Servo system speed loop parameter self-tuning method based on rotational inertia identification

Abstract

The invention belongs to the technical field of motion control, and discloses a servo system speed loop parameter self-tuning method based on rotational inertia identification, which comprises the following steps: constructing a reference adaptive model based on a variable gain model; constructing a parameter relation equation set of the open-loop cut-off frequency of the speed loop and the phase margin, and obtaining a function relation between the time domain proportional gain and the total rotational inertia identification value and between the time domain integral gain and the total rotational inertia identification value based on the parameter relation equation set; discretizing the time domain transfer function of the speed loop by adopting a bilinear method to obtain a discrete domain proportional gain and a discrete domain integral gain; and the speed and the electromagnetic torque under the motor operation condition are collected in real time, the total rotational inertia identification value is output in real time, and the speed loop proportional gain and the speed loop integral gain are dynamically adjusted according to the discrete domain proportional gain and the discrete domain integral gain. The problems that a traditional model reference self-adaptive rotational inertia identification algorithm cannot give consideration to speed performance and accuracy at the same time and a fusion type self-tuning algorithm is high in calculation complexity are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of motion control, and in particular relates to a method for self-tuning speed loop parameters of a servo system based on moment of inertia identification. Background Art

[0002] Permanent magnet synchronous servo motors (PMSMs), with their high power density, fast dynamic response, and high control accuracy, have become core drive components in industrial automation, robotics, electric vehicles, and other fields. Their control performance is highly dependent on the coordinated optimization of the current, velocity, and position loops within vector control technology. As the core component of each control loop, the PI controller (Proportional-Integral Controller) requires precise parameter tuning, which directly determines the system's steady-state accuracy and dynamic response characteristics. However, traditional PI controller parameter tuning relies on manual trial and error, which is not only inefficient but also makes it difficult to ensure robust control performance when load inertia or operating conditions change. In this context, parameter self-tuning technology has become a key breakthrough in improving the intelligence and adaptability of servo systems.

[0003] As a core parameter in the servo system's dynamic model, the accuracy of its identification directly impacts the effectiveness of PI controller parameter auto-tuning. The dynamic response of a motor-driven system (such as acceleration and braking characteristics) is closely related to the load's moment of inertia: excessive moment of inertia can lead to sluggish system response, while insufficient moment of inertia can easily cause overshoot or oscillation. However, traditional model-referenced adaptive moment of inertia identification algorithms employ fixed gain switching based on a set threshold. Improper threshold setting can lead to gain oscillation, affecting identification speed and failing to strike a good balance between speed and accuracy.

[0004] Among the current parameter self-tuning algorithms based on the fusion of models and rules, most algorithms need to simultaneously perform model parameter identification and data-driven rule optimization online, resulting in a surge in computing power requirements, a significant burden on the embedded platform, and a long time to complete the tuning process, resulting in slow drive control response. Summary of the Invention

[0005] The purpose of the present invention is to provide a servo system speed loop parameter self-tuning method based on moment of inertia identification, so as to solve the problems that the traditional model reference adaptive moment of inertia identification algorithm cannot take into account both speed and accuracy at the same time and the fusion type self-tuning algorithm has high computational complexity.

[0006] To achieve the above object, the technical solution adopted by the present invention is:

[0007] A method for self-tuning speed loop parameters of a servo system based on moment of inertia identification, comprising:

[0008] A reference adaptive model based on a variable gain model is constructed to identify the total moment of inertia of the servo system.

[0009] Establish the open-loop transfer function of the servo system velocity loop, construct the parameter relationship equation group of the velocity loop open-loop cutoff frequency and phase margin, and obtain the functional relationship between the time-domain proportional gain and the time-domain integral gain and the total moment of inertia identification value based on the parameter relationship equation group;

[0010] The time domain transfer function of the velocity loop is discretized using a bilinear method to obtain the discrete domain proportional gain and discrete domain integral gain.

[0011] The speed and electromagnetic torque of the motor are collected in real time during operation, and a reference adaptive model based on a variable gain model is used to output the total moment of inertia identification value in real time. The speed loop proportional gain and speed loop integral gain are dynamically adjusted according to the discrete domain proportional gain and discrete domain integral gain.

