DC power supply fault monitoring method and system based on edge calculation, and medium

Abstract

The invention relates to the technical field of power supply monitoring, in particular to a DC power supply fault monitoring method and system based on edge calculation and a medium, and the method comprises the steps: obtaining an abnormal time period of each electrical module in a DC power supply system and abnormal data corresponding to the abnormal time period; correcting the second abnormal time periods of the other electrical modules to obtain alignment time periods, and under the condition that the second abnormal time periods of the other electrical modules coincide with the corresponding alignment time periods, taking the other electrical modules as alternative nodes influencing the abnormality of the DC power supply module; determining a first possibility that the abnormal data of the DC power supply module is a real fault; under the condition that the first possibility is not less than a first threshold value, determining a second possibility that the abnormal data of the alternative node is a real fault; determining the abnormal significance of the DC power supply module; and re-determining whether the direct-current power supply module fails or not. Therefore, the accuracy of fault judgment is improved.

Description

Technical Field

[0001] The present invention relates to the field of power supply monitoring technology, and in particular to a direct current power supply fault monitoring method, system and medium based on edge computing. Background Art

[0002] The stable operation of data centers is highly dependent on continuous and reliable power supply. As one of the key infrastructures of data centers, the performance and reliability of the DC power system directly affect the overall operating efficiency and data security of the data center. Therefore, real-time monitoring of the operating status of the DC power system and timely detection and handling of potential power failures are important guarantees for ensuring the stable operation of the data center.

[0003] In some scenarios, the traditional DC power supply system monitoring method usually monitors each electrical module of the DC power supply, such as the rectifier module, inverter module, battery module, and DC power supply module, separately. When an electrical module is abnormal, the system will determine that the DC power supply system has a high risk of failure. However, this monitoring method has obvious limitations: the lack of a strongly associated intelligent management unit between the electrical modules leads to isolated operation between the electrical modules, and the inability to achieve effective integration and collaborative analysis of information. This isolated operation mode is prone to misjudgment of the status of the DC power supply module, such as mistaking a potential fault for a normal operating state, or misjudging a normal fluctuation as a fault risk. In this way, the traditional DC power supply system monitoring method has obvious deficiencies in the correlation between electrical modules and the accuracy of fault judgment, making it difficult to ensure the stable operation of the data center power system. Summary of the invention

[0004] In order to solve the technical problem that the traditional DC power supply system monitoring method has obvious deficiencies in the correlation between electrical modules and the accuracy of fault judgment, the purpose of the present invention is to provide a DC power supply fault monitoring method, system and medium based on edge computing. The technical solutions adopted are as follows: In a first aspect, an embodiment of the present invention provides a DC power supply fault monitoring method based on edge computing, comprising: obtaining an abnormal time period of each electrical module in a DC power supply system and abnormal data corresponding to the abnormal time period; correcting the second abnormal time period of the remaining electrical modules according to the first abnormal time period of the DC power supply module in the electrical module to obtain an alignment time period, and when the second abnormal time period of the remaining electrical modules and the corresponding alignment time period overlap, using the remaining electrical modules as candidate nodes affecting the DC power supply module abnormality; based on the second abnormal time period of the remaining electrical modules and the corresponding alignment time period, determining a first possibility that the abnormal data of the DC power supply module is a real fault; when the first possibility is not less than a first threshold, determining a second possibility that the abnormal data of the candidate node is a real fault; determining the abnormal significance of the DC power supply module according to the first possibility, the order of the candidate nodes in the DC power supply system and the corresponding second possibility; and re-determining whether the DC power supply module is faulty according to the abnormal significance and the first threshold.

[0005] Optionally, based on the second abnormal time period and the corresponding alignment time period of the remaining electrical modules, determining the first possibility that the abnormal data of the DC power supply module is a real fault includes: when the second abnormal time period and the corresponding alignment time period of the remaining electrical modules overlap, taking the remaining electrical modules as alternative nodes affecting the abnormality of the DC power supply module and marking them as the first numerical value; when the second abnormal time period and the corresponding alignment time period of the remaining electrical modules do not overlap, indicating that the remaining electrical modules are irrelevant to the abnormality of the DC power supply module and marked as the second numerical value, calculating the intersection between the second abnormal time period and the corresponding alignment time period, and the union between the second abnormal time period and the corresponding alignment time period; determining the first ratio between the intersection and the union as the overlap of the abnormal time of the remaining electrical modules; calculating the first product between the first numerical value or the second numerical value and the overlap, and normalizing the accumulated value of the first product using a positive correlation normalization function to obtain a normalized numerical value; determining the first difference between the predetermined numerical value and the normalized numerical value as the first possibility.

