Standstill auto-tuning method for motor

Abstract

The present invention relates to a standstill auto-tuning method for a motor, comprising the steps of: a) calculating a prediction error by using a voltage command, current input to the motor, and a computed coefficient vector; b) computing a coefficient vector of a next cycle by using the calculated prediction error; c) computing each of the dead time compensation time and motor parameters by using the coefficient vector acquired in step b); and d) calculating a compensation voltage command by applying the computed dead time compensation time to the voltage command.

Description

Stationary auto-tuning method for electric motors

[0001] The present invention relates to a static auto-tuning method for an electric motor, and more specifically, to an auto-tuning method capable of simultaneously tuning the parameters and dead time compensation time of the electric motor.

[0002] Generally, for high-performance operation of an inverter used to drive a motor, the inverter controller requires precise gain settings. High-performance operation can be understood as minimizing the difference between the speed command value and the current motor speed. To perform precise gain settings, it is required to determine accurate motor parameters.

[0003] In particular, when driving an electric motor using sensorless control without a position sensor, errors in the motor parameters have a significant impact on the motor's driving performance.

[0004] Motor parameters can be estimated using the inverter's output voltage and output current; however, motor drive inverters typically lack output voltage sensors, and even when using voltage values ​​calculated internally by the controller, errors exist due to the inverter's nonlinear characteristics.

[0005] Therefore, tuning motor parameters is significantly affected by voltage distortion characteristics, and conventionally, various tuning methods have been used.

[0006] The applicant of the present invention's registered patent No. 10-2547074 (registered on June 20, 2023, motor control device) describes auto-tuning, and in particular regarding the estimation of the parameters, the method by which the parameter estimation unit estimates the parameters is described as follows: the upper control unit applies voltage in various forms to a motor that is stopped, and the parameter estimation unit estimates the parameters based on various variables that are responses to this.

[0007] This method is significant in that it facilitates the estimation of parameters compared to previous technologies and allows for the setting of more accurate gains by applying this.

[0008] However, for high-performance operation, more accurate calculation of parameters is required, and an auto-tuning method capable of eliminating various sources of error, such as nonlinearity, is needed.

[0009] In addition, conventionally, to calculate the dead time compensation time, motor parameters were estimated and the dead time compensation time was calculated using the estimated parameters; conversely, to calculate the motor parameters, a method was used in which the dead time compensation time was estimated and the motor parameters were calculated using the estimated dead time compensation time.

[0010] This is because the calculation formulas for dead-time compensation and motor parameters are combined, and they could not be separated independently.

[0011] This conventional method had a problem in that it was impossible to calculate accurate dead-time compensation time or motor parameters because it used estimated values.

[0012] The problem that the present invention aims to solve, taking into account the market demands mentioned above, is to provide a static auto-tuning method for an electric motor capable of accurately calculating the motor parameters.

[0013] In addition, another problem that the present invention aims to solve is to provide a static auto-tuning method for a motor that can simultaneously tune the dead-time compensation time of an inverter and the motor parameters.

[0014] More specifically, the present invention aims to provide a static auto-tuning method for an electric motor that compensates for output voltage distortion by accurately tuning the dead-time compensation time and uses this to accurately tune the motor parameters.

[0015] In addition, another objective of the present invention is to provide a static auto-tuning method for an electric motor applicable even to inverters operating with a low switching frequency by implementing accurate digital modeling of the electric motor.

[0016] The static auto-tuning method of the present invention may include a) a step of calculating a prior error using a voltage command, a current input to the motor, and a calculated coefficient vector; b) a step of calculating a coefficient vector for the next cycle using the calculated prior error; c) a step of calculating a dead time compensation time and a motor parameter, respectively, using the coefficient vector obtained in step b); and d) a step of applying the calculated dead time compensation time to the voltage command to calculate a compensation voltage command.

[0017] In an embodiment of the present invention, the motor parameters may be stator resistance and stator inductance.

