Through a unified instruction interface, standardized control of motors of different models is achieved.

[0101] In the design of the motor adaptive controller, this embodiment innovatively introduces the load prediction result as the basis for optimizing the control parameters. The parameter matching unit selects the parameter group that best suits the current working condition from the motor control parameter library according to the predicted load state. These parameters include PID controller parameters, feedforward compensation coefficients, filter parameters, etc.

[0102] This embodiment adopts a parameter adaptive adjustment strategy based on load characteristics. When the load changes, the controller can dynamically adjust the control parameters according to the output of the load prediction model. For example, in the robot joint motor control, when the load is predicted to increase, the proportional coefficient of the torque control is automatically increased to improve the anti-interference ability of the system; when the load decreases, the control gain is appropriately reduced to avoid system oscillation.

[0103] The compensation adjustment unit adopts an innovative feedforward compensation strategy, which predictively compensates for the impact of load changes by establishing a load disturbance model. This compensation mechanism is particularly suitable for situations that require high precision, such as robot motion control. For example, when the robot end effector performs precision operations, feedforward compensation can effectively suppress the position deviation caused by load changes.

[0104] This embodiment adopts an integrated bus design for the transmission of control instructions, supporting a variety of industrial bus protocols, such as EtherCAT, PROFINET, etc. Through a unified communication interface, real-time and reliable transmission of control instructions is achieved. The bus system adopts a distributed clock synchronization mechanism to ensure the synchronization of instructions during the coordinated control of multiple motors.

[0105] In practical applications, this embodiment significantly improves the accuracy and reliability of motor control. For example, in the collaborative control of a multi-joint robot, the system can adjust the control strategy in real time according to the load state of each joint motor to ensure the smoothness and accuracy of the motion trajectory. When a joint is subjected to a sudden load change, the adaptive controller can respond quickly and maintain control performance through parameter adjustment and feedforward compensation.

[0106] The motor object abstraction layer of this embodiment simplifies the development and maintenance of the motor control system. Through unified parameter description and standardized interface, the repeated development work for different types of motors is greatly reduced. At the same time, the modular design makes the system easy to expand, and the access of new types of motors only requires the configuration of the corresponding parameter model.

[0107] Under complex working conditions, the adaptive control strategy of this embodiment shows excellent adaptability. Through load prediction and parameter adaptation, the system can maintain stable control performance under load changes. This prediction-based adaptive control scheme provides an intelligent solution for motor control systems.

[0108] This embodiment realizes the intelligentization and standardization of motor control through the organic combination of the motor object abstraction layer and the adaptive controller. This design not only improves the reliability and adaptability of the control system, but also provides an extensible technical framework for motor control in the field of industrial automation.

[0109] In one embodiment of the high-frequency real-time control method based on the integrated bus architecture of the present application, see Figure 7 , and can also include the following: Step S701: constructing a feedback signal processor, the feedback signal processor comprising a data acquisition unit and a signal analysis unit, the data acquisition unit periodically acquiring position information, speed information and current information output by a motor encoder through an integrated bus, setting a signal filter and a data verifier in the signal analysis unit, filtering high-frequency noise on the collected data through the signal filter, verifying data integrity through the data verifier, and encoding and packaging the processed feedback information according to a preset format; Step S702: Establish a data update manager, which includes a data classification module and a model optimization module. The data classification module performs time synchronization and feature extraction on the encoded feedback information and load data, and stores the processed data classifications in the corresponding data table of the task history database. The model optimization module periodically reads the newly added data samples, and uses an incremental learning algorithm to perform online fine-tuning on the network parameters of the load prediction model to achieve continuous optimization of the prediction model.

[0110] Optionally, this embodiment constructs a high-precision feedback signal processor, which realizes real-time monitoring and data optimization of the motor operating status through a multi-level data processing and verification mechanism. The data acquisition unit adopts a periodic sampling strategy to obtain key status information output by the motor encoder through an integrated bus.

[0111] In the signal acquisition process, this embodiment uses synchronous sampling technology to ensure the timing consistency of position information, speed information and current information. By configuring a reasonable sampling period, while ensuring the real-time performance of the data, the increase in system load caused by excessive sampling frequency is avoided. For example, in the robot joint control, according to the dynamic characteristics of different joints, an adaptive sampling strategy is adopted, and a higher sampling frequency is adopted for key joints, while the sampling frequency is appropriately reduced for secondary joints.

