A cloud collaborative data monitoring system and method applied to power equipment

Abstract

The application discloses a kind of cloud end collaborative data supervision systems and methods applied to electric power equipment, it is related to cloud end data management technical field, through cloud end acquisition electric power equipment data, the original data of acquisition is standardized, obtains operating characteristic, after cloud end issues order, the same operating characteristic of electric power equipment is recorded uniformly, according to the difference of operating characteristic preliminary alarm is carried out, call historical characteristic data, the operating characteristic anomaly of electric power equipment is classified, the abnormal association between the operating characteristic of electric power equipment is analyzed, the operating characteristic of electric power equipment is analyzed secondly, through cloud end to management personnel issues interlock abnormal alarm and key attention reminder, when monitoring cycle ends, the stability of equipment is analyzed, to confirm the electric power equipment needing maintenance, the application is through capturing the temperature variation rate of each key stage, predictive dynamic adjustment temperature parameter, into associated monitoring and risk early warning mechanism, the stability of electric power equipment management is enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cloud data management, in particular to a cloud collaborative data supervision system and method applied to power equipment. BACKGROUND

[0002] In the field of power equipment operation supervision, traditional monitoring technology has long been plagued by multiple technical bottlenecks. On the one hand, existing systems generally rely on single operating characteristic threshold judgment mechanisms, which can only set fixed alarm limits for basic parameters such as voltage and current of the equipment, and lack systematic analysis of the operating state of the equipment. This mode is easily disturbed by external factors such as environmental noise and data fluctuations, resulting in a large number of false positives or false negatives. For example, in industrial environments with strong electromagnetic interference, partial discharge monitoring equipment often misjudges transient noise as equipment anomalies. At the same time, single feature analysis cannot capture the correlation between multiple dimensions of equipment operating parameters, and cannot fully reflect the health status of the equipment. For example, capacitor bank failures often have complex correlations with multiple power quality indicators such as voltage deviation and harmonic distortion rate, but traditional methods cannot achieve multi-variable cross-analysis;

[0003] On the one hand, the traditional monitoring system generally presents a passive response characteristic, triggering an alarm only when the equipment parameters are significantly above the threshold, lacking the ability to predict potential risks. This mode often causes maintenance personnel to intervene only after the equipment fails, and cannot take preventive measures at the early stages of failure. For example, when a wind turbine generator is running under variable load, the vibration data collected by a single sensor may not be able to reflect the subtle wear and tear inside the gearbox in a timely manner, and the traditional system lacks the ability to correlate historical data, making it difficult to identify early signs of failure;

[0004] On the other hand, the equipment maintenance strategy has long relied on experience-driven periodic inspection or post-failure maintenance mode, which has significant problems of unreasonable resource allocation. Periodic inspection cannot accurately identify the actual health status of the equipment, leading to over-maintenance or insufficient maintenance. For example, some equipment in good operating condition is frequently subjected to unnecessary maintenance, while equipment that actually has hidden dangers is not given timely attention. At the same time, the post-failure maintenance mode lacks the ability to predict the probability of equipment failure, often causing maintenance resources to be overwhelmed in emergency situations. In addition, existing technologies for statistical analysis of equipment anomaly characteristics stop at the simple counting level and fail to build a comprehensive evaluation system. For example, factors such as the number of abnormal characteristics and the correlation degree of cascading abnormalities are not included in the maintenance priority judgment model, resulting in maintenance decisions lacking data support. SUMMARY

[0005] The present application aims to provide a cloud collaborative data supervision system and method applied to power equipment to solve the problems raised in the background.

[0006] In order to solve the above technical problems, the present application provides the following technical solutions: a cloud collaborative data monitoring method applied to power equipment, comprising the following steps:

[0007] S1, collecting power equipment data through the cloud, standardizing the collected raw data to obtain operation characteristics;

[0008] S2, after the cloud issues an instruction, uniformly recording the same operation characteristics of the power equipment, and preliminarily alarming according to the differences in the operation characteristics;

[0009] S3, calling historical characteristic data to classify the operation characteristic abnormalities of the power equipment;

[0010] S4, analyzing the abnormal correlations between the operation characteristics of the power equipment;

[0011] S5, secondarily analyzing the operation characteristics of the power equipment, and issuing a chain abnormality alarm and a key attention reminder to the management personnel through the cloud;

[0012] S6, analyzing the equipment stability at the end of the monitoring period, so as to confirm the power equipment that needs to be maintained.

