Online self-switching identification and compensation method and device for hyperlocal model scale factor

Abstract

The present invention relates to the field of model-free predictive current control, and specifically to an online self-switching identification and compensation method and device for a hyperlocal model proportional factor. The method and device first establish a hyperlocal model of a permanent magnet synchronous motor; perform mismatch performance analysis on the proportional factor designed in the hyperlocal model to ensure that the deviation between the designed proportional factor and the optimal proportional factor is lower than a preset threshold; then identify the designed proportional factor, input the designed reference value into the identification model and output the proportional factor observation value to ensure that the proportional factor observation value converges to the reference value; finally, perform online compensation and self-switching on the designed proportional factor to avoid current prediction errors caused by parameter fluctuations. The present invention can perform online compensation for the hyperlocal model proportional factor while realizing the identification of the hyperlocal model proportional factor.

Description

Technical Field

[0001] The present invention relates to the field of model-free predictive current control, and in particular to an online self-switching identification and compensation method and device for a super-local model proportional factor. Background Art

[0002] In order to improve the robustness of model-predictive current control, in addition to identifying the motor parameters, model-free predictive current control can predict the current at future moments without the need for motor parameters. In finite set model-free predictive current control, the prediction of the current at future moments depends on the current gradient corresponding to the inverter voltage vector. However, since only one voltage vector can be applied in each control cycle, there is a lag problem in the update of the current gradient, which affects the accuracy of the predicted current.

[0003] In order to increase the update frequency of the current gradient, a current gradient update strategy based on the hyperlocal model can be designed. However, in the hyperlocal model, in addition to using an observer to observe the parameters, a proportional factor needs to be designed to achieve the update of the current gradient. In the model-free current prediction based on the hyperlocal model, the proportional factor can be designed in combination with the nominal inductance parameters of the motor. However, due to changes in the motor operating conditions, the actual parameters of the motor will not match the nominal parameters, and the designed proportional factor may have errors. Therefore, it is necessary to study the online identification and compensation strategy of the hyperlocal model proportional factor to improve the accuracy of current prediction at future moments, thereby achieving high-performance control of the permanent magnet synchronous motor current loop. Summary of the invention

[0004] The embodiment of the present invention provides an online self-switching identification and compensation method and device for a hyperlocal model proportional factor, so as to at least solve the technical problem in the prior art that the compensation parameters cannot be switched when the working state of the motor changes.

[0005] According to an embodiment of the present invention, there is provided an online self-switching identification and compensation method for a hyperlocal model scale factor, comprising the following steps:

[0006] Establish a hyperlocal model of a permanent magnet synchronous motor;

[0007] Perform mismatch performance analysis on the designed scale factor in the hyperlocal model to ensure that the deviation between the designed scale factor and the optimal scale factor is below a preset threshold;

[0008] Identify the designed scale factor, input the designed reference value into the identification model and then output the scale factor observation value to ensure that the scale factor observation value converges to the reference value;

[0009] The designed proportional factor is compensated and switched online.

[0010] Furthermore, establishing a super-local model of the permanent magnet synchronous motor includes:

[0011] The expression of the hyperlocal model of a first-order single-input single-output system is:

[0012]

[0013] Among them, U(t) and Y(t) are the input and output of the system respectively; α is the design scaling factor; F(t) is the sum of the relevant terms of the system, including multiple parts of the system;

[0014] Based on formula (1), the discretized permanent magnet synchronous motor model is expressed as:

[0015]

[0016] in: i d (k), i d (k+1) are the d-axis currents of the (k)th sampling period and the (k+1)th sampling period respectively; i q (k), i q (k+1) are the q-axis currents of the (k)th sampling period and the (k+1)th sampling period respectively; R s , L s and ψ f are the nominal resistance, nominal inductance and nominal flux linkage of the permanent magnet synchronous motor respectively; ω r is the electrical angular velocity of the permanent magnet synchronous motor; T s is the sampling period of the permanent magnet synchronous motor current loop;

[0017] Δi s (k) = i s (k+1)-i s Substituting (k) into formula (2), the current gradient is:

[0018]

[0019] Where: u d (k) and u q (k) is the input voltage of the d and q axes in the kth control cycle.

[0020] Furthermore, the mismatch performance analysis of the designed scale factor in the hyperlocal model is performed to ensure that the deviation between the designed scale factor and the optimal scale factor is lower than a preset threshold, including:

[0021] Assuming that other parameters in the system remain unchanged, the proportional factor α designed according to the nominal parameters s The performance analysis of the mismatch with the actual system scale factor α0 is as follows:

[0022] Taking the q-axis current as the research object, the q-axis reference current Substituting into formula (2) we can get:

[0023] At the same time, the actual current i in the (k+1)th sampling period q (k+1) According to formula (2), we can get:

[0024] Compare formula (4) and formula (5), and replace F in formula (2) q Substituting in, we get:

[0025]

[0026] By performing z-transformation on formula (6), we can obtain the current transfer function H(z):

[0027]

[0028] At this time, the poles of the current prediction discrete system are:

[0029]

[0030] From formula (8), we can see that the stability condition of the discrete system is:

[0031]

[0032] Furthermore, the designed scale factor is identified, and the designed reference value is input into the identification model and then the scale factor observation value is output. Ensuring that the scale factor observation value converges to the reference value includes:

[0033] The sliding mode observer SMO is designed as follows:

[0034]

[0035] in: and are the d-axis current and parameter F d Observed value of is the observed value of the proportional factor; sign() is the sign function; g is the gain coefficient of the observer; η is the gain coefficient of the control law;

[0036] Simplifying formula (10), we can get and The relationship between them is:

[0037]

[0038] Will and are set as the input and output of the identification system respectively, Assuming that the external disturbance of the identification system is , the Laplace transform of formula (11) can obtain the identification model transfer function G(s):

[0039]

[0040] Design the PI controller as:

[0041]

[0042] Among them, E in is the input of the PI controller; k p is the proportional coefficient of the PI controller; k i is the integral coefficient of the PI controller;

[0043] The PI controller is designed around the d-axis current, and formula (2) gives:

[0044]

[0045] Furthermore, the output of the PI controller Input into the identification model shown in formula (12), and the model output When the observed value Converge to reference value When Equal to the actual scale factor α0 of the motor.