[0112] This embodiment designs an innovative filtering algorithm in the signal analysis unit. The signal filter adopts a multi-stage filtering structure, including a combination of low-pass filtering and median filtering, to effectively remove high-frequency noise and mutation interference. The filtering parameters are dynamically adjusted according to the signal characteristics. For example, for speed signals, a narrower filtering bandwidth is used in the low-speed range to improve the smoothness of the signal; the bandwidth is appropriately expanded in the high-speed range to ensure the rapid response characteristics of the signal.

[0113] The data verifier uses multiple verification mechanisms, including data format verification, range verification, and consistency verification. By setting reasonable verification rules, abnormal data can be discovered and filtered in a timely manner. For example, when it is detected that the change relationship between the position signal and the speed signal does not conform to the physical law, the exception handling mechanism is triggered to ensure the reliability of the data.

[0114] This embodiment innovatively designs a data encoding and packaging scheme, and uses compression encoding technology to reduce the amount of data transmission. According to the characteristics of different types of signals, appropriate encoding methods are selected, such as incremental encoding for position signals and fixed-point representation for speed signals, to achieve efficient data transmission.

[0115] In the data update manager, this embodiment implements intelligent data classification and model optimization. The data classification module first performs time alignment on the encoded feedback information and the load data to establish a time series association between different data sources. Through the feature extraction algorithm, key features reflecting the dynamic characteristics of the system are extracted, such as load change trends, motion pattern features, etc.

[0116] This embodiment adopts a layered architecture design for data storage. The task history database contains multiple dedicated data tables, each storing different types of data. By establishing indexes and associations, efficient query and management of data is achieved. For example, a timestamp-based index is established to support rapid retrieval of operating data for a specific time period.

[0117] The model optimization module adopts an online optimization strategy based on incremental learning. It evaluates the performance of the prediction model by regularly analyzing new data samples and fine-tunes the model parameters based on the evaluation results. This incremental learning method avoids the overhead of retraining the entire model in traditional batch learning while maintaining the prediction accuracy of the model.

[0118] In practical applications, this embodiment significantly improves the adaptability and reliability of the system. For example, during the long-term operation of an industrial robot, through continuous data accumulation and model optimization, the system can continuously improve its ability to predict load changes under different working conditions, providing a more accurate decision-making basis for adaptive control.

[0119] The data processing flow of this embodiment fully considers the actual needs of the industrial site. Through multi-level data processing and verification, practical problems such as sensor noise and data loss are effectively solved. At the same time, the incremental learning mechanism enables the system to continuously adapt to the dynamic changes in equipment performance and maintain a good control effect.

[0120] This embodiment builds a complete data processing and optimization closed loop through the collaborative work of the feedback signal processor and the data update manager. This design not only improves the operational stability of the system, but also provides strong support for equipment predictive maintenance and performance optimization.

[0121] In order to significantly improve the real-time performance and reliability of multi-motor coordinated control through adaptive task scheduling and load prediction compensation, the present application provides an embodiment of a high-frequency real-time control device based on an integrated bus architecture for implementing all or part of the high-frequency real-time control method based on an integrated bus architecture, see Figure 8 The high-frequency real-time control device based on the integrated bus architecture specifically includes the following contents: A solution module 10 is used to divide the control system into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to perform data exchange, the control module includes a plurality of motor control units, each of which is connected to a corresponding motor via an integrated bus, the integrated bus includes an industrial Ethernet bus and a field bus, the solution module includes a plurality of solution tasks, each of which is set with a corresponding task weight coefficient, and a task history database is established, wherein the task history database records the response time and resource occupancy of the motor in executing each solution task under different load conditions; A prediction module 20 is used to collect real-time load data of the motor through the integrated bus, wherein the real-time load data includes motor torque, motor speed and motor current, and input the real-time load data into a load prediction model, which is trained based on the task history database, and the load prediction model outputs a load prediction result, which includes a predicted load interval and a load change trend, and dynamically adjusts the task weight coefficient based on the load prediction result to generate a pre-adjusted weight coefficient, wherein the pre-adjusted weight coefficient changes nonlinearly with the change of the load prediction result, and prioritizes the solution tasks based on the pre-adjusted weight coefficient, and establishes an adaptive task scheduling strategy based on the priority ranking, wherein the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, and the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue; A control module 30 is used for the control module to receive the solution result, read the motor configuration file, wherein the motor configuration file includes motor model parameters, motor initialization parameters and motor control parameters, abstractly model the motor according to the motor model parameters, generate a motor abstraction layer, adaptively initialize the motor by calling the motor initialization parameters based on the motor abstraction layer, match the load prediction result with the motor control parameters, generate load adaptive control parameters, compensate and adjust the solution result according to the load adaptive control parameters, and send control instructions to the motor through the integrated bus. The control module receives the encoding feedback information of the motor through the integrated bus, and stores the encoding feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model.

