Abnormal power data management method and system

Abstract

The invention discloses an abnormal electric power data management method and system. The method comprises the steps of collecting electric power data, generating a spatial position map, recognizing abnormal data according to the electric power data, recognizing an abnormality generation reason based on the spatial position map and the abnormal data, and repairing the abnormal data. A monitor is used for monitoring multi-domain entropy increase, multiple domains comprise a physical domain, an information domain and a service domain, and the physical domain is physical equipment of the power data center; constructing a cross-domain entropy flow model according to the monitoring data of the physical domain, the information domain and the service domain; predicting an entropy increase trend according to the monitoring data of the physical domain, the information domain and the service domain; entropy increase is adjusted according to a cross-domain entropy flow model, a situation model is established according to the state of the power system, and abnormity is found and positioned in time by collecting and analyzing data in real time; through three-dimensional atlas and root cause decoupling analysis, abnormal root causes and propagation paths are accurately identified, and through a dynamic guiding function, the repair efficiency of operation and maintenance personnel is improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of computers, in particular to an abnormal power data management method and system. BACKGROUND

[0002] In today's digital era, the stability and reliability of the power data center, as the core infrastructure supporting numerous critical business operations, are of great importance. With the continuous expansion of the scale of the power data center and the increasing complexity of equipment, power data management faces many challenges. The power data center generates a large amount of power data, which covers equipment operation data, communication data, and environmental data in multiple dimensions. These data are not only large in quantity but also complex in type, containing rich information, but also may hide various abnormal situations. If abnormal power data cannot be discovered and processed in time, it may cause equipment failure, communication interruption, business system collapse and other serious consequences, thereby affecting the normal operation of the entire power data center and even causing huge economic losses.

[0003] Traditional power data management methods often rely on manual inspection and simple threshold alarms. This approach is not only inefficient, but also difficult to accurately identify the root cause and propagation path of abnormal data, and cannot meet the needs of modern power data centers for abnormal power data management. SUMMARY

[0004] The application provides an abnormal power data management method and system, which discovers and locates abnormalities in real time through real-time data collection and analysis, accurately identifies the root cause and propagation path of abnormalities through three-dimensional atlas and root cause decoupling analysis, improves the repair efficiency of operation and maintenance personnel through dynamic guidance function, realizes real-time positioning and root cause decoupling of concurrent abnormalities in the power data center, and improves the operation stability and reliability of the power data center.

[0005] The application provides an abnormal power data management method, comprising: S101, collecting power data, generating a spatial position atlas, identifying abnormal data according to the power data, identifying the cause of abnormality based on the spatial position atlas and the abnormal data, and repairing the abnormal data; S102, monitoring the multi-domain entropy increase using a monitor, the multi-domain including a physical domain, an information domain and a business domain, the physical domain being the physical equipment of the power data center; S103, constructing a cross-domain entropy flow model according to the monitoring data of the physical domain, the information domain and the business domain; S104, predicting the entropy increase trend according to the monitoring data of the physical domain, the information domain and the business domain; S105, adjusting the entropy increase according to the cross-domain entropy flow model, and establishing a situation model according to the power system state.

[0006] Preferably, the multi-domain entropy increase refers to a process in which a system in multiple domains transforms from an ordered state to a disordered state, that is, a process in which the entropy value of the system continuously increases.

[0007] Preferably, an entropy increase trend graph is generated according to the predicted entropy increase trend, and an algorithm is used to identify an entropy increase inflection point according to the entropy increase trend graph, wherein the entropy increase inflection point marks the transition of the system from a steady state to a latent unstable state.

[0008] Preferably, the central difference method is used to calculate the first-order derivative and the second-order derivative of the entropy increase curve, and the entropy increase inflection point is identified according to the first-order derivative and the second-order derivative. The calculation formula of the central difference method is: ,in, represents the first derivative of the entropy increase curve, Indicates a time point The entropy increase, Indicates a time point The entropy increase, Indicates a time interval.

[0009] Preferably, S201, obtaining the temperature of a physical device and forming a heat vector diagram based on the device surface temperature and spatial information; S202, calculating a thermal anomaly index based on the surface temperature of the physical device, and using the thermal anomaly index to identify a thermal anomaly vortex area; S203, dissipating heat from the device according to the abnormal thermal vortex area and the device surface temperature.

[0010] Preferably, the convection rate is calculated based on the surface temperature of the physical device, using the formula: ,in, represents the natural convection rate of air, is the empirical coefficient, Indicates the surface temperature of the device. Indicates the ambient temperature, Represents an exponent.

[0011] Preferably, the abnormal thermal vortex zone is used to identify an area where heat accumulates abnormally between equipment clusters.

