Method for suppressing thrust fluctuation of linear motor based on acceleration hybrid model

Abstract

The application provides a linear motor thrust fluctuation suppression method based on an acceleration hybrid model, and relates to the technical field of motor control. The method comprises the following steps: firstly, measuring a linear motor to obtain structural parameters of the linear motor, taking motor displacement as model input, taking thrust fluctuation as model output, establishing a thrust fluctuation mechanism model of end effect and slot effect, considering normal force fluctuation generated by the end effect and the slot effect, establishing a friction force mechanism model, collecting thrust data by using a sensor to obtain a thrust fluctuation data model, improving the thrust fluctuation suppression effect on the basis of suppressing the thrust fluctuation by using the mechanism model, and enhancing the connection between a control algorithm and structural parameters of the motor.

Description

Technical Field

[0001] This invention relates to the field of motor control technology, and in particular to a method for suppressing thrust fluctuations in linear motors based on an acceleration hybrid model. Background Technology

[0002] Precise testing of thrust fluctuation in linear motors is a crucial prerequisite for guiding motor design and control. Thrust fluctuations can cause vibration and noise, and at low speeds, the motor may resonate, deteriorating its operating characteristics. Therefore, accurately detecting the thrust fluctuation characteristics of linear motors is essential.

[0003] Currently, the observation of thrust fluctuations in linear motors employs data modeling methods. These models often utilize drag-and-drop experiments to obtain positioning force data. However, the torque fluctuations inherent in the rotating motor itself, as well as the transmission efficiency and accuracy of the connecting mechanism, significantly impact the detection accuracy during drag-and-drop experiments. Using an accelerometer to acquire thrust fluctuation data can avoid the influence of the rotating motor.

[0004] The data model contains high-frequency components, and the control algorithm and motor structural parameters fail to form an effective connection. Using a mechanistic model to describe the thrust fluctuation of the linear motor can solve the problems of the data model, but the mechanistic model itself suffers from inaccurate modeling. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to address the shortcomings of the prior art by providing a method for suppressing thrust fluctuations in linear motors based on an acceleration hybrid model. This method can solve the problem of the disconnect between the control algorithm and the motor mechanism, and also improve the measurement accuracy of the model.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A method for suppressing thrust fluctuations in a linear motor based on an acceleration hybrid model includes the following steps:

[0008] Step 1: Collect the structural parameters of the linear motor. Based on the structural parameters of the linear motor, establish a thrust fluctuation mechanism model, as shown in the following formula:

[0009] F j =F d +F c +F m

[0010] Among them, F j For the mechanistic model of thrust fluctuation, F d For the end magnetic resistance, F c For the cogging magnetic resistance, F m Friction;

[0011] Step 2: The suppression method uses current injection. The compensation current is obtained based on the linear motor thrust equation, and its formula is as follows:

[0012]

[0013] Among them, i qj Thrust fluctuation mechanism model compensation current, n p Let τ be the number of pole pairs of the linear motor, τ be the pole pitch length, and ψ be the number of pole pairs. PM For permanent magnet flux linkage;

[0014] Step 3: Thrust fluctuation is calculated using the electromagnetic thrust suppression mechanism model, and the formula is as follows:

[0015] F em +F j =0

[0016] Among them, F em Electromagnetic thrust to suppress thrust fluctuations in the mechanism model;

[0017] Step 4: After compensation by the thrust fluctuation mechanism model, thrust data is collected using sensors. Since the thrust fluctuation of the linear motor is a periodic function of position, the data is processed and fitted using Fourier series to establish a thrust fluctuation data model.

[0018] Step 5: Calculate the compensation current of the thrust fluctuation data model established after compensation by the thrust fluctuation mechanism model. The formula is as follows:

[0019]

[0020] Among them, F s For data model thrust fluctuations;

[0021] Step 6: Calculate the corresponding thrust fluctuation using the electromagnetic thrust suppression data model. The formula is as follows:

[0022] F em '+F s =0

[0023] Among them, F em 'Electromagnetic thrust to suppress thrust fluctuations in the data model.'

