A multi-step joint debugging system and method for equipment electrical control parameters based on intelligent algorithms
Patent Information
- Application Number
- CN202611312160.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-27
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies for debugging electrical control parameters in high-end manufacturing equipment suffer from long debugging cycles, low accuracy, poor consistency, inability to handle multi-action coupling interference, lack of multi-step coupling correlation modeling and adaptive algorithms, resulting in unreliable simulation results and failure to achieve overall equipment performance optimization.
A multi-step joint debugging system for equipment electrical control parameters based on intelligent algorithms is adopted. Through parameter acquisition unit, multi-step coupled modeling unit, adaptive intelligent optimization unit, digital twin simulation unit and real-time closed-loop feedback unit, a full-element digital twin model is constructed to achieve multi-objective optimization and real-time closed-loop debugging. The coupling weight is dynamically adjusted, and global and local optimization is achieved by combining genetic algorithm and particle swarm algorithm.
It improved debugging efficiency by 50%, accuracy by 30%, and parameter consistency by 85%, significantly reduced reliance on the experience of senior engineers, improved debugging efficiency by 70%, reduced the number of on-site debugging sessions by 80%, and significantly improved equipment adaptability and stability.
Smart Images

Figure CN123113344A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent debugging technology for electrical control parameters of high-end manufacturing equipment, specifically a multi-step joint debugging system and method for equipment electrical control parameters based on intelligent algorithms. It is applicable to multi-axis collaborative high-end equipment such as optical chip die bonders, silicon photocouplers, high-precision CNC machines, and placement machines, and can realize automated, high-precision, and high-efficiency joint debugging of electrical control parameters. Background Technology
[0002] As high-end intelligent manufacturing equipment develops towards multi-axis operation, high precision, and continuous multi-motion, the difficulty of debugging equipment electrical control parameters (PID, acceleration, velocity loop, position compensation, etc.) is increasing exponentially. Traditional debugging methods rely on senior engineers manually adjusting parameters one by one and optimizing each step independently, which has drawbacks such as long debugging cycles, low precision, poor consistency, and inability to handle multi-motion coupling interference.
[0003] Currently, the closest existing technologies include parameter-by-parameter debugging methods based on engineer experience, and improved step-by-step continuous parameter testing methods developed by enterprises. Both focus on single-parameter self-tuning and independent optimization of single actions, without addressing multi-action step coupling and correlation modeling. They also fail to disclose the criteria for adaptive algorithm dimension division, digital twin model construction methods, coupling weight quantization rules, or achieve closed-loop joint debugging between simulation and real-world testing. Specific shortcomings are as follows: If only single-parameter / single-action optimization is implemented, ignoring the coupling interference between multi-step actions, the error of the preceding action will be directly transmitted to the subsequent action, making it impossible to achieve the optimal overall performance of the device. Algorithm selection is based on experience and lacks clear dimensional division criteria, resulting in poor adaptability to high-dimensional / low-dimensional scenarios and an inability to balance optimization efficiency and accuracy. The digital twin model is simply applied without disclosing the construction and calibration methods, resulting in unreliable simulation results that cannot guide on-site debugging. Key technologies such as coupling weights, parameter constraints, and fine-tuning rules are not fully disclosed, their protection scope is unclear, and they cannot be industrialized. Without a complete closed-loop debugging mechanism, simulation and actual operation are disconnected, resulting in poor parameter consistency and stability.
[0004] In existing technologies, most solutions focus on real-time control parameter optimization (such as PID self-tuning) and do not have a dedicated solution for "multi-step joint debugging", which makes it difficult to resolve the core contradiction between debugging efficiency and overall performance optimization.
[0005] Therefore, developing a debugging system and method that can automatically combine multiple actions and intelligently optimize parameters has become the key to breaking through the debugging bottleneck of high-end equipment. Summary of the Invention
[0006] To address the shortcomings of existing technologies, this invention provides a system and method for multi-step joint debugging of equipment electrical control parameters based on intelligent algorithms. This solves the problem that existing solutions mostly focus on real-time control parameter optimization and do not design a specific solution for "multi-step joint debugging," making it difficult to resolve the core contradiction between debugging efficiency and overall performance optimization.