[0046] Furthermore, the online compensation and self-switching of the designed scale factor include:

[0047] When the speed tracking error is within ±2%, it is determined that the speed has reached a steady state; condition 1 is:

[0048]

[0049] Among them, E speed is the speed tracking error, is the reference speed;

[0050] Condition 2 is:

[0051] |S spee d|≤S min (16)

[0052] Among them, S speed is the slope of the velocity, S min is the set slope comparison value;

[0053] When speed tracking does not meet conditions 1 and 2, the output value of the PI controller is limited to the value at the last moment before the change;

[0054] The initial value of the proportional factor α based on the nominal parameters of the motor Make settings.

[0055] Furthermore, when the speed tracking satisfies both conditions 1 and 2, the system will automatically switch the proportional factor α to the identification result output by the PI controller. When the motor speed changes, the proportional factor α is automatically switched to the last value before the change until the system speed stabilizes again and meets conditions 1 and 2, and then automatically switches to the identification result output by the PI controller.

[0056] According to another embodiment of the present invention, there is provided an online self-switching identification and compensation device for a hyperlocal model scale factor, comprising:

[0057] A model building unit for building a super-local model of a permanent magnet synchronous motor;

[0058] A mismatch performance analysis unit, used for performing mismatch performance analysis on the designed scale factor in the hyperlocal model to ensure that the deviation between the designed scale factor and the optimal scale factor is lower than a preset threshold;

[0059] The scale factor identification unit is used to identify the designed scale factor, input the designed reference value into the identification model and then output the scale factor observation value to ensure that the scale factor observation value converges to the reference value;

[0060] The online compensation and self-switching unit is used for online compensation and self-switching of the designed proportional factor.

[0061] A processor is used to run a program, wherein when the program is run, any one of the above-mentioned online self-switching identification and compensation methods for the hyperlocal model scale factor is executed.

[0062] A processor is used to run a program, wherein when the program is run, any one of the above-mentioned online self-switching identification and compensation methods for the hyperlocal model scale factor is executed.

[0063] The online self-switching identification and compensation method and device of the hyperlocal model proportional factor in the embodiment of the present invention first establishes a hyperlocal model of a permanent magnet synchronous motor; performs mismatch performance analysis on the proportional factor designed in the hyperlocal model to ensure that the deviation between the designed proportional factor and the optimal proportional factor is lower than a preset threshold; then identifies the designed proportional factor, inputs the designed reference value into the identification model and outputs the proportional factor observation value to ensure that the proportional factor observation value converges to the reference value; finally, performs online compensation and self-switching on the designed proportional factor to avoid current prediction errors caused by parameter fluctuations. The present invention can perform online compensation on the hyperlocal model proportional factor while realizing the identification of the hyperlocal model proportional factor. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0065] Figure 1 It is a flow chart of the online self-switching identification and compensation method of the hyperlocal model scale factor of the present invention;

[0066] FIG. 2( a ) is a diagram of the online self-switching identification and compensation method of the hyperlocal model scale factor of the present invention. Zero-pole distribution diagram of discrete system for current prediction;

[0067] FIG. 2( b ) is a diagram of the online self-switching identification and compensation method of the hyperlocal model scale factor of the present invention. Pulse response diagram of discrete system predicted by current;

[0068] FIG. 3( a ) is a diagram of the online self-switching identification and compensation method of the hyperlocal model scale factor of the present invention. Zero-pole distribution diagram of discrete system for current prediction;

[0069] FIG. 3( b ) is a diagram of the online self-switching identification and compensation method of the hyperlocal model scale factor of the present invention. Pulse response diagram of discrete system predicted by current;

[0070] FIG. 4( a ) is a diagram of the online self-switching identification and compensation method of the hyperlocal model scale factor of the present invention. Zero-pole distribution diagram of discrete system for current prediction;

[0071] FIG. 4( b ) is a diagram of the online self-switching identification and compensation method of the hyperlocal model scale factor of the present invention.

[0072] Pulse response diagram of discrete system predicted by current;

[0073] Figure 5 It is a control block diagram of identification and self-switching strategy in the online self-switching identification and compensation method of the hyperlocal model scale factor of the present invention;

[0074] Figure 6 It is an output diagram of the proportional factor α in the online self-switching identification and compensation method of the hyperlocal model proportional factor of the present invention under different working conditions;

[0075] FIG. 7( a ) is a diagram showing q-axis current tracking under different working conditions without proportional factor compensation in the prior art;

[0076] FIG. 7( b ) is a diagram showing q-axis current tracking under different working conditions with proportional factor compensation in the online self-switching identification and compensation method of the ultra-local model proportional factor of the present invention;

[0077] Figure 8 It is a module diagram of the online self-switching identification and compensation device of the ultra-local model proportional factor of the present invention. DETAILED DESCRIPTION

[0078] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0079] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0080] Example 1

[0081] According to an embodiment of the present invention, a method for online self-switching identification and compensation of a hyperlocal model scale factor is provided. Figure 1 , including the following steps:

[0082] S100: establishing a super-local model of a permanent magnet synchronous motor;

[0083] S200: performing mismatch performance analysis on the scale factor designed in the hyperlocal model to ensure that the deviation between the designed scale factor and the optimal scale factor is lower than a preset threshold;

[0084] S300: Identify the designed scale factor, input the designed reference value into the identification model and then output the scale factor observation value, to ensure that the scale factor observation value converges to the reference value;

[0085] S400: Online compensation and self-switching of designed proportional factors.

[0086] The online self-switching identification and compensation method of the hyperlocal model proportional factor in the embodiment of the present invention first establishes a hyperlocal model of a permanent magnet synchronous motor; performs mismatch performance analysis on the proportional factor designed in the hyperlocal model to ensure that the deviation between the designed proportional factor and the optimal proportional factor is lower than a preset threshold; then identifies the designed proportional factor, inputs the designed reference value into the identification model and outputs the proportional factor observation value to ensure that the proportional factor observation value converges to the reference value; finally, performs online compensation and self-switching on the designed proportional factor to avoid current prediction errors caused by parameter fluctuations. The present invention can perform online compensation on the hyperlocal model proportional factor while realizing the identification of the hyperlocal model proportional factor.

