A method for full-velocity domain observer fusion switching of PMSM for drop test rigs with rotating mechanisms
By using a continuous Sigmoid function and a current stability evaluation factor to correct the observer fusion weights in the aircraft landing gear rotating mechanism, the problem of inconsistent current dynamic state during motor speed switching was solved, achieving smooth transition of the observer and stable control of the motor.
Patent Information
- Application Number
- CN202611421813.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-14
- Publication Date
- 2026-11-10
AI Technical Summary
Existing full-speed sensorless control methods in aircraft landing gear rotating mechanisms struggle to account for the dynamic state of the current during motor speed switching or fusion, leading to fluctuations in position estimation and electromagnetic torque, which affects the smoothness of continuous acceleration.
A continuous Sigmoid function is used to establish speed fusion weights. Combined with the real-time motor speed and q-axis current stability evaluation factors, the observer fusion weights are corrected to achieve continuous output ratio adjustment of low-speed and medium-speed observers, ensuring current stability.
By adjusting the observer output ratio, the impact of changes in position and speed estimation during the observer transition process on the q-axis current and electromagnetic torque is reduced, thereby improving the operational stability of the aircraft landing gear rotating mechanism.
Smart Images

Figure CN123074339A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of motor control technology, specifically relating to a PMSM full-speed domain observer fusion switching method for a drop test bench with rotating mechanism. Background Technology
[0002] Sensorless control technology for permanent magnet synchronous motors can reduce the use of mechanical position sensors and improve the reliability of the drive system. Full-speed-range sensorless control typically adopts a speed-range observation method, that is, high-frequency signal injection is used to obtain rotor position information in the low-speed range, while sliding mode observers and other methods are used to estimate rotor position and speed in the medium and high-speed ranges, and a transition zone is set to achieve switching between the two types of observers.
[0003] In the prior art, CN117792197A uses a combination of square wave injection and extended Kalman filtering, and achieves switching between different speed regions through a linear transition algorithm; CN115149866A uses high-frequency sinusoidal voltage injection and a sliding mode observer, and employs a weighted average method in the transition section.
[0004] The above method can achieve the transition between low-speed and medium-to-high-speed observation results, but its switching or fusion process is mainly determined based on the motor speed and the preset speed range. For aircraft landing gear drive mechanisms, during the continuous acceleration of high-inertia tires, when the motor enters the observer transition zone, the q-axis current and electromagnetic torque may still be in a dynamic adjustment state. If the output ratio of the two types of observers is determined solely based on the speed, the fusion process may not be consistent with the actual current state of the motor, thus causing fluctuations in position estimation, current, and electromagnetic torque, affecting the smoothness of the continuous acceleration process of the drive mechanism. Summary of the Invention
[0005] In order to overcome the problems existing in the prior art, the purpose of this invention is to provide a full-speed domain observer fusion switching method for a PMSM (Permanent Magnet Synchronous Motor) with rotating mechanism on a drop test bench, so as to solve the problem that the current dynamic state during continuous acceleration is difficult to take into account when the existing full-speed domain sensorless control mainly determines the observer switching or fusion ratio based on the motor speed. This enables continuous fusion of the outputs of low-speed observers and medium- and high-speed observers.
[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for fusion switching of PMSM full-velocity domain observers for drop test rigs with rotating mechanisms includes the following steps; Step 1: Obtain the estimation results from the low-speed observer and the medium-to-high-speed observer respectively. The low-speed observer outputs the position estimate. and speed estimate Position estimates output by medium- and high-speed observers and estimated rotational speed Within the observer fusion interval, the two types of observers operate in parallel, providing basic data for observer fusion; Step 2: Construct a speed fusion weight function based on the real-time motor speed; use a continuous Sigmoid function to establish the observer speed weights, so that the fusion weights change continuously with the motor speed, keeping the low-speed observer dominant in the low-speed region and gradually increasing the output ratio of the medium- and high-speed observers in the medium- and high-speed regions; avoid abrupt changes in position estimation caused by switching fixed speed thresholds. Step 3: Real-time sampling of the q-axis current of the permanent magnet synchronous motor with rotating mechanism on the drop test bench; and construction of current stability evaluation factor based on the change of q-axis current at adjacent sampling times to characterize the current regulation state of the motor. Step 4: Calculate the final fusion weight based on the speed fusion weight and the current stability evaluation factor, and use the current stability evaluation factor to correct the speed fusion weight so that the final fusion weight changes with the motor speed and the stable state of the q-axis current. Step 5: Based on the final fusion weights, the position estimates and rotational speed estimates output by the low-speed observer and the medium-to-high-speed observer are weighted and fused to obtain the fused position angle. and rotational speed ; Step 6, merge the position angles and rotational speed As feedback for the field-oriented control system of the permanent magnet synchronous motor, it performs closed-loop speed and closed-loop current regulation, and drives the permanent magnet synchronous motor through space vector pulse width modulation control of the inverter.
