Power supply rapid switching method and device based on minimum beat voltage characteristics

Abstract

The invention discloses a power supply rapid switching method and device based on minimum beat voltage characteristics, and the method comprises the steps: responding to the rapid switching start of a power supply, obtaining the voltage signal of each sampling point on a bus, and calculating the frequency, phase and amplitude of the bus voltage based on the Fourier phasor by adjusting the equal-angle sampling time according to the bus voltage sampling data; obtaining a bus voltage phase predicted value and a bus voltage amplitude predicted value according to the bus voltage phase, frequency and amplitude; and obtaining an estimated voltage phase difference according to the voltage phase predicted value and the voltage amplitude predicted value, predicting a moment when the beat voltage is minimum according to the estimated voltage phase difference, and determining a moment when a switching-on instruction is sent out so as to realize rapid switching of a power supply. Compared with the prior art, more accurate power supply rapid switching under adaptive equal-angle sampling adjustment is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of rapid power supply switching, and particularly to a rapid power supply switching method based on improved Fourier frequency measurement and minimum beat voltage feature prediction. Background Art

[0002] The rapid power supply switching technology is a key technology for ensuring the reliable power supply of asynchronous motors. When the main power supply loses power, due to the inertial rotation power generation of the asynchronous motor group and the asynchronous attenuation of the rotational speed, a residual voltage with a rapidly dropping amplitude will appear on the bus, and the phase difference Δθ and voltage vector difference ΔU between the residual voltage and the standby power supply are also changing rapidly. Precise switching needs to be carried out according to Δθ and ΔU to avoid excessive impact current. The capture synchronization switching method is a core technology to ensure a smooth transition during the power supply switching process by closing the switch as much as possible when the voltages of the residual voltage and the standby power supply are synchronized (Δθ≈0°). However, the closing angle error of the capture synchronization switching technology commonly used in the industry at present still reaches ±10°, and this error may lead to excessive impact current, thereby causing the switching to fail.

[0003] Therefore, how to optimize the rapid power supply switching strategy by improving the prediction accuracy of the closing angle is the key point of the technical improvement urgently needed in the present invention. Summary of the Invention

[0004] In order to overcome the technical problem of reducing the impact current during rapid power supply switching in the prior art, the present invention aims to propose a rapid power supply switching method and device based on the minimum beat voltage feature, adaptively adjust the equal-angle sampling time based on improved Fourier frequency measurement, obtain the voltage phase prediction value and the voltage amplitude prediction value, obtain the estimated voltage phase difference according to the voltage phase prediction value and the voltage amplitude prediction value, and perform a rapid power supply switching method with the minimum beat voltage feature, thereby realizing more precise rapid power supply switching under adaptive equal-angle sampling adjustment.

[0005] In order to achieve the above invention purpose, the present invention proposes the following technical solutions:

[0006] In the first aspect, the present invention proposes a rapid power supply switching method based on the minimum beat voltage feature, including:

[0007] Respond to the start of rapid power supply switching and obtain the voltage signals of each sampling point on the bus;

[0008] Perform adaptive adjustment of the equal-angle sampling time according to the voltage signals, and calculate the voltage frequency, voltage phase and voltage amplitude according to the discrete Fourier voltage phasor of the voltage signals within the adjusted equal-angle sampling time;

[0009] Obtain the voltage phase prediction value and the voltage amplitude prediction value according to the voltage amplitude, the voltage phase and the voltage frequency;

[0010] Based on the predicted voltage phase value and the predicted voltage amplitude value, an estimated voltage phase difference is obtained. According to the estimated voltage phase difference, the moment when the beat voltage is minimized is predicted, and the moment to issue the closing command is determined to achieve fast switching of the power supply.

