A rotating state detection method and system applied to a flywheel motor

Abstract

The application discloses a rotating state detection method and system applied to a flywheel motor, relates to the flywheel detection technical field, and comprises the following steps: superimposing a high-frequency voltage signal in the motor; detecting a response current flowing through a winding of the motor; demodulating the response current to obtain a low-frequency envelope signal; estimating rotor position information based on the low-frequency envelope signal; the rotor position information comprises a rotor electric angle and a rotor electric angular velocity; and performing dynamic eccentricity detection based on the low-frequency envelope signal and the estimated rotor position. Through the processing scheme of the application, the flywheel motor state can be monitored without additionally installing sensors, and sensorless control in zero-speed and low-speed states is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of flywheel detection, in particular to a rotating state detection method and system applied to a flywheel motor. BACKGROUND

[0002] As a kind of high efficiency, environmental protection, long service life physical energy storage device, flywheel energy storage system shows significant advantages in power grid frequency modulation, uninterrupted power supply (UPS), energy recovery and other fields. The core of the system is the high-speed rotating flywheel motor, whose speed can be as high as tens of thousands of revolutions per minute. The accurate detection of the rotating state of the rotor is the key prerequisite for realizing high-precision field-oriented control (FOC), ensuring stable and reliable operation of the system and maximizing energy storage efficiency.

[0003] At present, the detection of rotating state mainly uses photoelectric encoder or rotary transformer. Such sensors are directly installed on the motor shaft and can provide high-precision position signals. However, due to the extremely strict dynamic balance requirement of high-speed rotating flywheel, plus the mass of the sensor itself and its installation process, it is inevitable to introduce unbalance. This not only increases the risk of system vibration, but also limits the further improvement of the ultimate speed.

[0004] To overcome the drawbacks of physical sensors, sensorless control technology has emerged and become a mainstream research direction. The most widely used is the estimation method based on motor back electromotive force model, which mainly estimates the rotor position and speed by detecting the back electromotive force induced in the motor winding during operation. However, the magnitude of back electromotive force is proportional to the speed, and during the motor starting, low-speed or extremely low-speed operation stage, the amplitude of back electromotive force is extremely small, and the signal-to-noise ratio is extremely low, resulting in large position estimation error or even complete failure, leading to zero-speed / low-speed blind area.

[0005] In addition, whether it is a physical sensor or a traditional back electromotive force observer, its function is limited to providing basic position and speed feedback, and it cannot monitor the mechanical health status of the flywheel motor body online.

[0006] It can be seen that the above-mentioned rotating state detection method of the motor has the disadvantages of inconvenience and defects, and needs to be further improved. How to create a new rotating state detection method of the motor has become the goal of the current industry to improve. SUMMARY

[0007] Therefore, the present application provides a rotating state detection method applied to a flywheel motor, which at least partially solves the problems existing in the prior art.

[0008] In a first aspect, the present application provides a rotating state detection method applied to a flywheel motor, which comprises the following steps:

[0009] superimposing a high frequency voltage signal in the motor;

[0010] detecting a response current flowing through the motor winding;

[0011] demodulating the response current to obtain a low frequency envelope signal;

[0012] estimating rotor position information based on the low frequency envelope signal; the rotor position information comprises a rotor electrical angle and a rotor electrical angular velocity;

[0013] performing dynamic eccentricity detection based on the low frequency envelope signal and the estimated rotor position.

[0014] According to a specific implementation manner of the embodiment of the present disclosure, the high frequency voltage signal is defined based on the following method:

[0015] ;

[0016] wherein, is a high frequency voltage signal superimposed in a two-phase stationary coordinate system ; is an amplitude of the injected voltage; is an angular frequency of the injected signal, , is a frequency of the injected signal; is a time variable;

[0017] ;

[0018] ;

[0019] wherein, is a high frequency voltage signal varying in the axis; is a high frequency voltage signal varying in the axis;

[0020] According to a specific implementation manner of the embodiment of the present disclosure, the method further comprises:

[0021] mapping to a three-phase coordinate system c:

[0022] ;

[0023] ;

[0024] ;

[0025] wherein, is a high frequency voltage signal varying in the High-frequency voltage signal of the axis; for a three-phase coordinate system High-frequency voltage signal of the axis; for a three-phase coordinate system High-frequency voltage signal of the axis.

