A linear resonant motor back electromotive force detection method and system

Abstract

The embodiment of the application provides a linear resonant motor reverse electromotive force detection method and system, which can detect driving voltage V(t n ) and driving current I(t n ) of the linear resonant motor in real time without stopping the driving circuit, estimates reverse electromotive force and coil impedance of the linear resonant motor in real time by using an adaptive sliding mode observer (Sliding mode observer) or an adaptive Luenberger observer (Luenberger Observer), and thus calculates information such as resonator speed and temperature. Even if R changes due to factors such as interface plugging, environmental temperature change, motor long vibration temperature rise and the like, the change can be detected in real time, so that the BEMF detection value can be detected correctly.

Description

Technical Field

[0001] This invention relates to the field of back electromotive force detection technology, specifically to a method and system for detecting the back electromotive force of a linear resonant motor. Background Technology

[0002] Linear resonant actuators (LRAs) are commonly used to provide haptic feedback in portable devices. An LRA consists of components such as a spring, coil, and oscillator. It is driven by an LRA driver chip. The driver chip applies an excitation current to the coil, generating a magnetic field that propels the magnetically driven oscillator in a specific direction. When the direction of the excitation current changes, the magnetic field and the driving force also change. Therefore, if a periodic voltage signal is applied to the coil by the driver chip, the resulting periodic excitation current will drive the oscillator to oscillate back and forth, achieving the haptic feedback effect. Due to the resonant characteristics of the LRA, the amplitude of the oscillation exhibits a bandpass characteristic with the frequency of the driving signal. When the driving signal frequency is at the oscillator's natural frequency (F0), the amplitude of the oscillation reaches its maximum, and the vibration efficiency is optimal.

[0003] Existing methods for detecting the back electromotive force (BEMF) of an LRA (Low-Range Electric Ride) oscillator include: zero-crossing window voltage detection and current-voltage (IV) detection. The zero-crossing window voltage detection method detects the BEMF generated by the movement of the LRA oscillator by having the driver chip shut off the drive signal within a short time window near the zero-crossing point of the provided drive voltage waveform. When the current in the LRA coil inductance discharges to 0, the voltage across the LRA is the back electromotive force. The back electromotive force signal within the window is obtained by acquiring the voltage across the LRA through an acquisition circuit. When the drive circuit is restarted, the BEMF detection window closes, and BEMF detection stops. The current-voltage (IV) detection method, without stopping the drive voltage waveform signal, uses current and voltage detection circuits to obtain the voltage signal across the LRA and the current signal flowing through the LRA. The actual value of BEMF is calculated using the known motor coil resistance R and the formula BEMF = VI * R.

[0004] The zero-crossing window voltage detection method requires windowing detection of BEMF at the zero-crossing point of the chip's driving voltage waveform. During the zero-crossing period, the chip stops driving the circuit, causing additional harmonic components and resulting in high audio noise. Furthermore, because the chip stops driving the circuit transistors during the zero-crossing period, the average driving signal amplitude decreases as the zero-crossing window time increases, requiring additional compensation algorithms to adjust the output amplitude to achieve consistent vibration amplitude. This also reduces the maximum average driving signal amplitude under the rated power supply voltage. Additionally, while the chip stops driving the circuit, the current in the parasitic inductance of the motor coil cannot change abruptly, resulting in inductor freewheeling and causing the parasitic diode of the chip's output power transistor to conduct, affecting BEMF detection at the output. Therefore, it is necessary to wait for the parasitic inductance to discharge before detecting BEMF, increasing the zero-crossing waiting time. To avoid reliability issues caused by parasitic diode conduction in the chip's output power transistor and substrate debiasing, additional discharge circuits or increased device spacing are required, thereby increasing chip costs.

[0005] The current-voltage (IV) sensing method requires a known motor coil resistance. Portable devices typically measure the motor coil resistance during factory calibration or power-on calibration. However, in reality, the load impedance measured at the chip output port includes the motor coil resistance, PCB trace resistance, flexible printed circuit board (FPC) parasitic resistance, and interface contact resistance. After calibration, the load impedance changes due to factors such as interface insertion / removal, ambient temperature variations, and motor temperature rise during prolonged operation, causing errors in BEMF calculation. This results in inaccurate amplitude and frequency of the chip drive voltage waveform, and may even lead to chip malfunction. Summary of the Invention

[0006] Therefore, embodiments of the present invention provide a method and system for detecting the back electromotive force of a linear resonant motor to solve the above-mentioned technical problems in the prior art.

[0007] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0008] According to a first aspect of the present invention, this application provides a method for detecting the back electromotive force of a linear resonant motor, the method comprising:

[0009] Drive a linear resonant motor;

[0010] Real-time detection of t without stopping the drive. nThe driving voltage V(t) of the linear resonant motor at time t n ) and driving current I(t) n ), where n is the number of samples;

[0011] Based on the set initial estimate of the excitation current The first negative of the initial estimated values ​​of the motor coil resistance voltage drop and the initial estimated values ​​of the back electromotive force is obtained;

[0012] Using the driving voltage V(t1) at the first moment, the first inverse, and the initial estimated value of the voltage drop across the motor coil resistance, calculate the estimated value of the excitation current at the first moment.

[0013] Using the estimated values ​​of the drive current I(t1) and excitation current at the first moment Calculate the second inverse of the estimated voltage drop across the motor coil resistance and the estimated back electromotive force at the first moment;

[0014] Based on the second negative value calculated at the previous moment and the estimated value of the voltage drop across the motor coil resistance, combined with the current driving voltage V(t) n This allows us to obtain estimated values ​​of the excitation current at subsequent times.

[0015] Using the driving current I(t) at each subsequent moment n and estimated excitation current Calculate the second negative of the estimated values ​​of the motor coil resistance voltage drop and the back electromotive force at each subsequent time point;

[0016] Calculate the product of the second negative and -1 at each time step; and

[0017] After filtering out the ripple generated by the adaptive observer's chattering based on the first product, the estimated values ​​of the back electromotive force at each time moment are obtained and output.

[0018] Furthermore, based on the set initial estimate of the excitation current... Obtain the first negative of the initial estimated values ​​of the motor coil resistance voltage drop and the initial estimated values ​​of the back electromotive force, including:

[0019] Set the initial estimate of the excitation current =0;

[0020] Calculate the initial estimate of the excitation current. Compared with the preset initial estimate of the motor coil resistance The second product yields the initial estimate of the voltage drop across the motor coil resistance.

[0021] Calculate the initial drive current I(t0) and the initial estimated value of the excitation current. The difference is used to obtain the initial error of the excitation current estimation.

[0022] Using adaptive sliding mode observer gain K or adaptive Luenberger observer gain L d The initial error of the excitation current estimation Process it;

[0023] The product of the initial error of the processed excitation current estimate and the second gain L is calculated to obtain the first negative number of the initial estimate of the back electromotive force.