[0006] Optionally, determining the abnormal significance of the DC power module according to the first possibility, the order of the candidate nodes in the DC power system and the corresponding second possibility includes: obtaining a possibility sequence of each candidate node according to the order of the candidate nodes in the DC power system and the second possibility; selecting any candidate node from the possibility sequence as the node to be analyzed, taking the candidate node before the node to be analyzed as the first electrical module set upstream of the node to be analyzed, and taking the candidate node after the node to be analyzed as the second electrical module set downstream of the node to be analyzed; and determining the abnormal significance of the DC power module according to the second possibility corresponding to the node to be analyzed, the second possibility of the candidate node adjacent to the node to be analyzed, and the probability of the abnormal significance of the candidate node in the first electrical module set. The first average possibility of the fault of the node to be analyzed after correction is determined based on the first average possibility and the second average possibility in the second electrical module set; the maximum value in the third possibilities is selected as the reference possibility, and the electrical modules adjacent to the DC power module are selected as the third electrical module set of the DC power module; if the alternative node corresponding to the reference possibility belongs to the third electrical module set, the abnormal significance of the DC power module is determined according to the first possibility and the third average possibility in the third electrical module set; if the alternative node corresponding to the reference possibility does not belong to the third electrical module set, the abnormal significance of the DC power module is determined according to the first possibility, the reference possibility and the third average possibility.

[0007] Optionally, determining the third possibility of the fault of the node to be analyzed after correction based on the second possibility corresponding to the node to be analyzed, the second possibility of the alternative node adjacent to the node to be analyzed, the first average possibility in the first electrical module set, and the second average possibility in the second electrical module set includes: calculating the second difference between the second average possibility and the first average possibility, and the third difference between the second possibility of the latter alternative node adjacent to the node to be analyzed and the second possibility of the former alternative node adjacent to the node to be analyzed; calculating the absolute value of the first sum between the second difference and the third difference; and determining the second product between the absolute value of the first sum and the second possibility corresponding to the node to be analyzed as the third possibility of the fault of the node to be analyzed after correction.

[0008] Optionally, determining the abnormal significance of the DC power module based on the first possibility and the third average possibility in the third electrical module set includes: determining a second sum between the first possibility and the third average possibility in the third electrical module set as the abnormal significance of the DC power module.

[0009] Optionally, determining the abnormal significance of the DC power module based on the first possibility, the reference possibility and the third average possibility includes: calculating the absolute value of the reciprocal of the fourth difference between the reference possibility and the third average possibility, normalizing the absolute value of the reciprocal of the fourth difference using a positive correlation normalization function to obtain a normalized absolute value; calculating the third product between the normalized absolute value and the third average possibility; and determining the third sum between the third product and the first possibility as the abnormal significance of the DC power module.

[0010] Optionally, re-determining whether the DC power supply module is faulty based on the abnormality significance and the first threshold includes: normalizing the abnormality significance to obtain a normalized abnormality significance; when the normalized abnormality significance is not less than the first threshold, determining that the DC power supply module is faulty; when the normalized abnormality significance is less than the first threshold, determining that the DC power supply module is not faulty.

[0011] Optionally, normalizing the anomaly significance includes: determining a reference possibility of the DC power supply module; calculating a fourth sum between the reference possibility and the first possibility; and determining a second ratio between the anomaly significance and the fourth sum as the normalized anomaly significance.

[0012] In a second aspect, an embodiment of the present invention provides a DC power supply fault monitoring system based on edge computing, comprising: a processor and a memory; wherein the memory is used to store a computer program that can be run on the processor; and the processor is used to execute the program stored in the memory to implement the steps of the DC power supply fault monitoring method based on edge computing as mentioned in the first aspect.

[0013] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the DC power supply fault monitoring method based on edge computing as mentioned in the first aspect are implemented.

[0014] The present invention has the following beneficial effects: firstly, the abnormal time period of each electrical module in a DC power supply system and the abnormal data corresponding to the abnormal time period are obtained; then, the second abnormal time period of the remaining electrical modules is corrected according to the first abnormal time period of the DC power supply module in the electrical module to obtain the alignment time period, and when the second abnormal time period of the remaining electrical modules and the corresponding alignment time period overlap, the remaining electrical modules are used as candidate nodes affecting the abnormality of the DC power supply module; secondly, based on the second abnormal time period of the remaining electrical modules and the corresponding alignment time period, the first possibility that the abnormal data of the DC power supply module is a real fault is determined; when the first possibility is not less than a first threshold, the second possibility that the abnormal data of the candidate node is a real fault is determined; and the abnormal significance of the DC power supply module is determined according to the first possibility, the order of the candidate nodes in the DC power supply system and the corresponding second possibility; finally, whether the DC power supply module is faulty is re-determined according to the abnormal significance and the first threshold.

[0015] In this way, the embodiment of the present invention can obtain the possibility of each electrical module being a real fault according to the abnormal time period and abnormal data of each electrical module of the DC power supply system, and determine the abnormal significance of the DC power supply module in combination with the possibility of a real fault under the mutual influence between each electrical module. Thus, the fault of the DC power supply module is monitored and diagnosed multiple times through the mutual influence between each electronic module, the correlation between the electrical modules is strengthened, the coordinated management between the electrical modules is ensured, the accuracy of fault judgment is improved, and the stable operation of the central power system is ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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.

