CN104124908A - Inertia ratio on-line identifying system and method - Google Patents

Abstract

The invention provides an inertia ratio on-line identifying system and method. The inertia ratio on-line identifying system comprises a sampling unit, a first computing unit and a second computing unit, wherein the sampling unit is used for collecting rotation speeds and corresponding current of a plurality of servo motors in fixed sampling time when the servo motors are started or reversed; the first computing unit is used for calculating electromagnetic torque according to the current, and subjecting the collected rotation speeds and the calculated electromagnetic torque to iterative computation to obtain an identification vector; the second computing unit is used for calculating an inertia ratio according to the identification vector. According to the inertia ratio on-line identifying system and method, iterative computation of the identification vector from the sampled rotation speeds is performed, and the inertia ratio is calculated according to the identification vector, so that inertia ratio parameters of the servo motors can be obtained without off-line operation, and user's operation is simplified.

Description

System and method for online identification of inertia ratio

technical field

The invention relates to the field of servo motor control, more specifically, to a system and method for on-line identification of inertia ratio.

Background technique

In the servo system, the inertia ratio is an important control factor, and it is the basis for establishing the loop model. The inertia ratio is the ratio between the load inertia and the motor inertia. According to the inertia ratio, it can be estimated whether the acceleration and deceleration time of the servo system can meet the equipment process requirements.

At present, the most commonly used inertia ratio identification method is based on offline identification, that is, the acceleration change is obtained through acceleration and deceleration commands, and then the inertia ratio is calculated. Although the above-mentioned off-line inertia ratio identification can obtain a relatively accurate servo inertia ratio, it cannot cope with changes in the external inertia. If the control parameters do not follow the change of the external inertia ratio, it will often cause the control performance of the servo motor to decline and affect the use effect.

Friction force is an important external force disturbance in servo control, which has a decisive influence on the stop phenomenon of the servo motor during reverse operation. In order to make corresponding compensation, the friction model is often used to calculate the Coulomb friction force at the zero crossing point. When the current friction model calculates the Coulomb friction force, it is difficult to obtain correct results due to the influence of the accuracy of the friction model. However, if a complex model is used, the amount of calculation will increase greatly in practical application, making it difficult to realize real-time operation.

Contents of the invention

The technical problem to be solved by the present invention is to provide an online inertia ratio identification system and method for the above-mentioned problem that the offline inertia ratio identification cannot cope with external inertia changes.

The technical solution of the present invention to solve the above technical problems is to provide a system for on-line identification of inertia ratio, including a sampling unit, a first calculation unit and a second calculation unit, wherein: the sampling unit is used to start or reverse the servo motor In the time direction, the rotational speeds and corresponding currents of multiple servo motors are collected at a fixed sampling time; the first calculation unit is used to calculate the electromagnetic torque according to the currents, and use the collected rotational speeds and the calculated electromagnetic torques The torque iteratively calculates the identification vector; the second calculation unit is used to calculate the inertia ratio according to the identification vector.

In the system for identifying the inertia ratio online according to the present invention, the identification vector calculated by the first calculation unit is: $b\left(k\right)\≈\left[\begin{array}{ccc}1-\frac{\mathrm{bts}}{J}& \frac{\mathrm{Ts}}{J}& \frac{\mathrm{Ts}}{J}{T}_l(k-1)\end{array}\right],$ And iteratively calculate the identification vector through the following calculation formula: $b\left(k\right)=\frac{\mathrm{\β\ψ}(k-1)[\mathrm{the\; y}\left(k\right)-{\mathrm{\ψ}}^T(k-1)b(k-1)]}{1+{\mathrm{\ψ}}^T(k-1)\mathrm{\β\ψ}(k-1)}+b(k-1),$ Where ψ(k-1)=[y(k-1) T _e (k-1)-1] ^T , and y is the servo motor speed obtained by sampling, T _e is the electromagnetic torque, B is the friction coefficient, T _s is the sampling time, T _l is the load torque, J is the sum of motor and load inertia, and β is the recursive variable gain.

