Self-adaptive regulation and control method and equipment for press fitting process parameters of motor coil

Abstract

The invention discloses a motor coil press fitting process parameter self-adaptive regulation and control method and device, and relates to the technical field of motor vehicle motor manufacturing. The method comprises the steps that press fitting target parameters including a target termination position and target pressing force are obtained, press fitting is started, actual pressure and displacement values are collected in real time, a process state deduction model is constructed, whether double targets can be met at the same time when press fitting is finished in a current control mode or not is predicted, and an analysis result is output. If the conditions cannot be met at the same time, the predicted pressure deviation value and the displacement deviation value are taken as optimization targets, a process state compensation value is introduced, a multi-target optimization function is constructed to solve a control target weighting coefficient, a press fitting tail-section hybrid controller is reconstructed, output instructions of a position ring and a force ring are fused, and press fitting is executed. According to the invention, synchronous high-precision control of the pressing force and the termination position under material fluctuation and environmental interference is realized, and the adaptability and stability of coil press-fitting processes of various motors such as a new energy automobile driving motor and a household appliance motor are improved.

Description

Technical Field

[0001] This invention relates to the field of motor manufacturing technology, specifically to an adaptive control method and equipment for motor coil press-fitting process parameters. Background Technology

[0002] Press-fitting motor coils is a critical step in motor manufacturing, and its quality directly affects the motor's electrical performance and mechanical reliability. In high-performance applications, such as the flat-wire stator coils used in new energy vehicle drive motors or the round-wire stator coils commonly found in household appliance motors, the press-fitting process faces even more stringent requirements. The press-fitting process not only requires precise control of the axial press-fitting termination position of the coils to ensure that the geometric dimensions meet design requirements, but also necessitates the application of appropriate clamping force to guarantee the insulation reliability between coils and the overall structural stability.

[0003] However, existing press-fitting control methods are mostly based on preset programs or single feedback loops, such as switching control between position loops and force loops or fixed priority control strategies. These methods lack robustness when facing dynamic disturbances such as fluctuations in coil insulation material properties, changes in ambient temperature, and drift in the press-fitting equipment itself. Specifically, towards the end of the press-fitting process, the actual process results often fail to converge simultaneously to the preset target clamping force and target termination position, leading to process deviations where the force meets the target but the position is out of tolerance, or vice versa. Existing technologies fail to fully consider the impact of real-time state variables such as temperature on material deformation behavior and lack a multi-objective dynamic coordination mechanism based on process trend prediction. This results in limited process stability and makes it difficult to meet the stringent requirements for consistency, reliability, and high yield in the production of various types of motors. Summary of the Invention

[0004] This invention addresses the technical problem in existing technologies where dynamic interferences such as batch differences in materials, fluctuations in ambient temperature, and mold wear make it difficult to simultaneously and accurately meet the preset target pressing force and target termination position during the motor coil pressing process. It provides an adaptive control method and equipment for motor coil pressing process parameters.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] In a first aspect, the present invention provides an adaptive control method for motor coil press-fitting process parameters, comprising:

[0007] Obtain the press-fit target parameters of the target stator coil, wherein the press-fit target parameters include a preset target termination position and a target clamping force;

[0008] Start the pressing process and collect the actual pressure and displacement values ​​in real time during the current pressing process to generate a pressure-displacement monitoring sequence.

[0009] Based on the pressure-displacement monitoring sequence, a process state deduction model is constructed to predict whether the target termination position and the target clamping force are simultaneously satisfied when pressing continues until the end of pressing under the current control mode, and the trend analysis results are output.

[0010] If the trend analysis results show that they cannot be satisfied at the same time, then the predicted pressure deviation and the predicted displacement deviation are used as optimization objectives, and the real-time collected pressure head temperature and coil temperature are introduced as process state compensation quantities. A multi-objective optimization function containing weighted allocation variables is constructed, and the control objective bias coefficient is solved.

[0011] Based on the control target bias coefficient, the hybrid controller of the final pressing stage is reconstructed, the output commands of the position loop and force loop are dynamically weighted and fused, and based on the fused unified command, the pressing control of the target stator coil is continued.

[0012] Secondly, the present invention provides an adaptive control device for motor coil press-fitting process parameters, comprising:

[0013] The target parameter acquisition module is used to acquire the press-fit target parameters of the target stator coil, wherein the press-fit target parameters include a preset target termination position and a target clamping force;

[0014] The data acquisition and sequence generation module is used to start the pressing process, collect the actual pressure value and actual displacement value in real time during the current pressing process, and generate a pressure-displacement monitoring sequence.

[0015] The deduction and trend analysis module is used to construct a process state deduction model based on the pressure-displacement monitoring sequence, predict whether the target termination position and the target clamping force are simultaneously satisfied when pressing continues until the end of pressing under the current control mode, and output the trend analysis results.

[0016] The multi-objective optimization and solution module is used to construct a multi-objective optimization function containing weighted allocation variables when the trend analysis results show that they cannot be satisfied simultaneously. It takes the predicted pressure deviation and the predicted displacement deviation as optimization objectives, introduces the real-time collected pressure head temperature and coil temperature as process state compensation quantities, and solves the control objective bias coefficient.

[0017] The hybrid control and execution module is used to reconstruct the hybrid controller of the final pressing stage based on the control target bias coefficient, dynamically weight and fuse the output commands of the position loop and force loop, and continue to complete the pressing control of the target stator coil based on the fused unified command.

[0018] The beneficial effects of this invention are:

[0019] Compared to existing technologies, this invention firstly predicts the pressing trend online by constructing a process state deduction model, enabling early identification of risks to achieving the target during the pressing process. Secondly, it introduces the temperature of the pressure head and coil as real-time compensation quantities, and solves the control target bias coefficient by establishing a multi-objective optimization function, allowing the control strategy to dynamically adapt to changes in material properties and environmental disturbances. Thirdly, based on the solved bias coefficient, the hybrid controller is reconstructed to achieve dynamic weighted fusion of position loop and force loop commands, thereby enabling precise coordinated control at the end of the pressing process. In summary, this invention enables the pressing process to reliably and synchronously achieve the preset target pressing force and target termination position even under dynamic disturbances, improving the stability, control accuracy, and product consistency of the pressing process for various motor coils, such as those used in new energy vehicle drive motors and household appliance motors. Attached Figure Description

[0020] Figure 1 A flowchart illustrating an adaptive control method for motor coil press-fitting process parameters provided by the present invention;

[0021] Figure 2 This invention provides a schematic diagram of the structure of an adaptive control device for motor coil press-fitting process parameters;

[0022] Figure 3 This is a physical image of the motor coil pressing equipment provided by the present invention.

[0023] In the attached diagram, the components represented by each number are as follows:

[0024] The module includes: target parameter acquisition module 11, data acquisition and sequence generation module 12, deduction and trend analysis module 13, multi-objective optimization and solution module 14, and hybrid control and execution module 15. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0027] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0028] Example 1, as Figure 1 As shown, this embodiment of the invention provides an adaptive control method for motor coil press-fitting process parameters, including:

[0029] S10: Obtain the press-fit target parameters of the target stator coil, wherein the press-fit target parameters include a preset target termination position and a target clamping force;

[0030] First, the press-fit target parameters for the target stator coil are obtained. The target stator coil refers to a specific motor stator winding assembly to be press-fitted, such as the flat-wire stator coil used in new energy vehicle drive motors or the round-wire stator coil used in household appliance motors. The press-fit target parameters for the target stator coil are a set of pre-set specific process indicators expected to be achieved at the end of the press-fitting process, including the preset target termination position and target clamping force. Specifically, the target termination position defines the final axial geometric position that the stator coil needs to achieve in the press-fitting direction. Obtaining this target termination position ensures that the coil is pressed to the accurate depth specified in the design, which is related to the neatness of the coil arrangement and the control of the slot fill factor within the motor stator slots. The target clamping force defines the final axial mechanical load that the press-fitting device should apply to the stator coil when the target termination position is reached. Obtaining this target clamping force ensures that each turn of the coil receives sufficient clamping force and contact pressure, which is crucial for ensuring the insulation reliability, mechanical stability, and subsequent impregnation process effectiveness between coils.

