Permanent magnet synchronous motor remote state estimation method and system based on filtering forwarding relay mechanism and adaptive anti-outlier strategy

Abstract

The invention belongs to the technical field of state estimation of a permanent magnet synchronous motor in a complex industrial environment, and discloses a permanent magnet synchronous motor remote state estimation method and system based on a filtering forwarding relay mechanism and a self-adaptive anti-outlier strategy. Permanent magnet synchronous motor measurement data collected by a sensor is locally estimated at a repeater through a filtering forwarding relay mechanism, and is transmitted to an estimation center through a communication network for fusion; adjusting a saturation function threshold value by using a self-adaptive anti-outlier strategy, and limiting an abnormal residual error within a specified range; constructing a local recursion estimator according to a motor state equation and a measurement equation of a filtering forwarding relay mechanism and a self-adaptive anti-outlier strategy; gains of a local recursion estimator are solved; and transmitting the local estimation value to a remote estimation center for sequential inertial fusion to obtain a motor state. The method is beneficial for realizing long-distance transmission of signals, reducing the adverse influence of outliers, and improving the accuracy of long-distance state estimation of the permanent magnet synchronous motor in a complex environment.

Description

Technical Field

[0001] The present invention belongs to the technical field of state estimation of permanent magnet synchronous motors, and specifically relates to a method and system for remote state estimation of permanent magnet synchronous motors based on a filtering forwarding relay mechanism and an adaptive anti-wild value strategy, which is suitable for situations where signals are abnormal and transmission is restricted in complex environments. Background Art

[0002] As industrial machinery gradually develops towards energy saving, high efficiency, high reliability and high power, permanent magnet synchronous motors (PMSM) have been widely used due to their high power density and stable operation. However, the industrial environment in which motors operate is often special, complex and harsh, with high equipment pressure and difficult maintenance. Once an accident occurs, it will not only affect the economic benefits of the enterprise, but also threaten the safety of personnel. Therefore, reliable equipment monitoring is crucial to promote the stable operation of enterprises and ensure the safety of personnel and equipment, and state estimation provides a solution to this problem.

[0003] Most sensors in complex environments such as underground coal mines are under high voltage and are more prone to aging / failure. They are also affected by strong electromagnetic radiation and humid environments, which causes some measured values ​​to deviate seriously from the true measured values, i.e., measured wild values. In order to improve the accuracy of motor state estimation, an adaptive saturation function anti-wild value mechanism is adopted to reduce the negative impact of wild values ​​by limiting the residual to a set of saturation upper bounds. Compared with the traditional fixed saturation upper bound anti-wild value method, the adaptive anti-wild value strategy can adaptively adjust the saturation upper bound according to the size of the residual, avoiding the confusion between normal values ​​and wild values ​​to a certain extent.

[0004] For example, Patent Document 1 discloses a remote state estimation method for a neural network system with limited communication. The method uses a traditional saturation function anti-outlier strategy to limit the residual to the saturation upper bound, which reduces the impact of outliers on the estimation results to a certain extent and improves the accuracy of state estimation. However, the traditional saturation function anti-outlier strategy has a fixed saturation upper bound and cannot be dynamically adjusted according to the change of residual size at different times, which limits the flexibility of the mechanism in measuring outlier detection, resulting in inaccurate state estimation results.

[0005] Patent document 2 discloses a state estimation method for a complex network system with time delay under wild value detection. The method first transforms the state equation into a system equation without time delay, and then establishes a detection method for intermittent wild values ​​based on the system equation without time delay to detect all wild values ​​of the complex network system with time delay during operation. Although Patent document 2 takes into account the influence of intermittent wild values, the detection effect of continuously existing wild values ​​is poor, and it cannot be applied to the state estimation problem of permanent magnet synchronous motors, and the flexibility of the algorithm still needs to be improved.

[0006] In addition, the problem of long-distance signal transmission in complex environments has become particularly prominent. The use of repeaters for signal transmission has attracted widespread attention, among which filter forwarding relays have attracted much attention due to their better performance and the ability to perform a local estimation of state quantities. Considering that the monitoring center is farther away from the sensor collection point in actual engineering, the present invention proposes a permanent magnet synchronous motor remote state estimation method based on a filter forwarding relay mechanism and an adaptive anti-wild value strategy from the perspective of improving the signal long-distance transmission capability under the filter forwarding relay strategy.

