Operation and maintenance rapid response method and system under power market monitoring and medium

Abstract

The application discloses a power market monitoring-based operation and maintenance rapid response method and system and a medium, relates to the related technical field of power operation and maintenance, and comprises the following steps: continuously monitoring a power market to obtain a multi-source data set; calling a state sequence identification plan to identify and analyze the state data set; calling a composite current criterion plan to analyze the current data set; cooperating real-time state sequences and real-time leakage information to obtain a real-time fault list; obtaining predetermined equipment weights, combining an operation data set to analyze a first fault to obtain a first priority index; and performing operation and maintenance response on the first fault based on the first priority index. The application solves the technical problems of multi-source power data dispersion, fault identification lag, insufficient composite fault diagnosis precision and unclear operation and maintenance response priority in the prior art, and achieves the technical effects of real-time and accurate monitoring, rapid fault positioning, intelligent priority sorting and efficient operation and maintenance response.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric power operation and maintenance, and particularly relates to an operation and maintenance rapid response method and system under electric power market monitoring and a medium. BACKGROUND

[0002] The operation state monitoring and fault rapid response of electric power equipment are the key to guarantee the safe and stable operation of the power grid. The traditional electric power operation and maintenance is difficult to cope with the high-frequency, multi-source and dynamic change of the operation environment of the modern electric power market. Especially under the background of electric power marketization, factors such as electric power transaction, supply and demand fluctuation and distributed energy access further aggravate the uncertainty of power grid operation, which expands the probability and influence range of faults. At the same time, the power system monitoring relies on the scattered and heterogeneous data collected by SCADA (Supervisory Control And Data Acquisition) and PMU (Phasor Measurement Unit) devices, lacks effective collaborative analysis means, and leads to lagging fault identification or misjudgment. In addition, fault diagnosis is difficult to adapt to complex and variable operation conditions, especially in the composite fault scenarios such as electric leakage, short circuit and equipment overload, which is prone to miss or false reporting. The real-time monitoring and fault positioning of the operation state of the power system are affected.

[0003] Therefore, in the related art at present, there are technical problems of scattered multi-source electric power data, lagging fault identification, insufficient composite fault diagnosis accuracy and unclear operation and maintenance response priority. SUMMARY

[0004] The present application provides an operation and maintenance rapid response method and system under electric power market monitoring and a medium, which solves the technical problems of scattered multi-source electric power data, lagging fault identification, insufficient composite fault diagnosis accuracy and unclear operation and maintenance response priority in the prior art, and achieves the technical effects of real-time and accurate monitoring, rapid fault positioning, intelligent priority sorting and efficient operation and maintenance response.

[0005] The present application provides an operation and maintenance rapid response method under electric power market monitoring, which comprises: continuously monitoring the electric power market to obtain a multi-source data set; calling a state sequence identification plan to identify and analyze a state data set in the multi-source data set to obtain a real-time state sequence; calling a composite current criterion plan to criterion analyze a current data set in the multi-source data set to obtain real-time electric leakage information; cooperating the real-time state sequence with the real-time electric leakage information to obtain a real-time fault list of the electric power market, wherein the real-time fault list comprises a plurality of faults with device and degree identifiers; obtaining a predetermined device weight, and combining an operation data set in the multi-source data set to analyze a first fault to obtain a first priority index, wherein the first fault refers to any one of the plurality of faults with device and degree identifiers; and performing operation and maintenance response on the first fault based on the first priority index.

[0006] In a possible implementation, the operation and maintenance rapid response method under the power market monitoring further performs the following processing: performing state monitoring on the power system in the power market according to a physical memory mapping mechanism to obtain the state data set; obtaining a substation in the power market, and performing current information monitoring on a long-section power cable of the substation to obtain the current data set; performing operation monitoring on the power market to obtain the operation data set; and the state data set, the current data set, and the operation data set constitute the multi-source data set.

[0007] In a possible implementation, the operation and maintenance rapid response method under the power market monitoring further performs the following processing: obtaining secondary equipment of the power system; obtaining a memory address of the secondary equipment according to the physical memory mapping mechanism; and accessing the memory address to obtain the state data set.

[0008] In a possible implementation, the operation and maintenance rapid response method under the power market monitoring further performs the following processing: extracting a first state parameter in the state data set; performing standardization processing on the first state parameter according to a parameter standard strategy in the state sequence identification plan to obtain a first sequence value; and establishing the real-time state sequence based on the first sequence value.

[0009] In a possible implementation, the operation and maintenance rapid response method under the power market monitoring further performs the following processing: analyzing historical secondary equipment state records and establishing a historical database, and constructing a state prediction model according to the historical database; performing prediction analysis on the state data set through the state prediction model to obtain a prediction state sequence; analyzing the prediction state sequence to obtain a state sequence limit value, and verifying the real-time state sequence by using the state sequence limit value; wherein the processing includes: extracting a first data set of a first historical stage in the historical database, wherein the first data set includes a first historical state data set and a first historical fault condition; matching a first predetermined state sequence corresponding to the first historical fault condition, and establishing a training data set in combination with the first historical state data set; and performing supervised learning on the training data set to obtain the state prediction model.

[0010] In a possible implementation, the operation and maintenance rapid response method under the power market monitoring further performs the following processing: traversing the real-time state sequence in the historical database to obtain a traversal result; analyzing the traversal result to obtain a most matched data set, wherein the most matched data set includes a most matched fault condition; and adding the most matched fault condition to the real-time fault list.