[0011] In some embodiments, the step of adaptively adjusting the equal-angle sampling time according to the voltage signal and calculating the voltage frequency, voltage phase, and voltage amplitude based on the discrete Fourier voltage phasor of the voltage signal within the adjusted equal-angle sampling time further includes:

[0012] Taking the voltage signal as the input, setting the number of sampling points in each sampling sliding window, calculating the phasor of the initial bus voltage signal at each sampling point, and based on the improved adaptive Fourier frequency measurement: obtaining the predicted value of the voltage frequency at the next sampling point of the first sampling sliding window according to the historical voltage frequency before power failure;

[0013] Based on the predicted value of the voltage frequency at the next sampling point, adjusting the equal-angle sampling time series, determining the next equal-angle sampling time interval, and resampling the voltage signal at this sampling point by linear interpolation according to the time interval to achieve the next equal-angle sampling;

[0014] Repeat the above steps until the predicted values of the bus voltage frequencies at all sampling points are obtained based on the exponential fitting method within all sampling sliding windows.

[0015] In some embodiments, the predicted value of the voltage frequency at the next sampling point corresponds to the voltage frequency of the sampling point at the moment, where t k-1 is the time corresponding to the previous sampling point, N is the number of sampling points, and f k-1 is the voltage frequency corresponding to the previous sampling point.

[0016] In some embodiments, the voltage frequency, voltage phase, and voltage amplitude are respectively as follows:

[0017] The voltage amplitude is as shown in the following formula:

[0018]

[0019] where F(r + 1) is the discrete Fourier transform phasor of the voltage signal of the r-th sliding window, F(r) is the discrete Fourier transform phasor of the voltage signal of the r-th sliding window, f k is the actual frequency of the current sampling point, Δt is the time interval between adjacent sampling points, v* i+k-N is the voltage signal of the sampling points within the normalized equal-angle sampling window, N is the number of sampling points, and j is the imaginary number;

[0020] The voltage frequency fk As shown in the following formula:

[0021]

[0022] where f 0k is the predicted frequency value of the next sampling point

[0023] The voltage phase value is as shown in the following formula;

[0024]

[0025] where F(r + 1) is the discrete Fourier transform phasor of the voltage signal of the (r + 1)-th sliding window, and angle() is the phase angle calculation function.

[0026] In some embodiments, it further includes calculating the predicted beat voltage ΔU(t), and the formula is as follows:

[0027]

[0028] where U s is the amplitude of the power supply voltage, U m (t) is the amplitude of the residual voltage U L , Δθ(t) is the phase difference predicted based on exponential fitting, and ΔU(t) is the function of the beat voltage changing with time;

[0029] Calculate the minimum point of the beat voltage using the derivative of ΔU 2 (t) [ΔU 2 (t s )]' == 0, and the time of the minimum beat voltage is obtained as t s ;

[0030] According to the closing fixed time t c , determine that the time to issue the closing command is t s -t c .

[0031] In a second aspect, the present invention proposes a power supply fast switching device based on the minimum beat voltage feature, including a sampling module, a measurement module, a prediction module, and a determination module; where:

[0032] The sampling module is responsible for responding to the start of the power supply fast switching and obtaining the voltage signals of each sampling point on the bus;

[0033] The measurement module is responsible for adaptively adjusting the equal-angle sampling time according to the voltage signal, and calculating the voltage frequency, voltage phase, and voltage amplitude according to the discrete Fourier voltage phasor of the voltage signal within the adjusted equal-angle sampling time;

[0034] A prediction module, responsible for obtaining a predicted voltage phase value and a predicted voltage amplitude value according to the voltage amplitude, the voltage phase, and the voltage frequency;

[0035] A determination module, responsible for obtaining an estimated voltage phase difference according to the predicted voltage phase value and the predicted voltage amplitude value, predicting the moment when the beat voltage is minimized according to the estimated voltage phase difference, and determining the moment to issue a closing command to achieve fast switching of the power supply.

[0036] In some embodiments, the step of adaptively adjusting the equal-angle sampling time according to the voltage signal and calculating the voltage frequency, the voltage phase, and the voltage amplitude according to the discrete Fourier voltage phasor of the voltage signal within the adjusted equal-angle sampling time further includes:

[0037] Taking the voltage signal as an input, setting the number of sampling points in each sampling sliding window, calculating the phasor of the initial bus voltage signal at each sampling point, and based on an improved adaptive Fourier frequency measurement: considering the voltage frequency attenuation to obtain the voltage phase interval angle between adjacent bus sampling points, and obtaining the predicted value of the voltage frequency at the next sampling point of the first sampling sliding window according to the historical voltage frequency before power loss;

[0038] Based on the predicted value of the voltage frequency at the next sampling point, adjusting the equal-angle sampling time series, determining the next equal-angle sampling time interval, and resampling the voltage signal at this sampling point through linear interpolation according to the time interval to achieve the next equal-angle sampling;

[0039] Repeat the above steps until the predicted values of the bus voltage frequencies at all sampling points are obtained based on the exponential fitting method within all sampling sliding windows.