[0007] References Patent Document 1 China invention patent application publication number: CN111025914A, publication date: 2020.04.17; Patent document 2 Chinese invention patent application publication number: CN113315667A, publication date: 2021.08.27. Summary of the invention

[0008] The purpose of the present invention is to propose a remote state estimation method for a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-wild value strategy. This method aims at the problem of wild values ​​in the measurement data, and designs an adaptive saturation function anti-wild value strategy to reduce the impact of wild values. In addition, in view of the situation where the signal transmission distance is limited in a complex environment, the present invention designs a recursive state estimator with a filtering forwarding relay strategy to improve the accuracy of state estimation of the permanent magnet synchronous motor.

[0009] In order to achieve the above object, the present invention adopts the following technical scheme: A permanent magnet synchronous motor remote state estimation method based on a filter forwarding relay mechanism and an adaptive anti-outlier strategy includes the following steps: Step 1. Use the adaptive anti-outlier strategy to adjust the saturation function threshold and limit the abnormal residual to the specified range; Step 2. Construct a local recursive estimator based on the motor state equation and measurement equation of the filter-forward relay mechanism and the adaptive anti-outlier strategy; Step 3. Obtain the gain of the local recursive estimator; Step 4. Transmit the local estimated value to the remote estimation center for sequence fusion to obtain the motor state.

[0010] In addition, based on the permanent magnet synchronous motor remote state estimation method based on the filtering forwarding relay mechanism and the adaptive anti-wild value strategy, the present invention also proposes a permanent magnet synchronous motor remote state estimation system based on the filtering forwarding relay mechanism and the adaptive anti-wild value strategy adapted thereto, which adopts the following technical solutions: A permanent magnet synchronous motor remote state estimation system based on a filter forwarding relay mechanism and an adaptive anti-outlier strategy includes: The filter forwarding relay transmission module is used to locally estimate the permanent magnet synchronous motor measurement data collected by the sensor at the repeater and transmit it to the estimation center through the communication network for fusion; Adaptive anti-outlier module, using adaptive anti-outlier strategy to adjust the saturation function threshold and limit the abnormal residual to the specified range A local estimator building module is used to establish a system state model and a measurement model according to the characteristics of the permanent magnet synchronous motor, and to build a local recursive estimator according to the permanent magnet synchronous motor state model and the measurement model considering the filtering forwarding relay mechanism and the adaptive anti-outlier strategy; A gain calculation module, used for obtaining the gain of the local estimator; The state estimation module is used to feed the gain back to the local estimator to perform state estimation on the permanent magnet synchronous motor.

[0011] And the sequence fusion module is used to fuse the local estimation value and the local estimation error covariance to obtain the final state estimation value.

[0012] In addition, based on the method for estimating the safety status of a distribution network based on an event triggering mechanism, the present invention also proposes a computer device, which includes a memory and one or more processors.

[0013] The memory stores executable code, and when the processor executes the executable code, it is used to implement the steps of the permanent magnet synchronous motor remote state estimation method based on the filtering forwarding relay mechanism and the adaptive anti-outlier strategy mentioned above.

[0014] In addition, based on the method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-outlier strategy, the present invention also proposes a computer-readable storage medium on which a program is stored. When the program is executed by a processor, it is used to implement the steps of the method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-outlier strategy as described above.

[0015] The present invention has the following advantages: As described above, the present invention relates to a method and system for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-wild value strategy. The method of the present invention applies a recursive estimation algorithm to randomly occurring measured wild values, and the assumption that the wild values ​​appear randomly is more in line with engineering practice. The present invention adopts an adaptive anti-wild value strategy to automatically adjust the saturation upper bound according to the size of the residual. The flexibility of the saturation function anti-wild value mechanism is greatly improved. While reducing the impact of wild values, the state estimation accuracy is further improved, ensuring the safe and normal operation of permanent magnet synchronous motor equipment in complex industrial environments. In addition, the present invention considers that the long-distance transmission limitation of the signal affects the integrity of the communication data between the sensor and the remote estimation center, and proposes a filtering forwarding relay strategy and a recursive estimation method based on adaptive anti-wild value, which ensures the integrity of data transmission and reduces the adverse effects of wild values. The present invention is conducive to improving the accuracy of remote state estimation of permanent magnet synchronous motors in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The present invention is a flowchart of a method for remote state estimation of a permanent magnet synchronous motor based on a filtering and forwarding relay mechanism and an adaptive anti-outlier strategy in an embodiment of the present invention.