[0011] In a possible implementation, the operation and maintenance rapid response method under the power market monitoring further performs the following processing: obtaining a first element in the predetermined element, and matching a first current parameter corresponding to the first element in the current data set; forming a current vector based on the first current parameter, and calculating a current similarity index of the current vector and a reference current vector; if the current similarity index does not reach a predetermined similarity limit value, identifying and determining the real-time leakage information according to a composite current criterion plan; wherein the predetermined element at least includes a steady-state residual current element, a residual current increment element, a steady-state residual current differential element, and a single-phase residual current differential element.

[0012] In a possible implementation, the operation and maintenance rapid response method under the power market monitoring further performs the following processing: establishing a leakage database, wherein the leakage database includes a plurality of current vectors with leakage type and leakage degree identification; calculating a first similarity of the current vector and a first vector, wherein the first vector refers to any one of the plurality of current vectors with leakage type and leakage degree identification; if the first similarity reaches the predetermined similarity limit value, matching a first leakage type and a first leakage degree corresponding to the first vector; based on the first leakage type and the first leakage degree, obtaining the real-time leakage information.

[0013] The application further provides an operation and maintenance rapid response system under a power market monitoring, comprising: a continuity monitoring module configured to monitor the power market for continuity, and obtain a multi-source data set; an identification analysis module configured to call a state sequence identification plan to identify and analyze state data sets in the multi-source data set, and obtain a real-time state sequence; a criterion analysis module configured to call a composite current criterion plan to analyze current data sets in the multi-source data set, and obtain real-time leakage information; a real-time fault list obtaining module configured to obtain a real-time fault list of the power market in cooperation with the real-time state sequence and the real-time leakage information, wherein the real-time fault list includes a plurality of faults with device and degree identification; a first fault analysis module configured to obtain a predetermined device weight, and analyze a first fault in combination with operation data sets in the multi-source data set to obtain a first priority index, wherein the first fault refers to any one of the plurality of faults with device and degree identification; and an operation and maintenance response module configured to perform operation and maintenance response on the first fault based on the first priority index.

[0014] The application further provides a computer-readable storage medium, comprising: a computer program stored thereon, which is executed by a processor to implement the operation and maintenance rapid response method under the power market monitoring.

[0015] The power market monitoring operation and maintenance rapid response method, system and medium provided by the application can continuously monitor the power market, obtain a multi-source data set, call a state sequence identification plan to identify and analyze the state data set, call a composite current criterion plan to analyze the current data set, obtain a real-time fault list by cooperating with real-time state sequence and real-time leakage information, obtain a predetermined equipment weight, analyze the first fault by combining the operation data set to obtain a first priority index, and perform operation and maintenance response on the first fault based on the first priority index. The technical problems of scattered multi-source power data, lagging fault identification, insufficient composite fault diagnosis accuracy and unclear operation and maintenance response priority in the prior art are solved, and the technical effects of real-time and accurate monitoring, rapid fault positioning, intelligent priority sorting and efficient operation and maintenance response are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application will be briefly introduced below. In the present application, a flowchart is used to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously as needed. Meanwhile, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 The power market monitoring operation and maintenance rapid response method flowchart provided by the embodiments of the present application is shown.

[0018] Figure 2 The power market monitoring operation and maintenance rapid response system structure schematic diagram provided by the embodiments of the present application is shown.

[0019] The reference signs are explained as follows: the continuity monitoring module 10, the identification analysis module 20, the criterion analysis module 30, the real-time fault list obtaining module 40, the first fault analysis module 50, and the operation and maintenance response module 60. DETAILED DESCRIPTION

[0020] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the following specific embodiments of the present application can be implemented according to the content of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0021] In order to make the purposes, technical solutions and advantages of the present application more clear, the following will further describe the present application in combination with the drawings, and the described embodiments should not be regarded as limiting the present application. All other embodiments obtained by those skilled in the art without making creative efforts fall within the scope of protection of the present application.

[0022] In the following description, "some embodiments" are referred to, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict, and the term "first\second" referred to only distinguishes similar objects, and does not represent a specific order for the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0023] The embodiments of the present application provide a method for rapid response of operation and maintenance under power market monitoring, as shown in the method comprises: Figure 1 The method comprises the following steps:

[0024] Step S100, continuously monitoring the power market to obtain a multi-source data set.

[0025] Step S100 further comprises the following steps: Step S110, monitoring the state of the power system in the power market according to a physical memory mapping mechanism to obtain the state data set; Step S120, obtaining the transformer substation of the power market and monitoring the current information of the long section power cable of the transformer substation to obtain the current data set; Step S130, monitoring the operation of the power market to obtain the operation data set; and Step S140, the state data set, the current data set and the operation data set constitute the multi-source data set.

[0026] Preferably, the power market is continuously monitored, that is, full-time coverage monitoring is performed at the millisecond level (device state) and minute level (market transaction), and state data sets, current data sets and operation data sets are obtained. Specifically, according to a physical memory mapping mechanism, real-time state register values are read from embedded systems or FPGA / CPLD of power equipment through MMU (Memory Management Unit) or direct memory access (DMA), that is, the real-time running state of hardware devices (such as relay protection devices, smart meters, circuit breakers, transformers, etc.) in the power system is monitored, the state information in the registers, I / O ports or firmware of the devices is obtained, and state data sets are formed, which may include device switching states (such as opening / closing), protection device action signals (such as overcurrent protection triggering), communication states (such as PLC, fiber channel connectivity), and device health (such as temperature, vibration, insulation aging monitoring data), for example, fault flag bits of smart circuit breakers, oil temperature data of transformers, etc.

