An energy station intelligent operation and maintenance monitoring and management platform

Abstract

The application discloses an energy station intelligent operation and maintenance monitoring and management platform, and belongs to the technical field of energy station management systems; aims to solve the problems of equipment association ambiguity, collection strategy solidification, fault root cause positioning difficulty and disposal inefficiency in existing operation and maintenance; an equipment association modeling module integrates multidimensional data such as energy transmission and signal interaction, calculates comprehensive coupling degree and divides grades, and generates a core equipment coupling relationship matrix; a collection strategy optimization module dynamically adjusts the collection frequency of itself and associated equipment according to the equipment importance and operating state, and generates a time sequence associated data set; a fault diagnosis and root cause tracing module constructs a three-dimensional fault conduction topology graph, accurately determines faults, quantifies grades, locates root causes, outputs a diagnosis report and carries out operation and maintenance according to layered logic; the application realizes equipment association quantification, intelligent optimization of collection strategies and efficient disposal of the whole fault link, improves the intelligentization and refinement level of energy station operation and maintenance, and guarantees stable operation of equipment.

Description

Technical Field

[0001] This invention belongs to the technical field of energy station management systems, specifically relating to an intelligent operation and maintenance monitoring and management platform for energy stations. Background Technology

[0002] With the rapid development of the new energy industry, energy stations, as core hubs for energy conversion, storage, and transmission, are experiencing continuous expansion in equipment scale and an increasing variety of core equipment. The coupling relationships between equipment in energy transmission and signal interaction are becoming increasingly complex. Reliable operation, maintenance, monitoring, and management are crucial for ensuring the stable operation of energy stations and improving energy utilization efficiency. However, existing energy station operation and maintenance technologies still have significant limitations:

[0003] The lack of a multi-dimensional correlation and quantification mechanism that integrates equipment energy transmission, signal interaction, and anomaly propagation makes it impossible to accurately characterize equipment interaction characteristics, resulting in a lack of scientific basis for operation and maintenance decisions and difficulty in avoiding the risk of fault propagation.

[0004] The lack of device association features, real-time operating status and collection strategy linkage mechanism, fixed collection frequency, and insufficient data relevance and time sequence consistency not only waste resources, but also make it difficult to support accurate fault diagnosis.

[0005] Meanwhile, the lack of integration of equipment correlation and transmission characteristics to construct a complete fault chain system makes it difficult to locate the root cause of the fault, the classification of fault levels is vague, and the allocation of operation and maintenance resources is unbalanced, making it impossible to achieve rapid fault blocking and efficient handling. These problems seriously restrict the level of intelligence and precision of energy station operation and maintenance. To address this, we propose an intelligent operation and maintenance monitoring and management platform for energy stations. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent operation and maintenance monitoring and management platform for energy stations to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an intelligent operation and maintenance monitoring and management platform for energy stations, comprising:

[0008] Equipment Association Modeling Module: Mark core monitoring equipment, acquire coupling raw data, construct ordered pairs of equipment, analyze their energy transmission factors, signal factors and common anomaly factors and perform comprehensive processing to obtain the comprehensive coupling degree, determine the coupling level, and generate the core equipment coupling relationship matrix;

[0009] Data Acquisition Strategy Optimization Module: Determines importance based on the historical failure frequency and energy transmission intensity of core equipment, and matches the corresponding benchmark acquisition frequency; calculates operational evaluation indicators based on the percentage of abnormal operating parameters, and adjusts the acquisition frequency of the equipment itself and related equipment in combination with coupling relationships; collects data at the adjusted frequency, and generates a time-series associated dataset bound to the equipment identifier and a unified timestamp.

[0010] Fault diagnosis and root cause tracing module: Constructs a three-dimensional fault propagation topology map, determines equipment faults and classifies fault levels based on time-series correlation datasets, locates the root cause equipment and type of fault, integrates core diagnostic information to form a fault diagnosis report and sends it to the operation and maintenance scheduling terminal for hierarchical operation and maintenance.

[0011] Preferably, the specific process for constructing ordered pairs of devices is as follows:

[0012] Identify the operating equipment in the energy station that requires maintenance and monitoring, mark them as core equipment, and organize them into a core equipment set; obtain the coupled raw dataset for each core equipment, which includes:

[0013] Energy transmission topology file, signal linkage configuration file, rated transmission parameter matrix, and historical operation log;

[0014] Energy transfer topology file: records the energy transfer path between devices in the form of a quadruple;

[0015] Signal linkage configuration file: records the correlation of control or feedback signals between devices in the form of a quad tuple;

[0016] Rated transmission parameter matrix: Characterizes the rated transmission capacity or power between devices; the corresponding value is 0 when there is no transmission path.

[0017] Historical operation log: records the number of simultaneous anomalies between devices and the sequence of anomalies;

[0018] After data acquisition is completed, all core devices in the core device set are paired to obtain ordered pairs of devices. When pairing, combinations of the same device with itself are excluded.

[0019] Preferably, the specific process of comprehensive coupling degree analysis is as follows:

[0020] For each ordered pair of devices, three types of coupled judgment factors are constructed: energy transfer factor, signal factor, and common anomaly factor.

[0021] The energy transfer factor is determined based on the rated inter-device transfer data in the rated transfer parameter matrix and the maximum rated total transfer amount from the device to downstream devices. When there is no transfer path, the factor is 0.

[0022] The signal factor is determined based on whether there are control or feedback signals between devices;

[0023] The common anomaly factor is calculated based on the number of simultaneous anomalies and the number of sequential anomalies between equipment.