[0031] Obtaining the press-fit target parameters of the target stator coil can provide a clear comparison standard and optimization direction for subsequent real-time monitoring, trend prediction and control decisions.

[0032] S20: Start the pressing process, collect the actual pressure value and actual displacement value in real time during the current pressing process, and generate a pressure-displacement monitoring sequence;

[0033] Secondly, the pressing process is initiated, and the actual pressure and displacement values ​​during the pressing process are collected in real time. Specifically, after the pressing device begins to apply force to the target stator coil, the actual axial pressure value of the pressing head acting on the coil is continuously measured and recorded by a force sensor installed on the pressing actuator. At the same time, the actual axial displacement value of the pressing head or related actuators is continuously measured and recorded by a displacement measuring device.

[0034] The collected actual pressure and displacement values ​​are strictly synchronized in time, forming a data pair reflecting the instantaneous state of the pressing process. As the pressing process continues, the continuous data pairs arranged in chronological order form a pressure-displacement monitoring sequence. This pressure-displacement monitoring sequence digitally records the dynamic relationship between mechanical load and geometric position over time during the pressing process, providing a real-time and accurate data foundation for subsequent process analysis, model building, and decision control.

[0035] S30: Based on the pressure-displacement monitoring sequence, construct a process state deduction model, predict whether the target termination position and the target clamping force are simultaneously satisfied when pressing continues until the end of pressing under the current control mode, and output the trend analysis results;

[0036] Furthermore, based on the pressure-displacement monitoring sequence, a process state prediction model is constructed. This process state prediction model is a dynamic prediction model based on joint calculation of real-time process data and historical process knowledge. It maps the future trajectory of the pressure and displacement state of the pressing process over time under a given control strategy and current operating conditions. It can be used to predict in advance whether the pressing process can converge to the preset target termination position and target clamping force at the end if the current control mode remains unchanged.

[0037] Specifically, the construction steps of the process state deduction model include:

[0038] Based on the historical press-fitting process database, the standard pressure-displacement reference curve under standard working conditions is extracted, and combined with the real-time change trend of the pressure-displacement monitoring sequence, the basic deduction logic of the process state deduction model is constructed.

[0039] Configure the deduction termination conditions and accuracy verification mechanism of the process state deduction model, and set the prediction time step and calculation depth of the deduction process according to the target termination position and the target clamping force;

[0040] The pressure head temperature and coil temperature are introduced as real-time correction variables. The basic deduction logic is compensated online by a preset temperature-material deformation relationship to generate a process state deduction model suitable for the current pressing process.

[0041] First, based on the historical press-fitting process database, standard pressure-displacement reference curves recorded under standard operating conditions are extracted. This historical press-fitting process database is a component of the enterprise's Manufacturing Execution System (MES), and historical press-fitting process data can be obtained through a query interface, providing data support for the establishment of standard reference curves and model training. These standard pressure-displacement reference curves represent the typical correspondence between pressure and displacement under ideal material, stable environment, and wear-free equipment conditions. Simultaneously, by combining the real-time change trend of the current press-fitting process reflected by the pressure-displacement monitoring sequence, the real-time data is compared and fused with the historical reference curves, thereby constructing the basic deduction logic for the process state extrapolation model.

[0042] Specifically, the fundamental deduction logic is a hybrid prediction framework that integrates data-driven methods and physical mechanism models. First, by aligning and comparing the real-time acquired pressure-displacement monitoring sequence with historical standard pressure-displacement reference curves at the same displacement or pressure stage, the real-time offset or deviation rate of the current process relative to the standard operating condition is calculated. Second, the slope and curvature of the monitoring sequence within the most recent time window are analyzed to identify whether the current pressing process is in the linear compression stage, the material yielding transition stage, or the densification stage, and its dynamic characteristics are determined. Then, based on the deviation and stage characteristics identified through comparison, the kernel of the calculation model used for state deduction is adaptively selected or adjusted. This kernel may include a time series prediction part based on an autoregressive integral moving average model and a physical model approximation part based on the mechanical constitutive relationship of the pressing process. Finally, the selected calculation model kernel is combined with the latest state of the current monitoring sequence as initial conditions to form the fundamental deduction logic capable of iteratively calculating the predicted pressure and displacement values ​​at multiple future time steps, starting from the current time point and based on the input control mode command sequence. In summary, the basic deduction logic establishes the fundamental calculation rules and relational mappings for predicting future state changes starting from the current state.

[0043] Secondly, the process state simulation model is configured with simulation termination conditions and an accuracy verification mechanism. The simulation termination condition uses the target termination position and target clamping force as a joint judgment benchmark to determine when the process state simulation model stops forward calculations and predictions. The accuracy verification mechanism is a dynamic monitoring and correction process used to evaluate and ensure the consistency between the simulation output of the process state simulation model and the actual state of the pressing process. During the simulation, this accuracy verification mechanism identifies whether the model's prediction accuracy has decreased and, if necessary, triggers online adjustments to the model's internal compensation parameters, thereby maintaining the reliability of the simulation results.

[0044] Specifically, the process state simulation model is configured with simulation termination conditions and accuracy verification mechanisms, and the prediction time step and calculation depth of the simulation process are set based on the target termination position and the target clamping force, including:

[0045] The target termination position is taken as the endpoint of displacement calculation, and the target clamping force is taken as the endpoint of pressure calculation, which together constitute the joint judgment condition for the termination of calculation.

[0046] Establish an accuracy verification mechanism based on the real-time feedback of the pressure-displacement monitoring sequence, and perform rolling comparison and deviation correction between the predicted values ​​and the actual collected values ​​during the simulation process;

[0047] The prediction time step is dynamically adjusted based on the relative difference between the actual displacement value and the target termination position at the current pressing stage.

[0048] The computational depth of the simulation process is adaptively determined based on the real-time magnitudes of the predicted pressure deviation and the predicted displacement deviation.

[0049] First, the target termination position is used as the endpoint of displacement estimation, and the target clamping force is used as the endpoint of pressure estimation. These two factors together constitute the joint judgment condition for terminating the estimation. Specifically, when the predicted displacement value calculated by the process state estimation model reaches or exceeds the target termination position, and at the same time the predicted pressure value reaches or exceeds the target clamping force, the estimation termination condition is determined to be met, and the estimation stops.

[0050] Secondly, an accuracy verification mechanism based on real-time feedback from the pressure-displacement monitoring sequence is established. This accuracy verification mechanism runs continuously during the simulation process, comparing and correcting deviations between the predicted pressure and displacement values ​​output by the process state simulation model at the end of each prediction time step and the corresponding actual pressure and displacement values ​​in the pressure-displacement monitoring sequence at the same time.

[0051] Specifically, an accuracy verification mechanism based on the real-time feedback of the pressure-displacement monitoring sequence is established to perform rolling comparisons and deviation corrections between predicted and actual collected values ​​during the simulation process, including:

[0052] At the end of each prediction time step, the predicted pressure value and predicted displacement value output by the process state deduction model are compared with the actual pressure value and actual displacement value in the pressure-displacement monitoring sequence at the same time, and the instantaneous pressure deviation and instantaneous displacement deviation of the current step are calculated respectively.