[0027] Preferably, the substation of the power market is sampled in real time by using a Rogowski coil or a high-precision CT (current transformer), high-precision current monitoring is performed on long-distance power cables (such as 10kV / 35kV transmission lines) of the substation, abnormal currents (such as leakage, harmonics, short-circuit currents) are detected, and FFT (Fast Fourier Transform) analysis is combined to analyze harmonic components, or wavelet transform is combined to detect transient faults, forming current data sets, which may include three-phase current amplitude, zero-sequence current, current harmonic distortion rate and transient current waveform, such as short-circuit impulse current. The power market is synchronously monitored in a wide range of operation, including obtaining data of overall operation situation of the power market from SCADA / EMS (energy management system) or power trading platform API, including grid dispatching, trading, load forecasting and other information, forming operation data sets, which may include grid flow distribution, node voltage / frequency stability data, power market transaction information (such as real-time electricity price, supply and demand balance) and new energy (photovoltaic, wind power) grid-connected power fluctuation.

[0028] Preferably, the state data set is used to determine whether the device is abnormal (such as circuit breaker refusal), the current data set is used to locate the fault type and position (such as cable leakage point), and the operation data set is used to evaluate the influence of the fault on the overall power grid (such as whether it causes voltage collapse). The state data set, the current data set and the operation data set are fused to form a multi-source data set, which can significantly improve the comprehensiveness and response speed of fault detection.

[0029] Further, step S110 further includes step S111 of acquiring secondary equipment of the power system; step S112 of acquiring a memory address of the secondary equipment according to the physical memory mapping mechanism; and step S113 of accessing the memory address to obtain the state data set.

[0030] Preferably, the secondary equipment refers to intelligent electronic devices in the power system for monitoring, protection and control of the main equipment (primary equipment), mainly including protection devices (such as line protection devices, transformer differential protection devices), measurement and control devices (such as merging units, intelligent terminals), automation devices (such as automatic voltage control systems, synchronous phasor measurement devices) and communication devices (such as protocol converters, time synchronization devices); the memory address of the secondary equipment is obtained according to the physical memory mapping mechanism, specifically, each secondary equipment has a specific memory area inside for storing real-time running parameters and state flags, these memory addresses are usually predefined by the device manufacturer during development, corresponding to different function registers, a direct channel with the device memory is established through the underlying hardware interface (such as FPGA or special communication chip), bypassing the conventional communication protocol stack, realizing direct read-write operation at the register level; including identifying the memory mapping table of the target device, locating the memory address range of the key state register, finally accessing the memory address to read real-time data through direct memory access technology, obtaining the state data set, such as device running state (switching value state, protection start / action flag), real-time measurement value (current / voltage sampling value, power calculation value), event record (protection action event, communication exception event, device alarm event) and device health status (internal temperature, power state, memory state, etc.).

[0031] Step S200, calling the state sequence identification plan to identify and analyze the state data set in the multi-source data set, to obtain the real-time state sequence.

[0032] Preferably, the state data set in the multi-source data set is identified and analyzed through the pre-defined intelligent analysis rule (i.e. the state sequence identification plan), wherein the state sequence identification plan includes the pre-established device state evaluation strategy, which may contain the normal working condition parameter threshold values (such as temperature, insulation resistance, etc. standard values) of various power equipment and the characteristic patterns of typical abnormal states (such as the vibration waveform characteristics of circuit breaker mechanical failure), specifically, the data preprocessing is performed on the state data set in the multi-source data set, i.e. the standardization conversion is performed on the original state data obtained by memory mapping, including analog quantity normalization (converting sensor data of different units into standard dimension) and state quantity coding (transforming discrete switching signals into state machine recognizable codes), then multi-dimensional feature extraction is performed, such as extracting the change rate, fluctuation amplitude and signal spectrum distribution of the monitoring parameters; then the extracted features are compared with the standard rules in the state sequence identification plan, including checking whether the pre-set safety threshold is exceeded and judging whether the feature combination of the typical fault is met, and then outputting the state chain with time stamp, i.e. the real-time state sequence, for the accurate real-time operation and maintenance of the power equipment.

[0033] Further, step S200 further comprises step S210 of extracting a first state parameter in the state data set; step S220 of performing standardization processing on the first state parameter according to a parameter standard strategy in the state sequence identification plan to obtain a first sequence value; and step S230 of establishing the real-time state sequence based on the first sequence value.

[0034] Preferably, the original state data of the equipment operation is converted into a state evaluation index chain with time sequence characteristics and standard comparability by the parameter standard strategy in the state sequence identification plan, wherein the parameter standard strategy is a data standardization rule defined in the state sequence identification plan. Specifically, a first state parameter, such as a circuit breaker contact resistance, a transformer winding temperature, and a cable partial discharge amount, is randomly extracted from the state data set, matched in the parameter standard strategy, and standardized, such as calling a standard parameter corresponding to the equipment model and operation life in the parameter standard strategy to correct the first state parameter, and converting a continuous quantity into a discrete state level, such as converting a temperature value into normal / attention / abnormal / critical, to obtain a first sequence value. Finally, the real-time state sequence is established according to the first sequence value, so that the operation and maintenance personnel can intuitively master the dynamic evolution law of the equipment state.