[0013] Further, in step S1, any cloud monitors X sets of uniformly managed same type power equipment, the equipment set is {A1, A2, …, A x ,…,A X}, the uniformly managed same type power equipment means the same type power equipment that is uniformly instructed, receives information and processes abnormalities by the cloud, wherein A x represents the xth set of same type power equipment monitored by the cloud, Y operation characteristics of the xth set of power equipment are obtained after the raw data of the power equipment are standardized, and the operation characteristic set is {B x_1 ,B x_2 ,…,B x_y ,…,B x_Y}, wherein B x_yBx,y(t) represents the yth operation characteristic of the xth power equipment, an initial collection time interval T is set, and the operation characteristics collected at the same time are time stamped; the cloud centrally manages and monitors the same type of power equipment, can realize centralized operation of command issuance, information reception and abnormality processing, and greatly improves the management efficiency and response speed. Standardizing the original data of the equipment can eliminate the differences in data formats, make the operation characteristics of different equipment consistent and comparable, and lay a standard foundation for subsequent abnormality analysis. Setting a fixed collection time interval and time stamping the operation characteristics can construct a time-sequenced equipment operation data chain, which is convenient for accurately tracing the operation status of the equipment at different stages, and provides an orderly data support for subsequent feature correlation analysis and abnormal evolution tracking based on time dimension, making the equipment operation monitoring more systematic and traceable.

[0014] Further, in step S2, after the cloud issues an instruction to the power equipment, the yth operation characteristic of the power equipment is uniformly recorded, the yth operation characteristic of the Xth power equipment is {B 1_y , B 2_y , …, B x_y , …, B X_y}, preliminary abnormality monitoring is performed, the preliminary abnormality monitoring is that when (B x_y -B y ) / B y ≤ α y , it is judged that the yth operation characteristic of the xth power equipment is normal, wherein B y is the average value of the yth operation characteristic of the Xth power equipment, and α y is the allowable fluctuation range of the yth operation characteristic; when (B x_y -B y ) / B y > α y , it is judged that the yth operation characteristic of the xth power equipment is abnormal, and the cloud sends an abnormality alarm of the yth operation characteristic of the xth power equipment to the manager; after the cloud issues an instruction to the power equipment, the operation characteristics are uniformly recorded, which can realize centralized data collection and management of the same characteristics of the same type of equipment, and the dispersed equipment operation status can be systematically presented. By comparing the characteristics of a single device with the average value of the group, the normal operation range of the individual device can be accurately defined with the common reference of the same type of equipment, which not only avoids the one-sidedness of single device data judgment, but also relies on the stability of group data to improve the scientificity of abnormality identification. When the characteristic fluctuation exceeds the allowable range, the cloud alarm is triggered immediately, which enables the manager to master the equipment abnormality at the first time, provides an opportunity for rapid intervention in fault handling and prevents abnormal expansion, effectively guarantees the safety and reliability of equipment operation, and makes the equipment abnormality monitoring more real-time and targeted.

[0015] Further, in step S3, the historical power equipment features are called to the current time as the end point, the N times collected power equipment features are called to analyze the xth power equipment, and the y1th running feature is judged to be abnormal N times x_y1 ; when the y1th running feature is judged to be abnormal, the y2th running feature is judged to be abnormal N times x_y1_y2 ; when the y1th running feature is judged to be abnormal, the y2th running feature is judged to be abnormal n times in time sequence x_y1_y2 ; the calling of the historical power equipment features can mine the time sequence rules of equipment operation, and provide a long-term perspective for analyzing the evolution of equipment abnormalities through backtracking multiple rounds of collected data. The statistics of the abnormal times of a specific running feature can accurately measure the abnormal frequency of the feature, and the associated abnormal times of different features can find the potential correlation pattern between the running features of the equipment, especially the abnormal situation of the subsequent features in time sequence, which can capture the transmission chain and chain reaction of the abnormality in the equipment. This deep mining and layered statistics of historical data enable the equipment abnormality analysis to jump out of the limitation of single judgment, accumulate data support for equipment health status evaluation from the long-term correlation and conduction mechanism, and make the abnormality monitoring more prospective and systematic.

[0016] Further, in step S4, the synchronous abnormality coefficient of the y1th running feature and the y2th running feature of the xth equipment is recorded as C y1-y2 =N x_y1_y2 / N x_y1 , the chain abnormality coefficient of the y1th running feature and the y2th running feature of the xth equipment is recorded as c y1-y2 =n x_y1_y2 / N x_y1 , and the additional abnormality parameter D y1-y2 of the y1th running feature to the y2th running feature is obtained, D y1-y2 =(C y1-y2 +c y1-y2 ) / 2, and the D y1-y2is used to describe the degree of influence of the y1th operating feature anomaly on the y2th operating feature anomaly; by calculating the synchronization anomaly coefficient and the linkage anomaly coefficient between the operating features of the computing device, the correlation degree of different feature anomalies can be accurately quantified. The synchronization anomaly coefficient can reflect the frequency of feature collaborative anomaly, and the linkage anomaly coefficient can capture the probability of anomaly transmission over time. The additional anomaly parameter formed by the combination of the two can comprehensively depict the abnormal correlation strength between features. This quantitative analysis breaks the limitations of single feature independent judgment, and changes the device anomaly monitoring from isolated analysis to correlation analysis, providing a quantitative basis for feature linkage for subsequent anomaly judgment. Management personnel can master the transmission rules and collaborative relationship of feature anomalies inside the device, fully consider the mutual influence between features when abnormal warning, so that the anomaly identification is more in line with the actual operation logic of the device, effectively improving the accuracy and systematicness of fault prediction.