[0024] Furthermore, the thrust fluctuation mechanism model in step 1 includes the end effect mechanism model, the tooth groove effect mechanism model, and the friction mechanism model;

[0025] The formula for the end effect mechanism model is as follows:

[0026]

[0027] Among them, Kc δ is the air gap coefficient, and δ is the air gap length. The maximum magnetic flux passing through the longitudinal end edge of the mover core, μ0 is the free permeability, k1 is the flux compression coefficient, τ is the pole gap length, and l ef Where λ is the equivalent magnetic circuit length, x is the mover position, λ is the difference between the mover length and the multiple of the pole pitch, and n = 1, 2, 3, ...;

[0028] The formula for the tooth cogging effect mechanism model is as follows:

[0029]

[0030] Where z is the number of motor slots, h PM The height is the magnetization direction of the permanent magnet, p is the number of motor poles directly opposite the armature core length, and B is the height of the permanent magnet. rn λ represents the amplitude of the harmonic component of the permanent magnet remanent density. k The amplitudes of the relative air gap permeability harmonic components are given by k,n = 1, 2, 3, ...;

[0031] The formula for the friction mechanism model is as follows:

[0032]

[0033] Among them, f c For Coulomb friction, f m For the maximum static friction force, For the motioner velocity, The coefficient of lubrication. For the sign function, k v is the coefficient of viscous friction.

[0034] Furthermore, the formula for the maximum static friction force in the friction mechanism model is as follows:

[0035] f m =μF N

[0036] Where μ is the static friction coefficient; F N For positive pressure, its formula is as follows:

[0037] F N =F ds +F cs +F z

[0038] Among them, F ds For the end effect normal force, F cs For the cogging effect normal force, F z The force is the weight of the motor element itself.

[0039] Furthermore, the end effect normal force F dsThe mechanistic model formula is as follows:

[0040]

[0041] Among them, F dsL F is the normal force caused by the left end effect. dsR The normal force is caused by the right-end effect; n = 1, 2, 3, ...;

[0042] Cogging effect normal force F cs The mechanistic model formula is as follows:

[0043]

[0044] Where λ0 is the DC component of the relative air gap permeability, B r0 Let k,n be the DC component of the remanent magnetization of the permanent magnet, where k,n = 1, 2, 3, ...

[0045] Furthermore, the end effect normal force F ds In the mechanistic model, when λ=0, the normal force caused by the left-end effect is equal to the normal force caused by the right-end effect, and the net force received by the mover is twice the single-end normal force, as shown in the following formula:

[0046] F ds =F dsL +F dsR

[0047] When λ=τ / 2, the odd harmonic phases of the wave forces at the left and right ends are opposite, and the mover has a tendency for "longitudinal pitching motion". Therefore, the influence of the upper surface of the motor guide rail on the friction force needs to be considered. The mechanism model of static friction force is as follows:

[0048] f m =μ1(F dsL +F cs +F z )+μ2F dsR

[0049] Among them, f m μ1 represents the maximum static friction force, μ2 represents the static friction coefficient of the lower surface of the motor guide rail, and μ2 represents the static friction coefficient of the upper surface of the motor guide rail.

[0050] Furthermore, in step 4, the sensor is a force sensor or an acceleration sensor;

[0051] When the sensor is a force sensor, the thrust fluctuation data of the linear motor is obtained through the uniform velocity method. The data is then fitted, and a linear motor thrust fluctuation data model is established. The formula is as follows:

[0052]

[0053] Among them, A i The amplitude of the frequency component. The phase of the frequency component;

[0054] When the sensor is an acceleration sensor, the acceleration of the linear motor's mover is collected. An acceleration data model is used to fit the collected acceleration data, and then a linear motor thrust fluctuation data model is established. The formula is as follows:

[0055] F s =ma

[0056] Where m is the mass of the mover and a is the acceleration of the mover.

[0057] Furthermore, when the sensor is an accelerometer, the acceleration model is composed of Fourier series combinations, and the acceleration model structure is as follows:

[0058]

[0059] Calculate the fundamental frequency, which is the ratio of motor speed to pole pitch:

[0060]

[0061] Where f is the fundamental frequency and v is the mover velocity;

[0062] FFT analysis was performed on the acceleration data to obtain its frequency response plot. The remaining frequency components were obtained from the frequency response plot. All frequency points were then incorporated into the acceleration model and fitted using the least squares method to obtain the fitted acceleration data model.