[0087] Among them, the establishment of a super-local model of a permanent magnet synchronous motor includes:

[0088] The expression of the hyperlocal model of a first-order single-input single-output system is:

[0089]

[0090] Among them, U(t) and Y(t) are the input and output of the system respectively; α is the design scaling factor; F(t) is the sum of the relevant terms of the system, including multiple parts of the system;

[0091] Based on formula (1), the discretized permanent magnet synchronous motor model is expressed as:

[0092]

[0093] in: i d (k), i d (k+1) are the d-axis currents of the (k)th sampling period and the (k+1)th sampling period respectively; i q (k), i q (k+1) are the q-axis currents of the (k)th sampling period and the (k+1)th sampling period respectively; R s , L s and ψ f are the nominal resistance, nominal inductance and nominal flux linkage of the permanent magnet synchronous motor respectively; ω r is the electrical angular velocity of the permanent magnet synchronous motor; T s is the sampling period of the permanent magnet synchronous motor current loop;

[0094] Δi s (k) = i s (k+1)-i s Substituting (k) into formula (2), the current gradient is:

[0095]

[0096] Where: u d (k) and u q (k) is the input voltage of the d and q axes in the kth control cycle.

[0097] Among them, the mismatch performance analysis of the designed scale factor in the hyperlocal model is performed to ensure that the deviation between the designed scale factor and the optimal scale factor is lower than the preset threshold, including:

[0098] Assuming that other parameters in the system remain unchanged, the proportional factor α designed according to the nominal parameters s The performance analysis of the mismatch with the actual system scale factor α0 is as follows:

[0099] Taking the q-axis current as the research object, the q-axis reference current Substituting into formula (2) we can get:

[0100] At the same time, the actual current i in the (k+1)th sampling period q (k+1) According to formula (2), we can get:

[0101] Compare formula (4) and formula (5), and replace F in formula (2) q Substituting in, we get:

[0102]

[0103] By performing z-transformation on formula (6), we can obtain the current transfer function H(z):

[0104]

[0105] At this time, the poles of the current prediction discrete system are:

[0106]

[0107] From formula (8), we can see that the stability condition of the discrete system is:

[0108]

[0109] Among them, the scale factor of the design is identified, and the scale factor observation value is output after the design reference value is input into the identification model. Ensuring that the scale factor observation value converges to the reference value includes:

[0110] The sliding mode observer SMO is designed as follows:

[0111]

[0112] in: and are the d-axis current and parameter Fd Observed value of is the observed value of the proportional factor; sign() is the sign function; g is the gain coefficient of the observer; η is the gain coefficient of the control law.

[0113] Simplifying formula (10), we can get and The relationship between them is:

[0114]

[0115] Will and are set as the input and output of the identification system respectively, Assuming that the external disturbance of the identification system is , the Laplace transform of formula (11) can obtain the identification model transfer function G(s):

[0116]

[0117] Design the PI controller as:

[0118]

[0119] Among them, E in is the input of the PI controller; k p is the proportional coefficient of the PI controller; k i is the integral coefficient of the PI controller.

[0120] The PI controller is designed around the d-axis current, and formula (2) gives:

[0121]

[0122] The output of the PI controller is Input into the identification model shown in formula (12), and the model output When the observed value Converge to reference value When Equal to the actual scale factor α0 of the motor.

[0123] Among them, the online compensation and self-switching of the designed proportional factor include:

[0124] When the speed tracking error is within ±2%, it is determined that the speed has reached a steady state; condition 1 is:

[0125]

[0126] Among them, E speed is the speed tracking error, is the reference speed;

[0127] Condition 2 is:

[0128] S speed |≤S min (16)

[0129] Among them, S speed is the slope of the velocity, S min is the set slope comparison value;

[0130] When speed tracking does not meet conditions 1 and 2, the output value of the PI controller is limited to the value at the last moment before the change;

[0131] The initial value of the proportional factor α based on the nominal parameters of the motor Make settings.

[0132] When the speed tracking satisfies both condition 1 and condition 2, the system will automatically switch the proportional factor α to the identification result output by the PI controller. When the motor speed changes, the proportional factor α is automatically switched to the last value before the change until the system speed stabilizes again and meets conditions 1 and 2, and then automatically switches to the identification result output by the PI controller.

[0133] The following is a detailed description of the online self-switching identification and compensation method of the hyperlocal model scale factor of the present invention using a specific embodiment:

[0134] In view of the hysteresis problem of current gradient update in the model-free predictive current control strategy, a current gradient update strategy based on a hyperlocal model can be designed. However, since the parameter mismatch of the hyperlocal model will affect the current prediction accuracy, the present invention proposes an online self-switching identification and compensation method for the hyperlocal model proportional factor to improve the accuracy of the hyperlocal model proportional factor, and also analyzes the impact of the proportional factor mismatch on the current tracking performance. The method is based on finite control set model-free predictive current control and involves the field of finite control set predictive control of permanent magnet synchronous motors. The present invention can perform online compensation for the hyperlocal model proportional factor while realizing the identification of the hyperlocal model proportional factor. At the same time, the proposed self-switching strategy can ensure that the compensation parameters are switched when the working state of the motor changes, avoiding the current prediction error caused by parameter fluctuations.

[0135] The present invention specifically includes:

[0136] 1. Establish a super-local model of permanent magnet synchronous motor.

[0137] The expression of the hyperlocal model of a first-order single-input single-output system is:

[0138]

[0139] Among them, U(t) and Y(t) are the input and output of the system respectively; α is a designable proportional factor; F(t) is the sum of the related terms of the system, which includes multiple parts of the system.

[0140] Based on formula (1), the discretized permanent magnet synchronous motor model can be expressed as:

[0141]

[0142] in: i d (k), i d (k+1) are the d-axis currents of the (k)th sampling period and the (k+1)th sampling period respectively; i q (k), i q (k+1) are the q-axis currents of the (k)th sampling period and the (k+1)th sampling period respectively; R s , L s and ψ f are the nominal resistance, nominal inductance and nominal flux linkage of the permanent magnet synchronous motor respectively; ω r is the electrical angular velocity of the permanent magnet synchronous motor; T s is the sampling period of the permanent magnet synchronous motor current loop.

[0143] Δi s (k) = i s (k+1)-i s Substituting (k) into formula (2), the current gradient is:

[0144]

[0145] Where: u d (k) and u q (k) is the input voltage of the d and q axes in the kth control cycle.

[0146] In formula (3), F d (k) and F q (k) can be observed by an observer. However, when designing the observer, the proportional factor α will directly affect the accuracy of the observation result, and thus affect the accuracy of the current gradient corresponding to the inverter voltage vector. When the motor parameters are matched, the proportional factor α can be designed by the nominal inductance parameters of the motor. However, due to changes in the motor operating conditions and other reasons, the motor parameters are mismatched. At this time, the proportional factor α designed according to the nominal parameters s There will be a deviation from the actual optimal α of the motor. Therefore, the proportional factor α needs to be identified and compensated online to ensure the accuracy of the current gradient prediction.