[0035] Further, step S230 further comprises step S231 of analyzing historical secondary equipment state records and establishing a historical database, and constructing a state prediction model according to the historical database; step S232 of performing prediction analysis on the state data set through the state prediction model to obtain a predicted state sequence; and step S233 of analyzing the predicted state sequence to obtain a state sequence limit value, and verifying the real-time state sequence with the state sequence limit value; wherein, step a of extracting a first data set of a first historical stage in the historical database, wherein the first data set comprises a first historical state data set and a first historical fault condition; step b of matching a first predetermined state sequence corresponding to the first historical fault condition, and establishing a training data set in combination with the first historical state data set; and step c of performing supervised learning on the training data set to obtain the state prediction model.

[0036] Preferably, during the operation of the power system, the secondary equipment will continuously generate a large amount of operation state data, recording various parameter information of the equipment at different times and under different working conditions. By collecting, sorting and storing historical secondary equipment state records, including long-term operation records of secondary equipment (such as protection device action log, self-check report), fault event archives (including fault type, occurrence time, treatment measures) and environment related data (temperature, humidity, load rate at the time of failure, etc.), a historical database is established. Then, data of a specific time period (i.e. the first historical stage) is extracted from the historical database to establish a first data set containing typical operation stage data in the whole life cycle of the secondary equipment. The first data set includes the first historical state data set and the first historical fault condition. The first historical state data set covers various data of the secondary equipment in normal operation state in this stage, and the first historical fault condition records the time, type, performance, etc. of the equipment failure in this stage.

[0037] Preferably, the first historical fault condition is matched with the related state sequence in the first historical state data set to obtain the corresponding first predetermined state sequence, i.e. the specific sequence pattern presented by the equipment operation state before the failure occurs. By matching the first predetermined state sequence and integrating it with the corresponding first historical state data set, a training data set is formed, containing data of normal operation state and pre-failure state. Then, the relationship between the input data (historical state data in the training data set) and the output data (corresponding fault condition or state change) is determined. The training data set is input into the constructed machine learning model (LSTM, TCN, etc. time series neural network). According to the corresponding relationship between the input and the output, the parameters and structure of the model are adjusted to minimize the error between the predicted results and the actual results. After repeated training and optimization, a state prediction model that can accurately predict the equipment state is finally obtained, which is used to accurately predict the future equipment state according to the state data set, and output the predicted state sequence of the secondary equipment, including the future state prediction value and the failure probability.

[0038] Preferably, the state data set is analyzed by a state prediction model, that is, the real-time collected state data set (such as the voltage, current, temperature, switch state and other parameters of the secondary equipment) is input into the state prediction model to identify the trend characteristics (such as continuous rise of equipment temperature), periodic regularity (such as load fluctuation during daily peak electricity consumption period) and abnormal fluctuation mode in the data. The model predicts the equipment state at multiple future time nodes to generate a predicted state sequence containing future values of each parameter. Then, the predicted state sequence is analyzed, including descriptive statistics of the predicted state sequence, calculation of the mean and standard deviation of each parameter, and taking the mean ± k x standard deviation as the normal range limit value, where k is an empirical coefficient, for example, k = 2 covers 95% of the data, and thus the state sequence limit value is obtained. Finally, the real-time state sequence is checked by the state sequence limit value. Specifically, the real-time collected state sequence is aligned with the predicted state sequence according to the time stamp to ensure that the parameter values at the same time point can be compared, and the Mahalanobis distance between the real-time data and the predicted value is compared. If the real-time current value exceeds the upper limit of the predicted sequence limit value, it is determined as a potential fault and an early warning is triggered, and a comprehensive check result is generated.

[0039] Step S300, retrieve the composite current criterion plan to analyze the current data set in the multi-source data set to obtain real-time leakage information.

[0040] Preferably, the composite current criterion plan is a multi-dimensional and intelligent leakage detection scheme, which can analyze and judge the current data by comprehensively considering multi-dimensional characteristics such as steady-state current, transient current and harmonic component to obtain real-time leakage information. The multi-dimensional current criterion data is shown in Table 1:

[0041] Table 1 Multi-dimensional current criterion data table

[0042]

[0043] Preferably, the composite current criterion scheme performs criterion analysis on the current data set in the multi-source data set. Specifically, first, zero sequence current and residual current criterion analysis is performed. If the zero sequence current is greater than a set threshold (e.g., 50 mA) and the duration is > 500 ms, a zero sequence current warning is triggered, and if the residual current is greater than 30 mA, current mutation detection and spectral feature extraction are performed. Specifically, the current rate of change in the sliding window is calculated, and when it is greater than 10 A / ms and lasts for 2 sampling periods (40 ms), it is marked as a transient leakage pulse. Then, the current signal is subjected to fast Fourier transform (FFT), and the energy proportion in the 20 kHz frequency band is analyzed. If the energy in this frequency band increases by more than 30% compared to the historical average, it is determined to be partial discharge caused by insulation aging. Then, logical combination verification is performed. If the zero sequence current exceeds the standard and the residual current exceeds the standard, the leakage type is determined to be a metallic grounding fault. If the current mutation rate exceeds the standard and the spectrum is abnormal, the leakage type is determined to be a cable transient breakdown. If the three-phase unbalance degree is > 20% and the zero sequence current fluctuates, the leakage type is determined to be internal winding leakage of equipment. Finally, real-time leakage information is output, including the distance of the leakage point from the measurement end calculated by the current traveling wave positioning method (such as the double-ended traveling wave method), the leakage current size (such as 200 mA), the leakage duration (such as 2 minutes and 30 seconds), and the risk level, thereby improving the reliability and intelligent level of power system leakage protection.