[0017] Figure 2 It is a flowchart of the adaptive anti-outlier strategy and filtering forwarding relay mechanism in an embodiment of the present invention.

[0018] Figure 3 Schematic diagram of the original measured value of the d-axis current and the measured value affected by the wild value.

[0019] Figure 4 Schematic diagram of the original measured value of the q-axis current and the measured value affected by the wild value.

[0020] Figure 5 Schematic diagram of the original measured value of the motor speed and the measured value affected by the wild value.

[0021] Figure 6 Schematic diagram of the estimated values ​​based on recursive estimation of adaptive and traditional anti-outlier strategies.

[0022] Figure 7 Schematic diagram of the root mean square error of recursive estimation based on adaptive and traditional anti-outlier strategies. DETAILED DESCRIPTION

[0023] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments: Example 1

[0024] This embodiment 1 describes a method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-outlier strategy, such as Figure 1As shown, the permanent magnet synchronous motor remote state estimation method based on the filtering forwarding relay mechanism and the adaptive anti-outlier strategy includes the following steps: Step 1. Use the adaptive anti-outlier strategy to adjust the saturation function threshold and limit the abnormal residual to the specified range.

[0025] The present invention proposes an adaptive saturation function anti-outlier strategy, the principle of which is as follows: For any vector is a saturation function, defined as (1) (2) in, is an adaptive saturation upper bound that satisfies the following conditions: (3) in, (4) is a scalar with an initial value of , , represents the weight coefficient of the saturation threshold at the previous moment; , represents the coefficient of the saturation threshold adaptively adjusted according to the residual size, and Represents the weights of different local estimator saturation thresholds.

[0026] In order to solve the problem of insufficient flexibility of the saturation function anti-outlier mechanism, the present invention introduces an adaptive adjustment parameter , and , so that the saturation upper bound can be adaptively adjusted according to the residual size, which greatly improves the flexibility of the adaptive anti-outlier mechanism.

[0027] Step 2. Construct a local recursive estimator based on the motor state equation and measurement equation of the filter-forward relay mechanism and the adaptive anti-outlier strategy; Considering that long-distance transmission in complex industrial environments damages the integrity of communication data between sensors and remote estimation centers, the present invention uses a filtering forwarding relay strategy to perform a one-step local prediction on the measurement data collected by the sensor, and amplifies and forwards it to the remote estimation center.

[0028] The system state model is shown in formula (5); (5) in, , Respectively k +1 moment, k The state variables at the moment, is defined as , and are the components of the stator current along the d-axis and q-axis respectively, is the motor mechanical angular velocity; is the state transition function, is defined as , and are the components of the stator voltage along the d-axis and q-axis respectively; represents unknown but bounded process noise, with mean zero and variance Gaussian distribution.

[0029] The measurement data includes the motor stator d-axis current, q-axis current and motor mechanical angular velocity; the measurement data is defined as: (6) in, For the The input of the repeater, , n represents the total number of repeaters; the measured output of the permanent magnet synchronous motor is expressed as: (7) in, is the measurement transfer coefficient matrix, is the measurement noise, which has a mean of zero and a variance of Gaussian distribution.

[0030] For Repeaters, design a local recursive estimator with the following structure: (8) (9) in, and are the one-step predicted value and estimated value at time k+1 respectively, Indicator measurement With pre-measurement measurement The difference between is the local estimator gain at time k+1 to be designed, Represents the random uncertainty transformation of the filter gain, satisfying: (10) in, and is a real matrix of known appropriate dimension, It is a multiplicative noise with zero mean and unit covariance, which is uncorrelated with other noise signals; Will Multiply by the estimator gain As the correction amount, the state prediction amount is corrected to obtain the corrected estimate As the final local estimate, the measurement data collected by the sensor is processed by local filtering at the repeater, and then amplified and forwarded to the remote state estimation center.

[0031] Step 3. Obtain the gain of the local recursive estimator; Step 3.1. Transform the nonlinear state transfer function Linearization.

[0032] The nonlinear measurement function About state estimates The Taylor expansion is as follows: (11) in, Represents the state transition function The estimated value at time k, Represents the true value of the state at time k With state estimate The difference is called the estimation error. is the Jacobian matrix at time k, , is a known matrix of suitable dimension, a time-varying matrix satisfy ; Represents the identity matrix with appropriate dimensions.

