A motor position detection system
By combining data processing from eddy current modules, resolver modules, and photoelectric modules, and utilizing Kalman filters and error compensation algorithms, the problems of Hall sensor saturation under high magnetic fields and electromagnetic interference affecting resolvers were solved, achieving higher reliability and accuracy in motor position detection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- BEIJING AERONAUTIC SCI & TECH RES INST OF COMAC
- Filing Date
- 2025-06-18
- Publication Date
- 2026-07-03
AI Technical Summary
Existing motor position detection systems may experience Hall sensor saturation under high magnetic field strength, resolvers are affected by electromagnetic interference, and temperature changes cause position drift, resulting in poor equipment reliability. Furthermore, the sensor type limits its versatility, requiring calibration in a controlled environment to maintain accuracy.
By combining an eddy current module, a resolver module, and a photoelectric module with a data processing module, and using a Kalman filter and an error compensation algorithm, the eddy current data is corrected using resolver data and photoelectric data to achieve motor position detection.
It improves the accuracy and reliability of motor position detection, reduces dependence on environmental conditions, and enhances the versatility of the sensor and the overall reliability of the equipment.
Smart Images
Figure CN224459679U_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of electrical equipment technology, and in particular to a motor position detection system. Background Technology
[0002] Conventional methods for monitoring motor position include Hall effect sensors, resolvers, and optical encoders. Among these, some Hall effect sensors may saturate under high magnetic field strength, resulting in inaccurate measurement of magnetic field strength. Resolvers are affected by electromagnetic interference generated by the motor, and temperature changes may cause position drift, resulting in poor overall equipment reliability.
[0003] The document "A Calibration System and Method for a Motor Position Sensor" discloses relevant content on the calibration of a motor position sensor. However, this document has the following problems:
[0004] 1) Repeatability: After the motor has been running for a period of time, the sensor may need to be recalibrated to maintain its accuracy, which increases the workload of maintenance.
[0005] 2) Environmental factors: The calibration process may be affected by environmental conditions (such as temperature, humidity, and electromagnetic interference), and needs to be carried out in a controlled environment to ensure accuracy.
[0006] 3) Sensor type limitations: Some calibration systems may only be applicable to specific types of sensors, which may limit their versatility in different types of motor control systems. Utility Model Content
[0007] This specification provides a motor position detection system to address the technical problem of improving the motor position detection effect.
[0008] To address the aforementioned technical problems, the embodiments in this specification provide the following technical solutions:
[0009] This specification provides an embodiment of a motor position detection system, the system comprising:
[0010] Eddy current module, used to output eddy current data of the motor;
[0011] The resolver module is used to output the resolver data of the motor;
[0012] The photoelectric module is used to output photoelectric data from the motor.
[0013] The data processing module is used to obtain an initial position result based on the eddy current data, and to obtain a position calibration result based on the resolver data and photoelectric data; to calculate the residual between the position calibration result and the initial position result, to determine the cumulative error of the eddy current data based on the residual, and to correct the eddy current data using the cumulative error; and to obtain a motor position detection result based on the resolver data, photoelectric data, and the corrected eddy current data.
[0014] Preferably, the eddy current module is an eddy current sensor;
[0015] And / or,
[0016] The resolver module is a rotary transformer;
[0017] And / or,
[0018] The photoelectric module is a photoelectric encoder.
[0019] Preferably, the resolver module and the eddy current module are installed on the motor drive end, and the photoelectric module is installed on the non-drive end of the motor.
[0020] Preferably, the eddy current data includes angle measurement data;
[0021] The initial position result obtained based on the eddy current data includes:
[0022] The Kalman filter predicts the initial position based on the angle measurement data, historical data from the eddy current sensor, and the motion model.
[0023] Preferably, determining the cumulative error of the eddy current data based on the residual includes:
[0024] The residual is input into the Kalman filter to update the error covariance matrix of the Kalman filter;
[0025] The cumulative error of the eddy current sensor is calculated using the updated error covariance matrix from the error model.
[0026] Preferably, the resolver data includes an angle signal, and the photoelectric data includes an incremental positioning result of the angle;
[0027] The position calibration results obtained based on the rotary data and photoelectric data include:
[0028] The angle signal is fused with the incremental positioning result of the angle to form a position calibration result.
[0029] Preferably, the resolver data is output by the resolver module, and the eddy current data is output by the eddy current module.
[0030] Preferably, the output frequency of the photoelectric data is the motor's base frequency;
[0031] And / or,
[0032] The output frequency of the eddy current data is the motor's fundamental frequency multiplied by the number of pole pairs.