[0018] In an embodiment of the present invention, step a) can calculate a prior error using the following mathematical formula 1.

[0019] [Mathematical Formula 1]

[0020]

[0021] e pri is the prior error, e pos is the posterior error, v s* ds is the d-axis voltage command in the stationary coordinate system, i s ds is the d-axis current in the stationary coordinate system, and n is the order of the period, is the computed coefficient vector.

[0022] In an embodiment of the present invention, step b) can calculate the coefficient vector using the following mathematical formula 2.

[0023] [Mathematical Formula 2]

[0024]

[0025] is the computed coefficient vector, e pri is the prior error, e pos is the posterior error. k is the gain, "subject to~" means to perform the operation to satisfy ~, n is the order of the period, C and α are constants, = { , , }.

[0026] In an embodiment of the present invention, step c) can calculate the dead time compensation time using the following mathematical formula 3.

[0027] [Mathematical Formula 3]

[0028]

[0029] is the calculated dead time compensation time, is the computed coefficient vector, K pd is the proportional gain for the dead-time compensation time output, K id is the integral gain for the dead-time compensated time output, s is the differentiator (d / dt).

[0030] The present invention has the effect of shortening the tuning time of the inverter by simultaneously tuning the dead time compensation time of the inverter and the motor parameters, and also has the effect of enabling more accurate gain calculation and thus enabling high-performance operation of the inverter by ensuring that there is no influence of the dead time compensation time error on the calculation of the motor parameters.

[0031] In addition, the present invention can be implemented in a state where the motor is not rotating, and by implementing accurate digital modeling, it has the effect of being applicable to inverters operating at low switching frequencies.

[0032] Figure 1 is a diagram showing the configuration of a motor driving circuit to which the present invention is applied.

[0033] FIG. 2 is an example diagram of the configuration of a controller for implementing the present invention.

[0034] Figures 3 to 5 are graphs of the tuning results of the present invention.

[0035] The national research and development projects that supported this invention are as follows.

[0036] [Project ID] 2410004019

[0037] [Assignment No.] 20019455

[0038] [Ministry Name] Ministry of Trade, Industry and Energy

[0039] [Name of Project Management (Specialized) Agency] Korea Institute of Industrial Technology Evaluation and Management

[0040] [Research Project Name] Development of Materials and Components Technology

[0041] [Research Project Title] Development of 2MW Permanent Magnet Synchronous Motor and Drive Unit for Ship Propulsion

[0042] [Name of Project Performing Organization] LS Electric Co., Ltd.

[0043] [Research Period] April 1, 2022 ~ December 31, 2025

[0044] To fully understand the structure and effects of the present invention, preferred embodiments of the present invention are described with reference to the attached drawings. However, the present invention is not limited to the embodiments disclosed below, but can be implemented in various forms and various modifications can be made. The description of the embodiments is provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention. In the attached drawings, components are depicted enlarged from their actual size for convenience of explanation, and the proportions of each component may be exaggerated or reduced.

[0045] Terms such as 'first' and 'second' may be used to describe various components, but said components should not be limited by said terms. These terms may be used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, 'first component' may be named 'second component,' and similarly, 'second component' may be named 'first component.' Furthermore, singular expressions include plural expressions unless the context clearly indicates otherwise. Unless otherwise defined, terms used in the embodiments of the present invention may be interpreted in the sense commonly known to those skilled in the art.

[0046] Hereinafter, a static auto-tuning method for an electric motor according to one embodiment of the present invention will be described in detail with reference to the drawings.

[0047] FIG. 1 is an exemplary diagram of a motor driving circuit to which the present invention is applied, and FIG. 2 is an exemplary diagram of the configuration of the controller in FIG. 1.