[0017] Further, in step S5, when the abnormal alarm of the y1th operating feature anomaly of the xth power device is triggered in the preliminary anomaly monitoring, the y2th operating feature of the xth power device is judged again. When , it is judged that the y2th operating feature of the xth power device is normal, wherein B y2 is the average value of the y2th operating feature of the X power devices, and a y2 is the allowable fluctuation range of the y2th operating feature; when , it is judged that the y2th operating feature of the xth power device is linkage anomaly, and the linkage anomaly alarm of the y2th operating feature of the xth power device is sent to the management personnel through the cloud;

[0018] When all operating features are judged to be normal in the preliminary anomaly monitoring, the correlation monitoring of the operating features is performed, and the xth power device is analyzed. The set of operating features judged to be abnormal or linkage anomaly in the previous monitoring with the current time as the endpoint is {G x_1 ,G x_2 ,…,G x_m ,…,G x_M}, wherein G x_m represents the mth operating feature in {G x_1 ,G x_2 ,…,G x_m ,…,G x_M}, the calibration coefficient F y of the yth operating feature of the xth power device is calculated. The calibration coefficient F y is the sum of the values in the set {D G1-y ,D G2-y ,…,D Gm-y ,…,D GM-y}, wherein D Gm-y represents the operating feature G x_mFor the additional anomaly parameter of the y-th running feature, when When the x-th power device is in operation, the y-th operating characteristic is determined to be normal; when When an abnormal risk is identified in the y-th operating characteristic of the x-th power device, a key attention alert for the y-th operating characteristic of the x-th power device is sent to management personnel via the cloud. When an initial abnormality is triggered, a secondary assessment of related characteristics is performed by adding abnormal parameters. This fully utilizes the abnormal transmission patterns between device characteristics, avoiding the isolation of single-characteristic assessments. This reduces the risk of missed detections due to feature linkage and accurately identifies true cascading abnormalities, making cloud alerts more aligned with the actual fault logic of the equipment. When the initial assessment is normal, the correlation impact of historical abnormal characteristics is summarized by calibration coefficients. This uncovers potential abnormal risks behind the current normal characteristics, extending the monitoring perspective from the immediate state to historical correlation evolution. This allows management personnel to intervene early before obvious abnormalities appear, shifting from passive response to proactive prevention. This effectively improves the foresight and comprehensiveness of equipment abnormality early warnings, providing deeper decision support for equipment health management.

[0019] In step S6, at the end of any monitoring cycle, the number of abnormal features, cascading abnormal features, and key-concern features of any power equipment are counted. The stability of the power equipment is obtained by weighted summation of these numbers. Then, the β power equipment with the lowest stability is selected for maintenance, where β is the predetermined number of power equipment to be maintained. After the monitoring cycle ends, the number of various abnormal features of the equipment is counted and weighted to calculate the stability, allowing for a comprehensive quantitative assessment of the reliability of the equipment's operating status. This method, which comprehensively considers abnormal features, cascading abnormal features, and key-concern features, avoids the one-sidedness of single-indicator evaluation, making the assessment of equipment health status more consistent with actual operating conditions. Selecting maintenance targets based on stability ranking accurately identifies the equipment with the most prominent problems, prioritizing maintenance resources for the highest-risk areas. This changes the traditional blind maintenance model, improving maintenance efficiency and targeting, eliminating potential equipment hazards in advance, effectively reducing the failure rate, and ensuring the overall stable operation of the power equipment group.

[0020] A cloud-based collaborative data monitoring system for power equipment includes: a data acquisition and standardization module, a preliminary anomaly monitoring module, a historical data feature analysis module, a feature anomaly correlation analysis module, a secondary analysis and alarm module, and an equipment maintenance decision module.

[0021] The data acquisition and standardization module is used to collect power equipment data through the cloud, standardize the collected raw data, and obtain operating characteristics.

[0022] The preliminary anomaly monitoring module is used for recording the same operation characteristics of the power equipment after the instruction of the cloud, and performing preliminary alarm according to the difference of the operation characteristics;

[0023] The historical data characteristic analysis module is used for calling the historical characteristic data, and classifying the operation characteristic anomaly of the power equipment;

[0024] The characteristic anomaly correlation analysis module is used for analyzing the abnormal correlation between the operation characteristics of the power equipment;

[0025] The secondary analysis and alarm module is used for performing secondary analysis on the operation characteristics of the power equipment, and issuing a chain abnormal alarm and a key attention reminder to the manager through the cloud;

[0026] The equipment maintenance decision module is used for analyzing the stability of the equipment at the end of the monitoring period, so as to confirm the power equipment that needs to be maintained.