[0122] From the above description, it can be seen that the high-frequency real-time control device based on the integrated bus architecture provided by the embodiment of the present application can divide the system into a solution module and a control module, realize efficient data interaction through concurrent queues, adopt the integrated architecture of industrial Ethernet bus and field bus, train the load prediction model based on historical data, and realize dynamic adjustment of task weight coefficients. Adaptive initialization of motors of different models is achieved by establishing a motor abstraction layer, and control parameters are dynamically generated based on load prediction results. This method breaks through the limitations of traditional fixed scheduling strategies, and significantly improves the real-time and reliability of multi-motor collaborative control through adaptive task scheduling and load prediction compensation.

[0123] From the hardware level, in order to significantly improve the real-time performance and reliability of multi-motor coordinated control through adaptive task scheduling and load prediction compensation, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the high-frequency real-time control method based on an integrated bus architecture, and the electronic device specifically includes the following contents: Processor, memory, communication interface and bus; wherein the processor, memory and communication interface communicate with each other through the bus; the communication interface is used to realize information transmission between the high-frequency real-time control device based on the integrated bus architecture and the core business system, user terminal and related database and other related devices; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the high-frequency real-time control method based on the integrated bus architecture and the embodiment of the high-frequency real-time control device based on the integrated bus architecture in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0124] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0125] In practical applications, part of the high-frequency real-time control method based on the integrated bus architecture can be executed on the electronic device side as described above, or all operations can be completed in the client device. The specific selection can be based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor.

[0126] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0127] Fig. 9 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig. 9As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig. 9 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0128] In one embodiment, the high-frequency real-time control method function based on the integrated bus architecture can be integrated into the central processing unit 9100. The central processing unit 9100 can be configured to perform the following control: Step S101: dividing the control system into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to perform data exchange, the control module includes a plurality of motor control units, each of the motor control units is connected to a corresponding motor through an integrated bus, the integrated bus includes an industrial Ethernet bus and a field bus, the solution module includes a plurality of solution tasks, each of the solution tasks is set with a corresponding task weight coefficient, and a task history database is established, wherein the task history database records the response time and resource occupancy of the motor in executing each solution task under different load conditions; Step S102: collecting real-time load data of the motor through the integrated bus, wherein the real-time load data includes motor torque, motor speed and motor current, inputting the real-time load data into a load prediction model, the load prediction model is obtained by training based on the task history database, the load prediction model outputs a load prediction result, the load prediction result includes a predicted load interval and a load change trend, dynamically adjusting the task weight coefficient based on the load prediction result, generating a pre-adjustment weight coefficient, the pre-adjustment weight coefficient changes nonlinearly with the change of the load prediction result, prioritizing the solution tasks based on the pre-adjustment weight coefficient, establishing an adaptive task scheduling strategy based on the priority ranking, the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue; Step S103: The control module receives the solution result, reads the motor configuration file, wherein the motor configuration file includes motor model parameters, motor initialization parameters and motor control parameters, abstractly models the motor according to the motor model parameters, generates a motor abstraction layer, adaptively initializes the motor by calling the motor initialization parameters based on the motor abstraction layer, matches the load prediction result with the motor control parameters, generates load adaptive control parameters, compensates and adjusts the solution result according to the load adaptive control parameters, and sends control instructions to the motor through the integrated bus. The control module receives the encoding feedback information of the motor through the integrated bus, and stores the encoding feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model.

[0129] From the above description, it can be seen that the electronic device provided in the embodiment of the present application divides the system into a solution module and a control module, realizes efficient data interaction through concurrent queues, adopts an integrated architecture of industrial Ethernet bus and field bus, trains load prediction model based on historical data, and realizes dynamic adjustment of task weight coefficients. Adaptive initialization of motors of different models is achieved by establishing a motor abstraction layer, and control parameters are dynamically generated based on load prediction results. This method breaks through the limitations of traditional fixed scheduling strategies, and significantly improves the real-time and reliability of multi-motor collaborative control through adaptive task scheduling and load prediction compensation.