[0024] The three types of factors are weighted and summed according to preset weight coefficients to obtain the overall coupling degree of the ordered pairs of the device.

[0025] Preferably, the process for generating the core device coupling relationship matrix is ​​as follows:

[0026] Preset high coupling threshold and low coupling threshold, and determine the coupling level of ordered pairs of devices by comparing the comprehensive coupling degree with the two types of thresholds;

[0027] A combined coupling degree greater than the high coupling threshold is considered strong coupling, between the high and low coupling thresholds is considered medium coupling, and less than the low coupling threshold is considered no coupling.

[0028] Based on the coupling levels of all ordered pairs of devices, a matrix is ​​constructed with dimensions corresponding to the total number of core devices. Matrix elements are assigned values ​​according to their coupling levels: strong coupling is assigned a value of 1, medium coupling is assigned a value of 0.5, and no coupling is assigned a value of 0. The diagonal elements of the matrix are uniformly set to 0, thus obtaining the core device coupling relationship matrix.

[0029] Preferably, the specific process of determining importance based on the historical failure frequency and energy transmission intensity of core equipment and matching the corresponding reference acquisition frequency is as follows:

[0030] The historical failure frequency and energy transmission intensity of each core device are obtained and comprehensively analyzed to obtain the device importance assessment value;

[0031] Set several equipment importance levels, and preset a unique equipment importance evaluation value range and a unique benchmark collection frequency range for each level;

[0032] The importance assessment value of the core equipment is matched with the assessment value range corresponding to each level, and the benchmark acquisition frequency range corresponding to the core equipment is output.

[0033] Preferably, the specific process for adjusting the acquisition frequency of the device itself and associated devices is as follows:

[0034] Acquire the operating parameter data within the preset window of the core equipment, preset the normal operating range of each parameter, and count the percentage of times the parameter deviates from the range as the parameter anomaly percentage; calculate the operating evaluation index by combining the preset weight of each operating parameter.

[0035] Compare the operational evaluation indicators with the preset benchmark values: if the indicator is less than the benchmark value, it is determined that there is an abnormal trend. The sampling frequency of the device is adjusted to the upper limit of the benchmark range, and non-uncoupled related devices are extracted. Strongly coupled devices are adjusted to the upper limit of the benchmark range, and moderately coupled devices are adjusted to the median value. If the indicator meets the standard, the sampling frequency of the device is lowered to the lower limit of the benchmark range.

[0036] Preferably, the process of forming a time-series associated dataset is as follows:

[0037] The system monitors the operational evaluation indicators of each core device and the status of the core device coupling relationship matrix in real time. Based on the current acquisition frequency of each core device, it collects multimodal data of each core device. It uses a unified time synchronization mechanism to synchronize the timestamps of all acquisition nodes and binds each piece of acquired data with the corresponding device identifier and unified timestamp to form a time-series associated dataset.

[0038] Preferably, the process of constructing a three-dimensional fault propagation topology map and determining equipment faults based on a time-series correlation dataset is as follows:

[0039] Construct a historical fault case library, an equipment parameter threshold library, and an equipment physical connection map, respectively recording fault-related information, parameter fault thresholds, and equipment entity connection relationships;

[0040] Based on core equipment, its physical attributes and parameter fault thresholds are integrated to form topology nodes, and the equipment identification, type and key parameter thresholds are labeled; when the conditions of non-uncoupled, physical connection and effective energy transmission channel are met, the associated links are constructed, the link types are distinguished according to the coupling level, the link propagation weight is calculated by combining the coupling degree and the similarity of historical common faults, and a three-dimensional fault propagation topology map is integrated.

[0041] For each core device, the corresponding acquisition time and parameter value are extracted from the time-series correlation data. The abnormal intensity factor is calculated by matching the normal baseline value and fault threshold of the device. Combined with the historical fault type feature weight of the device type, the fault matching factor is obtained.

[0042] When the abnormal intensity factor of any parameter of the core device is ≥1 and the duration of this state exceeds the preset fault duration threshold, the core device is determined to have failed.

[0043] Preferably, the specific process for classifying fault levels and locating the root cause equipment and type of fault is as follows:

[0044] For the core equipment that has failed, the total number of related equipment with failure or abnormal trend is counted, the average value of the abnormal intensity factor during the failure duration is calculated, and the two are normalized and weighted according to the preset weight to obtain the fault level quantification value. The fault level is then matched with the preset level range to output the fault level.

[0045] In the three-dimensional fault propagation topology map, valid fault propagation paths with propagation weights reaching a preset threshold are retained, and the device with the first anomaly and the earliest timestamp is selected as the root cause device of the fault.

[0046] A fault anomaly determination window is set up with the first time the root cause device becomes abnormal. All historical fault types of its type are extracted, the comprehensive matching factor of each type in the window is calculated, and the historical fault type with the largest value is selected as the final root cause type.

[0047] Preferably, the specific process of generating a fault diagnosis report and sending it to the operation and maintenance dispatch terminal for tiered operation and maintenance is as follows:

[0048] The core diagnostic information, including fault device identifier, fault level, abnormal feature sequence, effective fault propagation path, root cause device and fault type, is integrated into a fault diagnosis report according to a preset standardized format and sent to the operation and maintenance dispatch terminal.