[0007] In step 1, a high-frequency signal is injected into the observer in the low-speed region to obtain the rotor position estimate. and estimated rotational speed In the medium-to-high speed region, a super-helical sliding mode observer combined with an adaptive phase-locked loop is used to obtain rotor position estimates. and estimated rotational speed Within the observer fusion interval, the low-speed observer and the medium-to-high-speed observer operate in parallel, continuously outputting the corresponding position and rotational speed estimates as inputs for subsequent fusion calculations.
[0008] In step 2, the velocity fusion weights are established using the Sigmoid function, and their calculation formula is as follows: in, This refers to the real-time speed of the motor. The observer fusion center rotation speed is denoted by k; k is the slope coefficient of the Sigmoid function. Speed fusion weights The output weights for the medium- and high-speed observers are determined based on the motor speed, while the corresponding output weights for the low-speed observers are... ; When the motor speed is lower than the observer fusion center speed hour, <0.5, low-speed observer output dominates; when motor speed equals hour, =0.5, the velocity weights of the two types of observers are equal; when the motor speed is higher than... hour, >0.5, the output of the medium-to-high speed observer is dominant; as the motor speed increases, As the speed increases, the output ratio of the medium- and high-speed observers gradually increases, while the output ratio of the low-speed observers gradually decreases.
[0009] In step 3, a current stability evaluation factor is established based on the q-axis current change in adjacent sampling periods. First, the q-axis current change in adjacent sampling periods is calculated: in, This represents the q-axis current during the k-th sampling period. Let q be the q-axis current in the (k-1)th sampling period. This represents the change in q-axis current between adjacent sampling periods.
[0010] Calculate the current stability evaluation factor based on the q-axis current change: in, Let be the q-axis current stability evaluation factor, and 'a' be the adjustment coefficient, where a > 0, therefore 0 < 0. ≤1. When When it increases, A decrease indicates an increase in the q-axis current fluctuation; when When decreasing, An increase that approaches 1 indicates that the q-axis current is stabilizing. This current stability evaluation factor is used to correct the velocity fusion weights.
[0011] In step 4, based on the velocity fusion weights and q-axis current stability evaluation factor Calculate the final fusion weights Its expression is: in, For the final fusion weight, For speed fusion weights, This is a factor for evaluating the stability of the q-axis current. This is the fusion correction factor; When the change in q-axis current increases Decrease, Correction Term Increase, so that the final fusion weights As the q-axis current decreases, the output ratio of the medium-to-high-speed observer also decreases accordingly; when the q-axis current change decreases... The weights gradually increase and approach 1, while the correction term gradually decreases, eventually leading to the fusion weights. Gradually approaching the speed fusion weight The output ratio of medium and high speed observers increases accordingly.
[0012] In step 5, based on the final fusion weights The position angles and rotational speeds output from the low-speed and medium-to-high-speed observers are weighted and fused. The formula for calculating the position angle fusion is as follows: The formula for calculating rotational speed fusion is: in, The position angle after fusion. The rotational speed after fusion. and These are the estimated position angle and rotational speed output from the low-speed observer, respectively. and These are the position angle and rotational speed estimates output by the medium- and high-speed observers, respectively. For the final fusion weight; When the final fusion weight When the weight is reduced, the output proportion of the low-speed observer in the fusion result increases; when the final fusion weight... As the speed increases, the output ratio of the medium- and high-speed observers in the fusion result increases, thereby achieving continuous fusion of the position angle and rotational speed outputs of the two types of observers.
