Distributed collaborative modular rapid detection method and device for box-type substation

Abstract

This invention discloses a distributed, collaborative, modular, and rapid detection method and device for prefabricated substations, comprising the following steps: each distributed module determines whether it has a potential fault based on collected operational data; the distributed module with a potential fault initiates an anomaly signaling; upon receiving the anomaly signaling, other distributed modules confirm whether to fill in their own correction feature parameters at the corresponding position in the anomaly signaling, and the anomaly signaling is sequentially transmitted among all other distributed modules; all obtained correction feature parameters are summarized, the distributed module that initiated the anomaly signaling calculates the posterior fault probability, and when the posterior fault probability exceeds a set threshold, a fault is determined to exist. The distributed architecture of this invention avoids the bottleneck of centralized systems, with each module processing in parallel, improving detection efficiency. Through a feature association model, interference from redundant information is avoided, improving the accuracy of fault determination.

Description

Technical Field

[0001] This invention belongs to the field of testing, and particularly relates to a modular rapid testing method and device for distributed collaborative testing of prefabricated substations. Background Art

[0002] With the expansion and increasing complexity of power systems, prefabricated substations are widely used in distribution networks and new energy access systems due to their advantages such as convenient installation, flexible expansion, and easy maintenance. However, due to their distributed and modular characteristics, there are complex correlations between the states and performance of each module. Traditional centralized fault diagnosis methods struggle to detect and locate faults in a timely and accurate manner, and cannot fully utilize the correlation information between modules. For example, patent CN117741305A discloses a fault detection system and method for substations, relating to the field of power testing technology. The system includes: a distributed data acquisition module, located in a key area of ​​the substation, comprising multiple electromagnetic induction sensors for collecting real-time operating status information of the substation; and a data processing module, connected to the distributed data acquisition module via a preset anti-interference communication line, for receiving the real-time operating status information and generating fault detection results based on the real-time operating status information. Therefore, there is an urgent need for a fault diagnosis method based on a distributed architecture and feature correlation to achieve rapid, efficient, and accurate detection and fault diagnosis of prefabricated substations. Summary of the Invention

[0003] To address the shortcomings of the existing technology, this invention provides a modular rapid detection method for distributed collaborative operation of prefabricated substations, the method comprising the following steps:

[0004] Each distributed module determines whether it is likely to fail based on the collected operational data;

[0005] A distributed module that is at risk of failure initiates an abnormal signaling message;

[0006] After receiving the abnormal signaling, other distributed modules confirm whether to fill in their own correction characteristic parameters in the corresponding position of the abnormal signaling. The abnormal signaling is then transmitted sequentially among all other distributed modules.

[0007] The distributed module that initiated the abnormal signaling calculates the posterior fault probability by summarizing all the obtained correction feature parameters, and determines that the distributed module has a fault when the posterior fault probability exceeds a set threshold.

[0008] Specifically, a distributed detection network is constructed for each of the distributed modules, including deploying intelligent detection nodes and building self-organizing network connections.

[0009] Each distributed module acquires the time-synchronized collection data and determines whether it has the possibility of failure based on the collection data.

[0010] Each distributed module independently collects its own operational data, performs preprocessing, and calculates and corrects the key feature parameters to obtain the corrected feature parameters.

[0011] Each distributed module determines whether it is likely to malfunction based on the aforementioned correction feature parameters;

[0012] A distributed module that is at risk of failure will set itself as the head node, initiate an abnormal signaling, and request other distributed modules to provide correction characteristic parameters.

[0013] The distributed modules include transformer modules, insulation performance modules, dielectric performance modules, high-voltage switch modules, and surge arrester modules.

[0014] Key characteristic parameters of transformer modules include DC resistance and turns ratio; key characteristic parameters of insulation performance modules include insulation resistance; key characteristic parameters of dielectric performance modules include dielectric loss factor; key characteristic parameters of high-voltage switch modules include operating time, speed, and acceleration; key characteristic parameters of surge arrester modules include reference voltage.

[0015] The corrected characteristic parameters obtained by calculating and correcting the key characteristic parameters satisfy the following relationship:

[0016]

[0017] In the formula, : Corrected DC resistance Measured DC resistance Temperature coefficient of resistance Reference temperature Measured temperature : Corrected transformer ratio Measured transformer ratio Temperature coefficient of change Corrected insulation resistance Measured insulation resistance Temperature influence coefficient of insulation resistance Insulation resistance humidity effect coefficient Insulation resistance influenced by atmospheric pressure. Measured and reference relative humidity Measured and reference atmospheric pressures Corrected dielectric loss factor Temperature influence coefficient of dielectric loss factor Measured dielectric loss factor Actual operation time Signal transmission and device response delay, : The time when the operation command was issued Contact action time : Corrected reference voltage Measured voltage These are the influence coefficients of voltage and temperature, voltage and atmospheric pressure, and voltage and humidity, respectively.

[0018] Upon receiving the abnormal signaling, the other distributed modules confirm their association with the head node.

[0019] The correction feature parameters of distributed modules whose correlation exceeds the correlation threshold are filled into the corresponding positions in the abnormal signaling.

[0020] The abnormal signaling is transmitted sequentially among other distributed modules whose association with the head node exceeds the association threshold, and is finally returned to the head node by the tail node.

[0021] Among them, the other distributed modules rely on the internal feature association matrix Confirm the degree of association between itself and the head node.