[0026] According to a specific implementation manner of the embodiment of the present disclosure, the detection of the response current flowing through the motor winding comprises:

[0027] Obtaining the current of the motor in a three-phase coordinate system;

[0028] Sampling the current based on an analog-to-digital converter and converting the sampled current into a digital quantity;

[0029] Converting the digital sampled current into a response current in a two-phase stationary coordinate system.

[0030] According to a specific implementation manner of the embodiment of the present disclosure, the conversion of the digital sampled current into a response current in a two-phase stationary coordinate system comprises:

[0031] Converting the digital sampled current into a response current in a two-phase stationary coordinate system based on the following method:

[0032] ;

[0033] ;

[0034] wherein, is a response current sequence of the axis in a two-phase stationary coordinate system ; is a response current sequence of the axis in a two-phase stationary coordinate system ; is a sampled current of the axis in a three-phase coordinate system ; is a sampled current of the axis in a three-phase coordinate system , ; is a sampled current of the axis in a three-phase coordinate system ; is a sampling index.

[0035] According to a specific implementation manner of the embodiment of the present disclosure, the demodulation of the response current to obtain a low-frequency envelope signal comprises:

[0036] Demodulating the response current based on the following method:

[0037] ;

[0038] wherein, is a mixed-frequency product signal; is the response current sequence of the axis in the two-phase stationary coordinate system; is the response current sequence of the axis in the two-phase stationary coordinate system; is the demodulation signal with the same frequency as the injected high-frequency voltage signal; is the angular frequency of the injected signal; is the time variable;

[0039] The low-frequency envelope signal is obtained based on the following method:

[0040] ;

[0041] wherein, is the low-frequency envelope signal; is the mixed-frequency product signal; is a low-pass filter.

[0042] According to a specific implementation manner of the embodiment of the present disclosure, the rotor position information is estimated based on the low-frequency envelope signal, comprising:

[0043] The low-frequency envelope signal is calculated based on the following method: function relationship between the low-frequency envelope signal and the rotor position:

[0044] ;

[0045] wherein, is a constant related to the amplitude of the injected voltage and the motor parameters; is the inductance difference; is the rotor position estimation value; is the actual rotor position value; is the actual rotor position value;

[0046] The is input to a phase-locked loop, so that when the position estimation error tends to zero, the rotor position estimation value is output.

[0047] According to a specific implementation manner of the embodiment of the present disclosure, the dynamic eccentricity detection is performed based on the low-frequency envelope signal and the estimated rotor position, comprising:

[0048] The instantaneous amplitude of the envelope signal is extracted based on the following method:

[0049] ;

[0050] wherein, is the instantaneous amplitude of the envelope signal; is the low-frequency envelope signal; is a Hilbert transform;

[0051] The instantaneous amplitude of the envelope signal​ performing a fast Fourier transform (FFT):

[0052] ;

[0053] wherein, is the distribution of the envelope signal amplitude in the frequency domain; is the instantaneous amplitude of the envelope signal; is a fast Fourier transform operator;

[0054] calculating a mechanical rotation frequency :

[0055] ;

[0056] wherein, is the mechanical rotation frequency; is an estimated value of the rotor electrical angular velocity; P is the number of motor pole pairs;

[0057] calculating an eccentricity index based on the following method:

[0058] ;

[0059] wherein, is the eccentricity index; is the amplitude at the mechanical rotation frequency ; is the sum of the amplitudes of all frequency components.

[0060] According to a specific implementation manner of the embodiment of the present disclosure, the method further comprises:

[0061] determining the severity of the fault according to the eccentricity index;

[0062] when , it is determined that the system is normal;

[0063] when , it is determined that there is slight eccentricity, a warning is given and monitoring is performed;

[0064] when , it is determined that there is obvious eccentricity, an alarm is given, and maintenance is performed according to a preset period;

[0065] when , it is determined that there is serious eccentricity, an emergency alarm is given, and immediate shutdown and inspection are performed.