[0130] In another embodiment, the high-frequency real-time control device based on the integrated bus architecture can be configured separately from the central processing unit 9100. For example, the high-frequency real-time control device based on the integrated bus architecture can be configured as a chip connected to the central processing unit 9100, and the function of the high-frequency real-time control method based on the integrated bus architecture can be realized through the control of the central processing unit.

[0131] like Fig. 9 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig. 9 In addition, the electronic device 9600 may also include Fig. 9 For components not shown, reference may be made to the prior art.

[0132] like Fig. 9 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0133] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0134] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0135] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0136] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0137] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0138] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module. The communication module 9110 (transmitter / receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0139] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the high-frequency real-time control method based on an integrated bus architecture in the above-mentioned embodiments, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps of the high-frequency real-time control method based on an integrated bus architecture in the above-mentioned embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented: Step S101: dividing the control system into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to perform data exchange, the control module includes a plurality of motor control units, each of the motor control units is connected to a corresponding motor through an integrated bus, the integrated bus includes an industrial Ethernet bus and a field bus, the solution module includes a plurality of solution tasks, each of the solution tasks is set with a corresponding task weight coefficient, and a task history database is established, wherein the task history database records the response time and resource occupancy of the motor in executing each solution task under different load conditions; Step S102: collecting real-time load data of the motor through the integrated bus, wherein the real-time load data includes motor torque, motor speed and motor current, inputting the real-time load data into a load prediction model, the load prediction model is obtained by training based on the task history database, the load prediction model outputs a load prediction result, the load prediction result includes a predicted load interval and a load change trend, dynamically adjusting the task weight coefficient based on the load prediction result, generating a pre-adjustment weight coefficient, the pre-adjustment weight coefficient changes nonlinearly with the change of the load prediction result, prioritizing the solution tasks based on the pre-adjustment weight coefficient, establishing an adaptive task scheduling strategy based on the priority ranking, the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue; Step S103: The control module receives the solution result, reads the motor configuration file, wherein the motor configuration file includes motor model parameters, motor initialization parameters and motor control parameters, abstractly models the motor according to the motor model parameters, generates a motor abstraction layer, adaptively initializes the motor by calling the motor initialization parameters based on the motor abstraction layer, matches the load prediction result with the motor control parameters, generates load adaptive control parameters, compensates and adjusts the solution result according to the load adaptive control parameters, and sends control instructions to the motor through the integrated bus. The control module receives the encoding feedback information of the motor through the integrated bus, and stores the encoding feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model.

[0140] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application divides the system into a solution module and a control module, realizes efficient data interaction through concurrent queues, adopts an integrated architecture of industrial Ethernet bus and field bus, trains load prediction model based on historical data, and realizes dynamic adjustment of task weight coefficients. Adaptive initialization of motors of different models is achieved by establishing a motor abstraction layer, and control parameters are dynamically generated based on load prediction results. This method breaks through the limitations of traditional fixed scheduling strategies, and significantly improves the real-time and reliability of multi-motor collaborative control through adaptive task scheduling and load prediction compensation.

