A power distribution network switch cabinet control method and system based on intelligent monitoring

Abstract

The application relates to the technical field of power systems, in particular to a power distribution network switch cabinet control method and system based on intelligent monitoring. Real-time environmental characteristic data is introduced into an index knowledge graph for pairing analysis and evaluation; if the power distribution control performance of a target switch cabinet at a preset time node is in a normal state, the target switch cabinet is not processed; if the power distribution control performance of the target switch cabinet at the preset time node is in a warning state, the index knowledge graph is used to trace back the warning working parameters of the target switch cabinet at the preset time node, and drift analysis is performed on the warning working parameters; if the drift analysis result of the warning working parameters is that the working parameters need to be regulated, the working parameters needing to be regulated are regulated and processed. The application realizes intelligent regulation and control of the switch cabinet, thereby ensuring the stability and reliability of the power distribution system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power systems, in particular to a switch cabinet control method and system for distribution networks based on intelligent monitoring. BACKGROUND

[0002] In the current field of power systems, the safe, stable and reliable operation of distribution networks is of great importance. As a key device in distribution networks, switch cabinets bear the important responsibility of control, protection and power distribution. However, with the increasing complexity of power systems and the growing load, the traditional switch cabinet control method gradually reveals its limitations. On the one hand, the traditional control method mainly relies on human experience and regular inspection, which is difficult to achieve real-time and comprehensive monitoring of the switch cabinet state, and cannot timely and accurately warn and handle potential faults and abnormal conditions. On the other hand, as the running time of switch cabinets increases, their internal components, insulation materials, etc. will age and deteriorate, and changes in operating environment and fluctuations in load will also affect the performance of switch cabinets, which may lead to various abnormalities and even failures in switch cabinets. The traditional control method is difficult to comprehensively analyze and effectively cope with these complex factors.

[0003] In addition, the existing technology often lacks sufficient mining and utilization of historical data when dealing with abnormal events of switch cabinets, and it is difficult to establish an effective correlation between abnormal events and device states, resulting in a lack of precision and foresight in judging and handling abnormal events. At the same time, there is a lack of in-depth research and application of the relationship between environmental characteristic data and device working parameters when abnormal events occur, which makes it impossible to fully utilize these data to improve the intelligent level of the control method. SUMMARY

[0004] The present application overcomes the low intelligent level of switch cabinet control in the prior art and provides a switch cabinet control method and system for distribution networks based on intelligent monitoring.

[0005] To achieve the above-mentioned purposes, the technical solution adopted by the present application is as follows:

[0006] The present application discloses a switch cabinet control method for distribution networks based on intelligent monitoring, comprising the following steps:

[0007] S102: Obtain various working environment abnormal events that have occurred to the target switch cabinet during operation, and perform correlation evaluation on the various working environment abnormal events that have occurred to the target switch cabinet during operation, to obtain the associated working parameters of the target switch cabinet when various working environment abnormal events occur;

[0008] S104: Obtain historical environment characteristic data of the target switch cabinet when various work environment abnormal events occur, and construct an index knowledge graph according to the historical environment characteristic data of the target switch cabinet when various work environment abnormal events occur and the associated work parameters;

[0009] S106: In the actual operation process of the target switch cabinet, real-time environment characteristic data of the target switch cabinet is collected at a preset time node, the real-time environment characteristic data is imported into the index knowledge graph for pairing analysis and evaluation, and the power distribution control performance state of the target switch cabinet at the preset time node is obtained through evaluation;

[0010] S108: If the power distribution control performance of the target switch cabinet at the preset time node is normal, the target switch cabinet is not processed, and the power distribution control performance continues to be analyzed and evaluated at the next preset time node;

[0011] S110: If the power distribution control performance of the target switch cabinet at the preset time node is a warning state, the warning work parameters of the target switch cabinet at the preset time node are traced back in the index knowledge graph, and drift analysis is performed on the warning work parameters to obtain a drift analysis result;

[0012] S112: If the drift analysis result of the warning work parameters is that the work parameters do not need to be adjusted, the warning work parameters are continuously monitored; if the drift analysis result of the warning work parameters is that the work parameters need to be adjusted, the work parameters that need to be adjusted are adjusted and processed.

[0013] More specifically, various work environment abnormal events that have occurred in the operation of the target switch cabinet are obtained, the correlation of the various work environment abnormal events that have occurred in the operation of the target switch cabinet is evaluated, the associated work parameters of the target switch cabinet when various work environment abnormal events occur are obtained, and specifically:

[0014] S202: Obtain the work record book of the target switch cabinet, and obtain various work environment abnormal events that have occurred in the operation of the target switch cabinet according to the work record book;

[0015] S204: Randomly extract a work environment abnormal event that has not been extracted, obtain various historical work parameters of the target switch cabinet when the work environment abnormal event occurs according to the work record book, calculate the parameter difference between each historical work parameter and the corresponding preset work parameter respectively, and obtain the offset amplitude of each historical work parameter of the target switch cabinet when the work environment abnormal event occurs;

[0016] S206: The historical work parameters corresponding to the offset amplitude greater than the preset threshold are marked as the associated work parameters of the target switch cabinet when the work environment abnormal event occurs;

[0017] S208: return to execute S204 to S206 to continue to perform correlation evaluation on the work environment abnormal events occurred to the target switch cabinet in operation until all work environment abnormal events are extracted to obtain the associated work parameters of the target switch cabinet when various work environment abnormal events occur.