[0044] Further, step S300 further includes step S310 of acquiring a first element in the predetermined element and matching a first current parameter corresponding to the first element in the current data set; step S320 of forming a current vector based on the first current parameter and calculating a current similarity index of the current vector and a reference current vector; and step S330 of identifying and determining the real-time leakage information according to the composite current criterion scheme if the current similarity index does not reach a predetermined similarity limit. The predetermined element at least includes a steady-state residual current element, a residual current increment element, a steady-state residual current differential element, and a single-phase residual current differential element.

[0045] Preferably, the predetermined element is a core parameter acquisition unit for leakage detection, including at least a steady-state residual current element, a residual current increment element, a steady-state residual current differential element, and a single-phase residual current differential element. The steady-state residual current element is used to monitor the residual current (i.e., the three-phase current vector sum) during normal operation of the power system, reflecting the leakage reference value under system steady state. The three-phase current vector sum is theoretically zero under normal conditions. If there is a steady-state residual current, it may indicate a slow leakage fault such as insulation aging. The residual current increment element focuses on the dynamic change of the residual current, i.e., the difference between the current residual current and the historical reference value, which is used to identify sudden leakage events such as current surges during equipment insulation breakdown. The steady-state residual current differential element is used to compare the difference in steady-state residual current at different positions (such as the beginning and end of the line). If the difference between the two ends exceeds the threshold, it may indicate that there is a leakage point in the interval (such as line damage leading to current loss). The single-phase residual current differential element compares the current difference between the phase line and the zero line for single-phase lines. Under normal conditions, the current of the two is equal in size and opposite in direction. If there is a differential current, it directly points to a single-phase ground fault.

[0046] Preferably, a detection element is randomly selected from the predetermined element as the first element, and the current monitoring data corresponding to the first element, i.e., the first current parameter, is matched and extracted from the current data set. If the steady-state residual current element is selected, the matched parameter is the vector sum of the three-phase current at a certain time. Then, the current parameters (or multi-dimensional parameters) of the same element at different time points are combined into a vector to obtain a current vector, which contains multiple continuously monitored residual current values or harmonic amplitudes of different frequency bands. The reference current vector is a standard vector constructed based on historical normal operation data, representing the current characteristics of the element when it is not faulty. For example, the residual current values of 1000 normal operation samples are extracted from the historical database, and the average value is taken to form the reference current vector.

[0047] Preferably, the vector similarity algorithm (such as cosine similarity) is used to calculate the matching degree of the current vector and the reference current vector to obtain a current similarity index. Specifically, the smaller the vector angle, the higher the similarity, indicating that the current characteristics are closer to the normal state. The larger the angle, the lower the similarity, indicating that there may be a leakage anomaly. If the current similarity index does not reach the predetermined similarity limit value (such as the threshold is set to 0.9), it indicates that the current characteristics are significantly different from the normal state, triggering a leakage warning. Combined with the composite current criterion plan, the similarity index results of multiple elements are integrated to determine the leakage location and degree. For example, if the similarity of the steady-state residual current element is low and the increment element shows a current surge, it may be determined as a gradual insulation fault. If the similarity of the steady-state residual current differential element is low and the current difference between the two ends exceeds the threshold, it directly locates the leakage point in the interval. Finally, it effectively distinguishes between normal load fluctuations and real leakage faults, reduces the false alarm rate, and determines real-time leakage information.

[0048] Further, step S330 further comprises step S331 of establishing an electric leakage database, wherein the electric leakage database comprises a plurality of current vectors with electric leakage type and electric leakage degree identification; step S332 of calculating a first similarity of the current vector and a first vector, wherein the first vector refers to any one of the plurality of current vectors with electric leakage type and electric leakage degree identification; step S333 of matching a first electric leakage type and a first electric leakage degree corresponding to the first vector if the first similarity reaches the predetermined similarity limit value; and step S334 of obtaining the real-time electric leakage information based on the first electric leakage type and the first electric leakage degree.

[0049] Preferably, the electric leakage database is established by a plurality of current vectors with electric leakage type and electric leakage degree identification in historical fault data and power system simulation data, wherein the current vector records multi-dimensional characteristics of current at the time of fault (such as amplitude, phase, harmonic component, etc.), usually in the form of time series or multi-dimensional parameter combination. Specifically, the electric leakage type identification is a classification of fault nature, which may include but is not limited to single-phase ground leakage (such as A-phase insulation damage grounding), phase-to-phase leakage (such as AB-phase short circuit), arc leakage (such as electric arc generated by poor line contact), and internal equipment leakage (such as transformer winding insulation breakdown). The electric leakage degree identification quantifies the severity level of the fault, which is usually divided into mild (such as slight damp of insulation, leakage current < 100 mA), moderate (such as partial discharge, leakage current 100 mA ~ 500 mA), and severe (such as metallic grounding, leakage current > 500 mA).

