Sliding-mode observer design method based on back electromotive force reconstruction

Abstract

The invention discloses a sliding-mode observer design method based on back electromotive force reconstruction, and relates to the field of motor control. In order to solve the problem in the prior art that the sliding mode control quantity buffeting problem generally exists in the existing sensorless motor control to influence the stability of the system, the technical scheme provided by the invention is as follows: the back electromotive force extraction method comprises the following steps: collecting the observation current of a target motor; obtaining a sliding mode control quantity according to the observation current and a preset sampling current; and obtaining back electromotive force information according to the sliding mode control quantity. And calculating position information and speed of a target motor rotor through a phase-locked loop or an arc tangent function according to the back electromotive force information. The sliding-mode observer design method based on back electromotive force reconstruction is suitable for being applied to sliding-mode observer design work.

Description

Technical Field

[0001] The invention relates to the field of motor control, and specifically to the design of a sliding mode observer based on back-EMF reconstruction. Background Art

[0002] In the field of motor control, position sensorless motor control technology has received extensive attention and research in recent years. In traditional motor control systems, position sensors (such as rotary encoders, Hall sensors, etc.) are usually relied on to obtain the position information and speed signals of the motor rotor. These sensors can directly and accurately provide the necessary data to achieve efficient motor drive and control. However, in some application scenarios, the use of position sensors has many limitations, such as easy damage in harsh environments, high cost, inconvenient installation, or impact on the volume and reliability of the system. Therefore, the development of control methods that do not rely on position sensors has become one of the important development directions of current motor control technology.

[0003] At present, the position sensorless control technology has made great progress, mainly including various methods based on back-EMF estimation, model reference adaptive system (MRAS), sliding mode observer and Kalman filter. Among them, the method based on back-EMF estimation indirectly calculates the position information of the rotor through the back-EMF signal of the motor stator coil. It has good response speed and accuracy, and does not rely on additional sensors. It is one of the more common position sensorless control schemes. The sliding mode observer is a method that uses the principle of system state estimation and introduces a sliding mode control function to make the system state approach the sliding surface, ultimately achieving the purpose of observing the motor rotor state. The sliding mode observer has strong anti-interference ability and robustness, and is particularly suitable for systems with high-frequency disturbances or strong uncertainties.

[0004] Although the above technologies have been applied in position sensorless control, there are still some obvious problems. For example, when the motor is running at a low speed, the back-EMF estimation method is difficult to effectively obtain the motor position and speed due to the weak back-EMF signal and low accuracy; and although the sliding mode observer can provide good robustness, its sliding mode control quantity often has high-frequency jitter, which will affect the stability and accuracy of the system. In addition, the existing sliding mode control strategy lacks sufficient adaptability for complex load conditions or environmental disturbances, and is prone to error accumulation and reduced control performance.

[0005] In practical applications, such as industrial motor drive systems, in order to reduce dependence on mechanical sensors, many researchers have tried to reduce jitter and improve low-speed control performance by improving filtering technology and sliding mode observer design. However, improving the sliding mode control function to reduce jitter usually affects the system's response speed and control robustness, resulting in reduced control accuracy; for filter design, if the parameters are not properly selected, it often leads to the weakening of useful signals. Therefore, how to reduce the jitter phenomenon of sliding mode control while ensuring system robustness and high-precision estimation has always been an important research topic in this field.

[0006] In the prior art, in position sensorless motor control, there is a common problem of sliding mode control variable chattering that affects the stability of the system. Summary of the invention

[0007] In order to solve the problem in the prior art that the chattering problem of the sliding mode control quantity in the existing position sensorless motor control affects the stability of the system, the technical solution provided by the present invention is as follows:

[0008] A back electromotive force extraction method, comprising:

[0009] The step of collecting the observed current of the target motor;

[0010] A step of obtaining a sliding mode control amount according to the observed current and the preset sampling current;

[0011] The step of obtaining back electromotive force information according to the sliding mode control quantity.

[0012] Furthermore, a preferred implementation is provided, in which a current observer is established through a target motor model to obtain an observed current.