[0013] In step 6, the fused position angles As a rotor position feedback quantity, it is used for Park transform and inverse Park transform; the fused speed is... As a speed feedback quantity, it is compared with a given rotational speed, and the resulting rotational speed deviation is used by the speed regulator to obtain the q-axis current reference value; the q-axis current reference value is compared with the actual q-axis current, and the resulting current deviation is used by the current regulator to obtain the q-axis voltage command, while the d-axis current is subjected to closed-loop regulation to obtain the d-axis voltage command; based on the fused position angle The d-axis and q-axis voltage commands are subjected to inverse Park transformation, and the inverter switching signal is generated by space vector pulse width modulation to control the inverter to drive the permanent magnet synchronous motor, thereby realizing the full-speed domain sensorless closed-loop control of the permanent magnet synchronous motor with rotating mechanism.
[0014] The beneficial effects of this invention are: In view of the characteristics of aircraft landing gear rotating mechanism driving tires, such as large load rotational inertia and need for continuous acceleration, this invention constructs speed fusion weights based on the rotational speed of permanent magnet synchronous motor of rotating mechanism of drop test bench, and establishes current stability evaluation factor based on the q-axis current change in adjacent sampling periods. The speed fusion weights are corrected by the current stability evaluation factor to obtain the final fusion weights.
[0015] When the change in q-axis current increases, the current stability evaluation factor decreases, and the final fusion weight decreases accordingly, thus reducing the output ratio of the medium- and high-speed observers. Conversely, when the change in q-axis current decreases, the current stability evaluation factor increases, and the final fusion weight gradually approaches the speed fusion weight, causing the output ratio of the medium- and high-speed observers to gradually increase with motor speed. Through this method, the observer fusion process simultaneously considers the motor speed and the steady state of the q-axis current, reducing the impact of changes in position and speed estimation results on the q-axis current and electromagnetic torque during the observer transition process, and improving the operational smoothness of the aircraft landing gear rotating mechanism during continuous acceleration.
[0016] This invention adds a fusion weight calculation step to the original structure of low-speed and medium-to-high-speed observers without changing the basic structure of the two types of observers. Attached image description: Figure 1 This is a schematic diagram of the PMSM full-speed domain observer fusion switching control for the aircraft landing gear rotating mechanism of the present invention. Detailed Implementation
[0017] The following is in conjunction with the appendix Figure 1 The specific embodiments of the present invention will be further described below.
[0018] like Figure 1 As shown, this embodiment uses a permanent magnet synchronous motor with a rotating mechanism on an aircraft landing gear as the controlled object, and employs a speed outer loop and a current inner loop to form a field-oriented control system. The three-phase currents are transformed to obtain the d-axis current and q-axis current, where the q-axis current is used to construct a current stability evaluation factor. In the low-speed region, a high-frequency signal is injected into the observer to obtain rotor position and speed estimates, while in the medium-to-high-speed region, a super-helical sliding mode observer is used to obtain rotor position and speed estimates. Within the observer fusion interval, the two types of observers operate in parallel, and their output position and speed estimates are input into the fusion module.
[0019] When the aircraft landing gear drive mechanism continuously accelerates the tires from low speed to the target speed, the tires and transmission components generate a large equivalent moment of inertia. During acceleration, the motor has a continuous torque output demand, and the q-axis current changes with the motor's operating state. This embodiment determines the speed fusion weight based on the motor speed and establishes a current stability evaluation factor based on the change in q-axis current. This evaluation factor is then used to correct the speed fusion weight, resulting in the final fusion weight. The fusion module performs weighted fusion of the position and speed estimates output by the two types of observers based on the final fusion weight, and feeds the fused position angle and speed back to the field-oriented control system for closed-loop control of the permanent magnet synchronous motor of the drive mechanism.
[0020] The specific implementation steps are as follows.