[0141] The embodiments of the present application also provide a computer program product capable of implementing all the steps of the high-frequency real-time control method based on an integrated bus architecture in the above-mentioned embodiments, where the execution subject is a server or a client. When the computer program / instruction is executed by a processor, the steps of the high-frequency real-time control method based on an integrated bus architecture are implemented. For example, the computer program / instruction implements the following steps: Step S101: dividing the control system into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to perform data exchange, the control module includes a plurality of motor control units, each of the motor control units is connected to a corresponding motor through an integrated bus, the integrated bus includes an industrial Ethernet bus and a field bus, the solution module includes a plurality of solution tasks, each of the solution tasks is set with a corresponding task weight coefficient, and a task history database is established, wherein the task history database records the response time and resource occupancy of the motor in executing each solution task under different load conditions; Step S102: collecting real-time load data of the motor through the integrated bus, wherein the real-time load data includes motor torque, motor speed and motor current, inputting the real-time load data into a load prediction model, the load prediction model is obtained by training based on the task history database, the load prediction model outputs a load prediction result, the load prediction result includes a predicted load interval and a load change trend, dynamically adjusting the task weight coefficient based on the load prediction result, generating a pre-adjustment weight coefficient, the pre-adjustment weight coefficient changes nonlinearly with the change of the load prediction result, prioritizing the solution tasks based on the pre-adjustment weight coefficient, establishing an adaptive task scheduling strategy based on the priority ranking, the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue; Step S103: The control module receives the solution result, reads the motor configuration file, wherein the motor configuration file includes motor model parameters, motor initialization parameters and motor control parameters, abstractly models the motor according to the motor model parameters, generates a motor abstraction layer, adaptively initializes the motor by calling the motor initialization parameters based on the motor abstraction layer, matches the load prediction result with the motor control parameters, generates load adaptive control parameters, compensates and adjusts the solution result according to the load adaptive control parameters, and sends control instructions to the motor through the integrated bus. The control module receives the encoding feedback information of the motor through the integrated bus, and stores the encoding feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model.

[0142] From the above description, it can be seen that the computer program product provided by the embodiment of the present application divides the system into a solution module and a control module, realizes efficient data interaction through concurrent queues, adopts an integrated architecture of industrial Ethernet bus and field bus, trains load prediction model based on historical data, and realizes dynamic adjustment of task weight coefficients. Adaptive initialization of motors of different models is achieved by establishing a motor abstraction layer, and control parameters are dynamically generated based on load prediction results. This method breaks through the limitations of traditional fixed scheduling strategies, and significantly improves the real-time and reliability of multi-motor collaborative control through adaptive task scheduling and load prediction compensation.

[0143] It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0144] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0145] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0146] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0147] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A high-frequency real-time control method based on an integrated bus architecture, characterized in that: The method comprises: The control system is divided into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to perform data exchange, the control module includes a plurality of motor control units, each of which is connected to a corresponding motor through an integrated bus, the integrated bus includes an industrial Ethernet bus and a field bus, the solution module includes a plurality of solution tasks, each of which is set with a corresponding task weight coefficient, and a task history database is established, wherein the task history database records the response time and resource occupancy of the motor in executing each solution task under different load conditions; The real-time load data of the motor is collected through the integrated bus, wherein the real-time load data includes motor torque, motor speed and motor current, the real-time load data is input into a load prediction model, the load prediction model is obtained by training based on the task history database, the load prediction model outputs a load prediction result, the load prediction result includes a predicted load interval and a load change trend, the task weight coefficient is dynamically adjusted based on the load prediction result, a pre-adjustment weight coefficient is generated, the pre-adjustment weight coefficient changes nonlinearly with the change of the load prediction result, the solution tasks are prioritized based on the pre-adjustment weight coefficient, an adaptive task scheduling strategy is established according to the priority ranking, the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue; The control module receives the solution result, reads a motor configuration file, wherein the motor configuration file includes motor model parameters, motor initialization parameters and motor control parameters, abstractly models the motor according to the motor model parameters, generates a motor abstraction layer, adaptively initializes the motor by calling the motor initialization parameters based on the motor abstraction layer, matches the load prediction result with the motor control parameters, generates load adaptive control parameters, compensates and adjusts the solution result according to the load adaptive control parameters, and sends control instructions to the motor through the integrated bus. The control module receives encoding feedback information of the motor through the integrated bus, and stores the encoding feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model.

2. The high-frequency real-time control method based on an integrated bus architecture according to claim 1, characterized in that: The control system is divided into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to exchange data, the control module includes a plurality of motor control units, each of which is connected to a corresponding motor through an integrated bus, and the integrated bus includes an industrial Ethernet bus and a field bus, including: A concurrent data queue structure is established based on a high-speed cache, a read-write controller and a data synchronizer are configured in the concurrent data queue structure, the read-write controller includes a mutex mechanism and a conditional variable, the data synchronizer is responsible for read-write synchronization and status monitoring of queue data, data transmission from a solution module to a control module is realized through the concurrent data queue structure, a data priority identification bit and a data check bit are set in the data queue, the data priority identification bit is used to mark the priority of data processing, and the data check bit is used to ensure the correctness of data transmission; A multi-channel motor control unit array is constructed in the control module, and an independent communication resource pool is allocated to each motor control unit. The communication resource pool includes an Ethernet communication interface and a field bus interface. A TCP / IP protocol stack and a real-time Ethernet protocol stack are configured in the Ethernet communication interface, and a CAN bus protocol is configured in the field bus interface. A corresponding communication protocol is selected according to the motor communication characteristics to establish a control link. The control link includes an uplink data channel and a downlink data channel. The sending of motor control instructions and the receiving of status feedback are realized through the control link.