[0022] Among them, other distributed modules look up the feature association matrix based on their internally stored feature association models. The degree of association between itself and the head node ;

[0023] if If the value is greater than the predetermined association threshold, it indicates that the module... With head node module It exhibits strong feature correlation and participates in data filling for abnormal signaling;

[0024] Otherwise, it is considered to be unrelated to the fault and will not participate in the data filling of abnormal signaling.

[0025] The head node summarizes all the corrected feature parameters. Then, using Bayes' theorem, combined with prior failure probabilities and observational data, the posterior failure probability is calculated. ;

[0026] A head node whose posterior failure probability exceeds a set threshold is considered to have a fault, triggering an alarm.

[0027] The head node is based on the posterior fault probability. To determine if you have a malfunction, follow these steps:

[0028] Determine the prior probability ;

[0029] Calculate the weights of the corrected characteristic parameters ;

[0030] Calculate the conditional probability density function of the corrected feature parameters;

[0031] Calculate the weighted log-likelihood function;

[0032] Calculate the likelihood ratio and apply Bayes' theorem to calculate the posterior failure probability. ;

[0033] Set a posterior probability threshold , like Then determine the distributed module. A malfunction occurred; if Then determine the distributed module Normal operation.

[0034] Among them, weight The calculation formula is:

[0035] ,

[0036] Correction characteristic parameters The weights;

[0037] Distributed module With correction characteristic parameters Belongs to distributed module Feature correlation degree;

[0038] Correction characteristic parameters Importance coefficient;

[0039] The calculation of the conditional probability density function for the corrected feature parameters includes determining:

[0040] Under fault conditions: ;

[0041] Under normal conditions: ;

[0042] in:

[0043] Correction characteristic parameters Conditional probability density functions PDF under fault and normal conditions respectively;

[0044] The actual measured value of the characteristic parameter;

[0045] Correction characteristic parameters Mean and standard deviation under fault conditions;

[0046] Correction characteristic parameters Mean and standard deviation under normal conditions;

[0047] The natural exponential function, i.e. ;

[0048] The calculation of the weighted log-likelihood function includes determining:

[0049] Log-likelihood function under fault conditions: ;

[0050] Log-likelihood function under normal conditions: ;

[0051] in:

[0052] : Observed under fault conditions The log-likelihood value;

[0053] Observed under normal conditions The log-likelihood value;

[0054] Correction characteristic parameters The weights;

[0055] and PDF values ​​of the feature parameters;

[0056] This involves calculating the likelihood ratio and applying Bayes' theorem to calculate the posterior failure probability. Specifically, it includes:

[0057] Calculate the likelihood ratio :

[0058] ,

[0059] in:

[0060] Likelihood ratio is the ratio of the probability of observing data under fault conditions to the probability of observing data under normal conditions.

[0061] Calculate the posterior probability of failure :

[0062] ,

[0063] in:

[0064] Given observation data Then, the distributed module The posterior probability of failure;

[0065] Likelihood ratio;

[0066] Distributed module The prior probability of failure occurring;

[0067] Distributed module The prior probability of normal operation;

[0068] Where, the likelihood function Indicates in the parameter Below, observed data possibility.

[0069] This invention discloses a modular rapid detection device for distributed collaborative operation of prefabricated substations, employing the aforementioned modular rapid detection method for distributed collaborative operation of prefabricated substations. The device includes:

[0070] The probability assessment module is used to determine the possibility of a fault based on the collected operational data.

[0071] The signaling initiation module is used to initiate abnormal signaling when there is a possibility of failure.

[0072] The signaling determination module is used to confirm whether its own correction characteristic parameters are filled in the corresponding position in the abnormal signaling;

[0073] The fault determination module is used to summarize all the obtained correction feature parameters, calculate the posterior fault probability, and determine that a fault exists when the posterior fault probability exceeds a set threshold.

[0074] The present invention also discloses a rapid detection system for prefabricated substations, the system comprising a communication module, a memory, and a processor, the memory storing a computer program, and the processor executing the computer program to implement the above-described method.

[0075] In this invention, each module independently acquires and preprocesses data, calculating key feature parameters in real time, reducing centralized data processing time and accelerating detection speed. Each module immediately determines the likelihood of a fault based on its own corrected feature parameters, without waiting for global information, resulting in a rapid response. When a fault possibility is detected, the module proactively initiates anomaly signaling, communicating only with associated modules, reducing communication latency and data transmission volume, and quickly acquiring relevant data. Utilizing Bayes' theorem, combined with its own and associated module's corrected feature parameters, the fault probability is rapidly calculated, achieving efficient fault determination.

[0076] This invention's distributed architecture avoids the bottlenecks of centralized systems, allowing modules to process data in parallel and improving detection efficiency. Through a feature association model, it focuses only on data from relevant modules, avoiding interference from redundant information and improving the accuracy of fault diagnosis. Attached Figure Description

[0077] The above and other objects, features, and advantages of exemplary embodiments of the present disclosure will become readily apparent upon reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of the present disclosure are illustrated by way of example and not limitation, and like or corresponding reference numerals denote like or corresponding parts, wherein:

[0078] Figure 1 This is a flowchart illustrating a modular rapid detection method for distributed coordination in a prefabricated substation according to an embodiment of the present invention.

[0079] Figure 2 This is a schematic diagram illustrating a modular rapid detection device for distributed coordination in a prefabricated substation according to an embodiment of the present invention. Detailed Implementation

[0080] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0081] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0082] It should be understood that although the terms first, second, third, etc., may be used to describe... in the embodiments of the present invention, these... should not be limited to these terms. These terms are only used to distinguish... For example, first... may also be referred to as second... without departing from the scope of the embodiments of the present invention, and similarly, second... may also be referred to as first...

