An embedded magnetic encoder decoding method and system based on a two-stage closed-loop ESO-PLL composite architecture
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-05-15
- Publication Date
- 2026-08-11
AI Technical Summary
[0003]然而，现有的基于同步旋转坐标系锁相环（SRF-PLL）的解码系统在低速运行时，内嵌式磁编码器信号中的低次谐波（如负一次、三次、五次）频率极低，与基波接近
[0036]本发明达到的有益效果：本发明的方法利用双级闭环 ESO-PLL 复合架构，实现了滤波与动态带宽的系统级解耦，与现有技术中单级锁相环不同，通过第一级SRF-PLL提取初始基准，利用ESO将低频谐波定义为集总扰动进行实时观测与剥离，解决了低速工况下谐波频率低导致的滤波延迟问题，使得基于锁相环的角度解码系统在深度抑制扰动的同时，维持了原有的高控制带宽与动态响应速度。本发明并非将ESO作为简单的开环前馈补偿，而是将第二级SRF-PLL解算出的高精度位置角作为观测反馈量，重新引入ESO的输入端形成全局闭环，这种自适应反馈机制能够自动克服传感器安装误差、增益不一致导致的信号直流偏置与不平衡，在恶劣的非理想工况下依然保证位置信号的极高纯净度。
Smart Images
Figure CN122553789A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power electronics and motor control technology, specifically to an embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture. Background Technology
[0002] High-performance control of permanent magnet synchronous motors relies on accurate rotor position detection. To achieve low-cost, high-power-density design, embedded magnetic encoders have become a preferred solution. Embedded magnetic encoders use linear Hall sensors, which are embedded inside the motor to directly sense the magnetic field of the permanent magnet.
[0003] However, in existing decoding systems based on Synchronous Rotating Coordinate System Phase-Locked Loops (SRF-PLL), the low-order harmonics (such as -1, -3, and -5) in the embedded magnetic encoder signal have extremely low frequencies, close to the fundamental frequency, when operating at low speeds. In existing single-stage SRF-PLL systems, there is a strong coupling between filtering capability and dynamic tracking bandwidth: to completely filter out low-frequency harmonics, the loop bandwidth must be drastically reduced, which will cause severe phase lag during motor start-up, shutdown, or speed changes, resulting in a rapid deterioration of dynamic performance; conversely, if a high bandwidth is maintained, harmonic disturbances directly penetrate the loop, leading to distortion in steady-state position calculation.
[0004] Existing technologies are mostly limited to open-loop compensation for specific errors, failing to resolve the coupling contradiction between bandwidth and filtering from the decoding architecture itself. It is necessary to break through the traditional single-stage decoding method and decouple the fundamental frequency extraction and disturbance suppression of the position signal at the system level, achieving deep suppression of complex low-frequency harmonics without sacrificing dynamic control bandwidth. Summary of the Invention
[0005] The technical problem to be solved by this invention is that the three-phase linear Hall signal built into the motor has obvious harmonic disturbances at low speeds, making it difficult for traditional phase-locked loops to balance filtering effect and dynamic response.
[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:
[0007] An embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture includes:
[0008] S1, the embedded magnetic encoder uses a linear Hall sensor to process the raw signal acquired by the three-phase linear Hall sensor. Perform Clarke transform to obtain two-phase orthogonal linear Hall signals. The orthogonal linear Hall signal The first-level SRF-PLL is used for preliminary calculation to obtain the estimated angle of the motor rotor position. ;
[0009] S2. The estimated angle of the motor rotor position extracted by the first-stage PLL. The input vector of the extended state observer ESO is constructed as the baseline input. The perturbation state observation is decoupled in real time through the discretized model of the extended state observer (ESO). Then the rotor position angle calculated by the second-stage PLL is... As the observation feedback vector of the extended state observer (ESO), based on the perturbation state observation... Reconstruct the fundamental state observations after the perturbation was removed ;
[0010] S3. Fundamental state observations reconstructed from the extended state observer ESO Input the second-level SRF-PLL to calculate the final motor rotor position angle. .
[0011] The aforementioned embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture, in step S1, converts the original magnetic flux density signal acquired by the three-phase linear Hall sensor into a digital encoder. The transformation from a three-phase stationary coordinate system to a two-phase stationary α-β axis is expressed as:
[0012]
[0013]
[0014]
[0015] in It is the Clarke transformation matrix. The amplitude of the input signal. The original magnetic flux density signal matrix, It is an orthogonal linear Hall signal matrix.
