Primary and secondary fusion complete ring main unit fault diagnosis method and system

Abstract

The invention relates to the technical field of electrical variable measurement, in particular to a primary and secondary fusion complete ring main unit fault diagnosis method and system, and the method comprises the steps: obtaining current curves of N windows and correction voltage curves of the N windows; determining a voltage correction error and generating a voltage correction error curve of the N windows according to each correction voltage value in the correction voltage curve of the N windows and a plurality of voltage values corresponding to the correction voltage values; correcting the voltage correction error curve of the ith window by using the fluctuation trend of the current curve of the ith window to obtain a target voltage correction error curve of the ith window; determining the current abnormal probability of the current curve of the ith window according to the current curve of the ith window and the warping path distance of the target voltage correction error curve of the ith window; and when the current abnormal probability meets a preset fault condition, determining that the primary and secondary fusion complete ring main unit has a fault. The accuracy of fault diagnosis is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of measuring electric variables, and in particular to a method and system for diagnosing faults of a primary-secondary integrated ring network box. Background Art

[0002] The primary and secondary integrated ring network box integrates the power system of primary and secondary equipment. It is an electric power equipment that organically integrates high-voltage primary equipment (such as circuit breakers, disconnectors, etc.) with secondary equipment (such as measurement, control, protection, devices, etc.). With the development of technology, the primary and secondary integrated ring network box will develop in the direction of being more intelligent, miniaturized, and efficient. The sensor is used as a secondary device to measure voltage, current and other data. Compared with the measurement using a transformer, the primary and secondary integrated ring network box is easier to achieve high integration because the sensor is small in size, easy to install, and has higher sampling accuracy and frequency.

[0003] The primary equipment in the primary-secondary integrated ring network box will cause strong electromagnetic interference to the secondary measurement equipment. Currently, the interference is removed by setting up an electromagnetic shielding cover or filtering. However, the electromagnetic interference generated by the primary equipment is strong and random. A single shielding cover cannot completely remove the interference. The use of filtering will cause the interference to change, resulting in the corrected measurement data being unable to fully reflect the actual changes, resulting in a low accuracy rate in the final fault diagnosis. Summary of the invention

[0004] In order to solve the technical problem that the fault diagnosis accuracy of the primary and secondary integrated ring network box is not high, the purpose of the present invention is to provide a primary and secondary integrated ring network box fault diagnosis method and system, the technical scheme adopted is as follows: Obtain current curves of N windows and corrected voltage curves of N windows; wherein the corrected voltage curve is generated according to the corrected voltage value, which is obtained by performing electromagnetic interference correction on multiple voltage values ​​of the primary and secondary integrated ring network box, and the multiple voltage values ​​are obtained by setting shielding layers with different coverage areas on multiple voltage sensors and collecting voltages of the primary and secondary integrated ring network box; Determine voltage correction errors of the N windows according to each correction voltage value in the correction voltage curve of the N windows and multiple voltage values ​​corresponding to the correction voltage value, and generate voltage correction error curves of the N windows; Using the fluctuation trend of the current curve of the i-th window, the voltage correction error curve of the i-th window is corrected to obtain the target voltage correction error curve of the i-th window; Determine the current abnormal probability of the current curve of the i-th window according to the normalized path distance of the current curve of the i-th window and the target voltage correction error curve of the i-th window; When the current abnormal probability of the current curve in the i-th window meets the preset fault condition, it is determined that the primary and secondary integrated ring network box has a fault; wherein N and i are positive integers.

[0005] Preferably, determining the voltage correction errors of the N windows according to each correction voltage value in the correction voltage curves of the N windows and a plurality of voltage values ​​corresponding to the correction voltage value comprises: For the j-th corrected voltage value, calculating an average value of multiple voltage values ​​corresponding to the j-th corrected voltage value; The difference between the average value and the j-th corrected voltage value is calculated to obtain the voltage correction error corresponding to the j-th corrected voltage value; wherein j is a positive integer.

[0006] Preferably, the method for determining the correction voltage value includes: Determining a relative interference change rate and an electromagnetic interference resistance strength of a shielding layer of a plurality of voltage sensors according to a plurality of voltage values; Determine a plurality of predicted voltage values ​​that are not affected by electromagnetic interference according to the relative interference change rate, the electromagnetic interference resistance strength of the shielding layers of the plurality of voltage sensors, and each voltage value; The average value of multiple predicted voltage values ​​is calculated to obtain a corrected voltage value.

