Multi-objective Cooperative Control Method and Device for Direct Drive Motor

Abstract

The present invention discloses a multi-objective collaborative control method and device for a direct drive motor, relating to the technical field of motor control. The method includes: obtaining a motor control scheme corresponding to a predetermined control scenario; establishing a motor control desired vector; obtaining a control desired deviation evaluation result; if the control desired deviation evaluation result does not satisfy the deviation evaluation multi-sided constraint, generating a first guiding factor for control optimization; generating a first motor control strategy set; performing multi-objective joint optimization on the first motor control strategy set to generate a motor control optimization strategy, and controlling the direct drive motor in combination with the predetermined control scenario. The technical problems existing in the prior art, such as the difficulty in coordinating the control of multiple objectives of a direct drive motor, the lack of an effective deviation evaluation and optimization mechanism for complex working conditions, and insufficient adaptability, resulting in insufficient control accuracy and poor operation reliability of the direct drive motor, are solved, and the technical effects of realizing the multi-objective collaborative control of the direct drive motor, improving the control accuracy and operation reliability are achieved.

Description

Technical Field

[0001] This application relates to the technical field of motor control, and particularly to a multi-objective cooperative control method and device for a direct drive motor. Background Art

[0002] Direct drive motors are widely used in fields such as numerical control machine tools, robots, aerospace, and new energy power generation. However, in the actual operation process, direct drive motors face complex and variable working scenarios. Different control tasks put forward multiple interrelated or even contradictory objective requirements for the motor performance. For example, when pursuing high-speed operation, the position control accuracy of the motor may be sacrificed; when emphasizing system stability, the dynamic response speed of the motor may be reduced. Existing direct drive motor control usually optimizes for a single objective, making it difficult to effectively coordinate between multiple objectives and unable to meet the requirements for high-performance and high-reliability operation of the motor under complex working conditions. Moreover, when the working environment changes or the motor fails, there is a lack of an effective evaluation and optimization mechanism for multi-objective deviation and the ability to make quick adjustments, making it difficult to achieve precise control and efficient operation of the direct drive motor under complex and variable working conditions.

[0003] Therefore, in the current related technologies, there are technical problems such as difficult coordinated control of multiple objectives of direct drive motors, lack of an effective deviation evaluation and optimization mechanism under complex working conditions, and insufficient adaptability, resulting in insufficient control accuracy and poor operation reliability of direct drive motors. Summary of the Invention

[0004] By providing a multi-objective cooperative control method and device for a direct drive motor, this application solves the technical problems in the prior art, such as difficult coordinated control of multiple objectives of direct drive motors, lack of an effective deviation evaluation and optimization mechanism under complex working conditions, and insufficient adaptability, resulting in insufficient control accuracy and poor operation reliability of direct drive motors, and achieves the technical effects of realizing multi-objective cooperative control of direct drive motors, improving control accuracy, and operation reliability.

[0005] The present application provides a multi-objective collaborative control method for a direct drive motor. The method includes: interacting with a control terminal of the direct drive motor to obtain a motor control scheme corresponding to a predetermined control scenario; performing multi-objective expectation prediction on the direct drive motor according to the predetermined control scenario to establish a motor control expectation vector; based on the motor control scheme, performing multi-objective deviation evaluation on the direct drive motor according to the motor control expectation vector to obtain a control expectation deviation evaluation result; if the control expectation deviation evaluation result does not satisfy the deviation evaluation multi-sided constraint, tracing control-related features according to the control expectation deviation evaluation result to generate a first guiding factor for control optimization; performing optimization adjustment on the motor control scheme according to the first guiding factor for control optimization to generate a first set of motor control strategies; performing multi-objective joint optimization on the first set of motor control strategies according to the motor control expectation vector and the deviation evaluation multi-sided constraint to generate a motor control optimization strategy, and controlling the direct drive motor in combination with the predetermined control scenario.

[0006] In a possible implementation manner, the multi-objective collaborative control method for a direct drive motor further performs the following processing: performing energy consumption expectation prediction on the direct drive motor according to the predetermined control scenario to obtain an energy consumption feature expectation; performing temperature rise feature expectation prediction on the direct drive motor according to the predetermined control scenario to obtain a temperature rise feature expectation; performing torque fluctuation expectation prediction on the direct drive motor according to the predetermined control scenario to obtain a torque fluctuation feature expectation; constructing the motor control expectation vector according to the energy consumption feature expectation, the temperature rise feature expectation, and the torque fluctuation feature expectation.

[0007] In a possible implementation manner, the multi-objective collaborative control method for a direct drive motor further performs the following processing: interconnecting motors of the same model according to the direct drive motor to obtain a cluster of motors; performing normal energy consumption sample retrieval on the cluster of motors according to the predetermined control scenario to obtain a set of scenario energy consumption samples; performing confidence evaluation on the set of scenario energy consumption samples to obtain an energy consumption confidence evaluation set; performing data fusion on the set of scenario energy consumption samples according to the energy consumption confidence evaluation set to generate the energy consumption feature expectation.

[0008] In a possible implementation, the multi-objective collaborative control method for the direct drive motor further performs the following processing: performing multi-objective prediction on the direct drive motor according to the motor control scheme, and establishing a motor control prediction vector; performing multi-objective comparison on the motor control prediction vector according to the motor control desired vector to obtain an energy consumption comparison result, a temperature rise comparison result, and a torque ripple comparison result; respectively performing deviation evaluation on the energy consumption comparison result, the temperature rise comparison result, and the torque ripple comparison result to obtain an energy consumption deviation evaluation coefficient, a temperature rise deviation evaluation coefficient, and a torque ripple deviation evaluation coefficient; performing deviation loss evaluation on the direct drive motor according to the energy consumption comparison result, the temperature rise comparison result, and the torque ripple comparison result to obtain a deviation loss evaluation coefficient; adding the energy consumption deviation evaluation coefficient, the temperature rise deviation evaluation coefficient, the torque ripple deviation evaluation coefficient, and the deviation loss evaluation coefficient to the control desired deviation evaluation result.

[0009] In a possible implementation, the multi-objective collaborative control method for the direct drive motor further performs the following processing: if the control desired deviation evaluation result does not satisfy the multi-sided constraint of deviation evaluation, determining an abnormal deviation evaluation result and a deviation evaluation abnormal element, performing sensitivity analysis on the motor control variable set of the motor control scheme according to the deviation evaluation abnormal element, and generating a control optimization sensitive variable; performing gap identification on the abnormal deviation evaluation result according to the multi-sided constraint of deviation evaluation to obtain a deviation evaluation gap feature; performing adjustment constraint configuration on the control optimization sensitive variable according to the deviation evaluation gap feature, and generating a first guiding factor for control optimization.

