Dual-power-supply rapid switching control method based on wavelet algorithm

Abstract

The invention relates to the technical field of power supply switching, in particular to a dual-power-supply rapid switching control method based on a wavelet algorithm, and aims to solve the problems that in the prior art, whether power supply switching is carried out or not is determined by detecting power supply abnormity through a controller and combining magnetic flux calculation, feature quantities reflecting the power supply state are not analyzed in detail, and power supply switching is not carried out. Whether the power supply is abnormal or not cannot be accurately judged, and whether a switching signal is generated or not cannot be determined by combining a matching judgment result when the power supply is abnormal; according to the invention, the wavelet transformation module and the characteristic quantity extraction module carry out wavelet transformation processing on the preprocessed signal, extract the characteristic quantity reflecting the state of the power supply, and process the extracted characteristic quantity through the abnormity judgment module, so that whether the power supply is abnormal or not can be accurately judged. The matching judgment module analyzes the operation parameters to judge whether the standby power supply is matched with the load or not, and whether a switching signal is generated or not is determined by integrating an abnormal judgment result and a matching judgment result.

Description

Technical Field

[0001] The present invention relates to the technical field of power switching, and more specifically to a dual power supply fast switching control method based on wavelet algorithm. Background Art

[0002] Dual power supply fast switching control technology is a technology that ensures continuous power supply to the load by quickly switching power supplies when power fails or the power grid is unstable. Its main purpose is to quickly switch to the backup power supply when the main power fails or fluctuates to avoid power interruption to the load equipment. This technology is widely used in uninterruptible power supply systems, data centers, medical equipment, communication equipment and other fields with high requirements for power supply reliability.

[0003] The patent application with reference publication number CN117595479A discloses a dual power switching system and a control method thereof, wherein the dual power switching system includes a first static transfer switch, a second static transfer switch, an inductive device and a controller, wherein the first static transfer switch is electrically coupled to a main power source, and the second static transfer switch is electrically coupled to a backup power source, and when the controller detects that an abnormality occurs in the main power source, the controller calculates a magnetic flux difference between a predicted magnetic flux of the inductive device and a current residual magnetic flux, and when the controller determines that the magnetic flux difference is less than or equal to a magnetic flux deviation value, the controller determines whether the output power meets a forced commutation condition, and when the controller determines that the forced commutation condition is met, the controller turns on the second static transfer switch and forcibly turns off the first static transfer switch;

[0004] However, the existing technology detects power supply abnormalities through a controller and determines whether to switch the power supply based on the calculation of magnetic flux. It does not perform a detailed analysis of the characteristic quantities reflecting the power supply status and cannot accurately determine whether the power supply is abnormal. When determining that the power supply is abnormal, it cannot determine whether to generate a switching signal based on the matching evaluation results. Direct switching can easily lead to a mismatch between the backup power supply and the load, thereby causing power supply abnormalities and equipment damage. At the same time, it is impossible to perform a detailed analysis of the dynamic response parameters during the switching process, and it is impossible to accurately determine whether the power supply is stable after the switch, and it is impossible to ensure the safe and stable operation of the power supply after the switch.

[0005] Therefore, we propose a dual power supply fast switching control method based on wavelet algorithm to address the above problems. Summary of the invention

[0006] The purpose of the present invention is to provide a dual power supply fast switching control method based on wavelet algorithm, which solves the problem that the prior art detects power supply abnormality through a controller and determines whether to switch the power supply in combination with magnetic flux calculation, does not perform detailed analysis on the characteristic quantity reflecting the power supply state, cannot accurately determine whether the power supply is abnormal, and cannot determine whether to generate a switching signal in combination with the matching evaluation result when determining the power supply abnormality. Direct switching can easily lead to mismatch between the backup power supply and the load, thereby causing power supply abnormality and equipment damage. At the same time, it is impossible to perform detailed analysis on the dynamic response parameters in the switching process, cannot accurately determine whether the power supply operation is stable after switching, and cannot ensure the safe and stable operation of the power supply after switching.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A dual power supply fast switching control method based on wavelet algorithm comprises the following steps:

[0009] Step 1: Collect the electrical operation signals of the main power supply and the backup power supply in real time, convert the collected electrical operation signals into digital signals, and pre-process the digital signals;

[0010] Step 2: Perform wavelet transform on the preprocessed signal to decompose the signal into different frequency scales, and analyze the signal at each scale to extract the characteristic quantity reflecting the power supply status;

[0011] Step 3: Based on the extracted feature quantity and the preset abnormality criterion, determine whether the main power supply is abnormal. If the main power supply is abnormal, start the fast switching process and prepare to switch to the backup power supply;

[0012] Step 4: Before starting the switch, monitor the operating parameters of the backup power supply and the load in real time, analyze the operating parameters, and judge whether the backup power supply and the load match;

[0013] Step 5: Generate a switching signal based on abnormal judgment and matching judgment, and send the switching signal to the fast switching switch. After the switch receives the signal, it quickly cuts off the abnormal power supply and connects the backup power supply;

[0014] Step 6: After the switching is completed, continue to monitor the electrical operation signals of the newly connected power supply and analyze the dynamic response parameters during the switching process.