[0147] 2. Analysis of scale factor α mismatch performance.

[0148] Assuming that other parameters in the system remain unchanged, the proportional factor α designed according to the nominal parameters s The performance analysis of the mismatch with the actual system scale factor α0 is as follows:

[0149] Taking the q-axis current as the research object, the q-axis reference current Substituting into formula (2) we can get:

[0150] At the same time, the actual current i in the (k+1)th sampling period q (k+1) According to formula (2), we can get:

[0151] Compare formula (4) and formula (5), and replace F in formula (2) q Substituting in, we get:

[0152]

[0153] By performing z-transformation on formula (6), we can obtain the current transfer function H(z):

[0154]

[0155] At this time, the poles of the current prediction discrete system are:

[0156]

[0157] The pole distribution of the discrete system determines the stability and current step response characteristics of the discrete system. According to the stability judgment condition of the discrete system in the z domain, when all the poles of the discrete system are distributed in the unit circle with the origin of the z domain as the center, the discrete system can remain stable. Therefore, from formula (8), it can be seen that the stability condition of the discrete system is:

[0158]

[0159] when When , the pole of the discrete system is located on the right half plane of the real axis of the unit circle. At this time, the discrete system is in a stable state. When the system input is a step reference signal, the output of the system will decay monotonically, and the closer the pole position is to the origin, the faster the response speed. When , the pole of the discrete system is located at the origin of the unit circle. At this time, the discrete system is in a stable state and can respond to the step reference current signal without a beat. When , the pole of the discrete system is located on the left half plane of the real axis of the unit circle. At this time, the discrete system is in a stable state. When the system input is a step reference signal, the system output will oscillate and decay. When , the poles of the discrete system are outside the unit circle, and the discrete system is unstable. At this time, it will not be able to effectively respond to the step input.

[0160] The zero-pole distribution and pulse response of the current prediction discrete system under different proportional factors are shown in Figures 2 to 4. When the designed proportional factor differs greatly from the actual parameters of the motor, the current output by the system will not be able to accurately track the input, thus affecting the control performance of the current loop. Therefore, there may be a deviation between the proportional factor designed based on the nominal parameters of the motor and the optimal proportional factor of the system. Designing an identification strategy to identify and compensate the proportional factor online is crucial to improving the accuracy of finite set model-free predictive current control.

[0161] 3. Identification strategy of the proportional factor α.

[0162] In order to identify the proportional factor α, the sliding mode observer (SMO) is designed as follows:

[0163]

[0164] in: and are the d-axis current and parameter F d Observed value of is the observed value of the proportional factor; sign() is the sign function; g is the gain coefficient of the observer; η is the gain coefficient of the control law.

[0165] Simplifying formula (10), we can get and The relationship between them is:

[0166]

[0167] Will and are set as the input and output of the identification system respectively, Assuming that the external disturbance of the identification system is , the Laplace transform of formula (11) can obtain the identification model transfer function G(s):

[0168]

[0169] It can be seen from formula (12) that the model of the identification system is a first-order inertia link. Therefore, in order to realize the identification of the proportional factor α, a PI controller is designed based on formula (12). The identification of the proportional factor α is realized through the convergence of the PI controller. The designed PI controller is:

[0170]

[0171] Among them, E in is the input of the PI controller; k p is the proportional coefficient of the PI controller; k i is the integral coefficient of the PI controller.

[0172] Since the d-axis current formula does not include the motor flux term and the d-axis current is set to 0, the PI controller is designed around the d-axis current. From formula (2), we can get:

[0173]

[0174] Since the d-axis current is set to 0, The term can be approximated to 0, so the output of the PI controller Input into the identification model shown in formula (12), and the model output When the observed value Converge to reference value That is ω r i q When It can be approximately equal to the real scale factor α0 of the motor, thereby realizing the identification of the scale factor of the hyperlocal model.

[0175] 4. Scale Factor Online compensation and self-switching strategy.

[0176] Through formula (13), the proportional factor α of the hyperlocal model can be identified by the PI controller. However, due to the reference value Related to the speed of the motor, so when the motor starts, the speed is 0 and the reference value is After calculation, the output of the PI controller is 0. is also 0, when the output of the PI controller is directly When used for current gradient prediction, the poles of the current prediction discrete system are outside the unit circle, and the output current cannot correctly respond to the reference current. At this time, the system cannot converge. In addition, since the PI controller takes time to converge, when the motor speed changes, the reference value As the motor speed changes, the output of the PI controller will change with the input. Only when the speed is stable can the PI controller output a reliable identification value. Therefore, the output value of the PI controller cannot be used directly for current prediction, and the output of the PI controller needs to be self-switched to ensure the accuracy of the proportional factor α of the hyperlocal model.

[0177] Combined with the characteristics of the PI controller, it is first necessary to determine whether the speed has reached a steady state. Based on the definition of the adjustment time, that is, the shortest time required for the speed to reach and remain within ±2% of the reference speed, it is determined whether the speed has reached a steady state. When the speed tracking error is within ±2%, it is determined that the speed has reached a steady state. Condition 1 is:

[0178]

[0179] Among them, E speed is the speed tracking error, is the reference speed.

[0180] Secondly, when there is overshoot in speed tracking, and the overshoot is greater than ±2% of the reference speed, the speed tracking error will be within ±2%, but the speed has not reached a steady state at this time. If the output value of the PI controller is applied, it will cause oscillation due to self-switching. Therefore, it is also necessary to determine it through the slope of the motor output speed. When the motor speed rises, its slope is greater than the speed slope in the steady state. By adding the conditions for determining the speed slope, the proportional factor can be avoided from switching when the speed overshoots. Condition 2 is:

[0181] |S speed |≤S min (16)

[0182] Among them, S speed is the slope of the velocity, S min is the set slope comparison value.

[0183] Again, when the reference speed of the motor changes, the reference value of the identification model is related to speed, so It will change with the change of the motor reference speed. At this time, the stable state established by the PI controller before the speed change will be broken, and the adjustment state will continue. During the adjustment process, the proportional factor α of the PI controller's output will fluctuate accordingly, thereby affecting the accuracy of the predicted current. Therefore, when the speed tracking does not meet conditions 1 and 2, it is necessary to limit the output value of the PI controller to the value at the last moment before the change, so as to avoid the fluctuation of the proportional factor α.

[0184] Finally, since the output value of the PI controller gradually reaches a steady state from 0, in order to achieve online compensation, the initial value of the proportional factor α is also required. The initial value is set to ensure that the motor can run normally and reach a steady state, so as to ensure that the PI controller can converge normally. In order to ensure the effectiveness of the initial value of the proportional factor α, it can be set according to the nominal parameters of the motor.