[0012] Preferably, S301, monitoring internal changes of the physical device and identifying misalignment and accumulation of metal connectors of the physical device; S302, correcting the heat vector diagram and the abnormal thermal vortex area according to the misalignment and accumulation of metal connectors; S303: Setting a heat dissipation method according to the staggered stacking of the metal connectors.

[0013] Preferably, the misalignment accumulation of the metal connectors is formed by a thermal hysteresis effect, which describes the phenomenon that the thermal response of a substance lags behind the temperature change when the temperature changes.

[0014] The application also provides an abnormal power data management system, comprising a heat flow field modeling system, a thermal anomaly vortex identification system, and a dynamic heat dissipation strategy formulation system, the heat flow field modeling system is used to generate a heat vector diagram, the thermal anomaly vortex identification system is used to calculate a thermal anomaly index and identify a thermal anomaly vortex, and the dynamic heat dissipation strategy formulation system is used to determine the heat dissipation demand of the power equipment, the heat flow field modeling system is electrically connected with the thermal anomaly vortex identification system, and the thermal anomaly vortex identification system is electrically connected with the dynamic heat dissipation strategy formulation system.

[0015] One or more technical solutions provided in the application have at least the following technical effects or advantages: abnormality is found and located in time through real-time data collection and analysis; the root cause and propagation path of the abnormality are accurately identified through three-dimensional mapping and root cause decoupling analysis; the repair efficiency of the operation and maintenance personnel is improved through dynamic guidance function; real-time positioning and root cause decoupling of concurrent abnormality in the power data center are realized; the operation stability and reliability of the power data center are improved; cross-domain entropy increase monitoring, entropy flow coupling analysis, entropy increase inflection point prediction, and cross-domain entropy flow scheduling are used to realize the balance of cross-domain entropy flow; the occurrence and propagation of abnormality are found and inhibited in advance through the situation model and abnormality suppression strategy; cross-domain entropy balance in the power data center is realized; the steady-state operation of the system is maintained; the system is actively intervened before implicit instability; and the occurrence of steady-state collapse is prevented; comprehensive system monitoring and abnormality suppression are realized by identifying the cross-domain entropy increase of the physical domain, the information domain, and the business domain. Through the steps of heat vector diagram, thermal anomaly vortex identification, dynamic heat dissipation strategy formulation, and heat dissipation evaluation and verification, precise, dynamic, and efficient heat dissipation management is realized. The final result is that the surface temperature of the power equipment is effectively controlled, the thermal anomaly vortex area is significantly reduced or disappears, the stability and reliability of the power data center are significantly improved, the abnormal accumulation of heat in the equipment cluster in the power data center is solved, and efficient heat dissipation and stable operation of the system are ensured. By monitoring the internal changes of the equipment materials, identifying the misalignment accumulation of the equipment metal connectors, and combining the heat dissipation optimization algorithm, more precise and efficient heat dissipation management is realized, the stability and heat dissipation efficiency of the power data center are improved, the heat dissipation energy consumption is reduced, the overall energy efficiency is improved, high-precision monitoring of the misalignment accumulation of the equipment metal connectors is realized, intelligent formulation and dynamic adjustment of the heat dissipation strategy are realized, the heat dissipation energy consumption is reduced, and the overall energy efficiency is improved. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 It is a flowchart of the abnormal power data management method and system of the application; Figure 2 Flowchart for forming heat vector diagram and heat anomaly vortex area of the present application; Figure 3 Flowchart for identifying misaligned accumulation of physical device metal connectors of the present application. DETAILED DESCRIPTION

[0017] In order to facilitate the understanding of the present application, the present application will be described in more detail below with reference to the relevant drawings; the preferred embodiments of the present application are shown in the drawings, but the present application can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0018] It should be noted that the terms "vertical", "horizontal", "upper", "lower", "left", "right" and similar expressions used herein are only for illustrative purposes and do not represent the only embodiment.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs; the terms used herein in the specification of the present application are only for the purpose of describing the specific embodiments and are not intended to limit the present application; the term "and / or" used herein includes any and all combinations of one or more related listed items.