[0083] It should be understood that the term "and / or" as used herein is merely a description of the relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0084] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”

[0085] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0086] The purpose of this invention is to provide a distributed, collaborative, modular, and rapid detection method for prefabricated substations. By constructing a distributed detection network and utilizing the correlation of characteristic parameters between distributed modules, a complex distributed fault diagnosis algorithm is designed to achieve rapid and accurate diagnosis and location of faults in each module, thus solving the problems of low detection efficiency and insufficient accuracy in existing technologies.

[0087] like Figure 1 As shown, this invention provides a modular rapid detection method for distributed collaborative operation of prefabricated substations, comprising the following steps:

[0088] Constructing a distributed detection network includes deploying intelligent detection nodes on distributed modules and building self-organizing network connections;

[0089] Obtain the original data of each distributed module after time synchronization;

[0090] Each distributed module determines whether it is likely to fail based on the collected operational data;

[0091] If the distributed module detects a potential failure, it initiates an exception signaling.

[0092] After receiving the abnormal signaling, other distributed modules confirm whether to fill in their own correction characteristic parameters in the corresponding position of the abnormal signaling. The abnormal signaling is then transmitted sequentially among all other distributed modules.

[0093] The distributed module that initiated the abnormal signaling calculates the posterior fault probability by summarizing all the correction feature parameters obtained by the head node. If the posterior fault probability exceeds a set threshold, the distributed module is determined to be faulty. Here, a distributed module refers to each module in a prefabricated substation; that is, a prefabricated substation includes multiple distributed modules.

[0094] This invention fully utilizes the efficiency of distributed architecture and the accuracy of Bayesian methods to achieve rapid fault detection in prefabricated substations, demonstrating the system's rapid response and intelligent diagnostic capabilities.

[0095] The distributed modules are: transformer comprehensive test module, insulation performance test module, dielectric performance test module, high voltage switch comprehensive test module, and surge arrester AC / DC test module.

[0096] Intelligent detection nodes are deployed on each module of the prefabricated substation (including transformer module, insulation performance module, dielectric performance module, high voltage switch module, surge arrester module, etc.). Each intelligent detection node has data acquisition, preprocessing, feature extraction and communication functions.

[0097] Each intelligent detection node is interconnected via wireless communication (such as ZigBee, LoRa, WiFiMesh) or wired communication to form a self-organizing network connection, that is, a self-organizing distributed detection network, to achieve real-time data transmission and sharing.

[0098] Each intelligent detection node collects raw data from the site, including voltage, current, temperature, humidity, and signal timestamps.

[0099] Each intelligent detection node uses local sensors to collect the operating data of the corresponding distributed module in real time, and uses Network Time Protocol (NTP) or Precision Time Protocol (PTP) to achieve time synchronization between nodes, and outputs the original data (original operating data) of each distributed module after time synchronization.

[0100] Data preprocessing includes: filtering and denoising the raw data to remove interference signals; detecting and eliminating outlier data points using statistical methods to ensure data quality; and normalizing the data to unify the data format and units.

[0101] Each distributed module independently collects its own operational data, preprocesses it, and calculates corrected key feature parameters (i.e., corrected feature parameters). Based on these key feature parameters, each distributed module determines whether it is likely to fail. If a distributed module is likely to fail, it sets itself as the head node, initiates an anomaly signaling, and requests corrected feature parameters from other distributed modules. Upon receiving the anomaly signaling, other distributed modules, based on their internal feature correlation matrices... The system checks its correlation with the head node. If the correlation is high, it fills in its own correction feature parameters in the corresponding position in the signaling. Abnormal signaling is transmitted sequentially among other distributed modules whose correlation with the head node exceeds the correlation threshold, and finally returned to the head node by the tail node. The head node summarizes all the correction feature parameters obtained. Then, using Bayes' theorem, combined with prior failure probabilities and observational data, the posterior failure probability is calculated. If the posterior fault probability exceeds the set threshold, the fault is determined to be valid, and an alarm is triggered.

[0102] In this invention, each distributed module independently collects and preprocesses data, calculating key feature parameters in real time, reducing centralized data processing time and accelerating detection speed. Each distributed module immediately determines the potential for a fault based on its own corrected feature parameters, without waiting for global information, resulting in a rapid response. When a potential fault is detected, the distributed module proactively initiates anomaly signaling, communicating only with associated distributed modules, reducing communication latency and data transmission volume, and quickly acquiring relevant data. Utilizing Bayes' theorem, combined with its own and associated module's corrected feature parameters, the fault probability is rapidly calculated, achieving efficient fault determination.

[0103] This invention's distributed architecture avoids the bottlenecks of centralized systems, with each distributed module processing in parallel, improving detection efficiency. Through a feature association model, it focuses only on data from relevant modules, avoiding interference from redundant information and improving the accuracy of fault diagnosis.

[0104] This invention leverages the efficiency of distributed architecture and the accuracy of Bayesian methods to achieve rapid fault detection in prefabricated substations, demonstrating the system's rapid response and intelligent diagnostic capabilities.

[0105] Each intelligent detection node calculates the key feature parameters of its distributed module based on the preprocessed data. The corresponding key feature parameters of this module are as follows:

[0106] Transformer module: DC resistor Transformation ratio .