[0040] In some embodiments, the predicted value of the voltage frequency at the next sampling point corresponds to the voltage frequency of the sampling point at the moment, where t k-1 is the time corresponding to the previous sampling point, N is the number of sampling points, and f k-1 is the voltage frequency corresponding to the previous sampling point.

[0041] In some embodiments, the voltage frequency, the voltage phase, and the voltage amplitude are respectively as follows:

[0042] The voltage amplitude is as shown in the following formula:

[0043]

[0044]

[0045] Among them, F(r + 1) is the discrete Fourier transform phasor of the voltage signal of the r-th sliding window, F(r) is the discrete Fourier transform phasor of the voltage signal of the r-th sliding window, f k is the actual frequency of the current sampling point, Δt is the time interval between adjacent sampling points, v* i+k-N is the voltage signal of the sampling points within the normalized equal-angle sampling window, N is the number of sampling points, and j is the imaginary number;

[0046] The voltage frequency f k is as shown in the following formula:

[0047]

[0048] Among them, f 0k is the frequency prediction value of the next sampling point

[0049] The voltage phase value is as shown in the following formula;

[0050]

[0051] Among them, F(r + 1) is the discrete Fourier transform phasor of the voltage signal of the (r + 1)-th sliding window, and angle() is the phase angle function.

[0052] In some embodiments, it further includes calculating the predicted beat voltage ΔU(t), and the formula is as follows:

[0053]

[0054] Among them, U s is the amplitude of the power supply voltage, U m (t) is the amplitude of the residual voltage U L , Δθ(t) is the phase difference predicted based on exponential fitting, and ΔU(t) is the function of the beat voltage changing with time;

[0055] Using the derivative of ΔU 2 (t) to calculate the minimum value point of the beat voltage [ΔU 2 (t s )]' == 0, and the moment of the minimum beat voltage is obtained as t s ;

[0056] According to the closing fixed time t c , the moment of sending the closing command is determined as t s -t c .

[0057] Based on the prior art, the present invention can obtain the following beneficial technical effects:

[0058] 1) Based on the improved adaptive (equal-angle) Fourier frequency measurement method, the predicted value of the voltage frequency at the next sampling point is measured in real time, the equal-angle sampling time series is adjusted, the next equal-angle sampling time interval is determined, and the voltage signal at this sampling point is obtained through linear interpolation resampling according to the time interval, realizing the next sampling at equal angles, which is convenient for considering the attenuation of the voltage frequency to obtain the angle of the voltage phase interval between adjacent bus sampling points, and helps to improve the sampling efficiency;

[0059] 2) Obtain the predicted voltage phase difference according to the predicted value of the voltage phase and the predicted value of the voltage amplitude, predict the moment when the beat voltage is the smallest according to the predicted voltage phase difference, and accurately predict the inherent closing delay of the circuit breaker in advance, which helps to achieve more accurate and rapid power source switching. Description of the Drawings

[0060] Figure 1 It is a flowchart of a method for rapid power source switching based on the characteristics of the minimum beat voltage of the present invention;

[0061] Figure 2 It is a flowchart of an embodiment of a method for rapid power source switching based on the characteristics of the minimum beat voltage of the present invention;

[0062] Figure 3 It is a schematic diagram of the residual voltage of the bus after the asynchronous motor loses power in the embodiment of the present invention;

[0063] Figure 4 It is a structural diagram of a device for rapid power source switching based on the characteristics of the minimum beat voltage of the present invention. Detailed Embodiments

[0064] The present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0065] As Figure 1 shown, a method for rapid power source switching based on the characteristics of the minimum beat voltage proposed by the present invention includes:

[0066] S1. In response to the start of rapid power source switching, obtain the voltage signal of the bus sampling point;