[0063] Furthermore, the acceleration model needs to perform a second integral on the acceleration to obtain position information, thus obtaining the relationship between acceleration and position, and further obtaining the relationship between thrust fluctuation and mover position, as shown in the following formula:

[0064]

[0065]

[0066] Where V(t) is velocity, S(t) is displacement, a(t) is acceleration, and a j Let v be the acceleration value at time j. j Let be the velocity value at time j, and Δt be the sampling period.

[0067] Furthermore, in the process of obtaining location information, if there are error factors that may affect the calculation results, the Kalman filter algorithm is used to weaken the impact of errors, and its formula is as follows:

[0068] State prediction:

[0069] xt =F t x t-1 +B t a t

[0070] Where, x t Let F be the state vector. t Let B be the state transition matrix. t For the control matrix, a t Accelerate for the current moment;

[0071] Predicted Observations:

[0072] Z t =Hx t

[0073] Among them, Z t Let H be the observation vector, H be the observation matrix, and x be the observation vector. t It is a state vector;

[0074] Covariance prediction:

[0075] P t - =F t P t-1 F t T +Q

[0076] Among them, P t - P is the inverse of the covariance matrix. t-1 Let F represent the covariance matrix at the previous time step. t Let Q be the state transition matrix, and let Q be the covariance matrix of the process noise.

[0077] Update Gains:

[0078] K t =P t - H T HP t - H T +R) -1

[0079] Among them, K t Let R be the Kalman gain matrix, and R be the covariance matrix of the observation noise.

[0080] Update status:

[0081] X t =x t +K t (Z t -Hx t )

[0082] Update the covariance matrix:

[0083]

[0084] Among them, P t Let be the covariance matrix.

[0085] Furthermore, the state prediction uses acceleration as the object motion control variable to express the relationship between the input state and the output state, and its formula is as follows:

[0086]

[0087] Where, d t v represents the current position. t Let a be the velocity at the current moment. t Acceleration at the current moment.

[0088] The beneficial effects of adopting the above technical solution are as follows: The linear motor thrust fluctuation suppression method based on an acceleration hybrid model provided by this invention utilizes an acceleration sensor to obtain thrust fluctuation data and establishes a thrust fluctuation data model. This avoids the influence of the rotating motor in traditional thrust fluctuation testing, improves the accuracy of the data model, and provides a new solution for detecting thrust fluctuations. By deriving the linear motor friction mechanism model, the relationship between motor structural parameters and friction force is described, filling the gap in linear motor friction mechanism models. Simultaneously, the mechanism model can compensate for the high-frequency component problem of the data model, improve the model's measurement accuracy, and reduce the computational cost of the data model. Combining the linear motor thrust fluctuation mechanism model with the data model improves the thrust fluctuation observation accuracy and provides a new solution for thrust fluctuation suppression algorithms. This not only solves the problem of the disconnect between the control algorithm and the motor mechanism but also improves the model's measurement accuracy. Attached Figure Description

[0089] Figure 1 A flowchart of a linear motor thrust fluctuation suppression method based on an acceleration hybrid model provided in an embodiment of the present invention;

[0090] Figure 2 A block diagram illustrating the connection between the mechanism model and the data model provided in this embodiment of the invention;

[0091] Figure 3 The control block diagram is provided for the linear motor thrust fluctuation suppression method in an embodiment of the present invention. Detailed Implementation

[0092] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0093] like Figure 1 As shown, the method of this embodiment is described below.

[0094] Step 1: Collect the structural parameters of the linear motor. Based on the structural parameters of the linear motor, establish a thrust fluctuation mechanism model, as shown in the following formula:

[0095] F j =F d +F c +F m

[0096] Among them, F j For the mechanistic model of thrust fluctuation, F d For the end magnetic resistance, F c For the cogging magnetic resistance, F m This is friction.

[0097] Factors contributing to thrust fluctuations include end effect, cogging effect, and friction. Therefore, thrust fluctuation mechanism models include end effect mechanism models, cogging effect mechanism models, and friction mechanism models. The formula for the end effect mechanism model is as follows:

[0098]

[0099] Among them, K c δ is the air gap coefficient, and δ is the air gap length. The maximum magnetic flux passing through the longitudinal end edge of the mover core, μ0 is the free permeability, k1 is the flux compression coefficient, τ is the pole gap length, and l ef λ is the equivalent magnetic circuit length, x is the mover position, λ is the difference between the mover length and the multiple of the pole pitch, and n = 1, 2, 3, ...