[0024] Furthermore, using the driving voltage V(t1) at the first moment, the first inverse, and the initial estimated value of the voltage drop across the motor coil resistance, the estimated value of the excitation current at the first moment is calculated. include:

[0025] Using the driving voltage V(t1) at the first moment and the first negative number as positive inputs, and the initial estimated value of the voltage drop across the motor coil resistance as negative inputs, a mixed addition and subtraction operation is performed to obtain the estimated value of the excitation voltage of the motor coil at the first moment.

[0026] The product of the estimated excitation voltage at the first moment and the first gain 1 / L is calculated to obtain the estimated rate of change of the excitation current at the first moment.

[0027] Using the estimated rate of change of excitation current at the first moment By performing integration, the estimated value of the excitation current at the first moment is obtained.

[0028] Furthermore, the estimated values ​​of the driving current I(t1) and the excitation current at the first moment are used. Calculate the second negative of the estimated voltage drop across the motor coil resistance and the estimated back electromotive force at the first moment, including:

[0029] Calculate the estimated values ​​of the driving current I(t1) and the excitation current at the first moment. The difference is used to obtain the excitation current estimation error at the first moment.

[0030] Using the excitation current estimate at the first moment And excitation current estimation error Iterative calculations are performed to detect changes in the motor coil resistance with external factors, and the estimated value of the motor coil resistance at the first moment is obtained and output.

[0031] Calculate the estimated value of the motor coil resistance at the first moment. With excitation current estimate The second product yields the estimated value of the motor coil resistance voltage drop at the first moment;

[0032] Using adaptive sliding mode observer gain K or adaptive Luenberger observer gain L d The error in estimating the excitation current at the first moment Process it;

[0033] The error in estimating the excitation current at the first moment after calculation and processing. The product of this and the second gain L yields the second negative of the estimated reverse electromotive force at the first moment.

[0034] Furthermore, based on the second inverse calculated at the previous moment and the estimated value of the motor coil resistance voltage drop, combined with the current driving voltage V(t)... n This allows us to obtain estimated values ​​of the excitation current at subsequent times. include:

[0035] The driving voltage V(t) at the current moment n The second negative number of the previous moment is used as the positive input, and the estimated value of the motor coil resistance voltage drop of the previous moment is used as the negative input. The addition and subtraction are mixed to obtain the estimated value of the excitation voltage of the motor coil at each subsequent moment.

[0036] The product of the estimated excitation voltage at each subsequent time step and the first gain 1 / L is calculated to obtain the estimated rate of change of the excitation current at each subsequent time step.

[0037] Using the estimated rate of change of excitation current at subsequent time points By performing integration, the estimated values ​​of the excitation current at subsequent time points are obtained.

[0038] Furthermore, the estimated values ​​of the excitation current change rate at each moment are used. The formula for integration is as follows:

[0039]

[0040] Among them, t n It is the current moment. This is the estimated rate of change of the excitation current at time τ, where d is the differential sign and τ is the integral variable. This is the initial estimated value of the excitation current, which is 0.

[0041] Furthermore, using the driving current I(t) at subsequent time points... n and estimated excitation current Calculate the second negative of the estimated motor coil resistance voltage drop and the estimated back electromotive force at subsequent time points, including:

[0042] Calculate the driving current I(t) at each subsequent time step. n ) and the estimated value of excitation current The difference is used to obtain the excitation current estimation error at subsequent times.

[0043] Using the excitation current estimates at subsequent time points And excitation current estimation error Iterative calculations are performed to detect changes in the motor coil resistance with external factors, and estimated values ​​of the motor coil resistance at subsequent time points are obtained and output.

[0044] Calculate the estimated values ​​of the motor coil resistance at each subsequent time point. With excitation current estimate The second product is used to obtain the estimated value of the motor coil resistance voltage drop at subsequent time points;

[0045] Using adaptive sliding mode observer gain K or adaptive Luenberger observer gain L d The estimation error of the excitation current at subsequent times Process it;

[0046] The calculation of the excitation current estimation error at each subsequent time point after processing. The product of this and the second gain L yields the second negative of the estimated back electromotive force at subsequent time points.

[0047] Furthermore, the estimated values ​​of the motor coil resistance at each time point. The iterative calculation formula is as follows:

[0048]

[0049] Where L is the coil inductance of the linear resonant motor; α is the iteration coefficient, which is a preset positive real number; This is a preset initial value, usually set to the nominal coil resistance value provided by the linear resonant motor manufacturer.

[0050] According to a second aspect of the present invention, this application provides a back electromotive force detection system for a linear resonant motor, the system comprising:

[0051] A linear resonant motor drive circuit is used to drive a linear resonant motor.

[0052] The signal detection section includes a voltage detection ADC and a current detection ADC, which are used to detect t in real time without stopping the drive.n The driving voltage V(t) of the linear resonant motor at time t n ) and driving current I(t) n ), where n is the number of samples;

[0053] An adaptive observer is used to perform the following steps:

[0054] Based on the set initial estimate of the excitation current The first negative of the initial estimated values ​​of the motor coil resistance voltage drop and the initial estimated values ​​of the back electromotive force is obtained;

[0055] Using the driving voltage V(t1) at the first moment, the first inverse, and the initial estimated value of the voltage drop across the motor coil resistance, calculate the estimated value of the excitation current at the first moment.

[0056] Using the estimated values ​​of the drive current I(t1) and excitation current at the first moment Calculate the second inverse of the estimated voltage drop across the motor coil resistance and the estimated back electromotive force at the first moment;

[0057] Based on the second negative value calculated at the previous moment and the estimated value of the voltage drop across the motor coil resistance, combined with the current driving voltage V(t) n This allows us to obtain estimated values ​​of the excitation current at subsequent times.

[0058] Using the driving current I(t) at each subsequent moment n and estimated excitation current Calculate the second negative of the estimated values ​​of the motor coil resistance voltage drop and the back electromotive force at each subsequent time point;

[0059] Calculate the product of the second negative and -1 at each time step; and

[0060] After filtering out the ripple generated by the adaptive observer's chattering based on the first product, the estimated values ​​of the back electromotive force at each time moment are obtained and output.

[0061] Furthermore, the adaptive observer includes: a hybrid arithmetic unit, a first amplifier, an integrator, a parameter identification module, a multiplier, a two-input subtractor, an observation unit, a second amplifier, a third amplifier, and a low-pass filter;

[0062] The multiplier calculates the initial estimate of the excitation current. Compared with the preset initial estimate of the motor coil resistance The second product yields the initial estimate of the voltage drop across the motor coil resistance.

[0063] The initial drive current I(t0) and the initial estimated value of the excitation current are calculated by the two-input subtractor. The difference is used to obtain the initial error of the excitation current estimation. The observation unit utilizes either the adaptive sliding mode observer gain K or the adaptive Lumberjack observer gain L. d The initial error of the excitation current estimation The process is performed; the product of the initial error of the processed excitation current estimate and the second gain L is calculated through the second amplifier to obtain the first negative number of the initial estimate of the back electromotive force.