[0049] Based on this report, the operation and maintenance dispatch terminal combines the core equipment coupling relationship matrix, the handling results of the corresponding root cause type in the historical fault case library, and the operating parameter standards in the equipment parameter threshold library. It follows the hierarchical operation and maintenance logic of prioritizing the handling of the root cause equipment, repairing the faulty equipment itself, and focusing on the prevention and control of related equipment: prioritizing the matching of standard solutions to repair the root cause equipment, repairing the faulty equipment according to the fault level requirements to bring its parameters back to the normal range, and implementing targeted monitoring of related equipment on the effective fault propagation path according to the coupling level.

[0050] Compared with the prior art, the beneficial effects of the present invention are:

[0051] (1) This intelligent operation and maintenance monitoring and management platform for energy stations integrates raw data such as energy transmission topology files, signal linkage configuration files, and historical operation logs through the equipment association modeling module. It constructs three types of coupling judgment factors: energy transmission, signal, and common anomalies. After weighted calculation, the comprehensive coupling degree is obtained and divided into three coupling levels: strong, medium, and none. Finally, a core equipment coupling relationship matrix is ​​generated. This matrix clearly presents the interaction strength between equipment, effectively solving the problem of ambiguous equipment association in traditional operation and maintenance. It provides accurate data support for subsequent acquisition strategy optimization and fault propagation prediction from the source, reducing the risk of fault spread.

[0052] (2) The intelligent operation and maintenance monitoring and management platform for energy stations first calculates the importance of core equipment based on historical fault frequency and energy transmission intensity through the data acquisition strategy optimization module, and matches the differentiated benchmark acquisition frequency; then, it calculates the evaluation index through the abnormal proportion of operating parameters, and adjusts the acquisition frequency of abnormal trend equipment and its strongly / medium coupled related equipment in a differentiated manner based on the coupling relationship matrix; finally, it uses a unified time synchronization mechanism to bind equipment identifiers and timestamps to generate time-series associated datasets. This process avoids the waste of resources from excessive acquisition of non-critical equipment, and ensures the full acquisition of core data of critical equipment, providing high-quality multimodal data with consistent time sequence and strong targeting for fault diagnosis.

[0053] (3) This intelligent operation and maintenance monitoring and management platform for energy stations first integrates data such as coupling relationship matrix and physical connection map through the fault diagnosis and root cause tracing module to construct a three-dimensional fault transmission topology map containing equipment nodes, associated links and transmission weights; then, based on the time-series associated dataset, it accurately determines the fault through the abnormal intensity factor and fault matching factor; then, it quantifies the fault level, filters the root cause equipment with the earliest timestamp through the effective transmission path, and determines the root cause type by combining the historical fault case library; finally, it performs hierarchical operation and maintenance according to the logic of "root cause priority, fault must be repaired, and associated prevention and control"; this process solves the problems of difficult fault root cause location and ambiguous level, greatly shortens the investigation time, optimizes the allocation of operation and maintenance resources, and realizes rapid fault blocking and efficient handling. Attached Figure Description

[0054] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Example: Please refer to Figure 1 This invention provides an intelligent operation and maintenance monitoring and management platform for energy stations, including: an equipment association modeling module, a data acquisition strategy optimization module, and a fault diagnosis and tracing module;

[0057] Equipment Association Modeling Module: This module marks core monitoring equipment, acquires raw coupling data, constructs ordered pairs of equipment, analyzes their energy transfer factors, signal factors, and common anomaly factors, performs comprehensive processing to obtain the comprehensive coupling degree, determines the coupling level, and generates a core equipment coupling relationship matrix. The specific process is as follows:

[0058] Identify the operating equipment in the energy station that requires operation and maintenance monitoring, and mark it as core equipment, including: energy storage PCS, inverters, transformers, switch cabinets, energy storage battery packs, etc., and organize them into a core equipment set;

[0059] Wherein, the set of core equipment is denoted as S, a single core equipment is denoted as si, the total number of core equipment is denoted as n, and i is the label of the core equipment i=1,2,…,n;

[0060] For each core device in the core device set, its coupled original dataset is obtained one by one. This set includes:

[0061] Energy transfer topology file: The file is recorded in the form of a quadruple of origin device - destination device - energy type - flow direction, which is used to characterize whether there is an energy transfer path between any two core devices;

[0062] Signal linkage configuration file: The file is recorded in the form of a quadruple of signal sending device - signal receiving device - signal type - response delay, which is used to characterize whether there is a control or feedback signal association between any two core devices.

[0063] The rated transmission parameter matrix P has an n×n dimension. Its matrix element Pij represents the rated capacity or rated power transmitted from core device si to core device sj. If there is no energy transmission path between the two, then Pij=0. i and j are both labels of the core devices.

[0064] Historical operation log: Records the number of times Cij occurs simultaneously between any two core devices si and sj within a preset period, and the number of times Eij occurs after core device si fails within a preset period.

[0065] After obtaining the original coupled dataset as described above, a total order pairwise combination is performed on all core devices in the core device set to obtain several ordered pairs of devices (si, sj); where i ≠ j, and the number of ordered pairs of devices is . indivual;

[0066] For each ordered pair of devices (si, sj), a coupling determination factor is constructed to analyze the coupling level between them, as follows:

[0067] To obtain the elements Pij in the rated transmission parameter matrix P and the rated transmission total Pi of the core device si to all downstream core devices, use the formula: We obtain the energy transfer factor Energy(i,j); where, if Pij=0, then Energy(i,j)=0. for The maximum value in;

[0068] The energy factor characterizes the correlation strength factor of energy transfer between two core devices in an ordered alignment of equipment;

[0069] Obtain the signal association records of core devices si and sj in the signal linkage configuration file, determine whether there is a control or feedback signal sent by core device si to core device sj, and obtain the signal factor Sig(i,j);

[0070] If a control or feedback signal exists, then Sig(i,j)=1; otherwise, Sig(i,j)=0.