[0066] In a second aspect, the embodiment of the present disclosure provides a rotating state detection system applied to a flywheel motor, the system comprising:

[0067] a voltage superposition module configured to superimpose a high-frequency voltage signal in the motor;

[0068] a current detection module configured to detect a response current flowing through the motor winding;

[0069] a demodulation module configured to demodulate the response current to obtain a low-frequency envelope signal;

[0070] a prediction module configured to estimate rotor position information based on the low-frequency envelope signal; the rotor position information comprising a rotor electrical angle and a rotor electrical angular velocity;

[0071] a detection module configured to perform dynamic eccentricity detection based on the low-frequency envelope signal and the estimated rotor position.

[0072] The rotating state detection method for a flywheel motor in the embodiments of the present disclosure synchronously realizes early online diagnosis of rotor dynamic eccentricity and other mechanical faults by injecting a high-frequency signal into the motor and demodulating the response thereof, by analyzing the response signal, realizes integration of control and health monitoring functions, and monitors the state of the flywheel motor without additionally installing sensors, thereby realizing sensorless control in a true zero-speed and low-speed state. BRIEF DESCRIPTION OF DRAWINGS

[0073] Figure 1 A rotating state detection method flowchart for a flywheel motor provided in the embodiments of the present disclosure;

[0074] Figure 2 A rotating state detection method flowchart for a flywheel motor provided in the embodiments of the present disclosure;

[0075] Figure 3 A rotating state detection system structure diagram for a flywheel motor provided in the embodiments of the present disclosure. DETAILED DESCRIPTION

[0076] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0077] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0078] It should be noted that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. Furthermore, this device and / or this method can be implemented using other structures and / or functionalities besides one or more of the aspects set forth herein.

[0079] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0080] This invention provides a method for detecting the rotational state of a flywheel motor. The method involves superimposing a high-frequency voltage signal into the motor; detecting the response current flowing through the motor windings; demodulating the response current to obtain a low-frequency envelope signal; estimating rotor position information based on the low-frequency envelope signal; the rotor position information including rotor electrical angle and rotor electrical angular velocity; and performing dynamic eccentricity detection based on the low-frequency envelope signal and the estimated rotor position. The low-frequency envelope signal can directly drive a phase-locked loop to estimate the rotor position for sensorless control; furthermore, its amplitude spectrum contains frequencies related to the mechanical rotation frequency. The relevant components can be used to quantitatively diagnose dynamic eccentricity faults in rotors.

[0081] Figure 1 This is a schematic diagram of the rotational state detection method for a flywheel motor provided in an embodiment of this disclosure.

[0082] Figure 2 To and Figure 1 The corresponding flowchart of the rotation state detection method applied to flywheel motors.

[0083] like Figure 1 As shown, in step S110, a high-frequency voltage signal is superimposed on the motor.

[0084] A high-frequency voltage signal is superimposed on the modulation signal of the three-phase voltage source inverter of the motor driver. This signal is injected synchronously with the motor's main drive signal, and the injection shaft is selected. axis.

[0085] In this embodiment of the invention, the high-frequency voltage signal is in a two-phase stationary coordinate system ( The coordinate system is defined as follows:

[0086] ;

[0087] in, is a high frequency voltage signal in the two-phase stationary coordinate system is a superimposed high frequency voltage signal; is the amplitude of the injected voltage, preferably 10-20 V; is the angular frequency of the injected signal, , is the frequency of the injected signal, is chosen much higher than the fundamental frequency of the motor and the switching frequency of the inverter, and much lower than the Nyquist limit of the inverter switching frequency, preferably 500 Hz-2500 Hz; is a time variable.

[0088] ;

[0089] ;

[0090] wherein, is a high frequency voltage signal in the two-phase stationary coordinate system axis; is a high frequency voltage signal in the two-phase stationary coordinate system axis.

[0091] In embodiments of the invention, the method further comprises mapping to a three-phase coordinate system c:

[0092] ;

[0093] ;

[0094] ;

[0095] wherein, is a high frequency voltage signal in the three-phase coordinate system axis; is a high frequency voltage signal in the three-phase coordinate system axis; is a high frequency voltage signal in the three-phase coordinate system axis.