[0012] Several optional methods are also provided below, but they are not intended to be additional limitations on the above-mentioned overall solution. They are merely further supplements or optimizations. Under the premise that there are no technical or logical contradictions, each optional method can be combined separately for the above-mentioned overall solution, or multiple optional methods can be combined.

[0013] Preferably, the reference adaptive model based on the variable gain model maintains an adaptive gain, and the adaptive gain is adjusted as follows:

[0014] If the relative change rate of the current total moment of inertia identification value is greater than the sum of the relative change rate threshold and the hysteresis band width, the maximum gain value is taken as the latest adaptive gain;

[0015] If the relative change rate of the current total moment of inertia identification value is less than or equal to the difference between the relative change rate threshold and the hysteresis band width, the difference between the maximum gain and the minimum gain is calculated, and the product of the relative change rate of the current total moment of inertia identification value and the difference is taken and added to the minimum gain to obtain the latest adaptive gain;

[0016] Otherwise, the last adaptive gain is used as the latest adaptive gain.

[0017] Preferably, the functional relationship between the time-domain proportional gain, the time-domain integral gain, and the total moment of inertia identification value obtained based on the parameter relationship equation group includes:

[0018]

[0019] Where, is the time domain proportional gain, is the total moment of inertia identification value, is the inverter carrier frequency, is the motor torque coefficient, is the time domain integral gain.

[0020] Preferably, the discrete domain proportional gain and discrete domain integral gain include:

[0021]

[0022] Where, is the discrete domain proportional gain, is the discrete domain integral gain, is the time domain proportional gain, is the time domain integral gain, is the speed loop control period.

[0023] Preferably, the dynamically adjusting the speed loop proportional gain and the speed loop integral gain according to the discrete domain proportional gain and the discrete domain integral gain includes:

[0024] If the total moment of inertia identification value is stable, proceed to the next step; otherwise, re-collect the speed and electromagnetic torque of the motor under operation, and use the reference adaptive model based on the variable gain model to output the total moment of inertia identification value in real time;

[0025] Calculate the speed error and dynamic error reference value based on the target speed and the actual speed collected in real time;

[0026] Dynamically adjust the speed loop proportional gain and speed loop integral gain as follows:

[0027]

[0028] Where, Indicates the The speed loop proportional gain of the speed loop control cycle is: Indicates the The speed loop integral gain of the speed loop control cycle is: is the discrete domain proportional gain, is the discrete domain integral gain, For the The speed error of a speed loop control cycle, To take the absolute value operation, For the Dynamic error reference value of a speed loop control cycle, For the The speed error change rate of the speed loop control cycle is: is the integral adjustment factor.

[0029] Preferably, the stability judgment process of the total moment of inertia identification value is as follows:

[0030] Calculate the relative rate of change of the total moment of inertia identification value. If the relative change rate within each speed loop control cycle is less than the change threshold, the total rotational inertia identification value is judged to be stable; otherwise, the total rotational inertia identification value is unstable.

[0031] Preferably, the speed error is the difference between the target speed and the actual speed collected in real time;

[0032] The dynamic error reference value is obtained as follows: the difference between the absolute value of the current speed error and half of the previous dynamic error reference value is taken, the product of the learning rate and the difference is added to the previous dynamic error reference value to obtain the current dynamic error reference value.

[0033] Compared with existing technologies, this invention offers the following significant advantages: 1) It proposes an adaptive gain switching mechanism based on the relative rate of change of identified inertia. When a large change in identified inertia is detected, it automatically switches to high-gain mode for rapid tracking. In steady-state or slightly changing operating conditions, it performs smooth adjustments based on the change in identified inertia, improving steady-state identification accuracy and effectively balancing speed and accuracy. It also introduces a hysteresis band width to suppress gain oscillation near the threshold. 2) It uses the parameter function relationship established by the moment of inertia identification results to generate the initial solution domain, and further designs a dynamic continuous adjustment mechanism to achieve a smooth transition between transient acceleration convergence and steady-state interference rejection. This allows for online adjustment, making it particularly suitable for embedded scenarios with limited computing power, and addresses the high computational complexity of traditional fusion algorithms. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of a method for self-tuning speed loop parameters of a servo system based on moment of inertia identification according to the present invention;

[0035] Figure 2 This is a schematic diagram of the principle of identifying the total moment of inertia of a servo system using a variable gain model reference adaptive method according to the present invention;

[0036] Figure 3 This is a flow chart of the present invention for dynamically adjusting the proportional integral gain of the speed loop. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0039] Example 1: This example provides a method for self-tuning the speed loop parameters of a servo system based on moment of inertia identification. Figure 1 As shown, the following steps are included:

[0040] Step 1: Construct a variable gain model reference adaptive method to identify the total moment of inertia of the servo system. The specific process is as follows: Figure 2 shown.