[0010] In a possible implementation, the multi-objective collaborative control method for the direct drive motor further performs the following processing: respectively performing random perturbation on the motor control scheme according to the motor control variable set to obtain a set of variable perturbation control schemes; respectively performing simulation control on the direct drive motor according to the set of variable perturbation control schemes to obtain a set of control simulation data; performing response trend analysis on the set of control simulation data according to the deviation evaluation abnormal element to obtain a response trend of each variable perturbation; performing sensitivity evaluation on the motor control variable set according to the response trend of each variable perturbation to obtain a sensitivity of each variable; based on the sensitivity of each variable, screening the motor control variable set according to a predetermined sensitivity to generate a control optimization sensitive variable.

[0011] In a possible implementation manner, the multi-objective collaborative control method for the direct drive motor further performs the following processing: Based on the motor control desired vector, perform multi-objective deviation evaluation on each motor control strategy in the first motor control strategy set to obtain the deviation evaluation results of each strategy; determine whether the deviation evaluation results of each strategy satisfy the deviation evaluation multi-sided constraint; if any one of the deviation evaluation results of the deviation evaluation results of each strategy satisfies the deviation evaluation multi-sided constraint, obtain the motor control optimization strategy.

[0012] In a possible implementation manner, the multi-objective collaborative control method for the direct drive motor further performs the following processing: If the deviation evaluation results of each strategy do not satisfy the deviation evaluation multi-sided constraint, identify the approximation trend of the deviation evaluation multi-sided constraint according to the deviation evaluation results of each strategy to obtain the evaluation approximation constraint trend; identify the control correlation feature of the first motor control strategy set according to the evaluation approximation constraint trend to obtain the approximation trend control correlation feature; optimize the control optimization first guiding factor according to the approximation trend control correlation feature to obtain the control optimization second guiding factor; adjust the first motor control strategy set according to the control optimization second guiding factor to obtain the second motor control strategy set; perform multi-objective joint optimization on the second motor control strategy set according to the motor control desired vector and the deviation evaluation multi-sided constraint to generate the motor control optimization strategy.

[0013] In a possible implementation manner, the multi-objective collaborative control method for the direct drive motor further performs the following processing: The deviation evaluation multi-sided constraint includes an energy consumption deviation evaluation constraint, a temperature rise deviation evaluation constraint, a torque ripple deviation evaluation constraint, and a deviation loss evaluation constraint.

[0014] The present application also provides a multi-objective collaborative control device for a direct drive motor. The device includes: a motor control scheme acquisition module for interacting with the control terminal of the direct drive motor to obtain a motor control scheme corresponding to a predetermined control scenario; a multi-objective expectation prediction module for performing multi-objective expectation prediction on the direct drive motor according to the predetermined control scenario to establish a motor control expectation vector; a multi-objective deviation evaluation module for performing multi-objective deviation evaluation on the direct drive motor based on the motor control scheme according to the motor control expectation vector to obtain a control expectation deviation evaluation result; a control-related feature tracing module for, if the control expectation deviation evaluation result does not meet the multi-sided constraints of the deviation evaluation, tracing control-related features according to the control expectation deviation evaluation result to generate a first guiding factor for control optimization; an optimization adjustment module for optimizing and adjusting the motor control scheme according to the first guiding factor for control optimization to generate a first set of motor control strategies; and an optimization strategy generation module for performing multi-objective joint optimization on the first set of motor control strategies according to the motor control expectation vector and the multi-sided constraints of the deviation evaluation to generate a motor control optimization strategy, and controlling the direct drive motor in combination with the predetermined control scenario.

[0015] It is intended to solve the technical problems existing in the prior art, such as the difficulty in coordinating the control of multiple objectives of a direct drive motor, the lack of an effective deviation evaluation and optimization mechanism under complex working conditions, and insufficient adaptability, resulting in insufficient control accuracy and poor operation reliability of the direct drive motor, by the multi-objective collaborative control method and device for a direct drive motor proposed in the present application, including obtaining a motor control scheme corresponding to a predetermined control scenario; establishing a motor control expectation vector; obtaining a control expectation deviation evaluation result; generating a first guiding factor for control optimization if the control expectation deviation evaluation result does not meet the multi-sided constraints of the deviation evaluation; generating a first set of motor control strategies; performing multi-objective joint optimization on the first set of motor control strategies to generate a motor control optimization strategy, and controlling the direct drive motor in combination with the predetermined control scenario. The technical effect of realizing the multi-objective collaborative control of the direct drive motor, improving the control accuracy and operation reliability is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. Flowcharts are used in the present application to illustrate the operations performed by the devices according to the embodiments of the present application. It should be understood that the operations before or below do not necessarily need to be executed precisely in sequence. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0017] Figure 1 It is a schematic flowchart of the multi-objective collaborative control method for a direct drive motor provided by the embodiment of the present application.

[0018] Figure 2 This is a schematic structural diagram of a multi-objective collaborative control device for a direct drive motor provided by an embodiment of the present application.

[0019] Explanation of reference numerals: Motor control scheme acquisition module 10, multi-objective expected prediction module 20, multi-objective deviation evaluation module 30, control correlation feature tracing module 40, optimization adjustment module 50, optimization strategy generation module 60. Specific implementation manners

[0020] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the specific implementation manners of the present application are specifically given below.

[0021] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0022] In the following descriptions, "some embodiments" are involved, which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first\second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product or server including a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0023] An embodiment of the present application provides a multi-objective collaborative control method for a direct drive motor, as Figure 1 shown. The method includes:

[0024] Step S100, interact with the control terminal of the direct drive motor to obtain a motor control scheme corresponding to a predetermined control scenario.

[0025] Preferably, for the control terminal of the interactive direct drive motor, the control terminal is a platform for information interaction with the direct drive motor control system, which may be an industrial control computer, a human-machine interface of a programmable logic controller (PLC), a touch screen operation panel, etc. Operators input various instructions, parameters, and operation requirements through the control terminal, and can also obtain the operating status, feedback information, etc. of the direct drive motor from the control terminal; then obtain the motor control scheme corresponding to the predetermined control scenario, where the predetermined control scenario is a working mode or operating condition preset according to different working conditions and task requirements in the actual application of the direct drive motor. For example, in robot applications, different operation scenarios such as carrying heavy objects, fine assembly, and rapid movement, and the performance requirements of each predetermined control scenario, such as speed, accuracy, torque, etc. are different; for different predetermined control scenarios, corresponding motor control schemes are set, usually including the selection of control algorithms, the setting of parameters, and the planning of control strategies, etc. Through interaction with the control terminal, operators can select the appropriate predetermined control scenario according to the current work task, and then obtain the corresponding motor control scheme to accurately control the direct drive motor.

[0026] Step S200, perform multi-objective expectation prediction on the direct drive motor according to the predetermined control scenario, and establish a motor control expectation vector.

[0027] Preferably, step S200 further includes step S210, perform energy consumption expectation prediction on the direct drive motor according to the predetermined control scenario, and obtain the energy consumption characteristic expectation; step S220, perform temperature rise characteristic expectation prediction on the direct drive motor according to the predetermined control scenario, and obtain the temperature rise characteristic expectation; step S230, perform torque fluctuation expectation prediction on the direct drive motor according to the predetermined control scenario, and obtain the torque fluctuation characteristic expectation; step S240, construct the motor control expectation vector according to the energy consumption characteristic expectation, the temperature rise characteristic expectation, and the torque fluctuation characteristic expectation.