[0096] Further, first, a continuously varying high frequency sinusoidal wave voltage signal is generated inside the controller.

[0097] The high frequency sinusoidal wave voltage signal is defined on the two-phase stationary coordinate system axis as:

[0098]

[0099]

[0100] wherein, is a high-frequency voltage signal varying only in axis; is the amplitude of the injected voltage, which needs to be large enough to generate a response current with a sufficient signal-to-noise ratio for subsequent detection, but must be small enough to avoid causing additional torque pulsation or audible noise, affecting the normal operation of the motor; is the frequency of the injected signal; t is the time variable, driven by the real-time clock of the controller.

[0101] outputs a high-frequency voltage signal sequence varying only in axis, axis is zero and .

[0102] Secondly, since the final motor is powered by three-phase voltage, the generated high-frequency signal needs to be mapped to the three-phase coordinate system (a-b-c coordinate system), and the coordinate transformation and signal mapping are realized through inverse Clarke transformation:

[0103]

[0104] Substituting , we get:

[0105]

[0106]

[0107]

[0108] Again, the high-frequency voltage signal is algebraically superimposed with the main drive voltage signal calculated by the motor control system :

[0109]

[0110]

[0111]

[0112] wherein, is the A-phase total voltage command; is the B-phase total voltage command; is the C-phase total voltage command; is the A-phase main drive voltage; is the B-phase main drive voltage; is the C-phase main drive voltage; is the high-frequency voltage signal injected into the A-phase; is the high-frequency voltage signal injected into the B-phase; is the high frequency voltage signal injected to phase C. The calculation method of the main driving voltage signal is not the focus of the present application, and is not described here.

[0113] Finally, the superimposed three-phase total voltage reference signal , , is subjected to space vector pulse width modulation (SVPWM) to generate a duty cycle signal, which in turn drives the six power switches of the three-phase inverter.

[0114] More specifically, next turn to step S120.

[0115] At step S120, the response current flowing through the motor winding is detected.

[0116] The original current sensor in the motor driver is used to detect the motor phase current at a sampling frequency much higher than , for example . Through the Clarke transformation, it is converted into the response current and in the two-phase stationary coordinate system.

[0117] In an embodiment of the present application, the response current flowing through the motor winding is detected, including: obtaining the current of the motor in the three-phase coordinate system; sampling the current based on an analog-to-digital converter, and converting the sampled current into a digital quantity; converting the digital sampled current into the response current in the two-phase stationary coordinate system.

[0118] In an embodiment of the present application, the digital sampled current is converted into the response current in the two-phase stationary coordinate system, including:

[0119] The digital sampled current is converted into the response current in the two-phase stationary coordinate system based on the following method:

[0120] ;

[0121] ;

[0122] wherein, is the response current sequence of the axis in the two-phase stationary coordinate system; is the response current sequence of the axis in the two-phase stationary coordinate system; is the sampled current of the axis in the three-phase coordinate system; is the sampled current of the axis in the three-phase coordinate system, ; is the sampled current of the​ the sampled current of the shaft; for the sampling index.

[0123] More specifically, first, using the current sensors already existing in the motor driver, the two-phase or three-phase current flowing through the motor winding is detected in real time. Sensor A is connected in series on the motor A-phase winding line to measure the A-phase current ; sensor B is connected in series on the motor B-phase winding line to measure the B-phase current ; according to Kirchhoff's current law , the C-phase current can be calculated.

[0124] The measured current contains a low-frequency main component generated by the main drive voltage , which is used to control the motor torque and speed, and a high-frequency response component generated by the injected high-frequency voltage , the amplitude and phase of which are modulated by the rotor position .

[0125] Second, to ensure the time consistency of the sampling points, the controller's ADC (analog-to-digital converter) synchronously samples the continuous current signal output by the sensor at a specific time and retains its voltage value. The sampling time is at the midpoint of the PWM switching period, at which time the current ripple and noise caused by the power tube switching action are minimal. The sampling frequency of the ADC satisfies the Nyquist sampling theorem, i.e. , the invention is preferred .