[0033] Preferably, the vortex data is data filtered by a convolutional filtering algorithm;
[0034] And / or,
[0035] The eddy current data is the data after being filtered by a convolutional filtering algorithm;
[0036] And / or,
[0037] The photoelectric data is the data after being filtered by an average filtering algorithm.
[0038] The above-described at least one technical solution used in the embodiments of this specification can achieve the following beneficial effects:
[0039] By using electronic signal processing technology, various observation values such as eddy current data, resolver data, and photoelectric data are collected. The eddy current data is corrected using resolver data and photoelectric data. The motor position detection result is obtained based on the corrected eddy current data, resolver data, and photoelectric data, which can effectively improve the motor position detection effect. Attached Figure Description
[0040] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings used in the description of the embodiments of this specification or the prior art will be briefly described below. Obviously, the drawings used in some embodiments of this application are only described below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0041] Figure 1 This is a schematic diagram of data acquisition in the embodiments of this specification.
[0042] Figure 2 This is a schematic diagram of the motor position detection process in the embodiments of this specification.
[0043] Figure 3 This is a schematic diagram of the data processing flow in the embodiments of this specification. Detailed Implementation
[0044] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments involved in the specific implementation are only a part of the embodiments of this application, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments in the specific implementation without creative effort should fall within the protection scope of this application.
[0045] This specification provides an embodiment of a motor position detection system, the system comprising:
[0046] Eddy current module, used to output eddy current data of the motor;
[0047] The resolver module is used to output the resolver data of the motor;
[0048] The photoelectric module is used to output photoelectric data from the motor.
[0049] The data processing module is used to obtain an initial position result based on the eddy current data, and to obtain a position calibration result based on the resolver data and photoelectric data; to calculate the residual between the position calibration result and the initial position result, to determine the cumulative error of the eddy current data based on the residual, and to correct the eddy current data using the cumulative error; and to obtain a motor position detection result based on the resolver data, photoelectric data, and the corrected eddy current data.
[0050] Preferably, the eddy current module is an eddy current sensor; and / or, the resolver module is a rotary transformer; and / or, the photoelectric module is a photoelectric encoder.
[0051] Preferably, the resolver module and the eddy current module are mounted on the motor drive end, while the photoelectric module is mounted on the non-drive end of the motor. The eddy current module and the resolver module are decoupled and operate independently. For example... Figure 1 As shown. For example, the stator of the resolver is connected to the motor housing, the rotor of the resolver is connected to the motor shaft via a key, the eddy current sensor is connected to the motor drive end via a threaded hole in the housing, and the photoelectric encoder is attached to the rear end cover of the motor. The specific installation methods for the eddy current module, resolver module, and photoelectric module are not limited in the embodiments described in this manual.
[0052] The following is a further explanation of the operation of the eddy current module:
[0053] In the embodiments of this specification, an eddy current module can be used to collect relevant data from the motor, and the data collected and output by the eddy current module is eddy current data; alternatively, the data collected by the eddy current module can be filtered, and the filtered data is eddy current data. For example, convolution filtering can be used on the data collected by the eddy current module, that is, the eddy current data is the data filtered by the convolution filtering algorithm;
[0054] Preferably, the output frequency of the eddy current data or the data acquisition frequency of the eddy current module is the motor's fundamental frequency multiplied by the number of pole pairs. In practical applications, the output frequency of the eddy current data is a high-frequency output, such as a kHz-level update rate.
[0055] In the embodiments described in this specification, the eddy current data may include angle measurement results or angle measurement data, which can be obtained through the calculation unit of the eddy current module itself.
[0056] The following is a further explanation of the working process of the resolver module:
[0057] In the embodiments of this specification, a resolver module can be used to collect relevant data from the motor, and the data collected and output by the resolver module is resolver data; alternatively, the data collected by the resolver module can be filtered, and the filtered data is resolver data. Specifically, convolution filtering can be used on the data collected by the resolver module, meaning the resolver data is the data filtered by a convolution filtering algorithm.
[0058] Resolver data can include angle signals, which can be obtained through the resolver module's own calculation unit.
[0059] The following is a further explanation of the working function of the optoelectronic module:
[0060] In the embodiments of this specification, a photoelectric module can be used to collect relevant data from the motor, and the data collected and output by the photoelectric module is photoelectric data; alternatively, the data collected by the photoelectric module can be filtered, and the filtered data is also photoelectric data. Specifically, the data collected by the photoelectric module can be filtered using an average filtering algorithm, meaning the photoelectric data is the data after being filtered by an average filtering algorithm.