[0041] Ignoring the motor friction torque, the permanent magnet synchronous motor rotor motion equation is as follows:

[0042]

[0043] Where, represents the electromagnetic torque of the motor, Represents the load torque of the motor, Indicates the motor speed About time The differential of Indicates the total moment of inertia of the servo system.

[0044] The permanent magnet synchronous motor rotor motion equations are respectively Moment and Discretization of time yields:

[0045]

[0046] Where, Indicates the The electromagnetic torque of a speed loop control cycle, Indicates the The electromagnetic torque of a speed loop control cycle, Indicates the The load torque of a speed loop control cycle, Indicates the The load torque of a speed loop control cycle, is the speed loop control period, Indicates the The motor speed of a speed loop control cycle, Indicates the The motor speed of a speed loop control cycle, Indicates the The motor speed during a speed loop control cycle.

[0047] Normally, during the motor enabling process, the load torque changes slightly in a speed loop control cycle. Moment and At this moment, the load torque remains unchanged. Subtracting the two equations above, we have:

[0048]

[0049] Define the electromagnetic torque difference , , then there is a reference model:

[0050]

[0051] Based on this reference model, an adjustable model is built. 、 The speed and electromagnetic torque of each speed loop control cycle are estimated The speed at the moment can be obtained:

[0052]

[0053] Where, It is The estimated speed of a speed loop control cycle, It is speed loop control cycle and The relevant estimate, estimated value.

[0054] Subtract the reference model from the adjustable model to obtain the velocity error expression of the two models:

[0055]

[0056] During the identification process, the goal is to reduce the deviation between the reference model and the adjustable model. When the deviation between the reference model and the adjustable model gradually decreases and approaches zero infinitely, that is, when the allowable error range is met, the true value can be estimated by the estimated value, thereby achieving the identification purpose.

[0057] According to Popov's hyperstability theory, the formula for constructing a reference adaptive model based on a variable gain model is as follows:

[0058]

[0059] Where, It is speed loop control cycle and The relevant estimates, It is The adaptive gain of the speed loop control cycle, For the The motor electromagnetic torque error during a speed loop control cycle is: For the The speed error of a speed loop control cycle.

[0060] When the adaptive gain When the adaptive gain increases, the convergence speed of inertia identification is accelerated, but the dynamic fluctuation amplitude will be larger; on the contrary, when the adaptive gain When it decreases, the convergence speed of inertia identification becomes slower, but the dynamic fluctuation amplitude of the system is smaller.

[0061] The traditional model reference adaptive moment of inertia identification algorithm performs fixed gain switching according to the set threshold. Improper threshold setting may cause gain oscillation, affect the identification speed, and fail to strike a good balance between speed and accuracy. Therefore, this embodiment proposes an adaptive gain switching mechanism based on the relative change rate of the identified inertia. When a large change in the identified inertia is detected, it automatically switches to the high gain mode to achieve fast tracking; in steady-state or slightly changing working conditions, it performs smooth adjustment according to the change in the identified inertia, improves the steady-state identification accuracy, and effectively balances speed and accuracy; at the same time, it also introduces the hysteresis loop width , suppressing the gain from oscillating near the threshold, the adaptive change law adjustment process of the gain coefficient is as follows:

[0062] If the relative change rate of the current total moment of inertia identification value is greater than the sum of the relative change rate threshold and the hysteresis band width, the maximum gain is taken as the latest adaptive gain; if the relative change rate of the current total moment of inertia identification value is less than or equal to the difference between the relative change rate threshold and the hysteresis band width, the difference between the maximum gain and the minimum gain is calculated, and the product of the relative change rate of the current total moment of inertia identification value and the difference is taken and added to the minimum gain to obtain the latest adaptive gain; otherwise, the last adaptive gain is taken as the latest adaptive gain. The formula is as follows:

[0063]

[0064] Where, is the relative rate of change of the identified value of the total moment of inertia, and , Indirectly reflects the degree of change of system inertia, For the The total moment of inertia identification value of a speed loop control cycle, For the Total moment of inertia identification value of a speed loop control cycle, relative change rate threshold , used to determine the degree of change in identification inertia and the width of the hysteresis loop make exist Lock the gain within the range to avoid adaptive gain Frequent switching. It is the maximum value of the adaptive gain (i.e. the maximum gain), which ensures the system's rapid response to large inertia changes and serves as the initial value of the adaptive gain. It is the minimum value of the adaptive gain (i.e. the minimum gain), which prevents the system from being overly sensitive to small inertia fluctuations.

[0065] Step 2: Establish the open-loop transfer function of the servo system speed loop , build the speed loop open loop cutoff frequency and phase margin The time domain proportional gain is analytically derived based on the parameter relationship equations of and time domain integral gain and total moment of inertia identification value The functional relationship is as follows:

[0066] Open-loop transfer function The formula is as follows:

[0067]

[0068] Where, is a complex variable, is the motor torque coefficient.

[0069] Open-loop cutoff frequency The frequency value corresponding to the amplitude of the open-loop frequency response function is 1; the phase margin is the open loop cutoff frequency The angle difference between the phase of the open-loop cutoff frequency response function and the critical phase (-180°) has an impact on the open-loop transfer function. Perform frequency response analysis, Substitution , and construct the speed loop open loop cutoff frequency according to the above definition and phase margin The mapping relationship equation between:

[0070]

[0071] Where, represents the open-loop frequency response function The amplitude of represents the open-loop frequency response function The phase angle, represents the imaginary unit, represents the arctan function. The time domain proportional gain can be derived as and time domain integral gain and total moment of inertia identification value The functional relationship is as follows:

[0072]

[0073] Where, Represents the cotangent function. This embodiment sets it based on experience. , , the total moment of inertia identification value Substituting in, the time domain parameter expression of the speed PI controller is as follows:

[0074]

[0075] Where, is the frequency of the inverter carrier.

[0076] Step 3: Use the bilinear method to discretize the time domain transfer function of the velocity loop to obtain the discrete domain proportional gain and discrete domain integral gain , as follows:

[0077] The time domain transfer function of the speed PI controller is , discretize it using the bilinear method, that is, Bring in Get the Z domain expression:

[0078]

[0079] Where, It represents the Z-domain output expression of the discretized speed PI controller, It represents the Z-domain input expression of the discretized speed PI controller.

[0080] Cross-multiply the Z-domain expressions and take the inverse Z transform to obtain the discretized differential equation of the speed PI controller:

[0081]

[0082] Where, Indicates the The PI controller output of a speed loop control cycle, Indicates the The PI controller output of a speed loop control cycle, Indicates the The speed error signal input by the PI controller of the speed loop control cycle is Indicates the The speed error signal is input to the PI controller of each speed loop control cycle.

[0083] According to the difference equation, the time domain proportional gain can be obtained and time domain integral gain and discrete domain parameters and The relationship is as follows:

[0084]

[0085] Step 4: Drive the motor to run, collect the speed and torque of the motor in real time, and use the variable gain model reference adaptive method to output the total moment of inertia identification value in real time, and dynamically adjust the speed loop proportional gain and speed loop integral gain according to the discrete domain proportional gain and discrete domain integral gain. Figure 3 The specific steps are as follows:

[0086] Step 4-1: Drive the motor to run, collect speed and electromagnetic torque in real time, and use the variable gain model reference adaptive method to output the total moment of inertia identification value in real time. , determine the total moment of inertia identification value Is it stable? If continuous Relative rate of change within (e.g. 6) speed loop control cycles If both are less than the change threshold (e.g. 2%, which can be adjusted by 1%), the total moment of inertia identification value is considered stable; otherwise, the total moment of inertia identification value is unstable. If stable, proceed to the next step; otherwise, repeat step 4-1.