[0027] Compared with the prior art, the beneficial effects achieved by the present application are: on the one hand, by constructing a multi-level anomaly monitoring system, the operation state of the power equipment is accurately controlled. The original data of the equipment is standardized and a time stamp is collected, and the fluctuation threshold is set according to the average value of the operation characteristics of the same equipment to make a preliminary abnormality judgment, and at the same time, the synchronous abnormality coefficient and the chain abnormality coefficient are introduced, the secondary judgment is made according to the preliminary abnormality characteristic correlation analysis, the alarm deviation caused by the single characteristic misjudgment is avoided, the accuracy and reliability of the abnormality identification are improved, and the cloud can more accurately locate the real abnormality of the equipment.

[0028] On the one hand, the correlation monitoring and risk early warning mechanism is innovatively integrated, that is, even if all the characteristics are normal in the preliminary monitoring, the additional influence of the historical abnormality or the chain abnormality characteristic on the current operation characteristic is calculated to evaluate the potential risk, and the key attention reminder is sent to the characteristics exceeding the risk threshold, so that the change from passive response to active risk prediction is realized, and the hidden problems in the equipment operation are found in advance, so that more sufficient decision time is provided for equipment failure prevention.

[0029] On the other hand, a stability evaluation system is constructed, the number of abnormal characteristics, chain abnormal characteristics and key attention characteristics of the equipment is comprehensively counted, and the stability is calculated by weighting, the equipment needing maintenance is accurately located according to the stability, the traditional blind maintenance mode is changed, the maintenance work is more targeted, the resource allocation is optimized, the power equipment maintenance efficiency is improved, and the stability and reliability of the power system operation are ensured. BRIEF DESCRIPTION OF DRAWINGS

[0030] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0031] Figure 1 is a structural diagram of a cloud collaborative data monitoring system applied to power equipment according to the present application;

[0032] Figure 2 is a flowchart of a cloud collaborative data monitoring method applied to power equipment according to the present application. DETAILED DESCRIPTION

[0033] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0034] Please refer to Figure 1 and Figure 2 , the present application provides a technical solution: a cloud collaborative data monitoring method applied to power equipment, comprising the following steps:

[0035] S1, collecting power equipment data through the cloud, standardizing the collected raw data to obtain running characteristics;

[0036] S2, after the cloud issues an instruction, uniformly recording the same running characteristics of the power equipment, and preliminarily alarming according to the differences in the running characteristics;

[0037] S3, calling historical characteristic data to classify the abnormal running characteristics of the power equipment;

[0038] S4, analyzing the abnormal correlations between the running characteristics of the power equipment;

[0039] S5, secondarily analyzing the running characteristics of the power equipment, and issuing a chain abnormality alarm and a key attention reminder to the management personnel through the cloud;

[0040] S6, analyzing the stability of the equipment at the end of the monitoring period, so as to confirm the power equipment that needs to be maintained.

[0041] In step S1, any cloud monitors X sets of uniformly managed same type power equipment, and the equipment set is {A1, A2, …, Ax, …, An}, which represents the same type power equipment that is uniformly instructed, receives information and processes abnormalities by the cloud, wherein Ax represents the xth same type power equipment monitored by the cloud, the xth power equipment is monitored, and after the raw data of the power equipment is standardized, Y running characteristics of the xth power equipment are obtained, and the running characteristic set is {B1, B2, …, Bx, …, By}. x X x x_1 x_2 ​​​​..., B x_y ..., B x_Y}, wherein B x_y represents the yth operating characteristic of the xth power equipment, an initial collection time interval T is set, and the operating characteristics collected at the same time are time-stamped; the cloud centrally manages and monitors the same type of power equipment, can realize centralized operation of command issuance, information reception, and abnormality processing, and greatly improves the management efficiency and response speed. Standardizing the original data of the equipment can eliminate the differences in data formats, make the operating characteristics of different equipment consistent and comparable, lay a standard foundation for subsequent abnormality analysis. Setting a fixed collection time interval and time-stamping the operating characteristics can construct a time-sequenced equipment operating data chain, facilitate accurate tracing of the operating status of the equipment at different stages, and provide orderly data support for subsequent feature correlation analysis and abnormality evolution tracking based on the time dimension, making the equipment operation monitoring more systematic and traceable.