[0017] Figure 1 A flow chart of a DC power supply fault monitoring method based on edge computing is provided for one embodiment of the present invention.

[0018] Figure 2 A schematic diagram of the relationship between an edge server and an edge node provided for one embodiment of the present invention.

[0019] Figure 3 A schematic diagram of the structure of a DC power supply fault monitoring system based on edge computing is provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0020] 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 DC power supply fault monitoring method, system and medium based on edge computing proposed by the present invention, its specific implementation method, structure, characteristics 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.

[0021] 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.

[0022] The following is a detailed description of a DC power supply fault monitoring method based on edge computing provided by the present invention in conjunction with the accompanying drawings. Example

[0023] See also Figure 1 , which shows a flow chart of a DC power supply fault monitoring method based on edge computing provided by an embodiment of the present invention, including: S101, obtaining an abnormal time period of each electrical module in a direct current power supply system and abnormal data corresponding to the abnormal time period.

[0024] Specifically, the DC power supply system of the data center contains electrical modules such as medium-voltage power distribution, transformers, low-voltage power distribution, uninterruptible power supply, and DC power supply. The embodiments of the present invention integrate edge computing modules into these electrical modules as edge nodes. Corresponding sensors are installed in each electrical module to transmit the collected electrical data to the corresponding edge node. Since different electrical modules in the DC power supply system react differently to different data anomalies, different sensors are installed to monitor each electrical module. The above-mentioned edge computing module has certain data processing capabilities, storage capabilities, and communication interfaces, such as edge computing boxes.

[0025] Furthermore, for various types of electrical module sensor installations, specifically: For medium-voltage power distribution electrical modules, high-precision voltage sensors and current sensors are installed on the medium-voltage incoming and outgoing lines to monitor the operating current and operating voltage. For transformer electrical modules, temperature sensors are installed on the windings and core of the transformer to monitor its operating temperature. For low-voltage power distribution electrical modules, residual current transformers are installed in the low-voltage outgoing line circuit to monitor the residual current in the line. For uninterruptible power supply electrical modules, voltage sensors, current sensors, etc. are installed at both ends of the battery to monitor the battery’s terminal voltage, charge and discharge current, internal resistance and other parameters. For DC power supply electrical modules, current sensors, voltage sensors, etc. are installed at the DC power supply to monitor the current and voltage at the input end, the current and voltage at the output end, and the ripple voltage of the DC power supply.

[0026] Furthermore, after the data collected by the sensors of each electrical module is transmitted to the corresponding edge node, the edge node obtains the data collected by the sensors of the electrical module and inputs the data into a machine learning module, such as a support vector machine learning model (SVM), and outputs the abnormal data of the electrical module at that moment. In the machine learning model, a threshold is set to compare the data of each electrical module with the threshold and divide it into normal data and abnormal data. If the data of the edge node is normal, the relevant data will not be stored or uploaded. If the result is abnormal, the abnormal data at that moment will be marked, and the duration of the abnormal data will be obtained, and the abnormal time period corresponding to the duration will be uploaded to the edge server. The edge computing module of each electrical module analyzes the collected data, transmits the abnormal data to the edge server, and obtains the abnormal time period corresponding to each electrical module and the abnormal data within the abnormal time period. For example, Figure 2 As shown, Figure 2 A schematic diagram of the relationship between an edge server and an edge node is provided for an embodiment of the present invention, wherein the edge computing module in each edge node uploads the data collected by the sensor in the corresponding electrical module to the edge server, and the edge server processes the abnormal data of the abnormal time period of each electrical module.

[0027] S102, according to the first abnormal time period of the DC power supply module in the electrical module, the second abnormal time period of the remaining electrical modules is corrected to obtain the alignment time period. When the second abnormal time period of the remaining electrical modules overlaps with the corresponding alignment time period, the remaining electrical modules are used as alternative nodes affecting the abnormality of the DC power supply module.

[0028] Specifically, each electronic module has a sequence in the DC power supply system, which is to realize the power supply demand of the load through the order of medium voltage distribution, transformer, low voltage distribution, uninterruptible power supply, and DC power supply. Therefore, if the DC power supply module fails, there will be a certain delay error in time. Therefore, the embodiment of the present invention sets the allowable delay error of adjacent edge nodes to tms, and the value of t in the embodiment of the present invention is 5ms.

[0029] Furthermore, the embodiment of the present invention starts from the edge node of the DC power module and obtains the upstream adjacent edge node corresponding to each edge node according to the order of the power supply line. For example, the upstream adjacent edge node of the DC power module is an uninterruptible power supply, followed by a low-voltage power distribution, etc. It is worth noting that the actual adjacent upstream node is determined by the composition of the DC power system, and the embodiment of the present invention does not limit this.