[0067] S2. Input the voltage signal, adjust the equal-angle sampling time interval according to the voltage signal, and calculate the frequency, phase and amplitude of the voltage based on the discrete Fourier phasor of the voltage signal; As Figure 2 shown, this process further specifically includes the steps:

[0068] S2.1. Take the voltage signal as the input, set the number of sampling points N for each sampling sliding window, calculate the phasor of the initial bus voltage signal at each sampling point, obtain the voltage phase interval angle between adjacent bus sampling points based on the improved adaptive Fourier frequency measurement, i.e., considering the voltage frequency attenuation, and obtain the voltage frequency prediction value of the N sampling points in the first sampling sliding window according to the historical voltage frequency before the power outage.

[0069] S2.2. Determine the next equal-angle sampling time interval according to the voltage frequency prediction value (if currently in the second sliding window, directly according to the power frequency), achieve approximate equal-angle sampling, and resample the voltage signal at this sampling point by linear interpolation according to the time interval.

[0070] S2.3. Set a sliding window and update the sliding window at each sampling point.

[0071] S2.4. Calculate the discrete Fourier transform voltage amplitude, and calculate the voltage amplitude, frequency and phase of the sampling point according to the included angle between the voltage phasor of the discrete Fourier transform in the next sliding window and the predicted voltage phasor of the discrete Fourier transform.

[0072] S2.5. Determine whether there is a next sampling sliding window.

[0073] S2.6. Update to the next sampling sliding window and continue the above process.

[0074] S2.7. Until the calculation of the voltage amplitude, voltage frequency and voltage phase of the current sampling point is completed.

[0075] S3. Based on the voltage amplitude, the voltage phase and the voltage frequency, obtain the voltage phase prediction value and the voltage amplitude prediction value by exponential fitting.

[0076] S4. According to the voltage phase prediction value and the voltage amplitude prediction value, predict the moment when the beat voltage is the smallest, and determine the moment to issue the closing command according to the fixed time of the circuit breaker closing; including according to the voltage phase prediction value and the voltage amplitude prediction value obtained in S3, according to the power supply fast switching strategy predicted based on the minimum beat voltage characteristic, predict the moment when the beat voltage is the smallest, and determine the moment to issue the closing command according to the fixed time of the circuit breaker closing.

[0077] Specifically, in step 1, parameters such as the bus voltage and current are monitored in real time. When it is detected that the three-phase bus voltages are all lower than the undervoltage start setting value and there is no current in the incoming line, it is determined that the bus has lost power, and at this time, the fast switching device needs to be started.

[0078] Specifically, in step 2, equal-angle sampling is achieved through linear interpolation, that is, the sampling time series is adaptively adjusted according to the frequency change of the voltage signal, so that the phase difference between sampling points is approximately equal to meet the Fourier full-cycle frequency measurement condition, thus adapting to the dynamic frequency measurement requirements.

[0079] The formal definition of the present invention is specifically as follows:

[0080] The initial voltage signal of the bus voltage signal at each sampling point is calculated and expressed as:

[0081] v(t) = v m (t)sin(φ(t)) (1)

[0082] where v(t) is the function of the voltage signal value with respect to time t, v m (t) is the function of the voltage amplitude value with respect to time t, and φ(t) is the function of the voltage phase value with respect to time t.

[0083] The phasor form corresponding to the initial voltage signal is expressed as: V(t) = v m (t)e jφ(t) , where V(t) is the function of the voltage signal phasor with respect to time t;

[0084] Let the number of sampling points in each sampling window be N, k be the sampling point serial number, v k represents the voltage signal value at the kth sampling point. The voltage signal values of the same-phase sampling points in adjacent cycles satisfy v k = v k-N , and the angle difference satisfies φ k - φ k-N = 2π. Let the sampling points be equally angularly distributed to achieve equal-angle sampling, then the voltage phase interval angle Δφ between adjacent bus sampling points is:

[0085]

[0086] The voltage frequency after power failure is set as f(t) = ae -bt , and the attenuation has an exponential law. Among them, f(t) is the function of the voltage frequency with respect to time t, a and b are the parameters of the exponential expression, t k is the time of the kth sampling point, and t k-1 is the time of the (k - 1)th sampling point;