[0020] Example one: Figure 1 is a flowchart of an abnormal power data management method of an embodiment of the present application, comprising: S101, collecting power data, generating a spatial position map, identifying abnormal data according to the power data, identifying the cause of abnormality based on the spatial position map and the abnormal data, and repairing the abnormal data; Further, by deploying sensors in the power data center, real-time collection of equipment operation data, communication data and environmental data, cleaning of collected raw data, standardization of cleaned data, unification of data format and unit, detailed investigation of the physical layout of the power data center, drawing of a three-dimensional model of the power data center, labeling of the specific position of each equipment in the three-dimensional model, including the cabinet, rack, row number, column number and other information of the equipment, using a unified naming rule and identifier to identify the equipment, identifying the adjacent relationship and connection relationship between the equipment, generating a spatial position map according to the equipment position label and adjacent relationship record, using a network monitoring tool to analyze the data transmission relationship between the equipment, identifying the sender and receiver of the data, recording the transmission path of the data from the source equipment to the target equipment, including the intermediate equipment and network nodes, using a path diagram to represent the data transmission path, generating a data flow map according to the data transmission relationship and transmission path, the data flow map is used to show the flow of data in the power data center, generating a topology connection map according to the connection relationship of the equipment, using a deep learning algorithm LSTM network to monitor the power data according to the specific needs of the power data center and the historical data characteristics, setting the threshold value of abnormal monitoring according to the historical data and business needs, comparing the monitored real-time data with the set threshold value to judge whether the data is out of the normal range, when the monitored real-time data is greater than the preset threshold value, it is determined as abnormal data, and the related information of the abnormal data is recorded, such as occurrence time, equipment ID, abnormal type, abnormal value, etc., triggering an abnormal alarm mechanism to notify the operation and maintenance personnel or related system to handle the abnormal data, using the spatial position model according to the related information of the abnormal data, locating the specific spatial position where the abnormality occurs, including the computer room, cabinet, equipment, etc., using the data flow model to analyze the flow path of the abnormal data in the power data center, including the source equipment, intermediate equipment and target equipment, using the topology connection model to confirm the connection relationship and connection strength between the abnormal equipment and other equipment, and judge whether the abnormality is caused by the connection problem.

[0021] The specific location of the abnormal occurrence, data flow path, involved device and its connection relationship and other information are input into the causal analysis algorithm, the propagation path of the abnormality between the physical layer, the communication layer and the service layer is analyzed, and the coupling strength threshold is set for quantifying the coupling relationship between the abnormality between the physical layer, the communication layer and the service layer. The coupling strength threshold can be set according to historical data and expert experience, and the coupling strength between the abnormality between the layers is calculated according to the set coupling strength threshold. The coupling strength refers to the degree of mutual association and influence between the abnormality between the layers (physical layer, communication layer, service layer) of the power data center. It can be measured by quantifying the propagation ability, influence range and mutual dependence degree between the abnormality between the layers, and the size of the coupling strength reflects the degree of association and propagation ability between the abnormality between the layers. The coupling strength calculation is prior art, and will not be described here. According to the results of the propagation path analysis and the coupling strength calculation, the final root cause of the abnormality is determined.

[0022] The physical layout of the power data center, device position information, network topology diagram and the like are acquired, and the current position of the operation and maintenance personnel is analyzed: the current position of the operation and maintenance personnel is determined through the positioning information of the operation and maintenance terminal, the layout information of the power data center and the current position of the operation and maintenance personnel are combined, and the optimal travel route of the operation and maintenance personnel to the abnormal occurrence position is calculated using the path planning algorithm A* algorithm and the like, according to the final root cause of the abnormality, a specific repair strategy is formulated, including the device to be repaired, the repair method, the expected effect and the like, the repair strategy is converted into specific repair suggestions, the faulty device is replaced, the configuration parameters are adjusted, the software vulnerabilities are repaired, and the generated repair suggestions and the optimal travel route are integrated into a guide information, the guide information is sent to the mobile device or the handheld terminal of the operation and maintenance personnel through the communication module of the operation and maintenance terminal, and the operation and maintenance personnel performs repair operation at the abnormal occurrence position according to the repair suggestions, and the system monitors the indicators related to the abnormality in real time to determine whether the abnormality has been successfully repaired.

[0023] S102, a plurality of domain entropy increments are monitored using a monitor, the plurality of domains including a physical domain, an information domain and a service domain; Specifically, entropy is a concept in thermodynamics, used to measure the degree of disorder or chaos of a system. The entropy increase refers to the process of the system changing from an ordered state to a disordered state, that is, the process of the entropy value of the system increasing. In a natural state, an isolated system always tends to increase entropy, that is, the system will spontaneously evolve from a low-entropy (ordered) state to a high-entropy (disordered) state. The increase in entropy is manifested in the physical domain, information domain and business domain: in the physical equipment of the power data center, as the equipment is used and aged, the performance of the equipment will gradually decline, and the probability of failure will increase. The disordered process of equipment performance is the entropy increase process in the physical domain; in the communication network, as the data flow increases and the network equipment ages, the chaos of data transmission (such as packet loss rate, delay, etc.) will increase. The disordered process of information transmission is the entropy increase process in the information domain; in the business system, as the business load increases and the business process becomes more complex, the load demand and power balance of the system will become difficult to control. The disordered process of business operation is the entropy increase process in the business domain; monitor the physical domain entropy increase, set sensors on the physical equipment to monitor the physical equipment status in real time, and transmit the monitored data to the data center monitoring platform through a wireless way; monitor the information domain, set network monitors on the network equipment to monitor network bandwidth, delay, packet loss rate and other indicators in real time, use the network analysis tool Wireshark to collect network monitoring data in real time, and evaluate the entropy increase of the information domain; set load monitors on the business system server to monitor the changes of load demand, power balance and other business parameters in real time, and identify the entropy increase of the business domain.