[0021] Step 1: Obtain the estimation results from low-speed and medium-to-high-speed observers. During the continuous acceleration of the aircraft landing gear-driven tires from low speed to the target speed by the rotating mechanism, high-frequency signals are injected into the observer in the low-speed region to obtain the rotor position estimate. and estimated rotational speed In the medium-to-high speed region, a super-helical sliding mode observer is used to obtain rotor position estimates. and estimated rotational speed .
[0022] When the motor enters the observer fusion interval, the low-speed observer and the medium- and high-speed observers operate in parallel, continuously outputting the corresponding position and speed estimates as inputs for subsequent fusion calculations.
[0023] Step 2: Construct speed fusion weights Speed fusion weights are established based on the real-time rotational speed of the motor.
[0024] During the continuous acceleration of the tires driven by the aircraft landing gear rotating mechanism, a speed fusion weight is constructed based on the real-time rotational speed of the motor. This embodiment uses the Sigmoid function to calculate the speed fusion weight, and its expression is: in, This refers to the real-time speed of the motor. Let be the rotational speed of the observer fusion center, and k be the slope coefficient of the Sigmoid function.
[0025] The speed fusion weight The corresponding output weights for medium- and high-speed observers are as follows: When the motor speed is lower than the fusion center speed hour, <0.5, low-speed observer output dominates; when motor speed equals hour, =0.5, the velocity weights of the two types of observers are equal; when the motor speed is higher than... hour, >0.5, the output of the medium-to-high speed observer dominates. As the speed of the tire driven by the rotating mechanism gradually increases, As the speed increases, the output ratio of the medium- and high-speed observers gradually increases.
[0026] Step 3: Establish q-axis current stability evaluation factors During the continuous acceleration of the aircraft landing gear-driven tires, the motor needs to continuously output electromagnetic torque, and the q-axis current changes with the motor's operating state. To reflect the stable state of the current during observer fusion, a current stability evaluation factor is established based on the change in q-axis current between adjacent sampling periods. First, the change in q-axis current between adjacent sampling periods is calculated: in, This represents the q-axis current during the k-th sampling period. Let q be the q-axis current in the (k-1)th sampling period. This represents the change in q-axis current between adjacent sampling periods.
[0027] Construct current stability evaluation factors based on q-axis current changes: in, Let be the q-axis current stability evaluation factor, and 'a' be the adjustment coefficient, where a > 0, therefore 0 < 0. ≤1. When When it increases, A decrease indicates an increase in the q-axis current fluctuation; when When decreasing, An increase that approaches 1 indicates that the q-axis current is stabilizing. This current stability evaluation factor is used to subsequently correct the velocity fusion weights.
[0028] Step 4: Calculate the observer fusion weights To ensure that the observer fusion process simultaneously reflects the steady state of both motor speed and q-axis current, based on the speed fusion weights... and q-axis current stability evaluation factor Calculate the final fusion weights Its expression is: in, For the final fusion weight, For speed fusion weights, This is a factor for evaluating the stability of the q-axis current. This is the fusion correction coefficient.
[0029] During the continuous acceleration of the tires driven by the landing gear rotating mechanism, when the change in q-axis current increases... Decrease, Correction Term Increase, so that the final fusion weights As the q-axis current decreases, the output ratio of the medium-to-high-speed observer also decreases accordingly; when the q-axis current change decreases... As the value increases and approaches 1, the correction term gradually decreases, eventually leading to the fusion of weights. Gradually approaching the speed fusion weight The output ratio of medium and high speed observers increases accordingly.
[0030] Through the above modifications, the observer output ratio is adjusted not only with the motor speed but also with the steady state of the q-axis current, thus preventing the medium- and high-speed observer from occupying a large output ratio too early when the q-axis current changes significantly.
[0031] Step 5: Calculation of position angle and rotation speed Based on the final fusion weight The position angle and rotational speed estimates output by the low-speed observer and the medium- and high-speed observer are weighted and fused.
[0032] The formula for calculating position angle fusion is as follows: The formula for calculating rotational speed fusion is as follows: in, The position angle after fusion. The rotational speed after fusion. and These are the estimated position angle and rotational speed output from the low-speed observer, respectively. and These are the position angle and rotational speed estimates output by the medium- and high-speed observers, respectively. This is the final fusion weight.