[0007] To achieve the above objectives, the present invention provides the following technical solution: A multi-step joint debugging system for equipment electrical control parameters based on intelligent algorithms, the system comprising a parameter acquisition unit, a multi-step coupled modeling unit, an adaptive intelligent optimization unit, a digital twin simulation unit, a parameter distribution and execution unit, and a real-time closed-loop feedback unit; The parameter acquisition unit is used to acquire multi-axis electrical parameters and motion status data of the device at a high synchronous acquisition frequency of ≥100Hz. The multi-step coupling modeling unit is used to model the continuous multi-action steps of the equipment as a multi-objective optimization model with coupling relationships, based on the Pearson correlation coefficient and the neural network training to quantify the coupling weights between the steps. The adaptive intelligent optimization unit is used to automatically match the algorithm according to the dimension of the optimization variable to complete the joint optimization of multi-objective electronic control parameters; The digital twin simulation unit is used to complete the virtual verification of optimal parameters and the prediction of coupling effect based on the full-element digital twin model of the equipment; The parameter sending execution unit is used to send optimized parameters to the equipment motion controller via the industrial bus protocol; The real-time closed-loop feedback unit is used to collect on-site operating data, compare the deviation with the simulation results, and perform local fine-tuning.
[0008] Furthermore, the digital twin model of the digital twin simulation unit is constructed by mapping five-dimensional data: the three-dimensional solid model of the equipment, kinematic calibration parameters, dynamic identification parameters, electronic control constraint thresholds, and mechanical transmission characteristics. After calibration with 10-30 sets of real-world operating data, the simulation error is ≤3%.
[0009] Furthermore, the algorithm selection criteria for the adaptive intelligent optimization unit are as follows: the optimization variable dimension is divided by the number of control axes of the device, and the number of axes ≤ 8 axes is low-dimensional optimization, and a genetic algorithm is used to achieve global optimal search; For axes greater than 8, which are considered high-dimensional optimizations, the particle swarm optimization (PSO) algorithm is used to improve the convergence speed. For high-dimensional and high-precision scenarios, a fusion strategy of genetic algorithm initialization and PSO local optimization is adopted. The division is based on the matching characteristics of the multi-objective optimization variable dimension and the algorithm time complexity and optimization accuracy.
[0010] Furthermore, the coupling weights of the multi-step coupling modeling unit are initially calculated using the Pearson correlation coefficient, and then input into the BP neural network for iterative training based on historical debugging data to dynamically update the weight values. The weight adjustment step size is 0.01-0.05. The coupling weights are used to quantify the degree of influence of the preceding action error on the accuracy of the subsequent action.
[0011] Furthermore, the adaptive intelligent optimization unit has built-in parameter constraint rules: the upper limit of acceleration is set to 80% of the rated value of the equipment, and this threshold is determined based on the mechanical characteristics of the servo motor, the fatigue life of the equipment transmission components, and the standards for operational stability; the optimization process automatically filters parameter combinations that exceed the hardware constraints.
[0012] Furthermore, the parameter sending execution unit supports EtherCAT, Profinet, and Modbus-TCP industrial real-time buses, with a parameter sending delay of ≤10ms; The deviation threshold of the real-time closed-loop feedback unit is 15%. If it exceeds this threshold, it will trigger a fine-tuning of the local parameters. The fine-tuning range is ±5% to ±10% of the original parameters.
[0013] Furthermore, a multi-step joint debugging method for equipment electrical control parameters based on intelligent algorithms is provided, comprising the following steps: Step S1: High-synchronization data acquisition; The parameter acquisition unit collects multi-axis PID parameters, current loop gain, motion amplitude, and vibration data of the equipment, and simultaneously records multi-action timing correlation data, which is then processed by Kalman filtering for noise reduction. Step S2: Multi-step coupled modeling; The continuous motion of the equipment is decomposed into a multi-step sequence, the coupling relationship and constraints between the steps are quantified, and a multi-objective optimization model is constructed with the goals of minimizing positioning error, minimizing vibration, and maximizing smoothness. Step S3: Adaptive intelligent optimization; The optimization dimension is determined based on the number of axes of the equipment. The genetic algorithm / PSO algorithm / fusion algorithm is adaptively selected and iterated for 50-200 generations. The optimal combination of electronic control parameters is output with the target error ≤ ±0.001mm as the termination condition. Step S4: Digital twin simulation verification; The optimal parameters are substituted into the calibrated digital twin model for virtual operation to predict positioning accuracy and vibration amplitude. If the threshold is not met, the process returns to step S3 to adjust the algorithm parameters. Step S5: On-site closed-loop commissioning; The qualified parameters are sent to the controller, the running data is collected in real time and compared with the simulation results, and the deviation exceeds the threshold for local fine-tuning, thus completing multi-step joint debugging.