[0013] Furthermore, a preferred implementation is provided, in which a sliding mode control variable is obtained according to a difference between the observed current and a preset sampled current.

[0014] Furthermore, a preferred implementation is provided, in which the sliding mode control variable is subjected to low-pass filtering to obtain back electromotive force information.

[0015] Based on the same inventive concept, the present invention also provides a back electromotive force extraction device, comprising:

[0016] A module for collecting the observed current of the target motor;

[0017] A module for obtaining a sliding mode control amount according to the observed current and the preset sampling current;

[0018] A module for obtaining back electromotive force information according to the sliding mode control quantity.

[0019] Based on the same inventive concept, the present invention also provides a sliding mode observer design method based on back-EMF reconstruction, comprising:

[0020] The step of obtaining back electromotive force information of the target motor according to the method described;

[0021] The step of calculating the position information and speed of the target motor rotor through a phase-locked loop or an inverse tangent function according to the back electromotive force information.

[0022] Based on the same inventive concept, the present invention also provides a sliding mode observer design device based on back-EMF reconstruction, comprising:

[0023] A module for obtaining back-EMF information of a target motor according to the method described;

[0024] A module for calculating the position information and speed of the target motor rotor through a phase-locked loop or an inverse tangent function according to the back-electromotive force information.

[0025] Based on the same inventive concept, the present invention also provides a computer storage medium for storing a computer program, and when the computer reads the computer program, the computer executes the described method.

[0026] Based on the same inventive concept, the present invention also provides a computer, including a processor and a storage medium. When the processor reads a computer program stored in the storage medium, the computer executes the method described.

[0027] Based on the same inventive concept, the present invention also provides a computer program product, which is a computer program. When the computer program is executed, the method described above is implemented.

[0028] Compared with the prior art, the technical solution provided by the present invention is beneficial in that:

[0029] The observed current is obtained by establishing a current observer model, which is the core foundation of the entire sliding mode observer. Compared with the existing data collection method that only relies on a single sensor, the current observer built by the motor model in this solution can achieve more accurate current estimation without the need for additional sensors, thus improving the simplicity and stability of the system.

[0030] The calculation of the sliding mode control quantity and its comparison with the sampled current is one of the key steps, which effectively feeds back and controls the error. Compared with the traditional sliding mode control, this scheme achieves the goal of the system state approaching the sliding mode surface more effectively through meticulous difference calculation, and improves the dynamic performance and response speed of the system.

[0031] One of the innovations in this solution is to perform low-pass filtering on the sliding mode control quantity to extract the back EMF. The high-frequency switching signal of the sliding mode control often brings about a more significant jitter phenomenon, and by designing an appropriate low-pass filter, the jitter degree of the system can be effectively reduced. Compared with the method of reducing jitter by using an improved sliding mode control function in the prior art, the application of low-pass filtering significantly improves the system stability while retaining the system robustness.

[0032] By calculating the position and speed of the motor rotor through a phase-locked loop or inverse tangent, this solution can obtain accurate rotor position information without a position sensor. Compared with the existing method of direct measurement using a position sensor, this has a higher anti-interference ability and is particularly suitable for environments where sensors are not suitable for installation. By using back electromotive force for calculation, this solution reduces the dependence on hardware while ensuring calculation accuracy.

[0033] The reconstruction of the back EMF is achieved through an arithmetic mean filter, which further improves the stability of the signal. Compared with the traditional observer that directly uses high-frequency back EMF signals for position estimation, the present invention significantly reduces the impact of jitter on the position and speed estimation of the system by smoothing the amplitude of the back EMF, especially when the motor works under complex load conditions, the stability of the system and the observation accuracy are significantly improved.

[0034] Feeding back the reconstructed back EMF to the sliding mode observer is an important innovation that distinguishes this solution from the existing technology. This step not only reduces the chattering phenomenon in the sliding mode control, but also further improves the system's estimation accuracy of the motor position and speed. Compared with the traditional unidirectional control mode, this feedback method uses the feedback signal reconstructed by the back EMF to enable the observer to more accurately reflect the actual motor state, ensuring high-precision and high-robustness observation performance.