[0087] Step 4-2: Obtain speed command (target speed) in each speed loop control cycle The actual speed of feedback Calculate speed error And dynamic error reference value , For the Dynamic error reference value of a speed loop control cycle, For the Dynamic error reference value of a speed loop control cycle, learning rate Control the convergence speed and limit ,in is the error expected by the user, and As a dynamic error reference value The initial value of (For example, 5) speed loop control cycles Sliding average filter calculates the speed error change rate after smoothing , maintain an error queue of length 5 to store historical values to , remove the oldest data each time you update and add the latest , calculate the average of adjacent error differences.

[0088] Step 4-3: Dynamically adjust the speed loop proportional gain and speed loop integral gain :

[0089]

[0090] Among them, the integral adjustment factor , integral adjustment factor The denominator 10 is not a fixed value, but an empirical parameter that can be flexibly adjusted according to system characteristics: when the denominator is reduced, The value increases, thereby increasing the speed loop integral gain Increase, the inhibitory effect is weakened, allowing stronger integral compensation to improve the dynamic response speed; on the contrary, increasing the denominator The value decreases, the speed loop integral gain It reduces and enhances the suppression effect, which can effectively suppress transient overshoot but may sacrifice response speed.

[0091] Dynamic error reference value Track the actual error amplitude through online learning to replace the fixed threshold. When the system is in a highly dynamic condition (such as a sudden load change), the dynamic error reference value Automatically increase and expand the gain adjustment range of large error areas; when the system is in steady state, the dynamic error reference value Reduce and improve the control accuracy in small error areas.

[0092] Transient process ( ): The system is in the fast response stage, the dynamic error reference value Because online learning lags behind speed error , has not yet been updated to a high position. At this time, the speed loop proportional gain Approaching , greatly improving the proportional effect to accelerate convergence; at the same time Increase, speed loop integral gain Reduce, suppress integral accumulation, and prevent overshoot. When it increases, the integral adjustment factor Increase, partially offset The influence of the integral action allows moderate compensation of nonlinear disturbances.

[0093] Steady-state process ( ): The system approaches the target state, the speed error Very small, dynamic error reference value It has converged to the minimum value through online learning. At this time, the speed loop proportional gain Approximate discrete domain proportional gain , restore the basic proportional gain to avoid oscillation caused by over-adjustment; Decrease, speed loop integral gain Approaching discrete domain integral gain , restore the integral action and eliminate the steady-state error; dynamic error reference value Reduction leads to integral adjustment factor Reduce and strengthen the integral suppression to avoid steady-state micro-oscillation.

[0094] The present invention proposes an adaptive gain switching mechanism with hysteresis characteristics, which dynamically adjusts the identification gain coefficient by monitoring the relative change rate of the identified inertia, taking into account both dynamic response speed and steady-state accuracy. The present invention solves the problems of parameter mismatch and high computational complexity of the optimization algorithm when the moment of inertia changes suddenly in traditional methods. It is particularly suitable for parameter adjustment of embedded servo systems, and has the characteristics of fast dynamic response, high steady-state accuracy, and low computing resource usage.

[0095] Example 2: The present invention also provides a servo system speed loop parameter self-tuning device based on rotational inertia identification, including a processor and a memory storing a plurality of computer instructions. When the computer instructions are executed by the processor, the steps of the servo system speed loop parameter self-tuning method based on rotational inertia identification are implemented.

[0096] Regarding the specific limitations of the servo system speed loop parameter self-tuning device based on rotational inertia identification, please refer to the limitations of the servo system speed loop parameter self-tuning method based on rotational inertia identification above, which will not be repeated here.

[0097] The memory and processor are electrically connected, directly or indirectly, to enable data transmission or interaction. For example, these components may be electrically connected via one or more communication buses or signal lines. The memory stores a computer program executable on the processor, and the processor implements the method of the present invention by executing the computer program stored in the memory.

[0098] The memory may be, but is not limited to, random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), etc.

[0099] The processor may be an integrated circuit chip with data processing capabilities. Such a processor may be a general-purpose processor, including a central processing unit (CPU) or a network processor (NP). It may implement or execute the methods, steps, and logic diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor.