[0033] Step 3.2. Establishing the local prediction error and local estimation error The specific expression of .

[0034] According to the system state model and measurement model in formula (5) and formula (7), we can get k +1 time local prediction error and local estimation error The specific expression is as follows: (20) (twenty one) in, Represents the coefficients obtained after the adaptive saturation function processes the residual.

[0035] Step 3.3. Establish the local prediction error covariance and the local estimation error covariance The specific expression of .

[0036] According to the prediction error and estimation error in formula (20) and formula (21), we can get k Local prediction error covariance at time +1 and the local estimation error covariance The specific expression is as follows: (twenty two) (twenty three) in, (twenty four) (25) (26) (27) (28) (29) Step 3.4. By minimizing the local estimation error covariance To obtain the local estimator gain .

[0037] Local estimation error covariance The upper bound of is: (30) in: (31) (32) Let the local estimation error covariance upper bound be Gain Taking partial derivatives we get: (33) Minimize the upper bound of the local estimation error covariance and obtain the estimator gain for: (34) in: (35) (36) The obtained gain of the local estimator is fed back to the local recursive estimator to perform local state estimation on the permanent magnet synchronous motor.

[0038] Step 4. Transmit the local estimated value to the remote estimation center for sequence fusion to obtain the motor state.

[0039] For the i-th fusion state value and estimated covariance, they are defined as: (37) (38) in, , which means that for a system with N nodes, N-1 fusions are performed; is such that the performance index The minimized weight coefficient is: (39) The initial state estimate is and the initial covariance They are defined as: , ; Finally, the permanent magnet synchronous motor fusion estimation value and estimation error covariance are obtained as follows: , .

[0040] like Figure 2 As shown, the sensor device collects quantity measurements from the permanent magnet synchronous motor in a complex environment. The obtained quantity measurements need to be processed by an adaptive anti-outlier mechanism and locally estimated and transmitted through a filtering forwarding relay.

[0041] If the current residual meets the saturation function restriction condition, that is, the measured data is normal at this time, the residual size remains unchanged; if the current residual exceeds the saturation upper limit, that is, the measured data is a wild value at this time, the current residual is limited to the adaptive saturation upper limit. Then, a local estimation is performed at the repeater, and the local estimation value is transmitted to the remote estimation center through the communication network for sequential fusion processing to obtain the final system state estimation value.

[0042] Figure 3-Figure 5 The influence of the measured wild value on the d-axis current, q-axis current and rotational speed is shown respectively.

[0043] Depend on Figure 3-Figure 5 It is not difficult to see that the randomly occurring measurement wild values ​​cause serious pollution to the measurement data of the permanent magnet synchronous motor. If not processed, it will greatly affect the estimation performance and even cause the filter to fail to converge.

[0044] Figure 6 and Figure 7 The root mean square error between the state estimation value and the true state value under the adaptive anti-outlier strategy and the traditional anti-outlier strategy, as well as the recursive state estimation based on the adaptive anti-outlier strategy and the traditional anti-outlier strategy are shown respectively.

[0045] Depend on Figure 6 and Figure 7 It is not difficult to see from the comparison that the adaptive anti-wilderness strategy can adaptively adjust the saturation threshold. Therefore, when the load torque increases and causes the permanent magnet synchronous motor torque to change, the proposed algorithm significantly reduces the estimation error compared with the traditional anti-wilderness strategy. This further verifies that the proposed recursive estimator based on the filtering forwarding relay mechanism and the adaptive anti-wilderness strategy can achieve high state estimation accuracy while realizing long-distance data transmission. Example 2

[0046] This embodiment 2 describes a permanent magnet synchronous motor remote state estimation system based on a filtering forwarding relay mechanism and an adaptive anti-wild value strategy. This system is based on the same inventive concept as the permanent magnet synchronous motor remote state estimation method based on a filtering forwarding relay mechanism and an adaptive anti-wild value strategy in the above-mentioned embodiment 1.