[0087] Substitute the voltage frequency attenuation, then the expression of the voltage phase interval angle between adjacent bus sampling points is as follows:

[0088]

[0089] According to the Taylor formula, it can be approximately deduced:

[0090]

[0091] Among them, Δt is the time interval between adjacent sampling points;

[0092] Let be f 0k and f 0k correspond to the time which can be approximated as Based on the historical frequency measurement values, perform exponential fitting to determine the predicted voltage frequency f 0k of the next sampling point k; where t k-1 is the time corresponding to the previous sampling point, N is the number of sampling points, and f k-1 is the voltage frequency corresponding to the previous sampling point;

[0093] According to this, the sampling time interval at this time can be obtained as:

[0094]

[0095] From the predicted frequency f 0k obtain the sampling time t k of the next sampling point as: Furthermore, through linear interpolation resampling, obtain the voltage signal value v k at this sampling point to achieve approximate equal-angle sampling.

[0096] Calculate the discrete Fourier transform phasor of the voltage signal as follows:

[0097]

[0098] where X is the complex representation of the voltage signal on the main frequency component, v k-N+1 , v k-N+2. ,..., v k、 are the voltage signal values of N equal-angle sampling points;

[0099] Using Euler's formula, we can obtain:

[0100]

[0101] Calculate the real part A as:

[0102]

[0103] Calculate the imaginary part B as:

[0104]

[0105] From this, calculate the modulus value of the complex representation X of the voltage signal on the main frequency component, that is, the voltage amplitude v m , as follows:

[0106]

[0107] During the power-off transient process of an induction motor, as Figure 3 shown is the schematic diagram of the busbar residual voltage after the induction motor loses power. Since the amplitude of the busbar residual voltage decays continuously, the frequency measurement result will be affected. Therefore, in the above process, the voltage phase information θ k and the amplitude information v k of the voltage signal at the k-th sampling point are calculated through Fourier transform for the phasor v mk , and it is normalized to v* k = v k / v mk to avoid the influence of amplitude decay on frequency measurement, thereby improving the accuracy of frequency calculation.

[0108] Set the step size of the sliding window to 1, that is, update the sliding window (equal-angle sampling window) at each sampling point. When the sampling point number is k - 1, it is located in the r = k - N-th sliding window, which contains (t k-N , v* k-N ), (t k-N+1 , v* k-N+1 )...(t k-1 , v* k-1 ) these N sampling points. Let the initial phase of v* k-N be φ k-N . According to the discrete Fourier transform, define the phasor F(r):

[0109]

[0110] where F(r) is the discrete Fourier transform phasor of the voltage signal of the r-th sliding window, and v* i+k-N = sin(φ k-N + iΔφ) (i = 0, 1,..., N - 1) is the voltage signal of the sampling point within the normalized equal-angle sampling window;

[0111] Derived from F(r), it is further expressed as:

[0112]

[0113] where V k-1 is the phasor of the voltage signal corresponding to the last sampling point of the sliding window after normalization;

[0114] Then the discrete Fourier transform voltage phasor of the (r + 1)-th sliding window containing the normalized k-th sampling point voltage phasor v* k has the expression:

[0115]

[0116] Among them, F(r + 1) is the discrete Fourier transform voltage phasor of the (r + 1)-th window;

[0117] Considering that the voltage phasor rotates according to the frequency, F(r + 1) can also be expressed as:

[0118]

[0119] where f k is the actual frequency of the current sampling point.

[0120] Let the estimated value of the discrete Fourier transform voltage phasor of the (r + 1)-th sliding window be:

[0121]

[0122] F0(r + 1) is the discrete Fourier transform voltage phasor of the (r + 1)-th window estimated according to f 0k

[0123] The phase difference between F0(r + 1) and F(r + 1) reflects the difference between the frequency estimated value f 0k and the actual frequency f of the current sampling point k . F(r + 1) has a leading angle relative to F0(r + 1). Calculate the included angle Δσ between the voltage phasor F(r + 1) of the discrete Fourier transform in the next sliding window and the estimated discrete Fourier transform voltage phasor F0(r + 1), and the expression is as follows:

[0124]

[0125] Simplify Equation (11), and the voltage frequency of the sampling point k can be obtained as shown in the following formula:

[0126]

[0127] Calculate the phase value of the current voltage signal as shown in the following formula;

[0128]

[0129] Through the above process derivation, the voltage amplitude, phase and frequency of the sampling point are obtained.

