A master-slave complementary mode power supply control circuit and power supply abnormality detection method thereof

Through the master-slave complementary mode power control circuit and neural network abnormality detection method, the problem of insufficient reliability in the event of failure of existing power systems is solved, and the continuity of power supply and system reliability are improved.

CN118783390BActive Publication Date: 2024-11-19BEIJING JIAJIE HENGXIN ENERGY TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411267286.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-11
Publication Date
2024-11-19
Estimated Expiration
2044-09-11

AI Technical Summary

Technical Problem

When existing power systems face sudden failures or abnormal conditions, they lack effective redundant control mechanisms, resulting in insufficient system reliability, especially at critical moments that may cause immeasurable losses.

Method used

The master-slave complementary mode power control circuit is adopted to ensure that the power supply continues to be provided when a fault occurs, and combined with the abnormality protection module and neural network abnormality detection method, real-time power output is achieved. Monitoring and fault identification.

Benefits of technology

Through the design of the master-slave complementary mode power control circuit, the continuity of power supply is ensured, the risk of downtime caused by power failure is reduced, and the reliability of the system is improved, which is particularly important in key application scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118783390B_ABST
    Figure CN118783390B_ABST
Patent Text Reader

Abstract

The present invention relates to a master-slave complementary mode power supply control circuit and a power supply anomaly detection method thereof, belonging to the field of power supply management technology, comprising: a master power supply, comprising a first control system and a first power system, the first control system and the first power system are communicatively connected; the first power system is used to convert alternating current into preset direct current for output according to the instruction of the first control system; a slave power supply, comprising a second control system and a second power system, the second control system and the second power system are communicatively connected; the second power system is used to convert alternating current into preset direct current for output according to the instruction of the second control system; when the master / slave power supply fails, it switches to the slave / master power supply to continue working. In the present invention, when the main power supply fails, the system can quickly switch to the slave power supply to ensure the continuity of power supply. This design greatly reduces the risk of downtime caused by power failure, which is particularly important in key application scenarios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power management, and in particular to a master-slave complementary mode power control circuit and a power anomaly detection method thereof. Background Art

[0002] With the continuous development of society and technology, power supply has become an indispensable part of modern industrial life, providing necessary energy support for various industrial equipment. With the continuous refinement of the industry, special industries (aerospace, medical) have higher and higher requirements for power supply, especially the reliability of power supply output. At critical moments, abnormal output of power supply will cause immeasurable disasters and losses. Although the existing power supply system can meet the basic power supply needs to a certain extent, it often lacks an effective redundant control mechanism in the face of sudden failures or abnormal situations, resulting in insufficient system reliability. Summary of the invention

[0003] To solve the above problems, an object of the embodiments of the present invention is to provide a master-slave complementary mode power supply control circuit and a power supply abnormality detection method thereof.

[0004] A master-slave complementary mode power supply control circuit, comprising:

[0005] A main power supply includes a first control system and a first power system, wherein the first control system and the first power system are in communication connection; the first power system is used to convert the alternating current into a preset direct current for output according to the instruction of the first control system;

[0006] The slave power supply includes a second control system and a second power system, wherein the second control system and the second power system are in communication connection; the second power system is used to convert the alternating current into a preset direct current for output according to the instruction of the second control system;

[0007] When the master / slave power fails, it switches to the slave / master power to continue working.

[0008] Preferably, the first control system and the second control system communicate through the CAN bus. When the control system completes initialization, the chip ID of the first control system and the chip ID of the second control system are compared, and the address of the control system corresponding to the larger chip ID is set to 1, and it is confirmed as the control system of the main power supply. The address of the control system corresponding to the smaller chip ID is set to 0, and it is confirmed as the control system of the slave power supply.

[0009] Preferably, it also includes:

[0010] The abnormal protection module is used to protect the master / slave power supply immediately when an abnormality occurs in the master / slave power supply, shut down the power output, and immediately send abnormal alarm information through the CAN bus. At the same time, after receiving the abnormal alarm information, the slave / master power supply immediately adjusts according to the instructions of the control system to convert AC power into preset DC power for output.