[0018] More specifically, historical environment feature data of the target switch cabinet when various work environment abnormal events occur is acquired, and an index knowledge graph is constructed according to the historical environment feature data of the target switch cabinet when various work environment abnormal events occur and the associated work parameters, specifically:

[0019] feature extraction processing is performed on various work environment abnormal events to acquire historical environment feature data of the target switch cabinet when various work environment abnormal events occur;

[0020] and the associated work parameters of the target switch cabinet when various work environment abnormal events occur are acquired;

[0021] A knowledge graph is constructed, the various work environment abnormal events are initialized as child nodes, and a plurality of index nodes are divided in the knowledge graph according to the child nodes; a first storage space and a second storage space are respectively created in each index node;

[0022] The historical environment feature data of the target switch cabinet when various work environment abnormal events occur are respectively stored in the first storage space of the corresponding index node, and the associated work parameters of the target switch cabinet when various work environment abnormal events occur are respectively stored in the second storage space of the corresponding index node;

[0023] After the historical environment feature data and the associated work parameters of the target switch cabinet when various work environment abnormal events occur are both stored, an index knowledge graph is constructed.

[0024] More specifically, the real-time environment feature data is imported into the index knowledge graph for pairing analysis and evaluation, and the power distribution control performance state of the target switch cabinet at the preset time node is evaluated, specifically:

[0025] The real-time environment feature data is imported into the index knowledge graph, and the Pearson correlation coefficients between the real-time environment feature data and the historical environment feature data in each first storage space are calculated;

[0026] If the Pearson correlation coefficients between the real-time environment feature data and the historical environment feature data in each first storage space are all not greater than a preset coefficient value, the power distribution control performance of the target switch cabinet at the preset time node is marked as normal;

[0027] If the Pearson correlation coefficient between the real-time environment feature data and the historical environment feature data in one or more first storage spaces is greater than a preset coefficient value, the power distribution control performance of the target switch cabinet at the preset time node is marked as a warning state.

[0028] More specifically, if the power distribution control performance of the target switch cabinet at the preset time node is in the warning state, the warning working parameter of the target switch cabinet at the preset time node is backtracked in the index knowledge graph, and the following steps are further included:

[0029] Meanwhile, if the Pearson correlation coefficient between the real-time environment feature data and the historical environment feature data in one or more first storage spaces is greater than a preset coefficient value, the first storage space corresponding to the Pearson correlation coefficient greater than the preset coefficient value is marked.

[0030] The index node to which the marked first storage space belongs is backtracked and marked, the associated working parameter of the second storage space in the marked index node is extracted, and the extracted associated working parameter is designated as the warning working parameter of the target switch cabinet at the preset time node.

[0031] More specifically, the warning working parameter is subjected to drift analysis to obtain a drift analysis result, specifically:

[0032] An actual parameter value of the warning working parameter of the target switch cabinet at the preset time node is obtained, a difference between the actual parameter value and a preset value of the warning working parameter of the target switch cabinet at the preset time node is calculated, and a drift value of the warning working parameter of the target switch cabinet at the preset time node is obtained.

[0033] It is determined whether the drift value of the warning working parameter of the target switch cabinet at the preset time node is greater than a preset drift value.

[0034] If the drift value of the warning working parameter of the target switch cabinet at the preset time node is greater than the preset drift value, the warning working parameter is designated as a work parameter that needs to be regulated.

[0035] If the drift value of the warning working parameter of the target switch cabinet at the preset time node is not greater than the preset drift value, the warning working parameter is designated as a work parameter that does not need to be regulated, and the warning working parameter is continuously monitored.

[0036] More specifically, if the drift analysis result of the warning working parameter is a work parameter that needs to be regulated, the work parameter that needs to be regulated is subjected to regulation processing, specifically:

[0037] Regulation measures corresponding to different amplitude ranges of drifts of each working parameter of the target switch cabinet during operation are formulated.

[0038] construct a database, and import the control measures corresponding to the drift of each working parameter of the formulated target switch cabinet in different amplitude ranges during the operation into the database to obtain a control scheme database; and periodically update the control scheme database;

[0039] If the drift analysis result of the early warning working parameter is a working parameter that needs to be controlled, the drift value of the working parameter that needs to be controlled is obtained;

[0040] The drift value of the working parameter that needs to be controlled is imported into the control scheme database for pairing to obtain a corresponding control measure;

[0041] The obtained control measure is sent to the controller of the target switch cabinet to control the target switch cabinet according to the obtained control measure.

[0042] The application further discloses a power distribution network switch cabinet control system based on intelligent monitoring, which comprises a memory and a processor, and the memory stores a power distribution network switch cabinet control method program.