[0015] As a preferred embodiment of the present invention, the specific process of preprocessing the digital signal is as follows:

[0016] Noise removal: Remove unnecessary frequency components by using a low-pass filter;

[0017] Normalization: Calculate the minimum and maximum values ​​of the signal and apply the normalization formula: x is the numerical sequence of the signal, so that the normalized signal value is in the range of [0,1];

[0018] Detrending and smoothing: Identify and subtract trends through polynomial fitting and use Savitzky-Golay filter to smooth the signal;

[0019] Time windowing: Set the window size and overlap ratio, traverse the signal, and cut it into multiple overlapping windows. The starting position of each window is set according to the overlap ratio.

[0020] As a preferred embodiment of the present invention, the specific process of performing wavelet transform on the preprocessed signal is as follows:

[0021] The preprocessed signal x(t) is subjected to discrete wavelet transform. The Daubechies wavelet is selected as the wavelet basis function. The number of decomposition layers J is determined according to the frequency range of the signal and the analysis requirements. The signal is decomposed into different frequency scales. The decomposition of each level is expressed as:

[0022]

[0023] Among them, A J (t) is the approximate coefficient of the Jth layer, D j (t) is the detail coefficient of the jth layer;

[0024] The calculation formula of the approximate coefficient is:

[0025]

[0026] The calculation formula of detail coefficient is:

[0027]

[0028] Among them, h[n] and g[n] are the low-pass filter coefficients and high-pass filter coefficients of the wavelet basis, respectively.

[0029] As a preferred embodiment of the present invention, the Daubechies wavelet is defined by two basic functions: a mother wavelet function ψ(t) and a scale function φ(t). The mother wavelet function ψ(t) and the scale function φ(t) of the Daubechies wavelet are generated by a set of filter coefficients h[n]. These filter coefficients satisfy the orthogonal condition and are constructed by a recursive relationship. The recursive relationship of the filter coefficients is:

[0030] Definition of scaling function φ(t):

[0031]

[0032] Definition of mother wavelet function ψ(t):

[0033]

[0034] Among them, h[n] is the filter coefficient of the scaling function, g[n] is the filter coefficient of the mother wavelet function, and they satisfy the following relationship:

[0035] g[n]=(-1) n h[2N-1-n].

[0036] As a preferred embodiment of the present invention, extracting the power state feature quantity specifically includes the following steps:

[0037] Energy feature: Calculate the energy of detail coefficients at each scale:

[0038] Energy characteristics can reflect the energy distribution of the signal at different frequencies;

[0039] Statistical characteristics: mean

[0040] variance

[0041] Skewness

[0042] Kurtosis

[0043] Spectral features: For each layer of detail coefficient D j (t), perform discrete Fourier transform to obtain its spectrum S j (f):

[0044]

[0045] Where DFT is discrete Fourier transform, f is the frequency, N is the length of the signal, and r is the imaginary unit;

[0046] Combine the features at each scale into a feature vector F:

[0047] F=[E1,E2,…,E j , μ1, μ2, …, μ j ,σ1,σ2,…,σ j ,γ1,γ2,…,

[0048] γ j ,κ1,κ2,…,κ j ].

[0049] As a preferred embodiment of the present invention, the specific process of judging whether the main power supply is abnormal based on the extracted feature quantity and the preset abnormality criterion is as follows:

[0050] Set the normal range of each feature quantity:

[0051] E j min ≤E j ≤E j max ;

[0052] μ j min ≤μ j ≤μ j max ;

[0053] σ j min ≤σ j ≤σ j max ;

[0054] γ j min ≤γ j ≤γ j max ;

[0055] κ j min ≤κ j ≤κ j max ;

[0056] The preset abnormality criteria are: at least one characteristic quantity exceeds the normal range;

[0057] Compare the extracted feature vector F with the preset anomaly criterion:

[0058] If all the feature quantities in the extracted feature vector F are within the normal range, it is determined that the main power supply is normal;

[0059] If at least one feature quantity in the extracted feature vector F exceeds the normal range, it is determined that the main power supply is abnormal.