[0185] In summary, in order to ensure the online identification and compensation of the proportional factor, the self-switching strategy can be summarized as follows: the proportional factor of the hyperlocal model will predict the current based on the initial value, thereby ensuring the normal operation of the motor. When the speed tracking meets both conditions 1 and 2, the system will automatically switch the proportional factor to the identification result output by the PI controller. When the motor speed changes, the proportional factor α is automatically switched to the last value before the change until the system speed stabilizes again and meets conditions 1 and 2, and then automatically switches to the identification result output by the PI controller. The self-switching strategy can ensure the online compensation of the identification strategy. At the same time, when the motor operating state changes, the self-switching strategy is used to avoid the fluctuation of the proportional factor α following the motor state, thereby improving the performance of model-free predictive current control. The control block diagram of the identification strategy and the self-switching strategy is shown in Figure 2. Figure 5 shown.

[0186] In order to verify the effectiveness of the online self-switching identification algorithm proposed in the present invention, different operating conditions of the motor are simulated by changing the motor speed and load step. Figure 6 As shown in the figure, the motor speed jumps from 50r / min to 100r / min at 2s, and the external load jumps from 1N·m to 2N·m at 4s. The dotted line in the figure is the nominal value of the proportional factor α determined according to the nominal parameters of the motor, the dotted line is the identification value of the proportional factor output by the identification strategy, and the solid line is the output value of the proportional factor α after the self-switching strategy. Figure 6 It can be seen that when the motor working condition changes, the speed fluctuates, causing the identification value of α to deviate from the steady-state value following the speed fluctuation. Therefore, the identification value output by the identification strategy cannot be directly loaded into the system for current prediction. In the self-switching strategy proposed in the present invention, when the motor starts, the identification value of α gradually enters the steady state from 0. When directly loaded, the system will not be able to operate normally at this time due to system instability. The preset value in the self-switching strategy can ensure that the motor starts smoothly and the speed reaches a steady state. When the switching conditions are met, the system will automatically switch to the identification value; at the same time, when the motor working condition changes, the self-switching strategy can ensure that when the speed fluctuates, the system can input a stable proportional factor, thereby reducing the impact of the fluctuation of α on the current prediction.

[0187] In order to verify the effectiveness of the online self-switching identification and compensation algorithm proposed in the present invention, the tracking of the q-axis current under different working conditions is compared as shown in Figure 7. In order to simulate the mismatch of parameters during motor operation, the simulation is performed by directly modifying the parameters of the motor model. The proportional factor in Figure 7 (a) is selected based on the nominal parameters of the motor. However, since the model parameters of the motor have been modified and mismatched with the nominal parameters, a large error is generated in the current prediction at future moments, which increases the ripple current and affects the tracking performance of the q-axis reference current. The selection of the proportional factor in Figure 7 (b) is based on the online self-switching identification and compensation algorithm proposed in the present invention. By identifying and compensating the proportional factor, the dependence on the nominal parameters is avoided, thereby improving the current tracking performance. Compared with Figure 7 (a), the current ripple and tracking error of the q-axis current in Figure 7 (b) are significantly smaller, and have better current tracking performance. However, it should be noted that since the convergence of the identification strategy takes time, since the set initial value cannot be truly determined, large initial fluctuations may occur. However, once the identification strategy converges, the designed self-switching strategy can switch the proportional factor to ensure the performance of the motor during operation. In addition, the initial value can be pre-tested through repeated experiments to improve the transient performance of the motor when it starts while ensuring the steady-state performance of the motor.

[0188] The present invention uses the detection signal-to-noise ratio as an evaluation criterion, establishes an iterative relationship between direct effects, indirect effects and weak correlation effects between global parameters, and constructs a multi-coupling detection model method between global parameters in a full detection link containing multiple physical quantities. In view of this, the present invention proposes an online self-switching identification and compensation method for a hyperlocal model scale factor based on a multi-coupling relationship, including detection signal-to-noise ratio, instantaneous field of view, number of samplings, information redundancy, spectral resolution, system / detector noise, spectral range, data refresh rate, imaging frame rate, platform vibration, attitude change, platform speed and other aviation parameters in the detection of an aviation coded aperture spectroscopic imager. By distinguishing the effects of the above physical quantities on the detection signal-to-noise ratio, the indirect effects between the physical quantities and the weak correlation under specific conditions, a multi-parameter multi-coupling detection model that can be widely accepted is constructed.

[0189] Example 2

[0190] According to another embodiment of the present invention, there is provided an online self-switching identification and compensation device for a hyperlocal model scale factor, see Figure 8 ,include:

[0191] A model building unit 201 is used to build a super-local model of the permanent magnet synchronous motor;

[0192] A mismatch performance analysis unit 202, used to perform mismatch performance analysis on the designed scale factor in the hyperlocal model to ensure that the deviation between the designed scale factor and the optimal scale factor is lower than a preset threshold;

[0193] The scale factor identification unit 203 is used to identify the designed scale factor, input the designed reference value into the identification model and then output the scale factor observation value, so as to ensure that the scale factor observation value converges to the reference value;

[0194] The online compensation and self-switching unit 204 is used to perform online compensation and self-switching on the designed proportional factor.

[0195] The online self-switching identification and compensation device of the hyperlocal model proportional factor in the embodiment of the present invention first establishes a hyperlocal model of a permanent magnet synchronous motor; performs mismatch performance analysis on the proportional factor designed in the hyperlocal model to ensure that the deviation between the designed proportional factor and the optimal proportional factor is lower than a preset threshold; then identifies the designed proportional factor, inputs the designed reference value into the identification model and outputs the proportional factor observation value to ensure that the proportional factor observation value converges to the reference value; finally, performs online compensation and self-switching on the designed proportional factor to avoid current prediction errors caused by parameter fluctuations. The present invention can perform online compensation on the hyperlocal model proportional factor while realizing the identification of the hyperlocal model proportional factor.