[0064] The hybrid arithmetic unit takes the driving voltage V(t1) at the first moment and the first inverse as positive inputs, and the initial estimated value of the voltage drop across the motor coil resistance as negative inputs, and performs mixed addition and subtraction operations to obtain the estimated value of the excitation voltage of the motor coil at the first moment. The first amplifier then calculates the product of the estimated excitation voltage at the first moment and the first gain 1 / L to obtain the estimated value of the excitation current change rate at the first moment. The integrator uses the estimated value of the excitation current change rate at the first moment. By performing integration, the estimated value of the excitation current at the first moment is obtained.

[0065] The two-input subtractor calculates the estimated values ​​of the driving current I(t1) and the excitation current at the first moment. The difference is used to obtain the excitation current estimation error at the first moment. The parameter identification module uses the estimated excitation current value at the first moment. And excitation current estimation error Iterative calculations are performed to detect changes in the motor coil resistance with external factors, and the estimated value of the motor coil resistance at the first moment is obtained and output. The multiplier calculates the estimated value of the motor coil resistance at the first moment. With excitation current estimate The second product yields the estimated value of the motor coil resistance voltage drop at the first moment;

[0066] The observation unit utilizes either the adaptive sliding mode observer gain K or the adaptive Lumberjack observer gain L. d The error in estimating the excitation current at the first moment Processing is performed; the excitation current estimation error at the first moment after processing by the second amplifier is calculated. The product of this and the second gain L yields the second negative of the estimated back electromotive force at the first moment;

[0067] The hybrid arithmetic unit is also used to convert the current driving voltage V(t) into a variable. nThe first amplifier takes the second negative number from the previous moment as a positive input and the estimated voltage drop across the motor coil resistance from the previous moment as a negative input, performing a mixed addition and subtraction operation to obtain the estimated excitation voltage of the motor coil at subsequent moments. The first amplifier is also used to calculate the product of the estimated excitation voltage at subsequent moments and the first gain 1 / L to obtain the estimated rate of change of the excitation current at subsequent moments. The integrator is also used to estimate the rate of change of the excitation current at subsequent time points. By performing integration, the estimated values ​​of the excitation current at subsequent time points are obtained.

[0068] The two-input subtractor is also used to calculate the drive current I(t) at subsequent time steps. n ) and the estimated value of excitation current The difference is used to obtain the excitation current estimation error at subsequent times. The parameter identification module is also used to utilize the estimated excitation current values ​​at subsequent time points. And excitation current estimation error Iterative calculations are performed to detect changes in the motor coil resistance with external factors, and estimated values ​​of the motor coil resistance at subsequent time points are obtained and output. The multiplier is also used to calculate the estimated values ​​of the motor coil resistance at subsequent time points. With excitation current estimate The second product is used to obtain the estimated value of the motor coil resistance voltage drop at subsequent time points;

[0069] The observation unit is also used to utilize the adaptive sliding mode observer gain K or the adaptive Luenberger observer gain L. d The estimation error of the excitation current at subsequent times The second amplifier is used to process the data; it is also used to calculate the excitation current estimation error at subsequent time points after processing. The product of this and the second gain L yields the second negative of the estimated back electromotive force at subsequent time points;

[0070] The third amplifier calculates the first product of the second inverse number and -1; after filtering out the ripple generated by the adaptive observer jitter based on the first product by the low-pass filter, the estimated value of the back electromotive force is obtained and output.

[0071] Compared with the prior art, the method and system for detecting the back electromotive force of a linear resonant motor provided in this application embodiment can detect the driving voltage V(t) of the linear resonant motor in real time without stopping the driving circuit. n ) and driving current I(t) nUsing an adaptive sliding mode observer or an adaptive Luenberger observer, the back electromotive force and coil impedance of the linear resonant motor are estimated in real time, thereby calculating information such as oscillator speed and temperature. Even if factors such as interface plugging and unplugging, changes in ambient temperature, and long-term motor vibration temperature rise cause changes in R, these can be detected in real time to ensure the accuracy of the BEMF detection value. Attached Figure Description

[0072] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0073] The structures, proportions, sizes, etc. illustrated in this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed herein, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0074] Figure 1 This is a schematic diagram of a linear resonant motor back electromotive force detection system provided in an embodiment of the present invention;

[0075] Figure 2 This is a schematic diagram of the structure of the adaptive observer provided in an embodiment of the present invention;

[0076] Figure 3 This is a flowchart illustrating a method for detecting the back electromotive force of a linear resonant motor, as provided in an embodiment of the present invention. Detailed Implementation

[0077] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0078] refer to Figure 1This application provides a linear resonant motor back electromotive force detection system, including: a linear resonant motor drive circuit 1, a signal detection section 2, and an adaptive observer 3.

[0079] Specifically, the linear resonant motor drive circuit 1 is used to drive the linear resonant motor 4; the linear resonant motor drive circuit is any excitation source, the output signal is any non-zero drive voltage waveform, and the linear resonant motor 4 is the detection object.

[0080] The signal detection section includes a voltage detection ADC 21 and a current detection ADC 22. Without stopping the drive, the voltage detection ADC 21 is used to detect t in real time. n The driving voltage V(t) of the linear resonant motor at any given time n The current-sensing ADC 22 is used for real-time detection of t. n The driving current I(t) of the linear resonant motor at any given time n ), where n is the number of samplings. That is, when the linear resonant motor drive circuit 1 is driven without interruption, the voltage detection ADC 21 detects the voltage waveform across the linear resonant motor 4, and the current detection ADC 22 detects the current waveform flowing through the linear resonant motor 4.

[0081] refer to Figure 2 The driving voltage V(t) of the linear resonant motor 4 n ) and driving current I(t) n The relationship between the input signals and the input signal is the driving voltage V(t) of the linear resonant motor. The input signal to the IV characteristic model of the linear resonant motor is the driving voltage V(t) of the linear resonant motor. n The output signal of the IV characteristic model of a linear resonant motor is the drive current I(t) of the linear resonant motor. n The transfer function of the IV characteristic model of a linear resonant motor is the admittance (I / V) of the linear resonant motor. Due to the resonant characteristics of the linear resonant motor, the vibration amplitude and vibration velocity of the linear resonant motor reach their maximum values ​​at its natural frequency. The back electromotive force of the linear resonant motor is proportional to the vibration velocity of the linear resonant motor, therefore the back electromotive force also reaches its maximum value at its natural frequency. From the formula BEMF = VI·R, at the input drive voltage V(t) n Under the condition of constant amplitude, the output (drive current I(t)) of the IV characteristic model of a linear resonant motor n The value reaches its minimum at the natural frequency, therefore, the IV characteristic model of the linear resonant motor exhibits band-stop characteristics similar to those of a notch filter.