[0030] Furthermore, the embodiment of the present invention first obtains the first abnormal time period in which the DC power module is abnormal. T 1, T 2], then the corrected alignment time period of the upstream edge node adjacent to the DC power module is [ T 1- t , T 2- t ]. Then, the upstream edge nodes are traversed from the corrected alignment time period to correct the time of each edge node and obtain the alignment time period of each edge node.

[0031] Further, if the remaining electrical modules of the DC power supply system fail, the abnormality of the DC power supply module is due to the failure of the remaining electrical modules, which affects the input of the downstream electrical modules. In this case, the remaining electrical modules except the DC power supply module also have abnormal performance within the allowable delay error time, so that they are synchronized in time. Therefore, the time when the adjacent edge nodes are abnormal is similar. Therefore, the failure of the DC power supply module in the actual DC power supply system is not isolated. Due to the failure of the upstream electrical module, the downstream electrical module is abnormal. Therefore, the abnormal time period that occurs in the remaining electrical modules shows a strong consistency with the alignment time period of the module. Based on this, the embodiment of the present invention selects an edge node of an electrical module that is not a DC power supply module, obtains the abnormal data uploaded by the edge node, and if the abnormal time period uploaded by the edge node overlaps with the alignment time period of the edge node, the edge node behaves as an alternative node that affects the abnormality of the DC power supply module, and is marked as the first value in the embodiment of the present invention. If the abnormal time period uploaded by the edge node does not overlap with the alignment time period of the edge node, it means that the abnormal fault of the DC power supply module is not related to the edge node, and is marked as the second value. In the embodiment of the present invention, the first value or the second value is represented by L i The first value and the second value can be determined according to actual conditions. In the embodiment of the present invention, the first value is 1 and the second value is 0.

[0032] S103: Determine a first possibility that the abnormal data of the DC power supply module is a real fault based on the second abnormal time periods and the corresponding alignment time periods of the remaining electrical modules.

[0033] Specifically, if the abnormal time period of the electrical modules of all non-DC power modules fail and their corresponding alignment time period overlap, the abnormal time period of the remaining edge nodes has a high consistency with the abnormal time period of the edge node where the DC power module is located. This means that the possibility that the abnormality of the edge nodes of the remaining electrical modules is affected by the failure of the DC power module is smaller, and the possibility that the fault in its upstream causes the abnormality in the monitoring of the DC power module is greater.

[0034] Further, as an optional embodiment of the present invention, based on the second abnormal time period and the corresponding alignment time period of the remaining electrical modules, determining the first possibility that the abnormal data of the DC power supply module is a real fault includes: when the second abnormal time period and the corresponding alignment time period of the remaining electrical modules overlap, the remaining electrical modules are used as alternative nodes affecting the abnormality of the DC power supply module and marked as a first numerical value; when the second abnormal time period and the corresponding alignment time period of the remaining electrical modules do not overlap, it indicates that the remaining electrical modules are irrelevant to the abnormality of the DC power supply module and are marked as a second numerical value; the intersection between the second abnormal time period and the corresponding alignment time period and the union between the second abnormal time period and the corresponding alignment time period are calculated; the first ratio between the intersection and the union is determined to be the overlap of the abnormal time of the remaining electrical modules; the first product between the first numerical value or the second numerical value and the overlap is calculated, and the accumulated value of the first product is normalized using a positive correlation normalization function to obtain a normalized numerical value; and the first difference between the predetermined numerical value and the normalized numerical value is determined to be the first possibility.

[0035] Specifically, the embodiment of the present invention specifically uses the following formula to calculate the overlap of the abnormal time of the remaining electrical modules:

[0036] In the above formula, Indicates The coincidence of abnormal time of edge nodes of the remaining electrical modules reflects the consistency in time. Indicates the first non-DC power module in the DC power system. i The second abnormal time period of the remaining electrical modules. Indicates the first non-DC power module in the DC power system. i The alignment time periods of the remaining electrical modules. ∩ indicates intersection and ∩ indicates union.

[0037] Further, in the embodiment of the present invention, the predetermined value may be 1, and the embodiment of the present invention specifically uses the following formula to calculate the first possibility:

[0038] In the above formula,p The first possibility that the abnormal data uploaded by the edge node of the DC power module is a real fault. Represents the positive correlation normalization function, which is used to Perform normalization. Indicates The value of whether the edge node of the remaining electrical modules is a candidate node can be a first value or a second value. Indicates The coincidence of abnormal time of edge nodes of the remaining electrical modules reflects the consistency in time. n Indicates the number of remaining electrical modules other than the DC power supply module.

[0039] S104: When the first possibility is not less than the first threshold, determine a second possibility that the abnormal data of the candidate node is a real fault.