[0014] Furthermore, the dynamic weights of the multi-objective optimization model in step S3 can be customized, with the sum of the weights of each objective being 1, the positioning error weight being ≥0.4, the vibration amplitude weight being ≥0.2, the running smoothness weight being ≥0.2, and the debugging efficiency weight being ≤0.2.
[0015] Furthermore, the algorithm adjustment rules in step S4 when the simulation does not meet the threshold are as follows: increase the number of iterations by 20%-50%, adjust the crossover probability by 0.1-0.2, and adjust the particle swarm learning factor by 0.05-0.1 until the simulation results meet the standards.
[0016] Furthermore, the termination conditions for the adaptive intelligent optimization also include: the rate of change of the objective function value for 10 consecutive generations is ≤0.5%, or the maximum number of iterations is reached, and parameter combinations that exceed the hardware limitations of the device are automatically filtered during the optimization process, such as acceleration ≤80% of the device's rated value.
[0017] This invention provides a multi-step joint debugging system and method for equipment electrical control parameters based on intelligent algorithms. It has the following beneficial effects: 1. This invention provides a multi-step joint debugging system and method for equipment electrical control parameters based on intelligent algorithms. It models multiple continuous action steps of the equipment as a coupled and related multi-objective optimization model, quantifies the error transmission weight between steps, and completely solves the pain points of existing technologies such as single-step independent optimization and inability to eliminate coupling interference. It also divides the applicable scenarios of the algorithm based on the number of control axes, clarifies the scientific basis of the 8-axis division, and achieves low-dimensional global optimization and high-dimensional fast convergence. Compared with existing technologies with fixed algorithms, the optimization efficiency is improved by 50% and the accuracy is improved by 30%.
[0018] 2. This invention provides a multi-step joint debugging system and method for equipment electrical control parameters based on intelligent algorithms. It discloses a method for constructing and calibrating a full-element digital twin model, realizing error closed-loop between virtual simulation and actual operation, filtering invalid parameters in advance, reducing the number of actual debugging times by 80%, and using Pearson correlation coefficient + BP neural network to dynamically calculate coupling weights instead of empirical assignment. The weight adjustment is precise and quantifiable, ensuring the scientific nature of multi-step joint optimization, clarifying the basis for the mechanical life and operational stability of equipment with acceleration ≤80% of the rated value, standardizing parameter constraints, and avoiding equipment overload damage. Existing technologies do not have clear constraint basis.
[0019] 3. This invention provides a multi-step joint debugging system and method for device electrical control parameters based on intelligent algorithms. It adopts a modular design and can quickly adapt to different controllers and device types through configuration files without modifying the core algorithm, thus reducing secondary development costs. Through data-driven and intelligent algorithm automatic optimization, it significantly reduces the dependence on the experience of senior engineers, improves debugging efficiency by more than 70%, and improves parameter consistency by more than 85%. Attached Figure Description
[0020] Figure 1 This is a schematic diagram of the overall architecture of the intelligent algorithm-based multi-step joint debugging system for equipment electrical control parameters of the present invention; Figure 2 This is a flowchart of the multi-step joint debugging method for equipment electrical control parameters based on intelligent algorithms according to the present invention; Figure 3 This is a schematic diagram of the multi-step action coupling modeling and weight quantization of the present invention; Figure 4 This is a schematic diagram of the digital twin model construction and calibration process of the present invention; Figure 5 This is a schematic diagram of the adaptive intelligent algorithm selection logic of the present invention. Detailed Implementation
[0021] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.
[0022] In the description of this invention, unless otherwise explicitly specified and limited, the terms "connected," "linked," and "fixed" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0023] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.
[0024] In the description of this embodiment, the terms "upper," "lower," "right," etc., refer to the orientation or positional relationship shown in the accompanying drawings. They are used only for ease of description and simplification of operation, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention. In addition, the terms "first" and "second" are used only for distinction in description and have no special meaning.
[0025] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.