[0028] Preferably, different control scenarios have different requirements for motor energy efficiency. According to the running speed, load condition, running time, running trajectory, etc. of the motor, calculate the theoretical energy consumption through the formula where, is the energy consumption expectation, is the instantaneous power, is the control algorithm efficiency, is the motor efficiency; the resistance loss of the motor winding may cause the temperature to rise, and too high a temperature rise will reduce the efficiency and affect the life. Adopt the thermal resistance network model to predict the temperature rise characteristic expectation considering factors such as copper loss, iron loss, and heat dissipation coefficient: where, is the temperature characteristic expectation, is the total loss power, is the thermal resistance, is the heating time constant of the motor; cogging effect, current harmonics, etc. of the motor may cause torque fluctuations and affect the system stability. Based on the motor electromagnetic model, the relationship between the current harmonic content and the cogging torque is analyzed: , where is the expected torque fluctuation, is the torque amplitude of each harmonic, is the electrical angular velocity. Finally, the sub-item prediction results are integrated into a multi-dimensional vector to obtain the motor control expected vector, which comprehensively describes the expectations for the control performance of the direct drive motor under the predetermined control scenario.

[0029] Furthermore, step S210 further includes step S211 of interconnecting motors of the same model according to the direct drive motor to obtain a cluster of motors; step S212 of retrieving normal energy consumption samples of the cluster of motors according to the predetermined control scenario to obtain a scenario energy consumption sample set; step S213 of performing a confidence evaluation according to the scenario energy consumption sample set to obtain an energy consumption confidence evaluation set; and step S214 of performing data fusion on the scenario energy consumption sample set according to the energy consumption confidence evaluation set to generate the energy consumption characteristic expectation.

[0030] Preferably, multiple direct drive motors of the same model are interconnected with the target direct drive motor to better analyze and control the target direct drive motor by using the operation data of multiple motors of the same model. For example, multiple direct drive motors of the same model are connected to a central control unit so that they can share operation status information, control instructions, etc., and then form a cluster of motors; then, for the predetermined control scenario, data samples with normal energy consumption are selected from the operation data of the cluster of motors. Specifically, the normal range of energy consumption is determined according to the technical specifications and historical operation data of the direct drive motor, and then from a large amount of operation data of the cluster of motors in this scenario, energy consumption data samples that meet the normal range are selected. The set of all energy consumption data samples obtained through the retrieval of normal energy consumption samples is the scenario energy consumption sample set, which contains various energy consumption data of the cluster of motors under the predetermined control scenario and reflects the normal energy consumption characteristics of motors of the same model in this scenario.

[0031] Preferably, according to the source reliability of the data, the accuracy of data collection, the consistency of the data with other relevant data, etc., a confidence evaluation is performed on each data sample in the scenario energy consumption sample set to determine its reliability and representativeness. For example, if a data sample is collected by a calibrated high-precision sensor and is relatively consistent with the data of other motors in the same cluster under similar operating conditions, the confidence of this data sample is relatively high. Finally, the confidence evaluation results of each data sample in the scenario energy consumption sample set are sorted into a set, that is, an energy consumption confidence evaluation set is obtained. Corresponding to the scenario energy consumption sample set, each element represents the confidence of the corresponding energy consumption data sample. For example, the energy consumption confidence evaluation set can be an array with the same number of elements as the scenario energy consumption sample set, where the value of each element is between 0 and 1, representing the confidence level of the corresponding energy consumption data sample. The closer the value is to 1, the higher the confidence.

[0032] Preferably, data fusion is performed on the scenario energy consumption sample set according to the energy consumption confidence evaluation set, that is, considering the confidence of each data sample comprehensively, the scenario energy consumption sample set is processed. For example, weighted averaging of the data samples is performed according to the confidence. Among them, for data samples with higher confidence, a larger weight is given, while for data samples with lower confidence, a smaller weight is given. Through calculation, the influence of abnormal data or low-confidence data on the result can be reduced, and a more accurate result that can better represent the motor energy consumption characteristics under the predetermined control scenario is obtained, that is, the expected energy consumption characteristics are finally obtained, which are used to evaluate whether the actual energy consumption of the motor meets the expectations, and the energy consumption of the motor is made as close as possible to this expected characteristic through the adjustment of the control strategy.

[0033] Step S300, based on the motor control scheme, perform a multi-objective deviation evaluation on the direct drive motor according to the motor control expectation vector to obtain a control expectation deviation evaluation result.

[0034] Step S300 further includes step S310, perform multi-objective prediction on the direct drive motor according to the motor control scheme to establish a motor control prediction vector; step S320, perform multi-objective comparison on the motor control prediction vector according to the motor control expectation vector to obtain an energy consumption comparison result, a temperature rise comparison result, and a torque fluctuation comparison result; step S330, respectively perform deviation evaluation on the energy consumption comparison result, the temperature rise comparison result, and the torque fluctuation comparison result to obtain an energy consumption deviation evaluation coefficient, a temperature rise deviation evaluation coefficient, and a torque fluctuation deviation evaluation coefficient; step S340, perform deviation loss evaluation on the direct drive motor according to the energy consumption comparison result, the temperature rise comparison result, and the torque fluctuation comparison result to obtain a deviation loss evaluation coefficient; step S350, add the energy consumption deviation evaluation coefficient, the temperature rise deviation evaluation coefficient, the torque fluctuation deviation evaluation coefficient, and the deviation loss evaluation coefficient to the control expectation deviation evaluation result.

[0035] Preferably, key indicators such as energy consumption, temperature rise characteristics, and torque ripple during the operation of the direct drive motor are predicted according to the motor control scheme. Specifically, a suitable prediction model is selected according to the prediction target. For example, regression trees, random forests, or long short-term memory networks (LSTM) are used to capture the non-linear relationship between energy consumption and load, speed, and time for energy consumption prediction; a hybrid neural network (such as a physics-informed neural network PINN) combined with a heat conduction physical model is used to consider the heat accumulation effect for temperature rise prediction; a recurrent neural network (RNN) or a convolutional neural network (CNN) is used to analyze the correlation between current harmonics and torque pulsation for torque ripple prediction; the true energy consumption, temperature rise, and torque ripple values under different control scenarios are extracted from the historical operation data of motors in the same cluster (interconnected motors of the same model) and labeled, and independent prediction models are trained for energy consumption, temperature rise, and torque ripple respectively; then, a feature vector is constructed based on the parameters (such as voltage, current, frequency, control algorithm parameters) and operating states (such as load torque, speed, ambient temperature) in the motor control scheme, that is, the three prediction results are integrated into a motor control prediction vector, including the energy consumption value (such as kW·h) predicted by the machine learning model, the predicted temperature rise characteristics (such as the highest temperature, temperature rise rate), and the predicted torque ripple amplitude (such as N·m) or the fluctuation frequency distribution, comprehensively describing the prediction status of each key indicator of the direct drive motor under the current motor control scheme.