[0053] The instantaneous pressure deviation and instantaneous displacement deviation are compared with the allowable process deviation threshold based on the target clamping force and the target termination position;

[0054] If the instantaneous pressure deviation or instantaneous displacement deviation exceeds the corresponding process deviation threshold, it is determined that the process state simulation model has decreased in simulation accuracy under the current working condition, and deviation correction is initiated.

[0055] The deviation correction includes fitting the deviation change trend based on the instantaneous pressure deviation and instantaneous displacement deviation sequence within the most recent consecutive prediction time steps, and adjusting the compensation parameters determined by the temperature-material deformation relationship in the process state inference model in reverse.

[0056] The process state simulation model is updated using the adjusted compensation parameters, and subsequent simulations are performed again based on the latest pressure-displacement monitoring sequence to complete this rolling verification and correction.

[0057] First, at the end of each prediction time step, the predicted pressure and displacement values ​​output by the process state simulation model at that moment are compared with the actual pressure and displacement values ​​recorded in the pressure-displacement monitoring sequence at the same moment. By calculating the difference between the predicted and actual values, the instantaneous pressure deviation and instantaneous displacement deviation of the current step are obtained. The prediction time step refers to the time interval corresponding to one complete forward prediction calculation performed by the process state simulation model. This prediction time step is dynamically set based on the relative difference between the actual displacement value and the target termination position at the current pressing stage. For example, a step size of 10 milliseconds is used when the relative difference is large at the initial stage of pressing, and a step size of 1 millisecond is used when the relative difference is small at the end of pressing.

[0058] Next, the calculated instantaneous pressure deviation and instantaneous displacement deviation are compared with preset process deviation thresholds. These process deviation thresholds are set based on the allowable process fluctuation range between the target clamping force and the target termination position. For example, the allowable deviation threshold for the target termination position can be set to ±5% of the target termination position, and the allowable deviation threshold for the target clamping force can be set to ±10% of the target clamping force.

[0059] If the instantaneous pressure deviation or instantaneous displacement deviation exceeds its corresponding process deviation threshold, it is determined that the process state simulation model has decreased in accuracy under the current operating conditions, and the deviation correction process must be initiated immediately.

[0060] Specifically, the deviation correction process includes obtaining the deviation trend over time through fitting analysis based on the instantaneous pressure deviation sequence and instantaneous displacement deviation sequence within the most recent consecutive prediction time steps. Based on this deviation trend, the compensation parameters determined by the temperature-material deformation relationship in the process state deduction model are adjusted in reverse. These compensation parameters mainly include the thermal expansion displacement coefficient of the indenter and the stiffness pressure coefficient of the coil insulation layer.

[0061] Specifically, the indenter thermal expansion displacement coefficient is a proportional factor used to quantify the influence of axial thermal expansion caused by indenter temperature changes on the displacement prediction value. This indenter thermal expansion displacement coefficient is calculated based on the linear expansion coefficient of the indenter material and the effective working length of the indenter. Its function is to convert the difference between the real-time collected indenter temperature and the standard reference temperature into a compensation adjustment amount for the displacement prediction component in the process state extrapolation model.

[0062] The coil insulation stiffness pressure coefficient is a proportional factor used to quantify the impact of changes in the equivalent stiffness of the insulation material caused by coil temperature variations on the pressure prediction value. This coefficient is calculated based on the temperature-elastic modulus relationship of the insulation material. Its function is to substitute the real-time collected coil temperature into a preset temperature-modulus curve to calculate the rate of change of the current real-time elastic modulus of the insulation layer relative to a standard reference modulus, and then convert this into a compensation adjustment amount for the pressure prediction component in the process state simulation model.

[0063] For example, if the displacement deviation shows a continuous increasing trend, it indicates that the thermal expansion displacement compensation of the pressure head is insufficient, and the thermal expansion displacement coefficient of the pressure head needs to be increased; if the pressure deviation shows a continuous decreasing trend, it may indicate that the stiffness pressure compensation of the coil insulation layer is excessive, and the stiffness pressure coefficient of the insulation layer needs to be increased accordingly to weaken its compensation effect.

[0064] Finally, the internal structure of the process state simulation model is updated using the adjusted compensation parameters, and subsequent simulation calculations are performed again based on the latest acquired pressure-displacement monitoring sequence. Through these steps, a complete rolling verification and real-time correction are completed, thereby continuously maintaining the consistency between the prediction accuracy of the process state simulation model and the actual pressing process.

[0065] Furthermore, the prediction time step is dynamically adjusted based on the relative difference between the actual displacement value and the target termination position at the current pressing stage. Specifically, the relative difference = (target termination position - actual displacement value) / target termination position × 100%. When the relative difference is large, it indicates that the pressing end point is still far away. In this case, a larger prediction time step is used for trend extrapolation to improve calculation efficiency. When the relative difference is small, it indicates that the pressing end point is approaching. In this case, a smaller prediction time step is used for refined prediction to improve control accuracy.

[0066] Finally, the computational depth of the simulation process is adaptively determined based on the real-time magnitudes of the predicted pressure and displacement deviations. The computational depth refers to the number of internal iterations performed by the process state simulation model to complete a single prediction. More iterations result in a more detailed simulation of complex operating conditions and, theoretically, higher prediction accuracy, but also a correspondingly increased computational load.

[0067] Therefore, the calculation depth is dynamically adjusted based on the magnitude of the real-time deviation. For example, a deviation judgment threshold is set: when the relative deviation of the predicted pressure or displacement is ≥8%, it is judged as a large deviation, and a larger calculation depth is set accordingly, such as 5 iterations; when the relative deviation is between 3% and 8%, it is judged as a medium deviation, and a medium calculation depth is set accordingly, such as 3 iterations; when the relative deviation is <3%, it is judged as a small deviation, and a smaller calculation depth is set accordingly, such as 1 iteration. Through this adaptive mechanism, the allocation of computing resources can be optimized while ensuring the necessary inference accuracy.

[0068] Finally, the real-time acquired pressure head temperature and coil temperature are introduced as real-time correction variables for the process state simulation model. Through preset temperature-material deformation relationships, such as the thermal expansion coefficient of the pressure head material and the temperature-elastic modulus characteristic curve of the coil insulation material, the aforementioned basic simulation logic is compensated and corrected online. This compensation process can dynamically adjust key parameters in the simulation calculation, enabling the process state simulation model to reflect the influence of the current temperature field on material deformation behavior and mechanical response. After temperature compensation, a process state simulation model with higher predictive adaptability is finally generated, suitable for the specific pressing conditions.

[0069] Following the construction and real-time compensation of the process state simulation model, the prediction and trend analysis phase begins. Using the constructed process state simulation model, the complete future path of the pressing process, under the current control strategy and parameters, is simulated from its current state to its natural termination. Specifically, the process state simulation model uses the latest pressure-displacement monitoring sequence as the initial state and, based on the control law corresponding to the current control mode, iteratively calculates the pressure and displacement values ​​at each predicted time step until the simulation termination condition is met.

[0070] Specifically, this prediction process is used to determine whether the target termination position and target clamping force can be simultaneously achieved at the end of the pressing process if the existing control mode remains unchanged. The process state simulation model outputs qualitative trend analysis results by comparing the endpoint of the predicted path with the preset dual targets. These trend analysis results include a core judgment conclusion and a set of quantitative prediction data. The core judgment conclusion is a binary determination of whether the target termination position and target clamping force can be simultaneously satisfied under the current control mode until the end of the pressing process. When the judgment conclusion is "yes," it indicates that the predicted trajectory will converge to both targets simultaneously, and the current control strategy can remain unchanged. When the judgment conclusion is "no," it indicates that the dual targets cannot be achieved simultaneously under the current path. In this case, the trend analysis result will simultaneously output the corresponding quantitative prediction data, namely, the predicted pressure deviation and the predicted displacement deviation. The predicted pressure deviation refers to the difference between the predicted endpoint pressure value and the target clamping force, and the predicted displacement deviation refers to the difference between the predicted endpoint displacement value and the target termination position.