[0024] S103, constructing a cross-domain entropy flow model according to the monitoring data of the physical domain, the information domain and the business domain; Further, the monitoring data of the physical domain, the information domain and the business domain are integrated through a data warehouse, the data warehouse includes a data source layer, a data storage layer, a data processing layer and a data access layer, the data source layer contains various data sources of the physical domain, the information domain and the business domain, the data storage layer uses a relational database as the storage of the data warehouse, sets the table structure of the data warehouse, the data processing layer is used for designing the data cleaning, conversion and loading (ETL) process, and the data access layer is used for providing a data access interface, cleaning the integrated data, calculating the correlation coefficients of the physical domain and the information domain, the physical domain and the business domain and the information domain and the business domain by using a Pearson correlation coefficient, integrating the calculated correlation coefficients into a matrix, the rows and columns of the matrix respectively represent different domains, and the elements in the matrix represent the correlation coefficients between the corresponding domains, according to actual requirements, setting a threshold value of the correlation coefficient, extracting domain pairs with correlation coefficients exceeding the threshold value from the correlation matrix, and the entropy increase between the domain pairs has a significant correlation, and the domain pairs with the correlation coefficients exceeding the threshold value are used as key coupling factors of the cross-domain entropy flow; time series data of the entropy increase of each domain is extracted from the data warehouse, the extracted time series data is trained by using a Bayesian network, a directed acyclic graph and a conditional probability table are obtained according to the training result, a causal relationship model is formed, the constructed causal relationship model is verified by using historical data, and the generalization ability of the model is evaluated by cross-validation.

[0025] According to the matrix and the causal relationship model, an associated path and a causal chain of the entropy increase between the physical domain, the information domain and the business domain are identified, nodes of the physical domain, the information domain and the business domain are respectively created, domain nodes with significant correlation are connected by dashed lines according to the coefficients, and positive correlation or negative correlation is marked on the lines, domain nodes with a causal relationship are connected by solid lines with arrows according to the causal relationship model, the direction of the arrow represents the causal direction, a concept diagram of the cross-domain entropy flow is formed, the flow direction and the coupling points of the entropy between different domains are displayed, the coupling points are intersections of the associated paths or places with significant interaction, the coupling points are intersection areas of the entropy increase between multiple domains, a cross-domain entropy flow model is constructed by using a differential equation model, the model is verified by using historical data, the prediction result of the model is compared with actual data, the value of a verification index is calculated, a preliminary framework of the model is built by using a modeling tool AnyLogic, the coupling relationship between the domains is described in the model framework, including direct coupling and indirect coupling, The coupling strength is represented by using parameters or weights, the entropy increase data of the physical domain, the information domain and the business domain are used as the input of the cross-domain entropy flow model, and the output of the cross-domain entropy flow model is the state of the cross-domain entropy flow and the entropy values of the domains.

[0026] S104, entropy increase trends are predicted according to the monitoring data of the physical domain, the information domain and the business domain. Specifically, the monitoring data of the physical domain, information domain and business domain is extracted from the data warehouse, the monitoring data is entropy increase data, a time series analysis tool Statsmodels library is used to identify long-term trends and short-term fluctuations of the entropy increase, a Fourier transform is used to convert the time series data from the time domain to the frequency domain to obtain a frequency spectrum, in the frequency spectrum, the frequency axis represents the inverse of the period, and the amplitude axis represents the strength of the corresponding frequency component, the frequency components with larger amplitudes are found, the periods corresponding to the frequency components are the periods of the entropy increase data, the time series data is divided into a training set and a test set, a long short-term memory (LSTM) model is trained using the training set, model parameters are adjusted to optimize the prediction performance, the trained model is used to predict the test set, errors between the predicted values and the actual values are calculated, the model is adjusted according to the errors, the trained long short-term memory (LSTM) model is used to predict the entropy increase, an entropy increase trend chart is generated, and an algorithm is used to identify an entropy increase inflection point, which marks the transition of the system from a steady state to a latent instability state. In the power data center, as the entropy increase of the physical domain, information domain and business domain accumulates, the system will gradually deviate from its initial steady state, and when the entropy increase reaches a certain level, i.e., reaches the entropy increase inflection point, the system will face the risk of latent instability. The algorithm combines first derivative, second derivative and periodic verification to identify the inflection point, and the time series data of the entropy increase is S={ , the first derivative of the entropy increase curve is calculated using the central difference method, and the formula is: , wherein represents the first derivative of the entropy increase curve, represents the entropy increase value at time point , and represents the entropy increase value at time point , and represents the time interval. The first derivative reflects the instantaneous change rate of the entropy increase. The first derivative is input into the central difference method to calculate the second derivative of the entropy increase curve. For the first derivative sequence, the extreme points are found by traversing the first derivative sequence. The extreme points include maximum points and minimum points, and the maximum points satisfy and , or the minimum points satisfy and . The extreme points correspond to the candidate positions of the inflection points of the entropy increase curve. The second derivative sequence is traversed to find the sign change points, and the sign change points satisfy , the sign change point corresponds to the change of the concave-convex of the entropy increase curve, according to the period of the above entropy increase data, for each inflection point candidate position, check whether it is consistent with the known periodic change, if the inflection point candidate position is not consistent with the periodic analysis result, the inflection point is corrected or eliminated, otherwise, the inflection point candidate position is the inflection point position, according to the historical data and expert experience, set the threshold value of entropy increase, when the entropy increase data monitored exceeds the early warning threshold, trigger the early warning information generation mechanism, the early warning information should include the occurrence time, position, predicted entropy increase amplitude and other key information of the entropy increase inflection point.