[0100] The formula for the tooth cogging effect mechanism model is as follows:

[0101]

[0102] Where z is the number of motor slots, h PM The height is the magnetization direction of the permanent magnet, p is the number of motor poles directly opposite the armature core length, and B is the height of the permanent magnet. rn λ represents the amplitude of the harmonic component of the permanent magnet remanent density. k The amplitudes of the relative air gap magnetic permeability harmonic components are given by k,n = 1, 2, 3, ...

[0103] The formula for the friction mechanism model is as follows:

[0104]

[0105] Among them, f c For Coulomb friction, f m For the maximum static friction force, For the motioner velocity, The coefficient of lubrication. For the sign function, k v The coefficient of viscous friction is given. The maximum static friction force is given by f. m The formula is as follows:

[0106] f m =μF N

[0107] Where μ is the static friction coefficient; F N For positive pressure, its formula is as follows:

[0108] F N =F ds +F cs +F z

[0109] Among them, F ds For the end effect normal force, F cs For the cogging effect normal force, F z The force is the weight of the motor element itself.

[0110] End effect normal force F ds The mechanistic model formula is as follows:

[0111]

[0112] Among them, F dsL F is the normal force caused by the left end effect. dsR The normal force is caused by the right-end effect; n = 1, 2, 3, ...

[0113] End effect normal force F ds In the mechanistic model, when λ=0, the normal force caused by the left-end effect is equal to the normal force caused by the right-end effect, and the net force received by the mover is twice the single-end normal force, as shown in the following formula:

[0114] F ds =F dsL +F dsR

[0115] When λ=τ / 2, the odd harmonic phases of the wave forces at the left and right ends are opposite, and the mover has a tendency for "longitudinal pitching motion". Therefore, the influence of the upper surface of the motor guide rail on the friction force needs to be considered. The mechanism model of static friction force is as follows:

[0116] f m =μ1(F dsL +F cs +F z )+μ2F dsR

[0117] Among them, f mμ1 represents the maximum static friction force, μ2 represents the static friction coefficient of the lower surface of the motor guide rail, and μ2 represents the static friction coefficient of the upper surface of the motor guide rail.

[0118] Cogging effect normal force F cs The mechanistic model formula is as follows:

[0119]

[0120] Where λ0 is the DC component of the relative air gap permeability, B r0 Let k,n be the DC component of the remanent magnetization of the permanent magnet, where k,n = 1, 2, 3, ...

[0121] By deriving the friction mechanism model of a linear motor, the relationship between motor structural parameters and friction force is described, filling a gap in the modeling of friction mechanism in linear motors. Furthermore, the mechanism model can address the issue of high-frequency components in the data model, reducing its computational cost.

[0122] In motor parameters, δ, τ, h PM z, l, and L can be obtained by measuring the motor structure, and μ0 and k v B is a constant. rn , λ k l ef K c k1 can be derived from the motor structure parameters.

[0123] Step 2: The suppression method uses current injection. The compensation current is obtained based on the linear motor thrust equation, and its formula is as follows:

[0124]

[0125] Among them, i qj Thrust fluctuation mechanism model compensation current, F j For the thrust fluctuation in the mechanism model, n p Let τ be the number of pole pairs of the linear motor, τ be the pole pitch length, and ψ be the number of pole pairs. PM It is a permanent magnet flux linkage.

[0126] Step 3: Thrust fluctuation is calculated using the electromagnetic thrust suppression mechanism model, and the formula is as follows:

[0127] F em +F j =0

[0128] Among them, F j For the mechanistic model of thrust fluctuation, F em Electromagnetic thrust to suppress thrust fluctuations in the mechanism model;

[0129] Step 4: After compensation using the thrust fluctuation mechanism model, thrust data is collected using sensors. Since the thrust fluctuation of the linear motor is a periodic function of position, the data is processed and fitted using Fourier series to establish a data model for the linear motor thrust fluctuation. The sensors used to collect the thrust data are force sensors or acceleration sensors.