[0033] When the final fusion weight When the size is small, low-speed observers account for a larger proportion of the fusion results; as the size increases... As the speed increases, the proportion of medium- and high-speed observers gradually increases. This leads to a unified position angle. and rotational speed , serving as the position and velocity feedback quantities for the field orientation control system of the permanent magnet synchronous motor.
[0034] Step 6: Complete full-speed domain closed-loop control The fusion position angle obtained in Step 5 As a rotor position feedback quantity, it is used for Park transform and inverse Park transform; it integrates the rotational speed. As a speed feedback quantity, it is compared with a given speed, and the resulting speed deviation is used by the speed regulator to obtain the q-axis current reference value.
[0035] The q-axis current reference value is compared with the actual q-axis current, and the resulting current deviation is used by the current regulator to obtain the q-axis voltage command; the d-axis current is adjusted in a closed loop according to the field-oriented control method to obtain the d-axis voltage command. Based on the fusion position angle... The d-axis and q-axis voltage commands are subjected to inverse Park transformation, and the inverter switching signals are generated through space vector pulse width modulation to control the inverter to drive the permanent magnet synchronous motor.
[0036] During the continuous acceleration of the aircraft landing gear-driven tires from low speed to medium-high speed via the rotation mechanism, the fusion position angle is... and fusion speed As a unified feedback quantity, it participates in the above closed-loop control, enabling the low-speed observer and the medium-to-high-speed observer to continuously transition within the fusion interval, thus completing the full-speed-domain sensorless control of the permanent magnet synchronous motor with rotating mechanism.
Claims
1. A PMSM full-velocity domain observer fusion switching method for a drop test rig with a rotating mechanism, characterized in that, Includes the following steps; Step 1: Obtain the estimation results from the low-speed observer and the medium-to-high-speed observer respectively. The low-speed observer outputs the position estimate. and estimated rotational speed Position estimates output by medium- and high-speed observers and estimated rotational speed The two types of observers operate in parallel within the observer fusion interval; Step 2: Establish the observer speed weights using a continuous Sigmoid function, so that the fusion weights change continuously with the motor speed. In the low-speed region, the low-speed observer is dominant, and in the medium- and high-speed region, the output ratio of the medium- and high-speed observer is gradually increased. Step 3: Sample the q-axis current of the permanent magnet synchronous motor in real time, and construct a current stability evaluation factor based on the change in q-axis current at adjacent sampling times; Step 4: Calculate the final fusion weight based on the speed fusion weight and the current stability evaluation factor, and use the current stability evaluation factor to correct the speed fusion weight so that the final fusion weight changes with the motor speed and the stable state of the q-axis current. Step 5: Based on the final fusion weights, the position estimates and rotational speed estimates output by the low-speed observer and the medium-to-high-speed observer are weighted and fused to obtain the fused position angle. and rotational speed ; Step 6, merge the position angles and rotational speed As feedback for the field-oriented control system of the permanent magnet synchronous motor, it performs closed-loop speed and closed-loop current regulation, and drives the permanent magnet synchronous motor through space vector pulse width modulation control of the inverter.
2. The PMSM full-velocity domain observer fusion switching method for a drop test rig with a rotating mechanism as described in claim 1, characterized in that, In step 1, a high-frequency signal is injected into the observer in the low-speed region to obtain the rotor position estimate. and estimated rotational speed In the medium-to-high speed region, a super-helical sliding mode observer combined with an adaptive phase-locked loop is used to obtain rotor position estimates. and estimated rotational speed ; Within the observer fusion interval, the low-speed observer and the medium-to-high-speed observer operate in parallel, continuously outputting corresponding position estimates and rotational speed estimates as inputs for subsequent fusion calculations.
