KR101576350B1 - Method for detecting fault of realtime e-actuatior using online parameter identification method - Google Patents

Abstract

The present invention relates to an e-actuator of a high efficiency diesel engine turbocharger, and more particularly, to an electronic actuator error detection method of an eVGT (electronic Variable Geometry Turbo-charger) of the present invention, Determining a parameter estimate for the modeling of the e-actuator using a parametric identification technique; setting a threshold value, defined as a maximum value and a minimum value, for each parametric estimate; Judging that it is in a normal state if it is within the threshold value range, and judging it to be in an error state otherwise. According to the present invention, an error of the e-actuator can be detected quickly and easily.

Description

[0001] The present invention relates to a method and apparatus for detecting an e-actuator error in a real-

The present invention relates to an e-actuator of a high efficiency diesel engine turbocharger, and more particularly, to a method of detecting an error of an e-actuator which is an electronic exhaust flow controller.

Recently, interest in high-efficiency diesel engines, which provide high fuel efficiency and performance, is increasing. High-efficiency diesel engine turbochargers can produce the same output with small size, and the turbocharger has attracted much attention in the downsizing engine.

The turbocharger supplies compressed air to the engine using exhaust gas, and generally includes a compressor wheel that supplies compressed air to the engine, a turbine wheel that drives the compressor using engine exhaust gas, A compressor housing, a turbine housing which is an exhaust gas passage, and a bearing for supporting the load.

WGT (Waste Gate Turbo-charger) was used in the turbocharger to prevent compressed air from overcharging the engine due to excessively high compressor pressure. A variable geometry turbocharger (VGT), which improves the torque by increasing the driving force of the turbine at an increased exhaust gas velocity while increasing the exhaust flow rate when the vehicle is at a low speed, Was adopted.

eVGT (electronic VGT) is the latest technology to electronically control the exhaust flow path and it is possible to control precisely by using micro process, and it shows excellent performance in improving engine performance, fuel consumption and exhaust gas reduction.

E-Actuator (electronic Actuator), which is an electronic exhaust flow controller, is a core part of eVGT that controls the exhaust flow path by receiving the command of Engine Engine (Engine Control Unit). It is a digital controller for controlling the position of the wing, And the digital controller is composed of MCU, CAN (Controller Area Network) communication, PWM (Pulse Width Modulation) communication module, motor drive and Hall sensor for DC motor control.

Actuator is composed of many mechanical parts and electric parts, and it has many possibilities of failure. It is difficult to grasp if a malfunction occurs because it is installed in a high temperature engine room. Failure of the e-Actuator will cause a sudden change in output, which can increase the risk of accidents. Therefore, it is necessary to study the real-time error detection technique that can detect the abnormality of e-actuator and prevent accidents.

The real - time error detection technique is a technique that minimizes an accident caused by a failure, and is applied to many fields requiring safety. A signal model based method, a process model based method, and a knowledge based method are widely used as error detection methods.

The signal model based method is applied to the case where only the output signal can be measured in a certain system. There are three methods of the band pass filter, spectral analysis and maximum entropy estimation.

The process model based method is applied when the input and output of a system including an operating device and a sensor can be measured. There are three methods of output observation method, parity equation, identification and parameter estimation.

A knowledge-based method is a method of detecting errors from characteristic values or functions. It detects errors by checking the average value of the system output and the variance of the variance according to the occurrence of errors. The method includes averaging and variance estimation, likelihood- ratio test, and run-sum test.

Since the e-actuator system applies the voltage applied to the actuating device to the internal MCU by using PWM, it can measure the input voltage and can measure the motor angle of the e-actuator using Hall sensor System.

The variable identification technique can estimate the model of the system in a system capable of measuring input and output, and the estimated system model is expressed by the variables constituting the system. In the event of a system failure, the system model will also change due to physical changes, and you can see if an error has occurred. However, in order to detect errors using the variable identification technique, the system needs to be able to observe the input and output, the storage space to store the steady state system model, and accurate system modeling are needed.

Korean Patent Publication No. 10-2000-0025482

SUMMARY OF THE INVENTION The present invention has been made in order to solve the above problems. In the present invention, among the process model-based methods in which errors of a dynamic system are projected on physical parameters (friction, mass, resistance, moment of inertia, etc.) The proposed method is applied to the e-actuator error detection method.

