JPH11194073A - Method for identifying inertia moment - Google Patents

Abstract

PROBLEM TO BE SOLVED: To compensate an error by means of a viscosity item. SOLUTION: When a dynamo torque command TLC is inputted to one inertia model process 11, dynamo speed ND is obtained in output. Dynamo speed ND obtained at the output of one intertia model process 11 us inputted to a multiplication integration process 12. When the sine wave sin(ωn t) of a similar phase against dynamo speed ND is multiplied by a cosine wave cos(ωn t) shifted by am 90 deg. (π/2) in phase and they arc integrated in the process 12, the binary DC of an inertia item and the viscosity item are obtained in a DC generation process 13. Obtained binary DC are compared in a binary DC comparison process 14 and are compared with the error by the viscosity item in the result of an inertia identification method and the error by the viscosity item is compensated in a compensation Process 15.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an inertia moment identification method for measuring engine inertia as a pre-operation when controlling an engine in an engine test apparatus.

[0002]

To describe Conventional engine inertia identification method, two-inertia serves as an inertial system in sufficiently lower frequency range than the anti-resonance frequency omega _1. Figure 4 is one inertial system block diagram, the engine torque T _E is always zero in during identification. As shown in FIG. 5, the identification of engine moment of inertia, enter a low-frequency sine wave sin (omega _n t) to a dynamo torque command T _LC, difference dynamo speed (actual speed) as one inertia model speed is zero The model inertia is adjusted by the identification mechanism so that

In FIG. 5, reference numeral 1 denotes a one-inertia model part.
Low-frequency sine wave sin (ω _n t) is supplied to the model unit 1. The low frequency sine wave is also supplied to the adjustable model inertia 2. Output one inertia model unit 1 and the model inertia unit 2 is inputted to the deviation portion 3 taken deviation, the deviation signal is multiplied by the sine wave and the 90 ° phase shifted signal cos (ω _n t) Then, it is input to the DC component generator 4. The DC component of the DC component generator 4 is supplied to the identification mechanism 5. The identification mechanism unit 5 adjusts the model inertia unit 2 according to the DC component and identifies the engine inertia moment.

Next, the operation of the method for identifying the moment of inertia of the engine configured as shown in FIG. 5 will be described. The speed error between the dynamo speed and the model speed in the one inertia system model unit 1 and the model inertia unit 2 is (1) due to the difference between the two inertia moments.
It becomes as shown in the formula.

[0005]

(Equation 1)

[0006] Expression (1) is represented by expression (2) when expressed by a frequency transfer function.

[0007]

(Equation 2)

[0008] Here, as the identification signal, dynamo torque command _{T LC = sin (ω n t} ) (ω n << ω 1, ω n: frequency components of the identification signal, omega _1: anti-resonance frequency) of an inertial system Model part 1
Is input to the model inertia unit 2, the error is given by the following equation (3)
become that way.

[0009]

(Equation 3)

[0010] By multiplying the equation (3) of 90 ° phase with respect to the identification signal shifted signal cos (omega _n t), it is integrated by the difference between the inertia term and viscosity term, the e (t) A DC component m (t) of the speed difference due to the inertia moment difference shown in the following equation (4) is generated.

[0011]

(Equation 4)

When the obtained DC component m (t) is zero, the engine moment of inertia is identified as a true value because the model and the true moment of inertia are equal and the dynamo inertia is also known.

[0013]

According to the method for identifying the moment of inertia constituted as described above, when the moment of inertia of the model is identified by the DC component m (t), it is necessary to adjust every cycle T (second). When the convergence is reduced and the DC component m (t) becomes zero, if the viscous term (mechanical loss) is included, the model inertia Jm identified by the viscous term is expressed by the following equation (5). And the identification result has a problem that an error is caused by the viscosity term of the second term on the right side of the equation (5).

[0014]

(Equation 5)

The present invention has been made in view of the above circumstances, and it is an object of the present invention to provide an inertia moment identification method capable of compensating for an error due to a viscosity term.