[0036] Preferably, the motor control prediction vector is compared with the motor control expectation vector, that is, by comparing each element in the prediction vector with the corresponding element in the expectation vector, the difference between the predicted value and the expected value is calculated. Through multi-objective comparison, the energy consumption comparison result, temperature rise comparison result, and torque ripple comparison result are obtained, respectively reflecting the gaps between the predicted energy consumption, temperature rise, and torque ripple and their respective expected target values. For example, the energy consumption comparison result is used to quantify the deviation degree between the prediction and the expectation in terms of energy consumption, and the temperature rise comparison result and torque ripple comparison result are the same, respectively used to quantify the deviation degree between the prediction and the expectation in terms of temperature rise and torque ripple. For the energy consumption comparison result, temperature rise comparison result, and torque ripple comparison result, deviation evaluations are respectively carried out, that is, according to specific evaluation criteria, the deviation degree between the predicted value and the expected value of each indicator is further quantified to more accurately evaluate the execution effect of the motor control scheme. For example, a threshold range is set. When the comparison result is within the threshold range, it is considered that the deviation degree is small and a lower deviation evaluation coefficient is given; when the comparison result exceeds the threshold range, a corresponding higher deviation evaluation coefficient is given according to the exceeding amplitude; and then the energy consumption deviation evaluation coefficient, temperature rise deviation evaluation coefficient, and torque ripple deviation evaluation coefficient are obtained, respectively used to intuitively reflect the deviation degree of each indicator. The larger the value, the greater the deviation degree between the predicted value and the expected value of the corresponding indicator, indicating that the control effect of the current motor control scheme on this indicator is less ideal.

[0037] Preferably, the deviation loss of the direct drive motor is evaluated according to the energy consumption comparison result, the temperature rise comparison result, and the torque ripple comparison result, that is, the comprehensive loss brought by the deviation of each index from the expected target to the motor operation is evaluated as a whole according to the importance weights of different indexes, the influence of the deviation degree on the motor performance and life, etc. For example, in an application scenario with high precision requirements, the deviation of torque ripple may have a greater impact on the system, and a larger weight will be given to the torque ripple comparison result in the deviation loss evaluation; while for some scenarios that are more sensitive to energy consumption, the weight of the energy consumption comparison result may be relatively high; thus, a deviation loss evaluation coefficient is obtained, which reflects the overall loss degree caused by the deviation of the direct drive motor from the expected target on multiple target indexes. Finally, the energy consumption deviation evaluation coefficient, the temperature rise deviation evaluation coefficient, the torque ripple deviation evaluation coefficient, and the deviation loss evaluation coefficient are added to the control expectation deviation evaluation result to more detailed and accurately reflect the deviation situation in each aspect and the overall deviation loss degree.

[0038] Step S400, if the control expectation deviation evaluation result does not meet the deviation evaluation multilateral constraint, trace the control-related features according to the control expectation deviation evaluation result to generate the first control optimization guiding factor.

[0039] Step S400 further includes that the deviation evaluation multilateral constraint includes the energy consumption deviation evaluation constraint, the temperature rise deviation evaluation constraint, the torque ripple deviation evaluation constraint, and the deviation loss evaluation constraint.

[0040] Step S400 further includes step S410. If the control expectation deviation evaluation result does not meet the deviation evaluation multilateral constraint, determine the abnormal deviation evaluation result and the deviation evaluation abnormal element. Step S420, perform sensitivity analysis on the motor control variable set of the motor control scheme according to the deviation evaluation abnormal element to generate the control optimization sensitive variable. Step S430, identify the gap according to the deviation evaluation multilateral constraint for the abnormal deviation evaluation result to obtain the deviation evaluation gap feature. Step S440, configure the adjustment constraint for the control optimization sensitive variable according to the deviation evaluation gap feature to generate the first control optimization guiding factor.

[0041] Preferably, the deviation evaluation multilateral constraint is a standard for measuring whether the deviation between the actual operation and the expected situation of the direct drive motor in multiple aspects is within an acceptable range, including the energy consumption deviation evaluation constraint, the temperature rise deviation evaluation constraint, the torque ripple deviation evaluation constraint, and the deviation loss evaluation constraint. Among them, the energy consumption deviation evaluation constraint is a constraint condition set for the energy consumption aspect of the direct drive motor, which stipulates the allowable deviation range between the actual energy consumption of the direct drive motor and the expected energy consumption characteristics under a predetermined control scenario; the temperature rise deviation evaluation constraint is used to measure the temperature rise situation of the direct drive motor, and clarifies the acceptable deviation between the actual temperature rise characteristics of the motor and the expected temperature rise characteristics under a specific control scenario; the torque ripple deviation evaluation constraint is a limiting condition set for the torque ripple of the direct drive motor, which limits the allowable fluctuation range between the actual torque ripple characteristics of the motor and the expected torque ripple characteristics under a given control scenario; the deviation loss evaluation constraint is a constraint condition for evaluating the deviation loss generated by comprehensively considering multiple factors such as energy consumption, temperature rise, and torque ripple, so as to measure the degree of deviation of the direct drive motor from the expected state in terms of overall performance.

[0042] Preferably, the control expectation deviation evaluation result is the comprehensive result of the deviation evaluation coefficients in aspects such as energy consumption, temperature rise, and torque ripple. Specifically, if the control expectation deviation evaluation result does not meet the deviation evaluation multilateral constraint, it means that there is a large deviation between the actual operation of the direct drive motor and the expected control target. At this time, determine the specific abnormal deviation evaluation result, that is, clarify which aspects of the deviation exceed the constraint range, such as excessive energy consumption deviation, abnormal temperature rise or torque ripple; at the same time, find out the corresponding abnormal deviation evaluation elements, that is, the specific factors causing the abnormality, such as an unreasonable setting of a certain control parameter or a problem with a certain component of the motor; then conduct a sensitivity analysis on the motor control variable set of the motor control scheme, that is, evaluate the influence degree of the control variable on the abnormal deviation evaluation element, and regard the control variable with a greater influence on the abnormal deviation as the control optimization sensitive variable. For example, for the abnormal deviation of the motor's energy consumption, through sensitivity analysis, it is determined that the change in voltage has a very significant impact on the energy consumption, and voltage is the control optimization sensitive variable.