[0060] As a preferred embodiment of the present invention, the specific process of analyzing the operating parameters and judging whether the backup power supply matches the load is as follows:

[0061] Collect the operating parameters of the backup power supply and the load in real time, the operating parameters include voltage, frequency and phase, calculate the difference between the real-time backup power supply voltage and the real-time load voltage, take the absolute value to obtain the real-time voltage difference, mark the real-time voltage difference as SDC, calculate the difference between the real-time backup power supply frequency and the real-time load frequency, take the absolute value to obtain the real-time frequency difference, mark the real-time frequency difference as SPC, calculate the difference between the real-time backup power supply phase and the real-time load phase, take the absolute value to obtain the real-time phase difference, mark the real-time phase difference as SXC;

[0062] By formula Obtain a matching evaluation value, wherein b1, b2, and b3 are preset proportional coefficients, and the values ​​of b1, b2, and b3 are all positive numbers, and compare and analyze the matching evaluation value PD with the preset matching evaluation value threshold value entered and stored internally;

[0063] If the matching evaluation value PD is less than or equal to the preset matching evaluation value threshold, it is determined that the backup power supply and the load are matched;

[0064] If the matching evaluation value PD is greater than the preset matching evaluation value threshold, it is determined that there is a mismatch between the backup power supply and the load, and the output of the backup power supply is immediately adjusted to be consistent with the load.

[0065] As a preferred embodiment of the present invention, the specific process of generating a switching signal according to abnormality judgment and matching judgment is as follows:

[0066] When it is determined that the main power supply is normal, no switching signal is generated;

[0067] When it is determined that the main power supply is abnormal, if the matching evaluation value PD is greater than the preset matching evaluation value threshold, no switching signal is generated;

[0068] When it is determined that the main power supply is abnormal, if the matching evaluation value PD is less than or equal to the preset matching evaluation value threshold, a switching signal is generated.

[0069] As a preferred implementation of the present invention, the specific process of analyzing the dynamic response parameters during the switching process is as follows:

[0070] A monitoring period is set, and the monitoring period is divided into i sub-time periods, where i is a natural number greater than zero. The dynamic response parameters include transient voltage deviation, transient current deviation, frequency fluctuation value, and power factor fluctuation value. The transient voltage deviation in each sub-time period is collected to construct a set A of transient voltage deviations, obtain the mean in set A, and mark the mean in set A as the transient voltage deviation mean SYP. The transient current deviation in each sub-time period is collected to construct a set B of transient current deviations, obtain the mean in set B, and mark the mean in set B as the transient current deviation mean SLP. The frequency fluctuation value in each sub-time period is collected to construct a set C of frequency fluctuation values, obtain the mean in set C, and mark the mean in set C as the frequency fluctuation mean PBJ. The power factor fluctuation value in each sub-time period is collected to construct a set D of power factor fluctuation values, obtain the mean in set D, and mark the mean in set D as the power factor fluctuation mean GBJ.

[0071] By formula Obtain a power switching assessment value, wherein c1, c2, c3, and c4 are preset proportional coefficients, and the values ​​of c1, c2, c3, and c4 are all positive numbers, and compare and analyze the power switching assessment value DQP with the preset power switching assessment value threshold value recorded and stored internally;

[0072] If the power switching assessment value DQP is greater than the preset power switching assessment value threshold, it indicates that the power supply after switching is unstable, and a display text indicating that the power supply is unstable and needs maintenance is generated and sent to the communication terminal of the administrator;

[0073] If the power switching rating value DQP is less than or equal to the preset power switching rating value threshold, it indicates that the switched power supply operates stably and no display text is generated.

[0074] As a preferred embodiment of the present invention, it is applied to a dual power supply fast switching control system based on wavelet algorithm, including a management platform, a data acquisition module, a data preprocessing module, a wavelet transformation module, a feature extraction module, an abnormality judgment module, a matching judgment module and a dynamic response analysis module;

[0075] The data acquisition module acquires the electrical operation signal and converts the acquired electrical operation signal into a digital signal;

[0076] The data preprocessing module performs preprocessing operations on the digital signal;

[0077] The wavelet transform module performs wavelet transform processing on the pre-processed signal;

[0078] The feature quantity extraction module is used to extract the feature quantity reflecting the power supply state;

[0079] The abnormality judgment module judges whether the main power supply is abnormal based on the extracted characteristic quantity and the preset abnormality judgment criterion;

[0080] The matching evaluation module analyzes the operating parameters to determine whether the backup power supply matches the load;

[0081] The dynamic response analysis module analyzes the dynamic response parameters during the switching process.

[0082] Compared with the prior art, the advantages of the present invention are:

[0083] (1) In the present invention, the pre-processed signal is subjected to wavelet transform processing by a wavelet transform module and a feature extraction module, and a feature reflecting the power supply state is extracted. The extracted feature is processed by an abnormality judgment module to accurately judge whether the power supply is abnormal. The matching evaluation module analyzes the operating parameters to judge whether the backup power supply matches the load. The abnormality judgment result and the matching evaluation result are combined to determine whether a switching signal is generated, thereby avoiding power supply abnormality and equipment damage caused by mismatch between the backup power supply and the load.