[0130] Specifically, in step 3, according to the law that the bus voltage frequency decays exponentially with time constant after power failure, and based on the measured historical voltage frequency and voltage phase, the future frequency change is predicted by exponential fitting. The frequency is expressed as:

[0131] f(t) = ae -bt (17)

[0132] ​Among them, f(t) is a function of the voltage frequency with respect to time t, and a and b are parameters of the exponential expression;

[0133] The exponential fitting transforms the exponential relationship into a linear problem ln(f) = ln(a) - bt through logarithmic transformation. Subsequently, by minimizing the sum of squared errors (the amount of frequency and time data is n), linear fitting is performed.

[0134] E is the sum of squared errors, and n is the size of the data volume.

[0135] Let the partial derivatives of a and b be 0, that is:

[0136]

[0137] The parameters a and b can be obtained, and based on this, the phase change of the bus residual voltage is predicted. The fitting method for the amplitude v m (t) is the same. Predicting the phase and amplitude based on exponential fitting is more in line with the transient mechanism after the asynchronous motor loses power, with higher prediction accuracy and lower computational complexity.

[0138] Assume the current time is t0. According to the phase of the bus residual voltage (t0) measured by Fourier and the phase of the standby power supply voltage δ(t0) measured by Fourier, the phase difference θ(t0) between the bus residual voltage and the standby power supply voltage at the current time can be calculated:

[0139] θ(t0) = δ(t0) - (t0) (19)

[0140] Find the angle of the phase change of the bus residual voltage within the time (t0, t) The angle of the phase change of the standby power supply voltage and the change amount of the phase difference between the bus residual voltage and the standby power supply voltage

[0141]

[0142]

[0143]

[0144] In the formula, f N is the rated frequency of the power supply voltage, usually taken as 50 Hz.

[0145] Furthermore, if the phase difference calculated by Fourier transform at this time is θ(t0), then the voltage phase difference Δθ(t) at time t can be estimated as: Δθ(t) is the expression of the phase difference between the bus residual voltage and the standby power supply voltage predicted by exponential fitting with respect to time

[0146] Specifically, in step 4, the beat voltage is directly related to the impact voltage and impact current generated during power supply switching. Since the amplitude attenuation degree of the bus residual voltage is usually large, the minimum point of the beat voltage often appears before the 0° closing point. Let the amplitude of the power supply voltage be U s , the residual voltage U L amplitude is v m (t). According to the cosine theorem, the magnitude of the beat voltage ΔU(t) can be obtained:

[0147]

[0148] Among them, ΔU(t) is the expression of the predicted beat voltage varying with time, Δθ(t) is the expression of the phase difference between the predicted bus residual voltage and the standby power supply voltage varying with time by exponential fitting, and U s is the amplitude of the standby power supply voltage.

[0149] Substituting the phase difference Δθ(t) predicted by exponential fitting and the amplitude U m (t) gives the function ΔU(t) of the beat voltage varying with time. Using the derivative of ΔU 2 (t), the minimum value point of the beat voltage can be calculated as [ΔU 2 (t s )]' == 0, that is, the moment of the minimum beat voltage is t s . According to the fixed closing time t c , the moment to issue the closing command is determined as t s - t c .

[0150] Embodiment 2: A fast power supply switching device based on the characteristics of the minimum beat voltage proposed by the present invention includes a sampling module 100, a measurement module 200, a prediction module 300, and a determination module 400. Among them:

[0151] The sampling module 100 is responsible for responding to the start of fast power supply switching and acquiring the voltage signals of each sampling point on the bus;

[0152] The measurement module 200 is responsible for adjusting the equal-angle sampling time according to the voltage signal, and calculating the voltage frequency, voltage phase, and voltage amplitude according to the discrete Fourier voltage phasor of the voltage signal within the adjusted equal-angle sampling time;

[0153] The prediction module 300 is responsible for obtaining the predicted voltage phase value and the predicted voltage amplitude value according to the voltage phase and the voltage frequency;

[0154] The determining module 400 is responsible for obtaining an estimated voltage phase difference according to the predicted voltage phase value and the predicted voltage amplitude value, predicting the moment when the beat voltage is minimized according to the estimated voltage phase difference, and determining the moment to issue a closing command, so as to achieve fast switching of the power supply.