[0042] In step S2, after the cloud issues a command to the power equipment, the yth operating characteristic of the power equipment is uniformly recorded, and the yth operating characteristic of the xth power equipment is {B 1_y ,B 2_y ..., B x_y ..., B X_y} is recorded, preliminary abnormality monitoring is performed, the preliminary abnormality monitoring is: when (B x_y -B y ) / B y ≤ α y , it is judged that the yth operating characteristic of the xth power equipment is normal, wherein B y is the average value of the yth operating characteristic of the xth power equipment, and α y is the allowable fluctuation range of the yth operating characteristic; when (B x_y -B y ) / B y > α yWhen the xth power equipment yth operation characteristic is determined to be abnormal, the cloud sends an alarm to the manager about the abnormality of the xth power equipment yth operation characteristic; the cloud records the operation characteristics after sending the instructions to the power equipment, and can realize centralized data collection and management of the same characteristics of the same equipment, so that the running state of the dispersed equipment can be systematically presented. By comparing the characteristics of a single device with the average value of the group, the normal operation range of the individual device can be accurately defined by relying on the common benchmark of similar devices, which not only avoids the one-sidedness of single device data judgment, but also improves the scientificity of abnormal identification by relying on the stability of group data. When the characteristic fluctuation exceeds the allowed amplitude, the cloud alarm is triggered immediately, so that the manager can master the equipment abnormality at the first time, which provides an opportunity for rapid intervention in fault handling and prevents the abnormality from being enlarged, effectively ensuring the safety and reliability of equipment operation, and making the equipment abnormality monitoring more real-time and targeted.

[0043] In step S3, the historical power equipment characteristics are called, the current time is taken as the end point, the power equipment characteristics collected N times are called, and the xth power equipment is analyzed. The number of times that the yth operation characteristic is determined to be abnormal is N x_y1 ; when the yth operation characteristic is determined to be abnormal, the number of times that the y2th operation characteristic is determined to be abnormal is N x_y1_y2 ; when the yth operation characteristic is determined to be abnormal, the number of times that the y2th operation characteristic is determined to be abnormal in the next collection in time sequence is n x_y1_y2 ; the calling of the historical power equipment characteristics can mine the time sequence rule of the equipment operation, and provide a long-term perspective for analyzing the evolution of equipment abnormalities by backtracking multiple rounds of collected data. The statistics of the number of times of a specific operation characteristic abnormality can accurately measure the abnormal frequency of the characteristic, and the associated number of times of different characteristic abnormalities can find the potential correlation pattern between the equipment operation characteristics. In particular, tracking the abnormality of the subsequent characteristics in time sequence can capture the transmission chain and chain reaction of the abnormality in the equipment. This deep mining and layered statistics of historical data enable the equipment abnormality analysis to break out of the limitations of single judgment, accumulate data support for equipment health status evaluation from the perspective of long-term correlation and conduction mechanism, and make the abnormality monitoring more prospective and systematic.

[0044] In step S4, the synchronous abnormality coefficient of the yth operation characteristic and the y2th operation characteristic of the xth equipment is recorded as C y1-y2 =N x_y1_y2 / N x_y1 , the chain abnormality coefficient of the yth operation characteristic and the y2th operation characteristic of the xth equipment is recorded as c y1-y2 =n x_y1_y2 / N x_y1 , and the additional abnormality parameter D of the yth operation characteristic to the y2th operation characteristic is obtained.y1-y2 = (C y1-y2 + c y1-y2 ) / 2; by calculating the synchronization abnormal coefficient and the chain abnormal coefficient between the features, the correlation degree of different feature abnormalities can be accurately quantified. The synchronization abnormal coefficient can reflect the frequency of feature cooperative abnormalities, and the chain abnormal coefficient can capture the probability of abnormal transmission by time. The additional abnormal parameters formed by the combination of the two can comprehensively depict the abnormal correlation strength between the features. This quantitative analysis breaks the limitations of single feature independent judgment, and changes the device abnormality monitoring from isolated analysis to correlation analysis, providing a quantitative basis for feature linkage for subsequent abnormal judgment. Management personnel can master the transmission rules and cooperative relationship of feature abnormalities inside the device, fully consider the mutual influence between features when abnormal warning, so that the abnormal identification is more in line with the actual operation logic of the device, and the accuracy and systematicness of fault prediction are effectively improved.

[0045] In step S5, when the abnormal alarm of the y1th running feature of the xth power device is triggered in the preliminary abnormality monitoring, the y2th running feature of the xth power device is judged again. When , it is judged that the y2th running feature of the xth power device is judged normally twice, wherein B y2 is the average value of the y2th running feature of the xth power device, and a y2 is the allowable fluctuation range of the y2th running feature; when , it is judged that the y2th running feature of the xth power device is in chain abnormality, and the cloud sends a chain abnormality alarm of the y2th running feature of the xth power device to the management personnel;