[0040] Specifically, the first threshold value can be taken according to the actual situation, and the embodiment of the present invention is not limited here. If the first possibility is less than the first threshold, it means that the heating alignment time period of each electrical module in the DC power supply system and the actual abnormal time period have a high overlap, which means that the failure of the monitored DC power supply module is caused by the adjustment of the normal power supply strategy of the DC power supply system, such as changes in normal power supply demand during peak hours, and is not a real failure. If the first possibility is not less than the first threshold, it means that the alignment effect is poor, and the overlap between the alignment time period and the abnormal time period is low. However, at this time, it may appear that an electrical module in the DC power supply system has a fault, resulting in basically normal power supply in the upstream of the electrical module, and a large degree of abnormality in the downstream of the electrical module. At this time, although the DC power supply module also has uploaded abnormal data, it actually manifests as a failure of the remaining electrical modules. Based on this, the embodiment of the present invention selects the above label as = the abnormal time period corresponding to the candidate node of the electrical module with the first value, the candidate node is taken as the node to be analyzed, and the same analysis is performed according to the first possibility method of calculating the abnormal data as a real fault recorded in the above embodiment of the present invention, and the second possibility that the abnormal data uploaded by the candidate node indicates that the electrical module is a real fault is obtained. It is worth noting that when calculating the second possibility, the number of corresponding electrical modules n The value is 1.

[0041] Furthermore, in the embodiment of the present invention, the electrical module whose edge node does not upload abnormal data is regarded as working normally. At this time, the possibility of the corresponding real fault is 0, thereby obtaining the possibility of the real fault of each electrical module in the DC power supply system during the abnormal time period when the DC power supply module has an abnormality.

[0042] S105: Determine the abnormal significance of the DC power module according to the first possibility, the order of the candidate nodes in the DC power system, and the corresponding second possibility.

[0043] Specifically, after obtaining the first possibility of the DC power module and the second possibilities of the remaining electrical modules, the embodiment of the present invention determines the final abnormal significance of the DC power module according to the mutual influence between the electrical modules.

[0044] Further, as an optional embodiment of the present invention, determining the abnormal significance of the DC power module according to the first possibility, the order of the candidate nodes in the DC power system and the corresponding second possibility includes: obtaining a possibility sequence of each candidate node according to the order of the candidate nodes in the DC power system and the second possibility; selecting any candidate node from the possibility sequence as the node to be analyzed, taking the candidate node before the node to be analyzed as the first electrical module set upstream of the node to be analyzed, and taking the candidate node after the node to be analyzed as the second electrical module set downstream of the node to be analyzed; according to the second possibility corresponding to the node to be analyzed, the second possibility of the candidate node adjacent to the node to be analyzed, the first electrical module set The first average possibility in the electrical module set and the second average possibility in the second electrical module set are used to determine the third possibility of the fault after correction of the node to be analyzed; the maximum value in the third possibilities is selected as the reference possibility, and the electrical modules adjacent to the DC power module are selected as the third electrical module set of the DC power module; if the alternative node corresponding to the reference possibility belongs to the third electrical module set, the abnormal significance of the DC power module is determined according to the first possibility and the third average possibility in the third electrical module set; if the alternative node corresponding to the reference possibility does not belong to the third electrical module set, the abnormal significance of the DC power module is determined according to the first possibility, the reference possibility and the third average possibility.

[0045] Specifically, in the embodiment of the present invention, according to the order of the electrical modules corresponding to the candidate nodes in the DC power supply system, a possibility sequence of the second possibility that the candidate node is a real fault is obtained. . Select the first The edge nodes are taken as the nodes to be analyzed, and the electrical modules of the DC power supply system are divided into two categories: The first electrical module set C1 upstream of the node to be analyzed is The second electrical module set C2 downstream of the node to be analyzed. At this time, if there is a significant change in the probability of failure before and after the node to be analyzed, and the change in the probability of failure of the edge node adjacent to the node to be analyzed is the most significant, it means that the electrical modules before the electrical module corresponding to the node to be analyzed are all in normal operation, and there are significant abnormalities in the subsequent electrical modules, which means that the subsequent electrical module abnormalities are affected by the abnormality of the node to be analyzed. If the fault of the electrical module monitoring is caused by the fault of the other electrical modules, the fault probability of the adjacent electrical modules is relatively large, and the difference is relatively small, so correction is achieved.

[0046] Further, as an optional embodiment of the present invention, determining the third possibility of the fault of the node to be analyzed after correction based on the second possibility corresponding to the node to be analyzed, the second possibility of the alternative node adjacent to the node to be analyzed, the first average possibility in the first electrical module set, and the second average possibility in the second electrical module set includes: calculating the second difference between the second average possibility and the first average possibility, and the third difference between the second possibility of the latter alternative node adjacent to the node to be analyzed and the second possibility of the former alternative node adjacent to the node to be analyzed; calculating the absolute value of the first sum between the second difference and the third difference; and determining the second product between the absolute value of the first sum and the second possibility corresponding to the node to be analyzed as the third possibility of the fault of the node to be analyzed after correction.