[0061] Preferably, the output frequency of the photoelectric data or the data acquisition frequency of the photoelectric module is the motor's base frequency. In practical applications, the output frequency of the photoelectric data is a low-frequency output, such as a Hz-level update rate.
[0062] Photoelectric data can include incremental positioning results of angles, which can be obtained through the calculation unit of the photoelectric module itself.
[0063] The following is a further explanation of the working content of the data processing module:
[0064] In the embodiments of this specification, an initial position result can be obtained based on eddy current data. Obtaining the initial position result from eddy current data can include: a Kalman filter predicting the initial position result based on the aforementioned angle measurement data, historical data from the eddy current sensor, and a motion model. Specifically, the Kalman filter can predict the current motor positioning data based on historical eddy current data and the constructed motion model (the motion model can be, for example, a uniform velocity / uniform acceleration model; the content or form of the motion model is not limited in the embodiments of this specification). This motor positioning data can be output as the initial position result.
[0065] If no photoelectric data or resolver data is collected, the initial position result can be used as the motor position detection result.
[0066] Having acquired resolver data and photoelectric data, a position calibration result can be obtained from the resolver data and photoelectric data. Obtaining the position calibration result from the resolver data and photoelectric data can include fusing the aforementioned angle signals with the incremental positioning results of the aforementioned angles to form the position calibration result.
[0067] In the embodiments described in this specification, the residual between the position calibration result and the initial position result can be calculated. For example, the position calibration result is compared with the initial position result, and the difference between the two (i.e., the residual) is calculated. This difference reflects the cumulative error of the eddy current module.
[0068] The cumulative error of the eddy current data can be determined based on the residuals described above. This determination can include: inputting the residuals as observation errors into a Kalman filter to update the Kalman filter's error covariance matrix, thereby correcting the Kalman filter; and using the updated error covariance matrix, calculating the cumulative error of the eddy current data or the eddy current module using an error model. The cumulative error can then be output as an error compensation feedback signal.
[0069] The error model may be, for example, a state-space model. The embodiments in this specification do not limit the content or form of the error model.
[0070] In the embodiments described in this specification, the accumulated error can be used to correct the acquired eddy current data. For example, the eddy current data can be corrected by resetting the integral drift or adjusting the calibration parameters.
[0071] The above-mentioned "calculating the residual between the position calibration result and the initial position result, determining the cumulative error of the eddy current data based on the residual, and correcting the eddy current data using the cumulative error" can be called error compensation.
[0072] The following is a more detailed explanation of the error compensation process:
[0073] 1. Define state variables and observation variables
[0074] State variables:
[0075] In the formula: θ k Indicates the actual angular position; ω k b represents angular velocity; k This represents the cumulative error (the core quantity to be estimated) of the eddy current module or eddy current data.
[0076] Observed variable: z v This represents the eddy current data output by the eddy current module (including high-frequency measurements with errors); z o This indicates the position calibration result (low-frequency absolute reference).
[0077] 2. Establish state equations (or process models).
[0078] The dynamic evolution of state variables is represented as: x k =Fx k-1 +w k ;
[0079] The state transition matrix F is expressed as:
[0080]
[0081] In the formula, Δt represents the sampling time interval.
[0082] The above cumulative error b k Modeled as a random walk process: (b k =b k-1 +w b ).
[0083] Process noise
[0084]
[0085] In the formula, q b Control the drift rate of the accumulated error (which needs to be calibrated according to the eddy current characteristics).
[0086] 3. Establish observation equations
[0087] Correlation between state variables and sensor output: z k =Hx k +v k ;
[0088] Eddy current observation equation: z v =θ k +b k +v v →Hv =[1 0 1];
[0089] Position calibration result observation equation: z o =θ k +v o →H o =[1 0 0].
[0090] 4. Calculate the cumulative error
[0091] After acquiring resolver and photoelectric data, perform the following steps:
[0092] (1) Calculate the residual
[0093]
[0094] In the formula, z o This indicates the position calibration result, i.e., the current absolute position of the photoelectric + resolver. This represents the angle value predicted by the Kalman spectroscopy (uncorrected), i.e., the initial position result.
[0095] (2) Error decomposition
[0096] Through Kalman gain K k Separate the error components in the residuals:
[0097]
[0098] In the formula, This represents the third row of the gain matrix (corresponding to the cumulative error b). k );Δb k This represents the estimated increment of the cumulative error at the current moment.