[0048] Referring to FIG. 1 and FIG. 2 respectively, the present invention outputs a gate signal of a switching element of an inverter (200) using a voltage command and a detected output current of an inverter (200), and when the motor (300) is in a stationary state, the dead time compensation time of the switching element and the stator resistance and stator inductance of the motor (300) are calculated, and an accurate dead time compensation time and an accurate gain setting are performed in the controller (100).

[0049] The controller (100) includes a dead time compensation unit (110), a pulse width modulation unit (120), a prior error calculation unit (130), a coefficient vector calculation unit (140), a dead time compensation time calculation unit (150), and a motor parameter calculation unit (160) to perform auto-tuning.

[0050] The controller (100) may be a processor or a control circuit including a processor.

[0051] The above dead time compensation unit (110), pulse width modulation unit (120), prior error calculation unit (130), count vector calculation unit (140), dead time compensation time calculation unit (150), and motor parameter calculation unit (160) may each be a functional module of auto-tuning software performed in a controller (100), and may be a functional module of hardware that performs the software as needed.

[0052] In the following description, the configuration and operation of the present invention will be specifically explained from the perspective of auto-tuning software performed in the controller (100).

[0053] The inverter (200) may be composed of a rectifier (210), a DC link capacitor (220), and an inverter unit (230) including a plurality of switching elements (231).

[0054] FIG. 2 shows a sequence for simultaneously tuning the dead time compensation time and motor parameters proposed in the present invention. The required inputs are voltage commands and current, and the final outputs are the dead time compensation time and motor parameters.

[0055] First, the dead time compensation unit (110) of the controller (100) outputs a compensation voltage command using the voltage command and the dead time compensation time information of the dead time compensation time calculation unit (150).

[0056] The above compensation voltage command is a command value that ensures the voltage command matches the output voltage of the inverter (200).

[0057] The present invention can be applied to both systems where voltage commands are generated arbitrarily and systems where they are generated automatically through current control.

[0058] Deadtime refers to the time during which one phase arm remains in an open state when the inverter unit (230) is driven. The switching element (231) may use an IGBT (Insulated Gate Bipolar Transistor).

[0059] In order to calculate the dead time compensation time of the dead time compensation time calculation unit (150), calculation of the coefficient vector of the coefficient vector calculation unit (140) is required, and in the present invention, in order to converge the prior error to zero, prior error information is calculated first through the prior error calculation unit (13) before the coefficient vector calculation.

[0060] The prior error calculation unit (130) takes the voltage command, current, and coefficient vector as inputs and calculates the prior error.

[0061] The vector control system of the electric motor uses the feedback-type current control system used in the present invention, and can be understood as being intended to account for modeling errors that may occur or differ from the mathematical model of the actual control target. This may include errors due to the linearity of the inverter (200).

[0062] The prior error calculated in the prior error calculation unit (130) above is provided to the coefficient vector calculation unit (140) along with the voltage command and current.

[0063] The coefficient vector of the coefficient vector calculation unit (140) is a vector to which the calculated prior error is applied, and this is commonly applied to the calculation of the dead time compensation time described above and the calculation of the motor parameter of the motor parameter calculation unit (160).

[0064] According to the compensation voltage command of the dead time compensation unit (110), the pulse width modulation unit (120) outputs the gate voltage of the switching elements (231) of the inverter unit (230) described above.

[0065] That is, the present invention can improve the control performance of the motor (300) by calculating the motor parameters and dead time compensation time as accurate values.

[0066] The features of the present invention are explained with more specific examples as follows.

[0067] Mathematical formula 1 below is a mathematical formula to explain the nonlinearity of the inverter (200).

[0068]

[0069] v in mathematical formula 1 an , v bn , v cn is the average value of the pole voltage during the switching cycle of each phase, and v* an , v* bn , v* cn is the input (polar) voltage command.

[0070] V dc is the voltage of the DC link capacitor (220), and T d is the dead time setting time, T on The turn-on time of the switching element (331), T off represents the turn-off time of the switching element (331), and T cmp is the dead time compensation time calculated by the dead time compensation time calculation unit (150). T s is the control period, V ce ε is the collector-emitter voltage drop of the switching device, V ak represents the anode-cathode voltage drop.