[0126] Third, the sampled analog signal is converted to a digital signal by the ADC to obtain a discrete-time sequence of three-phase current sampling values , where is the sampling index, , the corresponding sampling time is , , and the sampling period is .

[0127] Finally, the three-phase current sampling values are converted to current components and in the coordinate system by Clarke transformation:

[0128]

[0129]

[0130] Next, go to step S130.

[0131] In step S130, the response current is demodulated to obtain a low-frequency envelope signal.

[0132] In this embodiment of the invention, demodulating the response current to obtain a low-frequency envelope signal includes:

[0133] Will Shaft response current Demodulated signal at the same frequency Multiply, and pass the product through a low-pass filter (LPF) with its cutoff frequency set much lower than 1000 Hz. But higher than expected (For example, the cutoff frequency is 500 Hz).

[0134] The response current is demodulated using the following method:

[0135] ;

[0136] in, It is a mixed product signal; In a two-phase stationary coordinate system The response current sequence of the shaft; The demodulated signal is in sync with the injected high-frequency voltage signal; The angular frequency of the injected signal; It is a time variable;

[0137] The low-frequency envelope signal was obtained using the following method:

[0138] ;

[0139] in, It is a low-frequency envelope signal; It is a mixed product signal; It is a low-pass filter. A low-pass filter removes high-frequency components (around...). (Components), retaining only the low-frequency envelope signal related to rotor position. .

[0140] More specifically, firstly, within the digital controller, a high-frequency voltage signal is generated and injected. In this invention, a sinusoidal signal is selected from demodulated signals that are of the same frequency and in phase or out of phase. As a demodulation signal.

[0141]

[0142] in, It is a digitally demodulated signal; , The frequency of the injected signal.

[0143] Time of demodulation signal completely synchronized with the clock source of the high-frequency voltage signal.

[0144] Secondly, the response current of the axis is point-by-point multiplied with the digital demodulation signal:

[0145]

[0146] wherein, is the mixed-frequency product signal; is the response current sequence of the axis in the two-phase stationary coordinate system; is the angular frequency of the injected signal; is the sampling period; is the sampling point index. Thirdly, the mixed-frequency signal is filtered through a digital low-pass filter (LPF) to filter out the double-frequency and higher-frequency components and only retain the direct-current and low-frequency components, i.e., the required envelope signal. The filter herein usually uses a first-order or second-order infinite impulse response (IIR) filter or a finite impulse response (FIR) filter, such as a Butterworth filter.

[0147] The cut-off frequency of the LPF must satisfy:

[0148]

[0149]

[0150] Meanwhile,

[0151]

[0152] wherein, is the highest electrical frequency that the motor can have (usually the frequency corresponding to the highest speed).

[0153] The signal after filtering the cut-off is the demodulated envelope signal .

[0154] Next, go to step S140.

[0155] At step S140, estimate the rotor position information based on the low-frequency envelope signal; the rotor position information includes: rotor electrical angle and rotor electrical angular velocity.

[0156] The demodulated envelope signal has a functional relationship with the rotor position .

[0157] ​​​​​In the embodiments of the present application, the rotor position information is estimated based on the low-frequency envelope signal, comprising:

[0158] The low-frequency envelope signal is calculated based on the following method and the function relationship with the rotor position :

[0159] ;

[0160] wherein, is a constant positively correlated with the injected voltage amplitude and the motor parameters, and the specific value is calculated based on the test on the target motor; is the inductance difference (the difference between the quadrature axis inductance and the direct axis inductance), , is the quadrature axis inductance, is the direct axis inductance, which can be obtained by consulting the data manual or offline measurement method, ; is the rotor position estimation value; is the actual rotor position value;

[0161] The low-frequency envelope signal is input to the phase-locked loop (PLL), so that the position estimation error tends to zero. When the PLL is locked, at this time, the output of the PLL is the accurate rotor position estimation.

[0162] The rotor electrical angular velocity can be obtained by differentiating . This method is still effective when , because the response is generated by high-frequency excitation, overcoming the defects of traditional back-EMF method near zero speed.