[0047] Specifically, the embodiment of the present invention specifically uses the following formula to calculate the third possibility:

[0048] In the above formula, Indicates The third possibility of the corrected failure of the node to be analyzed. Indicates The second possibility of the node to be analyzed. Indicates The second possibility of the adjacent candidate node before the node to be analyzed. Indicates The second possibility of an adjacent candidate node after the node to be analyzed. p (C1) represents the first average likelihood in the first electrical module set C1. p (C2) represents the second average likelihood in the second electrical module set C2.

[0049] It should be noted that if there is no previous or next electrical module during the analysis of the node to be analyzed, the above analysis is not performed. Thus, the possibility of corrected faults of the node to be analyzed at the middle position is obtained, that is, the node to be analyzed is not the first or the last one of the data sequence.

[0050] Furthermore, when a DC power module fails, especially a short circuit failure, the current will increase sharply, which will have a great impact on the switches, fuses, lines, etc. in the low-voltage distribution module. Or when the output voltage of the DC power module is abnormal, the DC power module may not be able to charge normally or switch to the battery power mode, thereby reducing the reliability of the entire DC power system. In addition, if the fault of the DC power system causes abnormal current, it may also cause damage to the inverter, battery and other components of the DC power system. Therefore, the DC power system also affects the electrical modules close to it. And as the number of electrical modules in the middle increases, the impact of the DC power module on the remaining electrical modules decreases. If an electrical module close to the DC power module fails, although the DC power module is normal at this time, there is a large potential fault, and a higher probability of failure should also be given at this time.

[0051] Further, in the embodiment of the present invention, the third possibility after correction is selected The maximum value in , as a possible reference. The electrical modules adjacent to the DC power module are selected as the third electrical module set M, such as the previous electrical module adjacent to the DC power module and the next electrical module adjacent to the DC power module.

[0052] Furthermore, if The corresponding electrical module belongs to the third electrical module set M, which means that the electrical module with the most significant abnormality has a greater abnormal transmission to the DC power module, which may cause a failure of the DC power module. Or because the failure of the DC power module has a greater interference with the electrical module, the DC power module should be given a higher abnormal significance. As an optional embodiment of the present invention, determining the abnormal significance of the DC power module based on the first possibility and the third average possibility in the third electrical module set includes: determining the second sum between the first possibility and the third average possibility in the third electrical module set as the abnormal significance of the DC power module.

[0053] Specifically, the embodiment of the present invention specifically uses the following formula to calculate the abnormal significance of the DC power supply module:

[0054] In the above formula, z The DC power module is abnormally significant. pThe first possibility that the abnormal data uploaded by the edge node of the DC power module is a real fault. represents the third average likelihood in the third set M of electrical modules.

[0055] Furthermore, if The corresponding electrical module does not belong to the third electrical module set M, indicating that the abnormal performance of the low-voltage power distribution and DC power supply modules is not significant in comparison. If there is a large difference, and the abnormality of the third electrical module set M is obviously smaller, it means that although there are abnormalities in the low-voltage power distribution and DC power supply modules, they are caused by the abnormalities of the remaining electrical modules, and the abnormalities of the remaining electrical modules have little impact on the low-voltage power distribution and DC power supply modules. At this time, it is considered that there is no fault. The abnormal significance of the corresponding DC power supply module is small. As an optional embodiment of the present invention, determining the abnormal significance of the DC power supply module according to the first possibility, the reference possibility and the third average possibility includes: calculating the absolute value of the reciprocal of the fourth difference between the reference possibility and the third average possibility, and using the positive correlation normalization function to normalize the absolute value of the reciprocal of the fourth difference to obtain a normalized absolute value; calculating the third product between the normalized absolute value and the third average possibility; determining the third sum between the third product and the first possibility as the abnormal significance of the DC power supply module.

[0056] Specifically, the embodiment of the present invention specifically uses the following formula to calculate the abnormal significance of the DC power supply module:

[0057] In the above formula, z The DC power module is abnormally significant. p The first possibility that the abnormal data uploaded by the edge node of the DC power module is a real fault. represents the third average likelihood in the third set M of electrical modules. Indicates the reference possibility. Represents the positive correlation normalization function, which is used to Perform normalization.

[0058] S106: re-determine whether the DC power supply module is faulty according to the abnormality significance and the first threshold.

[0059] Specifically, after obtaining the abnormality significance, as an optional embodiment of the present invention, re-determining whether the DC power supply module is faulty according to the abnormality significance and the first threshold value includes: normalizing the abnormality significance to obtain the normalized abnormality significance; when the normalized abnormality significance is not less than the first threshold value, determining that the DC power supply module is faulty; when the normalized abnormality significance is less than the first threshold value, determining that the DC power supply module is not faulty.

[0060] Among them, when normalizing the abnormality significance, first determine the reference possibility of the DC power supply module; then calculate the fourth sum value between the reference possibility and the first possibility; finally determine the second ratio between the abnormality significance and the fourth sum value as the normalized abnormality significance.

[0061] Specifically, the embodiment of the present invention specifically uses the following formula to calculate the normalized abnormal significance: In the above formula, Represents the normalized outlier significance. z The DC power module is abnormally significant. p The first possibility that the abnormal data uploaded by the edge node of the DC power module is a real fault. Indicates the reference possibility.

