Direct-current power supply online detection system and method

Abstract

The invention discloses a direct-current power supply on-line detection system and method, and belongs to the field of power supply safety detection. The method comprises the following steps: collecting operation data of a direct-current power supply, wherein the operation data comprises an insulation resistance value, an internal resistance value, a temperature and a power deviation; carrying out standardization processing on the collected operation data; weighting each operation data to construct a fault diagnosis model; training the fault diagnosis model to adjust and optimize model parameters; inputting the standardized operation data into the trained fault diagnosis model, calculating a DC power supply fault score S, and setting a threshold value Z if Sgt; and Z, judging that the direct-current power supply fails. According to the method, the insulation resistance value, the internal resistance, the temperature and the power deviation are introduced to serve as indexes for detecting the direct-current power supply, and the fault judgment model is constructed by weighting the indexes, so that the fault of the direct-current power supply can be detected and judged more accurately.

Description

Technical Field

[0001] The invention belongs to the field of power supply safety detection, and in particular relates to a direct current power supply online detection system and method. Background Art

[0002] DC power supply online detection technology is an advanced means of power system maintenance. It can monitor the operating status of the DC power supply system in real time and detect potential faults and safety hazards in time. This technology is crucial to ensure the stable operation of the power system, especially in key facilities such as substations.

[0003] In the prior art, the detection of DC power supply mostly relies on the detection of the internal resistance of the power supply. When the internal resistance reaches a certain value, it can be determined that the DC power supply has a fault. However, there are many types of DC power supply faults, such as temperature, insulation, etc. Therefore, it is obviously not accurate enough to monitor the operating status of the DC power supply only by detecting the internal resistance. Therefore, a DC power supply online detection system and method are proposed. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention aims to provide a direct current power supply online detection system and method, which solves the problems in the prior art.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A method for online detection of a DC power supply comprises the following steps:

[0007] Collect the operating data of the DC power supply, including: insulation resistance value, internal resistance, temperature and power deviation;

[0008] Standardize the collected operation data;

[0009] Weighting each operating data to build a fault diagnosis model;

[0010] Train the fault diagnosis model to optimize the model parameters;

[0011] The standardized operating data is input into the trained fault diagnosis model, the DC power supply fault score S is calculated, and a threshold Z is set to compare with S. If S>Z, the DC power supply is judged to be faulty.

[0012] Furthermore, the formula for standardization is:

[0013]

[0014] In the formula, Y is the original operation data, Y′ is the standardized operation data, and Y min is the minimum value in the data set, Y maxis the maximum value in the data set.

[0015] Furthermore, the mathematical expression of the fault diagnosis model is:

[0016] S=ω1×IR+ω2×R+ω3×T+ω4×ΔP+b

[0017] Where S represents the DC power supply fault score, IR is the insulation resistance value, R is the internal resistance, T is the temperature, ΔP is the power deviation, b is the bias constant; ω1, ω2, ω3 and ω4 represent the weights of the insulation resistance value, internal resistance, temperature and power deviation respectively.

[0018] Furthermore, the training process of the fault diagnosis model is as follows:

[0019] S41, dividing the normalized running data into a training set and a test set in a ratio of 7:3;

[0020] S42, constructing a feature matrix X using the standardized running data:

[0021]

[0022] S43, randomly initialize weights and bias constants;

[0023] S44, use the cross entropy loss function to evaluate the gap between the model output and the true label:

[0024]

[0025] Where N is the number of samples, is the predicted probability of the model output, σ is the sigmoid function;

[0026] S44, calculate the fault score S:

[0027] S i =ω1×IR i +ω2×R i +ω3×T i +ω4×ΔP i +b

[0028] Apply the activation function to get the predicted probabilities:

[0029]

[0030] S45, calculate the gradient of the loss function with respect to weights and biases:

[0031]

[0032] Where j is the index of the feature;

[0033] S46, update weights and biases using gradient descent:

[0034]

[0035] Among them, α is the learning rate;

[0036] S47, use K-fold cross validation to evaluate the performance of the model, adjust the learning rate and number of iterations, and perform a systematic search for different learning rates and other hyperparameters to find the best combination.

[0037] Furthermore, the threshold Z is determined based on the distribution of the scores S of all samples when the DC power supply is in a normal operating state.

[0038] Furthermore, the insulation resistance value is measured by an insulation resistance tester; the internal resistance is measured by an internal resistance tester to measure the internal resistance of the power supply; the temperature is monitored and read by a temperature sensor to read the external temperature of the power supply; the power deviation is obtained by taking the absolute value after subtracting the actual output power of the monitored power supply from the rated power.

[0039] A direct current power supply online detection system, comprising:

[0040] Data acquisition module: collects the operating data of the DC power supply, including: insulation resistance value IR, internal resistance R, temperature T and power deviation ΔP;

[0041] Data processing module: standardize the collected operation data;

[0042] Model building module: weighting each operating data to build a fault diagnosis model;

[0043] Model training module: trains the fault diagnosis model to optimize the model parameters;

[0044] And, the fault diagnosis module: inputs the standardized operation data into the trained fault diagnosis model, calculates the DC power supply fault score S, and sets the threshold Z to compare with S. If S>Z, the DC power supply is judged to be faulty.