[0163] More specifically, first, the envelope signal is directly used as a measure of the rotor position estimation error.

[0164] Based on the high-frequency model of the permanent magnet synchronous motor, there is the following approximate relationship:

[0165]

[0166] wherein, is the position error signal; is the rotor position estimation value; is the actual rotor position value.

[0167] When the position estimation is accurate, i.e. , the error signal When there is an estimation error, the amplitude of is proportional to the sine of the error angle. Within a small error range , the relationship can be approximated as linear: .

[0168] Secondly, a digital phase-locked loop (PLL) is used as a closed-loop controller, taking the error signal as input to dynamically adjust the position estimate and the speed estimate until the true rotor state is locked.

[0169] The PLL typically consists of a proportional-integral (PI) controller and an integrator, which can process the error signal and output a speed correction .

[0170]

[0171] where is the proportional gain; is the integral gain; is the control period (T); is the integer index; is the error value at the th sampling time. The values of

[0172] and are calculated by the critical proportionality method: 1. Set

[0173] to 0.

[0174] 2. Gradually increase until the system starts to oscillate with constant amplitude (i.e., at the stable critical point). Record the gain at this time as (the critical gain), and measure the oscillation period .

[0175] 3. Set the parameters according to the Ziegler-Nichols formula:

[0176]

[0177]

[0178] Add the speed correction to the initial speed estimate or the feedforward value to obtain the current electrical angular speed estimate . When the initial speed is 0, ; the current feedforward value is​ Time, .

[0179] The electrical angular velocity estimate is integrated to obtain the position estimate.

[0180]

[0181] Since the motor angle is periodic, when exceeds , a modulo operation is needed to limit it within .

[0182] Again, the rotor electrical angular velocity estimate has been directly estimated as an internal state of the PLL.

[0183] The conversion relationship between the electrical angular velocity estimate , the mechanical angular velocity estimate and the rotational speed estimate is:

[0184]

[0185]

[0186] where P is the number of pole pairs of the motor.

[0187] Finally, the final estimated rotational state parameters are output to the upper control system and monitoring unit.

[0188] The output rotor electrical angle estimate is used for Park transformation (transforming current from stationary to rotating coordinate system) and inverse Park transformation (transforming voltage from rotating to stationary coordinate system); the rotor electrical angular velocity estimate and / or the mechanical rotational speed are used for speed closed-loop feedback, system state monitoring and fault diagnosis.

[0189] Next, go to step S150.

[0190] At step S150, dynamic eccentricity detection is performed based on the low-frequency envelope signal and the estimated rotor position.

[0191] In embodiments of the present application, the dynamic eccentricity detection based on the low-frequency envelope signal and the estimated rotor position comprises:

[0192] The instantaneous amplitude thereof is extracted from the envelope signal based on the following method:

[0193] ;

[0194] wherein, is the instantaneous amplitude of the envelope signal; is the low frequency envelope signal; is the Hilbert transform;

[0195] the instantaneous amplitude of the envelope signal is subjected to a fast Fourier transform (FFT):

[0196] ;

[0197] wherein, is the distribution of the envelope signal amplitude in the frequency domain; is the instantaneous amplitude of the envelope signal; is the fast Fourier transform operator;

[0198] calculating the mechanical rotation frequency

[0199] ;

[0200] wherein, is the mechanical rotation frequency; is the rotor electrical angular velocity estimate; P is the number of motor pole pairs;

[0201] calculating the eccentricity index based on the following method:

[0202] ;

[0203] wherein, is the eccentricity index; is the amplitude at the mechanical rotation frequency ; is the sum of the amplitudes of all frequency components.

[0204] In the embodiments of the present application, the method further comprises: judging the fault severity according to the eccentricity index; when , judging that the system is normal; when , judging that there is slight eccentricity, giving a warning and suggesting monitoring; when , judging that there is obvious eccentricity, giving an alarm and performing maintenance according to a preset period; when , judging that there is serious eccentricity, giving an urgent alarm and immediately stopping for inspection.

[0205] More specifically, first, the instantaneous amplitude of the envelope signal is extracted:

[0206]

[0207] wherein, This represents the instantaneous amplitude of the envelope signal; It is a low-frequency envelope signal; This is the Hilbert transform.