[0035] The present invention provides a sliding mode observer design method based on back-EMF reconstruction, which is suitable for application in sliding mode observer design work. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is the block diagram of the sliding mode observer based on back-EMF reconstruction;

[0037] Figure 2 The simulation waveforms of the actual current and estimated current in the α coordinate system of the motor when the sliding mode observer based on back-EMF reconstruction is adopted;

[0038] Figure 3 When using the sliding mode observer based on back-EMF reconstruction, the back-EMF reconstruction and estimated back-EMF simulation waveforms are shown;

[0039] Figure 4When the sliding mode observer based on back-EMF reconstruction is used, the back-EMF simulation waveform is estimated in the αβ coordinate system;

[0040] Figure 5 The back-EMF simulation waveform diagram for the sliding mode observer without back-EMF reconstruction in the αβ coordinate system;

[0041] Figure 6 The simulation waveforms of the actual rotor position and the estimated rotor position when the sliding mode observer based on back-EMF reconstruction is adopted;

[0042] Figure 7 The simulation waveforms of the rotor position estimation error for the sliding mode observer without back-EMF reconstruction and with back-EMF reconstruction. DETAILED DESCRIPTION

[0043] In order to make the advantages and benefits of the technical solution provided by the present invention more clearly reflected, the technical solution provided by the present invention is now further described in detail with reference to the accompanying drawings, specifically:

[0044] Embodiment 1: This embodiment provides a back electromotive force extraction method, including:

[0045] The step of collecting the observed current of the target motor;

[0046] A step of obtaining a sliding mode control amount according to the observed current and the preset sampling current;

[0047] The step of obtaining back electromotive force information according to the sliding mode control quantity.

[0048] Embodiment 2: This embodiment further limits the back-EMF extraction method provided in Embodiment 1. A current observer is established through a target motor model to obtain an observed current.

[0049] Implementation method three: This implementation method further limits the back electromotive force extraction method provided in implementation method one, and obtains the sliding mode control amount according to the difference between the observed current and the preset sampling current.

[0050] Embodiment 4: This embodiment further limits the back-EMF extraction method provided in Embodiment 1, and performs low-pass filtering on the sliding mode control amount to obtain the back-EMF information.

[0051] Embodiment 5: This embodiment provides a back electromotive force extraction device, including:

[0052] A module for collecting the observed current of the target motor;

[0053] A module for obtaining a sliding mode control amount according to the observed current and the preset sampling current;

[0054] A module for obtaining back electromotive force information according to the sliding mode control quantity.

[0055] Embodiment 6: This embodiment provides a sliding mode observer design method based on back-EMF reconstruction, including:

[0056] A step of obtaining back electromotive force information of a target motor according to the method provided in the first embodiment;

[0057] The step of calculating the position information and speed of the target motor rotor through a phase-locked loop or an inverse tangent function according to the back electromotive force information.

[0058] Specifically,

[0059] This embodiment relates to a sliding mode observer design method for estimating motor position and speed. The technical solution includes the following steps, which are arranged in a logical order:

[0060] Step 1: Build the current observer model

[0061] A current observer is constructed through the motor model to obtain the observed current of the motor.

[0062] Detailed description:

[0063] First, a current observer based on the mathematical model of the motor is established. The mathematical model of the motor can include the current state equation of the motor in the rotating coordinate system, which describes the dynamic behavior of the current. Using this model information, the current observer is designed. The task of the observed current observer is to estimate the current current of the motor in real time based on the input voltage signal and the current state equation of the motor. The output of the observer is the observed current of the motor, which will be used to calculate the sliding mode control quantity in the next step.

[0064] Output: Observed current.

[0065] Step 2: Calculate the sliding mode control quantity

[0066] The difference between the observed current and the sampled current is input into the sliding mode control function to calculate the sliding mode control quantity.

[0067] Detailed description:

[0068] Next, the observed current obtained by the current observer is compared with the sampled current and their difference is calculated. This difference indicates the error between the actual current and the observed current. Then, the error is processed using a sliding mode control function, which is generally based on a sign function or a saturation function to ensure that the system state can quickly approach the sliding surface and reach a stable state. The output of this process is the sliding mode control quantity, which will be used for feedback control of the system and further filtering to improve the stability of the system.