[0071] Signal factors characterize the existence factors of signal linkage and correlation between two core devices in the orderly alignment of the equipment;

[0072] To retrieve Cij and Eij from historical execution logs, use the following formula: The common anomaly factor CoAbn(i,j) is obtained; where 1 is added to the denominator to avoid division by zero. ;

[0073] Common anomaly factors characterize the transmission correlation factors of abnormal states between two core devices in an ordered alignment of equipment;

[0074] Obtain the energy factor Energy(i,j), signal factor Sig(i,j), and co-anomaly factor CoAbn(i,j) corresponding to the ordered pair (si, sj) of the device, and use the formula:

[0075] The overall coupling degree Cpl(i,j) is obtained.

[0076] Among them, w1, w2, and w3 are preset weight coefficients;

[0077] Preset high coupling threshold and low coupling threshold;

[0078] If the overall coupling degree corresponding to the ordered pair (si, sj) of devices is greater than the preset high coupling threshold, then the core devices si and sj are determined to be strongly coupled.

[0079] If the overall coupling degree corresponding to the ordered pair of devices (si, sj) is greater than or equal to the preset low coupling threshold and less than or equal to the preset high coupling threshold, then the core devices si and sj are determined to be moderately coupled.

[0080] If the overall coupling degree corresponding to the ordered pair (si, sj) of devices is less than the preset low coupling threshold, then the core devices si and sj are determined to be uncoupled.

[0081] Based on the coupling determination results of ordered pairs of each device, the core device coupling relationship matrix M is obtained by assigning values ​​to matrix elements;

[0082] Matrix dimension is The matrix element Mij is assigned a value according to the coupling determination result: Mij=1 when there is strong coupling, Mij=0.5 when there is medium coupling, and Mij=0 when there is no coupling. The diagonal element Mii of the matrix is ​​uniformly set to 0.

[0083] It should be noted that a coupling relationship system among core devices was constructed through multi-dimensional data integration and quantitative analysis.

[0084] By integrating raw data from multiple dimensions such as energy transmission, signal linkage, and historical anomalies, the comprehensive coupling degree is calculated through three types of coupling judgment factors, and three coupling levels of strong, medium, and none are clearly defined. Finally, the device association status is presented intuitively in matrix form, providing accurate data support for subsequent operation and maintenance.

[0085] The generated core device coupling relationship matrix clarifies the degree of correlation between devices, enabling the acquisition strategy optimization module to adjust the acquisition frequency of associated devices in a targeted manner, avoiding over-acquisition of uncoupled devices or under-acquisition of strongly coupled devices, thereby improving the efficiency and practicality of data acquisition.

[0086] The coupling relationships represented by the core equipment coupling relationship matrix provide a core basis for constructing the associated links of the three-dimensional fault propagation topology map, helping to quickly screen effective fault propagation paths, accurately locate the root cause equipment and propagation range of the fault, reduce fault investigation time, reduce the risk of fault expansion, and ensure the stable operation of energy station equipment.

[0087] The data acquisition strategy optimization module determines the importance of core equipment based on its historical failure frequency and energy transmission intensity, and matches the corresponding baseline acquisition frequency. It calculates operational evaluation indicators based on the percentage of abnormal operating parameters and adjusts the acquisition frequencies of the equipment itself and related equipment based on coupling relationships. Data is collected at the adjusted frequencies to generate a time-series associated dataset bound to the equipment identifier and a unified timestamp. The specific process is as follows:

[0088] For each core device, its historical fault frequency F and energy transmission intensity T are obtained, and after normalization and dimensionless processing, the following formula is used: The equipment importance assessment value Q is obtained, where a1 and a2 are preset weight coefficients;

[0089] Several device importance levels are preset, each level corresponds to a device importance assessment value range, and each device importance level corresponds to a benchmark sampling frequency range.

[0090] By matching the equipment importance assessment value of the core equipment with the assessment value range corresponding to the importance level of all equipment, the corresponding benchmark acquisition frequency range is output.

[0091] Acquire the operating parameter data of the core equipment within the preset window, including voltage, current, power, temperature, vibration parameters, etc.

[0092] Each operating parameter is pre-defined to correspond to a normal operating range. For the operating parameters of each core device, the proportion of the number of times the operating parameter deviates from the corresponding normal operating range within the pre-defined window is counted to the total number of collections, and recorded as the parameter abnormality ratio Rk, where k is the label of the operating parameter.

[0093] To obtain the percentage of abnormal parameters Rk corresponding to all operating parameters of the core equipment, use the formula: We obtain the operation evaluation index G, where m is the total number of operation parameters of the core equipment and Hk is the preset weight corresponding to the k-th operation parameter.

[0094] If the operation evaluation index of the core equipment is less than the corresponding preset benchmark value, it is determined that the core equipment has an abnormal trend, and the collection frequency of the core equipment is adjusted to the upper limit of the corresponding benchmark collection frequency range.

[0095] If the core device exhibits an abnormal trend, the associated device's data acquisition frequency adjustment process will be triggered, specifically as follows:

[0096] Extract all that satisfy The associated core device sj is marked as the device to be adjusted;

[0097] If the device to be adjusted is strongly coupled with the core device that currently exhibits an abnormal trend, then the acquisition frequency of the device to be adjusted should be adjusted to the upper limit of the corresponding reference acquisition frequency range.