[0011] The present invention also provides a power supply abnormality detection method, comprising:

[0012] Step 1: Obtain current sampling signals and voltage sampling signals at the output ends of the main power supply and the slave power supply;

[0013] Step 2: reconstructing the current sampling signal and the voltage sampling signal to form a reconstructed current sampling signal and a reconstructed voltage sampling signal;

[0014] Step 3: performing image quantization encoding on the reconstructed current sampling signal and the reconstructed voltage sampling signal to form a current encoding image and a voltage encoding image;

[0015] Step 4: splicing the current coded image and the voltage coded image to form a spliced ​​coded image;

[0016] Step 5: Setting a label for the spliced ​​coded image to indicate the fault type of the power supply;

[0017] Step 6: pre-process the spliced ​​coded images to form training samples, and input the training samples into a neural network for training to obtain a power anomaly detection model;

[0018] Step 7: Use the power anomaly detection model to monitor the main power supply and the slave power supply to obtain a power supply monitoring result.

[0019] Preferably, the step 2: reconstructing the current sampling signal and the voltage sampling signal to form a reconstructed current sampling signal and a reconstructed voltage sampling signal comprises:

[0020] Step 2.1: Perform wavelet transform on the current sampling signal and the voltage sampling signal to generate wavelet coefficients;

[0021] Step 2.2: Screening the wavelet coefficients based on a screening threshold to form screened wavelet coefficients; wherein the screening threshold is:

[0022]

[0023] in, represents the screening threshold, represents the preset coefficient, Indicates the length of the signal, represents the wavelet transform scale, represents the standard deviation of Gaussian noise;

[0024] Step 2.3: reconstruct the screened wavelet coefficients to form a reconstructed current sampling signal and a reconstructed voltage sampling signal.

[0025] Preferably, in step 2.2, when the absolute value of the wavelet coefficient is greater than or equal to the screening threshold, the wavelet coefficient is screened using a first screening formula to form screened wavelet coefficients; wherein the first screening formula is:

[0026]

[0027] When the absolute value of the wavelet coefficient is less than the screening threshold, the wavelet coefficient is screened using a second screening formula to form a screened wavelet coefficient;

[0028]

[0029] in, represents the kth wavelet coefficient under the jth transformation scale, represents the shrinkage factor, represents the convergence factor, Represents a symbolic function.

[0030] Preferably, the step 3: performing image quantization encoding on the reconstructed current sampling signal and the reconstructed voltage sampling signal to form a current encoding image and a voltage encoding image comprises:

[0031] Step 3.1: Obtain the value of the sampling point at each moment according to the reconstructed current sampling signal and the reconstructed voltage sampling signal;

[0032] Step 3.2: Fill the encoded pixel points into a box of a preset size from left to right and from top to bottom based on the encoding function to form a current encoding image and a voltage encoding image; wherein the encoding function is:

[0033]

[0034] In the formula, Represents the encoded pixel point, Indicates taking an integer. represents the value of the i-th sampling point, represents the minimum value of the sampled signal, Indicates the maximum value of the sampled signal.

[0035] Preferably, the step 6: preprocessing the spliced ​​coded images to form training samples, and inputting the training samples into a neural network for training to obtain a power anomaly detection model, comprises:

[0036] Step 6.1: Enhance the spliced ​​coded image to form a training sample; wherein the enhancement processing formula is:

[0037]

[0038] in, Indicates that the coded image after splicing is The pixel value at position, represents the first shape parameter, represents the second shape parameter, represents the third shape parameter, represents the standard deviation of the Gaussian distribution;

[0039] Step 6.2: Input the training samples into the neural network for training and use the loss function to verify the training effect of the neural network. When the loss value is less than the preset value, the power anomaly detection model is obtained.

[0040] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and wherein the computer program, when executed by the processor, implements the steps in the above-mentioned power anomaly detection method.

[0041] The present invention also provides a computer-readable storage medium on which a computer program is stored, characterized in that when the computer program is executed by a processor, the steps in the above-mentioned power anomaly detection method are implemented.

[0042] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0043] The present invention relates to a master-slave complementary mode power supply control circuit and a power supply anomaly detection method thereof. Compared with the prior art, when the master power supply fails, the system can quickly switch to the slave power supply to ensure the continuity of power supply. This design greatly reduces the risk of downtime caused by power failure, which is particularly important in critical application scenarios.