[0047] Specifically, the permanent magnet synchronous motor remote state estimation system based on the filtering forwarding relay mechanism and the adaptive anti-outlier strategy includes: The filter forwarding relay transmission module is used to locally estimate the permanent magnet synchronous motor measurement data collected by the sensor at the repeater and transmit it to the estimation center through the communication network for fusion; Adaptive anti-outlier module, using adaptive anti-outlier strategy to adjust the saturation function threshold and limit the abnormal residual to the specified range A local estimator building module is used to establish a system state model and a measurement model according to the characteristics of the permanent magnet synchronous motor, and to build a local recursive estimator according to the permanent magnet synchronous motor state model and the measurement model considering the filtering forwarding relay mechanism and the adaptive anti-outlier strategy; A gain calculation module, used for obtaining the gain of the local estimator; The state estimation module is used to feed the gain back to the local estimator to perform state estimation on the permanent magnet synchronous motor.

[0048] And the sequence fusion module is used to fuse the local estimation value and the local estimation error covariance to obtain the final state estimation value.

[0049] It should be noted that in the permanent magnet synchronous motor remote state estimation system based on the filtering forwarding relay mechanism and the adaptive anti-wild value strategy, the implementation process of the functions and effects of each functional module is specifically described in the implementation process of the corresponding steps in the method in the above embodiment 1, and will not be repeated here. Example 3

[0050] This embodiment 3 describes a computer device, which is used to implement the steps of the distribution network security status estimation method based on the event trigger mechanism described in the above embodiment 1.

[0051] The computer device includes a memory and one or more processors. The memory stores executable codes, and when the processor executes the executable codes, the steps of the permanent magnet synchronous motor remote state estimation method based on the filtering forwarding relay mechanism and the adaptive anti-wild value strategy are implemented.

[0052] In this embodiment, the computer device is any device or apparatus with data processing capability, which will not be described in detail here. Example 4

[0053] This embodiment 4 describes a computer-readable storage medium, which is used to implement the steps of the permanent magnet synchronous motor remote state estimation method based on the filtering forwarding relay mechanism and the adaptive anti-wild value strategy described in the above embodiment 1.

[0054] The computer-readable storage medium in this embodiment 4 stores a program thereon, which, when executed by a processor, is used to implement the steps of a method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-wild value strategy.

[0055] The computer-readable storage medium may be an internal storage unit of any device or apparatus with data processing capabilities, such as a hard disk or memory, or an external storage device of any device with data processing capabilities, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc., equipped on the device.

[0056] Of course, the above description is only a preferred embodiment of the present invention, and the present invention is not limited to the above embodiments. It should be noted that all equivalent substitutions and obvious deformation forms made by any technician familiar with the field under the guidance of this specification fall within the essential scope of this specification and should be protected by the present invention.

Claims

1. A permanent magnet synchronous motor remote state estimation method based on a filter forwarding relay mechanism and an adaptive anti-outlier strategy, characterized in that: The steps include: Step 1. Use the adaptive anti-outlier strategy to adjust the saturation function threshold and limit the abnormal residual to the specified range; Step 2. Construct a local recursive estimator based on the motor state equation and measurement equation of the filter-forward relay mechanism and the adaptive anti-outlier strategy; Step 3. Obtain the gain of the local recursive estimator; Step 4. Transmit the local estimated value to the remote estimation center for sequential fusion to obtain the motor state.

2. The method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-outlier strategy according to claim 1 is characterized in that: In step 1, the adaptive anti-outlier mechanism is as follows: For any vector , is a saturation function, defined as (1) （2） in, is an adaptive saturation upper bound that satisfies the following conditions: （3） in, （4） is a scalar with an initial value of , , represents the weight coefficient of the saturation threshold at the previous moment; , represents the coefficient of the saturation threshold adaptively adjusted according to the residual size, and Represents the weights of different local estimator saturation thresholds.

3. The method for remote state estimation of a permanent magnet synchronous motor based on a filtering and forwarding relay mechanism and an adaptive anti-outlier strategy according to claim 2 is characterized in that: In step 2, the system state model is as shown in formula (5); （5） in, , Respectively k +1 moment, k The state variables at the moment, is defined as , and are the components of the stator current along the d-axis and q-axis respectively, is the motor mechanical angular velocity; is the state transition function, is defined as , and are the components of the stator voltage along the d-axis and p-axis respectively; represents unknown but bounded process noise, with mean zero and variance Gaussian distribution.