[0082] The adaptive observer 3 is specifically used to implement the following steps: based on the set initial estimate of the excitation current. Obtain the first negative of the initial estimated values ​​of the motor coil resistance voltage drop and the initial estimated values ​​of the back electromotive force; using the driving voltage V(t1) at the first moment, the first negative value, and the initial estimated value of the motor coil resistance voltage drop, calculate the estimated value of the excitation current at the first moment. Using the estimated values ​​of the drive current I(t1) and excitation current at the first moment Calculate the second inverse of the estimated motor coil resistance voltage drop and the estimated back electromotive force at the first moment; based on the second inverse calculated at the previous moment and the estimated motor coil resistance voltage drop, combine the current driving voltage V(t)... n This allows us to obtain estimated values ​​of the excitation current at subsequent times. Using the driving current I(t) at each subsequent moment n and estimated excitation current Calculate the second inverse of the estimated values ​​of the motor coil resistance voltage drop and the estimated values ​​of the back electromotive force at each subsequent time point; calculate the first product of the second inverse of each time point and -1; and after filtering out the ripple generated by the adaptive observer jitter based on the first product, obtain and output the estimated values ​​of the back electromotive force at each time point.

[0083] Further, refer to Figure 2 The adaptive observer 3 includes: a hybrid arithmetic unit 31, a first amplifier 32, an integrator 33, a two-input subtractor 34, a parameter identification module 35, a multiplier 36, an observation unit 37, a second amplifier 38, a third amplifier 39, and a low-pass filter 310.

[0084] In this embodiment of the invention, the preset parameters of the adaptive observer 3 include: the coil inductance L of the linear resonant motor, which is a known quantity provided by the LRA manufacturer or can be measured by an impedance meter, with a typical value of 100uH; and the adaptive sliding mode observer gain K or the adaptive Luenberger observer gain L. d , is a preset positive real number. The iteration coefficient α of the parameter identification module is a preset positive real number. The low-pass filter's bandwidth, order, and other parameters are defined, with a typical bandwidth of 500Hz and a typical order of 4th.

[0085] Specifically, the multiplier 36 calculates the initial estimate of the excitation current. Compared with the preset initial estimate of the motor coil resistance The second product yields the initial estimate of the voltage drop across the motor coil resistance; the initial drive current I(t0) and the initial estimate of the excitation current are calculated by the two-input subtractor 34. The difference is used to obtain the initial error of the excitation current estimation. The observation unit 37 utilizes either the adaptive sliding mode observer gain K or the adaptive Lumberjack observer gain L. d Initial error in estimating excitation current The process is as follows: the product of the processed excitation current estimation initial error and the second gain L is calculated by the second amplifier 38 to obtain the first negative value of the initial estimate of the back electromotive force; the hybrid arithmetic unit 31 takes the driving voltage V(t1) at the first moment and the first negative value as positive inputs, and the initial estimate of the motor coil resistance voltage drop as negative inputs, and performs mixed addition and subtraction operations to obtain the estimated value of the excitation voltage of the motor coil at the first moment; the product of the estimated value of the excitation voltage at the first moment and the first gain 1 / L is calculated by the first amplifier 32 to obtain the estimated value of the excitation current change rate at the first moment. The integrator 33 uses the estimated value of the excitation current change rate at the first moment. By performing integration, the estimated value of the excitation current at the first moment is obtained. The two-input subtractor 34 calculates the estimated values ​​of the drive current I(t1) and the excitation current at the first moment. The difference is used to obtain the excitation current estimation error at the first moment. The parameter identification module 35 uses the estimated excitation current value at the first moment. And excitation current estimation error Iterative calculations are performed to detect changes in the motor coil resistance with external factors, and the estimated value of the motor coil resistance at the first moment is obtained and output. The estimated value of the motor coil resistance at the first moment is calculated using multiplier 36. With excitation current estimate The second product yields the estimated voltage drop across the motor coil resistance at the first moment; this is then calculated by observation unit 37 using either the adaptive sliding mode observer gain K or the adaptive Luenberger observer gain L. d The error in estimating the excitation current at the first moment Processing is performed; the excitation current estimation error at the first moment after processing by the second amplifier 38 is calculated. The product of the first gain L and the second negative value of the back electromotive force estimate at the first moment is obtained; the hybrid arithmetic unit 31 is also used to convert the current driving voltage V(t) into the second negative value of the back electromotive force estimate at the first moment. n The first amplifier 32 takes the second negative number from the previous moment as a positive input and the estimated voltage drop across the motor coil resistance from the previous moment as a negative input, performing a mixed addition and subtraction operation to obtain the estimated excitation voltage of the motor coil at subsequent moments. The first amplifier 32 is also used to calculate the product of the estimated excitation voltage at subsequent moments and the first gain 1 / L to obtain the estimated rate of change of the excitation current at subsequent moments. Integrator 33 is also used to estimate the rate of change of excitation current at subsequent time points. By performing integration, the estimated values ​​of the excitation current at subsequent time points are obtained. The two-input subtractor 34 is also used to calculate the drive current I(t) at subsequent time steps. n ) and the estimated value of excitation current The difference is used to obtain the excitation current estimation error at subsequent times. The parameter identification module 35 is also used to utilize the estimated excitation current values ​​at subsequent times. And excitation current estimation error Iterative calculations are performed to detect changes in the motor coil resistance with external factors, and the estimated values ​​of the motor coil resistance at subsequent time points are obtained and output. Multiplier 36 is also used to calculate the estimated values ​​of the motor coil resistance at subsequent time points. With excitation current estimate The second product is used to obtain the estimated value of the motor coil resistance voltage drop at subsequent time points; the observation unit 37 is also used to utilize the adaptive sliding mode observer gain K or the adaptive Lumberjack observer gain L d The estimation error of the excitation current at subsequent times The second amplifier 38 is used to process the data; it is also used to calculate the excitation current estimation error at subsequent time points after processing. The product of the second gain L and the second inverse is used to obtain the second inverse of the back electromotive force estimate at each subsequent time step. The third amplifier 39 calculates the first product of the second inverse and -1. After filtering out the ripple generated by the adaptive observer jitter based on the first product by the low-pass filter 310, the back electromotive force estimate is obtained and output.

[0086] Compared with the prior art, the linear resonant motor back electromotive force detection system provided in this application embodiment can detect the driving voltage V(t) of the linear resonant motor in real time without stopping the driving circuit. n ) and driving current I(t) n Using an adaptive sliding mode observer or an adaptive Luenberger observer, the back electromotive force and coil impedance of the linear resonant motor are estimated in real time, thereby calculating information such as oscillator speed and temperature. Even if factors such as interface plugging and unplugging, changes in ambient temperature, and long-term motor vibration temperature rise cause changes in R, these can be detected in real time to ensure the accuracy of the BEMF detection value.

[0087] Corresponding to the aforementioned linear resonant motor back electromotive force detection system, this invention also discloses a method for detecting the back electromotive force of a linear resonant motor. The following details a method for detecting the back electromotive force of a linear resonant motor disclosed in this invention, in conjunction with the aforementioned linear resonant motor back electromotive force detection system.