[0026] like Figures 1-5 As shown, this embodiment of the invention provides a multi-step joint debugging system for equipment electrical control parameters based on intelligent algorithms. The system includes a parameter acquisition unit, a multi-step coupled modeling unit, an adaptive intelligent optimization unit, a digital twin simulation unit, a parameter distribution and execution unit, and a real-time closed-loop feedback unit. The parameter acquisition unit is used to acquire multi-axis electrical parameters and motion status data of the high-synchronization equipment. It consists of a high-precision sensor, a data acquisition card, and a synchronization clock module. It acquires data such as servo PID, current loop gain, speed loop bandwidth, position error, acceleration, and vibration amplitude. The sampling frequency is ≥100Hz, the multi-axis data synchronization error is ≤1ms, and outliers are removed by Kalman filtering.
[0027] The multi-step coupling modeling unit is used to model the continuous multi-action steps of the equipment as a multi-objective optimization model with coupling relationships, based on the Pearson correlation coefficient and the neural network training to quantify the coupling weights between the steps. The continuous operation of the equipment is broken down into steps such as material handling, alignment, processing, and resetting. The initial coupling weight is calculated based on the Pearson correlation coefficient, and the calculation formula is as follows: ; In the formula: is the coupling correlation coefficient; n is the number of historical debugging samples; These are parameters for the preceding action; This is for errors in subsequent actions; , These are the sample means of the parameters and the errors, respectively. The coupling weight is obtained by taking the absolute value of the correlation coefficient. The weight values range from [0, 1]. The larger the weight, the stronger the coupling interference between the two actions. The coupling weights are then input into the BP neural network, and the network is iteratively trained based on historical debugging data. The weight values are dynamically updated, and the weight adjustment step size is 0.01-0.05.
[0028] Then, a backpropagation (BP) neural network is trained based on historical data, dynamically updating the coupling weights to quantify the impact of preceding actions on subsequent actions, thus constructing a multi-objective optimization model. The calculation formula is as follows: ; Constraints: ; In the formula: =Positioning error, =Vibration amplitude, =Running smoothness, a =Acceleration, a e =Rated acceleration, For dynamic weights.
[0029] The adaptive intelligent optimization unit is used to automatically match the algorithm according to the dimension of the optimization variable to complete the joint optimization of multi-objective electronic control parameters; Algorithm selection criteria: Optimization variable dimension = number of axes × number of action steps; Low-dimensional optimization (number of axes ≤ 8): Genetic algorithm, strong global optimization ability, suitable for scenarios with few variables; High-dimensional optimization (number of axes > 8): PSO algorithm, fast convergence speed, suitable for scenarios with many variables; High-precision scenarios: Genetic algorithm initialization + PSO local optimization fusion, taking into account both global and local optima.
[0030] Optimization termination condition: the target change rate is ≤0.5% for 10 consecutive generations, or the positioning error is ≤±0.001mm.
[0031] The digital twin simulation unit is used to complete the virtual verification of optimal parameters and the prediction of coupling effects based on the full-element digital twin model of the equipment; The model building steps are as follows: ① Import the equipment's 3D CAD model; ② Calibrate kinematic parameters (stroke, clearance, transmission ratio); ③ Identify dynamic parameters (inertia, damping, stiffness); ④ Mapping electronic control constraints (rated speed, acceleration, current limit); ⑤ Import 10-30 sets of field data for calibration, ensuring the simulation error is ≤3%; Simulation verification thresholds: positioning error ≤ 0.005 mm, vibration amplitude ≤ 0.1 g.
[0032] The parameter sending execution unit is used to send optimized parameters to the device motion controller via industrial bus protocols. It supports industrial real-time buses such as EtherCAT, Profinet, and Modbus-TCP, with a sending latency of ≤10ms and supports batch multi-device synchronous updates.
[0033] The real-time closed-loop feedback unit is used to collect on-site operation data, compare the deviation with the simulation results, and perform local fine-tuning. On-site data is collected every 5-10 action cycles and compared with the simulation results. If the deviation is greater than 15%, fine-tuning is triggered. The fine-tuning range is ±5% to ±10%, and there is no need to re-optimize the entire process.