3. The high-frequency real-time control method based on an integrated bus architecture according to claim 1, characterized in that: The solution module includes a plurality of solution tasks, each of which sets a corresponding task weight coefficient, and establishes a task history database, which records the response time and resource occupation of the motor in executing each solution task under different load conditions, including: A task manager is established in the solution module. The task manager adopts a hierarchical structure design, including a task scheduling layer, a task execution layer and a resource management layer. A task priority queue is established in the task scheduling layer. A unique task identifier and an initial weight coefficient are assigned to each solution task. The task identifier includes a task type code and timestamp information. The initial weight coefficient is determined based on the real-time requirements and computational complexity of the task. The creation, destruction and status management of the solution task are realized through the task manager. A task history database of a distributed data storage architecture is constructed. The task history database includes a real-time data table and a historical data table. The execution information of each solution task under the current load state is recorded in the real-time data table. The execution information includes the task start time, completion time, CPU occupancy and memory usage. A load state index is established in the historical data table. The task execution information under different load states is stored in time series. The historical data is regularly archived and cleaned up through a data compression algorithm.

4. The high-frequency real-time control method based on an integrated bus architecture according to claim 1, characterized in that: The real-time load data is input into a load prediction model, the load prediction model is obtained by training based on the task history database, the load prediction model outputs a load prediction result, the load prediction result includes a predicted load interval and a load change trend, the task weight coefficient is dynamically adjusted based on the load prediction result to generate a pre-adjusted weight coefficient, the pre-adjusted weight coefficient changes nonlinearly with the change of the load prediction result, and the solution tasks are prioritized based on the pre-adjusted weight coefficient, including: A load prediction model is constructed based on a long short-term memory neural network, wherein the load prediction model includes an input layer, a hidden layer, and an output layer, wherein a data normalization unit is set in the input layer to perform standardization processing on real-time load data, a forget gate and a memory gate are set in the hidden layer, invalid features are filtered through the forget gate, and important features are retained through the memory gate, an activation function is set in the output layer, the output of the neural network is mapped into a load prediction interval and a change trend, and the load prediction model is trained and parameter optimized using sample data in a task history database; A weight adjuster based on a fuzzy neural network is constructed. The weight adjuster receives the output result of the load prediction model, converts the load prediction result into a weight adjustment amount through a fuzzy rule base, and the fuzzy rule base includes a load interval rule set and a trend rule set. A pre-adjustment weight coefficient is calculated according to the weight adjustment amount, and the pre-adjustment weight coefficient is input into a task priority sorting module. The task priority sorting module uses an improved quick sorting algorithm to re-prioritize the solution tasks and generate a task execution sequence.

5. The high-frequency real-time control method based on an integrated bus architecture according to claim 1, characterized in that: The adaptive task scheduling strategy is established according to the priority sorting, the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue, including: Construct a multi-level feedback task scheduler, which includes a task allocation unit and a resource monitoring unit. Multiple priority queues are set in the task allocation unit. Each priority queue is configured with an independent time slice length and scheduling cycle. A dynamic time slice is set based on the response time in the task history database. A task switching threshold is set based on the resource occupancy rate. The resource monitoring unit collects the system resource utilization rate in real time. When the resource utilization rate exceeds the switching threshold, the task downgrade process is triggered. A parallel computing framework is established in the solution module, which includes a task decomposer and a result synthesizer. The task decomposer splits the solution task into multiple subtasks according to an adaptive task scheduling strategy, and allocates computing resources to process each subtask in parallel. The result synthesizer summarizes and verifies the calculation results of the subtasks, generates a final solution result, encapsulates the solution result into a data packet, and transmits the data packet to the control module through a concurrent queue mechanism.