4. The method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-outlier strategy according to claim 3 is characterized in that: In step 2, the measured data includes the motor stator d-axis current, q-axis current and motor mechanical angular velocity; the measured data is defined as: （6） in, For the The input of the repeater, , n represents the total number of repeaters; the measured output of the permanent magnet synchronous motor is expressed as: （7） in, is the measurement transfer coefficient matrix, is the measurement noise, which has a mean of zero and a variance of Gaussian distribution.

5. The method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-outlier strategy according to claim 4 is characterized in that: In step 2, for Repeaters, design a local recursive estimator with the following structure: （8） （9） in, and are the one-step predicted value and estimated value at time k+1 respectively, Indicator measurement With pre-measurement measurement The difference between is the local estimator gain at time k+1 to be designed, Represents the random uncertainty transformation of the filter gain, satisfying: （10） in, and is a real matrix of known appropriate dimension, It is a multiplicative noise with zero mean and unit covariance, which is uncorrelated with other noise signals; Will Multiply by the estimator gain As a correction amount, the state prediction amount Correction is performed to obtain the corrected estimate As the final local estimate, the measurement data collected by the sensor is processed by local filtering at the repeater, and then amplified and forwarded to the remote state estimation center.

6. The method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-outlier strategy according to claim 5 is characterized in that: The step 3 is specifically as follows: Step 3.

1. Transform the nonlinear state transfer function Linearization; The nonlinear measurement function About state estimates The Taylor expansion is as follows: （11） in, Represents the state transition function The estimated value at time k, Represents the true value of the state at time k With state estimate The difference is called the estimation error. is the Jacobian matrix at time k, , is a known matrix of suitable dimension, a time-varying matrix satisfy ; represents the identity matrix with appropriate dimensions; Step 3.

2. Establishing the local prediction error and local estimation error Specific expression of According to the system state model and measurement model in formula (5) and formula (7), we can get k +1 time local prediction error and local estimation error The specific expression is as follows: （20） （21） in, Represents the coefficient obtained after the adaptive saturation function processes the residual; Step 3.

3. Establish the local prediction error covariance and the local estimation error covariance Specific expression of According to the prediction error and estimation error in formula (20) and formula (21), we can get k Local forecast error covariance at time +1 and the local estimation error covariance The specific expression is as follows: （22） （23） in, （24） （25） （26） （27） （28） （29） Step 3.

4. By minimizing the local estimation error covariance To obtain the local estimator gain ; Local estimation error covariance The upper bound of is: （30） in: （31） （32） Let the local estimation error covariance upper bound be Gain Taking partial derivatives we get: （33） Minimize the upper bound of the local estimation error covariance and obtain the estimator gain for: （34） in: （35） （36） The obtained gain of the local estimator is fed back to the local recursive estimator to perform local state estimation on the permanent magnet synchronous motor.

7. The method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-outlier strategy according to claim 6 is characterized in that: In step 4, for The fusion state value and estimated covariance are defined as: （37） （38） in, , which means that for a system with N nodes, N-1 fusions are performed; is such that the performance index The minimized weight coefficient is: （39） The initial state estimate is and the initial covariance They are defined as: ， ； Finally, the permanent magnet synchronous motor fusion estimation value and estimation error covariance are obtained as follows: ， 。 8. The method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-outlier strategy according to claim 7 is characterized in that: include: The filter forwarding relay transmission module is used to locally estimate the permanent magnet synchronous motor measurement data collected by the sensor at the repeater and transmit it to the estimation center through the communication network for fusion; Adaptive anti-outlier module, using adaptive anti-outlier strategy to adjust the saturation function threshold and limit the abnormal residual to the specified range A local estimator building module is used to establish a system state model and a measurement model according to the characteristics of the permanent magnet synchronous motor, and to build a local recursive estimator according to the permanent magnet synchronous motor state model and the measurement model considering the filtering forwarding relay mechanism and the adaptive anti-outlier strategy; A gain calculation module, used for obtaining the gain of the local estimator; The state estimation module is used to bring the gain back to the local estimator to perform state estimation on the permanent magnet synchronous motor, and the sequence-inertia fusion module is used to fuse the local estimation value and the local estimation error covariance to obtain the final state estimation value.

9. A permanent magnet synchronous motor remote state estimation system, comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that: When the processor executes the executable code, the steps of the method for remote state estimation of a permanent magnet synchronous motor based on a filtering forwarding relay mechanism and an adaptive anti-outlier strategy as described in any one of claims 1 to 8 are implemented.

Images