[0007] Preferably, determining the relative interference change rate and the anti-electromagnetic interference strength of the shielding layers of the plurality of voltage sensors according to the plurality of voltage values ​​comprises: Calculate the voltage difference between each two voltage values ​​respectively; According to the ratio of the voltage difference and the difference of the corresponding anti-electromagnetic interference strength, an overdetermined set of equations is formed; The overdetermined equations are calculated using the least square method to obtain the relative interference change rate and the anti-electromagnetic interference strength of the shielding layers of multiple voltage sensors.

[0008] Preferably, the voltage correction error curve of the i-th window is corrected by using the fluctuation trend of the current curve of the i-th window to obtain the target voltage correction error curve of the i-th window, including: Performing an exponential operation on the difference between the variance of the current curve of the i-th window and the variance of the current curve of m windows before the i-th window, to obtain a current fluctuation coefficient; Calculate the product of the current fluctuation coefficient and the window length of the voltage correction error curve of the i-th window to obtain the correction length; The correction length and the window length of the voltage correction error curve of the i-th window are summed to obtain the target voltage correction error curve of the i-th window.

[0009] Preferably, when the current abnormal probability of the current curve in the i-th window meets the preset fault condition, determining that the primary and secondary fusion ring network box has a fault includes: According to the current abnormal probability and the historical abnormal probability of the current curve of the i-th window, the failure probability of the primary and secondary integrated ring network box corresponding to the i-th window is determined; When the fault probability is greater than a preset fault probability threshold, it is determined that the primary and secondary fused ring network box has a fault.

[0010] Preferably, according to the current abnormal probability and the historical abnormal probability of the current curve of the i-th window, the failure probability of the primary and secondary fusion ring network box corresponding to the i-th window is determined, including: Get the historical abnormal probability of the i-th window in the previous k days; The difference between the current abnormal probability and the historical abnormal probability of the i-th window of the previous k days is calculated respectively, and the k differences are summed to obtain the failure probability of the primary and secondary fusion ring network box corresponding to the i-th window; where k is a positive integer.

[0011] Preferably, when the current abnormal probability of the current curve of the i-th window meets the preset fault condition, after determining that the primary and secondary fusion ring network box has a fault, the method further includes: Obtain relevant curves of the primary and secondary fusion ring network box of the i-th window, wherein the relevant curves include current curve, voltage curve, temperature curve and humidity curve; The relevant curves are input into the preset anomaly detection model to predict the fault type, and the fault type of the primary and secondary integrated ring network box is output.

[0012] Preferably, the method further comprises: The fault type and related curves of the primary and secondary integrated ring network box are sent to the terminal equipment.

[0013] The present invention also provides a primary and secondary fusion complete ring network box fault diagnosis system, comprising: An acquisition module is used to acquire current curves of N windows and corrected voltage curves of N windows; wherein the corrected voltage curve is generated according to a corrected voltage value, which is obtained by performing electromagnetic interference correction on multiple voltage values ​​of a primary and secondary integrated ring network box, and the multiple voltage values ​​are obtained by setting shielding layers with different coverage areas on multiple voltage sensors and collecting voltages of the primary and secondary integrated ring network box; A generating module, used for determining voltage correction errors of the N windows according to each correction voltage value in the correction voltage curve of the N windows and a plurality of voltage values ​​corresponding to the correction voltage value, and generating voltage correction error curves of the N windows; A correction module, used to correct the voltage correction error curve of the i-th window by using the fluctuation trend of the current curve of the i-th window, so as to obtain the target voltage correction error curve of the i-th window; A calculation module, used to determine the current abnormal probability of the current curve of the i-th window according to the current curve of the i-th window and the normalized path distance of the target voltage correction error curve of the i-th window; The detection module is used to determine that there is a fault in the primary and secondary integrated ring network box when the current abnormal probability of the current curve in the i-th window meets the preset fault condition; wherein N and i are positive integers.