In the system for on-line identification of inertia ratio according to the present invention, the recursive variable gain $\mathrm{\β}\left(k\right)=\mathrm{\β}(k-1)-\frac{\mathrm{\β}(k-1)\mathrm{\ψ}(k-1){\mathrm{\ψ}}^T(k-1)\mathrm{\β}(k-1)}{\mathrm{\λ}+{\mathrm{\ψ}}^T(k-1)\mathrm{\β}(k-1)\mathrm{\ψ}(k-1)},$ Where λ is a constant greater than 0.

In the system for identifying the inertia ratio online according to the present invention, the first calculation unit includes a data conversion subunit and a floating-point number operation subunit, wherein: the data conversion subunit is used to convert the sampled servo motor speed and The corresponding current is converted into a power series expressed in hexadecimal, and each of the power series is converted into a custom data structure, and the custom data structure includes a character array after the floating-point number is split, and the effective bit length of the floating-point number , floating-point number power series, and floating-point number positive and negative four defined variables; the floating-point number operation subunit uses the self-defined data structure to replace the floating-point number and substitutes the corresponding calculation formula to complete the operation.

In the system for on-line identification of inertia ratio according to the present invention, the second calculation unit is also used to calculate the load torque according to the identification vector; the system also includes a compensation calculation unit for Perform forward friction compensation and reverse friction compensation.

The present invention also provides a method for online identification of servo inertia ratio, comprising the following steps:

(a) When the servo motor starts or reverses, collect multiple rotational speeds and corresponding currents of the servo motor at a fixed sampling time;

(b) Calculate the electromagnetic torque according to the current, and iteratively calculate the identification vector by using the collected rotational speed and the calculated electromagnetic torque;

(c) Calculating an inertia ratio based on the identification vector.

In the method for online identification of inertia ratio according to the present invention, in the step (b), the identification vector is: $b\left(k\right)\≈\left[\begin{array}{ccc}1-\frac{\mathrm{bts}}{J}& \frac{\mathrm{Ts}}{J}& \frac{\mathrm{Ts}}{J}{T}_l(k-1)\end{array}\right],$ And the identification vector is iteratively calculated by the following formula: $b\left(k\right)=\frac{\mathrm{\β\ψ}(k-1)[\mathrm{the\; y}\left(k\right)-{\mathrm{\ψ}}^T(k-1)b(k-1)]}{1+{\mathrm{\ψ}}^T(k-1)\mathrm{\β\ψ}(k-1)}+b(k-1),$ Where ψ(k-1)=[y(k-1) T _e (k-1) -1] ^T , y is the servo motor speed, T _e is the electromagnetic torque, B is the friction coefficient, T _s is the sampling time , T _l is the load torque, J is the sum of motor and load inertia, β is the recursive variable gain.

In the method for online identification of inertia ratio according to the present invention, the recursive variable gain $\mathrm{\β}\left(k\right)=\mathrm{\β}(k-1)-\frac{\mathrm{\β}(k-1)\mathrm{\ψ}(k-1){\mathrm{\ψ}}^T(k-1)\mathrm{\β}(k-1)}{\mathrm{\λ}+{\mathrm{\ψ}}^T(k-1)\mathrm{\β}(k-1)\mathrm{\ψ}(k-1)},$ Where λ is a constant greater than 0.

In the method for online identification of inertia ratio according to the present invention, said step (b) comprises the following steps:

(b1) converting the rotational speed of the sampled servo motor and the corresponding current into a power series expressed in hexadecimal;

(b2) converting each described power series into a custom data structure, the self-defined data structure including character arrays after floating-point number splitting, floating-point number effective bit length, floating-point number power series, floating-point number positive and negative four a defined variable;

(b3) Using the self-defined data structure to replace the floating-point number and substituting the corresponding calculation formula to complete the operation.

In the method for online identification of inertia ratio according to the present invention, the step (c) includes: calculating the load torque according to the identification vector; after the step (c), it includes: performing forward rotation according to the load torque Friction compensation and reverse friction compensation.