[0050] Preferably, one of the labeled current vectors in the electric leakage database is selected as the first vector for comparison with the current vector to calculate the similarity, i.e., the vector similarity algorithm (such as cosine similarity, Euclidean distance) is used to calculate the matching degree of the real-time current vector and the first vector to determine the first similarity. According to the historical data, a predetermined similarity limit value is set. If the similarity of the real-time vector and a certain vector in the database reaches the limit value, it is determined that they belong to the same type of electric leakage fault, and then the electric leakage type and electric leakage degree corresponding to the first vector in the electric leakage database are matched as the first electric leakage type and the first electric leakage degree. Finally, the first electric leakage type and the first electric leakage degree are matched and integrated into the real-time electric leakage information, thereby realizing rapid positioning and severity assessment of electric leakage fault.

[0051] Step S400, in cooperation with the real-time state sequence and the real-time electric leakage information, obtains a real-time fault list of the power market, wherein the real-time fault list comprises a plurality of faults with device and degree identification.

[0052] Preferably, the real-time state sequence is matched with the real-time leakage information, specifically, the current vector in the leakage information is matched with the equipment operation data in the real-time state sequence to locate the specific faulty equipment, and the influence range of the fault is evaluated by combining the real-time state of the equipment (such as the tripping signal of the switch, the temperature anomaly) and the leakage degree, so as to convert the abstract leakage information into the fault identification of the specific physical equipment, and then each fault record is formatted in the format of equipment identification + degree identification to obtain the real-time fault list of the power market, wherein an exemplary real-time fault list of the power market is shown in Table 2:

[0053] Table 2 Real-time fault list of power market (part of example)

[0054]

[0055] Step S400 further comprises step S410 of traversing the real-time state sequence in the historical database to obtain a traversal result; step S420 of analyzing the traversal result to obtain a most matched data set, wherein the most matched data set comprises a most matched fault condition; and step S430 of adding the most matched fault condition to the real-time fault list.

[0056] Preferably, the real-time state sequence is traversed in the historical database, that is, the current real-time state data of the power equipment is used to traverse and match in the fault case library (historical database) to determine the most similar historical fault and add it to the real-time fault list to assist in more accurate judgment and positioning of the current real-time fault. Specifically, the current real-time state data is compared with the historical fault data in the historical database one by one to determine which historical fault data features are similar, and the matching degree (such as similarity of 80%, 60%, etc.) is recorded. The most similar historical fault data in the traversal result and the real-time state are analyzed to form a most matched data set. For example, the real-time data is "line M current suddenly rises with intermittent arc characteristics", after traversing the historical database, it is found that there are three historical faults: "line Z tree obstacle short circuit (similarity 90%) ", "line X lightning trip (similarity 70%) ", and "transformer D internal discharge (similarity 40%) ". Therefore, "line Z tree obstacle short circuit" is taken as the most matched fault condition and added to the real-time fault list to ensure that the fault diagnosis is more intelligent and the processing is more efficient.

[0057] Step S500, acquiring a predetermined equipment weight, and combining the operation data set in the multi-source data set to analyze a first fault to obtain a first priority index, wherein the first fault refers to any one of the plurality of faults with equipment and degree identification.

[0058] Preferably, a predetermined equipment weight, i.e., an importance score of the power equipment, is set based on the grid topology (equipment location), power supply responsibility (number of users), equipment value, etc., reflecting the criticality of the equipment in the grid. For example, substations in the core area of ​​a city and main lines supplying power to residents have higher weights (e.g., 0.8, 0.9), while small branches in remote areas and non-critical auxiliary equipment have lower weights (e.g., 0.3, 0.4). The first fault is analyzed in combination with the operation data set in the multi-source data set, wherein the first fault refers to any one of multiple faults with equipment and severity identifiers. For example, the weight of the equipment itself and the fault severity reflected by the current operation data (e.g., real-time current exceeds the standard by 200%, triggering a protection alarm) are weighted and calculated. The priority index = equipment weight × basic coefficient + operation data abnormality × impact coefficient, wherein the basic coefficient reflects the inherent importance of the equipment and the impact coefficient reflects the impact of the current fault on the operation, thereby obtaining a first priority index corresponding to the first fault, thereby ensuring the response and operation efficiency of the power equipment fault.

[0059] Step S600: Perform an operation and maintenance response to the first fault based on the first priority index.

[0060] Preferably, the fault levels of all faults are determined based on the first priority index mapping, and a processing sequence table is obtained. The response speed, resource input, and handling process are matched to different fault levels. For example, if transformer T3 internal leakage (priority index 1.2) is determined to be an urgent fault, a repair team is dispatched within 10 minutes, a backup transformer and inspection vehicle are deployed, and the power supply to the faulty transformer is immediately remotely cut off (to avoid explosion). The backup transformer is replaced first to restore power, and then the faulty equipment is repaired. If switch station K5 busbar arc leakage (index 0.85) is determined to be a general fault, an operations and maintenance team is dispatched within 2 hours, equipped with conventional detection tools such as multimeters and infrared imaging. The fault is included in the next-day inspection plan and handled during a power outage, or at night when load is low, a short power outage can be arranged for repair. By dynamically calculating the priority index, critical faults are prioritized, grid restoration efficiency is improved, and the automation level and reliability of power equipment operation and maintenance are significantly enhanced, thereby ensuring the safe and efficient operation of smart grids and power markets.

[0061] In the above, refer to Figure 1 The operation and maintenance rapid response method under power market monitoring according to an embodiment of the present invention is described in detail. Figure 2 The operation and maintenance rapid response system under power market monitoring according to an embodiment of the present invention is described.