[0016]

According to the present invention, in order to achieve the above-mentioned object, a dynamo torque command is input to a one inertia model process to obtain a dynamo speed as an output, and the dynamo speed is in phase with the dynamo speed. After multiplying and integrating a sine wave and a cosine wave whose phase is shifted by 90 °, a binary DC component of an inertia term and a viscosity term is obtained, and the two DC components are compared to obtain a ratio. After comparing with the error due to the term, the error due to the viscosity term is compensated.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the accompanying drawings. In order to compensate for the error due to the above-mentioned viscosity term, a sine wave T _LC = si
When n (ω _n t) is inserted, ω is calculated from the detected dynamo speed N _D
DC components based on _n · J (ω _n : detected value of motor rotation speed,: true moment of inertia) and D (dynamo moment of inertia) are extracted by the same principle of inertia identification.

In FIG. 1, first, one inertia model process 1
1 can be approximated as in equation (6) from the relationship between the dynamo torque command and the dynamo speed.

[0019]

(Equation 6)

When the equation (6) is expressed by a frequency transfer function, the following equation (7) is obtained.

[0021]

(Equation 7)

[0022] Here, entering a dynamo torque command T _LC to an inertial model step 11, the output, the dynamo speed N _D as shown in the following (8) is obtained.

[0023]

(Equation 8)

The obtained dynamo speed N _D is input to the output of the one inertia model process 11 to the multiplication and integration process 12. In this process 12, the sine wave sin having the same phase as the dynamo speed N _{D is} obtained.
(ω _n t) and the cosine wave cos (ω
_n t) are multiplied and integrated, and the following equations (9) and (1) are obtained.
0), a binary DC component of an inertia term and a viscosity term is obtained in a DC component generation step 13.

[0025]

(Equation 9)

[0026]

(Equation 10)

When the binary DC components obtained by the above equations (9) and (10) are compared in the binary DC component comparison step 14, the following equation (11) is obtained.

[0028]

[Equation 11]

Here, when the error due to the viscosity term in the result of the inertial identification method described above is compared with the above equation (11), the following equation (12) is obtained.

[0030]

(Equation 12)

When the model inertia (5) identified using the above relation is rewritten, the following equation (13) is obtained.

[0032]

(Equation 13)

Therefore, the compensation step 1 for the error due to the viscosity term
By correcting the identified model inertia by the following equation (14), the true moment of inertia can be identified.

[0034]

[Equation 14]

The validity of the identification method using the viscous term compensation method according to the above equation (14) was verified by simulation to confirm the validity. In the real-inertia model shown in Fig. 2, the moment of inertia (unit moment of inertia: 1.89) and the viscosity term (2.
The identification simulation was performed as 0). The result is shown in FIG. It is clear from FIG. 3 that when the viscous term is included, even if the inertial model is set to be shifted by 1.5 times, it is identified as a true value.

[0036]

As described above, according to the present invention,
By inputting a sine wave to the dynamo torque command, detecting the DC component of the inertia term and the viscous term using the detected dynamo speed, and compensating for the error component from these, even when the viscous term is included, The inertial model has an advantage that a true value can be identified.

[Brief description of the drawings]

FIG. 1 is a process chart illustrating an embodiment of the present invention.

FIG. 2 is a block diagram showing a real one inertia model.

FIG. 3 is a characteristic diagram showing an identification simulation result using the embodiment.

FIG. 4 is a block diagram showing one inertial system.

FIG. 5 is a block diagram illustrating a method for identifying an engine inertia moment.

[Explanation of symbols]

 11: One inertia model process 12: Multiplication and integration process 13: DC component generation process 14: Binary DC component comparison process 15: Compensation process

Claims (1)

[Claims] An inertia model and an adjustable model inertia unit are provided. A deviation output signal between the model unit and the inertia unit is multiplied by a sine wave and a cosine wave whose phase is shifted by 90 ° to be integrated. When the DC component due to the term difference is obtained and the DC component is supplied to the identification mechanism to identify the model inertia part, in the method of identifying the moment of inertia in which an error component due to the viscous term occurs, After inputting and obtaining a dynamo speed at the output, the dynamo speed is multiplied by a sine wave having the same phase and a cosine wave whose phase is shifted by 90 °, integrated, and then a binary DC component of an inertia term and a viscous term is obtained. Comparing the binary DC component to obtain a ratio, comparing the ratio with the error due to the viscosity term, and then compensating for the error due to the viscosity term, wherein the moment of inertia is characterized. Method.