[0034] This debugging method includes the following core steps: Step S1: High-synchronization data acquisition; Device connection: Establish communication between the parameter acquisition module and the device controller and sensors, configure the sampling frequency to 100-500Hz, and the parameter type to be collected as electrical parameters + motion status data; Reference data acquisition: The control equipment completes multiple steps according to the preset process path, and collects data such as parameter reference values, positioning errors, and vibration amplitudes for each step to establish an original database; Data preprocessing: The collected data is filtered and denoised (using Kalman filtering), outlier removal is performed, and time series data from each step is synchronized to provide reliable input for modeling.
[0035] Step S2: Multi-step coupled modeling; Action breakdown: Break down the complete process flow of the equipment into steps of continuous action. For example, the steps of a die bonder are: die picking → alignment → mounting → reset. Clarify the electrical control parameters of each step, such as PID parameters, speed limits, and acceleration. Coupling relationship quantification: The degree of coupling between parameters in each step is quantified through correlation analysis, such as the correlation weight between crystal extraction speed and alignment error, to construct a multi-objective optimization model. The objective function is:
[0036] in, For positioning error, The amplitude of the vibration. For smooth operation, Dynamic weights; Constraint settings: Based on the hardware limits of the equipment and process requirements, set parameter constraints, such as speed ≤3m / s, acceleration ≤5m / s², and PID parameter range.
[0037] Step S3: Adaptive intelligent optimization; Algorithm selection: The system automatically selects or merges optimization algorithms based on the number of device axes and motion complexity. For devices with ≤8 axes, it is a low-dimensional optimization, and a genetic algorithm is selected to achieve global optimality. For devices with >8 axes, it is a high-dimensional optimization, and a particle swarm optimization (PSO) algorithm is selected to improve convergence speed. Iterative optimization: Initialize the parameter population, including parameters such as PID, velocity, and acceleration, and iteratively calculate according to the objective function. In each generation, retain the best 20% of individuals, and generate a new generation population through crossover and mutation. Termination Judgment: The optimization will terminate if any of the following conditions are met: the rate of change of the objective function is ≤0.5% for 10 consecutive generations, the positioning error is ≤0.003mm, or the maximum number of iterations of 200 generations is reached, and the optimal parameter combination is output.
[0038] Step S4: Digital twin simulation verification; Virtual operation: Substitute the optimal parameters into the digital twin model of the equipment to simulate the complete process of multi-step actions, and run it ≥3 times; Result evaluation: Analyze the positioning error, vibration amplitude, and running smoothness in the simulation results. If all of them meet the preset threshold (e.g., vibration ≤ 0.08g), then proceed to the parameter distribution stage. If the conditions are not met, adjust the algorithm parameters (such as increasing the number of iterations) and return to step three to re-optimize.
[0039] Step S5: On-site closed-loop commissioning Parameter distribution: Optimized parameters are automatically distributed to the device controller via a preset communication protocol, and the distribution time and version are recorded. In actual operation: The control equipment completes multiple actions according to optimized parameters, and the real-time feedback module collects operational data; Closed-loop fine-tuning: Compare the actual data with the simulation results. If the deviation is ≤15%, the debugging is complete; if the deviation is >15%, local parameter fine-tuning (such as adjusting the PID proportional coefficient) is triggered, without the need for remodeling and full-process optimization.
[0040] Implementation Case: Multi-step joint debugging of a 6-axis optical chip die bonder; ① The system configuration is as follows: Hardware configuration: Parameter acquisition module (sampling frequency 200Hz, supports current, voltage and position error acquisition), industrial computer (CPU Intel i7-12700H, memory 16GB), communication module (supports EtherCAT / Profinet protocol), equipment digital twin simulation platform; Software configuration: Intelligent optimization algorithm library (genetic algorithm, PSO, simulated annealing algorithm), multi-step action modeling tool, data acquisition and analysis software, parameter distribution and control software; Debugging object: A certain type of optical chip die bonder, 6-axis coordinated, the action steps are: die picking → alignment → placement → reset, the positioning accuracy requirement is ≤0.005mm; Adaptive intelligent optimization: For 6-axis to 8-axis systems, a genetic algorithm is used, iterating for 150 generations; Digital twin model: calibrated with 20 sets of real-world data, simulation error is 2.7%; Parameter delivery: EtherCAT bus, delay 6ms; Closed-loop feedback: deviation threshold 15%, fine-tuning range ±8%.