[0098] If the device to be adjusted is moderately coupled with the core device that currently exhibits an abnormal trend, then the sampling frequency of the device to be adjusted should be adjusted to the midpoint of the corresponding reference sampling frequency range.

[0099] If the operational evaluation indicators of the core device and all its associated core devices are greater than or equal to the corresponding preset benchmark values, the current acquisition frequency of the core device will be lowered to the lower limit of the benchmark frequency range.

[0100] The system monitors the operational evaluation indicators of each core device and the status of the core device coupling relationship matrix in real time. Based on the current acquisition frequency of each core device, it collects multimodal data corresponding to each core device, including electrical parameters, physical characteristic parameters, and environmental parameters. It also adopts a unified time synchronization mechanism to synchronize the timestamps of all acquisition nodes and binds each piece of acquired data with the corresponding device identifier and unified timestamp to obtain a time-series associated dataset.

[0101] It should be noted that the data acquisition strategy optimization module dynamically optimizes the data acquisition scheme based on the importance and operating status of the equipment, providing high-quality data support for the operation and maintenance of energy stations.

[0102] The importance of core equipment is determined by combining its historical failure frequency and energy transmission intensity, and a differentiated benchmark acquisition frequency is matched to ensure that key equipment receives more monitoring resources. At the same time, the frequency is dynamically adjusted according to abnormal operating parameters, so as not to blindly increase invalid acquisitions or miss key data of abnormal equipment, thereby improving acquisition efficiency.

[0103] Based on the coupling relationship matrix generated by the device association modeling module, when the core device shows an abnormal trend, the collection frequency of strongly coupled and moderately coupled associated devices is adjusted in a targeted manner to achieve early monitoring and tracking of anomalies, reduce the possibility of fault expansion, and buy time for operation and maintenance.

[0104] By binding device identifiers and timestamps through a unified time synchronization mechanism, the temporal consistency and traceability of collected data are ensured, providing standardized and complete multimodal data support for fault diagnosis and root cause tracing modules. This helps to accurately determine faults, locate root causes, and improve the overall intelligence level of operation and maintenance.

[0105] Fault Diagnosis and Root Cause Tracing Module: Constructs a 3D fault propagation topology map, determines equipment faults and classifies fault levels based on time-series correlation datasets, locates the root cause equipment and type of fault, integrates core diagnostic information to form a fault diagnosis report and sends it to the operation and maintenance scheduling terminal for layered operation and maintenance. The specific process is as follows:

[0106] Build a historical fault case library, an equipment parameter threshold library, and an equipment physical connection map;

[0107] Historical Fault Case Database: Records the fault types, multimodal characteristics, root causes, transmission paths, and handling results of each core device within a preset period in the past, including the mapping relationship between fault characteristics and root causes;

[0108] Equipment parameter threshold library: For each core device, corresponding fault thresholds are preset for various operating parameters;

[0109] Equipment physical connection diagram: Records the physical connection relationships between core devices (such as line connections, interface adaptation relationships, etc.), and clarifies the basis of physical interaction between devices;

[0110] Based on the core equipment coupling relationship matrix M, the equipment physical connection map, energy transmission paths, and equipment parameter threshold library, a three-dimensional fault transmission topology map of equipment nodes, associated links, and transmission weights is constructed by following the logical steps of clarifying the core information of equipment nodes → constructing associated links → calculating transmission weights. The specific process is as follows:

[0111] Equipment node construction: Based on each core device in the core equipment set, the entity attributes of the core devices in the equipment physical connection map (such as equipment model and installation location) are integrated with the preset fault thresholds corresponding to each operating parameter in the equipment parameter threshold library, and the core devices are transformed into independent equipment nodes in the topology map; each node is labeled with core information, including: equipment identifier, equipment type, and key parameter thresholds, forming the basic node system of the topology map;

[0112] Link construction: For any two core devices si and sj in the core device set, only if the following conditions are met: When three conditions are met—that si and sj have a physical connection in the equipment physical connection diagram, and that si and sj have an effective transmission channel in the energy transmission path—the link between the two is constructed.

[0113] Where Mij=1 (strong coupling) corresponds to a solid line link, and Mij=0.5 (medium coupling) corresponds to a dashed line link; if Mij=0 (no coupling), or si and sj have no physical connection, or there is no effective energy transmission channel, no associated link is constructed.

[0114] Transmission weight calculation: For each established connection link, define a transmission weight Wij (i.e., the probability of a fault propagating from core device si to core device sj), calculated using the following formula:

[0115] , where Sim(i,j) is the historical co-fault similarity between core device si and core device sj (calculated based on the frequency statistics of simultaneous or successive failures of the two in the historical fault case library), and λ is the preset weight coefficient;

[0116] The calculated Wij values ​​are assigned to the corresponding associated links to form a weight assignment system for the topology graph.

[0117] By integrating device nodes with complete labeled information, distinguishing associated links by coupling level, and quantifying the weight values ​​of propagation probability, a three-dimensional fault propagation topology map T is obtained.