The system and method for on-line identification of inertia ratio of the present invention can iteratively calculate the identification vector through the sampled rotational speed and current and obtain the inertia ratio through the calculation of the identification vector. The inertia ratio parameter of the servo motor can be obtained without offline operation, which simplifies the user operation. Moreover, the present invention performs friction compensation through the obtained inertia ratio parameters, which can reduce the pause when the servo motor reverses.

Description of drawings

Figure 1 is a schematic diagram of servo drive control.

Fig. 2 is a schematic diagram of an embodiment of the system for on-line identification of inertia ratio according to the present invention.

Figure 3 is a schematic diagram of friction compensation.

Fig. 4 is a schematic diagram of another friction compensation.

Fig. 5 is a schematic flowchart of an embodiment of a method for online identification of inertia ratio in the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

As shown in Figure 1, it is a schematic diagram of the control of the servo drive. In this servo driver, in order to simplify the calculation, the friction force can be regarded as a linear function related to the speed, so the motion equation of the servo driver driving the servo motor can be described by formula (1) (ω' is the motor rotor acceleration):

$\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; \text{'}</span><span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; B</span>}}{\mathrm{<span\; class="notranslate">\; J</span>}}\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; +</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; J</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; e</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

Among them, J is the sum of the motor inertia and the load inertia, T _e is the electromagnetic torque, T _l is the load torque, and B is the friction coefficient. Taking formula (1) as a part of state equation (2), and using the discretization method described in formula (3) for calculation, the discretization equation expressed by calculation formula (4) can finally be obtained.

$\left\{\begin{array}{c}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \text{'}</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; Ax</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; Bu</span>}\\ \mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; Cx</span>}\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

$\left\{\begin{array}{c}\mathrm{<span\; class="notranslate">\; \&Phi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; At</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; L</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; si</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; ]</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\\ \mathrm{<span\; class="notranslate">\; G</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&Phi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&Phi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>{<span\; class="notranslate">\; |</span>}_{\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; T</span>}}\\ \mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; \&Integral;</span>}_{\mathrm{<span\; class="notranslate">\; 0</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; At</span>}}\mathrm{<span\; class="notranslate">\; Bdt</span>}\\ \mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; G</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; kT</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; kT</span>}<span\; class="notranslate">\; )</span>\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

$\left\{\begin{array}{c}\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; bts</span>}}{\mathrm{<span\; class="notranslate">\; J</span>}}}\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; bts</span>}}{\mathrm{<span\; class="notranslate">\; J</span>}}}}{\mathrm{<span\; class="notranslate">\; B</span>}}{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; e</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; bts</span>}}{\mathrm{<span\; class="notranslate">\; J</span>}}}}{\mathrm{<span\; class="notranslate">\; B</span>}}{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\\ \mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 4</span>}<span\; class="notranslate">\; )</span>$

make $b\left(k\right)=\left[\begin{array}{ccc}{e}^{-\frac{\mathrm{bts}}{J}}& \frac{1-e^{-\frac{\mathrm{bts}}{J}}}{B}& \frac{1-e^{-\frac{\mathrm{bts}}{J}}}{B}{T}_l(k-1)\end{array}\right]$ To identify the vector, in order to simplify the calculation amount in the program, the identification vector can be obtained by using Taylor series expansion:

$\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&ap;</span>\left[\begin{array}{ccc}\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; bts</span>}}{\mathrm{<span\; class="notranslate">\; J</span>}}& \frac{\mathrm{<span\; class="notranslate">\; Ts</span>}}{\mathrm{<span\; class="notranslate">\; J</span>}}& \frac{\mathrm{<span\; class="notranslate">\; Ts</span>}}{\mathrm{<span\; class="notranslate">\; J</span>}}{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\end{array}\right]<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

Let ψ(k-1)=[ω(k-1) T _e (k-1) -1] ^T is the known input and output data, that is, ω(k-1) is the sampling speed (the last sampling time ), T _e (k-1) is the electromagnetic torque calculated according to the sampled current (the last sampling time). At this time, the recursive formula of calculation formula (6) can be used for identification:

$\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; \&beta;\&psi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&psi;</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span>}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&psi;</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; \&beta;\&psi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

In this way, after calculating and obtaining the value of the above-mentioned identification vector, a ternary linear equation system can be obtained by combining the calculation formula (5), and by solving the ternary linear equation system, the friction coefficient B, the motor and load inertia sum J, and the load Torque T _l to further obtain the inertia ratio.