[0014] The present invention has the following beneficial effects: Obtain current curves of N windows and corrected voltage curves of N windows; wherein the corrected voltage curve is generated based on a corrected voltage value, the corrected voltage value is obtained by performing electromagnetic interference correction on multiple voltage values ​​of a primary and secondary integrated ring network box, and the multiple voltage values ​​are obtained by setting shielding layers with different coverage areas on multiple voltage sensors and collecting voltages of the primary and secondary integrated ring network box; determine the voltage correction errors of the N windows based on each corrected voltage value in the corrected voltage curves of the N windows and multiple voltage values ​​corresponding to the corrected voltage value, and generate a voltage correction error curve for the N windows; correct the voltage correction error curve of the i-th window using the fluctuation trend of the current curve of the i-th window to obtain a target voltage correction error curve of the i-th window; determine the current abnormal probability of the current curve of the i-th window based on the normalized path distance of the current curve of the i-th window and the target voltage correction error curve of the i-th window; when the current abnormal probability of the current curve of the i-th window meets the preset fault condition, determine that the primary and secondary integrated ring network box has a fault. It can be seen that when the fluctuation of the voltage correction error curve is similar to the fluctuation of the current curve, there may be a fault. Therefore, the current abnormal probability of the current curve of the i-th window is determined by the normalized path distance of the current curve of the i-th window and the target voltage correction error curve of the i-th window, and then whether the primary and secondary integrated ring network box has a fault is further determined by whether the current abnormal probability meets the preset fault condition, so as to improve the accuracy of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are 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.

[0016] Figure 1 A method flow chart of a method for diagnosing faults of a primary and secondary fusion ring network box provided by an embodiment of the present invention; Figure 2 A specific flow chart of step S130 in a method for diagnosing faults of a primary and secondary fused ring network box provided by an embodiment of the present invention; Figure 3 A specific flow chart of step S150 in a method for diagnosing faults of a primary and secondary fused ring network box provided by an embodiment of the present invention; Figure 4 A schematic diagram of a primary-secondary integrated ring network box fault diagnosis system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the fault diagnosis method and system of a primary and secondary fusion ring network box proposed by the present invention, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The specific scheme of the primary and secondary fusion ring network box fault diagnosis method and system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a method flow chart of a method for diagnosing a fault of a primary and secondary integrated ring network box provided by an embodiment of the present invention. In an exemplary embodiment, a method for diagnosing a fault of a primary and secondary integrated ring network box is provided, comprising: S110, obtaining current curves of N windows and corrected voltage curves of N windows; wherein the corrected voltage curve is generated according to the corrected voltage value, the corrected voltage value is obtained by performing electromagnetic interference correction on multiple voltage values ​​of the primary and secondary integrated ring network box, and the multiple voltage values ​​are obtained by setting shielding layers with different coverage areas on multiple voltage sensors and collecting voltages of the primary and secondary integrated ring network box; S120, determining voltage correction errors of the N windows according to each correction voltage value in the correction voltage curves of the N windows and multiple voltage values ​​corresponding to the correction voltage value, and generating voltage correction error curves of the N windows; S130, using the fluctuation trend of the current curve of the i-th window, correcting the voltage correction error curve of the i-th window to obtain a target voltage correction error curve of the i-th window; S140, determining a current abnormal probability of the current curve of the i-th window according to the normalized path distance of the current curve of the i-th window and the target voltage correction error curve of the i-th window; S150. When the current abnormal probability of the current curve in the i-th window meets the preset fault condition, it is determined that there is a fault in the primary-secondary integrated ring network box; wherein N and i are positive integers.

[0021] In step S110, illustratively, the current curves of the N windows are obtained by windowing the current curves in a preset time period. The current curves in the preset time period are generated by measuring the current value of the ring network box by a current transformer.

[0022] The corrected voltage curves of N windows are obtained by windowing the corrected voltage curves within a preset time period, and the preset time period can be 1 day (i.e., 24 hours), 2 days, 3 days, etc., which are not limited here. The current curve within the preset time period and the corrected voltage curve within the preset time period are divided by windows in the same order. Specifically, multiple voltage values ​​are measured by using a capacitive voltage sensor in a resistive voltage division method to measure the voltage of a primary and secondary fused ring network box. In this embodiment, 4 voltage sensors are provided, and each voltage sensor is provided with a shielding layer of a different coverage area. The material and thickness of the shielding layer are the same, and only the coverage area is different. They are numbered 1, 2, 3, and 4 from small to large coverage area; the size of the coverage area is determined by the number of openings; the frequency of each sensor during measurement remains consistent.

[0023] Preferably, the method for determining the corrected voltage value includes: determining the relative interference change rate and the anti-electromagnetic interference strength of the shielding layers of multiple voltage sensors based on multiple voltage values; determining multiple predicted voltage values ​​that are not affected by electromagnetic interference based on the relative interference change rate, the anti-electromagnetic interference strength of the shielding layers of multiple voltage sensors and each voltage value; and calculating the average value of the multiple predicted voltage values ​​to obtain the corrected voltage value.