[0088] like Figure 2 and Figure 3 As shown in the figure, this application provides a method for detecting the back electromotive force of a linear resonant motor, which specifically includes the following steps.

[0089] The linear resonant motor 4 is driven by the linear resonant motor drive circuit 1; the linear resonant motor drive circuit is any excitation source, the output signal is any non-zero drive voltage waveform, and the linear resonant motor 4 is the detection object.

[0090] The signal detection section 2 detects the IV characteristics of the linear resonant motor 4 in real time. Specifically, the IV characteristics of the linear resonant motor 4 include the driving voltage V and the driving current I. The signal detection section 2 includes a voltage detection ADC 21 and a current detection ADC 22. Under the condition of not stopping the drive, the voltage detection ADC 21 detects the current I in real time. n The driving voltage V(t) of the linear resonant motor 4 at any given time n The current is then sent to the adaptive observer 3, where the current sensing ADC 22 detects the drive current I(t) of the linear resonant motor 4 in real time. n And send it to adaptive observer 3.

[0091] Adaptive observer 3 receives the currently detected driving voltage V(t) of linear resonant motor 4. n ) and driving current I(t) n First, set the initial estimate of the excitation current. If the value is 0, the first negative number of the initial estimated value of the motor coil resistance voltage drop and the initial estimated value of the back electromotive force is obtained.

[0092] Specifically, obtaining the initial estimate of the motor coil resistance voltage drop includes: setting the initial estimate of the excitation current. The value is 0; the initial estimate of the excitation current is calculated by multiplier 36. Compared with the preset initial estimate of the motor coil resistance The second product yields the initial estimate of the voltage drop across the motor coil resistance.

[0093] The first negative value of the aforementioned back electromotive force estimate includes: the initial drive current I(t0) calculated by the two-input subtractor 34 and the initial estimate of the excitation current. The difference is used to obtain the initial error of the excitation current estimation. The observation unit 37 utilizes either the adaptive sliding mode observer gain K or the adaptive Lumberjack observer gain L. d Initial error in estimating excitation current The process is performed; the product of the initial error of the processed excitation current estimate and the second gain L is calculated by the second amplifier 38 to obtain the first negative number of the initial estimate of the reverse electromotive force.

[0094] In this embodiment, the adaptive observer 3 is divided into two types: an adaptive sliding mode observer and an adaptive Lumberjack observer. When the adaptive observer 3 is an adaptive sliding mode observer, the observation unit 37 estimates the initial error of the excitation current. After performing the sgn() operation, multiply by K to obtain the output. The specific calculation formula for observation unit 37 is as follows: Where sgn is the sign function and K is the adaptive sliding mode observer gain. A larger K results in a faster tracking rate from the sliding mode observer and a more stable system, but also more severe chattering. When the adaptive observer 3 is an adaptive Lumberjack observer, the observation unit 37 directly estimates the initial error using the excitation current. Multiply by L d As the output, the specific calculation formula for observation unit 37 is: Among them, L d For the Luneburg observer gain, L d The larger the value, the faster the tracking rate of the adaptive Luneburg observer, but the worse its stability.

[0095] The adaptive observer 3 uses the driving voltage V(t1), the first inverse, and the initial estimated value of the voltage drop across the motor coil resistance at the first moment to calculate the estimated value of the excitation current at the first moment. Specifically, in this embodiment of the invention, the hybrid arithmetic unit 31 is a three-input addition and subtraction hybrid arithmetic unit, obtained by circuit analysis of the LRA coil loop using Kirchhoff's voltage law. The above calculation of the estimated excitation current value at the first moment... The specific steps include: the voltage detection ADC 21 detects the driving voltage V(t1) of the linear resonant motor 4 in real time at the first moment and sends it to the hybrid arithmetic unit 31; the first negative value of the initial estimate of the back electromotive force calculated by the second amplifier 38 is sent to the hybrid arithmetic unit 31; the initial estimate of the motor coil resistance voltage drop calculated by the multiplier 36 is sent to the hybrid arithmetic unit 31; the hybrid arithmetic unit 31 takes the driving voltage V(t1) and the first negative value at the first moment as positive inputs and the initial estimate of the motor coil resistance voltage drop as negative inputs, performs addition and subtraction operations to obtain the estimated value of the excitation voltage of the motor coil at the first moment and sends it to the first amplifier 32; the first amplifier 32 calculates the product of the estimated value of the excitation voltage at the first moment and the first gain 1 / L to obtain the estimated value of the excitation current change rate at the first moment. And send it to integrator 33; integrator 33 uses the excitation current change rate estimate at the first moment to estimate the value. By performing integration, the estimated value of the excitation current at the first moment is obtained.

[0096] The integrator 33 uses the excitation current change rate estimate at the first moment to estimate the value. The formula for integration is as follows:

[0097]

[0098] Among them, t n It is the current moment. This is the estimated rate of change of the excitation current at time τ, where d is the differential sign and τ is the integral variable. This is the initial estimate of the excitation current, which is 0. Due to the low-pass characteristic of integrator 33, integrator 33 can... The ripples on the signal generated by the jitter of the adaptive sliding mode observer or the adaptive Luneburg observer are smoothed and filtered.

[0099] The adaptive observer 3 uses the estimated values ​​of the driving current I(t1) and the excitation current at the first moment. Calculate the second negative of the estimated voltage drop across the motor coil resistance and the estimated back electromotive force at the first moment.

[0100] In this embodiment of the invention, the specific steps for calculating the estimated value of the motor coil resistance voltage drop at the first moment include: calculating the estimated values ​​of the drive current I(t1) and the excitation current at the first moment using the two-input subtractor 34. The difference is used to obtain the excitation current estimation error at the first moment. And the excitation current estimation error at the first moment The signal is sent to the parameter identification module 35; the parameter identification module 35 then uses the estimated excitation current value from the first moment. And excitation current estimation error Iterative calculations are performed to detect changes in the motor coil resistance with external factors, and the estimated value of the motor coil resistance at the first moment is obtained and output. The estimated value of the motor coil resistance at the first moment is calculated using multiplier 36. With excitation current estimate The second product yields the estimated value of the motor coil resistance voltage drop at the first moment.

[0101] Furthermore, the estimated values ​​of the motor coil resistance at each first moment... The iterative calculation formula is as follows:

[0102]

[0103] Where L is the coil inductance of the linear resonant motor; α is the iteration coefficient, which is a preset positive real number; This is a preset initial value, usually set to the nominal coil resistance value provided by the linear resonant motor manufacturer.

[0104] In addition, the specific steps for calculating the second negative of the estimated back electromotive force at the first moment include: calculating the estimated values ​​of the driving current I(t1) and the excitation current at the first moment using the two-input subtractor 34. The difference is used to obtain the excitation current estimation error at the first moment. And the excitation current estimation error at the first moment The signal is sent to observation unit 37; observation unit 37 then uses either the adaptive sliding mode observer gain K or the adaptive Lumberjack observer gain L. d The error in estimating the excitation current at the first moment Processing is performed; the excitation current estimation error at the first moment after processing by the second amplifier 38 is calculated. The product of this and the second gain L yields the second negative of the estimated reverse electromotive force at the first moment.