[0041] ②The debugging and implementation steps are as follows: Step 1: Data Collection; Connect the acquisition module to the die bonder controller, set the sampling frequency to 200Hz, and collect data such as PID parameters, position error, and vibration amplitude for each step of die picking, alignment, and mounting. After preprocessing, establish the original database. Step 2: Multi-step motion modeling; The four consecutive action steps were broken down, and the coupling weight of the crystal picking speed and the alignment error was quantified to 0.3. A multi-objective optimization model was constructed, and the constraints were set as follows: speed ≤ 2m / s, acceleration ≤ 4m / s², and positioning error ≤ 0.005mm. Step 3: Intelligent joint optimization; The system automatically selects a genetic algorithm, sets the number of iterations to 150 generations, the crossover probability to 0.7, the mutation probability to 0.08, and the termination condition is met when the iteration reaches 120 generations. The positioning error is 0.0028mm, and the optimal parameter combination is output. Step 4: Simulation verification; The digital twin model was used for virtual operation 5 times, with an average positioning error of 0.003mm and a vibration amplitude of 0.07g, which met the preset threshold. Step 5: Parameter distribution and real-time feedback; Parameters are sent via the EtherCAT protocol, and data is collected in real time during actual operation. The deviation from the simulation results is 10%, which does not require fine-tuning and the debugging is complete. Implementation effect
[0042] Debugging efficiency: Reduced from the traditional 3 days to 1.5 hours, an efficiency improvement of 75%; Adjustment accuracy: Positioning error decreased from 0.0047mm to 0.0028mm, a reduction of 40.4%; vibration amplitude decreased from 0.11g to 0.07g, a reduction of 36.4%. Parameter consistency: When the system was debugged by three engineers at different levels, the parameter deviation was ≤3%, which is an effective improvement of 83.3% compared to the 18% deviation of traditional methods. Equipment compatibility: Successfully adapted to the debugging of an 8-axis silicon photocoupler, reducing debugging time by 72% and positioning error by 38%.
[0043] The following points should be noted in this article: 1. The accompanying drawings of the embodiments disclosed herein only relate to the structures involved in the embodiments disclosed herein; other structures can be referred to in general design.
[0044] 2. Where there is no conflict, the embodiments of this disclosure and the features in the embodiments can be combined with each other to obtain new embodiments.
[0045] Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.
Claims
1. A multi-step joint debugging system for electric control parameters of a device based on an intelligent algorithm, characterized in that, The system includes a parameter acquisition unit, a multi-step coupled modeling unit, an adaptive intelligent optimization unit, a digital twin simulation unit, a parameter distribution and execution unit, and a real-time closed-loop feedback unit. The parameter acquisition unit is used to acquire multi-axis electrical parameters and motion status data of the device at a high synchronous acquisition frequency of ≥100Hz. The multi-step coupling modeling unit is used to model the continuous multi-action steps of the equipment as a multi-objective optimization model with coupling relationships, based on the Pearson correlation coefficient and the neural network training to quantify the coupling weights between the steps. The adaptive intelligent optimization unit is used to automatically match the algorithm according to the dimension of the optimization variable to complete the joint optimization of multi-objective electronic control parameters; The digital twin simulation unit is used to complete the virtual verification of optimal parameters and the prediction of coupling effects based on the full-element digital twin model of the equipment. The parameter sending execution unit is used to send optimized parameters to the equipment motion controller via the industrial bus protocol; The real-time closed-loop feedback unit is used to collect on-site operating data, compare the deviation with the simulation results, and perform local fine-tuning. 2.The intelligent algorithm-based multi-step joint debugging system for electric control parameters of a device according to claim 1, characterized in that, The digital twin model of the digital twin simulation unit is constructed by mapping five-dimensional data, including the three-dimensional solid model of the equipment, kinematic calibration parameters, dynamic identification parameters, electronic control constraint thresholds, and mechanical transmission characteristics. After calibration with 10-30 sets of real-world operating data, the simulation error is ≤3%. 3.The intelligent algorithm-based multi-step joint debugging system for electric control parameters of a device according to claim 1, characterized in that, The algorithm selection criteria for the adaptive intelligent optimization unit are as follows: the optimization variable dimension is divided according to the number of control axes of the equipment. The number of axes ≤ 8 is low-dimensional optimization, and the genetic algorithm is used to achieve global optimal search. For axes greater than 8, which are considered high-dimensional optimizations, the particle swarm optimization (PSO) algorithm is used to improve the convergence speed. For high-dimensional and high-precision scenarios, a fusion strategy of genetic algorithm initialization and PSO local optimization is adopted. The division is based on the matching characteristics of the multi-objective optimization variable dimension and the algorithm time complexity and optimization accuracy.