[0024] Optionally, determining the relative interference change rate and the anti-electromagnetic interference strength of the shielding layers of the plurality of voltage sensors according to the plurality of voltage values ​​includes: The voltage difference between each two voltage values ​​is calculated respectively; an overdetermined set of equations is formed according to the ratio of the voltage difference to the difference in the corresponding anti-electromagnetic interference strength. The overdetermined set of equations is calculated using the least squares method to obtain the relative interference change rate and the anti-electromagnetic interference strength of the shielding layer of multiple voltage sensors.

[0025] Specifically, generally high-power primary-side devices will generate higher electric and magnetic fields, and as the switch is closed and opened, and high-power electrical appliances are connected, the electromagnetic field will change to a certain extent. At the same time, it is believed that the electromagnetic signals received by the outside of the shielding layers of different voltage sensors are the same, and the only difference is caused by the shielding cover. The material of the shielding layer is the same, so the anti-interference signal change rate of different shielding covers relative to the voltage difference is the same.

[0026] Take the setting of 4 voltage sensors as an example, which are represented as voltage sensor No. 1, voltage sensor No. 2, voltage sensor No. 3 and voltage sensor No. 4. The isolation degree of voltage sensors No. 1 to No. 4 is from small to large, and the internal anti-electromagnetic interference strength is set in turn. , , , ; In contrast, the voltage values ​​collected by each voltage sensor are , , , .

[0027] So, for , The relative interference change rate of the voltage difference is:

[0028] in, Indicates the relative interference change rate between voltage sensor No. 1 and voltage sensor No. 2; , Respectively represent the voltage values ​​measured by No. 1 and No. 2 voltage sensors; , They respectively represent the anti-electromagnetic interference strength of the shielding covers of the No. 1 and No. 2 voltage sensors.

[0029] Similarly, different relative interference change rates of voltage differences are obtained according to any combination of voltage values ​​of four different isolation levels, such as , , , , It can be understood that the relative interference change rate is the same. Then they are combined to form an overdetermined set of equations, and the least squares method is used to approximate the results to obtain the anti-electromagnetic interference strength inside different shielding covers and the relative interference change rate.

[0030] According to the relative interference change rate, the voltage signal not affected by electromagnetic interference is calculated. Since the electromagnetic interference intensity at the location without electromagnetic interference is 0, the relative interference change rate can be calculated according to the formula:

[0031] in, Indicates the voltage value measured by voltage sensor No. 1; Indicates the predicted voltage value not affected by electromagnetic interference; Indicates the anti-electromagnetic interference strength inside the shielding cover of sensor No. 1; Represents the relative interference change rate. Simplifying the above formula, we can get: .

[0032] In order to avoid data differences caused by the least squares fitting solution, the mean of each is considered when calculating the unaffected voltage value, so as to obtain the final unaffected corrected voltage value:

[0033] in, Represents the voltage value measured by the i-th voltage sensor; represents the relative disturbance change rate; Indicates the anti-electromagnetic interference strength inside the i-th voltage sensor.

[0034] It should be noted that the data acquisition frequency of each sensor is the same, and the above process is calculated using data of the same frequency. A correction voltage curve is generated according to the correction voltage value within a preset time period, and window division is performed, for example, every 100 consecutive data is used as a window.

[0035] In step S120, exemplarily, the source of the drastic change in electromagnetic interference is mainly due to operations such as the connection of high-power electrical appliances and the closing of the primary device switch. The capacitive voltage sensor will affect the measurement result due to the change in capacitance, and the current transformer usually uses the law of electromagnetic induction through a closed iron core and winding to generate an induced electromotive force in the secondary winding to obtain current; therefore, if both the voltage and current produce similar fluctuating signals, it is considered to be caused by electromagnetic interference and the equipment is in normal condition, otherwise it is considered that the possibility of equipment failure is high.

[0036] The correction error can indicate the fluctuation of electromagnetic interference. If the error fluctuation in an interval is large, the fluctuation trend of the current signal should be large. Therefore, the closer the change trend of the voltage correction error and the current is, the higher the possibility of the impact caused by electromagnetic interference and the lower the possibility of equipment failure. Therefore, it is necessary to determine the voltage correction error.