As shown in FIG. 2 , it is a schematic diagram of an embodiment of the system for identifying the inertia ratio online in the present invention. The system of this embodiment includes a sampling unit 21, a first computing unit 22, and a second computing unit 23, and the sampling unit 21, the first computing unit 22, and the second computing unit 23 can be integrated into the digital signal processor of the servo driver, of course It can also be composed of one or more chips connected to the above-mentioned digital signal processor.

Sampling unit 21 is used for when the servomotor starts or reverses (that is, at the zero-crossing point of the servomotor as the start time of sampling, so that the calculated frictional force is Coulomb friction), collect the values of a plurality of servomotors with a fixed sampling time Ts speed and the corresponding current. The input end of the sampling unit 21 can be connected to the encoder on the servo motor, and the rotation speed of the servo motor can be calculated according to the output signal of the encoder. Moreover, the above-mentioned sampling unit 21 can directly obtain the current data by sampling the output of the AD conversion unit of the servo driver. In particular, the sampling unit 21 can sample 200 sets of servo motor speeds and corresponding current data to ensure the accuracy of inertia ratio identification (the more sampled data sets, the more accurate the identification, but the slower the relative identification speed). The sampling interval of the sampling unit 21 (ie, the sampling time Ts) may be about 1 millisecond.

The first calculation unit 22 calculates the electromagnetic torque according to the sampled current and the torque coefficient of the servo motor (each current corresponds to an electromagnetic torque), and iteratively calculates the identification vector with the collected rotational speed and the calculated electromagnetic torque. Specifically, the identification vector is: $b\left(k\right)\&ap;\left[\begin{array}{ccc}1-\frac{\mathrm{bts}}{J}& \frac{\mathrm{Ts}}{J}& \frac{\mathrm{Ts}}{J}{T}_l(k-1)\end{array}\right],$ Among them, T _s is the sampling time, T _l is the load torque, and J is the sum of motor and load inertia.

Specifically, the above identification vector can be obtained by iterative calculation through the following calculation formula: $b\left(k\right)=\frac{\mathrm{\&beta;\&psi;}(k-1)[\mathrm{the\; y}\left(k\right)-{\mathrm{\&psi;}}^T(k-1)b(k-1)]}{1+{\mathrm{\&psi;}}^T(k-1)\mathrm{\&beta;\&psi;}(k-1)}+b(k-1),$ And ψ(k-1)=[y(k-1) T _e (k-1) -1] ^T , y (ie y(k)) is the servo motor speed obtained by sampling, T _e (ie T _e ( k-1)) is the electromagnetic torque, and β is the recursive variable gain.

If the above β value is a constant value, then the recursion process will not correctly reflect the convergence process of the recursion (that is, iterative calculation). If there is a sudden change in the input data (that is, the speed and electromagnetic torque), the recursion result will jump , in order to avoid this situation, it is necessary to make the β value attenuate as the recursion proceeds, and its change process can be expressed by formula (7):

$\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; \&psi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; \&psi;</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&psi;</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; \&psi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

Where λ is a constant greater than 0, and in practical applications, the above λ can be directly set to 1.

The second calculation unit 23 is used to calculate the inertia ratio according to the identification vector, that is, according to the value combination of the identification vector calculated by the first calculation unit 22 $b\left(k\right)\&ap;\left[\begin{array}{ccc}1-\frac{\mathrm{bts}}{J}& \frac{\mathrm{Ts}}{J}& \frac{\mathrm{Ts}}{J}{T}_l(k-1)\end{array}\right]$ Obtain a ternary linear equation system and obtain the motor and load inertia sum J, load torque T _l and friction coefficient B by solving the ternary linear equation system, and then calculate the inertia ratio based on the motor and load inertia sum J and motor inertia .

The above-mentioned system for identifying the inertia ratio online can obtain the inertia ratio parameters for subsequent servo motor control without separate offline identification, which simplifies the operation of the servo drive.