The present invention proposes an on-line modeling technique and an input selection technique that maximizes the observability of variables in order to implement an on-line variable identification technique in an e-actuator environment in which real-time operation is required and resources such as memory are insufficient. .

The objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.

In order to achieve the above object, there is provided a method of detecting an electronic actuator error of an electronic Variable Geometry Turbo-charger (eVGT) according to the present invention, wherein the e-actuator is mathematically modeled and the e-actuator is modeled Determining a parameter estimate using a parameter identification scheme for each parameter estimate, setting a threshold defined by a maximum value and a minimum value for each parameter estimate, and if the parameter estimate is within the threshold range, , Otherwise it is determined to be in an error state.

에 대하여 각도

, 각속도

, 각가속도

의 측정치를 얻는다면,

이고, n≥3의 측정치에 대하여 최소자승법을 적용하면, 매개변수

의 추정치인,

을 구할 수 있다. In determining the parameter estimate, n inputs

The angle

, Angular velocity

, Angular acceleration

Lt; / RTI >

, And applying the least squares method to the measured value of n? 3,

Lt; / RTI >

Can be obtained.

α ₂ is the parameter associated only with the motor and gear, α ₁ is the parameter associated with the gear, motor and mechanical viscous friction, and α ₀ is the parameter associated with the torsion spring and motor.

라고 할 때, 상기 매개변수

를

의 수학식으로 나타낼 수 있다. N is the gear ratio, J _m is the motor inertia moment, K _m is the speed constant, K _t is the torque constant, K _e is the motor back electromotive force constant,

, The parameter

To

. ≪ / RTI >

In setting the threshold, a reference value of each parameter may be determined, and a threshold value may be set as a range in which each parameter estimate must exist.

)과 최소 값(

)으로 정의되고, 상기 매개변수 추정치가

의 범위 안에 있으면 정상으로 판단하고, 그렇지 않으면 오류로 판단할 수 있다.The threshold may be a maximum value for three parameters (

) And the minimum value (

), And the parameter estimate is defined as

It is judged as normal, otherwise it can be judged as an error.

In determining the parameter estimate, the parameter estimate may be determined using a recursive variable identification technique.

According to the present invention, an error of the e-actuator can be detected quickly and easily.

In addition, according to the present invention, there is an effect that the on-line modeling technique and the observability of variables can be maximized in order to implement an on-line variable identification technique in an e-actuator environment in which real-time operation is required and resources such as memory are insufficient.

1 is a view illustrating a driving unit of an e-actuator according to an embodiment of the present invention.
2 is a block diagram illustrating a configuration of a controller of an e-actuator according to an exemplary embodiment of the present invention.
3 is a model of an e-actuator system according to an embodiment of the present invention.
4 is a graph of input voltages selected in consideration of a constraint according to an embodiment of the present invention.
FIG. 5 is a chart comparing a computation amount and a memory usage amount of a batch type and a recursive type according to an embodiment of the present invention.
6 is a diagram illustrating an e-actuator error determination algorithm according to an embodiment of the present invention.
7 is a view showing a broken spur gear used in an experiment according to an embodiment of the present invention.
8 to 10 are graphs showing parameter estimates obtained by mounting broken gears and identifying parameters according to an embodiment of the present invention.
11 is a flowchart illustrating an e-actuator error detection method according to an embodiment of the present invention.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted in an ideal or overly formal sense unless expressly defined in the present application Do not.

In the following description of the present invention with reference to the accompanying drawings, the same components are denoted by the same reference numerals regardless of the reference numerals, and redundant explanations thereof will be omitted. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.

1 is a view illustrating a driving unit of an e-actuator according to an embodiment of the present invention.

Referring to FIG. 1, the driving unit of the e-actuator includes a DC motor 110, a pinion gear 120, spur gears 130 and 140, and an output gear 150.

In an embodiment of the present invention, the DC motor 110 may use a model having a maximum speed of 6,000 RPM of the Johnson motor company and an operating voltage of 14 V. The pinion gear 120 is connected to the shaft of the DC motor 110 and the output gear 150 is connected to the e-actuator valve shaft 160.