[0044] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 A schematic diagram of a master-slave complementary mode power supply control circuit in an embodiment provided by the present invention;

[0047] Figure 2 This is a working flow chart of the master-slave complementary mode power supply in the embodiment provided by the present invention. DETAILED DESCRIPTION

[0048] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0049] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0050] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0051] The power supply generally consists of two parts: power and control. The power part realizes the conversion of energy, converting the AC 380V alternating current into direct current for output; the control part controls the output voltage of the power supply through the control system, outputs power distribution (opening and closing of the output switch), detects parameters such as the voltage and current of the power supply, and realizes the overall monitoring of the power supply.

[0052] Generally speaking, the reliability of the power supply is determined by the overall reliability of each functional component inside the power supply. The reliability formula is P = P1 * P2 * … Pn, where P1, P2, … Pn are the reliabilities of each component. Improving the reliability of any part can improve the overall reliability. Therefore, by redundant design of the control part, the reliability of the power supply can be improved. Assuming the reliability of the control part is P (0 < P < 1), after overall redundancy, its reliability can reach P’ = 1 - (1 - P) * (1 - P) = 2P - P2. By calculating P’ - P = 2P - P2 - P = P - P2, since 0 < P < 1, so P - P2 > 0, it can be known that P’ > P. Therefore, the requirement of improving the reliability of the power supply can be achieved by redundant design of the control part.

[0053] The overall redundancy of the control part is not a simple parallel operation. The two controls need to be interconnected, cooperate with each other, and work together as a whole to achieve the integrated control and monitoring of the power supply, so that the power supply as a whole appears as one power supply externally.

[0054] Please refer to Figure 1 , a master-slave complementary mode power supply control circuit, including:

[0055] The main power supply includes a first control system and a first power system, and the first control system and the first power system are communicatively connected; the first power system is used to convert the alternating current into a preset direct current according to the instruction of the first control system and output it;

[0056] The slave power supply includes a second control system and a second power system, and the second control system and the second power system are communicatively connected; the second power system is used to convert the alternating current into a preset direct current according to the instruction of the second control system and output it;

[0057] When the main / slave power supply fails, it switches to the slave / master power supply to continue working.

[0058] a) In the embodiment of the present invention, there are two power systems inside the power supply of the present invention, namely power system 1 and power system 2. The power system is responsible for the conversion of AC to DC, receiving the control of the control system, and adjusting or outputting the voltage according to the requirements of the control system.

[0059] b) The power supply consists of two control systems, control system 1 and control system 2. The two control systems are connected to the two power systems through the CAN bus to control the output voltage of the power systems. The control system also monitors the output power status of the power system to obtain the actual output voltage and current.

[0060] c) The user can control the entire power system through Ethernet 1, or through Ethernet 2, or through redundancy, through both Ethernet 1 and Ethernet 2. In this case, if a network cable is accidentally disconnected, it will not affect the use of the power supply at all;

[0061] d) The power supply has two sets of human-machine interactive front panels and display screens. These two sets of human-machine interactive hardware are connected to the corresponding control system. When the network is not used, and the buttons and display screen are used for local operation, any human-machine interactive device can be used for operation. In this way, even if one human-machine interactive device fails, the power supply can still be operated. Through the human-machine interactive device, the output voltage of the power supply can be adjusted and the output on and off can be controlled.

[0062] e) The entire power supply system is presented as a single power supply to the outside world. Whether it is operated through the human-machine interaction device such as the button display screen on the local side, or through the network (one or two control channels are possible), the power supply can be controlled. And all control instructions will be sent to the two power systems. The two power systems maintain synchronous actions, start or stop at the same time, and adjust the voltage at the same time. Since the outputs of the two power systems are finally gathered together through hardware. Therefore, from the external perspective, the internal dual-power and dual-control design details are hidden, and from the outside, it is just one power supply.

[0063] f) Control method.

[0064] There are two control systems inside the power supply. The two control systems are exactly the same in hardware. During the initialization process, the software initializes itself into two different states based on the judgment of the hardware. We call the power supply running the software in state A the master power supply, and the software running the software in state B the slave power supply. During initialization, which one is set as the master power supply and which one is set as the slave power supply has no effect on the entire power supply system. Therefore, the two control systems run the same software, but the functions they perform are not exactly the same.