[0196] The following is a detailed description of the online self-switching identification and compensation device for the hyperlocal model scale factor of the present invention using a specific embodiment:

[0197] In view of the hysteresis problem of current gradient update in the model-free predictive current control strategy, a current gradient update strategy based on a hyperlocal model can be designed. However, since the parameter mismatch of the hyperlocal model will affect the current prediction accuracy, the present invention proposes an online self-switching identification and compensation device for the hyperlocal model proportional factor to improve the accuracy of the hyperlocal model proportional factor, and also analyzes the influence of the proportional factor mismatch on the current tracking performance. The device is based on finite control set model-free predictive current control, and involves the field of finite control set predictive control of permanent magnet synchronous motors. The present invention can perform online compensation for the hyperlocal model proportional factor while realizing the identification of the hyperlocal model proportional factor. At the same time, the proposed self-switching strategy can ensure that the compensation parameters are switched when the working state of the motor changes, avoiding the current prediction error caused by parameter fluctuations.

[0198] The present invention specifically includes:

[0199] 1. Establish a super-local model of permanent magnet synchronous motor.

[0200] The expression of the hyperlocal model of a first-order single-input single-output system is:

[0201]

[0202] Among them, U(t) and Y(t) are the input and output of the system respectively; α is a designable proportional factor; F(t) is the sum of the related terms of the system, which includes multiple parts of the system.

[0203] Based on formula (1), the discretized permanent magnet synchronous motor model can be expressed as:

[0204]

[0205] in: i d (k), i d (k+1) are the d-axis currents of the (k)th sampling period and the (k+1)th sampling period respectively; i q (k), i q (k+1) are the q-axis currents of the (k)th sampling period and the (k+1)th sampling period respectively; R s , L s and ψ f are the nominal resistance, nominal inductance and nominal flux linkage of the permanent magnet synchronous motor respectively; ω r is the electrical angular velocity of the permanent magnet synchronous motor; T s is the sampling period of the permanent magnet synchronous motor current loop.

[0206] Δi s (k) = i s (k+1)-i s Substituting (k) into formula (2), the current gradient is:

[0207]

[0208] Where: u d (k) and u q (k) is the input voltage of the d and q axes in the kth control cycle.

[0209] In formula (3), F d (k) and F q (k) can be observed by an observer. However, when designing the observer, the proportional factor α will directly affect the accuracy of the observation result, and thus affect the accuracy of the current gradient corresponding to the inverter voltage vector. When the motor parameters are matched, the proportional factor α can be designed by the nominal inductance parameters of the motor. However, due to changes in the motor operating conditions and other reasons, the motor parameters are mismatched. At this time, the proportional factor α designed according to the nominal parameters s There will be a deviation from the actual optimal α of the motor. Therefore, the proportional factor α needs to be identified and compensated online to ensure the accuracy of the current gradient prediction.

[0210] 2. Analysis of scale factor α mismatch performance.

[0211] Assuming that other parameters in the system remain unchanged, the proportional factor α designed according to the nominal parameters s The performance analysis of the mismatch with the actual system scale factor α0 is as follows:

[0212] Taking the q-axis current as the research object, the q-axis reference current Substituting into formula (2) we can get:

[0213] At the same time, the actual current i in the (k+1)th sampling period q (k+1) According to formula (2), we can get:

[0214] Compare formula (4) and formula (5), and replace F in formula (2) q Substituting in, we get:

[0215]

[0216] By performing z-transformation on formula (6), we can obtain the current transfer function H(z):

[0217]

[0218] At this time, the poles of the current prediction discrete system are:

[0219]

[0220] The pole distribution of the discrete system determines the stability and current step response characteristics of the discrete system. According to the stability judgment condition of the discrete system in the z domain, when all the poles of the discrete system are distributed in the unit circle with the origin of the z domain as the center, the discrete system can remain stable. Therefore, from formula (8), it can be seen that the stability condition of the discrete system is:

[0221]

[0222] when When , the pole of the discrete system is located on the right half plane of the real axis of the unit circle. At this time, the discrete system is in a stable state. When the system input is a step reference signal, the output of the system will decay monotonically, and the closer the pole position is to the origin, the faster the response speed. When , the pole of the discrete system is located at the origin of the unit circle. At this time, the discrete system is in a stable state and can respond to the step reference current signal without a beat. When , the pole of the discrete system is located on the left half plane of the real axis of the unit circle. At this time, the discrete system is in a stable state. When the system input is a step reference signal, the system output will oscillate and decay. When , the poles of the discrete system are outside the unit circle, and the discrete system is unstable. At this time, it will not be able to effectively respond to the step input.

[0223] The zero-pole distribution and pulse response of the current prediction discrete system under different proportional factors are shown in Figures 2 to 4. When the designed proportional factor differs greatly from the actual parameters of the motor, the current output by the system will not be able to accurately track the input, thus affecting the control performance of the current loop. Therefore, there may be a deviation between the proportional factor designed based on the nominal parameters of the motor and the optimal proportional factor of the system. Designing an identification strategy to identify and compensate the proportional factor online is crucial to improving the accuracy of finite set model-free predictive current control.

[0224] 3. Identification strategy of proportional factor α.

[0225] In order to identify the proportional factor α, the sliding mode observer (SMO) is designed as follows:

[0226]

[0227] in: and are the d-axis current and parameter F d Observed value of is the observed value of the proportional factor; sign() is the sign function; g is the gain coefficient of the observer; η is the gain coefficient of the control law.

[0228] Simplifying formula (10), we can get and The relationship between them is:

[0229]

[0230] Will and are set as the input and output of the identification system respectively, Assuming that the external disturbance of the identification system is , the Laplace transform of formula (11) can obtain the identification model transfer function G(s):

[0231]

[0232] It can be seen from formula (12) that the model of the identification system is a first-order inertia link. Therefore, in order to realize the identification of the proportional factor α, a PI controller is designed based on formula (12). The identification of the proportional factor α is realized through the convergence of the PI controller. The designed PI controller is:

[0233]

[0234] Among them, E in is the input of the PI controller; k p is the proportional coefficient of the PI controller; k i is the integral coefficient of the PI controller.

[0235] Since the d-axis current formula does not include the motor flux term and the d-axis current is set to 0, the PI controller is designed around the d-axis current. From formula (2), we can get:

[0236]

[0237] Since the d-axis current is set to 0, The term can be approximated to 0, so the output of the PI controller Input into the identification model shown in formula (12), and the model output When the observed value Converge to reference value That is ω r i q When It can be approximately equal to the real scale factor α0 of the motor, thereby realizing the identification of the scale factor of the hyperlocal model.

[0238] 4. Scale Factor Online compensation and self-switching strategy.