[0043] Preferably, referring again to the deviation evaluation multilateral constraint, the abnormal deviation evaluation result is compared with the constraint condition to identify the gap between the two, such as the numerical difference, the difference in the change trend, etc., which is used as the deviation evaluation gap feature. For example, the deviation evaluation multilateral constraint stipulates that the deviation of energy consumption cannot exceed ±5%, while the actual energy consumption deviation evaluation result reaches +10%, and the gap is 5%. At the same time, it is also determined that the energy consumption exceeds the upper limit of the constraint, showing a positive deviation trend. Finally, according to the identified deviation evaluation gap feature, it is determined how to adjust and constrain the control optimization sensitive variables. For example, if the deviation evaluation gap feature of energy consumption shows that the energy consumption is too high and the voltage is the control optimization sensitive variable, the set value of the voltage is reduced, and a reasonable adjustment range and constraint conditions are set to ensure that while reducing the energy consumption, it will not have an adverse impact on other performances of the motor. Among them, the adjustment constraint configuration of the control optimization sensitive variable is to generate the first guiding factor for control optimization. For example, the first guiding factor is a parameter set, including how to adjust the control variable, as well as the adjustment amplitude and range, etc., so that the motor operation can be closer to the expected control target and meet the deviation evaluation multilateral constraint.

[0044] Further, step S420 further includes step S421, randomly disturbing the motor control scheme according to the motor control variable set to obtain each variable disturbance control scheme set; step S422, respectively performing simulation control on the direct drive motor according to each variable disturbance control scheme set to obtain each control simulation data set; step S423, performing response trend analysis on each control simulation data set according to the deviation evaluation abnormal element to obtain each variable disturbance response trend; step S424, performing sensitivity evaluation on the motor control variable set according to each variable disturbance response trend to obtain each variable sensitivity; step S425, based on each variable sensitivity, screening the motor control variable set according to a predetermined sensitivity to generate the control optimization sensitive variable.

[0045] Preferably, randomly disturbing the motor control scheme according to the motor control variable set means that on the basis of the original motor control scheme, each variable is randomly adjusted, such as increasing or decreasing the value, or changing its value within a certain range, etc., to obtain different control schemes, that is, each variable disturbance control scheme set, which is used to observe the influence of variable changes on the motor operation. Then, using each variable disturbance control scheme set, respectively perform simulation operation control on the direct drive motor. During the simulation control process, record various relevant data, such as the data of the motor's energy consumption, temperature rise, torque, etc. changing with time, to form each control simulation data set, reflecting the actual operation performance of the direct drive motor under different variable disturbance conditions.

[0046] Preferably, for each control simulation dataset, by combining these deviation evaluation abnormal elements, analyze the variation trend of the part related to the abnormal elements in the data with the perturbation of variables. For example, if energy consumption is a deviation evaluation abnormal element, focus on analyzing the variation of energy consumption data in each control simulation dataset with different variable perturbations, whether it increases, decreases, or remains stable, as well as the amplitude of the variation, etc., so as to obtain the variable perturbation response trend. Then, according to the variable perturbation response trend, evaluate the sensitivity of each motor control variable to the deviation evaluation abnormal element. If a small perturbation of a variable can cause a large change in the deviation evaluation abnormal element (such as energy consumption, temperature rise, or torque fluctuation), it means that this variable has a high sensitivity to this abnormal element; conversely, if the perturbation of the variable has a small impact on the abnormal element, the sensitivity is low; finally, obtain the sensitivities of each variable, which are used to quantify the influence degree of each variable on the abnormal operation of the motor. Then, according to the preset sensitivity standard, screen the motor control variable set, select the variables with sensitivities higher than the preset sensitivity as the control optimization sensitive variables, and then adjust the control optimization sensitive variables to make them closer to the expected control target, reduce the influence of the deviation evaluation abnormal elements, and improve the stability and efficiency of the motor operation.

[0047] Step S500, perform an optimization adjustment on the motor control scheme according to the first control optimization guiding factor to generate a first set of motor control strategies.

[0048] Preferably, the first control optimization guiding factor clarifies the variables that need to be optimized keyly in the motor control scheme, the optimization direction (such as increasing or decreasing a certain parameter), and the optimization constraint conditions (such as the range of parameter adjustment). Specifically, according to the first control optimization guiding factor, formulate a specific optimization adjustment strategy. For example, perform optimization using the genetic algorithm, encode the control variables in the motor control scheme as chromosomes, set the fitness function according to the guiding factor, such as comprehensively calculating the fitness with evaluation indicators such as energy consumption, temperature rise, and torque fluctuation, and continuously iterate through operations such as selection, crossover, and mutation to gradually find a better control scheme; that is, adjust the control variables in the motor control scheme and follow the constraint conditions in the first control optimization guiding factor to ensure that the adjusted scheme is feasible and safe in actual operation. For example, the guiding factor stipulates the upper limit of the current. During the optimization adjustment, the value of the current variable cannot exceed this upper limit. After each adjustment of the control variable, conduct a simulation control test to evaluate the impact of the adjusted scheme on the operation performance of the direct-drive motor, and calculate the energy consumption, temperature rise, torque fluctuation, etc. under the new scheme; through continuous optimization adjustment, finally obtain multiple motor control schemes that meet the requirements of the first control optimization guiding factor and have better performance, which form the first set of motor control strategies, and may include schemes focusing on reducing energy consumption, schemes focusing on improving stability and reducing torque fluctuation, and schemes balancing energy consumption and temperature rise, etc.

[0049] Step S600: Based on the motor control desired vector and the deviation evaluation multi-sided constraint, perform multi-objective joint optimization on the first motor control strategy set to generate a motor control optimization strategy, and control the direct drive motor in combination with the predetermined control scenario.

[0050] Step S600 further includes step S610: Based on the motor control desired vector, perform multi-objective deviation evaluation on each motor control strategy in the first motor control strategy set to obtain the deviation evaluation results of each strategy; step S620: Determine whether the deviation evaluation results of each strategy satisfy the deviation evaluation multi-sided constraint; step S630: If any one of the deviation evaluation results of each strategy satisfies the deviation evaluation multi-sided constraint, obtain the motor control optimization strategy.

[0051] Preferably, for each motor control strategy in the first motor control strategy set, compare and analyze the various performance indicators during its actual operation with the corresponding indicators in the motor control desired vector, calculate the degree of deviation of each strategy from the desired goals in multiple objectives such as energy consumption, temperature rise, and torque ripple, so as to obtain the deviation evaluation results of each strategy, and then determine whether the deviation evaluation results of each strategy satisfy the deviation evaluation multi-sided constraint. Specifically, compare the deviation evaluation results of each strategy with conditions such as the energy consumption deviation evaluation constraint, the temperature rise deviation evaluation constraint, the torque ripple deviation evaluation constraint, and the deviation loss evaluation constraint one by one to determine whether the degree of deviation of each strategy is within the allowable range. For example, the energy consumption deviation evaluation constraint may stipulate that the energy consumption cannot exceed 20% of the desired energy consumption, and the temperature rise deviation evaluation constraint stipulates that the temperature rise cannot exceed 10 °C of the desired temperature rise. If the energy consumption and temperature rise deviations of a certain strategy are within their respective constraint ranges, and the torque ripple also meets the corresponding constraint, and the deviation loss evaluation also meets the requirements, it is considered that this strategy satisfies the deviation evaluation multi-sided constraint.