[0043] The application solves the technical defects in the background art and has the following beneficial effects: real-time environmental feature data is imported into an index knowledge graph for pairing analysis and evaluation; if the power distribution control performance of the target switch cabinet at the preset time node is normal, the target switch cabinet is not processed; if the power distribution control performance of the target switch cabinet at the preset time node is in an early warning state, the early warning working parameter of the target switch cabinet at the preset time node is traced back in the index knowledge graph, and the early warning working parameter is analyzed for drift; if the drift analysis result of the early warning working parameter is a working parameter that does not need to be controlled, the early warning working parameter is continuously monitored; and if the drift analysis result of the early warning working parameter is a working parameter that needs to be controlled, the working parameter that needs to be controlled is controlled. The application can comprehensively evaluate the running state of the switch cabinet by real-time monitoring and intelligent analysis of the target switch cabinet, can accurately locate and process the early warning working parameter by correlation evaluation and index knowledge graph construction of abnormal events, can realize fine management and active prevention and control of the switch cabinet, and can ensure the stability and reliability of the power distribution system. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0045] Figure 1 The overall method flowchart of the power distribution network switch cabinet control method;

[0046] Figure 2 The partial method flowchart of the power distribution network switch cabinet control method;

[0047] Figure 3 The system block diagram of the power distribution network switch cabinet control system. DETAILED DESCRIPTION

[0048] In order to more clearly illustrate the above-mentioned purposes, features and advantages of the present application, the following will further describe the present application in combination with the drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0049] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, however, the present application can also be implemented in other ways different from those described herein, therefore, the protection scope of the present application is not limited by the specific embodiments disclosed below.

[0050] As shown in Figure 1 The present application discloses a power distribution network switch cabinet control method based on intelligent monitoring, comprising the following steps:

[0051] S102: acquiring various working environment abnormal event occurred to the target switch cabinet during operation, performing correlation evaluation on the various working environment abnormal event occurred to the target switch cabinet during operation, and acquiring the associated working parameters of the target switch cabinet when the various working environment abnormal event occurs;

[0052] S104: acquiring the historical environment characteristic data of the target switch cabinet when the various working environment abnormal event occurs, and constructing an index knowledge graph according to the historical environment characteristic data and the associated working parameters of the target switch cabinet when the various working environment abnormal event occurs;

[0053] S106: in the actual operation process of the target switch cabinet, collecting real-time environment characteristic data of the target switch cabinet at a preset time node, importing the real-time environment characteristic data into the index knowledge graph for matching analysis and evaluation, and evaluating the power distribution control performance state of the target switch cabinet at the preset time node;

[0054] S108: If the power distribution control performance of the target switch cabinet at the preset time node is in a normal state, no processing is performed on the target switch cabinet, and the power distribution control performance continues to be analyzed and evaluated at the next preset time node;

[0055] S110: If the power distribution control performance of the target switch cabinet at the preset time node is in a pre-warning state, the pre-warning working parameters of the target switch cabinet at the preset time node are traced back in the index knowledge graph, and drift analysis is performed on the pre-warning working parameters to obtain a drift analysis result;

[0056] S112: If the drift analysis result of the pre-warning working parameters is that the working parameters do not need to be adjusted, the pre-warning working parameters are continuously monitored; if the drift analysis result of the pre-warning working parameters is that the working parameters need to be adjusted, the working parameters that need to be adjusted are adjusted and processed.

[0057] The present application can comprehensively evaluate the running state of the switch cabinet, discover potential problems in time, accurately locate and process the pre-warning working parameters through the correlation evaluation and construction of the index knowledge graph of abnormal events, and realize the fine management and active prevention and control of the switch cabinet, so as to ensure the stability and reliability of the power distribution system.

[0058] More specifically, the various working environment abnormal events that have occurred to the target switch cabinet during operation are obtained, the correlation of the various working environment abnormal events that have occurred to the target switch cabinet during operation is evaluated, the correlation working parameters of the target switch cabinet when the various working environment abnormal events occur are obtained, such as Figure 2 as shown, specifically:

[0059] S202: Obtain the working record book of the target switch cabinet, and obtain the various working environment abnormal events that have occurred to the target switch cabinet during operation according to the working record book;

[0060] The working environment abnormal events include but are not limited to temperature abnormality, humidity abnormality, electromagnetic interference abnormality, dust concentration abnormality, and air pressure abnormality.

[0061] S204: Randomly extract a working environment abnormal event that has not been extracted, obtain the historical working parameters of the target switch cabinet when the working environment abnormal event occurs according to the working record book, calculate the parameter difference between each historical working parameter and the corresponding preset working parameter respectively, and obtain the offset amplitude of each historical working parameter of the target switch cabinet when the working environment abnormal event occurs;

[0062] The historical working parameters include but are not limited to input voltage, output voltage, input current, output current, power factor, frequency parameter, and grounding resistance value.

[0063] S206: The historical working parameter corresponding to the offset amplitude greater than the preset threshold is calibrated as the associated working parameter of the target switch cabinet when the working environment abnormal event occurs;

[0064] S208: Return to execute S204 to S206 to continue the relevance evaluation of the working environment abnormal events occurred in the operation of the target switch cabinet until all the working environment abnormal events are extracted to obtain the associated working parameters of the target switch cabinet when various working environment abnormal events occur.