[0016] The aforementioned embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture, in step S1, transforms the α-β axis signals to a dq coordinate system that rotates synchronously with the estimated rotor position through Park transformation, thereby obtaining two-phase Hall signals in the synchronous coordinate system. :
[0017]
[0018] in, To input the true phase information of the Hall signal, To estimate phase information, This is the phase estimation error;
[0019] The phase estimation The error is processed by a proportional-integral loop filter, and then the first-stage PLL estimation angle of the motor rotor position is obtained through an integrator. The closed-loop transfer function is obtained. , represented as:
[0020]
[0021]
[0022] in, For the first-level PLL to estimate the angle, The estimated rotational speed of the first-stage PLL. It is a proportionality coefficient. It is the integral coefficient. The damping coefficient is... For bandwidth, It is a complex frequency.
[0023] The aforementioned embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture includes, in step S2:
[0024] The first-level PLL estimates the angle. As the input benchmark for the extended state observer ESO, the input vector of the extended state observer ESO is constructed. , represented as:
[0025]
[0026] , These are the orthogonal Hall signal input vectors of the extended state observer ESO.
[0027] The aforementioned embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture, in step S2, establishes the position angle calculated by the second-stage PLL. As the observation feedback vector of the extended state observer (ESO), the observation feedback vector of the extended state observer (ESO) is expressed as:
[0028]
[0029] , These are the observation values of the quadrature Hall signals constructed from the position signals calculated by the second-stage phase-locked loop.
[0030] In the aforementioned embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture, step S2 uses the following extended state observer (ESO) discretization model to estimate the angle of the first-stage PLL. The total disturbance is estimated in real time through observation:
[0031]
[0032]
[0033] in, and These are the orthogonal errors of the extended state observer (ESO) observation vector and input vector in the two-phase stationary coordinate system, respectively, corresponding to the orthogonal input of the Hall signal; , It is the fundamental wave state observation in a two-phase stationary coordinate system obtained by the Extended State Observer (ESO). , It is the disturbance state observation in a two-phase stationary coordinate system obtained by the Extended State Observer (ESO). , and , These are observer gain factor one and observer gain factor two, respectively. , , , , , These are the differential components of the corresponding letters.
[0034] A computer system includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to perform the steps of the method described above.
[0035] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.
[0036] The beneficial effects achieved by this invention are as follows: The method of this invention utilizes a two-stage closed-loop ESO-PLL composite architecture to achieve system-level decoupling of filtering and dynamic bandwidth. Unlike the single-stage phase-locked loop in existing technologies, it extracts the initial reference through the first-stage SRF-PLL and uses the ESO to define low-frequency harmonics as lumped disturbances for real-time observation and removal. This solves the filtering delay problem caused by low harmonic frequencies under low-speed operating conditions, enabling the phase-locked loop-based angle decoding system to maintain its original high control bandwidth and dynamic response speed while deeply suppressing disturbances. This invention does not use the ESO as a simple open-loop feedforward compensation, but rather utilizes the high-precision position angle calculated by the second-stage SRF-PLL. As an observation feedback quantity, the input of the ESO is reintroduced to form a global closed loop. This adaptive feedback mechanism can automatically overcome the DC bias and imbalance of the signal caused by sensor installation errors and inconsistent gain, and still ensure the extremely high purity of the position signal under harsh and non-ideal working conditions.
[0037] Without increasing hardware costs, the software architecture of this invention can still achieve high-precision angle tracking and delay-free dynamic start-up response in the low-speed range with electrical frequencies below 20Hz, significantly reducing the harmonic content of position calculation errors and providing accurate position feedback for the low-speed dual closed-loop control of permanent magnet synchronous motors. Attached Figure Description
[0038] Figure 1 This is a schematic diagram of an embedded magnetic encoder decoding method with disturbance suppression capability according to Embodiment 1 of the present invention;
[0039] Figure 2 This is a schematic diagram of the SRF-PLL small-signal model in Embodiment 1 of the present invention;
[0040] Figure 3a and Figure 3b These are schematic diagrams of Hall signal and harmonic analysis in Embodiment 1 of the present invention;
[0041] Figure 4 This is a schematic diagram of the Hall signals in the stationary coordinate system before and after disturbance suppression in Embodiment 1 of the present invention;
[0042] Figure 5 This is a schematic diagram of the position calculation signal in Embodiment 1 of the present invention;
[0043] Figure 6a This is a schematic diagram comparing the method in Embodiment 1 of the present invention with the existing SRF-PLL angle calculation results;
[0044] Figure 6b yes Figure 6a A schematic diagram of harmonic analysis of the angle calculation results recorded in the figure;
[0045] Figure 7 This is a schematic diagram of the dynamic experimental results in Embodiment 1 of the present invention. Detailed Implementation
[0046] The present invention will now be described in further detail with reference to the accompanying drawings and specific preferred embodiments.