[0084] (2) In the present invention, the dynamic response parameters are analyzed in detail through the dynamic response analysis module, so as to accurately determine whether the power supply operation is stable after switching, and generate a text display indicating that the power supply operation is unstable and needs maintenance to remind the management personnel after determining that the power supply operation is unstable, so that the management personnel can take targeted treatment measures in time, thereby ensuring the safe and stable operation of the power supply after switching. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 It is a block diagram of the dual power supply fast switching control method of the present invention;

[0086] Figure 2 is a system block diagram of the present invention;

[0087] Figure 3 It is a logical flow diagram of the present invention. DETAILED DESCRIPTION

[0088] The following will combine the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all the embodiments. All other embodiments obtained by ordinary technicians in this field without creative work based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0089] Embodiment 1: Figure 1 As shown, the present invention proposes a dual power supply fast switching control method based on wavelet algorithm, comprising the following steps:

[0090] Step 1: Collect the electrical operation signals of the main power supply and the backup power supply in real time. The electrical operation signals include voltage signals, current signals, power signals and frequency signals. Convert the collected electrical operation signals into digital signals and pre-process the digital signals.

[0091] The specific process of preprocessing digital signals is as follows:

[0092] Noise removal: Remove unnecessary frequency components by using a low-pass filter (such as Butterworth or Chebyshev);

[0093] Normalization: Calculate the minimum and maximum values ​​of the signal and apply the normalization formula: x is the numerical sequence of the signal, so that the normalized signal value is in the range of [0,1];

[0094] Detrending and smoothing: Identify and subtract trends through polynomial fitting (usually first- or second-order polynomials), and use Savitzky-Golay filters to smooth the signal;

[0095] Time windowing: set the window size (e.g. 512 samples) and overlap ratio (e.g. 50%), traverse the signal, cut it into multiple overlapping windows, and set the starting position of each window according to the overlap ratio;

[0096] The purpose of preprocessing digital signals is to improve signal quality, eliminate noise, optimize transformation effects, and ensure the effectiveness and accuracy of subsequent analysis. Through these processes, relevant information in the signal can be better extracted and the reliability of the overall analysis can be improved.

[0097] Step 2: Perform wavelet transform on the preprocessed signal to decompose the signal into different frequency scales, and analyze the signal at each scale to extract the characteristic quantity reflecting the power supply status;

[0098] The specific process of wavelet transform on the preprocessed signal is as follows:

[0099] The preprocessed signal x(t) is subjected to discrete wavelet transform, and the wavelet basis function is Daubechies wavelet. Daubechies wavelet is defined by two basic functions: mother wavelet function ψ(t) and scale function φ(t). The mother wavelet function ψ(t) and scale function φ(t) of Daubechies wavelet are generated by a set of filter coefficients h[n]. These filter coefficients satisfy the orthogonal condition and are constructed through a recursive relationship. The recursive relationship of the filter coefficients is:

[0100] Definition of scaling function φ(t):

[0101]

[0102] Definition of mother wavelet function ψ(t):

[0103]

[0104] Among them, h[n] is the filter coefficient of the scaling function, g[n] is the filter coefficient of the mother wavelet function, and they satisfy the following relationship:

[0105] g[n]=(-1) n h[2N-1-n];

[0106] Determine the number of decomposition layers J according to the frequency range of the signal and the analysis requirements, and decompose the signal into different frequency scales. The decomposition of each level is expressed as:

[0107]

[0108] Among them, A J (t) is the approximate coefficient of the Jth layer (low frequency part), D j (t) is the detail coefficient (high frequency part) of the jth layer;

[0109] The calculation formula of the approximate coefficient is:

[0110]

[0111] The calculation formula of detail coefficient is:

[0112]

[0113] Among them, h[n] and g[n] are the low-pass filter coefficient and high-pass filter coefficient of the wavelet basis respectively;

[0114] Extracting the power state feature quantity specifically includes the following steps:

[0115] Energy feature: Calculate the energy of detail coefficients at each scale:

[0116] Energy characteristics can reflect the energy distribution of the signal at different frequencies;

[0117] Statistical characteristics: mean

[0118] variance

[0119] Skewness

[0120] Kurtosis

[0121] Spectral features: For each layer of detail coefficient D j (t), perform discrete Fourier transform to obtain its spectrum S j (f):

[0122]

[0123] Where DFT is discrete Fourier transform, f is the frequency, N is the length of the signal, and r is the imaginary unit;