In one embodiment of the present invention, two spur gears 130 and 140 are connected between the toothed gear 120 and the output gear 150, and the overall gear ratio is 50.91322.

In an embodiment of the present invention, the torsion spring is connected to the output gear 150 by twisting 295 degrees, and the spring constant is 0.88 [Nmm / deg].

2 is a block diagram illustrating a configuration of a controller of an e-actuator according to an exemplary embodiment of the present invention.

2, the control unit of the e-actuator includes an MCU 210, a hall sensor 220, and a motor driver 230.

The MCU 210 of the e-Actuator receives a reference input command from the main ECU of the vehicle using CAN communication or PWM communication, and controls the DC motor to a desired position through the motor driver 230.

The Hall sensor 220 is located on the axis of the output gear and serves to measure the angle of the e-actuator valve. For example, the hall sensor 220 is connected to an MCU 210 of an e-actuator using an SPI (Serial Peripheral Interface) bus and has a resolution of 0.022 degrees.

11 is a flowchart illustrating an e-actuator error detection method according to an embodiment of the present invention.

Referring to FIG. 11, a method of detecting an electronic actuator error of an electronic Variable Geometry Turbo-charger (eVGT), includes mathematical modeling for an e-actuator, parameter identification technique for modeling an e- To determine a parameter estimate (S101).

Then, a threshold value defined by a maximum value and a minimum value is set for each parameter estimation value (S103).

If the parameter estimation value is within the threshold value range, it is determined that the parameter value is in the normal state. Otherwise, it is determined that the parameter value is in the error state (S105, S107, S109).

3 is a model of an e-actuator system according to an embodiment of the present invention. 3 is a modeling of an e-actuator system including a motor and a load, and is a modeling of an e-actuator including a gear and a spring effect on a linear model of an armature and a rotor of a general DC motor.

Equation 1 shows a DC motor armature model of the e-actuator.

Equation (2) shows a rotor model including a gear and a spring of the e-actuator.

는 각가속도,

는 각속도, θ는 각도를 나타내며, k는 스프링상수, θ_s는 스프링 초기 장착각도, N는 기어 비, K_t는 토크상수, v_a는 전기자에 걸리는 전압, i_a는 전기자에 흐르는 전류를 나타낸다.In Equation 1) and (L 2 is _a DC motor internal inductance, R is _a motor the internal resistance, back EMF constant K _e is the motor, the moment of inertia J _m is the motor, b are mechanical viscous friction, K _m is the rate constant,

Respectively,

Where k is the spring constant, θ _s is the spring initial mounting angle, N is the gear ratio, K _t is the torque constant, v _a is the voltage across the armature, and i _a is the current through the armature .

In Equations (1) and (2), since the in-motor inductance L _a is very small, the e-actuator model of Equation (3) is obtained by disregarding it.

,

,

이다. Here, the parameters? ₂ ,? ₁ ,? ₀ are

,

,

to be.

Although variables such as L _a , R _a , K _e , and K _m are used in the e-actuator system model, since it is very complicated to handle both of these variables, the parameters α ₂ , α ₁ , α ₀ Will be considered.

Therefore, the e-actuator error determination method of the present invention obtains α ₂ , α ₁ , and α ₀ by the parameter identification technique, and derives a change in the characteristics of the actual system from the changes.

Now, an e-actuator parameter identification technique will be described as follows.

에 대하여 각도

, 각속도

, 각가속도

의 측정치를 얻는다면, 수학식 4를 이용하여 매개변수

를 구할 수 있다.Identification of a system is a process used to identify the mathematical model of a system using input and output data from a physical physical system and to estimate various physical characteristics of the system in science and engineering. In the present invention, a method of measuring the output of a known input and identifying a parameter using the measured output is applied. That is,

The angle

, Angular velocity

, Angular acceleration

(4) < / RTI >< RTI ID = 0.0 >

Can be obtained.

If the least squares method is applied to the measurement value of n? 3, an estimate of the parameter can be obtained as shown in the following equation (5).

, 각가속도

를 측정하여야 하지만, e-액츄에이터에서는 각도 θ만 측정할 수 있으므로, 수치 미분으로 각속도

, 각가속도

를 구하여 수학식 5에 적용한다. 이에 따라 오차가 증폭되므로 이에 대한 고려가 필요하다. 이를 포함한 e-액츄에이터 시스템의 제약 조건과 해결 방법을 상세히 설명하기로 한다.
In order to apply Equation (5), the input voltage v, the angle?