[0130] When the sensor is a force sensor, the thrust fluctuation data of the linear motor is obtained through the uniform velocity method. The data is then fitted, and a linear motor thrust fluctuation data model is established. The formula is as follows:

[0131]

[0132] Among them, F s For the thrust fluctuations in the data model, A i The amplitude of the frequency component. The phase of the frequency component.

[0133] The sensor can also be an accelerometer to collect the acceleration of the linear motor's mover. An acceleration data model can then be used to fit the collected acceleration data. Since the thrust fluctuation of the linear motor is directly proportional to the mover acceleration, a data model for the thrust fluctuation of the linear motor can be established, with the following formula:

[0134] F s =ma

[0135] Where m is the mass of the mover and a is the acceleration of the mover.

[0136] Since the thrust fluctuation mechanism model uses position as input, the thrust fluctuation data model also needs to use position as input to facilitate the establishment of a hybrid model of linear motor thrust fluctuation mechanism data.

[0137] When the sensor is an accelerometer, the acceleration model is composed of a combination of Fourier series. Let the structure of the acceleration model be:

[0138]

[0139] Where a is the motioner acceleration, A i Let τ be the amplitude of the frequency component, and τ be the pole spacing. The phase of the frequency component;

[0140] Calculate the fundamental frequency, which is the ratio of motor speed to pole pitch:

[0141]

[0142] Where f is the fundamental frequency and v is the mover velocity;

[0143] FFT analysis was performed on the acceleration data to obtain its frequency response plot. The remaining frequency components were obtained from the frequency response plot. All frequency points were then incorporated into the acceleration model and fitted using the least squares method to obtain the fitted acceleration data model.

[0144] By performing a second integral on the acceleration, position information can be obtained, revealing the relationship between acceleration and position, and further, the relationship between thrust fluctuation and mover position, as shown in the following formula:

[0145]

[0146]

[0147] Where V(t) is velocity, S(t) is displacement, a(t) is acceleration, and a j Let v be the acceleration value at time j. j Let be the velocity value at time j, and Δt be the sampling period.

[0148] Measuring the thrust fluctuation of a linear motor using an accelerometer suffers from measurement errors such as zero-point drift and high-frequency noise. A Kalman filter algorithm is employed to mitigate these errors; the formula is as follows:

[0149] State prediction:

[0150] x t =F t x t-1 +B t a t

[0151] Where, x t Let F be the state vector. t Let B be the state transition matrix. t For the control matrix, a t The acceleration at the current moment;

[0152] Predicted Observations:

[0153] Z t =Hx t

[0154] Among them, Z t Let H be the observation vector, H be the observation matrix, and x be the observation vector. t It is a state vector;

[0155] Covariance prediction:

[0156] P t - =F t P t-1 F t T +Q

[0157] Among them, P t-1 Let P represent the covariance matrix at the previous time step. t - The inverse of the covariance prediction matrix, F t Let Q be the state transition matrix, and let Q be the covariance matrix of the process noise.

[0158] Update Gains:

[0159] K t =P t - H T HP t - H T +R) -1

[0160] Among them, K t Let R be the Kalman gain matrix, and R be the covariance matrix of the observation noise.

[0161] Update status:

[0162] X t =x t +K t (Z t -Hx t )

[0163] Update the covariance matrix:

[0164] P t =(IK t H)P t -

[0165] Among them, P t Let be the covariance matrix.

[0166] State prediction uses acceleration as the control variable for object motion, expressing the relationship between the input state and the output state. The formula is as follows:

[0167]

[0168] Where, d t v represents the current position. t Let a be the velocity at the current moment. t Let Δt be the acceleration at the current moment, and Δt be the sampling period.

[0169] Step 5: Calculate the compensation current of the thrust fluctuation data model established after compensation by the thrust fluctuation mechanism model. The formula is as follows:

[0170]

[0171] Among them, Fs For the thrust fluctuation data model, n p Let τ be the number of pole pairs of the linear motor, τ be the pole pitch length, and ψ be the number of pole pairs. PM For permanent magnet flux linkages. The connection between the mechanistic model and the data model is as follows: Figure 2 As shown.