[0155] It should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, various changes and modifications can be made to the present invention. Any modification, equivalent replacement or improvement made under the spirit and principle of the present invention shall fall within the protection scope of the present invention.

Claims

1. A power supply fast switching method based on the characteristics of the minimum beat voltage, characterized in that, including: responding to the startup of rapid power supply switching, and acquiring voltage signals of each sampling point on the bus; performing adaptive adjustment of equal-angle sampling time according to the voltage signals, and calculating the voltage frequency, voltage phase and voltage amplitude according to the discrete Fourier voltage phasor of the voltage signals within the adjusted equal-angle sampling time; obtaining a predicted voltage phase value and a predicted voltage amplitude value according to the voltage amplitude, the voltage phase and the voltage frequency; obtaining a predicted voltage phase difference according to the predicted voltage phase value and the predicted voltage amplitude value, predicting the moment when the beat voltage is minimum according to the predicted voltage phase difference, and determining the moment to issue a closing command so as to achieve rapid power supply switching.

2. The power supply rapid switching method based on the minimum beat voltage characteristic according to claim 1 is characterized in that, The step of performing adaptive adjustment of equal-angle sampling time according to the voltage signals and calculating the voltage frequency, voltage phase and voltage amplitude according to the discrete Fourier voltage phasor of the voltage signals within the adjusted equal-angle sampling time further includes: taking the voltage signals as inputs, setting the number of sampling points in each sampling sliding window, calculating the phasor of the initial bus voltage signal at each sampling point, and based on an improved adaptive Fourier frequency measurement: obtaining a predicted value of the voltage frequency at the next sampling point of the first sampling sliding window according to the historical voltage frequency before power loss; based on the predicted value of the voltage frequency at the next sampling point, adjusting the equal-angle sampling time sequence, determining the next equal-angle sampling time interval, and resampling the voltage signals at this sampling point by linear interpolation according to the time interval to achieve the next sampling at equal angles; repeating the above steps until the predicted values of the bus voltage frequencies at all sampling points are obtained based on the exponential fitting method within all sampling sliding windows.

3. The power supply fast switching method based on the minimum beat voltage characteristic according to claim 2, characterized in that, The voltage frequency estimation value of the next sampling point corresponds to The voltage frequency at the sampling point at time t k-1 is the time corresponding to the previous sampling point, N is the number of sampling points, f k-1 is the voltage frequency corresponding to the previous sampling point.

4. A power supply rapid switching method based on the minimum beat voltage feature according to claim 2, characterized in that The voltage frequency, voltage phase and voltage amplitude are respectively as follows: The voltage amplitude is as shown in the following formula: Among them, F(r + 1) is the discrete Fourier transform phasor of the voltage signal of the r-th sliding window, F(r) is the discrete Fourier transform phasor of the voltage signal of the r-th sliding window, f k is the actual frequency of the current sampling point, Δt is the time interval between adjacent sampling points, v* i+k-N is the voltage signal of the sampling points within the normalized equal-angle sampling window, N is the number of sampling points, and j is the imaginary number; The voltage frequency f k is as follows: where f 0k is the estimated frequency value of the next sampling point The voltage phase value is as shown in the following formula; where F(r + 1) is the discrete Fourier transform phasor of the voltage signal of the (r + 1)-th sliding window, and angle() is a function for calculating the phase angle.

5. A power supply fast switching method based on the minimum beat voltage characteristic according to claim 1, characterized in that, further including calculating the predicted beat voltage ΔU(t), and the formula is as follows: Among them, U s is the amplitude of the power supply voltage, U m (t) is the amplitude of the residual voltage U L , Δθ(t) is the phase difference predicted based on exponential fitting, and ΔU(t) is the function of the beat voltage changing with time; Using ΔU 2 (t) derivative to calculate the minimum point of the beat voltage [ΔU 2 (t s )]' == 0, the moment of the minimum beat voltage is obtained as t s ; According to the closing fixed time t c , determine that the moment to issue the closing command is t s -t c .