[0107] Insulation performance module: Insulation resistance .

[0108] Dielectric performance module: Dielectric loss factor .

[0109] High-voltage switch module: operating time ,speed acceleration .

[0110] Surge arrester module: reference voltage

[0111] Then, considering the influence of environmental factors such as temperature, humidity, and atmospheric pressure on the characteristic parameters, corrections are made to obtain the corrected characteristic parameters.

[0112] Optionally, the methods for determining and calibrating the key characteristic parameters of each distributed module include:

[0113] For the DC resistance of the transformer module The calculation and correction include:

[0114] The method for measuring DC resistance is as follows: Use a DC resistance tester to measure the DC resistance of the transformer windings to obtain the measured DC resistance.

[0115] The effect of temperature on resistance: Resistance changes with temperature, and can be corrected using a temperature coefficient. The correction method is as follows:

[0116] ,

[0117] in:

[0118] Corrected DC resistance

[0119] Measured DC resistance

[0120] Temperature coefficient of resistance (for) )

[0121] Reference temperature (take) )

[0122] : Actual measured temperature.

[0123] For the turns ratio of the transformer module The calculation and correction include:

[0124] The actual measurement method for turns ratio: Use a turns ratio tester to measure the primary voltage. and secondary voltage Calculate the measured transformer ratio: .

[0125] in:

[0126] Measured transformer ratio

[0127] Primary voltage

[0128] Secondary voltage (V)

[0129] Temperature can affect the winding dimensions and material properties, thus affecting the turns ratio. Correction methods:

[0130] ,

[0131] in:

[0132] Corrected turns ratio

[0133] Measured transformer ratio

[0134] The turns ratio temperature coefficient (determined based on the transformer material and structure)

[0135] Reference temperature (take) )

[0136] : Actual measured temperature.

[0137] For the insulation resistance of the insulation performance module The calculation and correction include:

[0138] The measured insulation resistance is obtained by applying a test voltage using an insulation resistance tester (megohmmeter). .

[0139] The influencing factor for environmental factor correction is temperature. ,humidity Atmospheric pressure The correction method is as follows:

[0140] ,

[0141] in:

[0142] Corrected insulation resistance

[0143] Measured insulation resistance ( )

[0144] Temperature influence coefficient

[0145] Humidity Influence Coefficient

[0146] Atmospheric pressure influence coefficient

[0147] Measured and reference relative humidity (%)

[0148] Measured and reference atmospheric pressure

[0149] Reference temperature (take) )

[0150] : Actual measured temperature.

[0151] For dielectric loss factor of dielectric performance module The calculation and correction include:

[0152] The dielectric loss factor represents the loss characteristics of a dielectric material and is defined as the ratio of dielectric loss current to capacitance current. It is measured using a dielectric loss tester to measure the active power of the sample under AC voltage. and voltage Current Calculation formula: or .

[0153] in:

[0154] Measured dielectric loss factor

[0155] Power loss (W)

[0156] Loss current (A)

[0157] Angular frequency ( ,in For frequency )

[0158] Capacitance (F)

[0159] Test voltage

[0160] Temperature affects the loss characteristics of the medium; correction formula:

[0161] .

[0162] in:

[0163] Corrected dielectric loss factor

[0164] Temperature influence coefficient of dielectric loss factor

[0165] Measured dielectric loss factor

[0166] Reference temperature (take) )

[0167] : Actual measured temperature.

[0168] Operating time of high voltage switch module The calculation and measurement method involves recording the time when the operation command is issued. and contact action time .

[0169] Calculation formula:

[0170] in:

[0171] Actual operation time (s)

[0172] Signal transmission and device response delay (s) can be obtained through pre-calibration.

[0173] speed and acceleration The calculation involves acquiring motion travel data via a displacement sensor to obtain the change in the displacement of the switch contacts over time. .

[0174] Speed ​​calculation formula: .

[0175] Discrete data computation: .

[0176] Acceleration calculation formula:

[0177] Discrete data computation:

[0178] The mechanical motion characteristics of high-voltage switches are generally less affected by ambient temperature, therefore no environmental correction is required.

[0179] Reference voltage of surge arrester module The calculation and correction involve applying a standard voltage to the surge arrester, measuring the current flowing through the arrester, and finding the voltage at a specific current. .

[0180] Factors affecting environmental correction: Temperature Atmospheric pressure ,humidity .

[0181] The correction formula is

[0182] in:

[0183] Corrected reference voltage

[0184] Measured voltage

[0185] These are the influence coefficients of voltage and temperature, voltage and atmospheric pressure, and voltage and humidity, respectively.

[0186] Other parameters are the same as above.

[0187] Each module collects and corrects the feature parameters, forming its own feature parameter set. :

[0188] Transformer module: ;

[0189] Insulation performance module: ;

[0190] Dielectric performance module: ;

[0191] High-voltage switch module: ;

[0192] Surge arrester module: ;

[0193] All environmental parameters , , All of these are obtained through real-time measurements using environmental sensors.

[0194] Reference conditions , Standard conditions are generally used: temperature Relative humidity Atmospheric pressure Influence coefficient , , , , , , , The specific requirements need to be determined based on the characteristics of the equipment, the properties of the materials, and the experimental data.

[0195] Using the above calculation formula, each detection node can calculate and correct the key feature parameters of its module based on the preprocessed data, thus obtaining the corrected feature parameter set. These parameters not only reflect the operating status of the equipment but also eliminate the influence of environmental factors, providing an accurate and reliable data foundation for subsequent feature correlation analysis and fault diagnosis.