[0100] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0101] The above-described embodiments merely illustrate several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that a person skilled in the art would be able to make numerous modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A servo system speed loop parameter self-tuning method based on moment of inertia identification, characterized in that: The servo system speed loop parameter self-tuning method based on moment of inertia identification includes: A reference adaptive model based on a variable gain model is constructed to identify the total moment of inertia of the servo system. Establish the open-loop transfer function of the servo system velocity loop, construct the parameter relationship equation group of the velocity loop open-loop cutoff frequency and phase margin, and obtain the functional relationship between the time-domain proportional gain and the time-domain integral gain and the total moment of inertia identification value based on the parameter relationship equation group; The time domain transfer function of the velocity loop is discretized using a bilinear method to obtain the discrete domain proportional gain and discrete domain integral gain. The speed and electromagnetic torque of the motor are collected in real time during operation, and a reference adaptive model based on a variable gain model is used to output the total moment of inertia identification value in real time. The speed loop proportional gain and speed loop integral gain are dynamically adjusted according to the discrete domain proportional gain and discrete domain integral gain.

2. The servo system speed loop parameter self-tuning method based on moment of inertia identification according to claim 1 is characterized in that: The reference adaptive model based on the variable gain model maintains an adaptive gain, and the adjustment process of the adaptive gain is as follows: If the relative change rate of the current total moment of inertia identification value is greater than the sum of the relative change rate threshold and the hysteresis band width, the maximum gain value is taken as the latest adaptive gain; If the relative change rate of the current total moment of inertia identification value is less than or equal to the difference between the relative change rate threshold and the hysteresis band width, the difference between the maximum gain and the minimum gain is calculated, and the product of the relative change rate of the current total moment of inertia identification value and the difference is taken and added to the minimum gain to obtain the latest adaptive gain; Otherwise, the last adaptive gain is used as the latest adaptive gain.

3. The servo system speed loop parameter self-tuning method based on moment of inertia identification according to claim 1 is characterized in that: The functional relationship between the time domain proportional gain, the time domain integral gain and the total moment of inertia identification value obtained based on the parameter relationship equation group includes: ； Where, is the time domain proportional gain, is the total moment of inertia identification value, is the inverter carrier frequency, is the motor torque coefficient, is the time domain integral gain.

4. The servo system speed loop parameter self-tuning method based on moment of inertia identification according to claim 1 is characterized in that: The discrete domain proportional gain and discrete domain integral gain include: ； Where, is the discrete domain proportional gain, is the discrete domain integral gain, is the time domain proportional gain, is the time domain integral gain, is the speed loop control period.

5. The servo system speed loop parameter self-tuning method based on moment of inertia identification according to claim 1 is characterized in that: The dynamically adjusting the speed loop proportional gain and the speed loop integral gain according to the discrete domain proportional gain and the discrete domain integral gain includes: If the total moment of inertia identification value is stable, proceed to the next step; otherwise, re-collect the speed and electromagnetic torque of the motor under operation, and use the reference adaptive model based on the variable gain model to output the total moment of inertia identification value in real time; Calculate the speed error and dynamic error reference value based on the target speed and the actual speed collected in real time; Dynamically adjust the speed loop proportional gain and speed loop integral gain as follows: ； Where, Indicates the The speed loop proportional gain of the speed loop control cycle is: Indicates the The speed loop integral gain of the speed loop control cycle is: is the discrete domain proportional gain, is the discrete domain integral gain, For the The speed error of a speed loop control cycle, To take the absolute value operation, For the Dynamic error reference value of a speed loop control cycle, For the The speed error change rate of the speed loop control cycle is: is the integral adjustment factor.

6. The method for self-tuning speed loop parameters of a servo system based on moment of inertia identification according to claim 5, characterized in that: The stability judgment process of the total moment of inertia identification value is as follows: Calculate the relative rate of change of the total moment of inertia identification value. If the relative change rate within each speed loop control cycle is less than the change threshold, the total moment of inertia identification value is judged to be stable; Otherwise, the total moment of inertia identification value will be unstable.

7. The method for self-tuning speed loop parameters of a servo system based on moment of inertia identification according to claim 5, characterized in that: The speed error is the difference between the target speed and the actual speed collected in real time; The dynamic error reference value is obtained as follows: the difference between the absolute value of the current speed error and half of the previous dynamic error reference value is taken, the product of the learning rate and the difference is added to the previous dynamic error reference value to obtain the current dynamic error reference value.

Images