[0208] Secondly, for Perform Fast Fourier Transform (FFT):

[0209]

[0210] in, This represents the distribution of the envelope signal amplitude in the frequency domain. This represents the instantaneous amplitude of the envelope signal; This is the Fast Fourier Transform operator.

[0211] Searching for the mechanical rotation frequency Spectral peaks at:

[0212]

[0213] in, The mechanical rotation frequency; is the estimated value of the rotor's electric angular velocity; P is the number of pole pairs of the motor, which can be obtained from the motor's nameplate.

[0214] Finally, the eccentricity index is calculated using the following method:

[0215]

[0216] in, Eccentricity index; To achieve the mechanical rotation frequency The amplitude at that point; This is the sum of the amplitudes of all frequency components.

[0217] Finally, the severity of the fault is determined based on the eccentricity index.

[0218] when When the system is in normal condition, it is determined that the system is functioning correctly.

[0219] when In case of slight eccentricity, issue an early warning and monitor the situation;

[0220] when If obvious eccentricity is detected, an alarm will be triggered, and maintenance will be performed according to the preset cycle;

[0221] when If severe eccentricity is detected, an emergency alarm will be triggered, and the machine will be immediately stopped for inspection.

[0222] The rotation state detection method applied to the flywheel motor provided by the application can not only provide high-precision rotation speed and position feedback, but also can realize deep insight into the mechanical health of the flywheel motor. The integration of state monitoring and fault early warning is realized without additional hardware.

[0223] Figure 3 The rotation state detection system 300 applied to the flywheel motor provided by the application is shown, which comprises a voltage superposition module 310, a current detection module 320, a demodulation module 330, a prediction module 340 and a detection module 350.

[0224] The voltage superposition module 310 is used for superimposing a high-frequency voltage signal in the motor;

[0225] The current detection module 320 is used for detecting a response current flowing through the winding of the motor;

[0226] The demodulation module 330 is used for demodulating the response current to obtain a low-frequency envelope signal;

[0227] The prediction module 340 is used for estimating rotor position information based on the low-frequency envelope signal; the rotor position information comprises a rotor electrical angle and a rotor electrical angular velocity;

[0228] The detection module 350 is used for performing dynamic eccentricity detection based on the low-frequency envelope signal and the estimated rotor position.

[0229] The above is only a specific embodiment of the present disclosure, but the protection scope of the present disclosure is not limited to this. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present disclosure, which should be covered within the protection scope of the present disclosure.

Claims

1. A method for detecting the rotational state of a flywheel motor, characterized in that, The method includes the following steps: A high-frequency voltage signal is superimposed on the motor; Detect the response current flowing through the motor windings; The response current is demodulated to obtain a low-frequency envelope signal; The rotor position information is estimated based on the low-frequency envelope signal; the rotor position information includes: rotor electrical angle and rotor electrical angular velocity; Dynamic eccentricity detection is performed based on the low-frequency envelope signal and the estimated rotor position. The demodulation of the response current to obtain the low-frequency envelope signal includes: The response current is demodulated using the following method: ； in, It is a mixed product signal; In a two-phase stationary coordinate system The response current sequence of the shaft; The demodulated signal is in sync with the injected high-frequency voltage signal; The angular frequency of the injected signal; It is a time variable; The low-frequency envelope signal was obtained using the following method: ； in, It is a low-frequency envelope signal; It is a mixed product signal; It is a low-pass filter.

2. The method for detecting the rotational state of a flywheel motor according to claim 1, characterized in that, The high-frequency voltage signal is defined based on the following method: ； in, In a two-phase stationary coordinate system Superimposed high-frequency voltage signals; The magnitude of the injected voltage; The angular frequency of the injected signal, , The frequency of the injected signal; It is a time variable; ； ； in, In order to be in High-frequency voltage signal with axis variation; In order to be in High-frequency voltage signal with axis variation.