[0099] (3) Error status update
[0100]
[0101] In the formula, This indicates the cumulative error of the prediction (from the previous time step or period);
[0102] This represents the optimal estimate of the corrected cumulative error.
[0103] (4) Covariance update
[0104] Update the error covariance matrix P k Quantitative estimation of uncertainty:
[0105] P k =(IK k H o )P k|k-1 .
[0106] 5. Long-term modeling of cumulative error
[0107] Establish a statistical model of error using historical data: b k =α·t+β+∈(t)
[0108] In the formula, α represents the drift coefficient (estimated through linear regression);
[0109] β represents the initial bias;
[0110] ∈(t) represents random fluctuation (variance determined by P) k [3,3] confirmed);
[0111] 6. Error compensation implementation
[0112] Compensation or correction of eddy current data is performed by estimating the cumulative error:
[0113]
[0114] In the formula, z v This represents the original measured value of the eddy current, i.e., the collected eddy current data;
[0115] This indicates the angle output after compensation or correction, i.e., the result after compensation or correction of eddy current data.
[0116] For examples of the specific relationships between modules in the embodiments of this specification, please refer to [reference]. Figure 3 The drive module drives the motor, the solver module in the resolver module acquires and outputs resolver signals, the solver unit in the eddy current sensor corrects the eddy current data, and the data processing module calculates and outputs the motor position results.
[0117] In the embodiments of this specification, the motor position detection result can be obtained based on resolver data, photoelectric data, and corrected eddy current data. For example, the corrected eddy current data and the position calibration result (i.e., "photoelectric + resolver" data) are fused by a Kalman filter to output the best estimate of the final positioning data as the motor position detection result.
[0118] The above process can be repeated cyclically to achieve a dynamic closed loop of "prediction → correction → adjustment". For example, at each moment (cycle), resolver data, eddy current data, and photoelectric data of the motor at the current moment (cycle) are collected. The initial position result for the current moment (cycle) is obtained based on the eddy current data. The position calibration result for the current moment (cycle) is obtained based on the resolver data and photoelectric data. The residual between the position calibration result and the initial position result is calculated. The cumulative error of the eddy current data at the current moment (cycle) is determined based on the residual. The cumulative error is used to correct the eddy current data at the current moment (cycle). The motor position detection result for the current moment (cycle) is obtained based on the resolver data, photoelectric data, and the corrected eddy current data. Of course, historical data can be used in the process of calculating residuals, etc., and the embodiments in this specification are not limited to this.
[0119] The embodiments described in this specification can achieve the following beneficial effects:
[0120] In the embodiments of this specification, the current motor positioning data (i.e., the initial position result) is inferred based on the measurement results in the eddy current data, and the position calibration result is determined based on the positioning results in the photoelectric data + resolver data. By comparing the difference between the two and establishing an error model, the cumulative error of the eddy current module or the eddy current data is estimated, and the error compensation is fed back to the eddy current module to correct the eddy current data. Then, the photoelectric data, resolver data and the corrected eddy current data are fused to output the best estimated value of the positioning data.
[0121] The above process is repeated cyclically, realizing the cyclic correction of eddy current data and calculating the motor position result using the photoelectric data, resolver data and the latest corrected eddy current data at the current time (cycle), effectively improving the reliability and accuracy of motor position detection and improving the motor position detection effect.
[0122] This specification describes embodiments that acquire resolver data, eddy current data, and photoelectric data using a resolver, eddy current sensor, and photoelectric encoder. The eddy current sensor directly transmits digital signals, exhibiting strong EMC resistance but slightly weaker environmental tolerance. When the eddy current sensor experiences data transmission gaps or fails, the resolver data serves as auxiliary data for the position signal, outputting angle information calculated by the resolver encoder's analytical algorithm. Considering the susceptibility of resolver signals to interference, incremental data from the photoelectric encoder is used to correct position calculation errors, thereby obtaining a more reliable motor position. When angle data acquisition is complete, the eddy current data, resolver data, and photoelectric auxiliary data are fused together to output the motor angle value, effectively improving the accuracy of motor position detection.
[0123] To ensure reliable motor position output, the embodiments in this specification employ a heterogeneous approach using different types of sensors. Through hardware redundancy and algorithms, the collected data are fused together to obtain highly reliable motor position information.