[0037] Preferably, the method for determining the voltage correction error includes: S1210, for the j-th corrected voltage value, calculating an average value of multiple voltage values ​​corresponding to the j-th corrected voltage value; S1220, calculating the difference between the average value and the j-th corrected voltage value, and obtaining a voltage correction error corresponding to the j-th corrected voltage value.

[0038] Exemplarily, four voltage values ​​measured by four voltage sensors at a sampling moment are obtained, and the average of the four voltage values ​​is calculated. A corrected voltage value is calculated, and the average value is subtracted from the corrected voltage value to obtain a voltage correction error at the sampling moment.

[0039] Specifically, the voltage correction error at each sampling moment of the voltage signal is calculated according to the following formula:

[0040] in, represents the voltage correction error corresponding to the j-th correction voltage value; represents the average value of multiple voltage values ​​corresponding to the j-th corrected voltage value; indicates the jth corrected voltage value.

[0041] Finally, the voltage correction error calculation is performed on the voltage value at each sampling moment to obtain multiple voltage correction errors. The multiple voltage correction errors are taken as a point, and each point is curve-fitted to finally obtain a voltage correction error curve.

[0042] In step S130, exemplarily, when the electromagnetic interference is relatively small, since the corrected voltage error signal curve is obtained based on the difference mean of voltage sensors under multiple different shielding conditions, the sensor signal with a higher shielding effect and the sensor signal with a lower shielding effect cancel each other out, which may cause the curve to have a smaller fluctuation trend in a certain window. Therefore, it is necessary to appropriately extend the comparison correction curve to highlight the fluctuation trend of the current window. The change in current can reflect the severity of electromagnetic interference to a certain extent. If the fluctuation trend of the current is larger, the longer the length of the selected error comparison curve is. Therefore, it is necessary to correct the voltage correction error curve corresponding to the current curve of the i-th window.

[0043] Preferably, if Figure 2 As shown, step S130 includes: S1310, performing an exponential operation on the difference between the variance of the current curve of the i-th window and the variance of the current curve of m windows before the i-th window, to obtain a fluctuation trend of the current curve of the i-th window; S1320, calculating the product of the fluctuation trend of the current curve of the i-th window and the window length of the voltage correction error curve of the i-th window to obtain the correction length; S1330 , performing a sum operation on the correction length and the window length of the voltage correction error curve of the i-th window to obtain a target voltage correction error curve of the i-th window.

[0044] Specifically, the calculation formula of the window length of the target voltage correction error curve of the i-th window is as follows:

[0045] in, represents the window length of the target voltage correction error curve corresponding to the i-th window current curve; represents the window length of the voltage correction error curve of the i-th window; represents the variance of the current curve of the i-th window; It represents the variance of the current curve of m windows before the i-th window.

[0046] The variance of the current curve of the i-th window is obtained by calculating the variance of the current value in the current curve of the i-th window. The variance of the current curve of the m windows before the i-th window is obtained by selecting multiple windows before the i-th window, for example, 10. The variance of the current value in the current curve of the m windows is calculated to obtain the variance of the current curve of the m windows, and the average value of the variance of the current curve of the m windows is calculated as the variance of the current curve of the m windows before the i-th window.

[0047] Furthermore, the corrected voltage curves of N windows are obtained by windowing the corrected voltage curves within a preset time period, and the difference in length between the window length of the target voltage correction error curve and the window length of the voltage correction error curve of the i-th window is calculated. Based on the corrected voltage curve, the voltage correction error curve of the i-th window is increased by the window length of the length difference to form the target voltage correction error curve of the i-th window. In step S140, illustratively, the similarity of the fluctuation trends of the lower current and the correction error of each window is calculated respectively. If the fluctuation trends of the two are more similar, it means that the possibility of the impact caused by normal conditions is greater and the probability of suspected abnormalities is lower. Therefore, the calculation method for the current abnormal probability of the current curve of the i-th window is:

[0048] in, represents the current abnormal probability of the current curve in the i-th window; represents the normalized path distance, i.e., the dynamic time warping distance; represents the current curve of the i-th window; Represents the target voltage correction error curve of the i-th window corresponding to the i-th window current curve. It should be noted that the higher the similarity between the current change trend and the target voltage correction error curve, the lower the suspected abnormality. The higher the similarity between the current change trend and the target voltage correction error curve, the smaller the normalization path distance.

[0049] In step S150, illustratively, if the current abnormal probability is similar to the historical performance, it means that the change in the current window is caused by normal demand and the failure probability is low. Therefore, the preset failure condition can be that the similarity between the current abnormal probability and the historical abnormal probability exceeds the preset similarity.