[0071] The output trend analysis results can provide key decision-making basis for subsequent steps. Specifically, if the trend analysis results show that both objectives can be met simultaneously, the current control mode will be maintained and pressing will continue; if the trend analysis results show that they cannot be met simultaneously, the subsequent multi-objective optimization and controller reconfiguration process will be triggered to achieve adaptive adjustment of process parameters.

[0072] S40: If the trend analysis results show that they cannot be satisfied at the same time, then the predicted pressure deviation and the predicted displacement deviation are used as optimization objectives, the real-time collected pressure head temperature and coil temperature are introduced as process state compensation quantities, a multi-objective optimization function containing weight allocation variables is constructed, and the control objective bias coefficient is solved.

[0073] Specifically, if the trend analysis results indicate that both conditions cannot be met simultaneously, then the predicted pressure deviation and predicted displacement deviation are used as optimization objectives. Real-time collected pressure head temperature and coil temperature are introduced as process state compensation quantities. A multi-objective optimization function containing weighted allocation variables is constructed, and the control objective bias coefficient is solved, including:

[0074] Minimizing the predicted pressure deviation and minimizing the predicted displacement deviation are used as common optimization objectives;

[0075] Based on the pressure head temperature and coil temperature, determine the proportional coefficient of the influence of the process state compensation on the pressure control target and the displacement control target;

[0076] Using the aforementioned proportional coefficient, a multi-objective optimization function is constructed with the weighted allocation variable as the solution object, wherein the multi-objective optimization function aims to minimize the expected total deviation after weighting;

[0077] Solve the multi-objective optimization function to obtain the optimal solution of the weight allocation variable, and normalize the optimal solution as the control objective bias coefficient.

[0078] Specifically, if the trend analysis results indicate that the target termination position and the target clamping force cannot be simultaneously satisfied, then the predicted pressure deviation and predicted displacement deviation are used as optimization objectives. Real-time collected pressure head temperature and coil temperature are introduced as process state compensation quantities. A multi-objective optimization function containing weighted allocation variables is constructed, and the control target bias coefficient is solved. This process includes the following steps:

[0079] First, minimizing the predicted pressure deviation and minimizing the predicted displacement deviation are established as common optimization objectives. Second, based on real-time acquired pressure head temperature and coil temperature, a proportionality coefficient is determined to determine the respective influence of process state compensation on the pressure control objective and displacement control objective. This proportionality coefficient reflects the relative influence of temperature changes on achieving the pressure and displacement objectives under the current temperature field conditions. For example, changes in pressure head temperature may primarily affect displacement accuracy, while changes in coil temperature may primarily affect pressure response.

[0080] Specifically, based on the pressure head temperature and coil temperature, the proportionality coefficient of the influence of the process state compensation on the pressure control target and the displacement control target is determined, including:

[0081] Based on the thermal expansion characteristics of the indenter material, the thermal expansion of the indenter temperature relative to standard operating conditions is calculated and mapped to a temperature influence factor on the displacement control target.

[0082] Based on the thermodynamic properties of the coil insulation material, the rate of change of the equivalent stiffness of the insulation layer caused by the coil temperature is calculated and mapped to a temperature influence factor on the pressure control target.

[0083] Based on the temperature influence factors on the displacement control target and the temperature influence factors on the pressure control target, a proportional coefficient characterizing the relative influence of the two is calculated through normalization.

[0084] First, based on the thermal expansion characteristics of the indenter material, the thermal expansion of the indenter temperature relative to standard operating conditions is calculated and mapped to a temperature influence factor on the displacement control target, including:

[0085] Using the reference temperature of the pressure head under standard working conditions in the historical pressing process database as a benchmark, the real-time temperature difference between the real-time collected pressure head temperature and the pressure head reference temperature is calculated.

[0086] Based on the preset linear expansion coefficient of the indenter material and the effective working length of the indenter in the displacement measurement direction, the real-time temperature difference is converted into the absolute thermal expansion compensation amount of the indenter in the axial direction.

[0087] The ratio of the absolute thermal expansion compensation to the target termination position is calculated to obtain the relative influence coefficient, which is then used as the temperature influence factor for the displacement control target.

[0088] First, based on the thermal expansion characteristics of the indenter material, the thermal expansion of the indenter temperature relative to standard operating conditions is calculated, and this physical quantity is mapped to a temperature influence factor on the displacement control target. This process specifically includes the following steps.

[0089] Specifically, firstly, using the standard operating condition pressure head reference temperature recorded in the historical press-fitting process database as a benchmark, the real-time temperature difference between the real-time collected pressure head temperature and the pressure head reference temperature is calculated. For example, if the pressure head reference temperature under standard operating conditions is 25℃ and the real-time collected pressure head temperature is 35℃, then the real-time temperature difference ΔT = 35℃ - 25℃ = 10℃.

[0090] Secondly, based on the preset linear expansion coefficient of the indenter material and the effective working length of the indenter in the displacement measurement direction, the real-time temperature difference is converted into the absolute thermal expansion compensation of the indenter in the axial direction. The linear expansion coefficient of the indenter material is determined according to the specific material of the indenter; for example, when the indenter is made of 45# steel, its linear expansion coefficient α = 11.5 × 10⁻⁶. -6 / ℃. Assuming the effective working length L0 in the displacement measurement direction is 100mm, and considering the aforementioned temperature difference of 10℃, according to the thermal expansion calculation formula, the axial absolute thermal expansion compensation amount = linear expansion coefficient × effective working length × real-time temperature difference, that is, the axial absolute thermal expansion compensation amount ΔL = α·L0·ΔT. The calculation yields the axial absolute thermal expansion compensation amount ΔL = 11.5 × 10 -6 / ℃×100mm×10℃=0.0115mm. This value indicates that due to the increase in the temperature of the indenter, it has a thermal expansion elongation of 0.0115mm in the axial direction. If this thermal expansion elongation is not compensated, it will directly lead to a systematic positive deviation in the displacement measurement value.

[0091] Finally, the calculated absolute thermal expansion compensation is compared with the target termination position to obtain a relative influence coefficient, which is then used as the temperature influence factor for the displacement control target. Specifically, the temperature influence factor = relative influence coefficient = absolute thermal expansion compensation / target termination position. For example, if the target termination position is 50 mm, then the temperature influence factor = 0.0115 / 50 = 0.00023. This temperature influence factor is a dimensionless value, and its physical meaning characterizes the relative proportion of the displacement offset caused by the thermal expansion of the pressure head to the total target displacement, thus quantifying the potential impact of the current pressure head temperature fluctuation on achieving the displacement control target.

[0092] Furthermore, based on the thermodynamic properties of the coil insulation material, the rate of change of the equivalent stiffness of the insulation layer caused by the coil temperature is calculated and mapped to a temperature influence factor on the pressure control target, including:

[0093] Obtain the reference elastic modulus of the insulating material under standard operating conditions from the historical press-fitting process database;

[0094] Based on the preset temperature-modulus characteristic relationship of the insulation material, and combined with the real-time collected coil temperature, the real-time elastic modulus of the insulation layer at the current temperature is calculated by interpolation.

[0095] The rate of change of the real-time elastic modulus relative to the reference elastic modulus is calculated as the rate of change of the equivalent stiffness of the insulation layer.