[0065] Overall, the working record book is obtained, and various working environment abnormal events occurred are determined. Then, for each abnormal event that has not been extracted, the corresponding historical working parameter is obtained according to the working record book, and the difference value with the preset working parameter is calculated to obtain the offset amplitude. The historical working parameter with a large offset amplitude is calibrated as the associated working parameter. This process is repeated until all abnormal events are processed to obtain comprehensive associated working parameters. Through this step, various working environment abnormal events occurring in the operation of the target switch cabinet and their association with the working parameters can be systematically and comprehensively sorted out, and the working parameters closely related to the abnormal events can be accurately found out, which helps to deeply understand the running state of the switch cabinet and the essential connection of the abnormal events, provides accurate and key data basis for subsequent construction of knowledge graph, evaluation of power distribution control performance and targeted measures, and improves the effectiveness of analysis and processing of switch cabinet abnormal conditions.

[0066] More specifically, the historical environment feature data of the target switch cabinet when various working environment abnormal events occur is obtained, and an index knowledge graph is constructed according to the historical environment feature data of the target switch cabinet when various working environment abnormal events occur and the associated working parameters, specifically:

[0067] The feature extraction processing is performed on various working environment abnormal events to obtain the historical environment feature data of the target switch cabinet when various working environment abnormal events occur;

[0068] Among them, first, all data information related to various work environment abnormal events are comprehensively collected, including but not limited to temperature, humidity, electromagnetic and other aspects of data and specific situation description of abnormal events. Then, these data are classified and arranged in detail, and different types of abnormal events and their related characteristics are distinguished. Then, by using data mining and pattern recognition technology, the data characteristics of each kind of abnormal event are analyzed in depth, and the key features representing the abnormal event are extracted, such as specific temperature range, humidity change trend, etc. At the same time, combined with the operation record and related parameters of the target switch cabinet, the historical environmental characteristic data closely related to the abnormal event are further screened and determined. Finally, the extracted features and screened data are integrated and summarized to form the historical environmental characteristic data set for each kind of work environment abnormal event, so as to be used for subsequent construction of index knowledge graph and in-depth analysis and evaluation;

[0069] and acquiring the associated working parameters of the target switch cabinet when various work environment abnormal events occur;

[0070] constructing a knowledge graph, initializing the various work environment abnormal events as sub-nodes, and cutting out a plurality of index nodes in the knowledge graph according to the sub-nodes; creating a first storage space and a second storage space in each index node respectively;

[0071] storing the historical environmental characteristic data of the target switch cabinet when various work environment abnormal events occur in the first storage space of the corresponding index node respectively; and storing the associated working parameters of the target switch cabinet when various work environment abnormal events occur in the second storage space of the corresponding index node respectively;

[0072] After the historical environmental characteristic data and the associated working parameters of the target switch cabinet when various work environment abnormal events occur are stored, the index knowledge graph is constructed.

[0073] Specifically, first, feature extraction is performed on various work environment abnormal events to obtain the corresponding historical environment feature data of the target switch cabinet, and associated work parameters are obtained. Then, a knowledge graph is constructed, various abnormal events are taken as sub-nodes, and index nodes are divided out, a first storage space for storing historical environment feature data and a second storage space for storing associated work parameters are created in each index node, the corresponding data is respectively stored, and after the data storage is completed, the index knowledge graph is constructed. Through this way of constructing the index knowledge graph, the work environment abnormal events, the corresponding historical environment feature data and the associated work parameters can be clearly associated and integrated to form a systematic and hierarchical knowledge structure. This makes it possible to quickly and accurately find relevant data according to a specific abnormal event in subsequent analysis and evaluation of the switch cabinet, which helps to deeply understand the relationship between the abnormal event and the state of the switch cabinet, provides strong data support and analysis foundation for accurate evaluation of power distribution control performance, early warning and taking measures, and improves the intelligent level and efficiency of the entire switch cabinet monitoring and control.

[0074] More specifically, the real-time environment feature data is imported into the index knowledge graph for pairing analysis and evaluation, and the power distribution control performance state of the target switch cabinet at the preset time node is evaluated, specifically:

[0075] The real-time environment feature data is imported into the index knowledge graph, and the Pearson correlation coefficient between the real-time environment feature data and the historical environment feature data in each first storage space is calculated.

[0076] If the Pearson correlation coefficient between the real-time environment feature data and the historical environment feature data in each first storage space is not greater than a preset coefficient value, the power distribution control performance of the target switch cabinet at the preset time node is marked as normal state.

[0077] If the Pearson correlation coefficient between the real-time environment feature data and the historical environment feature data in one or more first storage spaces is greater than a preset coefficient value, the power distribution control performance of the target switch cabinet at the preset time node is marked as early warning state.

[0078] Through this step, the current power distribution control performance state of the target switch cabinet can be accurately judged to be normal or in early warning state according to the comparison between the real-time environment feature data and the historical data. This helps to discover potential problems in time so as to take corresponding measures for intervention and adjustment, ensures the stable operation of the switch cabinet and the reliability of power distribution, and improves the accuracy of monitoring and evaluation of the operating state of the switch cabinet.