[0052] Preferably, if the deviation evaluation results of at least one strategy satisfy the deviation evaluation multi-sided constraint, this (or these) strategy(s) that meet the conditions is / are determined as the motor control optimization strategy, that is, a control strategy that can balance multiple performance indicators such as the energy consumption, temperature rise, and torque ripple of the motor to a certain extent to meet the pre-set constraint conditions and desired goals, and is used as the final motor control scheme for actual application. Apply the generated motor control optimization strategy to the predetermined control scenario to control the direct drive motor, and make appropriate adjustments and parameter configurations to the optimization strategy according to the characteristics and requirements of the specific scenario, so that the direct drive motor can achieve optimized operation under different predetermined control scenarios and meet various requirements of actual applications.

[0053] Further, step S600 further includes step S640. If the deviation evaluation results of the respective strategies do not satisfy the deviation evaluation multi-sided constraints, trend recognition of approaching the deviation evaluation multi-sided constraints is performed according to the deviation evaluation results of the respective strategies to obtain an evaluation approaching constraint trend; step S650, control correlation feature recognition is performed on the first motor control strategy set according to the evaluation approaching constraint trend to obtain an approaching trend control correlation feature; step S660, according to the approaching trend control correlation feature, the first control optimization guiding factor is optimized to obtain a second control optimization guiding factor; step S670, the first motor control strategy set is adjusted according to the second control optimization guiding factor to obtain a second motor control strategy set; step S680, multi-objective joint optimization is performed on the second motor control strategy set according to the motor control expected vector and the deviation evaluation multi-sided constraints to generate the motor control optimization strategy.

[0054] Preferably, if the deviation evaluation results of the respective strategies do not satisfy the deviation evaluation multi-sided constraints, that is, all the strategies in the current first motor control strategy set fail to meet the predetermined constraint requirements in multiple objectives such as energy consumption, temperature rise, and torque fluctuation, trend recognition of approaching the deviation evaluation multi-sided constraints is performed according to the deviation evaluation results of the strategies that do not satisfy the constraints, including analyzing the change in the gap between the deviation evaluation results of the respective strategies and the constraints, such as analyzing whether the degree of energy consumption deviation is increasing, fluctuating within a certain range, or whether the temperature rise deviation has a trend of gradually approaching the constraint boundary, etc., so as to obtain the evaluation approaching constraint trend, that is, to understand the trend information of each strategy approaching or departing from the constraints in different objectives.

[0055] Preferably, control correlation feature recognition is performed on the first motor control strategy set according to the evaluation approaching constraint trend, that is, the correlation relationship between the control variables and parameters related to the approaching trend is determined. For example, the approaching trend of energy consumption deviation is closely related to the voltage control variable of the motor, and the approaching trend of temperature rise is related to the speed control parameter of the cooling fan, etc., so as to determine the correlation features of the control variables and parameters, so as to understand which control factors affect the deviation evaluation results of the strategies and the interaction relationship between these factors. Then, according to the approaching trend control correlation feature, the first control optimization guiding factor is optimized, that is, the first guiding factor is improved. For example, the first guiding factor mainly refers to the influence of voltage on energy consumption, but according to the control correlation feature, it is found that current and the speed of the cooling fan also have an important impact on energy consumption and temperature rise. When optimizing, the new correlation factors are incorporated into the guiding factor, and the adjustment strategy and range of the control variables are adjusted, etc., so as to obtain the second control optimization guiding factor. Then, the first motor control strategy set is adjusted according to the second control optimization guiding factor, that is, according to the new adjustment direction and range of the control variables in the second guiding factor, each strategy in the first motor control strategy set is modified and adjusted to generate the second motor control strategy set.

[0056] Preferably, multi-objective joint optimization is performed on the second motor control strategy set according to the motor control desired vector and the deviation evaluation multi-sided constraint, that is, the deviation of multiple objectives (such as energy consumption, temperature rise, torque ripple, etc.) from the motor control desired vector is considered simultaneously, as well as whether the deviation evaluation multi-sided constraint is satisfied. Through the multi-objective joint optimization process, the optimal strategy is screened out from the second motor control strategy set to generate a motor control optimization strategy, which includes a control scheme that can make the direct drive motor reach the optimal operating state under a predetermined control scenario. For example, this strategy can make the motor operate with the highest efficiency and the most stable performance on the premise of meeting the constraints such as energy consumption, temperature rise, and torque ripple.

[0057] In the above text, reference is made to Figure 1 A multi-objective cooperative control method for a direct drive motor according to an embodiment of the present invention is described in detail. Next, reference will be made to Figure 2 Describe a multi-objective cooperative control device for a direct drive motor according to an embodiment of the present invention.

[0058] The multi-objective cooperative control device for a direct drive motor according to an embodiment of the present invention is used to solve the technical problems in the prior art that it is difficult to coordinate the multi-objectives of a direct drive motor, there is a lack of an effective deviation evaluation and optimization mechanism for complex working conditions, and the adaptability is insufficient, resulting in insufficient control accuracy and poor operation reliability of the direct drive motor. It achieves the technical effect of realizing the multi-objective cooperative control of the direct drive motor, improving the control accuracy and operation reliability. As Figure 2 shown, the multi-objective cooperative control device for a direct drive motor includes: a motor control scheme acquisition module 10, a multi-objective desired prediction module 20, a multi-objective deviation evaluation module 30, a control correlation feature tracing module 40, an optimization adjustment module 50, and an optimization strategy generation module 60.

[0059] The motor control scheme acquisition module 10 is used to interact with the control terminal of the direct-drive motor to obtain the motor control scheme corresponding to a predetermined control scenario; the multi-objective expectation prediction module 20 is used to perform multi-objective expectation prediction on the direct-drive motor according to the predetermined control scenario and establish a motor control expectation vector; the multi-objective deviation evaluation module 30 is used to perform multi-objective deviation evaluation on the direct-drive motor based on the motor control scheme according to the motor control expectation vector to obtain a control expectation deviation evaluation result; the control-related feature tracing module 40 is used to, if the control expectation deviation evaluation result does not meet the deviation evaluation multi-sided constraint, perform control-related feature tracing according to the control expectation deviation evaluation result and generate a first control optimization guiding factor; the optimization adjustment module 50 is used to perform optimization adjustment on the motor control scheme according to the first control optimization guiding factor and generate a first motor control strategy set; the optimization strategy generation module 60 is used to perform multi-objective joint optimization on the first motor control strategy set according to the motor control expectation vector and the deviation evaluation multi-sided constraint, generate a motor control optimization strategy, and control the direct-drive motor in combination with the predetermined control scenario.

[0060] Next, the specific configuration of the multi-objective expectation prediction module 20 will be described in detail. The multi-objective expectation prediction module 20 further includes: performing energy consumption expectation prediction on the direct-drive motor according to the predetermined control scenario to obtain an energy consumption feature expectation; performing temperature rise feature expectation prediction on the direct-drive motor according to the predetermined control scenario to obtain a temperature rise feature expectation; performing torque fluctuation expectation prediction on the direct-drive motor according to the predetermined control scenario to obtain a torque fluctuation feature expectation; and constructing the motor control expectation vector according to the energy consumption feature expectation, the temperature rise feature expectation, and the torque fluctuation feature expectation.