[0118] For time-series correlation datasets, perform fault diagnosis and root cause tracing as follows:

[0119] For each data point in the time-series correlated dataset, extract its associated core device si, acquisition time t, and corresponding k-th type of operating parameter acquisition value P(i,k,t). Simultaneously, based on the device identifier si and operating parameter type k associated with the data, match and extract the normal baseline value P0(i,k) and fault threshold Pth(i,k) of the k-th type of operating parameter corresponding to the core device si from the device parameter threshold library. After normalization and dimensionless processing, use the formula: This yields the anomaly intensity factor Abn(i,k,t) corresponding to the current data.

[0120] Where k is the label of the type of operating parameter, P0(i,k,t) is the normal baseline value of the k-th type of operating parameter corresponding to the core device si at the acquisition time t. If P(i,k,t) is within the normal operating range of the k-th type of parameter of si, then Abn(i,k,t)=0; if P(i,k,t) exceeds the fault threshold Pth(i,k), then Abn(i,k,t)≥1.

[0121] Based on the core device si and acquisition time t bound to the current data, determine the device type to which si belongs; extract all historical fault types f that match the type to which si belongs from the historical fault case library, and obtain the feature weights Wf(f,k) of each historical fault type f for each type of operating parameter (the feature weights are determined based on the frequency of association between the fault type f and the k-th type of parameter anomaly in the historical records); simultaneously, based on the anomaly intensity factor Abn(i,k,t) corresponding to the current data, and after normalization and dimensionless processing, use the formula: Thus, the fault matching factor Mat(i,f,t) is obtained;

[0122] in, The sum of the abnormal intensity factors Abn(i,k,t) corresponding to all operating parameters of the core device si at acquisition time t;

[0123] Mat(i,f,t)∈[0,1], the closer the value is to 1, the higher the similarity between the multimodal anomaly characteristics of the core equipment si at time t and the historical fault type f;

[0124] For each core device si, if there exists an abnormal intensity factor Abn(i,k,t) ≥ 1 corresponding to any type of operating parameter, and the duration of this state exceeds the preset fault duration threshold, then the core device si is determined to have failed.

[0125] For any core device si that has failed, based on the effective fault propagation path of si in the three-dimensional fault propagation topology graph T, the total number of devices that are associated with core device si and have failed or shown an abnormal trend up to the current time is counted and denoted as the number of fault-associated devices Ni.

[0126] For the core equipment si that has failed, calculate the mean value of the abnormal intensity factor AvgAbn(i) of its various operating parameters during the failure duration up to the current time.

[0127] After normalizing and dimensionlessly processing the number of fault-related devices Ni corresponding to the core equipment and the mean of the anomaly intensity factor AvgAbn(i), the formula is used. The fault level quantification value GDi is obtained, where b1 and b2 are preset weight coefficients;

[0128] Three fault levels are defined: Level 1 fault (minor fault), Level 2 fault (planned maintenance), and Level 3 fault (emergency handling). Each fault level corresponds to a unique maintenance handling requirement and a fault level quantification range.

[0129] Among them, the higher the fault level, the larger the lower limit and upper limit of the corresponding fault level quantification value range.

[0130] Furthermore, the handling requirements for each fault level can be as follows: Level 1 fault corresponds to a minor fault scenario, and the handling requirement is a minor online adjustment that does not affect the normal operation of the equipment;

[0131] Level 2 faults correspond to medium-level fault scenarios, and the handling requirement is to carry out shutdown maintenance according to the preset plan.

[0132] Level 3 faults correspond to severe fault scenarios, and the handling requirement is to immediately initiate emergency shutdown procedures to quickly prevent the risk of the fault escalating.

[0133] The corresponding fault level is output by matching the quantized value of the fault level corresponding to the core equipment that has failed with the quantized value range of all fault levels.

[0134] For any core device si that has failed, in the three-dimensional fault propagation topology map, only links with propagation weight Wij ≥ preset propagation threshold are retained and marked as valid fault propagation paths;

[0135] From all valid fault propagation paths, select the core device that first shows an anomaly (i.e., the anomaly intensity factor Abn(i,k,t) ≥ the preset anomaly trend threshold) and has the earliest timestamp, and record it as the root cause device Sroot.

[0136] Starting from the first abnormal time t1 of the root cause device Sroot, a fault anomaly determination window [t1, t2] is preset, where t2 is the end time of the fault anomaly determination window; from the historical fault case library, all historical fault types f corresponding to the device type to which the root cause device Sroot belongs are extracted; for each historical fault type f, the formula is used: To obtain the comprehensive matching factor ;

[0137] Where Et is the preset weighting coefficient corresponding to time t. This refers to the fault matching factor between the root cause device Sroot and the historical fault type f at time t in the fault anomaly determination window.

[0138] For the root cause device, select the historical fault type with the largest value from the comprehensive matching factors corresponding to all historical fault types f, and use it as the final root cause type.

[0139] The core diagnostic information, including the fault device identifier, fault level, abnormal feature sequence, effective fault propagation path, root cause device, and root cause type of the core equipment that has failed, is integrated and organized according to a preset standardized format to form a fault diagnosis report and sent to the operation and maintenance dispatch terminal.

[0140] The operation and maintenance dispatch terminal uses fault diagnosis reports as the core execution basis, combining the core equipment coupling relationship matrix, the handling results corresponding to the root cause type in the historical fault case library, and the preset normal range of operating parameters and fault thresholds in the equipment parameter threshold library. Following a layered operation and maintenance logic of prioritizing the handling of the root cause equipment, requiring the repair of the faulty equipment itself, and focusing on the prevention and control of related equipment, it carries out precise handling, specifically:

[0141] For the equipment that is the root cause of the failure, prioritize repair by matching the standard handling plan for the corresponding root cause type in the historical failure case library;

[0142] For core equipment that has malfunctioned, the abnormal state shall be repaired in accordance with the handling requirements corresponding to the fault level to ensure that its operating parameters return to the normal range preset in the equipment parameter threshold library;

[0143] For associated devices along the effective fault propagation path, targeted monitoring is implemented based on the coupling level determined by the core device coupling relationship matrix to prevent further fault propagation.