[0065] like Figure 2As shown, the two control systems can communicate with each other through the CAN bus (CAN3 in the figure). When the system is powered on and the software is initialized, it will send a frame containing its own chip ID to CAN3 (the chip ID has been set by the manufacturer during the chip production process, and the IDs of any two chips will not be the same). After receiving the frame, the software will compare the received chip ID with its own chip ID. If its own ID is smaller, it will set its own address to 0 and confirm that it is the main power control system. If its own chip ID is larger than the received ID, it will set its own address to 1 and confirm that it is the slave power control system.

[0066] If the software does not receive a frame confirming the chip ID within 1 minute after powering on, it is assumed that the power supply is not in redundant working mode, and a prompt message is issued to indicate to the user that the power supply is in independent working mode.

[0067] After the master-slave relationship is confirmed, the master system sends a timing instruction to the slave system to coordinate the two systems to use the same timestamp. After that, all control instructions will be based on the later ones as the final valid instructions.

[0068] The main power control system receives a voltage setting instruction U from the outside s After that, it immediately sends the voltage setting command U to the slave power supply through CAN3. s , and also sends a voltage setting instruction U to the power system of the main power supply 主= U s After receiving the voltage setting instruction from CAN3, the control system immediately sends the voltage setting instruction to the power system. The voltage setting instruction is U 从= U s –XV (the X value is determined by the output voltage).

[0069] The power control system receives a voltage setting instruction U from the outside s After that, it immediately sends the voltage setting command U to the main power supply through CAN3 s At the same time, a voltage setting instruction is sent to the power system of the slave power supply, and the voltage setting instruction is U 从 =U s -XV. After the main control system receives the voltage setting instruction from CAN3, it immediately sends a voltage setting instruction to the power system. The voltage setting instruction is U 主 = U s ;

[0070] The main power control system can receive control signals from the local or remote end through the power control system. The control signal from the local end is the control signal of the buttons or display screen on the local end; the control signal from the remote end is the network command signal received through the network. The control signal includes the power distribution control signal and the voltage adjustment signal.

[0071] The power supply has local control and remote control working modes. In the local control mode, the power supply responds to various operation instructions from the local end. For the instructions from the network end, only read operations are allowed, and write operations are not allowed. In the remote control operation mode, the power supply responds to various operation instructions from the network end, and any operation instructions are prohibited from the local end.

[0072] Regardless of the master power supply or the slave power supply, when the power supply performs the power distribution action, when it receives the power distribution command, it also sends a frame of power distribution action command to the CAN3 bus. After the other power supply receives the power distribution action command, it immediately performs the power distribution action. The power distribution action includes power-on action and power-off action.

[0073] It should be noted that the master-slave complementary mode redundant control power supply of the present invention further includes:

[0074] The abnormal protection module is used to protect itself immediately when the power supply is abnormal, shut down the power supply and output, and immediately send abnormal alarm information to the CAN3 bus. After the other power supply receives the abnormal information, it will display a prompt. When the main power supply is abnormally protected, the slave power supply will immediately adjust its own voltage setting after receiving the module abnormal information sent by the main power control system from the CAN3 bus, and change the voltage setting value from U 从 =U s -XV changed to U 从 =U s .

[0075] The present invention also provides a power supply abnormality detection method, comprising:

[0076] Step 1: Obtain current sampling signals and voltage sampling signals at the output ends of the main power supply and the slave power supply;

[0077] Step 2: reconstructing the current sampling signal and the voltage sampling signal to form a reconstructed current sampling signal and a reconstructed voltage sampling signal;

[0078] Furthermore, step 2 includes:

[0079] Step 2.1: Performing wavelet transform on the current sampling signal and the voltage sampling signal to generate wavelet coefficients;

[0080] Step 2.2: Screening the wavelet coefficients based on a screening threshold to form screened wavelet coefficients; wherein the screening threshold is:

[0081]

[0082] in, represents the screening threshold, Indicates the preset coefficient, Indicates the length of the signal, represents the wavelet transform scale, represents the standard deviation of Gaussian noise;

[0083] In step 2.2, when the absolute value of the wavelet coefficient is greater than or equal to the screening threshold, the wavelet coefficient is screened using the first screening formula to form the screened wavelet coefficient; wherein the first screening formula is:

[0084]

[0085] When the absolute value of the wavelet coefficient is less than the screening threshold, the wavelet coefficient is screened using a second screening formula to form a screened wavelet coefficient;

[0086]

[0087] in, represents the kth wavelet coefficient under the jth transformation scale, represents the shrinkage factor, represents the convergence factor, Represents a symbolic function.