6. The high-frequency real-time control method based on an integrated bus architecture according to claim 1, characterized in that: The method includes performing abstract modeling on the motor according to the motor model parameters, generating a motor abstraction layer, calling the motor initialization parameters based on the motor abstraction layer to adaptively initialize the motor, matching the load prediction result with the motor control parameter, generating a load adaptive control parameter, compensating and adjusting the solution result according to the load adaptive control parameter, and sending a control instruction to the motor through the integrated bus, including: Constructing a motor object abstraction layer, the motor object abstraction layer includes a parameter description module and a function interface module, establishing a motor characteristic model in the parameter description module, the motor characteristic model includes an electrical parameter model and a mechanical parameter model, configuring a basic parameter matrix according to motor model parameters, encapsulating a standardized control instruction set in the function interface module, the control instruction set includes motion control instructions and status query instructions, and realizing unified management of motors of different models through the motor object abstraction layer; A motor adaptive controller is established, wherein the adaptive controller includes a parameter matching unit and a compensation adjustment unit. The parameter matching unit receives a load prediction result, selects an optimal control parameter group from a motor control parameter library, and adaptively adjusts the control parameters according to the load state. The compensation adjustment unit corrects and compensates the solution result based on the load adaptive control parameter, optimizes the control performance by adopting a feedforward compensation strategy, and sends the corrected control instruction to the motor execution unit through an integrated bus.

7. The high-frequency real-time control method based on an integrated bus architecture according to claim 1, characterized in that: The control module receives the encoder feedback information of the motor through the integrated bus, and stores the encoder feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model, including: Construct a feedback signal processor, the feedback signal processor includes a data acquisition unit and a signal analysis unit, the data acquisition unit periodically acquires position information, speed information and current information output by the motor encoder through an integrated bus, a signal filter and a data verifier are set in the signal analysis unit, the collected data is filtered for high-frequency noise through the signal filter, the data integrity is verified through the data verifier, and the processed feedback information is encoded and packaged according to a preset format; A data update manager is established, which includes a data classification module and a model optimization module. The data classification module performs time synchronization and feature extraction on the encoded feedback information and load data, and stores the processed data classifications in the corresponding data table of the task history database. The model optimization module periodically reads the newly added data samples, and uses an incremental learning algorithm to perform online fine-tuning on the network parameters of the load prediction model to achieve continuous optimization of the prediction model.

8. A high-frequency real-time control device based on an integrated bus architecture, characterized in that: The device comprises: A solution module is used to divide the control system into a solution module and a control module, wherein the solution module and the control module use a concurrent queue method to perform data exchange, the control module includes a plurality of motor control units, each of which is connected to a corresponding motor via an integrated bus, the integrated bus includes an industrial Ethernet bus and a field bus, the solution module includes a plurality of solution tasks, each of which is set with a corresponding task weight coefficient, and a task history database is established, wherein the task history database records the response time and resource occupancy of the motor in executing each solution task under different load conditions; A prediction module, used for collecting real-time load data of the motor through the integrated bus, wherein the real-time load data includes motor torque, motor speed and motor current, inputting the real-time load data into a load prediction model, the load prediction model is obtained by training based on the task history database, the load prediction model outputs a load prediction result, the load prediction result includes a predicted load interval and a load change trend, dynamically adjusting the task weight coefficient based on the load prediction result to generate a pre-adjusted weight coefficient, the pre-adjusted weight coefficient changes nonlinearly with the change of the load prediction result, prioritizing the solution tasks based on the pre-adjusted weight coefficient, establishing an adaptive task scheduling strategy based on the priority ranking, the adaptive task scheduling strategy is optimized based on the response time and resource occupancy in the task history database, the solution module generates a solution result according to the adaptive task scheduling strategy, and transmits the solution result to the control module through the concurrent queue; A control module is used for the control module to receive the solution result, read the motor configuration file, wherein the motor configuration file includes motor model parameters, motor initialization parameters and motor control parameters, abstractly model the motor according to the motor model parameters, generate a motor abstraction layer, adaptively initialize the motor by calling the motor initialization parameters based on the motor abstraction layer, match the load prediction result with the motor control parameters, generate load adaptive control parameters, compensate and adjust the solution result according to the load adaptive control parameters, and send control instructions to the motor through the integrated bus. The control module receives the encoding feedback information of the motor through the integrated bus, and stores the encoding feedback information and corresponding load data in the task history database for online optimization and update of the load prediction model.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the high-frequency real-time control method based on an integrated bus architecture as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the high-frequency real-time control method based on an integrated bus architecture described in any one of claims 1 to 7 are implemented.

Images