[0144] It should be noted that the fault diagnosis and root cause tracing module constructs a full-process fault handling system through multi-dimensional data integration, visual topology modeling, and hierarchical operation and maintenance logic.

[0145] Based on a three-dimensional fault propagation topology map, integrating equipment coupling relationships, physical connections and energy transmission paths, and combining multimodal parameters in time-series correlation data, the system accurately determines equipment fault status through quantitative analysis of anomaly intensity factors and fault matching factors. At the same time, based on effective propagation paths and timestamp filtering, the system quickly identifies the root cause equipment and type, significantly shortening the fault investigation cycle and improving diagnostic efficiency.

[0146] The fault level is quantified by the number of fault-related devices and the average of the anomaly intensity. Dedicated handling requirements are matched for faults of different severities, avoiding excessive downtime for minor faults or delayed handling of serious faults. This allows maintenance resources to be tilted toward high-priority faults, improving the rationality and efficiency of handling.

[0147] The generated standardized fault diagnosis reports provide a clear basis for operation and maintenance scheduling. Following the logic of "root cause priority, fault repair, and related prevention and control", it not only quickly repairs the root cause equipment by matching historical solutions, but also monitors related equipment according to coupling level, blocking the fault transmission chain from the source, reducing the risk of equipment downtime and operation and maintenance costs, and helping the energy station to carry out intelligent and refined operation and maintenance.

[0148] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent operation and maintenance monitoring and management platform for energy stations, characterized in that, include: Equipment Association Modeling Module: Mark core monitoring equipment, acquire coupling raw data, construct ordered pairs of equipment, analyze their energy transmission factors, signal factors and common anomaly factors and perform comprehensive processing to obtain the comprehensive coupling degree, determine the coupling level, and generate the core equipment coupling relationship matrix; The specific process of constructing ordered pairs of devices is as follows: Identify the operating equipment in the energy station that requires maintenance and monitoring, mark them as core equipment, and organize them into a core equipment set; Obtain the coupled raw dataset for each core device one by one. This dataset includes: Energy transmission topology file, signal linkage configuration file, rated transmission parameter matrix, and historical operation log; Energy transfer topology file: records the energy transfer path between devices in the form of a quadruple; Signal linkage configuration file: records the correlation of control or feedback signals between devices in the form of a quad tuple; Rated transmission parameter matrix: Characterizes the rated transmission capacity or power between devices; the corresponding value is 0 when there is no transmission path. Historical operation log: records the number of simultaneous anomalies between devices and the sequence of anomalies; After data acquisition is completed, all core devices in the core device set are paired up to obtain ordered pairs of devices. When combining, the same device with itself is excluded. The specific process of comprehensive coupling degree analysis is as follows: For each ordered pair of devices, three types of coupled judgment factors are constructed: energy transfer factor, signal factor, and common anomaly factor. The energy transfer factor is a factor that characterizes the strength of the energy transfer correlation between two core devices in an ordered pair of devices. It is determined based on the rated transfer data between devices in the rated transfer parameter matrix and the maximum value of the rated total transfer of the device to the downstream device. When there is no transfer path, this factor is 0. Signal factor is a factor that characterizes whether there is signal linkage between two core devices in the orderly alignment of the equipment. It is determined based on whether there are control or feedback signals between the devices. The common anomaly factor is a factor that characterizes the correlation of the transmission of abnormal states between the two core devices in the orderly alignment of the equipment. It is calculated based on the number of simultaneous anomalies and the number of sequential anomalies between the devices. The three types of factors are weighted and summed according to preset weight coefficients to obtain the comprehensive coupling degree of the ordered pairs of the device; Data Acquisition Strategy Optimization Module: Determines importance based on the historical failure frequency and energy transmission intensity of core equipment, and matches the corresponding benchmark acquisition frequency; calculates operational evaluation indicators based on the percentage of abnormal operating parameters, and adjusts the acquisition frequency of the equipment itself and related equipment in combination with coupling relationships; collects data at the adjusted frequency, and generates a time-series associated dataset bound to the equipment identifier and a unified timestamp. Fault diagnosis and root cause tracing module: Constructs a three-dimensional fault propagation topology map, determines equipment faults and classifies fault levels based on time-series correlation datasets, locates the root cause equipment and type of fault, integrates core diagnostic information to form a fault diagnosis report and sends it to the operation and maintenance scheduling terminal for hierarchical operation and maintenance.

2. The intelligent operation and maintenance monitoring and management platform for energy stations according to claim 1, characterized in that: The process of generating the coupling relationship matrix of core equipment is as follows: Preset high coupling threshold and low coupling threshold, and determine the coupling level of ordered pairs of devices by comparing the comprehensive coupling degree with the two types of thresholds; A combined coupling degree greater than the high coupling threshold is considered strong coupling, between the high and low coupling thresholds is considered medium coupling, and less than the low coupling threshold is considered no coupling. Based on the coupling levels of all ordered pairs of devices, a matrix is ​​constructed with dimensions corresponding to the total number of core devices. Matrix elements are assigned values ​​according to their coupling levels: strong coupling is assigned a value of 1, medium coupling is assigned a value of 0.5, and no coupling is assigned a value of 0. The diagonal elements of the matrix are uniformly set to 0, thus obtaining the core device coupling relationship matrix.