[0045] A computer storage medium stores a readable program, and when the program is running, the above-mentioned direct current power supply online detection method can be executed.

[0046] An electronic device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other through the communication bus;

[0047] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned direct current power supply online detection method.

[0048] A computer program product includes computer instructions, wherein the computer instructions instruct a computing device to execute operations corresponding to the above-mentioned direct current power supply online detection method.

[0049] Beneficial effects of the present invention:

[0050] The present invention introduces insulation resistance value, internal resistance, temperature and power deviation as indicators for detecting DC power supply, and constructs a fault judgment model by weighting each indicator, so as to more accurately detect and judge the fault of DC power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] 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, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0052] Figure 1 It is a flow chart of the direct current power supply online detection method of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with 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 of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] Example 1

[0055] A method for online detection of a DC power supply comprises the following steps:

[0056] S1, collects the operating data of the DC power supply;

[0057] Operation data include: insulation resistance value IR, internal resistance R, temperature T and power deviation ΔP;

[0058] Among them, the insulation resistance value IR is measured by an insulation resistance tester (digital insulation resistance tester or pointer insulation resistance tester can be selected); the internal resistance R data is measured by the internal resistance tester to measure the internal resistance of the power supply; the temperature T is read by monitoring the external temperature of the power supply through the temperature sensor. The power deviation ΔP is obtained by subtracting the actual output power of the monitored power supply from the rated power (taking the absolute value).

[0059] S2, preprocessing the operation data;

[0060] Data preprocessing includes standardizing the running data to make it on the same scale to facilitate later input into the model;

[0061] In this embodiment, the Min-Max normalization method is used to pre-process the operating data;

[0062] The purpose of Min-Max normalization is to scale the data to the range of [0,1]. The formula is:

[0063]

[0064] In the formula, Y is the original operation data, Y′ is the standardized operation data, and Y min is the minimum value in the data set, Y max is the maximum value in the data set.

[0065] S3, weighting each operation data to build a fault diagnosis model;

[0066] The mathematical expression of the fault diagnosis model is:

[0067] S=ω1×IR+ω2×R+ω3×T+ω4×ΔP+b

[0068] Where S represents the DC power supply fault score, IR is the DC power supply insulation resistance, R is the DC power supply internal resistance, T is the DC power supply temperature, ΔP is the DC power supply power deviation, b is the bias constant of the fault diagnosis model; ω1, ω2, ω3 and ω4 represent the weights of the DC power supply insulation resistance, DC power supply internal resistance, DC power supply temperature and DC power deviation, respectively.

[0069] S4, training the fault diagnosis model to optimize the model parameters including: the weights ω1, ω2, ω3, ω4 of each operating data and the bias constant b;

[0070] The training process of the fault diagnosis model is:

[0071] S41, dividing the normalized running data into a training set and a test set in a ratio of 7:3;

[0072] S42, constructing a feature matrix X using the standardized running data:

[0073]

[0074] S43, randomly initialize weights and bias constants;

[0075] S44, define the loss function;

[0076] The cross entropy loss function is used to evaluate the gap between the model output and the true label:

[0077]

[0078] Where N is the number of samples, is the predicted probability of the model output, σ is the sigmoid function

[0079] S44, calculate the fault score S:

[0080] S i =ω1×IR i +ω2×R i +ω3×T i +ω4×ΔP i +b

[0081] Apply the activation function to get the predicted probabilities:

[0082]

[0083] S45, calculate the gradient of the loss function with respect to weights and biases:

[0084]

[0085] Where j is the index of the feature;

[0086] S46, update weights and biases using gradient descent:

[0087]

[0088] Among them, α is the learning rate;

[0089] S47, use K-fold cross validation to evaluate the performance of the model, adjust hyperparameters such as learning rate and number of iterations, and perform a systematic search for different learning rates and other hyperparameters to find the best combination.

[0090] S5, input the standardized operation data into the trained fault diagnosis model, calculate the DC power supply fault score S, and set the threshold Z to compare with S. If S>Z, the DC power supply is judged to be faulty and an alarm message is generated.

[0091] The threshold Z may be determined based on the distribution of the scores S of all samples of the DC power supply under normal operating conditions; in some embodiments, the threshold Z is set to the mean of the normal scores plus a certain multiple of the standard deviation.

[0092] Based on similar inventive concepts, an embodiment of the present invention further provides a computer storage medium storing a readable program, which can execute the above-mentioned DC power supply online detection method when the program is running.

[0093] Based on similar inventive concepts, an embodiment of the present invention provides an electronic device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other through the communication bus;

[0094] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned direct current power supply online detection method.