[0069] Output: sliding mode control quantity.

[0070] Step 3: Low-pass filter the sliding mode control quantity

[0071] The sliding mode control variable is low-pass filtered to extract the back EMF.

[0072] Detailed description:

[0073] Since the sliding mode control quantity is a high-frequency switching signal, it usually contains undesirable high-frequency noise components, so it needs to be processed using a low-pass filter. By filtering out these high-frequency components, the motor's back-EMF information can be effectively extracted. The low-pass filter should have an appropriate cutoff frequency to ensure that sufficient back-EMF signals are retained after filtering while eliminating high-frequency jitter noise. The output of the filtering process is a smoothed back-EMF signal, which will be used in the next step of motor position and speed calculation.

[0074] Output: Filtered back EMF.

[0075] Step 4: Calculate the Motor's Rotor Position and Speed

[0076] Based on the filtered back EMF, the position and speed of the motor rotor are calculated.

[0077] Detailed description:

[0078] By using the filtered back EMF signal, the position information and speed of the motor rotor can be calculated. This embodiment uses a phase-locked loop (PLL) or an inverse tangent function to achieve this calculation. A phase-locked loop is a control system used to synchronize the output phase of a motor with the phase of an input signal, thereby obtaining the phase and position of the motor rotor. The inverse tangent function can also be used to estimate the motor angle by obtaining the orthogonal components of the back EMF to calculate the instantaneous angle and speed of the motor. The calculated rotor position and speed will be used in the subsequent back EMF reconstruction step.

[0079] Output: position information and speed of the motor rotor.

[0080] Step 5: Reconstructing the Back EMF

[0081] The extracted back EMF is reconstructed to improve the signal stability.

[0082] Detailed description:

[0083] After acquiring the back EMF signal, the amplitude of the back EMF is reconstructed using an arithmetic mean filter. The arithmetic mean filter is used to average multiple sampling values ​​of the back EMF to reduce signal fluctuations caused by high-frequency noise and jitter. This can obtain a more stable back EMF signal and ensure that the system accurately estimates the actual state when the motor is running. The reconstructed back EMF will be used as a feedback signal to further improve the performance of the sliding mode observer.

[0084] Output: reconstructed back EMF.

[0085] Step 6: Improve the sliding mode observer

[0086] The reconstructed back-EMF feedback is added to the sliding mode observer to improve the system observation performance.

[0087] Detailed description:

[0088] Finally, the reconstructed back EMF is used as a feedback signal and added to the current observer to improve the performance of the sliding mode observer. This feedback can reduce the chattering problem in the sliding mode control, thereby improving the observation accuracy and system stability. Unlike the traditional sliding mode observer that only relies on unidirectional control, this implementation achieves more accurate observation through feedback regulation of the back EMF. Especially under complex load conditions, the system can maintain a good balance between high robustness and high accuracy to ensure the estimation accuracy of position and speed.

[0089] Output: The improved sliding mode observer can achieve high-precision estimation of motor position and speed.

[0090] In order to ensure the smooth implementation of the entire technical solution, the following contents are further improved:

[0091] 1. Establishment of current observer model: The core of current observer lies in correct modeling. Park transformation can be used to transform the three-phase current of the motor into a rotating coordinate system to simplify the control process. In addition, the current observer needs to respond to the input voltage and current signals in real time and continuously update the current estimation according to the state equation of the motor.

[0092] 2. Calculation of sliding mode control quantity: The selection of sliding mode control function is crucial to system performance. Although the sign function can ensure rapid approach to the sliding surface, its high-frequency switching characteristics will cause jitter problems. Therefore, a saturation function or a high-order sliding mode algorithm can be used to reduce jitter. At the same time, the design of the sliding surface must take into account the nonlinear dynamic characteristics of the system to ensure that the error can be effectively converged on the sliding surface.