[0124] Combine the features at each scale into a feature vector F:

[0125] F=[E1,E2,…,E j, μ1, μ2, …, μ j ,σ1,σ2,…,σ j ,γ1,γ2,…,

[0126] γ j ,κ1,κ2,…,κ j ];

[0127] Step 3: Based on the extracted feature quantity and the preset abnormality criterion, determine whether the main power supply is abnormal. If the main power supply is abnormal, start the fast switching process and prepare to switch to the backup power supply;

[0128] Based on the extracted feature quantity and the preset abnormality criterion, the specific process of judging whether the main power supply is abnormal is as follows:

[0129] Set the normal range of each feature quantity:

[0130] E j min ≤E j ≤E j max ;

[0131] μ j min ≤μ j ≤μ j max ;

[0132] σ j min ≤σ j ≤σ j max ;

[0133] γ j min ≤γ j ≤γ j max ;

[0134] κ j min ≤κ j ≤κ j max ;

[0135] The preset abnormality criteria are: at least one characteristic quantity exceeds the normal range;

[0136] Compare the extracted feature vector F with the preset anomaly criterion:

[0137] If all the feature quantities in the extracted feature vector F are within the normal range, it is determined that the main power supply is normal;

[0138] If at least one characteristic quantity in the extracted characteristic vector F exceeds the normal range, the main power supply is judged to be abnormal;

[0139] Step 4: Before starting the switch, monitor the operating parameters of the backup power supply and the load in real time, analyze the operating parameters, and judge whether the backup power supply and the load match;

[0140] The specific process of analyzing the operating parameters and judging whether the backup power supply matches the load is as follows:

[0141] Collect the operating parameters of the backup power supply and the load in real time, the operating parameters include voltage, frequency and phase, calculate the difference between the real-time backup power supply voltage and the real-time load voltage, take the absolute value to obtain the real-time voltage difference, mark the real-time voltage difference as SDC, calculate the difference between the real-time backup power supply frequency and the real-time load frequency, take the absolute value to obtain the real-time frequency difference, mark the real-time frequency difference as SPC, calculate the difference between the real-time backup power supply phase and the real-time load phase, take the absolute value to obtain the real-time phase difference, mark the real-time phase difference as SXC;

[0142] By formula The matching evaluation value is obtained. The larger the value of the matching evaluation value is, the greater the difference in voltage, frequency and phase between the backup power supply and the load is, and the more mismatched the backup power supply and the load are. b1, b2 and b3 are preset proportional coefficients, and the values ​​of b1, b2 and b3 are all positive numbers. The matching evaluation value PD is compared and analyzed with the preset matching evaluation value threshold value stored in the internal input;

[0143] If the matching evaluation value PD is less than or equal to the preset matching evaluation value threshold, it is determined that the backup power supply and the load are matched;

[0144] If the matching evaluation value PD is greater than the preset matching evaluation value threshold, it is determined that the backup power supply and the load do not match each other, and the output of the backup power supply is immediately adjusted to keep it consistent with the load;

[0145] Step 5: Generate a switching signal based on abnormal judgment and matching judgment, and send the switching signal to the fast switching switch. After the switch receives the signal, it quickly cuts off the abnormal power supply and connects the backup power supply;

[0146] The specific process of generating a switching signal based on abnormal judgment and matching judgment is as follows:

[0147] When it is determined that the main power supply is normal, no switching signal is generated;

[0148] When it is determined that the main power supply is abnormal, if the matching evaluation value PD is greater than the preset matching evaluation value threshold, no switching signal is generated;

[0149] When it is determined that the main power supply is abnormal, if the matching evaluation value PD is less than or equal to the preset matching evaluation value threshold, a switching signal is generated;

[0150] Step 6: After the switching is completed, continue to monitor the electrical operation signals of the newly connected power supply and analyze the dynamic response parameters during the switching process.

[0151] Embodiment 2: The technical solution of the embodiment of the present invention is different from that of embodiment 1 in that:

[0152] like Figure 2 As shown, the specific process of analyzing the dynamic response parameters during the switching process is as follows:

[0153] A monitoring period is set, and the monitoring period is divided into i sub-time periods, where i is a natural number greater than zero. The dynamic response parameters include transient voltage deviation, transient current deviation, frequency fluctuation value, and power factor fluctuation value. The transient voltage deviation in each sub-time period is collected to construct a set A of transient voltage deviations, obtain the mean in set A, and mark the mean in set A as the transient voltage deviation mean SYP. The transient current deviation in each sub-time period is collected to construct a set B of transient current deviations, obtain the mean in set B, and mark the mean in set B as the transient current deviation mean SLP. The frequency fluctuation value in each sub-time period is collected to construct a set C of frequency fluctuation values, obtain the mean in set C, and mark the mean in set C as the frequency fluctuation mean PBJ. The power factor fluctuation value in each sub-time period is collected to construct a set D of power factor fluctuation values, obtain the mean in set D, and mark the mean in set D as the power factor fluctuation mean GBJ.