[0062] The operation and maintenance rapid response system under the power market monitoring according to the embodiment of the present application is used to solve the technical problems of scattered multi-source power data, lagging fault identification, insufficient composite fault diagnosis precision and unclear operation and maintenance response priority in the prior art, and achieves the technical effects of real-time accurate monitoring, rapid fault positioning, intelligent priority sorting and efficient operation and maintenance response. Figure 2 As shown in FIG. 1, the operation and maintenance rapid response system under the power market monitoring includes a continuity monitoring module 10, an identification analysis module 20, a criterion analysis module 30, a real-time fault list obtaining module 40, a first fault analysis module 50 and an operation and maintenance response module 60.

[0063] The continuity monitoring module 10 is used to continuously monitor the power market to obtain a multi-source data set; the identification analysis module 20 is used to call a state sequence identification plan to perform identification analysis on a state data set in the multi-source data set to obtain a real-time state sequence; the criterion analysis module 30 is used to call a composite current criterion plan to perform criterion analysis on a current data set in the multi-source data set to obtain real-time leakage information; the real-time fault list obtaining module 40 is used to obtain a real-time fault list of the power market in cooperation with the real-time state sequence and the real-time leakage information, wherein the real-time fault list includes a plurality of faults with device and degree identifiers; the first fault analysis module 50 is used to obtain a predetermined device weight and, in combination with an operation data set in the multi-source data set, perform analysis on a first fault to obtain a first priority index, wherein the first fault refers to any one of the plurality of faults with device and degree identifiers; and the operation and maintenance response module 60 is used to perform operation and maintenance response on the first fault based on the first priority index.

[0064] Next, the specific configuration of the continuity monitoring module 10 will be described in detail. The continuity monitoring module 10 further includes: performing state monitoring on a power system in the power market according to a physical memory mapping mechanism to obtain the state data set; obtaining a substation of the power market and performing current information monitoring on a long section of power cable of the substation to obtain the current data set; performing operation monitoring on the power market to obtain the operation data set; and the state data set, the current data set and the operation data set constitute the multi-source data set.

[0065] Next, the specific configuration of the continuity monitoring module 10 will be described in detail. The continuity monitoring module 10 further includes: obtaining secondary equipment of the power system; obtaining a memory address of the secondary equipment according to the physical memory mapping mechanism; and accessing the memory address to obtain the state data set.

[0066] Below, the specific configuration of the identification analysis module 20 will be described in detail. The identification analysis module 20 further comprises: extracting a first state parameter in the state data set; standardizing the first state parameter according to the parameter standard strategy in the scenario identified from the state sequence, to obtain a first sequence value; and assembling the real-time state sequence based on the first sequence value.

[0067] Below, the specific configuration of the identification analysis module 20 will be described in detail. The identification analysis module 20 further comprises: analyzing historical secondary equipment state records and assembling a historical database, and constructing a state prediction model according to the historical database; performing prediction analysis on the state data set through the state prediction model to obtain a predicted state sequence; analyzing the predicted state sequence to obtain a state sequence limit value, and verifying the real-time state sequence with the state sequence limit value; wherein, it comprises: extracting a first data set of a first historical stage in the historical database, wherein the first data set comprises a first historical state data set and a first historical fault condition; matching a first predetermined state sequence corresponding to the first historical fault condition, and assembling a training data set in combination with the first historical state data set; and performing supervised learning on the training data set to obtain the state prediction model.

[0068] Below, the specific configuration of the real-time fault list obtaining module 40 will be described in detail. The real-time fault list obtaining module 40 further comprises: traversing the real-time state sequence in the historical database to obtain a traversal result; analyzing the traversal result to obtain a most matched data set, wherein the most matched data set comprises a most matched fault condition; and adding the most matched fault condition to the real-time fault list.

[0069] Below, the specific configuration of the criterion analysis module 30 will be described in detail. The criterion analysis module 30 further comprises: obtaining a first element in the predetermined element, and matching a first current parameter corresponding to the first element in the current data set; forming a current vector based on the first current parameter, and calculating a current similarity index of the current vector and a reference current vector;

[0070] If the current similarity index does not reach a predetermined similarity limit value, the real-time leakage information is identified and determined according to a composite current criterion scenario; wherein the predetermined element at least comprises a steady-state residual current element, a residual current increment element, a steady-state residual current differential element and a single-phase residual current differential element.

[0071] Next, the specific configuration of the criterion analysis module 30 will be described in detail. The criterion analysis module 30 further comprises: establishing an electric leakage database, wherein the electric leakage database comprises a plurality of current vectors with electric leakage type and electric leakage degree identification; calculating a first similarity of the current vector and a first vector, wherein the first vector refers to any one of the plurality of current vectors with electric leakage type and electric leakage degree identification; matching a first electric leakage type and a first electric leakage degree corresponding to the first vector if the first similarity reaches the predetermined similarity limit; and obtaining the real-time electric leakage information based on the first electric leakage type and the first electric leakage degree.

[0072] The operation and maintenance rapid response system under the power market monitoring provided in the embodiments of the present application can execute the operation and maintenance rapid response method under the power market monitoring provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0073] Based on the foregoing embodiments, the embodiments of the present application further provide a computer readable storage medium, which stores a computer program. When the computer program is executed by a processor, the operation and maintenance rapid response method under the power market monitoring as described in any of the foregoing embodiments can be implemented.

[0074] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server, and the various units and modules are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be implemented; in addition, the specific names of the functional units are only for easy mutual differentiation, and do not limit the protection scope of the present application.