[0196] Each distributed module (including transformer module, insulation performance module, dielectric performance module, high voltage switch module, surge arrester module, etc.) uses local sensors to collect its own operating data in real time, such as voltage, current, temperature, humidity, etc., and achieves time synchronization through Network Time Protocol (NTP) or Precision Time Protocol (PTP).

[0197] The collected raw data undergoes preprocessing, including filtering, noise reduction, outlier removal, and normalization, to ensure data quality and consistency. Correction feature parameters are calculated, including adjustments based on the preprocessed data and environmental factors (temperature, humidity, atmospheric pressure, etc.). This process calculates the key feature parameters for this module, resulting in a corrected set of feature parameters. .

[0198] Next, the likelihood of a fault is determined, including each module based on its own calibration characteristic parameters. The system determines whether a fault is possible. Specifically, it compares the calibrated characteristic parameters with pre-set normal operating thresholds or standard ranges. If all characteristic parameters are within the normal range, no fault is considered possible, and normal operation continues. If some characteristic parameters exceed the normal range and surpass a preset fault trigger threshold, a fault is considered possible.

[0199] In addition, if there is a possibility of failure, an abnormal signaling request needs to be initiated, including:

[0200] Identifying the faulty module as the head node: When a certain distributed module (denoted as distributed module) is identified as the head node... When it determines that there is a possibility of a fault, it sets itself as the "head node" and constructs abnormal signaling.

[0201] The created abnormal signaling message includes:

[0202] Header node identifier: indicates that the signaling was sent by a distributed module. Initiated.

[0203] Anomaly type and description: Indicates the detected abnormal characteristic parameters and the out-of-range conditions.

[0204] Data population structure: A predefined data structure (such as an array or list) used by other distributed modules to populate correction feature parameters. This data structure contains information related to the distributed modules. Location of other related distributed modules.

[0205] Send an exception signaling request to other distributed modules. Distributed modules Anomaly signals are sent to other distributed modules via a distributed detection network, following a specific path.

[0206] After the abnormal signaling message is constructed, it is sent to other distributed modules. These other distributed modules (denoted as modules)... Upon receiving an abnormal signaling message, the following actions will be taken:

[0207] Read header node information: Identify the header node identifier in the signaling. .

[0208] Feature correlation judgment:

[0209] Based on the internally stored feature association model, find the feature association matrix. The corresponding correlation .

[0210] if If the value is greater than the predetermined association threshold, it indicates that the distributed module... Distributed modules with head nodes There is a strong correlation between their features.

[0211] If the correlation is insufficient, it is considered to have no direct connection with the fault and will not be included in the data filling.

[0212] Enter the correction characteristic parameters:

[0213] For distributed modules associated with the head node , to its own correction characteristic parameters Fill in the corresponding position in the abnormal signaling.

[0214] After processing the abnormal signaling, the distributed module The signaling is then passed to the next distributed module, continuing the transmission of abnormal signaling.

[0215] The transmission methods for abnormal signaling can be:

[0216] Sequential transmission: Transmitted sequentially according to the pre-defined distributed module sequence number or topological order.

[0217] Route-based delivery: Select the next-hop distributed module based on network topology and routing protocol.

[0218] The termination condition for abnormal signaling transmission is that all distributed modules associated with the head node have received the abnormal signaling and filled in the data.

[0219] After the signaling transmission is completed, the last distributed module that processed the abnormal signaling (the tail node) returns the abnormal signaling to the head node module. This means the tail node returns an abnormal signaling.

[0220] The return method can be:

[0221] Direct return: Send directly to the head node over the network.

[0222] Alternatively, reverse transmission: return in reverse order of signaling transmission.

[0223] Head node corresponds to distributed module Upon receiving the returned abnormal signaling, extract the correction feature parameters filled in by all associated distributed modules to form a correction feature parameter set. That is, the head node receives and summarizes the data.

[0224] The head node comprehensively analyzes the set of correction feature parameters based on its own and the correction feature parameters of its associated distributed modules. This includes using feature association models to further analyze faults and calculate updated fault probabilities. .

[0225] If the update failure probability If the set fault confirmation threshold is exceeded, the distributed module is deemed to be faulty. A malfunction has occurred.

[0226] If the probability of failure decreases, it may be a false alarm, and you can continue monitoring or cancel the fault alarm.

[0227] If a fault is confirmed, the head node sends a fault alarm message to the system or operations and maintenance test personnel, including the fault type, scope of impact, and suggested handling measures.

[0228] For example, suppose the transformer module (distributed module) ) Detected its own DC resistance An abnormally high level indicates a potential malfunction. Perform the following procedures sequentially:

[0229] 1. Distributed module Initiate abnormal signaling:

[0230] Set itself as the head node, create an exception signaling, and describe... Abnormal situation.

[0231] 2. Signaling transmission:

[0232] Broadcast exception signals to other modules.

[0233] 3. Insulation performance module (distributed module) Receive signaling:

[0234] Read the header node identifier as a distributed module .

[0235] Find the correlation matrix It was found that it had a high correlation with the transformer module.

[0236] Fill in its own insulation resistance Correction characteristic parameters.

[0237] 4. Handling abnormal signaling in dielectric performance modules, high-voltage switch modules, etc.:

[0238] Based on the degree of correlation, fill in the respective correction feature parameters.