[0027] In S105, the entropy increase is adjusted according to the cross-domain entropy flow model, and a situation model is established according to the system state. Further, the cross-domain entropy flow model describes the flow rule and mutual influence relationship of entropy between different domains, according to the real-time monitored entropy increase data and entropy increase trend, the scheduling is dynamically adjusted, when the entropy increase of a certain domain suddenly rises and the trend analysis shows that the entropy increase will continue to rise, the device running parameters or network communication path of the domain are immediately adjusted to suppress the entropy increase, at the same time, the adjusted strategy is fed back to the cross-domain entropy flow model to update the model parameters and improve the accuracy of the model; the state of the system is monitored in real time, combined with the historical data of the system, the patterns and trends in the historical data are identified using time series analysis, the real-time monitoring data and the historical data are fused, and a situation model is constructed using a dynamic Bayesian network method, when the model predicts that the device state of a certain domain will reach the failure threshold, the standby device is started in advance to ensure the continuous operation of the system; when the model predicts that the network traffic will be congested, the communication path is adjusted to optimize the network resource configuration, according to the severity and influence range of the anomaly, the priority of the anomaly suppression strategy is set. Ensure that when multiple anomalies occur at the same time, the anomaly with the greatest impact on the system can be processed first.

[0028] The technical solutions in the embodiments of the present application have at least the following technical effects or advantages: by real-time data acquisition and analysis, anomalies are found and located in time; by three-dimensional atlas and root cause decoupling analysis, the root cause and propagation path of the anomaly are accurately identified, by dynamic guidance function, the repair efficiency of the operation and maintenance personnel is improved, the real-time positioning and root cause decoupling of concurrent anomalies in the power data center are realized, and the operation stability and reliability of the power data center are improved; by cross-domain entropy increase monitoring, entropy flow coupling analysis, entropy increase inflection point prediction and cross-domain entropy flow scheduling, the balance of cross-domain entropy flow is realized, by the situation model and the anomaly suppression strategy, the occurrence and propagation of anomalies are found and suppressed in advance, the cross-domain entropy balance of the power data center is realized, and the steady-state operation of the system is maintained, the system is actively intervened before implicit instability, and the occurrence of steady-state collapse is prevented, by identifying the cross-domain entropy increase of the physical domain, information domain and business domain, comprehensive system monitoring and anomaly suppression are realized.

[0029] Embodiment two: based on the above embodiment one, this embodiment overheat flow field modeling, thermal anomaly vortex identification, dynamic heat dissipation strategy formulation and heat dissipation evaluation verification, realize the efficient heat dissipation of power equipment cluster, ensure the stable operation of power data center, such as Figure 2

[0030] S201, obtaining the temperature of the physical device, forming a heat vector diagram according to the device surface temperature and space information; Specifically, the heat power of the power equipment is obtained in the physical domain, the heat power is obtained through the operation log of the equipment, the space information is obtained according to the physical layout of the power data center, the environmental temperature monitoring point is set in the power data center, the environmental temperature is monitored, the heat flow field modeling system is built, the obtained equipment heat power and equipment space information are input into the heat flow field modeling system, the optical fiber temperature sensor is installed on the surface of the equipment, the surface temperature of the equipment is monitored in real time, and the data is transmitted to the heat flow field modeling system, the difference between the surface temperature of the equipment and the environmental temperature is calculated, the convection rate of natural convection field is calculated through the difference, and the calculation formula of the convection rate is: wherein, represents the natural convection rate of air, reflecting the speed of air flow in natural convection state, is an empirical coefficient, which is obtained through experiment, represents the surface temperature of the equipment, which is the actual temperature reached by the surface of the equipment in the running process, represents the environmental temperature, which is the air temperature of the environment around the equipment in the power data center, represents the index, which represents the influence degree of the difference between the surface temperature of the equipment and the environmental temperature on the natural convection rate of air, usually takes 1 / 3 or 1 / 4, and 1 / 3 is taken in this embodiment.