[0154] By formula The power switching assessment value is obtained. The larger the value of the power switching assessment value is, the more unstable the power supply operation is after switching, and the more unreliable the power supply is. Among them, c1, c2, c3, and c4 are preset proportional coefficients, and the values ​​of c1, c2, c3, and c4 are all positive numbers. The power switching assessment value DQP is compared and analyzed with the preset power switching assessment value threshold value recorded and stored internally;

[0155] If the power switching assessment value DQP is greater than the preset power switching assessment value threshold, it indicates that the power supply after switching is unstable, and a display text indicating that the power supply is unstable and needs maintenance is generated and sent to the communication terminal of the administrator;

[0156] If the power switching evaluation value DQP is less than or equal to the preset power switching evaluation value threshold, it indicates that the power supply after switching is running stably, and no display text is generated;

[0157] The invention is applied to a dual power supply fast switching control system based on wavelet algorithm, comprising a management platform, a data acquisition module, a data preprocessing module, a wavelet transform module, a feature extraction module, an abnormality judgment module, a matching evaluation module and a dynamic response analysis module; the management platform is connected to the data acquisition module, the matching evaluation module and the dynamic response analysis module in one-way communication, the matching evaluation module is also connected to the abnormality judgment module in one-way communication, the data acquisition module is connected to the data preprocessing module in one-way communication, the data preprocessing module is connected to the wavelet transform module in one-way communication, the wavelet transform module is connected to the feature extraction module in one-way communication, and the feature extraction module is connected to the abnormality judgment module in one-way communication;

[0158] The data acquisition module collects electrical operation signals and converts the collected electrical operation signals into digital signals. The data acquisition module collects electrical operation signals, and the data collection is accurate and comprehensive, which facilitates the subsequent analysis of the power supply status.

[0159] Data preprocessing module performs preprocessing operations on digital signals; preprocessing digital signals through the data preprocessing module is convenient for improving signal quality, eliminating noise, optimizing transformation effects, and ensuring the effectiveness and accuracy of subsequent analysis;

[0160] The wavelet transform module performs wavelet transform processing on the pre-processed signal; the wavelet transform module decomposes the signal into multiple scales to facilitate the subsequent extraction of feature quantities reflecting the power supply status;

[0161] A feature extraction module, used to extract feature quantities reflecting power supply status;

[0162] The abnormality judgment module judges whether the main power supply is abnormal based on the extracted feature quantity and the preset abnormality judgment criterion; the abnormality judgment module processes and analyzes the extracted feature quantity in combination with the abnormality judgment criterion to judge whether the power supply is abnormal;

[0163] The matching evaluation module analyzes the operating parameters to determine whether the backup power supply and the load match; the matching evaluation module determines whether the backup power supply and the load match by analyzing the operating parameters;

[0164] The dynamic response analysis module analyzes the dynamic response parameters during the switching process; the dynamic response analysis module performs a detailed analysis of the dynamic response parameters to determine whether the power supply is stable after the switching, and generates a text display indicating that the power supply is unstable and needs maintenance after determining that the power supply is unstable, so that the management personnel can take targeted measures in time to ensure that the power supply can operate safely and stably after the switching.

[0165] Working principle of the present invention: When in use, in the present invention, the pre-processed signal is subjected to wavelet transform processing by a wavelet transform module and a feature extraction module, and feature quantities reflecting the power supply state are extracted. The extracted feature quantities are processed by the abnormality judgment module to accurately judge whether the power supply is abnormal. The operation parameters are analyzed by the matching evaluation module to judge whether the backup power supply matches the load. The abnormality judgment result and the matching evaluation result are combined to decide whether to generate a switching signal, so as to avoid power supply abnormality and equipment damage caused by mismatch between the backup power supply and the load. The dynamic response analysis module performs a detailed analysis of the dynamic response parameters, so as to accurately judge whether the power supply after switching is stable, and after judging that the power supply operation is unstable, a display text is generated to remind the management personnel that the power supply operation is unstable and needs maintenance, so as to facilitate the management personnel to take targeted treatment measures in time, thereby ensuring the safe and stable operation of the power supply after switching.

[0166] The above are only preferred specific implementation modes of the present invention; however, the protection scope of the present invention is not limited thereto; any technician familiar with the technical field within the technical scope disclosed by the present invention; any equivalent replacement or change based on the technical solution and improved concept of the present invention shall be covered within the protection scope of the present invention.