[0079] More specifically, if the power distribution control performance of the target switch cabinet at the preset time node is in the early warning state, the early warning working parameters of the target switch cabinet at the preset time node are traced back in the index knowledge graph, and the following steps are further included:

[0080] Meanwhile, if the Pearson correlation coefficient between the real-time environment feature data and the historical environment feature data in one or more first storage spaces is greater than a preset coefficient value, the first storage space corresponding to the Pearson correlation coefficient greater than the preset coefficient value is marked;

[0081] The index node to which the marked first storage space belongs is traced back and marked, and the associated working parameters of the second storage space in the marked index node are extracted, and the extracted associated working parameters are designated as the early warning working parameters of the target switch cabinet at the preset time node.

[0082] When it is determined that the power distribution control performance of the target switch cabinet at the preset time node is in the early warning state, first, it is further determined which historical environment feature data in the first storage space has a greater Pearson correlation coefficient. Then, the first storage space corresponding to the Pearson correlation coefficient greater than the preset coefficient value is marked. Next, since the index nodes in the index knowledge graph have a corresponding relationship with the first storage space and the second storage space, the index node to which the marked first storage space belongs is traced back and marked. Finally, the associated working parameters are extracted from the second storage space in the marked index node, and these associated working parameters are determined as the early warning working parameters of the target switch cabinet at the preset time node. Through this step, after determining the early warning state, the specific index node corresponding to the specific historical environment feature data related to the early warning state can be accurately found, and the associated working parameters can be extracted, which makes the analysis of the early warning situation more accurate and targeted, and the specific working parameters that cause the early warning can be determined, so that the corresponding working parameters can be diagnosed, greatly shortening the diagnosis time, and the switch cabinet can be targeted for regulation and processing in a shorter time, effectively improving the efficiency and accuracy of the abnormal situation processing of the switch cabinet, and ensuring the stability and safety of the power distribution control of the switch cabinet.

[0083] More specifically, the early warning working parameters are subjected to drift analysis to obtain a drift analysis result, specifically:

[0084] An actual parameter value of the early warning working parameters of the target switch cabinet at the preset time node is obtained, a difference between the actual parameter value of the early warning working parameters of the target switch cabinet at the preset time node and a preset value is calculated, and a drift value of the early warning working parameters of the target switch cabinet at the preset time node is obtained;

[0085] It is determined whether the drift value of the early warning working parameters of the target switch cabinet at the preset time node is greater than a preset drift value;

[0086] If the drift value of the early warning working parameter of the target switch cabinet at the preset time node is greater than the preset drift value, the early warning working parameter is marked as a working parameter needing regulation;

[0087] If the drift value of the early warning working parameter of the target switch cabinet at the preset time node is not greater than the preset drift value, the early warning working parameter is marked as a working parameter not needing regulation, and the early warning working parameter is continuously monitored.

[0088] Specifically, first, the actual parameter value of the early warning working parameter of the target switch cabinet at the preset time node is obtained, and then the drift value of the early warning working parameter is obtained by calculating the difference between the actual parameter value and the preset value. Then, if the drift value is greater than the preset drift value set in advance, it indicates that the early warning working parameter needs to be regulated, and it is marked as a working parameter needing regulation. If the drift value is not greater than the preset drift value, the early warning working parameter is marked as a working parameter not needing regulation, and it is continuously monitored to observe its subsequent changes. The deviation of the actual state of the early warning working parameter from the normal state can be accurately analyzed, and whether the early warning working parameter needs to be regulated or only needs to be continuously monitored can be accurately determined by comparing the drift value with the preset drift value. This helps to timely and accurately take corresponding measures, and the working parameter needing regulation can be timely processed to avoid potential problems from worsening, and the working parameter not needing regulation can also be continuously monitored to ensure real-time grasp of its state, thereby improving the management and control level of the working parameters of the switch cabinet as a whole, and ensuring the stable and safe operation of the switch cabinet.

[0089] More specifically, if the drift analysis result of the early warning working parameter is a working parameter needing regulation, the working parameter needing regulation is regulated and processed, specifically:

[0090] Formulate regulation measures corresponding to the drift of each working parameter of the target switch cabinet in different amplitude ranges during operation;

[0091] Construct a database, import the regulation measures corresponding to the drift of each working parameter of the target switch cabinet in different amplitude ranges during operation into the database, obtain a regulation scheme database, and regularly update the regulation scheme database;

[0092] If the drift analysis result of the early warning working parameter is a working parameter needing regulation, the drift value of the working parameter needing regulation is obtained;

[0093] The drift value of the working parameter needing regulation is imported into the regulation scheme database for matching to obtain corresponding regulation measures;

[0094] The obtained regulation measure is transmitted to the controller of the target switch cabinet to regulate and control the target switch cabinet according to the obtained regulation measure.

[0095] Specifically, the related technical personnel first formulate the regulation measures corresponding to the drift of each working parameter of the target switch cabinet in different amplitude ranges (for example, by increasing or decreasing the output voltage value to adapt to the load change or correct the voltage drift; when the power factor is too low, increase the capacitor group to improve the power factor) during operation, then build a database, import these formulated measures into the database to form a regulation scheme database, and update the database regularly to ensure its timeliness and adaptability.