[0061] Next, the specific configuration of the multi-objective expectation prediction module 20 will be further described in detail. The multi-objective expectation prediction module 20 further includes: interconnecting motors of the same model according to the direct-drive motor to obtain a cluster of motors; retrieving normal energy consumption samples of the cluster of motors according to the predetermined control scenario to obtain a scenario energy consumption sample set; performing confidence evaluation on the scenario energy consumption sample set to obtain an energy consumption confidence evaluation set; and performing data fusion on the scenario energy consumption sample set according to the energy consumption confidence evaluation set to generate the energy consumption feature expectation.

[0062] Next, the specific configuration of the multi-objective deviation evaluation module 30 will be described in detail. The multi-objective deviation evaluation module 30 further includes: performing multi-objective prediction on the direct drive motor according to the motor control scheme to establish a motor control prediction vector; performing multi-objective comparison on the motor control prediction vector according to the motor control expected vector to obtain an energy consumption comparison result, a temperature rise comparison result, and a torque ripple comparison result; respectively performing deviation evaluation on the energy consumption comparison result, the temperature rise comparison result, and the torque ripple comparison result to obtain an energy consumption deviation evaluation coefficient, a temperature rise deviation evaluation coefficient, and a torque ripple deviation evaluation coefficient; performing deviation loss evaluation on the direct drive motor according to the energy consumption comparison result, the temperature rise comparison result, and the torque ripple comparison result to obtain a deviation loss evaluation coefficient; adding the energy consumption deviation evaluation coefficient, the temperature rise deviation evaluation coefficient, the torque ripple deviation evaluation coefficient, and the deviation loss evaluation coefficient to the control expected deviation evaluation result.

[0063] Next, the specific configuration of the control-related feature tracing module 40 will be described in detail. The control-related feature tracing module 40 further includes: if the control expected deviation evaluation result does not satisfy the deviation evaluation multi-sided constraint, determining an abnormal deviation evaluation result and a deviation evaluation abnormal element, performing sensitivity analysis on the motor control variable set of the motor control scheme according to the deviation evaluation abnormal element to generate a control optimization sensitive variable; performing gap identification on the abnormal deviation evaluation result according to the deviation evaluation multi-sided constraint to obtain a deviation evaluation gap feature; performing adjustment constraint configuration on the control optimization sensitive variable according to the deviation evaluation gap feature to generate the first control optimization guiding factor.

[0064] Next, the specific configuration of the control-related feature tracing module 40 will be further described in detail. The control-related feature tracing module 40 further includes: respectively performing random perturbation on the motor control scheme according to the motor control variable set to obtain a set of variable perturbation control schemes; respectively performing simulation control on the direct drive motor according to the set of variable perturbation control schemes to obtain a set of control simulation data; performing response trend analysis on the set of control simulation data according to the deviation evaluation abnormal element to obtain a response trend of each variable perturbation; performing sensitivity evaluation on the motor control variable set according to the response trend of each variable perturbation to obtain a sensitivity of each variable; based on the sensitivity of each variable, screening the motor control variable set according to a predetermined sensitivity to generate the control optimization sensitive variable.

[0065] Next, the specific configuration of the optimization strategy generation module 60 will be described in detail. The optimization strategy generation module 60 further includes: performing a multi-objective deviation evaluation on each motor control strategy in the first motor control strategy set based on the motor control desired vector to obtain the deviation evaluation results of each strategy; determining whether the deviation evaluation results of each strategy satisfy the multi-sided deviation evaluation constraint; if any one of the deviation evaluation results of each strategy satisfies the multi-sided deviation evaluation constraint, obtaining the motor control optimization strategy.

[0066] Next, the specific configuration of the optimization strategy generation module 60 will be described in detail. The optimization strategy generation module 60 further includes: if the deviation evaluation results of each strategy do not satisfy the multi-sided deviation evaluation constraint, identifying the approximation trend of the multi-sided deviation evaluation constraint according to the deviation evaluation results of each strategy to obtain the evaluation approximation constraint trend; identifying the control correlation feature of the first motor control strategy set according to the evaluation approximation constraint trend to obtain the approximation trend control correlation feature; optimizing the first control optimization guiding factor according to the approximation trend control correlation feature to obtain the second control optimization guiding factor; adjusting the first motor control strategy set according to the second control optimization guiding factor to obtain the second motor control strategy set; performing multi-objective joint optimization on the second motor control strategy set according to the motor control desired vector and the multi-sided deviation evaluation constraint to generate the motor control optimization strategy.

[0067] Next, the specific configuration of the multi-objective deviation evaluation module 30 will be described in detail. The multi-objective deviation evaluation module 30 further includes: the multi-sided deviation evaluation constraint includes an energy consumption deviation evaluation constraint, a temperature rise deviation evaluation constraint, a torque ripple deviation evaluation constraint, and a deviation loss evaluation constraint.

[0068] The multi-objective collaborative control device for a direct drive motor provided by an embodiment of the present invention can execute the multi-objective collaborative control method for a direct drive motor provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0069] Although various references are made to certain modules in the device according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or the server. The included various units and modules are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0070] The above specific embodiments do not constitute a limitation on the protection scope of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of this application shall be included within the protection scope of this application.

Claims

1. A multi-objective collaborative control method for a direct drive motor, characterized in that, The method includes: Controlling the terminal of the interactive direct drive motor to obtain a motor control scheme corresponding to a predetermined control scenario; Performing multi-objective expectation prediction on the direct drive motor according to the predetermined control scenario to establish a motor control expectation vector; Based on the motor control scheme, performing multi-objective deviation evaluation on the direct drive motor according to the motor control expectation vector to obtain a control expectation deviation evaluation result; If the control expectation deviation evaluation result does not satisfy the deviation evaluation multi-sided constraint, tracing the control-related features according to the control expectation deviation evaluation result to generate a first guiding factor for control optimization; Performing optimization adjustment on the motor control scheme according to the first guiding factor for control optimization to generate a first set of motor control strategies; Performing multi-objective joint optimization on the first set of motor control strategies according to the motor control expectation vector and the deviation evaluation multi-sided constraint to generate a motor control optimization strategy, and controlling the direct drive motor in combination with the predetermined control scenario, where the multi-objectives refer to energy consumption, temperature rise, and torque ripple; Among them, performing multi-objective expectation prediction on the direct drive motor according to the predetermined control scenario to establish a motor control expectation vector includes: Performing energy consumption expectation prediction on the direct drive motor according to the predetermined control scenario to obtain an energy consumption feature expectation; Performing temperature rise feature expectation prediction on the direct drive motor according to the predetermined control scenario to obtain a temperature rise feature expectation; Performing torque ripple expectation prediction on the direct drive motor according to the predetermined control scenario to obtain a torque ripple feature expectation; Constructing the motor control expectation vector according to the energy consumption feature expectation, the temperature rise feature expectation, and the torque ripple feature expectation; Among them, performing energy consumption expectation prediction on the direct drive motor according to the predetermined control scenario to obtain an energy consumption feature expectation includes: Interconnecting motors of the same model with the direct drive motor to obtain a cluster of motors; Retrieving normal energy consumption samples of the cluster of motors according to the predetermined control scenario to obtain a set of scenario energy consumption samples; Performing confidence evaluation on the set of scenario energy consumption samples to obtain a set of energy consumption confidence evaluation; Performing data fusion on the set of scenario energy consumption samples according to the set of energy consumption confidence evaluation to generate the energy consumption feature expectation; Among them, performing multi-objective joint optimization on the first set of motor control strategies according to the motor control expectation vector and the deviation evaluation multi-sided constraint to generate a motor control optimization strategy includes: Based on the motor control expectation vector, performing multi-objective deviation evaluation on each motor control strategy in the first set of motor control strategies to obtain deviation evaluation results of each strategy; Judging whether the deviation evaluation results of each strategy satisfy the deviation evaluation multi-sided constraint; If any one of the deviation evaluation results of each strategy deviation evaluation result satisfies the deviation evaluation multi-sided constraint, obtaining the motor control optimization strategy.