Claims

1. A dual power supply fast switching control method based on wavelet algorithm, characterized in that: The following steps are involved: Step 1: Collect the electrical operation signals of the main power supply and the backup power supply in real time, convert the collected electrical operation signals into digital signals, and pre-process the digital signals; Step 2: Perform wavelet transform on the preprocessed signal to decompose the signal into different frequency scales, and analyze the signal at each scale to extract the characteristic quantity reflecting the power supply status; Step 3: Based on the extracted feature quantity and the preset abnormality criterion, determine whether the main power supply is abnormal. If the main power supply is abnormal, start the fast switching process and prepare to switch to the backup power supply; Step 4: Before starting the switch, monitor the operating parameters of the backup power supply and the load in real time, analyze the operating parameters, and judge whether the backup power supply and the load match; Step 5: Generate a switching signal based on abnormal judgment and matching judgment, and send the switching signal to the fast switching switch. After the switch receives the signal, it quickly cuts off the abnormal power supply and connects the backup power supply; Step 6: After the switching is completed, continue to monitor the electrical operation signals of the newly connected power supply and analyze the dynamic response parameters during the switching process.

2. The dual power supply fast switching control method based on wavelet algorithm according to claim 1 is characterized in that: The specific process of preprocessing digital signals is as follows: Noise Removal: Remove unnecessary frequency components by using a low-pass filter; Normalization: Calculate the minimum and maximum values ​​of the signal and apply the normalization formula: x is the numerical sequence of the signal, so that the normalized signal value is in the range of [0,1]; Detrending and smoothing: Identify and subtract trends through polynomial fitting and use Savitzky-Golay filter to smooth the signal; Time windowing: Set the window size and overlap ratio, traverse the signal, and cut it into multiple overlapping windows. The starting position of each window is set according to the overlap ratio.

3. The dual power supply fast switching control method based on wavelet algorithm according to claim 1 is characterized in that: The specific process of wavelet transform on the preprocessed signal is as follows: The preprocessed signal x(t) is subjected to discrete wavelet transform. The Daubechies wavelet is selected as the wavelet basis function. The number of decomposition layers J is determined according to the frequency range of the signal and the analysis requirements. The signal is decomposed into different frequency scales. The decomposition of each level is expressed as: Among them, A J (t) is the approximate coefficient of the Jth layer, D j (t) is the detail coefficient of the jth layer; The calculation formula of the approximate coefficient is: The calculation formula of detail coefficient is: Among them, h[n] and g[n] are the low-pass filter coefficients and high-pass filter coefficients of the wavelet basis, respectively.

4. The dual power supply fast switching control method based on wavelet algorithm according to claim 3 is characterized in that: The Daubechies wavelet is defined by two basic functions: the mother wavelet function ψ(t) and the scaling function φ(t). The mother wavelet function ψ(t) and the scaling function φ(t) of the Daubechies wavelet are generated by a set of filter coefficients h[n]. These filter coefficients satisfy the orthogonal condition and are constructed through a recursive relationship. The recursive relationship of the filter coefficients is: Definition of scaling function φ(t): Definition of mother wavelet function ψ(t): Among them, h[n] is the filter coefficient of the scaling function, g[n] is the filter coefficient of the mother wavelet function, and they satisfy the following relationship: g[n]=(-1) n h[2N-1-n]。 5. The dual power supply fast switching control method based on wavelet algorithm according to claim 4 is characterized in that: Extracting the power state feature quantity specifically includes the following steps: Energy feature: Calculate the energy of detail coefficients at each scale: Energy characteristics can reflect the energy distribution of the signal at different frequencies; Statistical characteristics: mean μ j : variance Skewness γ j : Kurtosis κ j : Spectral features: For each layer of detail coefficient D j (t), perform discrete Fourier transform to obtain its spectrum S j (f): Where DFT is discrete Fourier transform, f is the frequency, N is the length of the signal, and r is the imaginary unit; Combine the features at each scale into a feature vector F: F=[E1,E2,…,E j ,μ1,μ2,…,μ j ,σ1,σ2,…,σ j ,γ1,γ2,…, c j ,k1,k2,…,k j ]。 6. The dual power supply fast switching control method based on wavelet algorithm according to claim 5 is characterized in that: Based on the extracted feature quantity and the preset abnormality criterion, the specific process of judging whether the main power supply is abnormal is as follows: Set the normal range of each feature quantity: AND j min ≤E j ≤E j max ; m j min ≤μ j ≤μ j max ; s j min ≤σ j ≤σ j max ; c j min ≤γ j ≤γ j max ; k j min ≤κ j ≤κ j max ; The preset abnormality criteria are: at least one characteristic value exceeds the normal range; Compare the extracted feature vector F with the preset anomaly criterion: If all the feature quantities in the extracted feature vector F are within the normal range, it is determined that the main power supply is normal; If at least one feature quantity in the extracted feature vector F exceeds the normal range, it is determined that the main power supply is abnormal.