[0239] 5. Signaling return header node:

[0240] The last module returns the exception signaling to the distributed module. .

[0241] 6. Distributed Module Perform fault analysis:

[0242] The resulting set of correction feature parameters .

[0243] By combining the feature correlation model, the insulation resistance was found This is also abnormal; further investigation is needed to determine the source of the fault.

[0244] Calculate the updated failure probability The threshold is exceeded.

[0245] 7. Fault diagnosis and handling:

[0246] The transformer module has been confirmed to be faulty.

[0247] An alarm message was sent, suggesting that the transformer insulation be checked.

[0248] Among them, the feature association model is used to describe the degree of association between feature parameters of different distributed modules. Association Matrix elements The larger the value, the larger the module. With modules The higher the correlation between the features, the better. The correlation can be determined based on historical data statistics, physical mechanism analysis, or expert experience.

[0249] For example, the following is a specific association matrix. Format:

[0250] The distributed modules and their key characteristic parameters are as follows:

[0251] Transformer Module (Distributed Module 1): DC Resistance , transformation ratio .

[0252] Insulation performance module (distributed module 2): Insulation resistance .

[0253] Dielectric performance module (distributed module 3): Dielectric loss factor .

[0254] High-voltage switch module (distributed module 4): Operation time ,speed acceleration .

[0255] Surge arrester module (distributed module 5): Reference voltage .

[0256] Feature Correlation Matrix It is A matrix, where elements Represents a distributed module With distributed modules The degree of correlation between features. This correlation can be represented numerically, typically ranging from [value missing]. The larger the value, the higher the correlation.

[0257] The following is an example matrix:

[0258]

[0259] Among them, self-association (diagonal elements) : indicates that each module has a correlation degree of 1 with itself, meaning it is fully correlated.

[0260] Degree of correlation between distributed modules:

[0261] Distributed Module 1 (Transformer Module) and Distributed Module 2 (Insulation Performance Module): The insulation performance of a transformer is crucial to its operation, and the DC resistance of the transformer... and ratio It may be affected by the insulation condition.

[0262] Insulation resistance It reflects the insulation condition between the transformer windings and the core, and directly affects the transformer's performance.

[0263] Therefore, the correlation between the two is relatively high.

[0264] Distributed Module 2 (Insulation Performance Module) and Distributed Module (Dielectric property module): Insulation performance and dielectric performance are closely related, and the dielectric loss factor... It reflects the dielectric loss of the insulating material. Insulation resistance. and Both are important parameters for measuring insulation status. Therefore, they are highly correlated.

[0265] Distributed Module 1 (Transformer Module) and Distributed Module 3 (Dielectric Performance Module): .

[0266] The dielectric properties of a transformer affect its insulation state, and thus its DC resistance. and ratio .

[0267] Although not as significant as the direct impact on insulation performance, there is still a strong correlation.

[0268] Distributed Module 1 and Distributed Module 4 (High Voltage Switch Module): The operating status of a transformer is somewhat related to the operation of a high-voltage switch; for example, the high-voltage switch may trip under overload or fault conditions. However, the correlation is relatively low.

[0269] Distributed Module 4 and Distributed Module 5 (Surge Arrester Module): .

[0270] High-voltage switches and surge arresters are both related to overvoltage protection.

[0271] In the event of a lightning strike or overvoltage, the surge arrester will trip, and the high-voltage switch may need to disconnect the circuit.

[0272] The two are related to some extent.

[0273] Distributed Module 1 and Module 5, which is attacked from both sides: The correlation between transformers and surge arresters is low.

[0274] Other low-correlation:

[0275] Distributed Module 2 and Distributed Module 4: ,

[0276] Distributed Module 3 and Distributed Module 4: ,

[0277] Distributed Module 2 and Distributed Module 5: ,

[0278] Distributed Module 3 and Distributed Module 5: ,

[0279] The correlation between these distributed modules is low because there is no strong direct relationship between their functions and the correction characteristic parameters.

[0280] Optionally, based on physical relationships and professional knowledge, the physical connections and working principles between the distributed modules are analyzed to determine the correlation between the correction characteristic parameters. For example, insulation performance and dielectric performance both involve the state of the insulating material and have a high degree of correlation.

[0281] Optionally, historical operational data can be used to calculate the correlation coefficient of the calibration characteristic parameters between distributed modules. If the calibration characteristic parameters of two distributed modules frequently change synchronously historically, it indicates a high degree of correlation.

[0282] In one embodiment, the head node is based on the failure probability. To determine if there is a fault, the following steps are involved:

[0283] Step 1: Determine the prior probability .

[0284] Based on historical data or equipment reliability analysis, determine the distributed module. Prior probability of failure And calculate the prior probability of normal operation. .

[0285] For example, assuming that based on historical statistics, distributed modules The prior probability of failure is Then the prior probability of normal operation is: .

[0286] Step 2: Calculate the weights of the correction feature parameters .

[0287] Weight The calculation formula is:

[0288] ,

[0289] Correction characteristic parameters The weight of the feature reflects its importance in fault diagnosis.

[0290] Distributed module With correction characteristic parameters Belongs to distributed module Feature correlation, with a value range of .

[0291] Correction characteristic parameters Importance coefficient, with a value range of The determination is based on the degree of influence of the characteristics on fault diagnosis.

[0292] The weighted sum of all corrected feature parameters is used for normalization, so that the sum of the weights is 1.

[0293] Step 3: Calculate the conditional probability density function of the corrected feature parameters.