[0095] Based on similar inventive concepts, an embodiment of the present invention further provides a computer program product, including computer instructions, wherein the computer instructions instruct a computing device to execute operations corresponding to the above-mentioned DC power supply online detection method.

[0096] Example 2

[0097] In this embodiment, a DC power supply online detection system is proposed, which specifically includes:

[0098] Data acquisition module: collects the operating data of the DC power supply, including: insulation resistance value IR, internal resistance R, temperature T and power deviation ΔP;

[0099] Data processing module: standardize the collected operation data

[0100] Model building module: weighting each operating data to build a fault diagnosis model;

[0101] Model training module: trains the fault diagnosis model to optimize the model parameters;

[0102] And, the fault diagnosis module: inputs the standardized operation data into the trained fault diagnosis model, calculates the DC power supply fault score S, and sets the threshold Z to compare with S. If S>Z, the DC power supply is judged to be faulty.

[0103] The method of the present invention may be implemented in hardware, firmware, or as software or computer code that may be stored in a recording medium (such as a CDROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code that is originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded over a network and will be stored in a local recording medium, so that the method described herein may be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It is understood that a computer, processor, microprocessor controller, or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, processor, or hardware, the method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the method shown herein, the execution of the code converts the general-purpose computer into a dedicated computer for executing the method shown herein.

[0104] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A method for online detection of a DC power supply, characterized in that: The following steps are involved: Collect the operating data of the DC power supply, including: insulation resistance value, internal resistance, temperature and power deviation; Standardize the collected operation data; Weighting each operating data to build a fault diagnosis model; Train the fault diagnosis model to optimize the model parameters; The standardized operating data is input into the trained fault diagnosis model, the DC power supply fault score S is calculated, and a threshold Z is set to compare with S. If S>Z, the DC power supply is judged to be faulty.

2. A DC power supply online detection method according to claim 1, characterized in that: The formula for standardization is: In the formula, Y is the original operation data, Y′ is the standardized operation data, and Y min is the minimum value in the data set, Y max is the maximum value in the data set.

3. A DC power supply online detection method according to claim 1, characterized in that: The mathematical expression of the fault diagnosis model is: S=ω1×IR+ω2×R+ω3×T+ω4×ΔP+b Where S represents the DC power supply fault score, IR is the insulation resistance value, R is the internal resistance, T is the temperature, ΔP is the power deviation, b is the bias constant; ω1, ω2, ω3 and ω4 represent the weights of the insulation resistance value, internal resistance, temperature and power deviation respectively.

4. A DC power supply online detection method according to claim 3, characterized in that: The training process of the fault diagnosis model is as follows: S41, dividing the normalized running data into a training set and a test set in a ratio of 7:3; S42, constructing a feature matrix X using the standardized running data: S43, randomly initialize weights and bias constants; S44, use the cross entropy loss function to evaluate the gap between the model output and the true label: Where N is the number of samples, is the predicted probability of the model output, σ is the sigmoid function; S44, calculate the fault score S: S i =ω1×IR i +ω2×R i +ω3×T i +ω4×ΔP i +b Apply the activation function to get the predicted probabilities: S45, calculate the gradient of the loss function with respect to weights and biases: Where j is the index of the feature; S46, update weights and biases using gradient descent: Among them, α is the learning rate; S47, use K-fold cross validation to evaluate the performance of the model, adjust the learning rate and number of iterations, and perform a systematic search for different learning rates and other hyperparameters to find the best combination.

5. A DC power supply online detection method according to claim 1, characterized in that: The threshold Z is determined based on the distribution of the scores S of all samples when the DC power supply is in a normal operating state.

6. A DC power supply online detection method according to claim 1, characterized in that: The insulation resistance value is measured by an insulation resistance tester; the internal resistance is measured by an internal resistance tester to measure the internal resistance of the power supply; the temperature is monitored and read by a temperature sensor to read the external temperature of the power supply; the power deviation is obtained by taking the absolute value after subtracting the actual output power of the monitored power supply from the rated power.

7. A DC power supply online detection system, characterized in that: include: Data acquisition module: collects the operating data of the DC power supply, including: insulation resistance value IR, internal resistance R, temperature T and power deviation ΔP; Data processing module: standardize the collected operation data; Model building module: weighting each operating data to build a fault diagnosis model; Model training module: trains the fault diagnosis model to optimize the model parameters; And, the fault diagnosis module: inputs the standardized operation data into the trained fault diagnosis model, calculates the DC power supply fault score S, and sets the threshold Z to compare with S. If S>Z, the DC power supply is judged to be faulty.

8. A computer storage medium storing a readable program, characterized in that: When the program is running, it can execute the direct current power supply online detection method described in any one of claims 1 to 6.

9. An electronic device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform an operation corresponding to a DC power supply online detection method according to any one of claims 1 to 6.

10. A computer program product comprising computer instructions, characterized in that: The computer instructions instruct the computing device to execute operations corresponding to the method for online detection of a DC power supply as described in any one of claims 1-6.

Images