, Angular acceleration

However, since only the angle &thetas; can be measured in the e-actuator,

, Angular acceleration

Is applied to Equation (5). Since the error is amplified, consideration should be given to this. The constraints and solution methods of the e-actuator system including this will be described in detail.

The following five constraints must be considered to apply the parameter identification technique in the e-Actuator system.

1) Dead zone

The DC motor does not respond to voltages below a certain level, and the input signal must be set in consideration of this. For example, since the dead zone of the motor of Johnson Motor Co. used in the experiment of the present invention is 0.6V, an input signal larger than 0.6V is used.

2) Calculation of angular velocity and angular acceleration by numerical differentiation

Angular velocity and angular velocity are used as numerical differentiation of angle measurement, so they are affected by noise included in angle measurement. In the present invention, a moving average that averages 10 samples is used to reduce the influence of noise. Since this requires 10 times the sampling time, the sampling period is selected considering this.

3) Ease of calculation of inverse matrix and calculation error

의 역행렬을 계산해야 한다. 역행렬 계산의 정확도는 행렬

의 조건계수(condition number)에 영향을 받는다. 예를 들어 입력 전압이 일정한 경우 각도, 각속도, 각가속도가 모두 일정한 값이며 이 경우

가 특이행렬(singular matrix)이 되어 역행렬을 구할 수 없다. 이를 해결하기 위하여 본 발명에서는 입력의 변화가 크게 나타나는 사인파 입력을 사용한다. 또한 샘플링 주파수와 입력 사인파의 주기에 따라 행렬

의 조건계수가 크게 나타나 역행렬 계산에 오차가 증가할 수 있다. 본 발명에서 샘플링 주파수를 10KHz로 사인파는 4Hz의 주기를 갖도록 하여 이를 해결한다.
When the parameter is estimated using Equation (5)

The inverse of The accuracy of inverse matrix computation depends on the matrix

And is influenced by the condition number of the system. For example, if the input voltage is constant, the angles, angular velocities, and angular velocities are all constant values.

Can be a singular matrix and inverse matrix can not be obtained. In order to solve this problem, the present invention uses a sine wave input in which a change in input is remarkable. Also, depending on the sampling frequency and the period of the input sinusoid,

The error of the inverse matrix calculation can be increased. In the present invention, a sampling frequency is set to 10 KHz and a sine wave is set to have a period of 4 Hz.

4) e-Actuator operating range constraint

The e-Actuator considered in the present invention operates only between 10 ° and 90 ° on the mechanical structure. Therefore, the size of the sine wave input is selected so that the e-Actuator operates between 15 ° and 85 °.

5) Limited memory and processing speed of MCU

The ATmega32M1 MCU considered in the present invention has 8 bits and the internal memory has 2 KB. Most e-Actuators will be composed of MCUs with similar specifications.

, 각가속도

은 각각 데이터 하나당 4Byte의 실수 형 변수를 할당하며, 이 경우 n개의 측정치에 대하여 수학식 5를 구현하기 위하여 필요한 메모리 용량은 16n(4×4×n)Byte 이다.
And, in the implementation of the batch type parameter identification technique, the input voltage v, the angle θ, the angular velocity

, Angular acceleration

(4 × 4 × n) Bytes in order to implement Equation (5) for n measurements, respectively.

4 is a graph of input voltages selected in consideration of a constraint according to an embodiment of the present invention.

로 입력 전압을 선정한다. 즉, 데드존을 회피하기 위하여 입력 전압의 최소 전압이 0.6V 이상 나오도록 바이어스 전압을 1.65V로 설정하며, 15°∼85° 범위를 동작하는 4Hz의 주파수를 가진 사인파 전압을 설정한다.
As shown in FIG. 4, in the present invention,

The input voltage is selected. That is, in order to avoid the dead zone, the bias voltage is set to 1.65V so that the minimum voltage of the input voltage is 0.6V or more, and the sinusoidal voltage having the frequency of 4Hz to operate in the range of 15 ° to 85 ° is set.