[0046] When all running features are judged to be normal in the preliminary abnormality monitoring, the running features are monitored in correlation, and the xth power device is analyzed. The set of running features judged to be abnormal or in chain abnormality in the previous monitoring with the current time as the end point is {G x_1 ,G x_2 ,…,G x_m ,…,G x_M}, wherein G x_m represents the mth running feature in {G x_1 ,G x_2 ,…,G x_m ,…,G x_M}, the calibration coefficient F y of the yth running feature of the xth power device is calculated. The calibration coefficient F y is the sum of the values in the set {D G1-y ,D G2-y ,…,D Gm-y ,…,D GM-y}, wherein D Gm-y represents the running feature Gx_m For the additional anomaly parameter of the y-th running feature, when When the x-th power device is in operation, the y-th operating characteristic is determined to be normal; when When an abnormal risk is identified in the y-th operating characteristic of the x-th power device, a key attention alert for the y-th operating characteristic of the x-th power device is sent to management personnel via the cloud. When an initial abnormality is triggered, a secondary assessment of related characteristics is performed by adding abnormal parameters. This fully utilizes the abnormal transmission patterns between device characteristics, avoiding the isolation of single-characteristic assessments. This reduces the risk of missed detections due to feature linkage and accurately identifies true cascading abnormalities, making cloud alerts more aligned with the actual fault logic of the equipment. When the initial assessment is normal, the correlation impact of historical abnormal characteristics is summarized by calibration coefficients. This uncovers potential abnormal risks behind the current normal characteristics, extending the monitoring perspective from the immediate state to historical correlation evolution. This allows management personnel to intervene early before obvious abnormalities appear, shifting from passive response to proactive prevention. This effectively improves the foresight and comprehensiveness of equipment abnormality early warnings, providing deeper decision support for equipment health management.

[0047] In step S6, at the end of any monitoring cycle, the number of abnormal features, cascading abnormal features, and key-concern features of any power equipment are counted. The stability of the power equipment is obtained by weighted summation of these numbers. Then, the β power equipment with the lowest stability is selected for maintenance, where β is the predetermined number of power equipment to be maintained. After the monitoring cycle ends, the number of various abnormal features of the equipment is counted and weighted to calculate the stability, allowing for a comprehensive quantitative assessment of the reliability of the equipment's operating status. This method, which comprehensively considers abnormal features, cascading abnormal features, and key-concern features, avoids the one-sidedness of single-indicator evaluation, making the assessment of equipment health status more consistent with actual operating conditions. Selecting maintenance targets based on stability ranking accurately identifies the equipment with the most prominent problems, prioritizing maintenance resources for the highest-risk areas. This changes the traditional blind maintenance model, improving maintenance efficiency and targeting, eliminating potential equipment hazards in advance, effectively reducing the failure rate, and ensuring the overall stable operation of the power equipment group.

[0048] A cloud-based collaborative data monitoring system for power equipment includes: a data acquisition and standardization module, a preliminary anomaly monitoring module, a historical data feature analysis module, a feature anomaly correlation analysis module, a secondary analysis and alarm module, and an equipment maintenance decision module.

[0049] The data acquisition and standardization module is used to collect power equipment data through the cloud, standardize the collected raw data, and obtain operating characteristics;

[0050] The preliminary anomaly monitoring module is used to record the same operation characteristics of the power equipment after the cloud issues an instruction, and to preliminarily alarm according to the differences in the operation characteristics;

[0051] The historical data feature analysis module is used to call historical feature data to classify the operation characteristic anomalies of the power equipment;

[0052] The feature anomaly correlation analysis module is used to analyze the abnormal correlation between the operation characteristics of the power equipment;

[0053] The secondary analysis and alarm module is used to perform secondary analysis on the operation characteristics of the power equipment, and to issue a chain abnormal alarm and a focus attention reminder to the manager through the cloud;

[0054] The equipment maintenance decision module is used to analyze the stability of the equipment at the end of the monitoring period, so as to confirm the power equipment that needs to be maintained.

[0055] Embodiment 1: In a cloud collaborative data supervision system applied to power equipment, real-time supervision needs to be performed on the same type of power equipment in multiple substations under jurisdiction. Taking a transformer as an example, the cloud system uniformly manages multiple transformers of the same type distributed in different sites, and continuously collects operation data through sensors deployed at the equipment end.

[0056] The cloud system starts data collection at a fixed time interval, and performs standardization processing on the original operation data (such as temperature, current, voltage and other parameters) of each transformer. For example, when collecting the oil temperature data of a transformer, the system converts the original signal transmitted by the sensor into a unified format temperature value, and at the same time, marks all operation parameters collected at the same time point with a time label, forming time-sequenced operation characteristic records. These characteristics cover multiple key indicators of the equipment, ensuring that the operation states of different transformers are comparable.

[0057] When the cloud issues a state query instruction to all transformers, the system automatically aggregates the same type of operation characteristics of each device (such as the real-time oil temperature of all transformers), and calculates the group average value of the same type of characteristics. If the deviation of the oil temperature of a transformer from the group average value exceeds the pre-set reasonable fluctuation range, the system immediately determines that the characteristic is abnormal, and sends an alarm to the manager. For example, when the oil temperature of a transformer is significantly higher than the average level of the same type of equipment, the cloud interface will highlight the equipment number, synchronously push the abnormal type and location information, and remind the operation and maintenance personnel to pay attention.