[0294] Assume that the correction characteristic parameters follow a normal distribution under both fault and normal conditions.

[0295] Under fault conditions:

[0296] Under normal conditions:

[0297] in:

[0298] Correction characteristic parameters Probability density function (PDF) under fault and normal conditions.

[0299] The actual measured value of the correction characteristic parameter.

[0300] Correction characteristic parameters Mean and standard deviation under fault conditions.

[0301] Correction characteristic parameters Mean and standard deviation under normal conditions.

[0302] The natural exponential function, i.e. .

[0303] Step 4: Calculate the weighted log-likelihood function, including calculating:

[0304] Log-likelihood function under fault conditions: .

[0305] Log-likelihood function under normal conditions: .

[0306] in:

[0307] : Observed under fault conditions The log-likelihood value.

[0308] Observed under normal conditions The log-likelihood value.

[0309] Correction characteristic parameters The weight.

[0310] and : Correct the PDF value of the characteristic parameters (calculated in step 3).

[0311] Step 5: Calculate the likelihood ratio and apply Bayes' theorem to calculate the posterior probability. Specifically, it includes:

[0312] Step 5.1: Calculate the likelihood ratio :

[0313] ,

[0314] in:

[0315] Likelihood ratio is the ratio of the probability of observing data under fault conditions to the probability of observing data under normal conditions.

[0316] The likelihood ratio, by comparing the probability of observing data under two conditions, reflects which hypothesis the data supports more. The likelihood ratio is obtained by converting the difference in log-likelihoods into exponential form.

[0317] Step 5.2: Calculate the posterior probability .

[0318] ,

[0319] in:

[0320] Given observation data Then, the distributed module The posterior probability of a failure occurring.

[0321] Likelihood ratio (already calculated).

[0322] Distributed module The prior probability of failure.

[0323] Distributed module Prior probability of normal operation.

[0324] Where, the likelihood function Indicates in the parameter Below, observed data possibility.

[0325] In this embodiment, the likelihood functions are as follows:

[0326] : Assuming a fault occurs (parameter is) In the case of ), the observed data possibility.

[0327] Assuming the fault does not occur (parameter is...) In the case of ), the observed data possibility.

[0328] Likelihood ratio Meaning: ,

[0329] Likelihood ratio The probability of the observed data under two hypotheses was compared.

[0330] If Λ > 1, it means the data supports the hypothesis that the fault occurred.

[0331] If Λ < 1, it means the data more strongly supports the assumption that the system is functioning correctly.

[0332] Using Bayes' theorem, combined with the likelihood ratio and prior probability, the posterior probability is calculated.

[0333] Posterior probability reflects the likelihood of a distributed module failing after data has been observed.

[0334] Step 6: Fault diagnosis.

[0335] Set a posterior probability threshold The value is determined based on the specific application scenario and risk appetite. It is generally between 0.5 and 0.8.

[0336] Judgment criteria:

[0337] like Then determine the distributed module. A malfunction has occurred.

[0338] like Then determine the distributed module Normal operation.

[0339] For example, the weights of the correction feature parameters The process is as follows:

[0340] The following modules and their key characteristic parameters are identified:

[0341] Distributed Module 1 (Transformer Module):

[0342] Characteristic parameter 1: DC resistance (recorded as) )

[0343] Correction characteristic parameter 2: turns ratio (recorded as) )

[0344] Distributed Module 2 (Insulation Performance Module):

[0345] Correction characteristic parameter 3: Insulation resistance (recorded as) )

[0346] Distributed Module 3 (Dielectric Performance Module):

[0347] Correction characteristic parameter 4: Dielectric loss factor (recorded as) )

[0348] Distributed Module 4 (High Voltage Switch Module):

[0349] Correction characteristic parameter 5: Operation time (recorded as) )

[0350] Correction characteristic parameter 6: Velocity (recorded as) )

[0351] Correction characteristic parameter 7: Acceleration (recorded as) )

[0352] Distributed Module 5 (Surge Arrester Module):

[0353] Correction characteristic parameter 8: Reference voltage (recorded as) )

[0354] Assuming that, based on professional knowledge and experience, each correction feature parameter is assigned an importance coefficient as follows: :

[0355]

[0356] Determine the distributed module to which the feature parameters belong. The module number to which each feature parameter belongs is as follows:

[0357]

[0358] Next, calculate the weights. ,include:

[0359] For the target module Suppose we are assessing whether distributed module 1 (transformer module) has failed.

[0360] Calculate the correlation degree of each feature parameter multiplied by the importance coefficient. .

[0361] Based on the given feature correlation matrix and importance coefficient ,calculate:

[0362] Correction characteristic parameters (Distributed module 1, itself) , .

[0363] Correction characteristic parameters (Distributed module 1, itself) , .

[0364] Correction characteristic parameters (Distributed Module 2) , .

[0365] Correction characteristic parameters (Distributed Module 3) .

[0366] Correction characteristic parameters (Distributed Module 4) , .

[0367] Correction characteristic parameters (Distributed Module 4) , .

[0368] Correction characteristic parameters (Distributed Module 4) , .

[0369] Correction characteristic parameters (Distributed Module 5) , .

[0370] Calculate the denominator: the weighted sum of all corrected characteristic parameters. .

[0371] Calculate the weight of each correction feature parameter .

[0372] weight .

[0373] Correction characteristic parameters weight .

[0374] Correction characteristic parameters weight .

[0375] Correction characteristic parameters weight .