7. The dual power supply fast switching control method based on wavelet algorithm according to claim 6 is characterized in that: The specific process of analyzing the operating parameters and judging whether the backup power supply matches the load is as follows: Collect the operating parameters of the backup power supply and the load in real time, the operating parameters include voltage, frequency and phase, calculate the difference between the real-time backup power supply voltage and the real-time load voltage, take the absolute value to obtain the real-time voltage difference, mark the real-time voltage difference as SDC, calculate the difference between the real-time backup power supply frequency and the real-time load frequency, take the absolute value to obtain the real-time frequency difference, mark the real-time frequency difference as SPC, calculate the difference between the real-time backup power supply phase and the real-time load phase, take the absolute value to obtain the real-time phase difference, mark the real-time phase difference as SXC; By formula Obtain a matching evaluation value, wherein b1, b2, and b3 are preset proportional coefficients, and the values ​​of b1, b2, and b3 are all positive numbers, and compare and analyze the matching evaluation value PD with the preset matching evaluation value threshold value entered and stored internally; If the matching evaluation value PD is less than or equal to the preset matching evaluation value threshold, it is determined that the backup power supply and the load are matched; If the matching evaluation value PD is greater than the preset matching evaluation value threshold, it is determined that there is a mismatch between the backup power supply and the load, and the output of the backup power supply is immediately adjusted to be consistent with the load.

8. The dual power supply fast switching control method based on wavelet algorithm according to claim 7 is characterized in that: The specific process of generating a switching signal based on abnormal judgment and matching judgment is as follows: When it is determined that the main power supply is normal, no switching signal is generated; When it is determined that the main power supply is abnormal, if the matching evaluation value PD is greater than the preset matching evaluation value threshold, no switching signal is generated; When the main power supply is judged to be abnormal, if the matching evaluation value PD is less than or equal to the preset matching evaluation value threshold, Generates a switching signal.

9. The dual power supply fast switching control method based on wavelet algorithm according to claim 1 is characterized in that: The specific process of analyzing the dynamic response parameters during the switching process is as follows: A monitoring period is set, and the monitoring period is divided into i sub-time periods, where i is a natural number greater than zero. The dynamic response parameters include transient voltage deviation, transient current deviation, frequency fluctuation value, and power factor fluctuation value. The transient voltage deviation in each sub-time period is collected to construct a set A of transient voltage deviations, obtain the mean in set A, and mark the mean in set A as the transient voltage deviation mean SYP. The transient current deviation in each sub-time period is collected to construct a set B of transient current deviations, obtain the mean in set B, and mark the mean in set B as the transient current deviation mean SLP. The frequency fluctuation value in each sub-time period is collected to construct a set C of frequency fluctuation values, obtain the mean in set C, and mark the mean in set C as the frequency fluctuation mean PBJ. The power factor fluctuation value in each sub-time period is collected to construct a set D of power factor fluctuation values, obtain the mean in set D, and mark the mean in set D as the power factor fluctuation mean GBJ. By formula Obtain a power switching assessment value, wherein c1, c2, c3, and c4 are preset proportional coefficients, and the values ​​of c1, c2, c3, and c4 are all positive numbers, and compare and analyze the power switching assessment value DQP with the preset power switching assessment value threshold value recorded and stored internally; If the power switching assessment value DQP is greater than the preset power switching assessment value threshold, it indicates that the power supply after switching is unstable, and a display text indicating that the power supply is unstable and needs maintenance is generated and sent to the communication terminal of the administrator; If the power switching rating value DQP is less than or equal to the preset power switching rating value threshold, it indicates that the switched power supply operates stably and no display text is generated.

10. A dual power supply fast switching control method based on wavelet algorithm according to any one of claims 1 to 9, characterized in that: Applied to a dual power supply fast switching control system based on wavelet algorithm, including management platform, data acquisition module, data preprocessing module, wavelet transform module, feature extraction module, abnormality judgment module, matching judgment module and dynamic response analysis module; The data acquisition module acquires the electrical operation signal and converts the acquired electrical operation signal into a digital signal; The data preprocessing module performs preprocessing operations on the digital signal; The wavelet transform module performs wavelet transform processing on the pre-processed signal; The feature quantity extraction module is used to extract the feature quantity reflecting the power supply state; The abnormality judgment module judges whether the main power supply is abnormal based on the extracted characteristic quantity and the preset abnormality judgment criterion; The matching evaluation module analyzes the operating parameters to determine whether the backup power supply matches the load; The dynamic response analysis module analyzes the dynamic response parameters during the switching process.

Images