[0172] Step 6: Calculate the corresponding thrust fluctuation using the electromagnetic thrust suppression data model. The formula is as follows:

[0173] F em '+F s =0

[0174] Among them, F s For the thrust fluctuations in the data model, F em 'Electromagnetic thrust to suppress thrust fluctuations in the data model.'

[0175] The control block diagram of the linear motor thrust fluctuation suppression method in this embodiment is as follows: Figure 3 As shown, the method of obtaining a compensation current by observing thrust fluctuations and using the electromagnetic thrust obtained by the compensation current to suppress thrust fluctuations has the advantages of being easy to implement and simple.

[0176] This invention derives a mechanism model for thrust fluctuation disturbance components, thereby describing the relationship between motor structural parameters and thrust fluctuation, filling a gap in friction mechanism models and providing a new basis for motor design. Furthermore, the design of a mechanism-data hybrid model achieves the complementary advantages of both models, offering a new approach to thrust fluctuation suppression control algorithms.

[0177] Because existing technologies inevitably introduce the influence of rotating motors in thrust fluctuation detection, affecting the accuracy of data models, using accelerometers can avoid the influence of rotating motors and improve the measurement accuracy of data models. However, the data model is not linked to the structural parameters of the motor, resulting in a disconnect between the motor control algorithm and the motor mechanism. Mechanism models can effectively solve the above problems, but inaccurate modeling in mechanistic models affects measurement accuracy. Therefore, this invention establishes a mechanism-data hybrid model, which uses a data model as an error compensator, thus solving the disconnect between the control algorithm and the motor mechanism and improving the measurement accuracy of the model.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope defined by the claims of the present invention.

Claims

1. A method for suppressing thrust fluctuations in a linear motor based on an acceleration hybrid model, characterized in that: Includes the following steps: Step 1: Collect the structural parameters of the linear motor. Based on the structural parameters of the linear motor, establish a thrust fluctuation mechanism model, as shown in the following formula: in, For the mechanistic model of thrust fluctuation, For end magnetic resistance, For coarse magnetic resistance, Friction; The thrust fluctuation mechanism model includes an end effect mechanism model, a tooth groove effect mechanism model, and a friction mechanism model; The formula for the end effect mechanism model is as follows: in, The air gap coefficient, δ The length of the air gap. The maximum magnetic flux passing through the longitudinal end edge of the mover core. The permeability of free space, is the flux compression factor. τ The polar distance is the length. The equivalent magnetic circuit length, x For the position of the mover, This is the difference between the mover length and the multiple of the pole moment. ; The formula for the tooth cogging effect mechanism model is as follows: in, z The number of slots for the motor. The height of the magnetization direction of the permanent magnet. p The number of poles of the motor relative to the armature core length. The amplitude of the harmonic component of the permanent magnet remanent density is given. This represents the amplitude of the relative air gap permeability harmonic component; ; The formula for the friction mechanism model is as follows: in, For Coulomb friction, For the maximum static friction force, For the motioner velocity, The coefficient of lubrication. For symbolic functions, It is the coefficient of viscous friction; Step 2: The suppression method uses current injection. The compensation current is obtained based on the linear motor thrust equation, and its formula is as follows: in, Thrust fluctuation mechanism model compensation current, This represents the number of pole pairs of the linear motor. τ The polar distance is the length. For permanent magnet flux linkage; Step 3: Thrust fluctuation is calculated using the electromagnetic thrust suppression mechanism model, and the formula is as follows: in, Electromagnetic thrust to suppress thrust fluctuations in the mechanism model; Step 4: After compensation by the thrust fluctuation mechanism model, thrust data is collected using sensors. Since the thrust fluctuation of the linear motor is a periodic function of position, the data is processed and fitted using Fourier series to establish a thrust fluctuation data model; the sensors are force sensors or acceleration sensors. When the sensor is a force sensor, the thrust fluctuation data of the linear motor is obtained through the uniform velocity method. The data is then fitted, and a linear motor thrust fluctuation data model is established. The formula is as follows: in, A i The amplitude of the frequency component. φ i The phase of the frequency component; When the sensor is an acceleration sensor, the acceleration of the linear motor's mover is collected. An acceleration data model is used to fit the collected acceleration data, and then a linear motor thrust fluctuation data model is established. The formula is as follows: in, m For the mass of the mover, a For the acceleration of the moving part; Step 5: Calculate the compensation current of the thrust fluctuation data model established after compensation by the thrust fluctuation mechanism model. The formula is as follows: in, For data model thrust fluctuations; Step 6: Calculate the corresponding thrust fluctuation using the electromagnetic thrust suppression data model. The formula is as follows: in, Electromagnetic thrust to suppress thrust fluctuations in the data model.