3. The PMSM full-velocity domain observer fusion switching method for a drop test rig with a rotating mechanism as described in claim 2, characterized in that, In step 2, the velocity fusion weights are established using the Sigmoid function, and their calculation formula is as follows: in, This refers to the real-time speed of the motor. The observer fusion center rotation speed is denoted by k; k is the slope coefficient of the Sigmoid function. Speed fusion weights The output weights for the medium- and high-speed observers are determined based on the motor speed, while the corresponding output weights for the low-speed observers are... ; When the motor speed is lower than the observer fusion center speed hour, <0.5, low-speed observer output dominates; when motor speed equals hour, =0.5, the velocity weights of the two types of observers are equal; when the motor speed is higher than hour, >0.5, the output of the medium-to-high speed observer is dominant; as the motor speed increases, As the speed increases, the output ratio of the medium- and high-speed observers gradually increases, while the output ratio of the low-speed observers gradually decreases.
4. The PMSM full-velocity domain observer fusion switching method for a drop test rig with a rotating mechanism as described in claim 3, is characterized in that, In step 3, a current stability evaluation factor is established based on the q-axis current change in adjacent sampling periods. First, the q-axis current change in adjacent sampling periods is calculated: in, This represents the q-axis current during the k-th sampling period. Let q be the q-axis current in the (k-1)th sampling period. This represents the change in q-axis current between adjacent sampling periods. Calculate the current stability evaluation factor based on the q-axis current change: in, Let be the q-axis current stability evaluation factor, and 'a' be the adjustment coefficient, where a > 0, therefore 0 < 0. ≤1. When When it increases, A decrease indicates an increase in the q-axis current fluctuation; when When decreasing, An increase that approaches 1 indicates that the q-axis current tends to stabilize. The current stability evaluation factor is used to correct the speed fusion weights.
5. The PMSM full-velocity domain observer fusion switching method for a drop test rig with a rotating mechanism according to claim 4, characterized in that, In step 4, based on the velocity fusion weights and q-axis current stability evaluation factor Calculate the final fusion weights Its expression is: in, For the final fusion weight, For speed fusion weights, This is a factor for evaluating the stability of the q-axis current. This is the fusion correction factor; When the change in q-axis current increases Decrease, Correction Term Increase, so that the final fusion weights As the q-axis current decreases, the output ratio of the medium-to-high-speed observer also decreases accordingly; when the q-axis current change decreases... The weights gradually increase and approach 1, while the correction term gradually decreases, eventually leading to the fusion weights. Gradually approaching the speed fusion weight .
6. The PMSM full-velocity domain observer fusion switching method for a drop test rig with a rotating mechanism according to claim 5, characterized in that, In step 5, based on the final fusion weights The position angles and rotational speeds output from the low-speed and medium-to-high-speed observers are weighted and fused. The formula for calculating the position angle fusion is as follows: The formula for calculating rotational speed fusion is: in, The position angle after fusion. The rotational speed after fusion. and These are the estimated position angle and rotational speed output from the low-speed observer, respectively. and These are the estimated position angle and rotational speed output from the medium-to-high speed observer, respectively. For the final fusion weight; When the final fusion weight When the weight is reduced, the output proportion of the low-speed observer in the fusion result increases; when the final fusion weight... As the speed increases, the output ratio of the medium- and high-speed observers in the fusion result increases, enabling continuous fusion of the position angle and rotational speed outputs of the two types of observers.
7. The PMSM full-velocity domain observer fusion switching method for a drop test rig with a rotating mechanism according to claim 6, characterized in that, In step 6, the fused position angles As a rotor position feedback quantity, it is used for Park transform and inverse Park transform; the fused speed As a speed feedback quantity, it is compared with a given rotational speed, and the resulting rotational speed deviation is used by the speed regulator to obtain the q-axis current reference value; the q-axis current reference value is compared with the actual q-axis current, and the resulting current deviation is used by the current regulator to obtain the q-axis voltage command, while the d-axis current is subjected to closed-loop regulation to obtain the d-axis voltage command; based on the fused position angle The d-axis and q-axis voltage commands are subjected to inverse Park transformation, and the inverter switching signal is generated by space vector pulse width modulation to control the inverter to drive the permanent magnet synchronous motor, thereby realizing the full-speed domain sensorless closed-loop control of the permanent magnet synchronous motor with rotating mechanism.
Citation Information
Patent Citations
Full-speed-domain position-sensorless vector control method for permanent magnet synchronous motor
CN115149866A
Full-speed-domain permanent magnet synchronous motor sensorless control method and system
CN117792197A