[0239] Through formula (13), the proportional factor α of the hyperlocal model can be identified by the PI controller. However, due to the reference value Related to the speed of the motor, so when the motor starts, the speed is 0 and the reference value is After calculation, the output of the PI controller is 0. is also 0, when the output of the PI controller is directly When used for current gradient prediction, the poles of the current prediction discrete system are outside the unit circle, and the output current cannot correctly respond to the reference current. At this time, the system cannot converge. In addition, since the PI controller takes time to converge, when the motor speed changes, the reference value As the motor speed changes, the output of the PI controller will change with the input. Only when the speed is stable can the PI controller output a reliable identification value. Therefore, the output value of the PI controller cannot be used directly for current prediction, and the output of the PI controller needs to be self-switched to ensure the accuracy of the proportional factor α of the hyperlocal model.

[0240] Combined with the characteristics of the PI controller, it is first necessary to determine whether the speed has reached a steady state. Based on the definition of the adjustment time, that is, the shortest time required for the speed to reach and remain within ±2% of the reference speed, it is determined whether the speed has reached a steady state. When the speed tracking error is within ±2%, it is determined that the speed has reached a steady state. Condition 1 is:

[0241]

[0242] Among them, E speed is the speed tracking error, is the reference speed.

[0243] Secondly, when there is overshoot in speed tracking, and the overshoot is greater than ±2% of the reference speed, the speed tracking error will be within ±2%, but the speed has not reached a steady state at this time. If the output value of the PI controller is applied, it will cause oscillation due to self-switching. Therefore, it is also necessary to determine it through the slope of the motor output speed. When the motor speed rises, its slope is greater than the speed slope in the steady state. By adding the conditions for determining the speed slope, the proportional factor can be avoided from switching when the speed overshoots. Condition 2 is:

[0244] |S speed |≤S min (16)

[0245] Among them, S speed is the slope of the velocity, S min is the set slope comparison value.

[0246] Again, when the reference speed of the motor changes, the reference value of the identification model is related to speed, so It will change with the change of the motor reference speed. At this time, the stable state established by the PI controller before the speed change will be broken, and the adjustment state will continue. During the adjustment process, the proportional factor α of the PI controller's output will fluctuate accordingly, thereby affecting the accuracy of the predicted current. Therefore, when the speed tracking does not meet conditions 1 and 2, it is necessary to limit the output value of the PI controller to the value at the last moment before the change, so as to avoid the fluctuation of the proportional factor α.

[0247] Finally, since the output value of the PI controller gradually reaches a steady state from 0, in order to achieve online compensation, the initial value of the proportional factor α is also required. The initial value is set to ensure that the motor can run normally and reach a steady state, so as to ensure that the PI controller can converge normally. In order to ensure the effectiveness of the initial value of the proportional factor α, it can be set according to the nominal parameters of the motor.

[0248] In summary, in order to ensure the online identification and compensation of the proportional factor, the self-switching strategy can be summarized as follows: the proportional factor of the hyperlocal model will predict the current based on the initial value, thereby ensuring the normal operation of the motor. When the speed tracking meets both conditions 1 and 2, the system will automatically switch the proportional factor to the identification result output by the PI controller. When the motor speed changes, the proportional factor α is automatically switched to the last value before the change until the system speed stabilizes again and meets conditions 1 and 2, and then automatically switches to the identification result output by the PI controller. The self-switching strategy can ensure the online compensation of the identification strategy. At the same time, when the motor operating state changes, the self-switching strategy is used to avoid the fluctuation of the proportional factor α following the motor state, thereby improving the performance of model-free predictive current control. The control block diagram of the identification strategy and the self-switching strategy is shown in Figure 2. Figure 5 shown.

[0249] In order to verify the effectiveness of the online self-switching identification algorithm proposed in the present invention, different operating conditions of the motor are simulated by changing the motor speed and load step. Figure 6 As shown in the figure, the motor speed steps from 50r / min to 100r / min at 2s, and the external load steps from 1Nm to 2Nm at 4s. The dotted line in the figure is the nominal value of the proportional factor α determined according to the nominal parameters of the motor, the dotted line is the identification value of the proportional factor output by the identification strategy, and the solid line is the output value of the proportional factor α after the self-switching strategy. Figure 6 It can be seen that when the motor working condition changes, the speed fluctuates, causing the identification value of α to deviate from the steady-state value following the speed fluctuation. Therefore, the identification value output by the identification strategy cannot be directly loaded into the system for current prediction. In the self-switching strategy proposed in the present invention, when the motor starts, the identification value of α gradually enters the steady state from 0. When directly loaded, the system will not be able to operate normally at this time due to system instability. The preset value in the self-switching strategy can ensure that the motor starts smoothly and the speed reaches a steady state. When the switching conditions are met, the system will automatically switch to the identification value; at the same time, when the motor working condition changes, the self-switching strategy can ensure that when the speed fluctuates, the system can input a stable proportional factor, thereby reducing the impact of the fluctuation of α on the current prediction.

[0250] In order to verify the effectiveness of the online self-switching identification and compensation algorithm proposed in the present invention, the tracking of the q-axis current under different working conditions is compared as shown in Figure 7. In order to simulate the mismatch of parameters during motor operation, the simulation is performed by directly modifying the parameters of the motor model. The proportional factor in Figure 7 (a) is selected based on the nominal parameters of the motor. However, since the model parameters of the motor have been modified and mismatched with the nominal parameters, a large error is generated in the current prediction at future moments, which increases the ripple current and affects the tracking performance of the q-axis reference current. The selection of the proportional factor in Figure 7 (b) is based on the online self-switching identification and compensation algorithm proposed in the present invention. By identifying and compensating the proportional factor, the dependence on the nominal parameters is avoided, thereby improving the current tracking performance. Compared with Figure 7 (a), the current ripple and tracking error of the q-axis current in Figure 7 (b) are significantly smaller, and have better current tracking performance. However, it should be noted that since the convergence of the identification strategy takes time, since the set initial value cannot be truly determined, large initial fluctuations may occur. However, once the identification strategy converges, the designed self-switching strategy can switch the proportional factor to ensure the performance of the motor during operation. In addition, the initial value can be pre-tested through repeated experiments to improve the transient performance of the motor when it starts while ensuring the steady-state performance of the motor.

[0251] The present invention uses the detection signal-to-noise ratio as an evaluation criterion, establishes an iterative relationship between direct effects, indirect effects and weak correlation effects between global parameters, and constructs a multi-coupling detection model method between global parameters in a full detection link containing multiple physical quantities. In view of this, the present invention proposes an online self-switching identification and compensation method for a hyperlocal model scale factor based on a multi-coupling relationship, including detection signal-to-noise ratio, instantaneous field of view, number of samplings, information redundancy, spectral resolution, system / detector noise, spectral range, data refresh rate, imaging frame rate, platform vibration, attitude change, platform speed and other aviation parameters in the detection of an aviation coded aperture spectroscopic imager. By distinguishing the effects of the above physical quantities on the detection signal-to-noise ratio, the indirect effects between the physical quantities and the weak correlation under specific conditions, a multi-parameter multi-coupling detection model that can be widely accepted is constructed.