[0093] 3. Low-pass filter design: The parameter setting of the low-pass filter is very critical, especially the cutoff frequency of the filter. The cutoff frequency of the filter should be slightly higher than the main frequency component of the back EMF in order to effectively retain the useful signal. It can be implemented using an analog filter or a digital filter, and the specific choice depends on the system hardware conditions.

[0094] 4. Phase-locked loop or inverse tangent function calculation: When a phase-locked loop is used for position and speed calculation, it needs to include a phase detector, a low-pass filter, and a voltage-controlled oscillator to ensure accurate locking of the motor phase. The inverse tangent function calculation is suitable for fast and simple implementation and is suitable for application scenarios that require reduced hardware overhead.

[0095] 5. Back EMF reconstruction: When the arithmetic mean filter reconstructs the back EMF, in order to further improve the system's responsiveness, a higher weight can be given to the most recent sampled value, i.e., weighted average filtering. This method can ensure the smoothing effect while making the system more adaptable to rapidly changing loads or working conditions.

[0096] 6. Improved sliding mode observer: The reconstructed back EMF is added to the sliding mode observer to improve the observer. This feedback method can significantly reduce the impact of jitter in the presence of complex loads or external disturbances and enhance the observer's accurate tracking of the actual motor state. In this way, the system can still maintain a high-precision estimation effect even under the dynamic changes of the motor operation.

[0097] Embodiment 7: This embodiment provides a sliding mode observer design device based on back-EMF reconstruction, including:

[0098] A module for obtaining back-EMF information of a target motor according to the method provided in the first embodiment;

[0099] A module for calculating the position information and speed of the target motor rotor through a phase-locked loop or an inverse tangent function according to the back-electromotive force information.

[0100] Embodiment 8: This embodiment provides a computer storage medium for storing a computer program. When the computer reads the computer program, the computer executes the method provided in embodiment 1.

[0101] Embodiment 9: This embodiment provides a computer, including a processor and a storage medium. When the processor reads a computer program stored in the storage medium, the computer executes the method provided in embodiment 1.

[0102] Embodiment 10: This embodiment provides a computer program product, which is a computer program. When the computer program is executed, the method provided in embodiment 1 is implemented.

[0103] Implementation Method XI: Combination Figure 1-7 This embodiment is described in detail by using specific examples to further describe the technical solution provided above. Specifically:

[0104] Combination Figure 1:Including current observer model, sliding mode control function, low-pass filter, arithmetic mean filter, inverse tangent angle calculation and other links. The current observer is built through the motor model to obtain the observed current. The observed current is subtracted from the sampled current, and after passing through the sliding mode control function, the sliding mode control quantity is obtained. When the system reaches the sliding mode surface, according to the equivalent principle of sliding mode control, it can be considered that the control quantity at this time is the back electromotive force of the motor. After low-pass filtering the control quantity, the position and speed of the motor rotor can be calculated through a phase-locked loop or inverse tangent calculation. The amplitude of the obtained back electromotive force is passed through an arithmetic mean filter to achieve reconstruction of the back electromotive force. The reconstructed back electromotive force is added to the aforementioned sliding mode observer to obtain an improved sliding mode observer.

[0105] The specific principles and implementation steps are as follows:

[0106] The motor current state equation is written as follows:

[0107]

[0108] Where i α 、i β is the current of the motor in the αβ coordinate system, e α 、e β is the back EMF of the motor in the αβ coordinate system.

[0109] According to the motor current state equation, the current observer is designed as follows:

[0110]

[0111] In the formula The current in the αβ coordinate system estimated by the sliding mode observer.

[0112] Subtract the current observer equation from the motor current state equation to obtain:

[0113]

[0114] When the state variable of the current observer Arrival at sliding surface After that, according to the equivalent control principle, the control quantity at this time can be regarded as the equivalent control quantity, that is,

[0115]

[0116] Since the sliding mode control quantity is a high-frequency switching signal, a low-pass filter is needed to extract the back electromotive force. The design of the low-pass filter is as follows:

[0117]

[0118] The phase of the low-pass filter is 0 and the amplitude is 1 at ω=ω0, which can effectively extract the value of the back electromotive force. The control quantity processed by the low-pass filter can be used to calculate the position and speed of the motor through the inverse tangent calculation:

[0119]

[0120] In order to further reduce the impact of sliding mode chattering, the back EMF is reconstructed. First, the amplitude of the back EMF is extracted:

[0121]

[0122] Perform arithmetic mean filtering on the back EMF amplitude. Perform arithmetic mean filtering on the N sampled back EMF amplitudes:

[0123]

[0124] In the formula Represents the back EMF amplitude of the nth sampling.