In the embodiment of the present invention, since the ATmega32M1 is used as the MCU in the e-Actuator, in order to implement the on-line variable identification technique using the MCU, the computation amount and memory constraints must be considered.

The computation amount and the memory required to implement the e-actuator by the variable identification method of the batch processing method using Equation (5) are calculated as follows.

와

을 이루기 위한 각 성분이 저장될 공간 12×4Byte와

의 역행렬에 행렬식과 각 성분이 저장될 공간 10×4Byte를 포함하여 총 (4n+22)×4Byte의 저장 공간이 필요하다. For n samples, if each sample is stored in 4-byte real numbers, 4n × 4 bytes of memory is required

Wow

12 x 4 Bytes for each component to be stored

(4n + 22) × 4 Bytes in total, including the matrix and the space for storing each component in the inverse matrix of 10 × 4 Bytes.

의 역행렬에 (54n+21)회의 곱셈과 (60n-36)회의 덧셈이 소요되며,

에 대하여 3n회의 곱셈과 (4n-3)회의 덧셈이 소요되며,

^{- 1}와

사이의 행렬 곱에서 9회의 곱셈과 6회의 덧셈이 발생하여 총 (57n+30)의 곱셈과 (64n-33)회의 덧셈이 소요된다. The amount of computation is

(54n + 21) multiplications and (60n-36) additions are required for the inverse matrix of

(3n) multiplications and (4n-3)

^{- 1} and

9 multiplications and 6 additions take place in the matrix multiplication between (57n + 30) and (64n-33) additions.

In the present invention, a recursive variable identification technique is proposed to solve the limitation of memory and computation amount. The proposed recursive variable identification scheme can be implemented using the following equations (6) and (7).

The recursive variable identification technique proposed in the present invention is a structure for adding only the changed parts to new measurement values each time, so that the memory and the calculation amount can be greatly reduced.

와

가 있다면 n번째 측정치 v_an, θ_n,

_n,

_n 에 대한 변화 값을 더해주면 된다. 예를 들면

의 첫 번째 성분은

을 이용하여 계산할 수 있다. 이때

는 (n-1)번째까지 측정치로 계산되어 있는 값이므로 1개의 곱셈과 1개의 덧셈으로 계산할 수 있다. 이는 n개의 곱셈과 n-1개의 덧셈으로 계산하던 이전의 일괄 처리 기법에 비하여 크게 연산량을 줄일 수 있음을 나타낸다. 또한 메모리 사용량도 n×4Byte에서 2×4Byte로 대폭 줄일 수 있음을 알 수 있다.If (n-1) < th >

Wow

If there is an nth measure v _an , θ _n ,

_n ,

_We can add the change value for _n . For example

The first component of

. ≪ / RTI > At this time

Is calculated as the (n-1) -th measurement value, so it can be calculated by one multiplication and one addition. This shows that the computational complexity can be greatly reduced compared to the previous batch processing which is performed by n multiplications and n-1 additions. Also, it can be seen that the memory usage can be drastically reduced from n × 4 bytes to 2 × 4 bytes.

과 수학식 7의 벡터

의 모든 성분에 대하여 필요한 연산량과 메모리 사용량은 다음과 같이 계산할 수 있다. 메모리에서 측정치의 개수는

의 9개,

의 3개이고, 총 12개의 성분에 대하여 현재 값과 직전 값의 저장과, 4개의 현재 측정치 v_an, θ_n,

_n,

_n 의 저장을 위한 28×4Byte와, 역행렬 계산을 위한 저장 공간 10×4Byte의 합인 38×4Byte가 사용된다. 이는 시간과 측정치의 개수에 상관없이 항상 일정한 값으로 (4n+22)×4Byte인 일괄 처리 방법에 비하여 월등한 차이를 나타낸다. 이는 도 5의 도표에서 확인할 수 있다. The matrix of equation (6)

And the vector of equation 7

The required amount of computation and the amount of memory used for all components of the memory can be calculated as follows. The number of measurements in memory is

9,

For the total of 12 components, the storage of the current value and the immediately preceding value, and the four current measurements v _an , &thetas; _n ,

_n ,

28 x 4 Bytes for storing _n and 38 x 4 Bytes, which is a sum of 10 x 4 Bytes for storage of inverse matrix. This is superior to the batch processing method (4n + 22) × 4 bytes, which is always constant regardless of the number of times and the number of measurements. This can be seen in the chart of FIG.