3. The intelligent operation and maintenance monitoring and management platform for energy stations according to claim 2, characterized in that: The specific process of determining importance based on the historical failure frequency and energy transmission intensity of core equipment and matching it with the corresponding reference acquisition frequency is as follows: The historical failure frequency and energy transmission intensity of each core device are obtained and comprehensively analyzed to obtain the device importance assessment value; Set several equipment importance levels, and preset a unique equipment importance evaluation value range and a unique benchmark collection frequency range for each level; The importance assessment value of the core equipment is matched with the assessment value range corresponding to each level, and the benchmark acquisition frequency range corresponding to the core equipment is output.

4. The intelligent operation and maintenance monitoring and management platform for energy stations according to claim 3, characterized in that: The specific process for adjusting the acquisition frequency of the device itself and associated devices is as follows: Acquire the operating parameter data within the preset window of the core equipment, preset the normal operating range of each parameter, and count the percentage of times the parameter deviates from the range as the parameter abnormality percentage. Calculate the operational evaluation indicators by combining the preset weights of each operational parameter; Compare the operational evaluation indicators with the preset benchmark values: if the indicator is less than the benchmark value, it is determined that there is an abnormal trend. Adjust the sampling frequency of the device to the upper limit of the benchmark interval, and extract non-uncoupled related devices. For strongly coupled devices, adjust the frequency to the upper limit of the benchmark interval, and for moderately coupled devices, adjust the frequency to the median value. If the target is met, the sampling frequency of the device will be lowered to the lower limit of the baseline range.

5. The intelligent operation and maintenance monitoring and management platform for energy stations according to claim 4, characterized in that: The process of forming a time-series associated dataset is as follows: The system monitors the operational evaluation indicators of each core device and the status of the core device coupling relationship matrix in real time. Based on the current acquisition frequency of each core device, it collects multimodal data of each core device. It uses a unified time synchronization mechanism to synchronize the timestamps of all acquisition nodes and binds each piece of acquired data with the corresponding device identifier and unified timestamp to form a time-series associated dataset.

6. The intelligent operation and maintenance monitoring and management platform for energy stations according to claim 5, characterized in that: The process of constructing a 3D fault propagation topology map and determining equipment faults based on a time-series correlation dataset is as follows: Construct a historical fault case library, an equipment parameter threshold library, and an equipment physical connection map, respectively recording fault-related information, parameter fault thresholds, and equipment entity connection relationships; Based on core equipment, its physical attributes and parameter fault thresholds are integrated to form topology nodes, and the equipment identification, type and key parameter thresholds are labeled; when the conditions of non-uncoupled, physical connection and effective energy transmission channel are met, the associated links are constructed, the link types are distinguished according to the coupling level, the link propagation weight is calculated by combining the coupling degree and the similarity of historical common faults, and a three-dimensional fault propagation topology map is integrated. For each core device, the corresponding acquisition time and parameter value are extracted from the time-series correlation data. The abnormal intensity factor is calculated by matching the normal baseline value and fault threshold of the device. Combined with the historical fault type feature weight of the device type, the fault matching factor is obtained. When the abnormal intensity factor of any parameter of the core device is ≥1 and the duration of this state exceeds the preset fault duration threshold, the core device is determined to have failed.

7. The intelligent operation and maintenance monitoring and management platform for energy stations according to claim 6, characterized in that: The specific process for classifying fault levels and identifying the root cause equipment and type of fault is as follows: For the core equipment that has failed, the total number of related equipment with failure or abnormal trend is counted, the average value of the abnormal intensity factor during the failure duration is calculated, and the two are normalized and weighted according to the preset weight to obtain the fault level quantification value. The fault level is then matched with the preset level range to output the fault level. In the three-dimensional fault propagation topology map, valid fault propagation paths with propagation weights reaching a preset threshold are retained, and the device with the first anomaly and the earliest timestamp is selected as the root cause device of the fault. A fault anomaly determination window is set up with the first time the root cause device becomes abnormal. All historical fault types of its type are extracted, the comprehensive matching factor of each type in the window is calculated, and the historical fault type with the largest value is selected as the final root cause type.

8. The intelligent operation and maintenance monitoring and management platform for energy stations according to claim 7, characterized in that: The specific process of generating a fault diagnosis report and sending it to the operation and maintenance dispatch terminal for tiered operation and maintenance is as follows: The core diagnostic information, including fault device identifier, fault level, abnormal feature sequence, effective fault propagation path, root cause device and fault root cause type, is integrated into the core equipment that has failed. The information is then organized into a fault diagnosis report according to a preset standardized format and sent to the operation and maintenance dispatch terminal. Based on this report, the operation and maintenance dispatch terminal combines the core equipment coupling relationship matrix, the handling results of the corresponding root cause type in the historical fault case library, and the operating parameter standards in the equipment parameter threshold library. It follows the hierarchical operation and maintenance logic of prioritizing the handling of the root cause equipment, repairing the faulty equipment itself, and focusing on the prevention and control of related equipment: prioritizing the matching of standard solutions to repair the root cause equipment, repairing the faulty equipment according to the fault level requirements to bring its parameters back to the normal range, and implementing targeted monitoring of related equipment on the effective fault propagation path according to the coupling level.

Images