[0071] sgn(i as ), sgn(i bs ), sgn(i cs ) indicates the direction of current for each phase, and becomes a value of +1 or -1 depending on whether it is forward or reverse.

[0072] Most preferably, the dead time compensation time (T) is such that the average of the pole voltages matches the voltage command. cmp It is desirable for ) to be determined, but V ce , V ak Since variables such as these are uncertain values, it is difficult to accurately set the precise dead time compensation time.

[0073] Therefore, in the present invention, the result of converting the above mathematical formula 1 into a stationary coordinate system d-axis voltage equation considering the voltage distortion of a synchronous motor in a stationary state is as shown in the following mathematical formula 2.

[0074]

[0075] v s* ds is the d-axis voltage command in the stationary coordinate system, i s ds is the d-axis current in the stationary coordinate system, and R s ε₀ represents the stator resistance, and Ls represents the stator inductance. The phase a current is positive.

[0076] If an impulse-invariant Z-transform is performed on Equation 2 for low-frequency digital control, Equation 3 below can be obtained.

[0077]

[0078] In Equation 3, k is the order of the sampling period, and the k+1th current defines the relationship between the kth sampling current and the k-1th sampling voltage. w is the coefficient vector.

[0079] w is a column vector with elements w1, w2, and w3, expressed as w={w1, w2, w3}.

[0080] In mathematical formula 3, w is expressed as in mathematical formula 4.

[0081]

[0082] In addition, Equation 5 below is a calculated coefficient vector that meets the conditions ( Represents the calculation formula of ). ^ means the calculated value.

[0083] This is performed in the coefficient vector operation unit (140).

[0084] As previously mentioned, the coefficient vector calculation unit (140) takes the voltage command, current, and prior error as inputs and calculates the coefficient vector using them.

[0085]

[0086] In mathematical equation 5, e pri is the prior error, e pos is the posterior error. k is the gain vector.

[0087] k is displayed in bold font and is distinguished from k, which represents the sampling order explained earlier.

[0088] n represents a sequential operation in the controller (100) and is a value expressed as a positive integer similar to k, but differs in meaning.

[0089] In mathematical equation 5, "subject to~" means to perform the operation to satisfy ~. C and α are assumed to be constants.

[0090] Calculated coefficient vector obtained in a specific period ( ) is also provided to the prior error calculation unit (130), and the prior error calculation unit (130) calculates the prior error (e) according to the following mathematical formula 6. pri ) can be obtained.

[0091]

[0092] Additionally, the calculated coefficient vector is provided to the dead time compensation time calculation unit (150) and the motor parameter calculation unit (160).

[0093] The motor parameter calculation unit (160) calculates the coefficient vector ( The stator resistance and inductance of the motor (300) are calculated as parameters from ). ^ represents the calculated value as previously explained.

[0094] The calculation formula of the motor value calculation unit (160) can be expressed as the following mathematical formula 7.

[0095]

[0096] In mathematical formula 7, w1 and w2, marked with ^ as the result of the operation, are each considered to be values ​​that make up w.

[0097] In other words, resistance and inductance can be accurately calculated using the calculated values ​​of w1 and w2.

[0098] As such, the present invention can accurately estimate motor parameters using a calculation method, thereby enabling the motor parameters to be obtained without using other higher-level controllers, and based on this, inverter tuning can be performed to enable more accurate drive control.

[0099] β is an additional operation expression, and its definition means that it is operated to satisfy a finite constant J.

[0100] In addition, the dead time compensation time calculation unit (150) uses the calculated coefficient vector to calculate the dead time compensation time (T cmp ) can be produced.

[0101] The formula for calculating the dead time compensation time is shown in Equation 8.