[0031] According to the convection rate and the equipment space information, the heat conduction and convection heat transfer model is used to calculate the heat conduction and convection heat transfer, the heat conduction and convection heat transfer model is iterated continuously, the heat diffusion path is formed, the vector data is generated according to the heat diffusion path, the size of the vector represents the amount of heat, which is usually related to the intensity of heat diffusion, the direction of the vector represents the diffusion direction of the heat, which is consistent with the flow direction of the air, the generated vector data is drawn in the form of vector diagram by using drawing software, in the vector diagram, each vector is represented by an arrow, the length of the arrow represents the size of the vector, and the direction of the arrow represents the direction of the vector, in the vector diagram, the area with small vector and disordered direction is identified, the heat diffusion in these areas is hindered, forming heat retention, and the identified heat retention area is marked on the heat flow vector diagram.

[0032] S202, calculating the thermal anomaly index according to the surface temperature of the physical equipment, and identifying the thermal anomaly vortex area using the thermal anomaly index; ​Further, the temperature rise rate is calculated according to the device surface temperature measured in step S201, and the formula is: wherein, represents the temperature rise rate, represents the end temperature of the device surface, represents the initial temperature of the device surface, represents the time interval, the air flow stagnation index is calculated by setting the air flow rate sensor around the device, and the formula is: wherein, represents the air flow stagnation index, represents the duration of the time window in which the peripheral air flow rate is less than 0.1 m / s, represents the total time, the thermal anomaly index is calculated according to the temperature rise rate and the air flow stagnation index, and the formula is: wherein, is the thermal anomaly index, is the temperature rise rate, is the air flow stagnation index; a threshold value is set according to the heat dissipation characteristics of the power device, and based on the calculated thermal anomaly index, when the thermal anomaly index is greater than the set threshold value for a continuous period of time, the area is a thermal anomaly vortex area. The thermal anomaly vortex area is used to identify the area where heat is abnormally accumulated between the device cluster. These areas are prone to local overheating due to air flow organization failure, affecting the normal operation of the device. The thermal anomaly vortex area is marked on the thermal flow vector diagram, and the position, size and severity of the thermal anomaly vortex area are recorded.

[0033] S203, according to the thermal anomaly vortex area and the device surface temperature, the device is cooled; Specifically, according to the type, material and operating condition of the device, a safety threshold is set, when the thermal anomaly vortex area is formed or the device surface temperature approaches the safety threshold, it is determined that the device needs to be cooled, according to the severity of the thermal anomaly vortex area, the severity is divided into three levels: mild, moderate and severe. For the thermal anomaly vortex area with mild severity, plan short-time cooling (10 minutes), for the thermal anomaly vortex area with moderate severity, plan long-time cooling (30-60 minutes), during the cooling process, the changes of the device surface temperature and the thermal anomaly vortex area are monitored in real time. If the device surface temperature drops quickly or the thermal anomaly index of the thermal anomaly vortex area decreases significantly, the cooling time can be shortened appropriately; on the contrary, if the temperature drops slowly or the thermal anomaly index changes is not obvious, the cooling time is appropriately extended. Use the fan to cool, adjust the device arrangement density, optimize the air flow channel and other means to cool.

[0034] The technical solutions in the embodiments of the present application have at least the following technical effects or advantages: through the steps of heat vector diagram, thermal anomaly vortex identification, dynamic heat dissipation strategy formulation, and heat dissipation evaluation verification, precise, dynamic, and efficient heat dissipation management is achieved. The final result is that the surface temperature of the power equipment is effectively controlled, the thermal anomaly vortex area is significantly reduced or disappears, the stability and reliability of the power data center are significantly improved, the problem of abnormal accumulation of heat in the power data center between equipment clusters is solved, and efficient heat dissipation and stable operation of the system are ensured.

[0035] Embodiment three: on the basis of embodiment two, this embodiment solves the problem caused by the thermal hysteresis effect of metal fatigue accumulation by dynamically modeling the heat flow field and thermal anomaly vortex, ensures the stable operation of the power data center and the safety of the equipment, as shown in Figure 3 .