[0124] The embodiments in this specification combine the advantages and disadvantages of each of the rotary transformer, photoelectric encoder and Hall sensor, and use electronic signal processing technology to fuse the measurement values of multiple sensors to obtain more accurate motor position measurement results.
[0125] In the embodiments described in this specification, the eddy current data is corrected for errors in real time through a compensation signal to ensure that it remains highly reliable even when there is no absolute positioning signal.
[0126] The embodiments in this specification can adapt to the development trend of high switching frequency in electric drive systems, and provide a solution to further meet the needs of motor position acquisition within a limited space and weight range.
[0127] In the embodiments of this specification, the photoelectric encoder signal is used to correct the error of the resolver, thereby approximating and compensating for electromagnetic interference and achieving stable angle tracking.
[0128] In the embodiments of this specification, the redundancy of the main position sensor (eddy current sensor) and the auxiliary position sensor (rotary transformer and photoelectric encoder) is used to estimate and compensate for uncertainties such as electromagnetic interference and environmental interference inside the high-frequency drive system, and to compensate for the estimation error.
[0129] The embodiments in this specification can form a combined settlement system (e.g., by setting up corresponding settlement modules and architectures) in the embodiments. Figure 3 The system shown is reconstructed to form the motor position detection system provided in the embodiments of this specification.
[0130] In the embodiments described in this specification, multiple angle sensors can be deployed to cover the measurement range in different scenarios, and redundancy can be achieved through sensor heterogeneity.
[0131] The embodiments in this specification can be matched with application scenarios requiring high precision. By arranging sensors on the axial and radial sides of the motor, position data is obtained by measuring the rotation angle of the shaft from different regions and orientations, thereby improving the motor position detection effect.
[0132] In the embodiments of this specification, the collected signal data is integrated and processed, and a filter is used to remove noise and interference to improve the signal-to-noise ratio. A real-time processing algorithm is used to perform real-time calculation on the signal to improve the motor position detection effect.
[0133] In the embodiments of this specification, the outputs of multiple sensors or data sources, such as the eddy current sensor, rotary transformer, and photoelectric encoder, are assigned different weights according to the accuracy and reliability of different sensors. By combining the advantages of low-pass and high-pass filters, data in different frequency ranges are processed to improve the motor position detection effect.
[0134] The above description is merely an embodiment of this specification and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.
Claims
1. A motor position detection system, characterized by, The system includes: Eddy current module, used to output eddy current data of the motor; The resolver module is used to output the resolver data of the motor; The photoelectric module is used to output photoelectric data from the motor. The data processing module is used to obtain an initial position result based on the eddy current data, and to obtain a position calibration result based on the resolver data and photoelectric data; to calculate the residual between the position calibration result and the initial position result, to determine the cumulative error of the eddy current data based on the residual, and to correct the eddy current data using the cumulative error; and to obtain a motor position detection result based on the resolver data, photoelectric data, and the corrected eddy current data.
2. The system of claim 1, wherein, The eddy current module is an eddy current sensor; And / or, The resolver module is a rotary transformer; And / or, The photoelectric module is a photoelectric encoder.
3. The system of claim 1 or 2, wherein, The resolver module and the eddy current module are installed on the motor drive end, and the photoelectric module is installed on the non-drive end of the motor.
4. The system of claim 1, wherein, The eddy current data includes angle measurement data; The initial position result obtained based on the eddy current data includes: The Kalman filter predicts the initial position based on the angle measurement data, historical data from the eddy current sensor, and the motion model.
5. The system of claim 1, wherein, The cumulative error of the eddy current data determined based on the residual includes: The residual is input into the Kalman filter to update the error covariance matrix of the Kalman filter; The cumulative error of the eddy current sensor is calculated using the updated error covariance matrix from the error model.
6. The system of claim 1, wherein, The resolver data includes angle signals, and the photoelectric data includes incremental positioning results of the angle. The position calibration results obtained based on the rotary data and photoelectric data include: The angle signal is fused with the incremental positioning result of the angle to form a position calibration result.
7. The system of claim 1, wherein, The resolver data is output by the resolver module, and the eddy current data is output by the eddy current module.
8. The system of claim 1 or 7, wherein, The output frequency of the photoelectric data is the motor's base frequency; And / or, The output frequency of the eddy current data is the motor's base frequency multiplied by the number of pole pairs.
9. The system of claim 1, wherein, The spin data is the data after being filtered by a convolutional filtering algorithm; And / or, The eddy current data is the data after being filtered by a convolutional filtering algorithm; And / or, The photoelectric data is the data after being filtered by an average filtering algorithm.