[0050] Preferably, if Figure 3 As shown, step S150 includes: S1510, determining the failure probability of the primary and secondary fusion ring network box corresponding to the i-th window according to the current abnormal probability and the historical abnormal probability of the current curve of the i-th window; S1520: When the fault probability is greater than a preset fault probability threshold, determine that the primary and secondary integrated ring network box has a fault. The preset fault probability threshold is set according to actual conditions and is not limited here. For example, 0.27.

[0051] Step S1510 includes: obtaining the historical abnormal probability of the i-th window of the previous k days; calculating the difference between the current abnormal probability and the historical abnormal probability of the i-th window of the previous k days, and summing the k differences to obtain the failure probability of the primary and secondary fused ring network box corresponding to the i-th window.

[0052] Specifically, the calculation formula for the failure probability of the primary and secondary integrated ring network box is:

[0053] in, represents the failure probability of the primary and secondary integrated ring network box corresponding to the i-th window; 5 represents the size of the selected historical cycle number. In this embodiment, the historical abnormal probability of 5 days is selected; represents the historical anomaly probability of the i-th window in the previous k days.

[0054] If the failure probability of the primary and secondary fusion ring network box corresponding to the i-th window is greater than the preset failure probability threshold, it means that the primary and secondary fusion ring network box fails.

[0055] In the technical solution of the present application, current curves of N windows and corrected voltage curves of N windows are obtained; wherein, the corrected voltage curve is generated according to the corrected voltage value, which is obtained by performing electromagnetic interference correction on multiple voltage values ​​of a primary and secondary integrated ring network box, and the multiple voltage values ​​are obtained by setting shielding layers with different coverage areas on multiple voltage sensors and collecting voltages of the primary and secondary integrated ring network box; according to each corrected voltage value in the corrected voltage curve of the N windows, and multiple voltage values ​​corresponding to the corrected voltage value, the voltage correction error of the N windows is determined, and a voltage correction error curve of the N windows is generated; the voltage correction error curve of the i-th window is corrected by using the fluctuation trend of the current curve of the i-th window to obtain the target voltage correction error curve of the i-th window; according to the current curve of the i-th window and the normalized path distance of the target voltage correction error curve of the i-th window, the current abnormal probability of the current curve of the i-th window is determined; when the current abnormal probability of the current curve of the i-th window meets the preset fault condition, it is determined that there is a fault in the primary and secondary integrated ring network box. It can be seen that when the fluctuation of the voltage correction error curve is similar to the fluctuation of the current curve, there may be a fault. Therefore, the current abnormal probability of the current curve of the i-th window is determined by the normalized path distance of the current curve of the i-th window and the target voltage correction error curve of the i-th window, and then whether the primary and secondary integrated ring network box has a fault is further determined by whether the current abnormal probability meets the preset fault condition, so as to improve the accuracy of fault diagnosis.

[0056] In one embodiment, when the current abnormal probability of the current curve in the i-th window meets the preset fault condition, after determining that the primary and secondary fusion ring network box has a fault, the method further includes: Obtain relevant curves of the primary and secondary fusion ring network box of the i-th window, wherein the relevant curves include current curve, voltage curve, temperature curve and humidity curve; The relevant curves are input into the preset anomaly detection model to predict the fault type, and the fault type of the primary and secondary integrated ring network box is output.

[0057] Furthermore, the fault type and related curves of the primary and secondary integrated ring network box are sent to the terminal equipment.

[0058] Specifically, when receiving the instruction that the primary and secondary integrated ring network box has a fault, the current curve, voltage curve, temperature curve and humidity curve are classified through the abnormal detection model, and the fault type is output. The final fault type and related curve data are sent to the maintenance control personnel terminal.

[0059] Among them, the basic architecture of the anomaly detection model is the CNN architecture, the core is the fault type classification, the data input is various measurement curves of the day, the data format is [voltage curve, current curve, temperature curve, humidity curve], and the output is the fault type; the training set is the manually labeled fault types.

[0060] It can be seen that in this embodiment, the sensor module is set to obtain measurement data, and the influence of electromagnetic interference is removed according to the voltage signal with different shielding effects. The voltage and current signals are compared with the historical signals to determine the suspected fault signal, and then the possibility of the final fault is determined by analyzing the matching of the voltage and current of the suspected fault signal; finally, the fault type is used to identify the network output fault type. In this way, the influence of electromagnetic interference caused by the primary equipment of the ring network box on the secondary equipment fault detection can be reduced, and the accuracy of fault diagnosis can be improved, which has certain practical value.