[0096] The equivalent stiffness change rate of the insulation layer is correlated with the pressure change sensitivity of the standard pressure-displacement reference curve at the end of the press fitting process to obtain the temperature influence factor on the pressure control target.

[0097] Furthermore, based on the thermodynamic properties of the coil insulation material, the rate of change of the equivalent stiffness of the insulation layer caused by the coil temperature is calculated, and this rate of change of the equivalent stiffness of the insulation layer is mapped to a temperature influence factor on the pressure control target. This process specifically includes the following steps.

[0098] First, the reference elastic modulus of the insulating material, measured under standard operating conditions, is obtained from the historical press-fitting process database. This reference elastic modulus is a benchmark value of the inherent mechanical properties exhibited by the insulating material at a specific standard temperature. For example, the historical press-fitting process database records that the reference elastic modulus of the insulating material is 3.2 GPa at a standard temperature of 25 degrees Celsius.

[0099] Secondly, based on the preset temperature-modulus characteristic relationship of the insulation material and combined with the real-time acquired coil temperature, the real-time elastic modulus of the insulation layer at the current temperature is calculated through interpolation. The temperature-modulus characteristic relationship is usually characterized in the form of a fitted curve, which describes the law of change of the elastic modulus of the insulation material with temperature. By inputting the real-time acquired coil temperature into this temperature-modulus characteristic relationship fitted curve, the real-time elastic modulus of the insulation layer under the current actual temperature conditions can be calculated. For example, if the real-time acquired coil temperature is 48℃, the real-time elastic modulus of the insulation layer at the current temperature can be calculated as 2.587 GPa through curve interpolation.

[0100] Then, the rate of change of the real-time elastic modulus relative to the reference elastic modulus is calculated, and this rate of change is taken as the rate of change of the equivalent stiffness of the insulation layer. Specifically, the rate of change of the equivalent stiffness of the insulation layer = (real-time elastic modulus - reference elastic modulus) / reference elastic modulus. This rate of change of the equivalent stiffness of the insulation layer is a dimensionless value. The positive or negative sign indicates the increase or decrease in stiffness relative to the standard operating condition, and its absolute value quantifies the severity of the stiffness change. For example, when the real-time elastic modulus is 2.587 GPa and the reference elastic modulus is 3.2 GPa, the rate of change of the equivalent stiffness of the insulation layer = (2.587 - 3.2) / 3.2 = -0.1916. This negative value indicates that compared to the standard operating condition, the equivalent stiffness of the insulation layer at the current temperature has decreased by approximately 19.16%.

[0101] Finally, the calculated equivalent stiffness change rate of the insulation layer is correlated with the pressure change sensitivity of the standard pressure-displacement reference curve extracted from the historical database at the end of the pressing process to obtain the temperature influence factor for the pressure control target. The pressure change sensitivity reflects the coupling gradient between force and displacement at the end of the pressing process. For example, from the standard pressure-displacement reference curve, in the final section between 48mm and 50mm displacement, a 2mm displacement change corresponds to a 2kN pressure change; therefore, the calculated pressure change sensitivity = 2kN / 1kN = 1kN / mm. The correlation calculation multiplies the stiffness change rate by the sensitivity and introduces a normalization coefficient calibrated based on historical data for calibration. For example, setting the normalization coefficient to 0.5, the temperature influence factor = |-0.1916×1|×0.5≈0.0958. This temperature influence factor accurately characterizes the interference intensity of insulation material softening on achieving the target clamping force at the current coil temperature, providing a quantitative basis for the adaptive adjustment of subsequent control parameters.

[0102] Finally, based on the temperature influence factors for the displacement control target and the pressure control target calculated above, a proportionality coefficient characterizing the relative influence of the two is calculated through normalization. Specifically, the normalization process first calculates the sum of the two temperature influence factors, then divides one of the temperature influence factors by the sum, resulting in a single proportionality coefficient with a value between 0 and 1. This proportionality coefficient is the ratio of the temperature influence factor for the displacement control target to the sum, used to quantify the relative weight of the displacement control target in the coordination of the two targets.

[0103] For example, when the temperature influence factor for the displacement control target is 0.00023 and the temperature influence factor for the pressure control target is 0.0958, the proportionality coefficient = 0.00023 / (0.00023+0.0958) = 0.00239. This proportionality coefficient is a single value between 0 and 1, used to indicate the relative importance of the process state compensation amount to the displacement control target and the pressure control target under the current temperature field conditions. In this example, the proportionality coefficient is approximately 0.00239, close to 0, indicating that under the current operating conditions, the temperature compensation amount has a much greater impact on the pressure control target than on the displacement control target. Therefore, in subsequent multi-objective optimization and controller reconfiguration, the control strategy should be tilted towards the force loop control target. This proportionality coefficient provides a direct quantitative basis for the weight allocation in the subsequent construction of the multi-objective optimization function.

[0104] Furthermore, using the aforementioned proportionality coefficients, a multi-objective optimization function is constructed, with the weighted allocation variable as the solution object. Mathematically, this multi-objective optimization function is expressed as a weighted sum of pressure and displacement deviations, with the core objective of minimizing this weighted sum, i.e., the expected value of the total deviation. The weighted allocation variable is the unknown quantity to be solved, and its value determines the relative emphasis to be placed on pressure control and displacement control in subsequent control. By solving this multi-objective optimization function, the optimal solution for the weighted allocation variable can be obtained. This optimal solution represents the best balance point between the pressure control objective and the displacement control objective to minimize the expected total deviation under the current temperature compensation conditions.

[0105] Finally, the obtained optimal solution is normalized, with its value constrained to the range of 0-1, and the processed value is used as the control target bias coefficient. This control target bias coefficient will be directly used for the subsequent reconstruction of the hybrid controller to guide the fusion ratio of the position loop and force loop output commands. The closer its value is to 1, the more the control strategy focuses on the force loop target; the closer it is to 0, the more it focuses on the position loop target, thereby achieving precise quantitative guidance for the control behavior of the final stage of press fitting.

[0106] S50: Based on the control target bias coefficient, reconstruct the hybrid controller of the final pressing stage, dynamically weight and fuse the output commands of the position loop and force loop, and continue to complete the pressing control of the target stator coil based on the fused unified command.

[0107] Specifically, based on the control target bias coefficient, the hybrid controller of the final pressing stage is reconstructed, the output commands of the position loop and force loop are dynamically weighted and fused, and based on the fused unified command, the pressing control of the target stator coil is continued, including:

[0108] The control target bias coefficient is used as the initial fusion weight of the force loop command, and the initial fusion weight is dynamically fine-tuned based on the real-time proportional relationship between the predicted pressure deviation and the predicted displacement deviation to obtain the dynamic weight of the force loop.

[0109] The corresponding complementary value is calculated based on the dynamic weight and used as the dynamic weight of the position loop;

[0110] Within each control cycle, the position adjustment command generated by the position loop based on the actual displacement value and the force adjustment command generated by the force loop based on the actual pressure value are synchronously acquired.

[0111] Based on the dynamic weights of the force loop and the dynamic weights of the position loop, the position adjustment command and the force adjustment command are weighted and summed to generate a unified command to drive the actuator.

[0112] Repeat the above dynamic weighted fusion process until the pressing process ends, while simultaneously satisfying the target termination position and the target clamping force.