[0096] When it is determined that the drift analysis result of a certain early warning working parameter is a working parameter that needs to be regulated, the drift value of the working parameter that needs to be regulated (i.e., the difference between the actual parameter value of the early warning working parameter of the target switch cabinet at the preset time node and the preset value) is obtained, then the drift value is matched with the regulation scheme database to find the corresponding regulation measure, and finally the found regulation measure is transmitted to the controller of the target switch cabinet, and the controller regulates and controls the switch cabinet according to the measures. It can be seen that through this step, the appropriate response measure can be quickly and accurately found from the pre-constructed and updated regulation scheme database for the working parameter that needs to be regulated due to drift, and is timely applied to the target switch cabinet for regulation, thereby effectively avoiding problems that may be caused by abnormal drift of the working parameter, ensuring stable and reliable operation of the switch cabinet, improving the management accuracy of the working parameters of the switch cabinet and the ability to respond to sudden situations, and ensuring that the switch cabinet can maintain good operating state and performance under various conditions.

[0097] In actual application process, after collecting the real-time environmental characteristic data of the target switch cabinet at the preset time node, a step of denoising the real-time environmental characteristic data can be further included:

[0098] A voxel space is constructed, and a plurality of voxel arrays of equal size are cut out in the voxel space, and real-time environmental characteristic data of the target switch cabinet collected at a preset time node are obtained;

[0099] Each collected real-time environmental characteristic data is separately imported into a blank voxel array, after the import is completed, a fuzzy algorithm is introduced, and the fuzziness between the real-time environmental characteristic data in each voxel array is calculated based on the fuzzy algorithm;

[0100] The fuzziness between the real-time environmental characteristic data in each voxel array is compared with a preset fuzziness threshold;

[0101] Fuse the voxel array attached to the real-time environment feature data with a degree of fuzziness greater than the preset fuzziness threshold, to obtain a plurality of voxel blocks;

[0102] Obtain the real-time environment feature data attached to each voxel block, calculate the variance of the real-time environment feature data attached to each voxel block, and determine the voxel center of each voxel block according to the calculated variance;

[0103] Calculate the Euclidean distance between the real-time environment feature data attached to each voxel block and the voxel center, sum the Euclidean distances between the real-time environment feature data attached to each voxel block and the voxel center, and then take the average to obtain the density coefficient of each voxel block;

[0104] If the density coefficient of a certain voxel block is greater than the preset density coefficient value, the voxel block is marked as a high-density voxel block; if the density coefficient of a certain voxel block is not greater than the preset density coefficient value, the voxel block is marked as a low-density voxel block;

[0105] Mark the real-time environment feature data attached to the low-density voxel block as noise data, and completely delete the real-time environment feature data marked as noise data to obtain the denoised real-time environment feature data.

[0106] In general, first, a voxel space is constructed and voxel arrays of equal size are divided, and then real-time environment feature data is imported into these arrays. Next, a fuzzy algorithm is used to calculate the degree of fuzziness between the data, and compared with a preset threshold, data with high fuzziness is fused into voxel blocks. Then, the variance of the data in each voxel block is calculated, the voxel center is determined, and the Euclidean distance between the data and the voxel center is calculated, and the average is obtained after summation to obtain the density coefficient. According to the comparison of the density coefficient with the preset value, the voxel block is marked as high-density or low-density. Finally, the data in the low-density voxel block is marked as noise data and deleted, thereby obtaining the denoised real-time environment feature data.

[0107] The reason why the data in the low-density voxel block is marked as noise data is that these data points are relatively sparse in the voxel space, meaning that they have low similarity with other data points and have not formed obvious clustering or patterns. In the fuzzy algorithm, these data points have low fuzziness with other data points, indicating that they have large differences in features with other data points. Since noise data refers to data points that are inconsistent with normal data patterns or have no clear association, the data points in the low-density voxel block are likely to be caused by measurement errors, transmission interference or other abnormal factors, rather than part of the normal operation of the switchgear. Therefore, marking these data points as noise data and deleting them can reduce the adverse effects on the analysis of the switchgear operating state, and improve the accuracy and reliability of data analysis.

[0108] In summary, through the denoising processing, the noise and abnormal values in the real-time environmental feature data can be effectively removed, the accuracy and reliability of the data are improved, the operation state of the switch cabinet can be more accurately analyzed and evaluated, and more effective regulation and optimization measures can be taken to ensure the stable operation of the switch cabinet and the reliability of power distribution.