[0075] The specific embodiments described above do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A rapid operation and maintenance response method under power market monitoring, characterized in that: include: Conduct continuous monitoring of the electricity market to obtain multi-source data sets; Retrieving a state sequence recognition plan to perform recognition and analysis on the state data set in the multi-source data set to obtain a real-time state sequence; Retrieving a composite current criterion plan to perform criterion analysis on the current data set in the multi-source data set to obtain real-time leakage information; Cooperating the real-time status sequence with the real-time leakage information to obtain a real-time fault list of the power market, wherein the real-time fault list includes a plurality of faults with equipment and degree identifiers; Obtaining a predetermined device weight, and analyzing a first fault in combination with an operating data set in the multi-source data set to obtain a first priority index, wherein the first fault is any one of the multiple faults with device and degree identifiers; Performing an operation and maintenance response to the first fault based on the first priority index; Continuous monitoring of the electricity market, generating multi-source datasets including: Performing status monitoring on the power system in the power market according to a physical memory mapping mechanism to obtain the status data set; Obtaining a substation in the power market, and monitoring current information of a long section of a power cable in the substation to obtain the current data set; Performing operation monitoring on the power market to obtain the operation data set; The state data set, the current data set and the operation data set constitute the multi-source data set; Performing status monitoring on the power system in the power market according to the physical memory mapping mechanism to obtain the status data set includes: obtaining secondary equipment of the power system; Acquire the memory address of the secondary device according to the physical memory mapping mechanism; Accessing the memory address to obtain the state data set; Retrieving a state sequence identification plan to identify and analyze the state data set in the multi-source data set to obtain a real-time state sequence, including: extracting a first state parameter from the state data set; Standardize the first state parameter according to the parameter standard strategy in the state sequence identification plan to obtain a first sequence value; Build the real-time status sequence based on the first sequence value; The current data set includes current parameters of predetermined components, and a composite current criterion plan is called to perform criterion analysis on the current data set in the multi-source data set to obtain real-time leakage information, including: Acquire a first element from the predetermined elements, and match a first current parameter corresponding to the first element in the current data set; forming a current vector based on the first current parameter, and calculating a current similarity index between the current vector and a reference current vector; If the current similarity index does not reach the predetermined similarity limit, identifying and determining the real-time leakage information according to the composite current criterion plan; Wherein, the predetermined elements include at least a steady-state residual current element, a residual current increment element, a steady-state residual current differential element and a single-phase residual current differential element; If the current similarity index does not reach the predetermined similarity limit, the real-time leakage information is identified and determined according to the composite current criterion plan, including: Establishing a leakage database, wherein the leakage database includes a plurality of current vectors with leakage type and leakage degree identification; Calculating a first similarity between the current vector and a first vector, wherein the first vector is any one of the multiple current vectors having leakage type and leakage degree identifiers; If the first similarity reaches the predetermined similarity limit, matching the first leakage type and the first leakage degree corresponding to the first vector; The real-time leakage information is obtained based on the first leakage type and the first leakage degree.

2. The method for rapid operation and maintenance response under power market monitoring according to claim 1, characterized in that: After forming the real-time status sequence based on the first sequence value, the method further includes: Analyze historical secondary equipment status records and build a historical database, and construct a status prediction model based on the historical database; Performing prediction analysis on the state data set using the state prediction model to obtain a predicted state sequence; Analyzing the predicted state sequence to obtain a state sequence limit value, and verifying the real-time state sequence with the state sequence limit value; Among them, include: Extracting a first data set of a first historical stage from the historical database, wherein the first data set includes a first historical state data set and a first historical fault condition; Matching a first predetermined state sequence corresponding to the first historical fault condition, and forming a training data set in combination with the first historical state data set; Supervised learning is performed on the training data set to obtain the state prediction model.

3. The method for rapid operation and maintenance response under power market monitoring according to claim 2, characterized in that: The real-time status sequence and the real-time leakage information are coordinated to obtain a real-time fault list of the power market, including: Traversing the real-time state sequence in the historical database to obtain a traversal result; Analyzing the traversal results to obtain a best-matching data set, wherein the best-matching data set includes a best-matching fault condition; The best matching fault condition is added to the real-time fault list.

4. A rapid response system for operation and maintenance under power market monitoring, characterized in that: The system is used to implement the operation and maintenance rapid response method under power market monitoring according to any one of claims 1 to 3, and the system includes: Continuity monitoring module, used to continuously monitor the power market and obtain multi-source data sets; An identification and analysis module is used to call a state sequence identification plan to identify and analyze the state data set in the multi-source data set to obtain a real-time state sequence; A criterion analysis module is used to call a composite current criterion plan to perform criterion analysis on the current data set in the multi-source data set to obtain real-time leakage information; A real-time fault list obtaining module, configured to coordinate the real-time status sequence and the real-time leakage information to obtain a real-time fault list of the power market, wherein the real-time fault list includes a plurality of faults with equipment and degree identifiers; a first fault analysis module, configured to obtain a predetermined device weight and analyze a first fault in combination with an operating data set in the multi-source data set to obtain a first priority index, wherein the first fault is any one of the multiple faults having device and degree identifiers; An operation and maintenance response module is used to perform an operation and maintenance response to the first fault based on the first priority index.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the operation and maintenance rapid response method under power market monitoring according to any one of claims 1 to 3 is implemented.

Images