[0105] Similarly, in this embodiment, the adaptive observer 3 is divided into two types: an adaptive sliding mode observer and an adaptive Lumberjack observer. When the adaptive observer 3 is an adaptive sliding mode observer, the observation unit 37 estimates the excitation current error. After performing the sgn() operation, multiply by K to obtain the output. The specific calculation formula for observation unit 37 is as follows: Where sgn is the sign function and K is the adaptive sliding mode observer gain. A larger K results in a faster tracking rate from the sliding mode observer and a more stable system, but also more severe chattering. When the adaptive observer 3 is an adaptive Luenberger observer, the observation unit 37 directly uses the excitation current to estimate the error. Multiply by L d As the output, the specific calculation formula for observation unit 37 is: Among them, L d For the Luneburg observer gain, L d The larger the value, the faster the tracking rate of the adaptive Luneburg observer, but the worse its stability.

[0106] The adaptive observer 3 calculates the second inverse number based on the previous moment and the estimated value of the motor coil resistance voltage drop, combined with the current driving voltage V(t). n This allows us to obtain estimated values ​​of the excitation current at subsequent times. At this point, n>1.

[0107] In this embodiment of the invention, the above-mentioned estimated values ​​of excitation current at subsequent times are obtained. The specific steps include: the voltage detection ADC 21 continuously detecting the driving voltage V(t) of the linear resonant motor 4 at each subsequent moment. nThe second inverse calculated by the second amplifier 38 at the previous moment is sent to the hybrid arithmetic unit 31. The estimated value of the motor coil resistance voltage drop calculated by the multiplier 36 at the previous moment is also sent to the hybrid arithmetic unit 31. The hybrid arithmetic unit 31 then sends the current driving voltage V(t) to the hybrid arithmetic unit 31. n The second negative number from the previous moment is used as the positive input, and the estimated value of the motor coil resistance voltage drop from the previous moment is used as the negative input. A mixed addition and subtraction operation is performed to obtain the estimated excitation voltage of the motor coil at subsequent moments. This estimated excitation voltage is then sent to the first amplifier 32. The first amplifier 32 calculates the product of the estimated excitation voltage at each subsequent moment and the first gain 1 / L to obtain the estimated rate of change of the excitation current at each subsequent moment. And send it to integrator 33; integrator 33 uses the excitation current change rate estimate at subsequent time points to estimate the value. By performing integration, the estimated values ​​of the excitation current at subsequent time points are obtained.

[0108] Furthermore, the estimated excitation current change rate at subsequent time points is used. The formula for integration is as follows:

[0109]

[0110] Among them, t n It is the current moment. This is the estimated rate of change of the excitation current at time τ, where d is the differential sign and τ is the integral variable. It is the initial estimate of the excitation current. The initial estimate of the excitation current is 0, and n>1.

[0111] The adaptive observer 3 utilizes the driving current I(t) at subsequent time points. n and estimated excitation current Calculate the second negative of the estimated values ​​of the motor coil resistance voltage drop and the estimated value of the back electromotive force at each subsequent time point.

[0112] In this embodiment of the invention, the specific steps for calculating the second inverse of the estimated values ​​of the motor coil resistance voltage drop and the back electromotive force at subsequent time points include: calculating the drive current I(t) at subsequent time points using the two-input subtractor 34. n ) and the estimated value of excitation current The difference is used to obtain the excitation current estimation error at subsequent times. The data is then sent to the parameter identification module 35; the parameter identification module 35 uses the estimated excitation current values ​​at subsequent times. And excitation current estimation error Iterative calculations are performed to detect changes in the motor coil resistance with external factors, and the estimated values ​​of the motor coil resistance at subsequent time points are obtained and output. The multiplier 36 calculates the estimated values ​​of the motor coil resistance at subsequent time points. With excitation current estimate The second product yields the estimated voltage drop across the motor coil resistance at subsequent time points; this is then calculated by observation unit 37 using either the adaptive sliding mode observer gain K or the adaptive Lumberjack observer gain L. d The estimation error of the excitation current at subsequent times The data is processed; the second amplifier 38 calculates the estimated error of the excitation current at subsequent time points after processing. The product of this and the second gain L yields the second negative of the estimated back electromotive force at subsequent time points.

[0113] Furthermore, the estimated values ​​of the motor coil resistance at subsequent time points... The iterative calculation formula is as follows:

[0114]

[0115] Where L is the coil inductance of the linear resonant motor; α is the iteration coefficient, which is a preset positive real number; This is a preset initial value, usually set to the nominal coil resistance value provided by the linear resonant motor manufacturer.

[0116] The second inverse of the back electromotive force estimate at each subsequent time point, calculated by the second amplifier 38, is sent to the third amplifier 39. The third amplifier 39 calculates the first product of the second inverse of the back electromotive force estimate at each subsequent time point and -1, and sends it to the low-pass filter 310. The low-pass filter 310 filters out the ripple generated by the adaptive observer jitter based on the first product, and then obtains and outputs the back electromotive force estimate at each time point.

[0117] Compared with the prior art, the method for detecting the back electromotive force of a linear resonant motor provided in this application embodiment can detect the driving voltage V(t) of the linear resonant motor in real time without stopping the driving circuit. n ) and driving current I(t) n Using an adaptive sliding mode observer or an adaptive Luenberger observer, the back electromotive force and coil impedance of the linear resonant motor are estimated in real time, thereby calculating information such as oscillator speed and temperature. Even if factors such as interface plugging and unplugging, changes in ambient temperature, and long-term motor vibration temperature rise cause changes in R, these can be detected in real time to ensure the accuracy of the BEMF detection value.

[0118] The method for detecting the back electromotive force of a linear resonant motor provided in this application has the following main advantages:

[0119] It eliminates high-order harmonics in the drive voltage waveform caused by zero-crossing window detection;

[0120] It eliminates the inconsistency in the amplitude of the drive voltage waveform caused by zero-crossing window detection;

[0121] It eliminates the attenuation of drive voltage amplitude caused by zero-crossing window detection, and increases the maximum output amplitude by 10% to 20% under the same power supply voltage;

[0122] The parameter identification module is used to estimate and detect the motor coil resistance in real time, which solves the problem of the detection error of BEMF caused by the drift of motor coil resistance in traditional IV detection.

[0123] Adaptive sliding mode observers or adaptive Luneburg observers are used to track BEMF voltages, reducing computational load, increasing tracking speed, and improving system stability.

[0124] By using the parameter identification module to estimate the motor coil resistance in real time, and considering the temperature coefficient of the motor coil resistance, an estimated value of the real-time temperature of the motor coil resistance can be calculated. When the motor coil overheats, appropriate vibration stop protection or attenuation of the drive voltage waveform amplitude can be implemented to prevent motor coil failure due to overheating. When the motor coil temperature is low, the drive voltage waveform amplitude is increased to compensate for the attenuation of the oscillator amplitude caused by the increased viscosity coefficient of the magnetic fluid at low temperatures.