4. The intelligent algorithm-based multi-step joint debugging system for electric control parameters of a device according to claim 1, characterized in that, The coupling weights of the multi-step coupling modeling unit are initially calculated using the Pearson correlation coefficient, and then input into the BP neural network for iterative training based on historical debugging data to dynamically update the weight values. The weight adjustment step size is 0.01-0.
05. The coupling weights are used to quantify the degree of influence of the error of the preceding action on the accuracy of the subsequent action. 5.The intelligent algorithm-based multi-step joint debugging system for electric control parameters of a device according to claim 1, characterized in that, The adaptive intelligent optimization unit has built-in parameter constraint rules: the upper limit of acceleration is set to 80% of the rated value of the equipment. This threshold is determined based on the mechanical characteristics of the servo motor, the fatigue life of the equipment transmission components, and the standards for operational stability; the optimization process automatically filters parameter combinations that exceed the hardware constraints.
6. The multi-step joint debugging system for equipment electrical control parameters based on intelligent algorithms according to claim 1, characterized in that, The parameter sending execution unit supports EtherCAT, Profinet, and Modbus-TCP industrial real-time buses, with a parameter sending delay of ≤10ms. The deviation threshold of the real-time closed-loop feedback unit is 15%. If it exceeds this threshold, it will trigger a fine-tuning of the local parameters. The fine-tuning range is ±5% to ±10% of the original parameters.
7. A multi-step joint debugging method for equipment electrical control parameters based on intelligent algorithms, characterized in that, The method, applied to the system of any one of claims 1-6, includes the following steps: Step S1: High-Synchronization Data Acquisition The parameter acquisition unit collects multi-axis PID parameters, current loop gain, motion amplitude, and vibration data of the equipment, and simultaneously records multi-action timing correlation data, which is then processed by Kalman filtering for noise reduction. Step S2: Multi-step coupling modeling The continuous motion of the equipment is decomposed into a multi-step sequence, the coupling relationship and constraints between the steps are quantified, and a multi-objective optimization model is constructed with the goals of minimizing positioning error, minimizing vibration, and maximizing smoothness. Step S3: Adaptive Intelligent Optimization The optimization dimension is determined based on the number of axes of the equipment. The genetic algorithm / PSO algorithm / fusion algorithm is adaptively selected and iterated for 50-200 generations. The optimal combination of electronic control parameters is output with the target error ≤ ±0.001mm as the termination condition. Step S4: Digital Twin Simulation Verification The optimal parameters are substituted into the calibrated digital twin model for virtual operation to predict positioning accuracy and vibration amplitude. If the threshold is not met, the process returns to step S3 to adjust the algorithm parameters. Step S5: On-site closed-loop commissioning The qualified parameters are sent to the controller, the running data is collected in real time and compared with the simulation results, and the deviation exceeds the threshold for local fine-tuning, thus completing multi-step joint debugging.
8. The system and method for multi-step joint debugging of equipment electrical control parameters based on intelligent algorithms according to claim 7, characterized in that, In step S3, the dynamic weights of the multi-objective optimization model can be customized. The sum of the weights of each objective is 1, the weight of positioning error is ≥0.4, the weight of vibration amplitude is ≥0.2, the weight of running smoothness is ≥0.2, and the weight of debugging efficiency is ≤0.
2.
9. The system and method for multi-step joint debugging of equipment electrical control parameters based on intelligent algorithms according to claim 7, characterized in that, The algorithm adjustment rules in step S4 when the simulation does not meet the threshold are as follows: increase the number of iterations by 20%-50%, adjust the crossover probability by 0.1-0.2, and adjust the particle swarm learning factor by 0.05-0.1 until the simulation results meet the standards.
10. The multi-step joint debugging system and method for equipment electrical control parameters based on intelligent algorithms according to claim 7, characterized in that, The termination conditions for the adaptive intelligent optimization also include: the rate of change of the objective function value for 10 consecutive generations is ≤0.5%, or the maximum number of iterations is reached, and parameter combinations that exceed the hardware limitations of the device are automatically filtered during the optimization process.