[0058] The system can call historical operation data collected for a transformer in recent times, and deeply analyze the abnormal frequency of specific features. For example, if it is found that the oil temperature of a transformer is frequently abnormal in recent monitoring, the system will further count the number of times other associated features (such as winding temperature) are abnormal when this feature is abnormal, as well as the abnormal transmission in the time series before and after. By tracing back the historical data, the evolution law of equipment abnormality can be mined, and whether there are signs of coordinated abnormality or chain reaction between different features can be judged.

[0059] Based on the statistical results of historical data, the system calculates the synchronous abnormal frequency and chain abnormal probability between different operating features to form a quantitative additional abnormality parameter. For example, when the oil temperature is abnormal, the proportion of the number of times the winding temperature is synchronously abnormal to the total number of times the oil temperature is abnormal, and the probability that the winding temperature is abnormal after the oil temperature is abnormal, combined to generate the abnormal association parameter of oil temperature on winding temperature. This parameter is used to describe the mutual influence strength between features, and provides linkage basis for subsequent abnormality judgment.

[0060] When the oil temperature of a transformer triggers a preliminary abnormality alarm, the system automatically makes a secondary judgment on its winding temperature, and adjusts the judgment threshold combined with the pre-calculated abnormal association parameter. If the winding temperature deviation is within the adjusted range, it is determined to be normal; if it is beyond the range, it is determined to be a chain abnormality and an alarm is issued. In addition, when all features are preliminarily judged to be normal, the system will review the features that have appeared abnormal or chain abnormal in the previous monitoring period, and calculate the comprehensive influence coefficient of these historical abnormal features on the current features. If a feature is affected by the comprehensive influence of historical abnormal features and its deviation exceeds the risk threshold, even if the current data seems normal, the management personnel will be reminded to pay special attention, and potential faults will be warned in advance.

[0061] After each monitoring period, the system counts the number of abnormal features, the number of chain abnormal features and the number of features of special attention for each transformer, and calculates the equipment stability index through weighted calculation. For example, the oil temperature of a transformer frequently appears abnormal and causes winding temperature chain abnormality, and many features are listed in the special attention list, so the stability score of the transformer will be significantly reduced. The system sorts by stability, and selects the worst devices in terms of state to be included in the maintenance plan, so as to ensure that maintenance resources are accurately invested in the devices with the highest risk, realize the change from passive maintenance to active prevention, and ensure the overall safe operation of the power equipment group.

[0062] Through the above whole-process collaborative supervision, the cloud system realizes multi-dimensional monitoring of the running state of power equipment, abnormal association analysis and accurate maintenance decision-making, effectively improves the efficiency and reliability of equipment management, and provides data support for stable operation of power grid.

[0063] It will be apparent to those skilled in the art that the application is not limited to the details of the above-exemplified embodiments and that the present application can be implemented in other particular forms without departing from the spirit or essential characteristics of the present application. The embodiments should therefore be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the above description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. No reference signs in the claims should be considered as limiting the scope of the claims with respect to the figures of the patent document.

Claims

1. A cloud collaborative data monitoring method applied to power equipment, characterized in that: The method comprises the following steps: S1, collecting power equipment data through the cloud, standardizing the collected raw data, and obtaining operation characteristics; S2, after the cloud issues an instruction, uniformly recording the same operation characteristics of the power equipment, and preliminarily alarming according to the differences in the operation characteristics; S3, calling historical characteristic data to classify the operation characteristic abnormalities of the power equipment; S4, analyzing the abnormal correlations between the operation characteristics of the power equipment; S5, performing secondary analysis on the operation characteristics of the power equipment, and issuing a chain abnormality alarm and a focus attention reminder to the manager through the cloud; S6, analyzing the stability of the equipment at the end of the monitoring period, thereby confirming the power equipment that needs to be maintained; In step S5, when the abnormal alarm of the y1th operating feature of the xth power device is triggered in the preliminary abnormality monitoring, the y2th operating feature of the xth power device is judged again, and when B y2 is the average value of the y2th operating feature of the X power devices, and α y2 is the allowable fluctuation range of the y2th operating feature; when the y2th operating feature of the xth power device is judged to be abnormal, the cloud sends a chain abnormality alarm of the y2th operating feature of the xth power device to the manager. When all operating characteristics are judged to be normal in the preliminary abnormality monitoring, the operating characteristics are monitored in association, the xth power equipment is analyzed, and the operating characteristic set judged to be abnormal or in a chain abnormality in the previous monitoring with the current time as the end point is {G x_1 ,G x_2 ,…,G x_m ,…,G x_M} where G x_m represents the mth operating characteristic in {G x_1 ,G x_2 ,…,G x_m ,…,G x_M}, the calibration coefficient F y of the yth operating characteristic of the xth power equipment is calculated, the calibration coefficient F y is the sum of the values in the set {D G1-y ,D G2-y ,…,D Gm-y ,…,D GM-y}, and D Gm-y represents the additional abnormality parameter of the operating characteristic G x_m to the yth operating characteristic, when D , the yth operating characteristic of the xth power equipment is judged to be normal; when D , the yth operating characteristic of the xth power equipment is judged to have an abnormality risk, and a key attention reminder of the yth operating characteristic of the xth power equipment is sent to the manager through the cloud, where D y1-y2 represents the additional abnormality parameter of the y1th operating characteristic to the y2th operating characteristic, B x_y represents the yth operating characteristic of the xth power equipment, B y is the average value of the yth operating characteristic of the X power equipment, and α y is the allowable fluctuation amplitude of the yth operating characteristic.