[0125] Although the present invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A linear resonant motor back electromotive force detection method, characterized by, The method comprises: driving a linear resonant motor; Real-time detection of t without stopping the drive. n The driving voltage V(t) of the linear resonant motor at time t n ) and driving current I(t) n ), where n is the number of samples; Based on the set initial estimate of the excitation current Obtaining the first negative of the initial estimated values ​​of the motor coil resistance voltage drop and the initial estimated values ​​of the back electromotive force includes: calculating the initial drive current I(t0) and the initial estimated value of the excitation current. The difference is used to obtain the initial error of the excitation current estimation. Using adaptive sliding mode observer gain K or adaptive Luenberger observer gain L d The initial error of the excitation current estimation The process involves processing the initial error of the excitation current estimation by the observation unit. After performing the sgn() operation, multiply by K and use the result as the output, or the observation unit can directly estimate the initial error using the excitation current. Multiply by L d As the output, the product of the processed excitation current estimation initial error and the second gain L is calculated to obtain the first negative number of the initial estimate of the back electromotive force. calculating an excitation current estimate for the first time instant using the drive voltage V(t1) at the first time instant, the first phase opposition number, and the initial estimate of the motor coil resistance voltage drop Using the estimated values ​​of the drive current I(t1) and excitation current at the first moment Calculate the second inverse of the estimated values ​​of the motor coil resistance voltage drop and the back electromotive force at the first moment, which includes: calculating the estimated values ​​of the drive current I(t1) and the excitation current at the first moment. The difference is used to obtain the excitation current estimation error at the first moment. Using adaptive sliding mode observer gain K or adaptive Luenberger observer gain L d The error in estimating the excitation current at the first moment The processing involves, specifically, the observation unit estimating the excitation current error. After performing the sgn() operation, multiply by K and use the result as the output, or the observation unit can directly estimate the error using the excitation current. Multiply by L d As output, the excitation current estimation error at the first moment after processing is calculated. The product of this and the second gain L yields the second negative of the estimated back electromotive force at the first moment; Based on the second opposite number and the motor coil resistance voltage drop estimation value calculated at the previous moment, combined with the driving voltage V(t n ) at the current moment, the excitation current estimation value at each subsequent moment is obtained using the drive current I(t n ) and the estimated value of the field current calculating a second phase opposition of the estimated value of the motor coil resistance voltage drop and the estimated value of the back electromotive force at each of the subsequent time instants; calculating a first product of a second reciprocal number and -1 at each time point; and obtaining and outputting an estimated value of a back electromotive force at each time point after filtering out ripple generated by the adaptive observer based on the first product.

2. A method of detecting back EMF in a linear resonant motor as defined in claim 1, wherein, Further comprising: Setting an initial estimate of the field current is 0; computing the initial estimate of the excitation current a second product of the initial estimate of the motor coil resistance to obtain the initial estimate of the motor coil resistance voltage drop.

3. A method of detecting back EMF in a linear resonant motor as defined in claim 2, wherein, calculating an excitation current estimate for the first time instant using the drive voltage V(t1) for the first time instant, the first sign, and the initial estimate of the motor coil resistance voltage drop comprising: obtaining an estimated value of an excitation voltage of the motor coil at the first time point by taking the driving voltage V(t1) and the first reciprocal number at the first time point as positive inputs and taking the initial estimated value of the motor coil resistance voltage drop as a negative input; calculating a product of the excitation voltage estimation value at the first time and the first gain 1 / L to obtain the excitation current rate of change estimation value at the first time The first time point of the excitation current rate of change estimate value Integral operation is performed to obtain the first time point of the excitation current estimate value 4. A method of detecting back EMF in a linear resonant motor as defined in claim 1, wherein, Further comprising: using the excitation current estimate value at the first time and the excitation current estimation error performing iterative calculation, detecting the change of the motor coil resistance with external factors, obtaining and outputting the motor coil resistance estimate value at the first time Computing a motor coil resistance estimate at the first time instant a second product of the field current estimate to obtain a motor coil resistance voltage drop estimate at the first time instant.

5. A method of detecting back EMF in a linear resonant motor as defined in claim 4, wherein, Based on the second opposite number and the motor coil resistance voltage drop estimation value calculated at the previous moment, combined with the driving voltage V(t n ) at the current moment, the excitation current estimation value at each subsequent moment is obtained includes: The driving voltage V(t n ) at the current time and the second opposite number at the previous time are taken as the positive input, and the motor coil resistance voltage drop estimate value at the previous time is taken as the negative input to obtain the excitation voltage estimate value of the motor coil at each subsequent time. The product of the excitation voltage estimation value at each subsequent time and the first gain 1 / L is calculated to obtain the excitation current rate of change estimation value at each subsequent time The rate of change of the field current at each of the subsequent times is estimated The integral operation is performed to obtain the field current estimate at each of the subsequent times 6. A method of detecting back EMF in a linear resonant motor as defined in claim 5, wherein, The rate of change of the field current at each time is estimated The integral operation formula for performing the integral operation is as follows: where t n is the current time, is the τ-time magnetizing current rate of change estimate, d is the differential symbol, and τ is the integration variable, is the initial magnetizing current estimate, which is 0.

7. A method of detecting back EMF in a linear resonant motor as defined in claim 6, wherein, calculating a second phase opposition of the motor coil resistance voltage drop estimate and the back electromotive force estimate for the subsequent time instant using the drive current I(t n ) and the field current estimate calculating a second phase opposition of the motor coil resistance voltage drop estimate and the back electromotive force estimate for the subsequent time instant using the drive current I(t n ) and the field current estimate the driving current I(t) at each of the subsequent time points n the difference between the estimated value of the excitation current and the estimated value of the excitation current at each of the subsequent time points using the excitation current estimate value at each of the subsequent times and the excitation current estimate error performing iterative calculations to detect changes in the motor coil resistance due to external factors, and outputting motor coil resistance estimate values at each of the subsequent times the motor coil resistance estimate for the subsequent time instant a second product of the field current estimate the motor coil resistance voltage drop estimate for the subsequent time instant Utilizing an adaptive sliding mode observer gain K or an adaptive Luenberger observer gain L d The excitation current estimation error for each subsequent time instant is processed; computing an excitation current estimation error for each of the subsequent time instants a second inverse number of the back emf estimate for each of the subsequent time instants by multiplying the second gain L 8. A method of detecting back EMF in a linear resonant motor as defined in claim 2, wherein, motor coil resistance estimate at each time The iterative calculation formula of the above equation is as follows: Wherein, L is the coil inductance of the linear resonant motor; a is an iteration coefficient, the iteration coefficient is a preset positive real number; is a preset initial value, and is usually set as the nominal value of the coil resistance provided by the linear resonant motor manufacturer.