[0102]

[0103] Calculated Dead Time Compensation Time ( ) is provided to the dead time compensation unit (110), which is provided to the pulse width modulation unit (120) as a voltage command compensated by applying the previously described mathematical formula 2, and controls the switching element of the inverter unit (230) with the gate signal, which is the pulse width modulation signal of the pulse width modulation unit (120).

[0104] Mathematical Equation 8 is proportional-integral control, The purpose is to control it to 0.

[0105] K pd is the proportional gain for the dead-time compensation time output, K id is the integral gain for the dead-time compensated time output, and s is the differentiator (d / dt).

[0106] To verify the present invention with such a configuration, the stator resistance (R s ) and inductance (L sThe setting value of ) was set to have a 50% error, the initial dead time compensation time was initially set to 0μs, the switching frequency of the inverter was set to 1kHz, and the operating frequency was set to 0Hz, so that auto-tuning was performed while the motor (300) was in a stationary state.

[0107] FIG. 3 is a graph comparing the stator resistance obtained as an auto-tuning result of the present invention with the actual stator resistance, FIG. 4 is a graph comparing the inductance value obtained as a tuning result with the actual stator inductance, and FIG. 5 is a graph of the dead-time compensation time.

[0108] Referring to FIGS. 3 to 5 respectively, it can be seen that the present invention is performed starting from 6 seconds, and the stator resistance, stator inductance, and dead time compensation time are all tuned simultaneously.

[0109] In other words, it can be confirmed that the motor parameters converge to an accurate value even at a low switching frequency and without an initial dead-time compensation time being set.

[0110] By using the present invention, the stator resistance, stator inductance, and dead-time compensation time required for driving a high-performance motor can all be tuned simultaneously.

[0111] Through this, motor driving performance can be improved, and it can be particularly effective for sensorless control that is significantly affected by motor parameters.

[0112] The present invention can be implemented in a stationary state without rotating the motor, can be applied to an inverter operating at a low switching frequency, and has the characteristic of not requiring additional hardware devices.

[0113] Although embodiments according to the present invention have been described above, they are merely illustrative and those skilled in the art will understand that various modifications and equivalent embodiments are possible therefrom. Accordingly, the true technical scope of protection of the present invention should be determined by the following claims.

Claims

1. As an auto-tuning method performed in a controller, a) A step of calculating a prior error using a voltage command, a current input to the motor, and a calculated coefficient vector; b) A step of calculating the coefficient vector of the next period using the calculated prior error; c) a step of calculating the dead time compensation time and the motor parameters, respectively, using the coefficient vector obtained in step b) above; and d) A static auto-tuning method for an electric motor comprising the step of applying a calculated dead-time compensation time to a voltage command to calculate a compensation voltage command.

2. In Paragraph 1, The above motor parameters are, A static auto-tuning method for an electric motor characterized by stator resistance and stator inductance.

3. In Paragraph 2, Step a) above is, A static auto-tuning method for an electric motor characterized by calculating a prior error using the following mathematical formula 1. [Mathematical Formula 1] e pri is the prior error, e pos is the posterior error, v s* ds is the d-axis voltage command in the stationary coordinate system, i s ds is the d-axis current in the stationary coordinate system, and n is the order of the period, is the calculated coefficient vector 4. In Paragraph 2, The above step b) is, A static auto-tuning method for an electric motor characterized by calculating a coefficient vector using the following mathematical formula 2. [Mathematical Formula 2] is the computed coefficient vector, e pri is the prior error, e pos is the posterior error. k is the gain, "subject to~" means to perform the operation to satisfy ~, n is the order of the period, C and α are constants, = { , , } 5. In Paragraph 2, The above step c) is, A static auto-tuning method for an electric motor characterized by calculating the dead time compensation time using the following mathematical formula 3. [Mathematical Formula 3] is the calculated dead time compensation time, is the computed coefficient vector, K pd is the proportional gain for the dead-time compensation time output, K id is the integral gain for dead-time compensated time output, s is the differentiator (d / dt)

Images