[0061] See also Figure 4 , which shows a schematic diagram of a primary and secondary integrated ring network box fault diagnosis system provided by an embodiment of the present invention. In an exemplary embodiment, a primary and secondary integrated ring network box fault diagnosis system is provided, comprising: The acquisition module 410 is used to acquire the current curves of N windows and the corrected voltage curves of N windows; wherein the corrected voltage curve is generated according to the corrected voltage value, which is obtained by performing electromagnetic interference correction on multiple voltage values ​​of the primary and secondary fused ring network box, and the multiple voltage values ​​are obtained by setting shielding layers with different coverage areas on multiple voltage sensors and collecting voltages of the primary and secondary fused ring network box; A generating module 420, for determining voltage correction errors of the N windows according to each correction voltage value in the correction voltage curve of the N windows and a plurality of voltage values ​​corresponding to the correction voltage value, and generating a voltage correction error curve of the N windows; A correction module 430, used to correct the voltage correction error curve of the i-th window by using the fluctuation trend of the current curve of the i-th window to obtain a target voltage correction error curve of the i-th window; A calculation module 440, for determining a current abnormal probability of the current curve of the i-th window according to the normalization path distance of the current curve of the i-th window and the target voltage correction error curve of the i-th window; The detection module 450 is used to determine that there is a fault in the primary and secondary integrated ring network box when the current abnormal probability of the current curve in the i-th window meets the preset fault condition; wherein N and i are positive integers.

[0062] In other embodiments, a computer program product is also provided. When the computer program product runs on a computer, the computer executes the above-mentioned related steps to implement a primary-secondary integrated ring network box fault diagnosis method and system provided in the above embodiments.

[0063] In other embodiments, a computer-readable storage medium is also provided, in which a computer program code is stored. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to implement a primary and secondary fused ring network box fault diagnosis method and system provided in the above-mentioned embodiment.

[0064] Among them, the provided systems, devices, computer program products, and computer-readable storage media are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above and will not be repeated here.

[0065] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0066] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for diagnosing faults of a primary and secondary fusion ring network box, characterized in that: The method comprises: Obtain current curves of N windows and corrected voltage curves of N windows; wherein the corrected voltage curve is generated according to the corrected voltage value, which is obtained by performing electromagnetic interference correction on multiple voltage values ​​of the primary and secondary integrated ring network box; the multiple voltage values ​​are obtained by setting shielding layers with different coverage areas on multiple voltage sensors and collecting voltages of the primary and secondary integrated ring network box; Determine voltage correction errors of the N windows according to each correction voltage value in the correction voltage curve of the N windows and multiple voltage values ​​corresponding to the correction voltage value, and generate voltage correction error curves of the N windows; Using the fluctuation trend of the current curve of the i-th window, the voltage correction error curve of the i-th window is corrected to obtain the target voltage correction error curve of the i-th window; Determine the current abnormal probability of the current curve of the i-th window according to the normalized path distance of the current curve of the i-th window and the target voltage correction error curve of the i-th window; When the current abnormal probability of the current curve in the i-th window meets the preset fault condition, it is determined that the primary and secondary integrated ring network box has a fault; wherein N and i are positive integers.

2. The method for diagnosing faults of primary and secondary fusion ring network boxes according to claim 1 is characterized in that: Determining voltage correction errors of the N windows according to each correction voltage value in the correction voltage curves of the N windows and multiple voltage values ​​corresponding to the correction voltage value includes: For the j-th corrected voltage value, calculating an average value of multiple voltage values ​​corresponding to the j-th corrected voltage value; The difference between the average value and the j-th corrected voltage value is calculated to obtain the voltage correction error corresponding to the j-th corrected voltage value; wherein j is a positive integer.

3. The method for diagnosing faults of primary and secondary fusion ring network boxes according to claim 1 is characterized in that: Methods for determining the correction voltage value include: Determining a relative interference change rate and an electromagnetic interference resistance strength of a shielding layer of a plurality of voltage sensors according to a plurality of voltage values; Determine a plurality of predicted voltage values ​​that are not affected by electromagnetic interference according to the relative interference change rate, the electromagnetic interference resistance strength of the shielding layers of the plurality of voltage sensors, and each voltage value; The average value of multiple predicted voltage values ​​is calculated to obtain a corrected voltage value.