[0047] Example 1
[0048] like Figure 1As shown, this embodiment provides an embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture. It employs a composite disturbance suppression structure combining an extended state observer (ESO) and a two-stage synchronous rotating coordinate system phase-locked loop (SRF-PLL), and includes the following steps:
[0049] S1, the embedded magnetic encoder uses a linear Hall sensor to process the raw magnetic flux density signal acquired by the three-phase linear Hall sensor. Preprocessing includes:
[0050] The Clarke transform is performed on the raw magnetic flux density signal acquired by the three-phase linear Hall sensor. Transforming from a three-phase stationary coordinate system to a two-phase stationary α-β axis yields a two-phase orthogonal linear Hall signal. , represented as:
[0051]
[0052]
[0053]
[0054] in It is the Clarke transformation matrix. The original magnetic flux density signal matrix, It is an orthogonal linear Hall signal matrix. The amplitude of the input signal.
[0055] Orthogonal linear Hall signals in two-phase stationary coordinate systems The first-level SRF-PLL is input for preliminary calculation to obtain the estimated angle of the motor rotor position without disturbance suppression and compensation. .
[0056] The SRF-PLL is theoretically a standard second-order system. The α-β axis signals are transformed into a dq coordinate system that rotates synchronously with the estimated rotor position using the Park transform, resulting in two-phase Hall signals in the synchronous coordinate system. :
[0057]
[0058] in, To input the true phase information of the Hall signal, To estimate phase information, This represents the phase estimation error. In phase-locked mode, the q-axis component... It should approach zero; a non-zero value represents the phase tracking error. For example... Figure 2 As shown, the phase estimation The error is processed by a proportional-integral (PI) loop filter, and then the initial PLL estimate of the motor rotor position is obtained through an integrator. Finally, the closed-loop transfer function is obtained. , represented as:
[0059]
[0060]
[0061] in, For the first-level PLL to estimate the angle, The estimated rotational speed of the first-stage PLL. To input the true phase information of the Hall signal, It is a proportionality coefficient. It is the integral coefficient. The damping coefficient is... For bandwidth, It is a complex frequency, representing the Laplace domain.
[0062] S2. Existing Extended State Observer (ESO) relies on high-precision input, while this invention creatively extracts the estimated angle of the perturbation-containing motor rotor position from the first-stage PLL. The input vector of the extended state observer ESO is constructed as the baseline input. In the extended state observer (ESO) discretized model, the orthogonal linear Hall signal is defined as the fundamental state observation. Non-ideal factors such as low-frequency harmonics and DC bias are defined as extended lumped disturbance state observations. The perturbation state observations are decoupled in real time using the discretized model of the extended state observer (ESO). Then the rotor position angle calculated by the second-stage PLL is... As the observation feedback vector of the extended state observer (ESO), the perturbation state observation... Reconstruct the high-purity fundamental state observations after stripping the perturbation. The specific process includes:
[0063] The first-level PLL estimates the angle. As input, the extended state observer (ESO) is used for perturbation estimation and compensation, including:
[0064] To suppress the angle estimation contained in the first-level PLL The disturbance in the first-stage PLL will estimate the angle. As the input benchmark for the extended state observer ESO, the input vector of the extended state observer ESO is constructed. , represented as:
[0065]
[0066] , These are the orthogonal Hall signal input vectors of the extended state observer ESO;
[0067] At the same time, a more precise position angle is established through subsequent second-level PLL calculation. As the observation feedback vector of the extended state observer (ESO), the observation feedback vector of the extended state observer (ESO) is expressed as:
[0068]
[0069] , These are the observations of the quadrature Hall signals constructed from the position signals calculated by the second-stage phase-locked loop;
[0070] The following extended state observer (ESO) discretization model is used to estimate the angle of the first-stage PLL. The total disturbance is estimated in real time through observation:
[0071]
[0072]
[0073] in, and These are the orthogonal errors of the extended state observer (ESO) observation vector and input vector in the two-phase stationary coordinate system, respectively, corresponding to the orthogonal input of the Hall signal; , It is the fundamental wave state observation in a two-phase stationary coordinate system obtained by the extended state observer (ESO), and its essence is the orthogonal Hall signal state quantity obtained after disturbance suppression. , It is the disturbance state observation in a two-phase stationary coordinate system obtained by the Extended State Observer (ESO). , and , The observer gain coefficients one and two, respectively, determine the dynamics and accuracy of the disturbance estimation. , , , , , These are the differential components of the corresponding letters.