[0062] After calculating the normalized abnormal significance, the edge server sends the normalized abnormal significance to the edge node where the DC power module is located. The edge node where the DC power module is located re-compares the normalized abnormal significance sent with the first threshold. If the normalized abnormal significance is not less than the first threshold, it is considered that the DC power module is faulty. If the normalized abnormal significance is less than the first threshold, there is no DC power failure.

[0063] The embodiment of the present invention can obtain the possibility of each electrical module being a real fault according to the abnormal time period and abnormal data of each electrical module of the DC power supply system, and determine the abnormal significance of the DC power supply module in combination with the possibility of a real fault under the mutual influence between each electrical module. Thus, the fault of the DC power supply module is monitored and diagnosed multiple times through the mutual influence between each electronic module, the correlation between the electrical modules is strengthened, the coordinated management between the electrical modules is ensured, the accuracy of fault judgment is improved, and the stable operation of the central power system is ensured.

[0064] Embodiment 2: Corresponding to the edge computing-based DC power supply fault monitoring method provided in the above embodiment, based on the same technical concept, an embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by the processor, the steps of the edge computing-based DC power supply fault monitoring method mentioned in the above embodiment are implemented.

[0065] It should be noted that the computer-readable storage medium provided in an embodiment of the present invention and the DC power supply fault monitoring method based on edge computing provided in an embodiment of the present invention are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned DC power supply fault monitoring method based on edge computing, and has the same or similar beneficial effects, and the repeated parts will not be repeated.

[0066] Embodiment three: Corresponding to the DC power supply fault monitoring method based on edge computing provided in the above embodiment, based on the same technical concept, an embodiment of the present invention further provides a DC power supply fault monitoring system based on edge computing, and the DC power supply fault monitoring system based on edge computing is used to execute the above DC power supply fault monitoring method based on edge computing. Figure 3 A schematic diagram of a DC power supply fault monitoring system based on edge computing is provided in another embodiment of the present invention, as shown in FIG. Figure 3 The DC power supply fault monitoring system based on edge computing may have relatively large differences due to different configurations or performances, and may include one or more processors 301 and memory 302, the memory 302 is used to store computer programs that can be run on the processor 301, and the processor 301 is used to execute the program stored in the memory 302 to achieve the above Figure 1 The various steps in the method embodiment. Among them, the memory 302 can be a temporary storage or a permanent storage. The application stored in the memory 302 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the DC power supply fault monitoring system based on edge computing.

[0067] Furthermore, the processor 301 may be configured to communicate with the memory 302, and execute a series of computer executable instructions in the memory 302 on the DC power supply fault monitoring system based on edge computing. The DC power supply fault monitoring system based on edge computing may also include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input and output interfaces 305, and one or more keyboards 306.

[0068] Specifically in this embodiment, the DC power supply fault monitoring system based on edge computing includes a processor, a communication interface, a memory and a communication bus; wherein the processor, the communication interface and the memory communicate with each other through the bus; the memory is used to store computer programs; the processor is used to execute the programs stored in the memory to achieve the above Figure 1 The various steps in the method embodiment are similar to those in the method embodiment, and have the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of the present invention will not be described again here.

[0069] It should be noted that the DC power supply fault monitoring system based on edge computing provided in an embodiment of the present invention and the DC power supply fault monitoring method based on edge computing provided in an embodiment of the present invention are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned DC power supply fault monitoring method based on edge computing, and has the same or similar beneficial effects, and the repeated parts will not be repeated.

[0070] 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.

[0071] 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 DC power supply fault monitoring method based on edge computing, characterized in that: The DC power supply fault monitoring method based on edge computing includes: Acquire the abnormal time period of each electrical module in the DC power supply system and the abnormal data corresponding to the abnormal time period; According to the first abnormal time period of the DC power module in the electrical module, the second abnormal time period of the remaining electrical modules is corrected to obtain the alignment time period, and when the second abnormal time period of the remaining electrical modules overlaps with the corresponding alignment time period, the remaining electrical modules are used as candidate nodes affecting the abnormality of the DC power module; Based on the second abnormal time period and the corresponding alignment time period of the remaining electrical modules, determining a first possibility that the abnormal data of the DC power supply module is a real fault; When the first possibility is not less than the first threshold, determining a second possibility that the abnormal data of the candidate node is a real fault; Determining the abnormal significance of the DC power module according to the first possibility, the order of the candidate nodes in the DC power system and the corresponding second possibility; Re-determine whether the DC power supply module is faulty based on the abnormal significance and the first threshold.