4. The method for diagnosing faults of a primary and secondary fusion ring network box according to claim 3 is characterized in that: Determining a relative interference change rate and an electromagnetic interference resistance strength of a shielding layer of a plurality of voltage sensors according to a plurality of voltage values ​​includes: Calculate the voltage difference between each two voltage values ​​respectively; According to the ratio of the voltage difference and the difference of the corresponding anti-electromagnetic interference strength, an overdetermined set of equations is formed; The overdetermined equations are calculated using the least square method to obtain the relative interference change rate and the anti-electromagnetic interference strength of the shielding layers of multiple voltage sensors.

5. The method for diagnosing faults of primary and secondary fusion ring network boxes according to claim 1 is characterized in that: The voltage correction error curve of the i-th window is corrected by using the fluctuation trend of the current curve of the i-th window to obtain the target voltage correction error curve of the i-th window, including: Performing an exponential operation on the difference between the variance of the current curve of the i-th window and the variance of the current curve of m windows before the i-th window, to obtain a current fluctuation coefficient; Calculate the product of the current fluctuation coefficient and the window length of the voltage correction error curve of the i-th window to obtain the correction length; The correction length and the window length of the voltage correction error curve of the i-th window are summed to obtain the target voltage correction error curve of the i-th window.

6. The method for diagnosing faults of a primary and secondary fusion ring network box according to claim 1 is characterized in that: When the current abnormal probability of the current curve in the i-th window meets the preset fault condition, it is determined that the primary and secondary integrated ring network box has a fault, including: According to the current abnormal probability and the historical abnormal probability of the current curve of the i-th window, the failure probability of the primary and secondary integrated ring network box corresponding to the i-th window is determined; When the fault probability is greater than a preset fault probability threshold, it is determined that the primary and secondary fused ring network box has a fault.

7. The method for diagnosing faults of a primary and secondary fusion ring network box according to claim 1 is characterized in that: According to the current abnormal probability and historical abnormal probability of the current curve of the i-th window, the failure probability of the primary and secondary integrated ring network box corresponding to the i-th window is determined, including: Get the historical abnormal probability of the i-th window in the previous k days; The difference between the current abnormal probability and the historical abnormal probability of the i-th window of the previous k days is calculated respectively, and the k differences are summed to obtain the failure probability of the primary and secondary fusion ring network box corresponding to the i-th window; where k is a positive integer.

8. The method for diagnosing faults of a primary and secondary fused ring network box according to any one of claims 1 to 7, characterized in that: When the current abnormal probability of the current curve of the i-th window meets the preset fault condition, after determining that the primary and secondary integrated ring network box has a fault, the method further includes: Obtain relevant curves of the primary and secondary fusion ring network box of the i-th window, wherein the relevant curves include current curve, voltage curve, temperature curve and humidity curve; The relevant curves are input into the preset anomaly detection model to predict the fault type, and the fault type of the primary and secondary integrated ring network box is output.

9. The method for diagnosing faults of a primary and secondary fused ring network box according to claim 8 is characterized in that: The method further comprises: The fault type and related curves of the primary and secondary integrated ring network box are sent to the terminal equipment.

10. A primary and secondary integrated ring network box fault diagnosis system, characterized in that: The system comprises: An acquisition module is used to acquire current curves of N windows and corrected voltage curves of N windows; wherein the corrected voltage curve is generated according to a corrected voltage value, which is obtained by performing electromagnetic interference correction on multiple voltage values ​​of a primary and secondary integrated ring network box, and the multiple voltage values ​​are obtained by setting shielding layers with different coverage areas on multiple voltage sensors and collecting voltages of the primary and secondary integrated ring network box; A generating module, used for determining voltage correction errors of the N windows according to each correction voltage value in the correction voltage curve of the N windows and a plurality of voltage values ​​corresponding to the correction voltage value, and generating voltage correction error curves of the N windows; A correction module, used to correct the voltage correction error curve of the i-th window by using the fluctuation trend of the current curve of the i-th window, so as to obtain the target voltage correction error curve of the i-th window; A calculation module, used to determine the current abnormal probability of the current curve of the i-th window according to the current curve of the i-th window and the normalized path distance of the target voltage correction error curve of the i-th window; The detection module is used to determine that there is a fault in the primary and secondary integrated ring network box when the current abnormal probability of the current curve in the i-th window meets the preset fault condition; wherein N and i are positive integers.

Images