[0109] In addition, the method can further include the following steps:

[0110] acquiring, through a big data network, dynamic working data of the target switch cabinet in a preset time period before the fault state, and performing discretization processing on the dynamic working data of the target switch cabinet in the preset time period before the fault state to obtain working data of each timestamp of the target switch cabinet in the preset time period before the fault state;

[0111] introducing a hidden Markov chain, taking the working data of each timestamp of the target switch cabinet in the preset time period before the fault state as an observation state, and calculating a state transition probability value of each observation state transitioning to another observation state based on the hidden Markov chain;

[0112] if the state transition probability value is greater than a preset probability value, updating the observation state of the corresponding timestamp to another observation state; if the state transition probability value is not greater than the preset probability value, maintaining the observation state of the corresponding timestamp unchanged;

[0113] after updating the observation state of each timestamp, constructing an observation state feature matrix according to the observation state of each timestamp;

[0114] constructing a hidden Markov model based on a deep learning network, embedding the observation state feature matrix in the hidden Markov model for back propagation learning until the prediction accuracy of the model meets a preset requirement;

[0115] In the actual operation process of the target switch cabinet, working data of the target switch cabinet is collected at a plurality of time nodes to obtain actual dynamic working data of the target switch cabinet;

[0116] importing the actual dynamic working data of the target switch cabinet into the hidden Markov model for prediction to obtain observation states of the target switch cabinet at a plurality of future timestamps;

[0117] analyzing the observation states of the target switch cabinet at the plurality of future timestamps, if the observation state of one or more future timestamps is a fault state, generating a fault warning information, controlling the target switch cabinet to stop working, and sending the fault warning information to a preset terminal for display.

[0118] It should be noted that the dynamic working data of the target switch cabinet in the preset time period before the fault state is obtained through the big data network, and these data are discretely processed to facilitate subsequent analysis. Then, the hidden Markov chain is introduced to calculate the state transition probability value between the working data (observation state) of each timestamp. According to these probability values, the observation state is updated, and if the probability value is greater than the preset value, the state is updated; otherwise, the state remains unchanged. After updating, the observation state feature matrix is constructed, and the hidden Markov model is constructed based on the deep learning network. The model is learned and optimized through back propagation until the prediction accuracy meets the preset requirements.

[0119] In actual operation, the working data of the target switch cabinet is collected, and the trained hidden Markov model is used for prediction to obtain the observation state of the future timestamp. If the prediction result shows that the observation state of a future timestamp is a fault state, a fault warning information is generated, and these information is sent to the preset terminal display. By combining big data, hidden Markov chain and deep learning technology, the fault state of the target switch cabinet can be effectively predicted, and a warning can be given before the fault occurs, which helps to take measures in advance to prevent the occurrence of faults, improves the reliability and safety of the switch cabinet, and reduces potential losses and risks.

[0120] As shown in Figure 3 The application further discloses a power distribution network switch cabinet control system 6 based on intelligent monitoring, which comprises a memory 41 and a processor 52, and the memory 41 stores a power distribution network switch cabinet control method program. When the power distribution network switch cabinet control method program is executed by the processor 52, the power distribution network switch cabinet control method steps of any one of the embodiments are realized.

[0121] In the several embodiments provided by the present application, it should be understood that the disclosed device and method can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. There can be another division manner in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the various components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0122] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units; they can be located in one place or distributed on multiple network units; and some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.

[0123] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be separately implemented as a single unit, or two or more units can be integrated in one unit; the integrated unit can be implemented in the form of hardware or hardware plus software function unit.

[0124] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware, and the foregoing program can be stored in a computer readable storage medium, and the program executes the steps including the above-mentioned method embodiments when executed; and the foregoing storage medium includes mobile storage equipment, read-only memory (ROM), random access memory (RAM), magnetic disc or optical disc and various storage program codes.

[0125] Alternatively, the integrated unit of the present application, if implemented in the form of a software function module and sold or used as an independent product, can also be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the embodiments of the method of the present application. The foregoing storage medium includes mobile storage equipment, ROM, RAM, magnetic disc or optical disc and various storage program codes.