2. The cloud collaborative data monitoring method for power equipment according to claim 1, characterized in that: In step S1, any cloud monitoring X table unified management of the same kind of power equipment, equipment set is {A1, A2, …, A x ,…,A X}, the unified management of the same kind of power equipment indicates that the same kind of power equipment is commanded, information is received and handled by the cloud, wherein A x represents the xth cloud monitoring of the same kind of power equipment, the xth power equipment is monitored, and the original data of the power equipment is standardized to obtain Y running characteristics of the xth power equipment. The running characteristic set is {B x_1 ,B x_2 ,…,B x_y ,…,B x_Y}, wherein B x_y represents the yth running characteristic of the xth power equipment, and the initial collection time interval T is set. The running characteristics collected at the same time are time stamped.

3. The cloud collaborative data monitoring method for power equipment according to claim 2, characterized in that: In step S2, after the cloud issues an instruction to the power equipment, the yth running characteristic of the power equipment is uniformly recorded, and the yth running characteristic of the X power equipment is {B 1_y , 2_y …, B x_y , X_y} A preliminary anomaly monitoring is performed, which is: when (B x_y -B y ) / B y ≤α y , it is judged that the yth running characteristic of the xth power equipment is normal, wherein B y is the average value of the yth running characteristic of the X power equipment, and α y is the allowable fluctuation range of the yth running characteristic; when (B x_y -B y ) / B y >α y , it is judged that the yth running characteristic of the xth power equipment is abnormal, and an abnormal alarm of the yth running characteristic of the xth power equipment is sent to the manager in the cloud.

4. The cloud collaborative data monitoring method for power equipment according to claim 3, characterized in that: In step S3, the historical power equipment features are called, with the current time as the end point, the N times collected power equipment features are called, the xth power equipment is analyzed, the y1th running feature is judged to be abnormal for N times x_y1 ; when the y1th running feature is judged to be abnormal, the y2th running feature is judged to be abnormal for N times x_y1_y2 ; when the y1th running feature is judged to be abnormal, the y2th running feature next collected in time sequence is judged to be abnormal for n times x_y1_y2 .

5. The cloud collaborative data monitoring method for power equipment according to claim 4, characterized in that: In step S4, the synchronization abnormality coefficient of the y1th operating feature and the y2th operating feature of the xth device is recorded as C y1-y2 =N x_y1_y2 / N x_y1 , the linkage abnormality coefficient of the y1th operating feature and the y2th operating feature of the xth device is c y1-y2 =n x_y1_y2 / N x_y1 , and then the additional abnormality parameter D of the y1th operating feature to the y2th operating feature is obtained y1-y2 , D y1-y2 =(C y1-y2 +c y1-y2 ) / 2.

6. The cloud collaborative data monitoring method for power equipment according to claim 5, characterized in that: In step S6, at the end of any monitoring period, the number of abnormal characteristics, the number of chain abnormal characteristics, and the number of focus attention characteristics of any power equipment are counted, the number of abnormal characteristics, the number of chain abnormal characteristics, and the number of focus attention characteristics are weighted and summed to obtain the stability of the power equipment, and then the beta power equipment with the lowest stability is selected for maintenance, wherein beta is the number of power equipment to be maintained.

7. A cloud collaborative data monitoring system applied to power equipment, the system being applied to the cloud collaborative data monitoring method applied to power equipment according to any one of claims 1-6, characterized in that: The system comprises a data collection and standardization module, a preliminary abnormality monitoring module, a historical data characteristic analysis module, a characteristic abnormality correlation analysis module, a secondary analysis and alarm module, and an equipment maintenance decision module; The data collection and standardization module is used to collect power equipment data through the cloud, standardize the collected raw data, and obtain operation characteristics; The preliminary abnormality monitoring module is used to uniformly record the same operation characteristics of the power equipment after the cloud issues an instruction, and preliminarily alarm according to the differences in the operation characteristics; The historical data characteristic analysis module is used to call historical characteristic data to classify the operation characteristic abnormalities of the power equipment; The characteristic abnormality correlation analysis module is used to analyze the abnormal correlations between the operation characteristics of the power equipment; The secondary analysis and alarm module is used to perform secondary analysis on the operation characteristics of the power equipment, and issue a chain abnormality alarm and a focus attention reminder to the manager through the cloud; The equipment maintenance decision module is used to analyze the stability of the equipment at the end of the monitoring period, thereby confirming the power equipment that needs to be maintained.

Images