9. A linear resonant motor back electromotive force detection system, characterized by, The system comprises: a linear resonant motor driving circuit for driving a linear resonant motor; The signal detection section includes a voltage detection ADC and a current detection ADC, which are used to detect t in real time without stopping the drive. n The driving voltage V(t) of the linear resonant motor at time t n ) and driving current I(t) n ), where n is the number of samples; an adaptive observer for performing the following steps: Based on a set initial estimate of the field current Obtaining a first negative number of the initial estimate of the motor coil resistance voltage drop and the initial estimate of the back electromotive force, which includes: calculating the difference between the initial driving current I(t0) and the initial estimate of the field current , obtaining the initial error of the field current estimate Using the adaptive sliding mode observer gain K or the adaptive Luenberger observer gain L d , the initial error of the field current estimate is processed, specifically, the observation unit multiplies the initial error of the field current estimate after the sgn() operation by K as the output, or the observation unit directly multiplies the initial error of the field current estimate by L as the output d ; calculating the product of the processed initial error of the field current estimate and the second gain L, obtaining the first negative number of the initial estimate of the back electromotive force calculating an excitation current estimate for the first time instant using the drive voltage V(t1) for the first time instant, the first phase opposition number, and the initial estimate of the motor coil resistance voltage drop Using the estimated values ​​of the drive current I(t1) and excitation current at the first moment Calculate the second inverse of the estimated values ​​of the motor coil resistance voltage drop and the back electromotive force at the first moment, which includes: calculating the estimated values ​​of the drive current I(t1) and the excitation current at the first moment. The difference is used to obtain the excitation current estimation error at the first moment. Using adaptive sliding mode observer gain K or adaptive Luenberger observer gain L d The error in estimating the excitation current at the first moment The processing involves, specifically, the observation unit estimating the excitation current error. After performing the sgn() operation, multiply by K and use the result as the output, or the observation unit can directly estimate the error using the excitation current. Multiply by L d As output, the excitation current estimation error at the first moment after processing is calculated. The product of this and the second gain L yields the second negative of the estimated back electromotive force at the first moment; Based on the second opposite number and the motor coil resistance voltage drop estimation value calculated at the previous moment, combined with the driving voltage V(t n ) at the current moment, the excitation current estimation value at each subsequent moment is obtained using the drive current I(t n ) and the estimated value of the field current calculating a second phase opposition of the estimated value of the motor coil resistance voltage drop and the estimated value of the back electromotive force at the subsequent time instant; calculating a first product of a second reciprocal number and -1 at each time point; and obtaining and outputting an estimated value of a back electromotive force at each time point after filtering out ripple generated by the adaptive observer based on the first product.

10. A linear resonant motor back EMF detection system as claimed in claim 9, wherein, The adaptive observer comprises a hybrid operator, a first amplifier, an integrator, a parameter identification module, a multiplier, a two-input subtractor, an observation unit, a second amplifier, a third amplifier, and a low-pass filter. calculating the initial estimate of the field current from the product of the initial estimate of the motor coil resistance and the initial estimate of the motor coil current calculating the initial estimate of the motor coil resistance from the product of the initial estimate of the motor coil resistance and the initial estimate of the motor coil current calculating the initial estimate of the motor coil resistance from the product of the initial estimate of the motor coil resistance and the initial estimate of the motor coil current a difference between the initial drive current I(t0) and the initial estimate of the field current to obtain an initial error of the field current estimate is processed by the observation unit using adaptive sliding mode observer gain K or adaptive Luenberger observer gain L d , the initial error of the field current estimate is processed; a product of the processed initial error of the field current estimate and the second gain L is calculated by the second amplifier to obtain a first inverse of the initial estimate of the back electromotive force the first time point and the first phase opposition number as positive inputs, and the motor coil resistance voltage drop initial estimation value as a negative input, to obtain an excitation voltage estimation value of the motor coil at the first time point the excitation current change rate estimation value at the first time point as an input, to obtain an excitation current estimation value at the first time point ​ The difference between the driving current I(t1) at the first time and the excitation current estimation value calculated by the two-input subtractor is obtained to obtain the excitation current estimation error at the first time The excitation current estimation error at the first time is obtained The excitation current estimation value at the first time and the excitation current estimation error are used by the parameter identification module to perform iterative calculation The excitation current estimation value at the first time and the excitation current estimation error are used by the parameter identification module to perform iterative calculation The motor coil resistance estimation value at the first time is obtained and output by detecting the change of the motor coil resistance with external factors The second product of the motor coil resistance estimation value at the first time and the excitation current estimation value is calculated by the multiplier to obtain the motor coil resistance voltage drop estimation value at the first time The second product of the motor coil resistance estimation value at the first time and the excitation current estimation value is calculated by the multiplier to obtain the motor coil resistance voltage drop estimation value at the first time ​ by the observation unit using an adaptive sliding mode observer gain K or an adaptive Luenberger observer gain L d , an excitation current estimation error at the first time point , the excitation current estimation error at the first time point after processing is calculated by the second amplifier , a second inverse number of the reverse electromotive force estimation value at the first time point is obtained by multiplying the excitation current estimation error at the first time point by the second gain L The mixing operator is also configured to receive a current time instant driving voltage V(t n ) and a second opposite number at a previous time instant as positive inputs, and receive a motor coil resistance voltage drop estimation value at the previous time instant as a negative input, to obtain an excitation voltage estimation value of the motor coil at each subsequent time instant; and the first amplifier is also configured to calculate a product of the excitation voltage estimation value at each subsequent time instant and a first gain 1 / L, to obtain an excitation current rate of change estimation value at each subsequent time instant The integrator is also configured to perform an integral operation on the excitation current rate of change estimation value at each subsequent time instant to obtain an excitation current estimation value at each subsequent time instant The two-input subtractor is also configured to calculate a difference between the driving current I(t n ) and the estimated excitation current at each of the subsequent time points, to obtain an estimated excitation current error at each of the subsequent time points. The parameter identification module is also configured to perform iterative calculation using the estimated excitation current and the estimated excitation current error at each of the subsequent time points, to detect a change in the motor coil resistance caused by external factors, and to obtain and output an estimated motor coil resistance value at each of the subsequent time points. The multiplier is also configured to calculate a second product of the estimated motor coil resistance value and the estimated excitation current at each of the subsequent time points, to obtain an estimated motor coil resistance voltage drop value at each of the subsequent time points. The observation unit is also configured to utilize an adaptive sliding mode observer gain K or an adaptive Luenberger observer gain L d The excitation current estimation error of the subsequent time The second amplifier is also configured to calculate the excitation current estimation error of the subsequent time after processing The second amplifier is also configured to calculate the excitation current estimation error of the subsequent time after processing The second amplifier is also configured to calculate the excitation current estimation error of the subsequent time after processing The third amplifier calculates a first product of the second reciprocal number and -1, and the low-pass filter obtains and outputs an estimated value of a back electromotive force after filtering out ripple generated by the adaptive observer based on the first product.

Images