6. A power supply fast switching device based on the characteristics of minimum beat voltage, characterized in that, including a sampling module, a measurement module, a prediction module and a determination module; the sampling module is responsible for responding to the startup of rapid power supply switching and acquiring voltage signals of each sampling point on the bus; the measurement module is responsible for performing adaptive adjustment of equal-angle sampling time according to the voltage signals and calculating the voltage frequency, voltage phase and voltage amplitude according to the discrete Fourier voltage phasor of the voltage signals within the adjusted equal-angle sampling time; the prediction module is responsible for obtaining a predicted voltage phase value and a predicted voltage amplitude value according to the voltage amplitude, the voltage phase and the voltage frequency; the determination module is responsible for obtaining a predicted voltage phase difference according to the predicted voltage phase value and the predicted voltage amplitude value, predicting the moment when the beat voltage is minimum according to the predicted voltage phase difference, and determining the moment to issue a closing command so as to achieve rapid power supply switching.

7. The power supply fast switching device based on the minimum beat voltage characteristic according to claim 6, characterized in that, The step of performing adaptive adjustment of equal-angle sampling time according to the voltage signals and calculating the voltage frequency, voltage phase and voltage amplitude according to the discrete Fourier voltage phasor of the voltage signals within the adjusted equal-angle sampling time further includes: Taking the voltage signal as the input, setting the number of sampling points for each sampling sliding window, calculating the phasor of the initial bus voltage signal at each sampling point, and based on the improved adaptive Fourier frequency measurement: considering the voltage frequency attenuation to obtain the angle of the voltage phase interval between adjacent bus sampling points, and obtaining the predicted value of the voltage frequency at the next sampling point of the first sampling sliding window according to the historical voltage frequency before the power failure; Based on the predicted value of the voltage frequency at the next sampling point, adjusting the equal-angle sampling time series, determining the next equal-angle sampling time interval, and resampling the voltage signal at this sampling point by linear interpolation according to the time interval to achieve the next sampling at equal angles; Repeat the above steps until the predicted values of the bus voltage frequencies at all sampling points are obtained based on the exponential fitting method within all sampling sliding windows.

8. A power supply fast switching device based on the minimum beat voltage characteristic according to claim 7, characterized in that, The predicted voltage frequency value of the next sampling point corresponds to the voltage frequency of the sampling point at the k-1 moment, where t k-1 is the time corresponding to the previous sampling point, N is the number of sampling points, and f k-1 is the voltage frequency corresponding to the previous sampling point.

9. The power supply rapid switching device based on the minimum beat voltage characteristic according to claim 7, wherein The voltage frequency, voltage phase, and voltage amplitude are as follows: The voltage amplitude is as shown in the following formula: Among them, F(r + 1) is the discrete Fourier transform phasor of the voltage signal of the r-th sliding window, F(r) is the discrete Fourier transform phasor of the voltage signal of the r-th sliding window, f k is the actual frequency of the current sampling point, Δt is the time interval between adjacent sampling points, v* i+k-N is the voltage signal of the sampling point within the equal-angle sampling window, N is the number of sampling points, and j is the imaginary number; The voltage frequency f k is as follows: where f 0k is the predicted frequency value of the next sampling point The voltage phase value is as shown in the following formula; where F(r + 1) is the discrete Fourier transform phasor of the voltage signal in the (r + 1)-th sliding window, and angle() is the phase angle calculation function.

10. A power supply rapid switching device based on the minimum beat voltage characteristic according to claim 6, characterized in that, It further includes calculating the predicted beat voltage ΔU(t), and the formula is as follows: Among them, U s is the amplitude of the power supply voltage, U m (t) is the amplitude of the residual voltage U L , Δθ(t) is the phase difference predicted based on exponential fitting, and ΔU(t) is the function of the beat voltage changing with time; Using ΔU 2 (t) derivative to calculate the minimum point of the beat voltage [ΔU 2 (t s )]' == 0, and the time when the minimum beat voltage is obtained is t s ; According to the closing fixed time t c , determine that the moment to issue the closing command is t s -t c .

Images