[0113] First, a control target bias coefficient is used as the initial weight for the force loop output command to participate in the fusion. This control target bias coefficient reflects the overall emphasis of the force control target relative to the displacement control target after considering temperature compensation. Simultaneously, this initial weight is dynamically fine-tuned based on the real-time proportional relationship between the predicted pressure deviation and the predicted displacement deviation, resulting in the force loop dynamic weight. For example, the force loop dynamic weight = control target bias coefficient + k × [(predicted pressure deviation / predicted displacement deviation - 1) / (predicted pressure deviation / predicted displacement deviation + 1)]. Here, k is a fine-tuning gain coefficient set based on process experience, and its value range is usually small, for example, k = 0.05 to 0.1, to ensure smooth and stable adjustment. When the predicted pressure deviation is significantly greater than the predicted displacement deviation, the force loop dynamic weight is appropriately increased to strengthen the adjustment of the force loop; conversely, it is appropriately decreased.

[0114] Secondly, the complementary value is calculated based on the dynamic weight of the force loop, and this complementary value is used as the dynamic weight of the position loop. The complementary value = 1 - the dynamic weight of the force loop, thus ensuring that the sum of the weights of the two loops is always 1. This fully preserves the contributions of the two control loops in command fusion, and the weight allocation clearly reflects the real-time bias of the control objective.

[0115] Furthermore, within each control cycle, position adjustment commands generated by the position loop based on actual displacement values ​​and force adjustment commands generated by the force loop based on actual pressure values ​​are synchronously acquired. Based on the calculated dynamic weights of the force loop and position loop, the acquired position adjustment commands and force adjustment commands are weighted and summed to generate a single, coordinated unified command. This unified command is sent to the pressing execution device to drive it to complete the pressing action of the current control cycle.

[0116] Finally, the above dynamic weighted fusion process is repeated, that is, data is continuously collected, dynamic weights are calculated, dual-ring instructions are obtained and weighted fusion is performed until the pressing process ends, and the actual pressure value and the actual displacement value simultaneously satisfy the target clamping force and the target termination position.

[0117] It should be noted that after completing the press-fit control of the target stator coil, the control process also includes subsequent continuous monitoring and adaptive maintenance.

[0118] During subsequent pressing, control is continuously performed based on the reconstructed hybrid controller and dynamic weighted fusion strategy. Simultaneously, the actual pressure value, actual displacement value, pressure head temperature, and coil temperature are monitored in real time. It is then determined whether these monitored values ​​exceed the safety threshold range preset based on the pressing target parameters and process specifications.

[0119] If any parameter value is detected to exceed its corresponding safety threshold, the event is recorded, and a warning is issued based on the trend analysis results, such as issuing an audible and visual alarm or recording a fault code. Simultaneously, before the start of the next pressing cycle, the status of parameters related to this event, such as temperature, pressure, displacement values ​​at the time of the event and their changing trends, is fed back to the process state simulation model. This is used to adaptively correct the basic simulation logic of the model, thereby improving the model's prediction accuracy and robustness for similar future operating conditions and achieving continuous self-optimization of the control system.

[0120] In summary, the embodiments of this application have at least the following technical effects:

[0121] Compared to existing technologies, the adaptive control method and equipment for motor coil pressing process parameters provided in this application achieves early and accurate prediction and trend judgment of the final state of the pressing process by constructing a process state deduction model that integrates real-time data and historical knowledge. By introducing the pressure head temperature and coil temperature as key process state compensation quantities, and constructing an optimization function with prediction deviation as the objective to solve for the control target bias coefficient, the control strategy can dynamically adapt to material property fluctuations and environmental disturbances. Furthermore, based on this bias coefficient, a hybrid controller is reconstructed, and the position loop and force loop commands are dynamically weighted and fused, achieving precise coordinated control of both pressure and displacement objectives at the end of the pressing process.

[0122] In summary, this application effectively overcomes the control mismatch problem caused by batch differences in materials, temperature changes, and equipment state drift, and improves the stability, control accuracy, and product consistency of coil pressing processes for various motors such as drive motors for new energy vehicles and motors for household appliances.

[0123] Example 2, as Figure 2 As shown, based on the same inventive concept as the adaptive control method for motor coil pressing process parameters provided in Embodiment 1, this embodiment of the invention also provides an adaptive control device for motor coil pressing process parameters, comprising:

[0124] The target parameter acquisition module 11 is used to acquire the press-fit target parameters of the target stator coil, wherein the press-fit target parameters include a preset target termination position and a target clamping force;

[0125] The data acquisition and sequence generation module 12 is used to start the pressing process, collect the actual pressure value and actual displacement value in the current pressing process in real time, and generate a pressure-displacement monitoring sequence.

[0126] The deduction and trend analysis module 13 is used to construct a process state deduction model based on the pressure-displacement monitoring sequence, predict whether the target termination position and the target clamping force are simultaneously satisfied when the pressing continues until the end of the pressing under the current control mode, and output the trend analysis results.

[0127] The multi-objective optimization and solution module 14 is used to construct a multi-objective optimization function containing weighted allocation variables when the trend analysis results show that they cannot be satisfied at the same time, with the predicted pressure deviation and the predicted displacement deviation as optimization objectives, the real-time collected pressure head temperature and coil temperature as process state compensation quantities, and the control objective bias coefficient is solved.

[0128] The hybrid control and execution module 15 is used to reconstruct the hybrid controller of the final pressing stage based on the control target bias coefficient, dynamically weight and fuse the output commands of the position loop and force loop, and continue to complete the pressing control of the target stator coil based on the fused unified command.

[0129] The target parameter acquisition module 11 is specifically used for:

[0130] Obtain the press-fit target parameters of the target stator coil, wherein the press-fit target parameters include a preset target termination position and a target clamping force.

[0131] The data acquisition and sequence generation module 12 is specifically used for:

[0132] Initiate the pressing process and collect the actual pressure and displacement values ​​in real time to generate a pressure-displacement monitoring sequence.

[0133] Specifically, the deduction and trend analysis module 13 is used for:

[0134] The steps for constructing the process state deduction model include:

[0135] Based on the historical press-fitting process database, the standard pressure-displacement reference curve under standard working conditions is extracted, and combined with the real-time change trend of the pressure-displacement monitoring sequence, the basic deduction logic of the process state deduction model is constructed.

[0136] Configure the deduction termination conditions and accuracy verification mechanism of the process state deduction model, and set the prediction time step and calculation depth of the deduction process according to the target termination position and the target clamping force;

[0137] The pressure head temperature and coil temperature are introduced as real-time correction variables. The basic deduction logic is compensated online by a preset temperature-material deformation relationship to generate a process state deduction model suitable for the current pressing process.

[0138] Specifically, the process state simulation model is configured with simulation termination conditions and accuracy verification mechanisms, and the prediction time step and calculation depth of the simulation process are set based on the target termination position and the target clamping force, including:

[0139] The target termination position is taken as the endpoint of displacement calculation, and the target clamping force is taken as the endpoint of pressure calculation, which together constitute the joint judgment condition for the termination of calculation.

[0140] Establish an accuracy verification mechanism based on the real-time feedback of the pressure-displacement monitoring sequence, and perform rolling comparison and deviation correction between the predicted values ​​and the actual collected values ​​during the simulation process;

[0141] The prediction time step is dynamically adjusted based on the relative difference between the actual displacement value and the target termination position at the current pressing stage.

[0142] The computational depth of the simulation process is adaptively determined based on the real-time magnitudes of the predicted pressure deviation and the predicted displacement deviation.

[0143] Specifically, an accuracy verification mechanism based on the real-time feedback of the pressure-displacement monitoring sequence is established to perform rolling comparisons and deviation corrections between predicted and actual collected values ​​during the simulation process, including:

[0144] At the end of each prediction time step, the predicted pressure value and predicted displacement value output by the process state deduction model are compared with the actual pressure value and actual displacement value in the pressure-displacement monitoring sequence at the same time, and the instantaneous pressure deviation and instantaneous displacement deviation of the current step are calculated respectively.