[0252] Example 3

[0253] A storage medium stores a program file capable of realizing any one of the above-mentioned online self-switching identification and compensation methods for the hyperlocal model scale factor.

[0254] Example 4

[0255] A processor is used to run a program, wherein when the program is run, any one of the above-mentioned online self-switching identification and compensation methods for the hyperlocal model scale factor is executed.

[0256] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0257] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0258] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the system embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0259] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0260] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0261] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0262] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An online self-switching identification and compensation method for a hyperlocal model scale factor, characterized in that: The following steps are involved: Establish a hyperlocal model of a permanent magnet synchronous motor; Perform mismatch performance analysis on the designed scale factor in the hyperlocal model to ensure that the deviation between the designed scale factor and the optimal scale factor is below a preset threshold; Identify the designed scale factor, input the designed reference value into the identification model and then output the scale factor observation value to ensure that the scale factor observation value converges to the reference value; The designed proportional factor is compensated and switched online.

2. The online self-switching identification and compensation method of the hyperlocal model scale factor according to claim 1 is characterized in that: The establishing of the super local model of the permanent magnet synchronous motor comprises: The expression of the hyperlocal model of a first-order single-input single-output system is: Among them, U(t) and Y(t) are the input and output of the system respectively; α is the design scaling factor; F(t) is the sum of the relevant terms of the system, including multiple parts of the system; Based on formula (1), the discretized permanent magnet synchronous motor model is expressed as: in: i d (k), i d (k+1) are the d-axis currents of the (k)th sampling period and the (k+1)th sampling period respectively; i q (k), i q (k+1) are the q-axis currents of the (k)th sampling period and the (k+1)th sampling period respectively; R s , L s and ψ f are the nominal resistance, nominal inductance and nominal flux linkage of the permanent magnet synchronous motor respectively; ω r is the electrical angular velocity of the permanent magnet synchronous motor; T s is the sampling period of the permanent magnet synchronous motor current loop; Δi s (k) = i s (k+1)-i s Substituting (k) into formula (2), the current gradient is: Where: u d (k) and u q (k) is the input voltage of the d and q axes in the kth control cycle.

3. The online self-switching identification and compensation method of the hyperlocal model scale factor according to claim 2 is characterized in that: The mismatch performance analysis of the designed scale factor in the hyperlocal model to ensure that the deviation between the designed scale factor and the optimal scale factor is lower than a preset threshold comprises: Assuming that other parameters in the system remain unchanged, the proportional factor α designed according to the nominal parameters s The performance analysis of the mismatch with the actual system scale factor α0 is as follows: Taking the q-axis current as the research object, the q-axis reference current Substituting into formula (2) we can get: At the same time, the actual current i in the (k+1)th sampling period q (k+1) According to formula (2), we can get: Compare formula (4) and formula (5), and replace F in formula (2) q Substituting in, we get: By performing z-transformation on formula (6), we can obtain the current transfer function H(z): At this time, the poles of the current prediction discrete system are: From formula (8), we can see that the stability condition of the discrete system is:

4. The online self-switching identification and compensation method of the hyperlocal model scale factor according to claim 3 is characterized in that: The identification of the designed scale factor, inputting the designed reference value into the identification model and then outputting the scale factor observation value, and ensuring that the scale factor observation value converges to the reference value includes: The sliding mode observer SMO is designed as follows: in: and are the d-axis current and parameter F d Observed value of is the observed value of the proportional factor; sign(·) is the sign function; g is the gain coefficient of the observer; η is the gain coefficient of the control law; Simplifying formula (10), we can get and The relationship between them is: Will and are set as the input and output of the identification system respectively, Assuming that the external disturbance of the identification system is , the Laplace transform of formula (11) can obtain the identification model transfer function G(s): Design the PI controller as: Among them, E in is the input of the PI controller; k p is the proportional coefficient of the PI controller; k i is the integral coefficient of the PI controller; The PI controller is designed around the d-axis current, and formula (2) gives:

5. The online self-switching identification and compensation method of the hyperlocal model scale factor according to claim 4 is characterized in that: The output of the PI controller Input into the identification model shown in formula (12), and the model output When the observed value Converge to reference value When Equal to the actual scale factor α0 of the motor.

6. The online self-switching identification and compensation method of the hyperlocal model scale factor according to claim 1, characterized in that: The online compensation and self-switching of the designed proportional factor include: When the speed tracking error is within ±2%, it is determined that the speed has reached a steady state; condition 1 is: Among them, E speed is the speed tracking error, is the reference speed; Condition 2 is: |S spee d|≤S min (16) Among them, S speed is the slope of the velocity, S min is the set slope comparison value; When speed tracking does not meet conditions 1 and 2, the output value of the PI controller is limited to the value at the last moment before the change; The initial value of the proportional factor α based on the nominal parameters of the motor Make settings.

7. The online self-switching identification and compensation method of the hyperlocal model scale factor according to claim 6, characterized in that: When the speed tracking meets both conditions 1 and 2, the system will automatically switch the proportional factor α to the identification result output by the PI controller. When the motor speed changes, the proportional factor α is automatically switched to the last value before the change until the system speed stabilizes again and meets conditions 1 and 2, and then automatically switches to the identification result output by the PI controller.

8. An online self-switching identification and compensation device for a hyperlocal model scale factor, characterized in that: include: A model building unit for building a super-local model of a permanent magnet synchronous motor; A mismatch performance analysis unit, used for performing mismatch performance analysis on the designed scale factor in the hyperlocal model to ensure that the deviation between the designed scale factor and the optimal scale factor is lower than a preset threshold; The scale factor identification unit is used to identify the designed scale factor, input the designed reference value into the identification model and then output the scale factor observation value to ensure that the scale factor observation value converges to the reference value; The online compensation and self-switching unit is used for online compensation and self-switching of the designed proportional factor.

9. A storage medium, characterized in that: The storage medium stores a program file capable of implementing the online self-switching identification and compensation method of the hyperlocal model scale factor described in any one of claims 1 to 7.

10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the online self-switching identification and compensation method of the hyperlocal model scale factor according to any one of claims 1 to 7.

Images