FIG. 5 is a chart comparing a computation amount and a memory usage amount of a batch type and a recursive type according to an embodiment of the present invention.

^-1을 구하기 위한 곱셈 54회, 덧셈 84회와

를 구하기 위하여 곱셈 3회, 덧셈 4회와,

^{- 1}와

사이의 행렬 곱에 곱셈 9회과 덧셈 6회를 수행하기 때문이다. The computation amount of the recursive scheme proposed in the present invention is 66 multiplications and 94 additions regardless of the number of measurements. However,

Multiplication 54 times to obtain ^-1 , addition 84 times

Three times of multiplication, four times of addition,

^{- 1} and

The multiplication is performed 9 times and the addition is performed 6 times.

This is much smaller than the batch processing technique and can be seen in FIG. In particular, since the built-in SRAM of the used MCU is 2KByte, the maximum measurement value is about 500, but it is reduced to less than half, except for the program and data area used in actual implementation.

The recursive variable identification technique proposed in Fig. 5 is more advantageous in terms of memory usage when the number of samples is two or more than that of the conventional batch type variable identification technique, and when the number of samples is more than ten, it is advantageous. Especially, since it has a constant amount of computation and memory usage irrespective of the number of measured values, it can be implemented efficiently using a low-cost MCU.

Now, an e-actuator element abnormality determination method according to the present invention will be described.

Since eVGT is installed in a high temperature vehicle engine room, there is a high probability of failure and it is difficult to grasp if a failure occurs. Therefore, a technique to detect and report failure of eVGT in a short time is needed. In the present invention, among the many causes of failure of the eVGT, a method of detecting an anomaly by considering only a gear, a spring, and a DC motor directly associated with an e_actuator is proposed.

는 e-액츄에이터의 구동부의 DC 모터, 기어 그리고 스프링의 특성이 반영되어 나타난다. 여기서 DC 모터와 관련된 매개변수를

로 두면

는 다음 수학식과 같이 간략히 나타낼 수 있다.The parameter estimates obtained by the variable identification technique described above

Is reflected by the characteristics of the DC motor, gear and spring of the driving part of the e-actuator. Here are the parameters related to the DC motor

If you leave

Can be briefly expressed as the following equation.

In equation (8), α ₂ is a parameter associated only with the motor and gear, α ₁ is the parameter associated with the gear and motor and mechanical viscous friction, and α ₀ is the parameter associated with the torsion spring and motor. In the present invention, a technique for determining an error of an e-actuator from a change in parameter estimates is presented.

When the first e-actuator is manufactured, it is possible to determine the reference value of the parameter through the test and set a threshold value from which the parameter estimates must be present.

)과 최소 값(

)으로 정의하며 이를 EEPROM에 저장해 둔다. The threshold is the maximum value for the three parameters (

) And the minimum value (

) And stores it in EEPROM.

의 범위 안에 있으면 정상상태라 판단하고, 이 범위를 벗어나면 오류로 판단한다. 예를 들어, 수학식(8)에서 만약 모터에 문제가 있을 경우, 모든 시스템 매개변수

가 영향을 받는다. 만약 기어에 문제가 있으며 두 개의 매개변수 α₂,α₁가 영향을 받고, 스프링에 문제가 있으면 오직 α₀의 매개변수만 영향을 받음을 알 수 있다. 이와 같은 방법으로 추정된 매개변수를 이용하여 오류를 판별할 수 있으며, 이를 도 6에 자세히 나타낸다. Then, when the e-Actuator is operating, the system parameters are determined using the variable identification technique, and the obtained parameter estimates

, It is judged to be in a normal state, and if it is out of this range, it is judged as an error. For example, in equation (8), if there is a problem with the motor, all system parameters

Is affected. If there is a problem with the gear and the two parameters α ₂ , α ₁ are affected and there is a problem with the spring, then only the parameters of α ₀ are affected. An error can be identified using the parameters estimated in this way, which is shown in detail in FIG.

6 is a diagram illustrating an e-actuator error determination algorithm according to an embodiment of the present invention.