[0088] Step 2.3: reconstruct the screened wavelet coefficients to form a reconstructed current sampling signal and a reconstructed voltage sampling signal.

[0089] The present invention can decompose the signal into components of different frequencies by performing wavelet transform on the sampled signal, and can effectively identify and remove noise by using a screening formula, thereby improving the quality of the reconstructed sampled signal.

[0090] Step 3: performing image quantization encoding on the reconstructed current sampling signal and the reconstructed voltage sampling signal to form a current encoding image and a voltage encoding image;

[0091] Furthermore, step 3 includes:

[0092] Step 3.1: Obtain the value of the sampling point at each moment according to the reconstructed current sampling signal and the reconstructed voltage sampling signal;

[0093] Step 3.2: Fill the encoded pixel points into a box of a preset size from left to right and from top to bottom based on the encoding function to form a current encoding image and a voltage encoding image; wherein the encoding function is:

[0094]

[0095] In the formula, Represents the encoded pixel point, Indicates taking an integer. represents the value of the i-th sampling point, represents the minimum value of the sampled signal, Indicates the maximum value of the sampled signal.

[0096] The processing method of the present invention for converting signals into images by using a coding function can convert time-frequency data into a pixel matrix with image properties without any parameter definition and human intervention, and can enhance invisible information in the original signal to facilitate subsequent processing.

[0097] Step 4: splicing the current coded image and the voltage coded image to form a spliced ​​coded image;

[0098] Step 5: Setting a label for the spliced ​​coded image to indicate the fault type of the power supply;

[0099] Step 6: pre-process the spliced ​​coded images to form training samples, and input the training samples into a neural network for training to obtain a power anomaly detection model;

[0100] Furthermore, step 6 includes:

[0101] Step 6.1: Enhance the spliced ​​coded image to form a training sample; wherein the enhancement processing formula is:

[0102]

[0103] in, Indicates that the coded image after splicing is The pixel value at position, represents the first shape parameter, represents the second shape parameter, represents the third shape parameter, represents the standard deviation of the Gaussian distribution;

[0104] In the embodiment of the present invention, a, b, and c are parameters introduced to adjust the shape of the mapping function transformation curve. By adjusting a, b, and c, the contrast of the image can be improved. By performing Gaussian enhancement on the encoded image, noise and unnecessary details can be reduced, thereby helping the neural network to better learn features and reduce the risk of overfitting.

[0105] Step 6.2: Input the training samples into the neural network for training and use the loss function to verify the training effect of the neural network. When the loss value is less than the preset value, the power anomaly detection model is obtained.

[0106] Step 7: Use the power anomaly detection model to monitor the main power supply and the slave power supply to obtain a power supply monitoring result.

[0107] The present invention adopts a master-slave redundancy design, so that even if the main power supply fails, the slave power supply can continue to provide power, thereby ensuring the continuous operation of the system and improving the overall reliability.

[0108] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and the computer program is characterized in that when the computer program is executed by the processor, the steps in the above-mentioned power anomaly detection method are implemented. Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as the beneficial effects of the master-slave complementary mode power control circuit described in the above-mentioned technical solution, and will not be described in detail here.

[0109] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps in the above-mentioned power anomaly detection method. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the master-slave complementary mode power supply control circuit described in the above technical solution, and are not described in detail here.