2. The multi-objective collaborative control method for a direct drive motor according to claim 1, characterized in that Based on the motor control scheme, performing multi-objective deviation evaluation on the direct drive motor according to the motor control expectation vector to obtain a control expectation deviation evaluation result includes: Performing multi-objective prediction on the direct drive motor according to the motor control scheme to establish a motor control prediction vector; Perform multi-objective comparison on the motor control prediction vector according to the motor control desired vector to obtain an energy consumption comparison result, a temperature rise comparison result, and a torque ripple comparison result; Respectively perform deviation evaluation on the energy consumption comparison result, the temperature rise comparison result, and the torque ripple comparison result to obtain an energy consumption deviation evaluation coefficient, a temperature rise deviation evaluation coefficient, and a torque ripple deviation evaluation coefficient; Perform deviation loss evaluation on the direct drive motor according to the energy consumption comparison result, the temperature rise comparison result, and the torque ripple comparison result to obtain a deviation loss evaluation coefficient; Add the energy consumption deviation evaluation coefficient, the temperature rise deviation evaluation coefficient, the torque ripple deviation evaluation coefficient, and the deviation loss evaluation coefficient to the control desired deviation evaluation result.

3. The multi-objective collaborative control method for a direct drive motor according to claim 1, characterized in that, If the control desired deviation evaluation result does not meet the multi-sided deviation evaluation constraint, trace the control-related features according to the control desired deviation evaluation result to generate a first guiding factor for control optimization, including: If the control desired deviation evaluation result does not meet the multi-sided deviation evaluation constraint, determine the abnormal deviation evaluation result and the abnormal deviation evaluation elements; Perform sensitivity analysis on the motor control variable set of the motor control scheme according to the abnormal deviation evaluation elements to generate control optimization sensitive variables; Perform gap identification on the abnormal deviation evaluation result according to the multi-sided deviation evaluation constraint to obtain the deviation evaluation gap feature; Configure adjustment constraints for the control optimization sensitive variables according to the deviation evaluation gap feature to generate the first guiding factor for control optimization.

4. The multi-objective collaborative control method for a direct drive motor according to claim 3, characterized in that, Perform sensitivity analysis on the motor control variable set of the motor control scheme according to the abnormal deviation evaluation elements to generate control optimization sensitive variables, including: Perform random perturbation on the motor control scheme respectively according to the motor control variable set to obtain a set of variable perturbation control schemes; Perform simulation control on the direct drive motor respectively according to the set of variable perturbation control schemes to obtain a set of control simulation data; Perform response trend analysis on the set of control simulation data according to the abnormal deviation evaluation elements to obtain the response trends of variable perturbations; Perform sensitivity evaluation on the motor control variable set according to the response trends of variable perturbations to obtain the sensitivities of variables; Based on the sensitivities of variables, screen the motor control variable set according to a predetermined sensitivity to generate the control optimization sensitive variables.

5. The multi-objective collaborative control method for a direct drive motor according to claim 1, characterized in that, Judge whether the deviation evaluation results of each strategy meet the multi-sided deviation evaluation constraint, including: If the deviation evaluation results of each strategy do not meet the multi-sided deviation evaluation constraint, perform an approximation trend identification on the multi-sided deviation evaluation constraint according to the deviation evaluation results of each strategy to obtain an evaluation approximation constraint trend. The approximation trend identification includes analyzing the change of the gap between the deviation evaluation results of each strategy and the constraint, and the evaluation approximation constraint trend refers to understanding the trend information of each strategy approaching or departing from the constraint on different objectives; Perform control-related feature identification on the first motor control strategy set according to the evaluation approximation constraint trend to obtain the approximation trend control-related feature; Control the associated features according to the approximation trend, optimize the first guiding factor of the control optimization, and obtain the second guiding factor of the control optimization; Adjust the first motor control strategy set according to the second guiding factor of the control optimization to obtain the second motor control strategy set; Perform multi-objective joint optimization on the second motor control strategy set according to the motor control desired vector and the deviation evaluation multi-sided constraint to generate the motor control optimization strategy.

6. The multi-objective collaborative control method for a direct drive motor according to claim 1, characterized in that, The deviation evaluation multi-sided constraint includes an energy consumption deviation evaluation constraint, a temperature rise deviation evaluation constraint, a torque fluctuation deviation evaluation constraint, and a deviation loss evaluation constraint.

7. Multi-objective cooperative control device for direct drive motor, characterized in that, The device is used to implement the multi-objective cooperative control method for a direct drive motor according to any one of claims 1 to 6. The device includes: A motor control scheme acquisition module, configured to interact with the control terminal of the direct drive motor to obtain a motor control scheme corresponding to a predetermined control scenario; A multi-objective expectation prediction module, configured to perform multi-objective expectation prediction on the direct drive motor according to the predetermined control scenario and establish a motor control desired vector; A multi-objective deviation evaluation module, configured to perform multi-objective deviation evaluation on the direct drive motor based on the motor control scheme according to the motor control desired vector to obtain a control desired deviation evaluation result; A control associated feature tracing module, configured to, if the control desired deviation evaluation result does not meet the deviation evaluation multi-sided constraint, trace the control associated features according to the control desired deviation evaluation result to generate a first guiding factor for control optimization; An optimization adjustment module, configured to perform optimization adjustment on the motor control scheme according to the first guiding factor for control optimization to generate a first motor control strategy set; An optimization strategy generation module, configured to perform multi-objective joint optimization on the first motor control strategy set according to the motor control desired vector and the deviation evaluation multi-sided constraint to generate a motor control optimization strategy, and control the direct drive motor in combination with the predetermined control scenario. The multi-objectives refer to energy consumption, temperature rise, and torque fluctuation.

Images