2. The method for suppressing thrust fluctuations in a linear motor based on an acceleration hybrid model according to claim 1, characterized in that: The formula for the maximum static friction force in the friction mechanism model is as follows: in, The coefficient of static friction; For positive pressure, its formula is as follows: in, For end effect normal force, This is the normal force due to the cogging effect. The force is the weight of the kinetic particle itself.

3. The method for suppressing thrust fluctuations in a linear motor based on an acceleration hybrid model according to claim 2, characterized in that: End effect normal force and cogging effect normal force The mechanistic model formula is as follows: in, The normal force is caused by the left-end effect. This is the normal force caused by the right-end effect; ; Cogging effect normal force The mechanistic model formula is as follows: in, This represents the DC component of the relative air gap permeability. This represents the DC component of the remanent magnetization of the permanent magnet. .

4. The method for suppressing thrust fluctuations in a linear motor based on an acceleration hybrid model according to claim 3, characterized in that: The end effect normal force In the mechanistic model, when When the ratio is 0, the normal force caused by the left-end effect is equal to the normal force caused by the right-end effect. The net force received by the mover is twice the single-end normal force, as shown in the following formula: when λ = τ When the phase is 2, the odd harmonics of the wave force at the left and right ends are out of phase, and the mover has a tendency for "longitudinal pitching motion". Therefore, the influence of the upper surface of the motor guide rail on the friction force needs to be considered. The mechanism model of static friction force is as follows: in, For the maximum static friction force, The static friction coefficient of the lower surface of the motor guide rail is... Let be the static friction coefficient of the upper surface of the motor guide rail.

5. The method for suppressing thrust fluctuations in a linear motor based on an acceleration hybrid model according to claim 1, characterized in that: When the sensor is an accelerometer, the acceleration model is composed of Fourier series combinations. The acceleration model structure is as follows: Calculate the fundamental frequency, which is the ratio of motor speed to pole pitch: Where f is the fundamental frequency and v is the mover velocity; FFT analysis was performed on the acceleration data to obtain its frequency response plot. The remaining frequency components were obtained from the frequency response plot. All frequency points were then incorporated into the acceleration model and fitted using the least squares method to obtain the fitted acceleration data model.

6. The method for suppressing thrust fluctuations in a linear motor based on an acceleration hybrid model according to claim 5, characterized in that: The acceleration model requires a quadratic integration of the acceleration to obtain position information, thus revealing the relationship between acceleration and position, and further deriving the relationship between thrust fluctuation and mover position, as shown in the following formula: in, For speed, For displacement, For acceleration, Let j be the acceleration value at time j. Let j be the velocity value at time j. The sampling period.

7. The method for suppressing thrust fluctuations in a linear motor based on an acceleration hybrid model according to claim 6, characterized in that: In the process of obtaining location information, if there are error factors that may affect the calculation results, the Kalman filter algorithm is used to reduce the impact of errors. The formula is as follows: State prediction: in, For state vectors, Here is the state transition matrix. For the control matrix, Accelerate for the current moment; Predicted Observations: in, For the observation vector, For the observation matrix, It is a state vector; Covariance prediction: in, It is the inverse of the covariance matrix. Let the covariance matrix at the previous time step be denoted as . Here is the state transition matrix. Let be the covariance matrix of the process noise; Update Gains: in, Here is the Kalman gain matrix. The covariance matrix of the observation noise; Update status: Update the covariance matrix: in, Let be the covariance matrix.

8. The method for suppressing thrust fluctuations in a linear motor based on an acceleration hybrid model according to claim 7, characterized in that: The state prediction uses acceleration as the object motion control variable to express the relationship between the input state and the output state, and its formula is as follows: in, This represents the current position. The speed at the current moment, Accelerate to the current moment, The sampling period.

Images