[0110] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technical solution that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A power supply abnormality detection method, characterized in that: include: Step 1: Obtain current sampling signals and voltage sampling signals at the output ends of the main power supply and the slave power supply; Step 2: reconstructing the current sampling signal and the voltage sampling signal to form a reconstructed current sampling signal and a reconstructed voltage sampling signal; Step 3: performing image quantization encoding on the reconstructed current sampling signal and the reconstructed voltage sampling signal to form a current encoding image and a voltage encoding image; Step 4: splicing the current coded image and the voltage coded image to form a spliced ​​coded image; Step 5: Setting a label for the spliced ​​coded image to indicate the fault type of the power supply; Step 6: pre-process the spliced ​​coded images to form training samples, and input the training samples into a neural network for training to obtain a power anomaly detection model; Step 7: Use the power anomaly detection model to monitor the main power supply and the slave power supply to obtain a power supply monitoring result; The step 2: reconstructing the current sampling signal and the voltage sampling signal to form a reconstructed current sampling signal and a reconstructed voltage sampling signal, comprises: Step 2.1: Perform wavelet transform on the current sampling signal and the voltage sampling signal to generate wavelet coefficients; Step 2.2: Screening the wavelet coefficients based on a screening threshold to form screened wavelet coefficients; wherein the screening threshold is: Where t represents the screening threshold, ξ represents the preset coefficient, l represents the length of the signal, N represents the wavelet transform scale, and σ represents the standard deviation of Gaussian noise; Step 2.3: reconstructing the screened wavelet coefficients to form a reconstructed current sampling signal and a reconstructed voltage sampling signal; In step 2.2, when the absolute value of the wavelet coefficient is greater than or equal to the screening threshold, the wavelet coefficient is screened using the first screening formula to form the screened wavelet coefficient; wherein the first screening formula is: When the absolute value of the wavelet coefficient is less than the screening threshold, the wavelet coefficient is screened using a second screening formula to form a screened wavelet coefficient; Among them, w j,k represents the kth wavelet coefficient under the jth transformation scale, α represents the shrinkage factor, m represents the convergence factor, and sgn represents the sign function; The step 3: performing image quantization encoding on the reconstructed current sampling signal and the reconstructed voltage sampling signal to form a current encoding image and a voltage encoding image, comprises: Step 3.1: Obtain the value of the sampling point at each moment according to the reconstructed current sampling signal and the reconstructed voltage sampling signal; Step 3.2: Fill the encoded pixel points into a box of a preset size from left to right and from top to bottom based on the encoding function to form a current encoding image and a voltage encoding image; wherein the encoding function is: In the formula, P(m,n) represents the encoded pixel point, round represents an integer, and x i represents the value of the i-th sampling point, x min Indicates the minimum value of the sampled signal, x max Indicates the maximum value of the sampled signal; The step 6: pre-processing the spliced ​​coded images to form training samples, and inputting the training samples into a neural network for training to obtain a power anomaly detection model, including: Step 6.1: Enhance the spliced ​​coded image to form a training sample; wherein the enhancement processing formula is: Wherein, f(x, y) represents the pixel value of the spliced ​​coded image at the position (x, y), a represents the first shape parameter, b represents the second shape parameter, c represents the third shape parameter, and ζ represents the standard deviation of the Gaussian distribution; Step 6.2: Input the training samples into the neural network for training and use the loss function to verify the training effect of the neural network. When the loss value is less than the preset value, the power anomaly detection model is obtained.

2. An electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, wherein: When the computer program is executed by the processor, the steps in the power anomaly detection method according to claim 1 are implemented.

3. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the power anomaly detection method according to claim 1 are implemented.

4. A master-slave complementary mode power supply control circuit using the power supply anomaly detection method according to claim 1, characterized in that: include: A main power supply includes a first control system and a first power system, wherein the first control system and the first power system are in communication connection; the first power system is used to convert the alternating current into a preset direct current for output according to the instruction of the first control system; The slave power supply includes a second control system and a second power system, wherein the second control system and the second power system are in communication connection; the second power system is used to convert the alternating current into a preset direct current for output according to the instruction of the second control system; When the master / slave power fails, it switches to the slave / master power to continue working.

5. A master-slave complementary mode power supply control circuit according to claim 4, characterized in that: The first control system and the second control system communicate through the CAN bus. When the control system completes initialization, the chip ID of the first control system and the chip ID of the second control system are compared, and the address of the control system corresponding to the larger chip ID is set to 1, and it is confirmed as the control system of the main power supply. The address of the control system corresponding to the smaller chip ID is set to 0, and it is confirmed as the control system of the slave power supply.

6. A master-slave complementary mode power supply control circuit according to claim 5, characterized in that: Also includes: The abnormal protection module is used to protect the master / slave power supply immediately when an abnormality occurs in the master / slave power supply, shut down the power output, and immediately send abnormal alarm information through the CAN bus. At the same time, after receiving the abnormal alarm information, the slave / master power supply immediately adjusts according to the instructions of the control system to convert AC power into preset DC power for output.