3. The method for detecting the rotational state of a flywheel motor according to claim 2, characterized in that, The method further includes: Will Mapped to the three-phase coordinate system abc: ； ； ； in, In a three-phase coordinate system High-frequency voltage signal of the shaft; In a three-phase coordinate system High-frequency voltage signal of the shaft; In a three-phase coordinate system High-frequency voltage signal of the shaft.

4. The method for detecting the rotational state of a flywheel motor according to claim 1, characterized in that, The detection of the response current flowing through the motor windings includes: Obtain the motor current in a three-phase coordinate system; The current is sampled using an analog-to-digital converter, and the sampled current is converted into a digital quantity. The digitally sampled current is converted into the response current in a two-phase stationary coordinate system.

5. The method for detecting the rotational state of a flywheel motor according to claim 4, characterized in that, The process of converting the digitally sampled current into a response current in a two-phase stationary coordinate system includes: The digitally sampled current is converted into a response current in a two-phase stationary coordinate system using the following method: ； ； in, In a two-phase stationary coordinate system The response current sequence of the shaft; In a two-phase stationary coordinate system The response current sequence of the shaft; In a three-phase coordinate system The sampling current of the shaft; In a three-phase coordinate system The sampling current of the shaft, ; In a three-phase coordinate system The sampling current of the shaft; This is the sampling index.

6. The method for detecting the rotational state of a flywheel motor according to claim 1, characterized in that, Estimating rotor position information based on the low-frequency envelope signal includes: The low-frequency envelope signal is calculated using the following method. With rotor position Functional relationship: ； in, To match the injection voltage amplitude Constants related to motor parameters; Poor inductance; This is the estimated rotor position. This represents the actual rotor position. Will The input to the phase-locked loop reduces the position estimation error. When it approaches zero, the output rotor position estimate .

7. The method for detecting the rotational state of a flywheel motor according to claim 1, characterized in that, The dynamic eccentricity detection based on the low-frequency envelope signal and the estimated rotor position includes: Based on the following method, from the envelope signal Extract its instantaneous amplitude: ； in, This represents the instantaneous amplitude of the envelope signal; It is a low-frequency envelope signal; For Hilbert transform; Instantaneous amplitude of the envelope signal Perform Fast Fourier Transform (FFT): ； in, This represents the distribution of the envelope signal amplitude in the frequency domain. This represents the instantaneous amplitude of the envelope signal; For Fast Fourier Transform operators; Calculate the rotational frequency of the machine : ； in, The mechanical rotation frequency; Here is the estimated rotor electric angular velocity; P is the number of pole pairs of the motor. The eccentricity index is calculated using the following method: ； in, Eccentricity index; To achieve the mechanical rotation frequency The amplitude at that point; This is the sum of the amplitudes of all frequency components.

8. The method for detecting the rotational state of a flywheel motor according to claim 7, characterized in that, The method further includes: The severity of the fault is determined by the eccentricity index; when When the system is in normal condition, it is determined that the system is functioning correctly. when In case of slight eccentricity, issue an early warning and monitor the situation; when If obvious eccentricity is detected, an alarm will be triggered, and maintenance will be performed according to the preset cycle; when If severe eccentricity is detected, an emergency alarm will be triggered, and the machine will be immediately stopped for inspection.

9. A rotational state detection system for a flywheel motor, characterized in that, The system includes: A voltage superposition module is configured to superimpose high-frequency voltage signals into a motor; A current detection module is configured to detect the response current flowing through the motor windings; The demodulation module is configured to demodulate the response current to obtain a low-frequency envelope signal; The demodulation of the response current to obtain the low-frequency envelope signal includes: The response current is demodulated using the following method: ； in, It is a mixed product signal; In a two-phase stationary coordinate system The response current sequence of the shaft; The demodulated signal is in sync with the injected high-frequency voltage signal; The angular frequency of the injected signal; It is a time variable; The low-frequency envelope signal was obtained using the following method: ； in, It is a low-frequency envelope signal; It is a mixed product signal; It is a low-pass filter; The prediction module is configured to estimate rotor position information based on the low-frequency envelope signal; the rotor position information includes: rotor electrical angle and rotor electrical angular velocity; The detection module is configured to perform dynamic eccentricity detection based on the low-frequency envelope signal and the estimated rotor position.

Images