[0074] The extended state observer (ESO) model in this invention differs from the traditional second-order extended state observer (ESO), which is represented as follows: = + + If the input is not a differential component, and the input is an orthogonal signal, then the state observation is the integral of the orthogonal signal after perturbation, expressed as... = The system needs to include rotational speed-related values. Introducing rotational speed into the closed-loop system would complicate its structure and reduce its dynamic characteristics. Therefore, this invention improves the traditional Extended State Observer (ESO) model into a discretized model, processing the signal in a two-phase stationary coordinate system. This eliminates the need for complex integration or differentiation and is not simply a combination of a phase-locked loop and an ESO model. It is suitable for position calculation.
[0075] In the prior art, the extended state observer (ESO) replaces the PI in the phase-locked loop (PLL) to calculate the position signal through direct input. However, this still cannot change the trade-off between bandwidth and filtering. The extended state observer (ESO) of this invention separates bandwidth and filtering. The parameter tuning of the extended state observer (ESO) only affects filtering, while the parameter tuning of the PLL only affects position calculation.
[0076] S3. High-precision fundamental state observation after reconstruction of the extended state observer (ESO) Input the second-level SRF-PLL to calculate the final high-precision motor rotor position angle. .
[0077] To ensure the global convergence and robustness of the observer, this invention constructs an outer closed-loop feedback: the final rotor position angle is... Transform into an ideal observation vector ( =cos( ), =sin( ), and feed it back to the ESO's observation terminal.
[0078] The aforementioned closed-loop mechanism continuously calculates the actual output. With internal state error Real-time correction of disturbance state observations The observation weights.
[0079] Finally, a purified orthogonal signal with strong anti-interference capability is output, completing the transformation from single-point filtering to a system-level adaptive anti-interference architecture. The specific process is as follows:
[0080] The orthogonal Hall signal state variables constructed above and The data is fed into the second-stage SRF-PLL to calculate the position angle. The second-stage SRF-PLL has the same structure as the first-stage SRF-PLL, but due to the significantly improved input signal quality, the calculated position angle is... The accuracy is much higher than Forming a key closed loop: achieving high-precision position angles The observed data is fed back to the ESO in the form of an observation vector, and the difference between the observed data and the input vector is used to update the perturbation state observation in real time. and The disturbance state observations are applied to the input vector. , Compensation is performed to construct a fundamental wave state observation with lower harmonic content and smaller disturbance. and .in and This means outputting a purer and more stable two-phase stationary coordinate system orthogonal signal to the second-stage phase-locked loop, such as... Figure 2 As shown, the output is after the second-stage phase-locked loop. .
[0081]
[0082] The closed-loop structure enables the embedded linear Hall rotor position detection system to adaptively suppress low-frequency disturbances.
[0083] This embodiment provides a specific set of parameters for calculating the position of the parameters, including:
[0084] First, the motor was controlled at a constant speed of 100 rpm, at which point the fundamental frequency was 11.67 Hz. The oscilloscope simultaneously recorded the three-phase linear Hall signals, as shown in Figure 3(a). FFT decomposition of the three-phase signals was then performed, and the results are shown in Figure 3(b). It can be seen that there are DC errors and amplitude imbalances among the three-phase Hall signals, with the third and fifth harmonics being the dominant frequencies.
[0085] To preprocess the signals from the embedded magnetic encoder in the motor, this software first uses Clarke transform to convert the three-phase magnetic flux density signals from the three linear Hall sensors. Transforming to a two-phase stationary coordinate system yields a set of orthogonal signals. Subsequently, the system performs preliminary calculations on the orthogonal signal based on a phase-locked loop (SRF-PLL) to extract the initial rotor angle, which contains basic position information but is affected by harmonic disturbances. This angle signal is used to construct the input reference for the Extended State Observer (ESO). In the disturbance suppression stage, the system uses the final position angle, which is then compensated by the ESO and precisely calculated again through a phase-locked loop. As an observational feedback quantity of ESO. Through real-time comparison With position angle The ESO can accurately estimate and compensate for the observed disturbances in the signal. Through the above closed-loop processing, disturbances introduced by factors such as harmonics are effectively suppressed, ultimately outputting a purer and more stable set of orthogonal signals in two-phase stationary coordinate systems. Its purified waveform is as follows Figure 4 As shown in the figure, this signal provides a reliable input for subsequent high-precision position calculation and control.