[0145] The instantaneous pressure deviation and instantaneous displacement deviation are compared with the allowable process deviation threshold based on the target clamping force and the target termination position;

[0146] If the instantaneous pressure deviation or instantaneous displacement deviation exceeds the corresponding process deviation threshold, it is determined that the process state simulation model has decreased in simulation accuracy under the current working condition, and deviation correction is initiated.

[0147] The deviation correction includes fitting the deviation change trend based on the instantaneous pressure deviation and instantaneous displacement deviation sequence within the most recent consecutive prediction time steps, and adjusting the compensation parameters determined by the temperature-material deformation relationship in the process state inference model in reverse.

[0148] The process state simulation model is updated using the adjusted compensation parameters, and subsequent simulations are performed again based on the latest pressure-displacement monitoring sequence to complete this rolling verification and correction.

[0149] The multi-objective optimization and solution module 14 is specifically used for:

[0150] If the trend analysis results indicate that both conditions cannot be met simultaneously, then the predicted pressure deviation and predicted displacement deviation are used as optimization objectives. Real-time collected pressure head temperature and coil temperature are introduced as process state compensation quantities. A multi-objective optimization function containing weighted allocation variables is constructed, and the control objective bias coefficient is solved, including:

[0151] Minimizing the predicted pressure deviation and minimizing the predicted displacement deviation are used as common optimization objectives;

[0152] Based on the pressure head temperature and coil temperature, determine the proportional coefficient of the influence of the process state compensation on the pressure control target and the displacement control target;

[0153] Using the aforementioned proportional coefficient, a multi-objective optimization function is constructed with the weighted allocation variable as the solution object, wherein the multi-objective optimization function aims to minimize the expected total deviation after weighting;

[0154] Solve the multi-objective optimization function to obtain the optimal solution of the weight allocation variable, and normalize the optimal solution as the control objective bias coefficient.

[0155] Among them, the proportional coefficient for determining the influence of the process state compensation on the pressure control target and the displacement control target based on the pressure head temperature and the coil temperature includes:

[0156] Based on the thermal expansion characteristics of the indenter material, the thermal expansion of the indenter temperature relative to standard operating conditions is calculated and mapped to a temperature influence factor on the displacement control target.

[0157] Based on the thermodynamic properties of the coil insulation material, the rate of change of the equivalent stiffness of the insulation layer caused by the coil temperature is calculated and mapped to a temperature influence factor on the pressure control target.

[0158] Based on the temperature influence factors on the displacement control target and the temperature influence factors on the pressure control target, a proportional coefficient characterizing the relative influence of the two is calculated through normalization.

[0159] Specifically, based on the thermal expansion characteristics of the indenter material, the thermal expansion of the indenter temperature relative to standard operating conditions is calculated and mapped to a temperature influence factor on the displacement control target, including:

[0160] Using the reference temperature of the pressure head under standard working conditions in the historical pressing process database as a benchmark, the real-time temperature difference between the real-time collected pressure head temperature and the pressure head reference temperature is calculated.

[0161] Based on the preset linear expansion coefficient of the indenter material and the effective working length of the indenter in the displacement measurement direction, the real-time temperature difference is converted into the absolute thermal expansion compensation amount of the indenter in the axial direction.

[0162] The ratio of the absolute thermal expansion compensation to the target termination position is calculated to obtain the relative influence coefficient, which is then used as the temperature influence factor for the displacement control target.

[0163] Specifically, based on the thermodynamic properties of the coil insulation material, the rate of change of the equivalent stiffness of the insulation layer caused by the coil temperature is calculated and mapped to a temperature influence factor on the pressure control target, including:

[0164] Obtain the reference elastic modulus of the insulating material under standard operating conditions from the historical press-fitting process database;

[0165] Based on the preset temperature-modulus characteristic relationship of the insulation material, and combined with the real-time collected coil temperature, the real-time elastic modulus of the insulation layer at the current temperature is calculated by interpolation.

[0166] The rate of change of the real-time elastic modulus relative to the reference elastic modulus is calculated as the rate of change of the equivalent stiffness of the insulation layer.

[0167] The equivalent stiffness change rate of the insulation layer is correlated with the pressure change sensitivity of the standard pressure-displacement reference curve at the end of the press fitting process to obtain the temperature influence factor on the pressure control target.

[0168] The hybrid control and execution module 15 is specifically used for:

[0169] Based on the control target bias coefficient, the hybrid controller of the final pressing stage is reconstructed, and the output commands of the position loop and force loop are dynamically weighted and fused. Based on the fused unified command, the pressing control of the target stator coil is continued, including:

[0170] The control target bias coefficient is used as the initial fusion weight of the force loop command, and the initial fusion weight is dynamically fine-tuned based on the real-time proportional relationship between the predicted pressure deviation and the predicted displacement deviation to obtain the dynamic weight of the force loop.

[0171] The corresponding complementary value is calculated based on the dynamic weight and used as the dynamic weight of the position loop;

[0172] Within each control cycle, the position adjustment command generated by the position loop based on the actual displacement value and the force adjustment command generated by the force loop based on the actual pressure value are synchronously acquired.

[0173] Based on the dynamic weights of the force loop and the dynamic weights of the position loop, the position adjustment command and the force adjustment command are weighted and summed to generate a unified command to drive the actuator.

[0174] Repeat the above dynamic weighted fusion process until the pressing process ends, while simultaneously satisfying the target termination position and the target clamping force.

[0175] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0176] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0177] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for adaptive adjustment of motor coil press-fitting process parameters, characterized in that, The method includes: Obtain the press-fit target parameters of the target stator coil, wherein the press-fit target parameters include a preset target termination position and a target clamping force; Start the pressing process and collect the actual pressure and displacement values ​​in real time during the current pressing process to generate a pressure-displacement monitoring sequence. Based on the pressure-displacement monitoring sequence, a process state deduction model is constructed to predict whether the target termination position and the target clamping force are simultaneously satisfied when pressing continues until the end of pressing under the current control mode, and the trend analysis results are output. If the trend analysis results show that they cannot be satisfied at the same time, then the predicted pressure deviation and the predicted displacement deviation are used as optimization objectives, and the real-time collected pressure head temperature and coil temperature are introduced as process state compensation quantities. A multi-objective optimization function containing weighted allocation variables is constructed, and the control objective bias coefficient is solved. Based on the control target bias coefficient, the hybrid controller of the final pressing stage is reconstructed, the output commands of the position loop and force loop are dynamically weighted and fused, and based on the fused unified command, the pressing control of the target stator coil is continued.

2. The adaptive control method for motor coil press-fitting process parameters according to claim 1, characterized in that, The steps for constructing the process state deduction model include: Based on the historical press-fitting process database, the standard pressure-displacement reference curve under standard working conditions is extracted, and combined with the real-time change trend of the pressure-displacement monitoring sequence, the basic deduction logic of the process state deduction model is constructed. Configure the deduction termination conditions and accuracy verification mechanism of the process state deduction model, and set the prediction time step and calculation depth of the deduction process according to the target termination position and the target clamping force; The pressure head temperature and coil temperature are introduced as real-time correction variables. The basic deduction logic is compensated online by a preset temperature-material deformation relationship to generate a process state deduction model suitable for the current pressing process.

3. The adaptive control method for motor coil press-fitting process parameters according to claim 2, characterized in that, Configure the deduction termination conditions and accuracy verification mechanism of the process state deduction model, and set the prediction time step and calculation depth of the deduction process based on the target termination position and the target clamping force, including: The target termination position is taken as the endpoint of displacement calculation, and the target clamping force is taken as the endpoint of pressure calculation, which together constitute the joint judgment condition for the termination of calculation. Establish an accuracy verification mechanism based on the real-time feedback of the pressure-displacement monitoring sequence, and perform rolling comparison and deviation correction between the predicted values ​​and the actual collected values ​​during the simulation process; The prediction time step is dynamically adjusted based on the relative difference between the actual displacement value and the target termination position at the current pressing stage. The computational depth of the simulation process is adaptively determined based on the real-time magnitudes of the predicted pressure deviation and the predicted displacement deviation.