Since there are many variables involved in the calculation process of the identification vector of the first calculation unit 22, and the calculation range is uncertain, considering that the servo control is usually realized by fixed-point DSP, and the above calculation process cannot be directly calculated by scaling, it can be Floating-point arithmetic is used in fixed-point DSP to complete the above-mentioned calculation process. In order to reduce the calculation amount of floating-point numbers, the sampling data can be regarded as a power series expressed in hexadecimal, and the following data structure is defined:

And complete the addition, subtraction, multiplication and division operations with the above data structure, and finally calculate the identification vector. That is, the first calculation unit 22 includes a data conversion subunit and a floating-point number operation subunit, wherein the data conversion subunit is used to convert the rotational speed of the servo motor sampled and the corresponding current into a power series expressed in hexadecimal, and convert each A power series is converted into a custom data structure, which includes four definition variables: character array after floating-point number splitting, effective bit length of floating-point number, power series of floating-point number, positive and negative of floating-point number; floating-point number operator The unit uses a custom data structure instead of a floating point number and substitutes the corresponding calculation formula to complete the operation.

In the above-mentioned system for identifying the inertia ratio on-line, a compensation calculation unit may also be included, which is used to calculate the obtained load torque T _l according to the second calculation unit 23, and use the load torque T _l as the servo The friction force at the zero point of the motor is compensated. In order to simplify the calculation, the compensation calculation unit can set the forward friction compensation value a and the reverse friction compensation value -a according to the command speed direction, as shown in FIG. 3 .

The premise of the above compensation method is that it is assumed that the identified load torque is equal to the friction force when the servo motor is running forward and reverse, but this is only a situation in practical application. Another situation is that when the servo motor is installed vertically, the load torque calculated according to the identified inertia ratio includes gravity, because the servo motor has the following expression when running up and down:

$\left\{\begin{array}{c}\mathrm{<span\; class="notranslate">\; Te</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; Ja</span>}}^{<span\; class="notranslate">\; +</span>}\\ \mathrm{<span\; class="notranslate">\; Te</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; Ja</span>}}^{<span\; class="notranslate">\; -</span>}\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 8</span>}<span\; class="notranslate">\; )</span>$

Among them, T _g is the gravitational force, T _f is the frictional force, a ⁺ is the acceleration of the load rising, and a ^- is the acceleration of the load falling, so the identified load torque can be expressed as:

$\left\{\begin{array}{c}{\mathrm{<span\; class="notranslate">\; Tl</span>}}^{<span\; class="notranslate">\; +</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}\\ {\mathrm{<span\; class="notranslate">\; Tl</span>}}^{<span\; class="notranslate">\; -</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 9</span>}<span\; class="notranslate">\; )</span>$

Among them, Tl ⁺ is the load torque when the load is rising, and Tl ^- is the load torque when the load is falling. At this time, it is necessary to combine forward friction compensation a and reverse friction compensation -a with gravity compensation, and the compensation method is shown in Figure 4.

As shown in FIG. 5 , it is a schematic flowchart of an embodiment of a method for online identification of servo inertia ratio in the present invention. The method can be run directly in the servo drive and includes the following steps:

Step S51 : when the servo motor starts or reverses, collect multiple rotational speeds and corresponding currents of the servo motor with a fixed sampling time Ts.

In this step, the zero-crossing point of the servo motor is used as the starting time of sampling, so that the calculated friction force is Coulomb friction. And the speed of the servo motor can be realized by sampling the output signal of the encoder installed on the servo motor, while the current signal can be directly obtained by sampling the output of the AD sampling unit of the servo drive. In particular, in this step, 200 sets of servo motor speeds and corresponding current data can be sampled to ensure the accuracy of inertia ratio identification (the more sampled data sets, the more accurate the identification, but the slower the relative identification speed), and The sampling interval may be about 1 millisecond.

Step S52: Calculate the electromagnetic torque according to the current, and iteratively calculate the identification vector by using the collected rotational speed and the calculated electromagnetic torque.

Step S53: Calculate the inertia ratio, load torque and friction coefficient according to the identification vector.