In order to test the performance of the technique proposed in the present invention, experiments were conducted on gear failure which is easy to reproduce among various errors of the e-actuator.

7 is a view showing a broken spur gear used in an experiment according to an embodiment of the present invention.

As shown in FIG. 7, the teeth of the two spur gears are broken and mounted in the e-actuator, and the validity of the error determination technique using the estimates obtained by the recursive parameter identification technique is confirmed.

8 to 10 are graphs showing parameter estimates obtained by mounting broken gears and identifying parameters according to an embodiment of the present invention.

)를 나타낸 그래프이다.Figs. 8 to 10 are graphs showing the relationship between the parameter estimates obtained by mounting the broken spur gears and parameter identification

FIG.

8 to 10, the blue line represents the parameter estimate for the steady state e-actuator, the red line represents the parameter estimate when only the first spur gear is broken, the green line represents the second parameter And the black line represents the parameter estimate when both spur gears are broken. The parameters were estimated by collecting the first 1,000 samples and then increasing the number of samples by 10 units to a total of 7,000 samples.

FIG. 8 is a graph showing the estimated value of the parameter? ₂ . In Fig. 8, it can be seen that the estimated value of? ₂ shows a large difference with respect to the breakage of the gear. This indicates that the error can be sufficiently detected by using the technique proposed in the present invention.

8 shows a similar result when one gear is broken regardless of the position of the gear, and it can be seen that when two gears are all broken, there is a large difference. As the number of measurements increases, it converges to a constant value and converges after about 3,000 samples. Also, the threshold value can be easily determined using the converged value.

9 is a graph showing an estimate of the parameter? ₁ .

It can be seen that an error can be sufficiently detected even by using the estimated value of the parameter? ₁ in FIG. In this case, there is a remarkable difference when both gears are broken, and it can be seen that even if only one gear is broken, it can be sufficiently distinguished.

10 is a graph showing the estimated value of the parameter? ₀ .

The parameter? ₀ in FIG. 10 is not influenced by the gear, so that the difference between the steady state and the case where the gear is broken can not be confirmed.

In the experiment of the present invention, a total of 7,000 samples were processed at a sampling period of 0.1 msec. However, the shorter the time, the more advantageous it is. In FIG. 8, an estimated value converged to about 3,000 samples can be obtained, which indicates that a gear failure error can be detected within 0.3 seconds. The thresholds for practical application and the number of samples to be used for error detection should be determined through more extensive testing.

While the present invention has been described with reference to several preferred embodiments, these embodiments are illustrative and not restrictive. It will be understood by those skilled in the art that various changes and modifications may be made therein without departing from the spirit of the invention and the scope of the appended claims.

210 MCU
220 Hall sensors
230 Motor Driver

Claims (4)

A method of detecting an electronic actuator error of an eVGT (electronic Variable Geometry Turbo-charger)
Performing mathematical modeling on the e-actuator and determining parameter estimates using parameter identification techniques for modeling the e-actuator;
Setting a threshold defined by a maximum value and a minimum value for each parameter estimate; And
Determining that the parameter estimate is in a normal range if the parameter estimate is within the threshold range, and otherwise determining an error condition;
In determining the parameter estimate,
n inputs

The angle

, Angular velocity

, Angular acceleration

Lt; / RTI >

ego,
If the least squares method is applied to the measurement of n? 3,

Lt; / RTI >

Can be obtained,
Where α ₂ is the parameter associated only with the motor and gear, α ₁ is the parameter associated with the gear, motor and mechanical viscous friction, and α ₀ is the parameter associated with the torsion spring and motor e-actuator error detection method.
delete The method according to claim 1,
N is the gear ratio, J _m is the motor inertia moment, K _m is the speed constant, K _t is the torque constant, K _e is the motor back electromotive force constant,

When you say,
The parameter

Lt;

, ≪ / RTI >
In setting the threshold, a reference value of each parameter is determined, from which a threshold, which is a range in which each parameter estimate must exist,
The threshold may be a maximum value for three parameters (

) And the minimum value (

),
If the parameter estimate is

The controller determines that the e-actuator is in error, otherwise, it determines that the e-actuator is in error.
The method according to claim 1,
In determining the parameter estimate,
Wherein the parameter estimates are determined using a recursive variable identification technique.

Images