[0126] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A power distribution network switchgear control method based on intelligent monitoring, characterized in that, The method comprises the following steps: S102: acquiring various working environment abnormal events occurred to the target switch cabinet during operation, performing correlation evaluation on the various working environment abnormal events occurred to the target switch cabinet during operation, and acquiring associated working parameters of the target switch cabinet when the various working environment abnormal events occur; S104: acquiring historical environment feature data of the target switch cabinet when the various working environment abnormal events occur, and constructing an index knowledge graph according to the historical environment feature data and the associated working parameters of the target switch cabinet when the various working environment abnormal events occur; S106: in the actual operation process of the target switch cabinet, real-time environment feature data of the target switch cabinet is collected at a preset time node, the real-time environment feature data is imported into the index knowledge graph for pairing analysis and evaluation, and a power distribution control performance state of the target switch cabinet at the preset time node is obtained through evaluation; S108: if the power distribution control performance of the target switch cabinet at the preset time node is normal, the target switch cabinet is not processed, and the power distribution control performance is continuously analyzed and evaluated at the next preset time node; S110: if the power distribution control performance of the target switch cabinet at the preset time node is a warning state, the warning working parameters of the target switch cabinet at the preset time node are traced back in the index knowledge graph, and drift analysis is performed on the warning working parameters to obtain a drift analysis result; S112: if the drift analysis result of the warning working parameters is that the working parameters do not need to be regulated, the warning working parameters are continuously monitored; if the drift analysis result of the warning working parameters is that the working parameters need to be regulated, the working parameters that need to be regulated are regulated and processed; The S102 is specifically: S202: obtaining a work record book of the target switch cabinet, and acquiring various working environment abnormal events occurred to the target switch cabinet during operation according to the work record book; S204: randomly extracting a working environment abnormal event that has not been extracted, acquiring various historical working parameters of the target switch cabinet when the working environment abnormal event occurs according to the work record book, respectively calculating parameter differences between the various historical working parameters and corresponding preset working parameters, and obtaining offset amplitudes of the various historical working parameters of the target switch cabinet when the working environment abnormal event occurs; S206: the historical working parameters corresponding to the offset amplitudes greater than a preset threshold are marked as the associated working parameters of the target switch cabinet when the working environment abnormal event occurs; S208: returning to execute the steps of S204 to S206 to continue to perform correlation evaluation on the working environment abnormal events occurred to the target switch cabinet during operation until all the working environment abnormal events are extracted, and the associated working parameters of the target switch cabinet when the various working environment abnormal events occur are obtained; The S104 is specifically: performing feature extraction processing on the various working environment abnormal events, acquiring the historical environment feature data of the target switch cabinet when the various working environment abnormal events occur, and acquiring the associated working parameters of the target switch cabinet when the various working environment abnormal events occur; and Construct a knowledge graph, initialize the various work environment abnormal events as child nodes, and divide a plurality of index nodes in the knowledge graph according to the child nodes; create a first storage space and a second storage space in each index node respectively; Store the historical environment characteristic data of the target switch cabinet when various work environment abnormal events occur in the first storage space of the corresponding index node respectively; and store the associated work parameters of the target switch cabinet when various work environment abnormal events occur in the second storage space of the corresponding index node respectively; After the historical environment characteristic data and the associated work parameters of the target switch cabinet when various work environment abnormal events occur are stored, an index knowledge graph is constructed; The S106 is specifically: Import the real-time environment characteristic data into the index knowledge graph, and calculate the Pearson correlation coefficients between the real-time environment characteristic data and the historical environment characteristic data in each first storage space; If the Pearson correlation coefficients between the real-time environment characteristic data and the historical environment characteristic data in each first storage space are all not greater than a preset coefficient value, mark the power distribution control performance of the target switch cabinet at a preset time node as normal; If the Pearson correlation coefficient between the real-time environment characteristic data and the historical environment characteristic data in one or more first storage spaces is greater than the preset coefficient value, mark the power distribution control performance of the target switch cabinet at the preset time node as a warning state.

2. The power distribution network switchgear control method based on intelligent monitoring according to claim 1, characterized in that, The S110 further comprises the following steps: At the same time, if the Pearson correlation coefficient between the real-time environment characteristic data and the historical environment characteristic data in one or more first storage spaces is greater than the preset coefficient value, mark the first storage space corresponding to the Pearson correlation coefficient greater than the preset coefficient value; Backtrack mark the index node to which the marked first storage space belongs, extract the associated work parameters of the second storage space in the marked index node, and mark the extracted associated work parameters as the warning work parameters of the target switch cabinet at the preset time node.

3. The power distribution network switchgear control method based on intelligent monitoring according to claim 1, characterized in that, Perform drift analysis on the warning work parameters to obtain a drift analysis result, specifically: Obtain the actual parameter value of the warning work parameters of the target switch cabinet at the preset time node, calculate the difference between the actual parameter value and a preset value to obtain the drift value of the warning work parameters of the target switch cabinet at the preset time node; Determine whether the drift value of the warning work parameters of the target switch cabinet at the preset time node is greater than a preset drift value; If the drift value of the warning work parameters of the target switch cabinet at the preset time node is greater than the preset drift value, mark the warning work parameters as a work parameter that needs to be regulated; If the drift value of the warning work parameters of the target switch cabinet at the preset time node is not greater than the preset drift value, mark the warning work parameters as a work parameter that does not need to be regulated, and continuously monitor the warning work parameters.

4. The power distribution network switchgear control method based on intelligent monitoring according to claim 3, characterized in that, If the drift analysis result of the warning work parameters is a work parameter that needs to be regulated, regulate the work parameter that needs to be regulated, specifically: Formulate the corresponding control measures of the target switch cabinet in the running process of each working parameter drift in different amplitude range; Construct a database, and import the corresponding control measures of the target switch cabinet in the running process of each working parameter drift in different amplitude range into the database to obtain a control scheme database; and update the control scheme database regularly; If the drift analysis result of the early warning working parameter is a working parameter that needs to be controlled, obtain the drift value of the working parameter that needs to be controlled; Import the drift value of the working parameter that needs to be controlled into the control scheme database for matching to obtain the corresponding control measures; The obtained control measures are sent to the controller of the target switch cabinet to control the target switch cabinet according to the obtained control measures.

5. A smart monitoring based switchgear control system for power distribution network characterized by, The switch cabinet control system of the power distribution network comprises a memory and a processor, and the memory stores a power distribution network switch cabinet control method program. When the power distribution network switch cabinet control method program is executed by the processor, the power distribution network switch cabinet control method steps of any one of claims 1 to 4 are realized.

Images