[0036] S301, monitor the internal changes of the physical equipment, and identify the mispositioning accumulation of the metal connectors of the physical equipment; Further, the internal changes of the equipment material are micro changes, the mispositioning accumulation of the metal connectors is identified by using X-ray diffraction (XRD) technology. If the diffraction peak of the metal connector is offset and the intensity of the peak is reduced, it indicates that the metal connector has mispositioning accumulation. The mispositioning accumulation of the metal connector increases the scattering probability of electrons in the metal, resulting in an increase in resistivity. When electrons move in the crystal, they will collide with defects such as mispositioning, hindering the free flow of electrons, thereby increasing the resistance. The resistance of the metal connector is monitored in real time by using a four-wire resistance measuring instrument. By measuring the change of the resistance value, in combination with the size and temperature of the connector and other parameters, the change of the resistivity is calculated. When the resistivity exceeds the threshold value, the connector has mispositioning accumulation. The mispositioning accumulation of the metal connector is formed by the thermal hysteresis effect. The thermal hysteresis effect describes the phenomenon that the thermal response of a substance lags behind the temperature change when the temperature changes. In the metal connector, when the temperature changes, due to the thermal hysteresis effect, the motion state of the atoms inside the metal will not immediately adjust to the new stable state with the change of the temperature, resulting in mispositioning accumulation of the metal connector.

[0037] S302, correct the heat vector diagram and the thermal anomaly vortex area according to the mispositioning accumulation of the metal connector; Specifically, the misalignment accumulation of the metal connecting piece will cause friction heat, resulting in a heating power including the heating power during normal operation of the equipment and the power of the friction heat, so that the total heating power increases, and the thermal resistance also increases. The corrected heating power and thermal resistance parameters are input into the numerical simulation software ANSYS Fluent. In the simulation, the misalignment accumulation area is set as an obstacle by setting the boundary conditions in the misalignment accumulation area, increasing the local resistance coefficient, simulating the situation that the air flow is blocked to form a vortex, and the numerical simulation software ANSYS Fluent outputs the natural convection field distribution. The natural convection field is calculated by using the heat flow calculation module in the numerical simulation software, the heat flow distribution of the connecting piece inside and surface is obtained by solving the coupled equation set of heat conduction equation and heat convection equation, the change trend of the heat flow vector is analyzed, the heat flow vector distribution diagram is drawn, and the heat flow concentration area and the heat flow weak area are found out by observing the size and direction of the heat flow vector. The heat flow concentration area usually corresponds to the high temperature area of the connecting piece or the vicinity of the heat source, while the heat flow weak area may appear in the parts with large thermal resistance or poor air flow.

[0038] S303, setting a heat dissipation mode according to the misalignment accumulation of the metal connecting piece; Further, by monitoring the temperature change rate, pressure change rate or heat flow change rate, when the monitored temperature change rate, pressure change rate or heat flow change rate exceeds the preset threshold value, it is determined that the thermal hysteresis effect causes the metal connecting piece to form misalignment accumulation. If the local heat flow direction of a region is disordered, the temperature gradient is large, and the heat transfer is blocked, the region is a thermal abnormal vortex area. When the metal connecting piece forms misalignment accumulation and the thermal abnormal vortex area is detected, heat dissipation is needed. The severity of the misalignment accumulation of the metal connecting piece is identified, the area, volume and distance of the thermal abnormal vortex area from the key part of the connecting piece are measured, and the heat dissipation time is planned according to the severity of the misalignment accumulation of the metal connecting piece and the characteristics of the thermal abnormal vortex area. For the equipment with serious misalignment accumulation of the metal connecting piece or the case with large thermal abnormal vortex area, the heat dissipation time is increased, for example, if the comprehensive score of the misalignment accumulation of the metal connecting piece is high or the vortex area is large, the heat dissipation time is set to 1.5-2 times of that in normal case. The heat dissipation intensity is measured by parameters such as power, wind speed or water flow speed of the heat dissipation equipment. According to the severity of the misalignment accumulation of the metal connecting piece and the characteristics of the thermal abnormal vortex area, the parameters of the heat dissipation equipment are adjusted to increase the heat dissipation intensity, for example, for the area with serious misalignment accumulation of the metal connecting piece, the speed of the fan or the water flow speed of the water cooling system can be increased.

[0039] The technical solutions in the embodiments of the present application have at least the following technical effects or advantages: by monitoring the internal changes of the equipment material, identifying the mispositioned accumulation of the equipment metal connecting piece, combining with the heat dissipation optimization algorithm, more accurate and efficient heat dissipation management is realized, the stability and heat dissipation efficiency of the power data center are improved, the heat dissipation energy consumption is reduced, the overall energy efficiency is improved, high-precision monitoring of the mispositioned accumulation of the equipment metal connecting piece is realized, intelligent formulation and dynamic adjustment of the heat dissipation strategy are realized, the heat dissipation energy consumption is reduced, and the overall energy efficiency is improved.