In this step, the above identification vector is: $b\left(k\right)\&ap;\left[\begin{array}{ccc}1-\frac{\mathrm{bts}}{J}& \frac{\mathrm{Ts}}{J}& \frac{\mathrm{Ts}}{J}{T}_l(k-1)\end{array}\right],$ And the identification vector is iteratively calculated by the following formula: $b\left(k\right)=\frac{\mathrm{\&beta;\&psi;}(k-1)[\mathrm{the\; y}\left(k\right)-{\mathrm{\&psi;}}^T(k-1)b(k-1)]}{1+{\mathrm{\&psi;}}^T(k-1)\mathrm{\&beta;\&psi;}(k-1)}+b(k-1),$ Where ψ(k-1)=[y(k-1) T _e (k-1) -1] ^T , y is the servo motor speed, T _e is the electromagnetic torque, B is the friction coefficient, T _s is the sampling time , T _l is the load torque, J is the sum of motor and load inertia, β is the recursive variable gain. Moreover, in order to avoid sudden changes in the input sampling data and cause jumps in the recursive results, the above-mentioned recursive variable gain can be made $\mathrm{\&beta;}\left(k\right)=\mathrm{\&beta;}(k-1)-\frac{\mathrm{\&beta;}(k-1)\mathrm{\&psi;}(k-1){\mathrm{\&psi;}}^T(k-1)\mathrm{\&beta;}(k-1)}{\mathrm{\&lambda;}+{\mathrm{\&psi;}}^T(k-1)\mathrm{\&beta;}(k-1)\mathrm{\&psi;}(k-1)},$ Where λ is a constant greater than 0.

Moreover, in order to ensure the computing efficiency of the fixed-point DSP in the servo drive, the sampled servo motor speed and the corresponding current can be converted into a power series expressed in hexadecimal, and each power series can be converted into a custom data structure, And use the above-mentioned custom data structure to substitute into the corresponding calculation formula to complete the operation. The above-mentioned custom data structure includes four definition variables of character array after floating-point number splitting, effective bit length of floating-point number, power series of floating-point number, and positive and negative of floating-point number.

In addition, after step S53, it may further include: performing forward friction compensation and reverse friction compensation according to the identified load torque.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present invention can easily think of changes or Replacement should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (10)