[0086] The magnetic flux density signal, after ESO disturbance suppression, is then processed by a next-stage phase-locked loop for more accurate position signal extraction. The results demonstrate that this method can still achieve accurate angle detection at electrical frequencies below 20Hz. As shown in Figure 6, the embedded magnetic encoder decoding software with disturbance suppression capabilities achieves better detection accuracy, further reducing the harmonics of the position signal error.
[0087] In addition to analyzing the steady-state characteristics of the algorithm software, dynamic experiments were also conducted, such as... Figure 7 As shown in the figure, the results indicate that the algorithm has good dynamic characteristics and a good response speed when the motor starts.
[0088] Example 2
[0089] A computer system includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to perform the steps of the method as described in Embodiment 1.
[0090] Example 3
[0091] A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the method as described in Example 1.
[0092] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0093] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0094] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0095] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.
Claims
1. A decoding method for an embedded magnetic encoder based on a two-stage closed-loop ESO-PLL composite architecture, characterized in that, include: S1, the embedded magnetic encoder uses a linear Hall sensor, performs Clarke transformation on the raw signal collected by the three-phase linear Hall sensor to obtain a two-phase orthogonal linear Hall signal; the orthogonal linear Hall signal is input into the first-stage SRF-PLL for preliminary calculation to obtain the estimated angle of the motor rotor position; S2. The estimated angle of the motor rotor position extracted by the first-stage PLL is used as the reference input of the extended state observer (ESO). The input vector of the extended state observer (ESO) is constructed. The disturbance state observation is decoupled in real time through the discretized model of the extended state observer (ESO). The rotor position angle calculated by the second-stage PLL is used as the observation feedback vector of the extended state observer (ESO). The fundamental state observation after the disturbance is removed is reconstructed based on the disturbance state observation. S3. Input the fundamental state observation reconstructed by the extended state observer (ESO) into the second-stage SRF-PLL to calculate the final motor rotor position angle.
2. The embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture according to claim 1, characterized in that, In step S1, the raw magnetic flux density signal acquired by the three-phase linear Hall sensor is... The transformation from a three-phase stationary coordinate system to a two-phase stationary α-β axis is expressed as: in It is the Clarke transformation matrix. The amplitude of the input signal. The original magnetic flux density signal matrix, It is an orthogonal linear Hall signal matrix.
3. The embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture according to claim 1, characterized in that, In step S1, the α-β axis signals are converted to the dq coordinate system, which rotates synchronously with the estimated rotor position, using the Park transform, to obtain the two-phase Hall signals in the synchronous coordinate system. : in, To input the true phase information of the Hall signal, To estimate phase information, This is the phase estimation error; The phase estimation The error is processed by a proportional-integral loop filter, and then the first-stage PLL estimation angle of the motor rotor position is obtained through an integrator. The closed-loop transfer function is obtained. , is represented as: in, For the first-level PLL to estimate the angle, The estimated rotational speed of the first-stage PLL. It is a proportionality coefficient. It is the integral coefficient. The damping coefficient is... For bandwidth, It is a complex frequency.
4. The embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture according to claim 1, characterized in that, Step S2 includes: The first-level PLL estimates the angle. As the input benchmark for the extended state observer ESO, the input vector of the extended state observer ESO is constructed. , is represented as: , These are the orthogonal Hall signal input vectors of the extended state observer ESO.
5. The embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture according to claim 1, characterized in that, In step S2, the position angle calculated by the second-level PLL is established. As the observation feedback vector of the extended state observer (ESO), the observation feedback vector of the extended state observer (ESO) is expressed as: , These are the observation values of the quadrature Hall signals constructed from the position signals calculated by the second-stage phase-locked loop.
6. The embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture according to claim 1, characterized in that, In step S2, the following extended state observer (ESO) discretization model is used to estimate the angle of the first-stage PLL. The total disturbance is estimated in real time through observation: in, and These are the orthogonal errors of the extended state observer (ESO) observation vector and input vector in the two-phase stationary coordinate system, respectively, corresponding to the orthogonal input of the Hall signal; , It is the fundamental wave state observation in a two-phase stationary coordinate system obtained by the Extended State Observer (ESO). , It is the disturbance state observation in a two-phase stationary coordinate system obtained by the Extended State Observer (ESO). , and , These are observer gain factor one and observer gain factor two, respectively. , , , , , These are the differential components of the corresponding letters.
7. A computer system comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the embedded magnetic encoder decoding method based on a two-stage closed-loop ESO-PLL composite architecture as described in any one of claims 1-6.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of an embedded magnetic encoder decoding method based on a dual-level closed-loop ESO-PLL composite architecture as described in any one of claims 1-6.