2. The DC power supply fault monitoring method based on edge computing according to claim 1 is characterized in that: The first possibility of determining that the abnormal data of the DC power supply module is a real fault based on the second abnormal time period and the corresponding alignment time period of the remaining electrical modules includes: In the case where the second abnormal time period of the remaining electrical modules overlaps with the corresponding alignment time period, the remaining electrical modules are taken as candidate nodes affecting the abnormality of the DC power supply module and are marked as the first value. In the case where the second abnormal time period of the remaining electrical modules does not overlap with the corresponding alignment time period, it indicates that the remaining electrical modules are irrelevant to the abnormality of the DC power supply module and are marked as the second value. Calculating the intersection between the second abnormal time period and the corresponding alignment time period, and the union between the second abnormal time period and the corresponding alignment time period; Determining a first ratio between the intersection and the union as a coincidence of abnormal times of the remaining electrical modules; Calculating a first product between the first value or the second value and the coincidence condition, and normalizing the accumulated value of the first product using a positive correlation normalization function to obtain a normalized value; A first difference between a predetermined value and the normalized value is determined as the first possibility.

3. The DC power supply fault monitoring method based on edge computing according to claim 1 is characterized in that: The determining of the abnormal significance of the DC power module according to the first possibility, the order of the candidate nodes in the DC power system and the corresponding second possibility includes: Obtaining a possibility sequence of each candidate node according to the order of the candidate nodes in the DC power supply system and the second possibility; Select any candidate node from the possible sequence as the node to be analyzed, use the candidate node before the node to be analyzed as the first electrical module set upstream of the node to be analyzed, and use the candidate node after the node to be analyzed as the second electrical module set downstream of the node to be analyzed; Determine a third possibility of a corrected fault of the node to be analyzed based on the second possibility corresponding to the node to be analyzed, the second possibility of the candidate node adjacent to the node to be analyzed, the first average possibility in the first electrical module set, and the second average possibility in the second electrical module set; Selecting a maximum value among the third possibilities as a reference possibility, and selecting an electrical module adjacent to the DC power module as a third electrical module set of the DC power module; If the candidate node corresponding to the reference possibility belongs to the third electrical module set, determining the abnormal significance of the DC power supply module according to the first possibility and the third average possibility in the third electrical module set; If the candidate node corresponding to the reference possibility does not belong to the third electrical module set, the abnormal significance of the DC power supply module is determined according to the first possibility, the reference possibility and the third average possibility.

4. The DC power supply fault monitoring method based on edge computing according to claim 3 is characterized in that: Determining a third possibility of a corrected fault of the node to be analyzed according to the second possibility corresponding to the node to be analyzed, the second possibility of the candidate node adjacent to the node to be analyzed, the first average possibility in the first electrical module set, and the second average possibility in the second electrical module set includes: Calculating a second difference between the second average likelihood and the first average likelihood, and a third difference between a second likelihood of a subsequent candidate node adjacent to the node to be analyzed and a second likelihood of a previous candidate node adjacent to the node to be analyzed; calculating an absolute value of a first sum between the second difference and the third difference; A second product of the absolute value of the first sum and the second possibility corresponding to the node to be analyzed is determined as a third possibility of the fault of the node to be analyzed after correction.

5. The DC power supply fault monitoring method based on edge computing according to claim 3 is characterized in that: Determining the abnormal significance of the DC power supply module according to the first possibility and the third average possibility in the third set of electrical modules includes: A second sum value between the first possibility and a third average possibility in the third set of electrical modules is determined as the abnormal significance of the DC power supply module.

6. The DC power supply fault monitoring method based on edge computing according to claim 3 is characterized in that: The determining the abnormal significance of the DC power supply module according to the first possibility, the reference possibility and the third average possibility includes: Calculating an absolute value of a reciprocal of a fourth difference between the reference likelihood and the third average likelihood, and normalizing the absolute value of the reciprocal of the fourth difference using a positive correlation normalization function to obtain a normalized absolute value; calculating a third product between the normalized absolute value and the third average likelihood; A third sum between the third product and the first possibility is determined as the abnormal significance of the DC power supply module.

7. The DC power supply fault monitoring method based on edge computing according to any one of claims 1 to 6, characterized in that: The re-determining whether the DC power supply module is faulty according to the abnormal significance and the first threshold value comprises: Normalizing the abnormal significance to obtain normalized abnormal significance; When the normalized abnormal significance is not less than the first threshold, determining that the DC power supply module is faulty; When the normalized abnormal significance is less than the first threshold, it is determined that the DC power supply module is not faulty.

8. The DC power supply fault monitoring method based on edge computing according to claim 7 is characterized in that: The normalizing process of the abnormal significance includes: Determining a reference possibility of the DC power supply module; calculating a fourth sum value between the reference likelihood and the first likelihood; A second ratio between the anomaly significance and the fourth sum is determined as the normalized anomaly significance.

9. A DC power supply fault monitoring system based on edge computing, characterized in that: include: A processor and a memory; wherein the memory is used to store a computer program that can be run on the processor; A processor is used to execute a program stored in a memory to implement the steps of a DC power supply fault monitoring method based on edge computing as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the DC power supply fault monitoring method based on edge computing as described in any one of claims 1 to 8 are implemented.

Images