1. A system for on-line identification of inertia ratio, characterized in that: it includes a sampling unit, a first calculation unit and a second calculation unit, wherein: the sampling unit is used to start or reverse the servo motor, with a fixed The sampling time collects the rotational speeds and corresponding currents of multiple servo motors; the first calculation unit is used to calculate the electromagnetic torque according to the currents, and iteratively calculates the identification vector by using the collected rotational speeds and the calculated electromagnetic torques ; The second calculation unit is used to calculate the inertia ratio according to the identification vector. 2. The system for online identification of inertia ratio according to claim 1, characterized in that: the identification vector calculated by the first calculation unit is: $b\left(k\right)\&ap;\left[\begin{array}{ccc}1-\frac{\mathrm{bts}}{J}& \frac{\mathrm{Ts}}{J}& \frac{\mathrm{Ts}}{J}{T}_l(k-1)\end{array}\right],$ And iteratively calculate the identification vector through the following calculation formula: $b\left(k\right)=\frac{\mathrm{\&beta;\&psi;}(k-1)[\mathrm{the\; y}\left(k\right)-{\mathrm{\&psi;}}^T(k-1)b(k-1)]}{1+{\mathrm{\&psi;}}^T(k-1)\mathrm{\&beta;\&psi;}(k-1)}+b(k-1),$ Where ψ(k-1)=[y(k-1) T _e (k-1)-1] ^T , and y is the servo motor speed obtained by sampling, T _e is the electromagnetic torque, B is the friction coefficient, T _s is the sampling time, T _l is the load torque, J is the sum of motor and load inertia, and β is the recursive variable gain. 3. The system for online identification of inertia ratio according to claim 2, characterized in that: the recursive variable gain $\mathrm{\&beta;}\left(k\right)=\mathrm{\&beta;}(k-1)-\frac{\mathrm{\&beta;}(k-1)\mathrm{\&psi;}(k-1){\mathrm{\&psi;}}^T(k-1)\mathrm{\&beta;}(k-1)}{\mathrm{\&lambda;}+{\mathrm{\&psi;}}^T(k-1)\mathrm{\&beta;}(k-1)\mathrm{\&psi;}(k-1)},$ Where λ is a constant greater than 0. 4. The system for online identification of inertia ratio according to claim 2 or 3, characterized in that: the first calculation unit includes a data conversion subunit and a floating-point number operation subunit, wherein: the data conversion subunit is used for Converting the rotational speed of the sampled servo motor and the corresponding current into a power series expressed in hexadecimal, and converting each of the power series into a custom data structure, the custom data structure includes floating-point numbers after splitting Character array, effective bit length of floating-point number, power series of floating-point number, positive and negative of floating-point number four definition variables; said floating-point number operation subunit uses said self-defined data structure instead of floating-point number and substitutes into corresponding calculation formula to complete the operation. 5. The system for online identification of inertia ratio according to claim 1, characterized in that: the second calculation unit is also used to calculate the load torque according to the identification vector; the system also includes a compensation calculation unit for Forward friction compensation and reverse friction compensation are performed according to the load torque. 6. A method for online identification of servo inertia ratio, characterized in that: comprising the following steps: (a) When the servo motor starts or reverses, collect multiple rotational speeds and corresponding currents of the servo motor at a fixed sampling time; (b) Calculate the electromagnetic torque according to the current, and iteratively calculate the identification vector by using the collected rotational speed and the calculated electromagnetic torque; (c) Calculating an inertia ratio based on the identification vector. 7. The method for online identification of inertia ratio according to claim 6, characterized in that: in the step (b), the identification vector is: $b\left(k\right)\&ap;\left[\begin{array}{ccc}1-\frac{\mathrm{bts}}{J}& \frac{\mathrm{Ts}}{J}& \frac{\mathrm{Ts}}{J}{T}_l(k-1)\end{array}\right],$ And the identification vector is iteratively calculated by the following formula: $b\left(k\right)=\frac{\mathrm{\&beta;\&psi;}(k-1)[\mathrm{the\; y}\left(k\right)-{\mathrm{\&psi;}}^T(k-1)b(k-1)]}{1+{\mathrm{\&psi;}}^T(k-1)\mathrm{\&beta;\&psi;}(k-1)}+b(k-1),$ Where ψ(k-1)=[y(k-1) T _e (k-1) -1] ^T , and y is the servo motor speed, T _e is the electromagnetic torque, B is the friction coefficient, T _s is the sampling Time, T _l is the load torque, J is the sum of motor and load inertia, β is the recursive variable gain. 8. The method for online identification of inertia ratio according to claim 7, characterized in that: the recursive variable gain $\mathrm{\&beta;}\left(k\right)=\mathrm{\&beta;}(k-1)-\frac{\mathrm{\&beta;}(k-1)\mathrm{\&psi;}(k-1){\mathrm{\&psi;}}^T(k-1)\mathrm{\&beta;}(k-1)}{\mathrm{\&lambda;}+{\mathrm{\&psi;}}^T(k-1)\mathrm{\&beta;}(k-1)\mathrm{\&psi;}(k-1)},$ Where λ is a constant greater than 0. 9. The method for online identification of inertia ratio according to claim 7 or 8, characterized in that: said step (b) comprises the following steps: (b1) converting the rotational speed of the sampled servo motor and the corresponding current into a power series expressed in hexadecimal; (b2) converting each described power series into a custom data structure, the self-defined data structure including character arrays after floating-point number splitting, floating-point number effective bit length, floating-point number power series, floating-point number positive and negative four a defined variable; (b3) Using the self-defined data structure to replace the floating-point number and substituting the corresponding calculation formula to complete the operation. 10. The method for online identification of inertia ratio according to claim 6, characterized in that: said step (c) includes: calculating load torque according to said identification vector; after said step (c) includes: according to said The load torque performs forward friction compensation and reverse friction compensation.