[0040] Embodiment four: the present embodiment provides an abnormal power data management system, including a heat flow field modeling system, a thermal anomaly vortex identification system and a dynamic heat dissipation strategy formulation system, the heat flow field modeling system includes a data input interface, a model calculation module and a result output module, the data input interface receives various data from the power equipment monitoring system, including temperature data, current data, voltage data and equipment structure parameters of the power equipment, the model calculation module is used for numerical calculation, and the result output module is used for outputting the calculation result of the calculation module and generating a heat vector diagram; the thermal anomaly vortex identification system integrates a thermal anomaly index calculation module and a black hole determination rule, the thermal anomaly index calculation module is used for calculating the thermal anomaly index, and the black hole determination rule is a set of judgment standard formulated based on the heat entropy theory and practical experience, and is used for identifying whether there is a thermal anomaly vortex area in the power equipment; the dynamic heat dissipation strategy formulation system includes a heat dissipation demand judgment module, a heat dissipation time planning module and a heat dissipation means selection module, the heat dissipation demand judgment module is used for dynamically judging the heat dissipation demand of the power equipment according to the determination result of the thermal anomaly vortex identification system, the heat dissipation time planning module is used for planning the time arrangement of heat dissipation according to the result of the heat dissipation demand judgment module, and the heat dissipation means selection module is used for selecting the heat dissipation means according to the heat dissipation demand and the heat dissipation time planning, the heat flow field modeling system is electrically connected with the thermal anomaly vortex identification system, and the thermal anomaly vortex identification system is electrically connected with the dynamic heat dissipation strategy formulation system.

[0041] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for managing abnormal power data, characterized in that: include: S101, collecting power data, generating a spatial location map, identifying abnormal data based on the power data, identifying the cause of the abnormality based on the spatial location map and the abnormal data, and repairing the abnormal data; S102, using a monitor to monitor entropy increase in multiple domains, where the multiple domains include a physical domain, an information domain, and a business domain, and the physical domain is a physical device in a power data center; S103, constructing a cross-domain entropy flow model based on monitoring data of the physical domain, information domain, and business domain; S104, predicting an entropy increase trend based on monitoring data of the physical domain, information domain, and business domain; S105: Adjust the entropy increase according to the cross-domain entropy flow model, and establish a situation model according to the power system state.

2. The abnormal power data management method according to claim 1, characterized in that: The multi-domain entropy increase refers to the process in which systems in multiple domains transform from an ordered state to a disordered state, that is, the process in which the entropy value of the system continuously increases.

3. The abnormal power data management method according to claim 1, characterized in that: An entropy increase trend graph is generated according to the predicted entropy increase trend. An algorithm is used to identify an entropy increase inflection point based on the entropy increase trend graph, where the entropy increase inflection point marks a transition of the system from a steady state to a latent unstable state.

4. The abnormal power data management method according to claim 3, characterized in that: The central difference method is used to calculate the first-order derivative and second-order derivative of the entropy increase curve. The entropy increase inflection point is identified based on the first-order derivative and the second-order derivative. The calculation formula of the central difference method is: ,in, represents the first-order derivative of the entropy increase curve, Indicates a time point The entropy increase, Indicates a time point The entropy increase, Indicates a time interval.

5. The abnormal power data management method according to claim 1, characterized in that: S201, obtaining the temperature of a physical device and forming a heat vector diagram based on the device surface temperature and spatial information; S202, calculating a thermal anomaly index based on the surface temperature of the physical device, and using the thermal anomaly index to identify a thermal anomaly vortex area; S203, dissipating heat from the device according to the abnormal thermal vortex area and the device surface temperature.

6. The abnormal power data management method according to claim 5, characterized in that: The convection rate is calculated based on the surface temperature of the physical equipment using the formula: ,in, represents the natural convection rate of air, is the empirical coefficient, Indicates the surface temperature of the device. Indicates the ambient temperature, Represents an exponent.

7. The abnormal power data management method according to claim 5, characterized in that: The abnormal thermal vortex area is used to identify an area where heat accumulates abnormally between equipment clusters.

8. The abnormal power data management method according to claim 6, characterized in that: S301, monitoring internal changes of the physical device and identifying misalignment and accumulation of metal connectors of the physical device; S302, correcting the heat vector diagram and the abnormal thermal vortex area according to the misalignment and accumulation of metal connectors; S303: Setting a heat dissipation method according to the staggered stacking of the metal connectors.

9. The abnormal power data management method according to claim 8, characterized in that: Misalignment accumulation in metal connectors is caused by thermal hysteresis, which describes the phenomenon that the thermal response of a material lags behind the temperature change.

10. An abnormal power data management system, applied to the abnormal power data management method according to any one of claims 1 to 9, characterized in that: It includes a thermal flow field modeling system, a thermal anomaly vortex identification system and a dynamic heat dissipation strategy formulation system. The thermal flow field modeling system is used to generate a heat vector diagram, the thermal anomaly vortex identification system is used to calculate the thermal anomaly index and identify thermal anomaly vortices, and the dynamic heat dissipation strategy formulation system is used to judge the heat dissipation requirements of power equipment. The thermal flow field modeling system is electrically connected to the thermal anomaly vortex identification system, and the thermal anomaly vortex identification system is electrically connected to the dynamic heat dissipation strategy formulation system.

Images