4. The adaptive control method for motor coil press-fitting process parameters according to claim 3, characterized in that, An accuracy verification mechanism based on the real-time feedback of the pressure-displacement monitoring sequence is established to perform rolling comparisons and deviation corrections between predicted and actual collected values ​​during the simulation process, including: At the end of each prediction time step, the predicted pressure value and predicted displacement value output by the process state deduction model are compared with the actual pressure value and actual displacement value in the pressure-displacement monitoring sequence at the same time, and the instantaneous pressure deviation and instantaneous displacement deviation of the current step are calculated respectively. The instantaneous pressure deviation and instantaneous displacement deviation are compared with the allowable process deviation threshold based on the target clamping force and the target termination position; If the instantaneous pressure deviation or instantaneous displacement deviation exceeds the corresponding process deviation threshold, it is determined that the process state simulation model has decreased in simulation accuracy under the current working condition, and deviation correction is initiated. The deviation correction includes fitting the deviation change trend based on the instantaneous pressure deviation and instantaneous displacement deviation sequence within the most recent consecutive prediction time steps, and adjusting the compensation parameters determined by the temperature-material deformation relationship in the process state inference model in reverse. The process state simulation model is updated using the adjusted compensation parameters, and subsequent simulations are performed again based on the latest pressure-displacement monitoring sequence to complete this rolling verification and correction.

5. The adaptive control method for motor coil press-fitting process parameters according to claim 1, characterized in that, If the trend analysis results indicate that both conditions cannot be met simultaneously, then the predicted pressure deviation and predicted displacement deviation are used as optimization objectives. Real-time collected pressure head temperature and coil temperature are introduced as process state compensation quantities. A multi-objective optimization function containing weighted allocation variables is constructed, and the control objective bias coefficient is solved, including: Minimizing the predicted pressure deviation and minimizing the predicted displacement deviation are used as common optimization objectives; Based on the pressure head temperature and coil temperature, determine the proportional coefficient of the influence of the process state compensation on the pressure control target and the displacement control target; Using the aforementioned proportional coefficient, a multi-objective optimization function is constructed with the weighted allocation variable as the solution object, wherein the multi-objective optimization function aims to minimize the expected total deviation after weighting; Solve the multi-objective optimization function to obtain the optimal solution of the weight allocation variable, and normalize the optimal solution as the control objective bias coefficient.

6. The adaptive control method for motor coil press-fitting process parameters according to claim 5, characterized in that, Based on the pressure head temperature and coil temperature, determine the proportional coefficient of the influence of the process state compensation on the pressure control target and the displacement control target, including: Based on the thermal expansion characteristics of the indenter material, the thermal expansion of the indenter temperature relative to standard operating conditions is calculated and mapped to a temperature influence factor on the displacement control target. Based on the thermodynamic properties of the coil insulation material, the rate of change of the equivalent stiffness of the insulation layer caused by the coil temperature is calculated and mapped to a temperature influence factor on the pressure control target. Based on the temperature influence factors on the displacement control target and the temperature influence factors on the pressure control target, a proportional coefficient characterizing the relative influence of the two is calculated through normalization.

7. The adaptive control method for motor coil press-fitting process parameters according to claim 6, characterized in that, Based on the thermal expansion characteristics of the indenter material, the thermal expansion of the indenter temperature relative to standard operating conditions is calculated and mapped to a temperature influence factor on the displacement control target, including: Using the reference temperature of the pressure head under standard working conditions in the historical pressing process database as a benchmark, the real-time temperature difference between the real-time collected pressure head temperature and the pressure head reference temperature is calculated. Based on the preset linear expansion coefficient of the indenter material and the effective working length of the indenter in the displacement measurement direction, the real-time temperature difference is converted into the absolute thermal expansion compensation amount of the indenter in the axial direction. The ratio of the absolute thermal expansion compensation to the target termination position is calculated to obtain the relative influence coefficient, which is then used as the temperature influence factor for the displacement control target.

8. The adaptive control method for motor coil press-fitting process parameters according to claim 6, characterized in that, Based on the thermodynamic properties of the coil insulation material, the rate of change of the equivalent stiffness of the insulation layer caused by the coil temperature is calculated and mapped to a temperature influence factor on the pressure control target, including: Obtain the reference elastic modulus of the insulating material under standard operating conditions from the historical press-fitting process database; Based on the preset temperature-modulus characteristic relationship of the insulation material, and combined with the real-time collected coil temperature, the real-time elastic modulus of the insulation layer at the current temperature is calculated by interpolation. The rate of change of the real-time elastic modulus relative to the reference elastic modulus is calculated as the rate of change of the equivalent stiffness of the insulation layer. The equivalent stiffness change rate of the insulation layer is correlated with the pressure change sensitivity of the standard pressure-displacement reference curve at the end of the press fitting process to obtain the temperature influence factor on the pressure control target.

9. The adaptive control method for motor coil press-fitting process parameters according to claim 1, characterized in that, Based on the target bias coefficient, the hybrid controller at the end of the press-fitting stage is reconstructed. The output commands of the position loop and force loop are dynamically weighted and fused. Based on the fused unified command, the press-fitting control of the target stator coil is continued, including: The control target bias coefficient is used as the initial fusion weight of the force loop command, and the initial fusion weight is dynamically fine-tuned based on the real-time proportional relationship between the predicted pressure deviation and the predicted displacement deviation to obtain the dynamic weight of the force loop. The corresponding complementary value is calculated based on the dynamic weight and used as the dynamic weight of the position loop; Within each control cycle, the position adjustment command generated by the position loop based on the actual displacement value and the force adjustment command generated by the force loop based on the actual pressure value are synchronously acquired. Based on the dynamic weights of the force loop and the dynamic weights of the position loop, the position adjustment command and the force adjustment command are weighted and summed to generate a unified command to drive the actuator. Repeat the above dynamic weighted fusion process until the pressing process ends, while simultaneously satisfying the target termination position and the target clamping force.

10. An adaptive control device for motor coil press-fitting process parameters, characterized in that, An adaptive control method for motor coil press-fitting process parameters as described in any one of claims 1-9 includes: The target parameter acquisition module is used to acquire the press-fit target parameters of the target stator coil, wherein the press-fit target parameters include a preset target termination position and a target clamping force; The data acquisition and sequence generation module is used to start the pressing process, collect the actual pressure value and actual displacement value in real time during the current pressing process, and generate a pressure-displacement monitoring sequence. The deduction and trend analysis module is used to construct a process state deduction model based on the pressure-displacement monitoring sequence, predict whether the target termination position and the target clamping force are simultaneously satisfied when pressing continues until the end of pressing under the current control mode, and output the trend analysis results. The multi-objective optimization and solution module is used to construct a multi-objective optimization function containing weighted allocation variables when the trend analysis results show that they cannot be satisfied simultaneously. It takes the predicted pressure deviation and the predicted displacement deviation as optimization objectives, introduces the real-time collected pressure head temperature and coil temperature as process state compensation quantities, and solves the control objective bias coefficient. The hybrid control and execution module is used to reconstruct the hybrid controller of the final pressing stage based on the control target bias coefficient, dynamically weight and fuse the output commands of the position loop and force loop, and continue to complete the pressing control of the target stator coil based on the fused unified command.

Images