[0125] The back EMF is reconstructed as follows:

[0126]

[0127] Where e RC is the reconstructed back EMF.

[0128] Add the reconstructed back EMF to the sliding mode observer:

[0129]

[0130] Where K RC is the reconstructed back EMF coefficient, and 0 <K RC .

[0131] Figure 2 The actual current and estimated current simulation waveform diagram of the motor in the α coordinate system when the sliding mode observer based on back-EMF reconstruction is adopted in this embodiment;

[0132] Figure 3 When the sliding mode observer based on back-EMF reconstruction is used in the present embodiment, a back-EMF reconstruction and back-EMF simulation waveform diagram is provided;

[0133] Figure 4 When the sliding mode observer based on back-EMF reconstruction is used in this embodiment, the back-EMF simulation waveform diagram estimated in the αβ coordinate system;

[0134] Figure 5 The back-EMF simulation waveform diagram for the sliding mode observer without back-EMF reconstruction in the αβ coordinate system;

[0135] Figure 6The actual rotor position and the estimated rotor position simulation waveform diagram when the sliding mode observer based on back-EMF reconstruction is adopted in this embodiment;

[0136] Figure 7 The simulation waveforms of the rotor position estimation error for the sliding mode observer without back-EMF reconstruction and with back-EMF reconstruction.

[0137] It can be seen from the above simulation waveforms that the sliding mode observer based on back-EMF reconstruction can effectively solve the chattering problem of the sliding mode and has good performance in the accuracy of back-EMF and position estimation.

[0138] The technical solution provided by the present invention is further described in detail above through several specific implementation modes in order to highlight the advantages and benefits of the technical solution provided by the present invention. However, the several specific implementation modes described above are not intended to be used as limitations on the present invention. Any reasonable modification and improvement of the present invention, combination of implementation modes and equivalent substitution within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A back electromotive force extraction method, characterized in that: include: The step of collecting the observed current of the target motor; A step of obtaining a sliding mode control amount according to the observed current and the preset sampling current; The step of obtaining back electromotive force information according to the sliding mode control quantity.

2. A back electromotive force extraction method according to claim 1, characterized in that: A current observer is established through the target motor model to obtain the observed current.

3. A back electromotive force extraction method according to claim 1, characterized in that: A sliding mode control amount is obtained according to the difference between the observed current and the preset sampling current.

4. A back electromotive force extraction method according to claim 1, characterized in that: The sliding mode control amount is subjected to low-pass filtering to obtain back electromotive force information.

5. A back electromotive force extraction device, characterized in that: include: A module for collecting the observed current of the target motor; A module for obtaining a sliding mode control amount according to the observed current and the preset sampling current; A module for obtaining back electromotive force information according to the sliding mode control quantity.

6. A sliding mode observer design method based on back-EMF reconstruction, characterized in that: include: The step of obtaining back EMF information of the target motor according to the method of claim 1; The step of calculating the position information and speed of the target motor rotor through a phase-locked loop or an inverse tangent function according to the back electromotive force information.

7. A sliding mode observer design device based on back-EMF reconstruction, characterized in that: include: A module for obtaining back-EMF information of a target motor according to the method of claim 1; A module for calculating the position information and speed of the target motor rotor through a phase-locked loop or an inverse tangent function according to the back-electromotive force information.

8. A computer storage medium for storing a computing program, characterized in that: When the computer reads the computer program, the computer executes the method of claim 1.

9. A computer, comprising a processor and a storage medium, characterized in that: When the processor reads the computer program stored in the storage medium, the computer executes the method of claim 1 .

10. A computer program product, being a computer program, characterized in that When the computer program is executed, the method of claim 1 is implemented.

Images