[0376] Correction characteristic parameters weight .

[0377] Correction characteristic parameters weight .

[0378] Correction characteristic parameters weight .

[0379] Correction characteristic parameters weight .

[0380] In this invention, each module independently acquires and preprocesses data, calculating key feature parameters in real time, reducing centralized data processing time and accelerating detection speed. Each module immediately determines the potential for a fault based on its own corrected feature parameters, without waiting for global information, resulting in a rapid response. When a potential fault is detected, the module proactively initiates anomaly signaling, communicating only with associated modules, reducing communication latency and data transmission volume, and quickly acquiring relevant data. Utilizing Bayes' theorem, combined with its own and associated module's corrected feature parameters, the fault probability is rapidly calculated, achieving efficient fault determination.

[0381] This invention's distributed architecture avoids the bottlenecks of centralized systems, allowing modules to process data in parallel and improving detection efficiency. Through a feature association model, it focuses only on data from relevant modules, avoiding interference from redundant information and improving the accuracy of fault diagnosis.

[0382] like Figure 2 As shown, this invention discloses a modular rapid detection device for distributed collaborative operation of prefabricated substations, employing the aforementioned modular rapid detection method for distributed collaborative operation of prefabricated substations. The device includes:

[0383] The probability assessment module is used to determine the possibility of a fault based on the collected operational data.

[0384] The signaling initiation module is used to initiate abnormal signaling when there is a possibility of failure.

[0385] The signaling determination module is used to confirm whether its own correction characteristic parameters are filled in the corresponding position in the abnormal signaling;

[0386] The fault determination module is used to summarize all the obtained correction feature parameters, calculate the posterior fault probability, and determine that a fault exists when the posterior fault probability exceeds a set threshold.

[0387] The present invention also discloses a rapid detection system for prefabricated substations, the system comprising a communication module, a memory, and a processor, the memory storing a computer program, and the processor executing the computer program to implement the above-described method.

[0388] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0389] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0390] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0391] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0392] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0393] The preferred embodiments of the present invention have been described above to make the spirit of the present invention clearer and easier to understand, and are not intended to limit the present invention. All modifications, substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope summarized by the appended claims.

Claims

1. A modular rapid detection method for distributed collaborative operation in a prefabricated substation, the method comprising the following steps: Each distributed module independently collects its own operational data, performs preprocessing, and calculates and corrects key feature parameters to obtain corrected feature parameters. Each distributed module determines whether it is likely to fail based on the correction feature parameters; if a distributed module is likely to fail, it sets itself as the head node, initiates an abnormal signaling, and requests other distributed modules to provide correction feature parameters. After receiving the abnormal signaling, the other distributed modules confirm their correlation with the head node; and fill the correction feature parameters of the distributed modules whose correlation exceeds the correlation threshold into the corresponding positions in the abnormal signaling. The abnormal signaling is transmitted sequentially among other distributed modules whose correlation with the head node exceeds the correlation threshold, and is finally returned to the head node by the tail node; The distributed module that initiates the abnormal signaling calculates the posterior fault probability by summarizing all the obtained correction feature parameters, and determines that a fault exists when the posterior fault probability exceeds a set threshold.

2. The modular rapid detection method for distributed collaborative operation of a prefabricated substation as described in claim 1, characterized in that, A distributed detection network is constructed for each of the distributed modules, including deploying intelligent detection nodes and building self-organizing network connections; Each distributed module acquires the operational data collected after time synchronization, and determines whether it has the possibility of failure based on the operational data.

3. The modular rapid detection method for distributed collaborative operation of a prefabricated substation as described in claim 1, characterized in that, The other distributed modules determine their correlation with the head node based on their internal feature association matrix A; Other distributed modules, based on their internally stored feature association models, search for the association degree A between themselves and the head node in the feature association matrix A. ij .

4. The modular rapid detection method for distributed collaborative operation of a prefabricated substation as described in claim 1, characterized in that, The set of all corrected feature parameters X obtained by summarizing the head nodes i Then, using Bayes' theorem, combined with prior failure probabilities and observational data, the posterior failure probability P(F) is calculated. i |X i ); A head node whose posterior failure probability exceeds a set threshold is considered to have a fault, triggering an alarm.

5. The modular rapid detection method for distributed collaborative operation of a prefabricated substation as described in claim 4, characterized in that, The head node is based on the posterior fault probability P(F). i |X i To determine if there is a fault, the following steps are involved: Determine the prior probability P(F) i ); Calculate the weights w of the corrected characteristic parameters j ; Calculate the conditional probability density function of the corrected feature parameters; Calculate the weighted log-likelihood function; Calculate the likelihood ratio and apply Bayes' theorem to calculate the posterior failure probability P(F). i |X i ); Set the posterior probability threshold P th If P(F) i |X i )≥P th If P(F) is faulty, then the distributed module i is determined to have failed; i |X i ) <P th Then it is determined that distributed module i is running normally.

6. The modular rapid detection method for distributed collaborative operation of a prefabricated substation as described in claim 5, characterized in that, in, weight w j The calculation formula is: w j Correction characteristic parameter x j The weights; A ik(j) Distributed module i and correction feature parameter x j The feature correlation degree of the distributed module k(j); s j Correction characteristic parameter x j Importance coefficient; The calculation of the conditional probability density function for the corrected feature parameters includes determining: Under fault conditions: Under normal conditions: in: Correction characteristic parameter x j Conditional probability density functions PDF under fault and normal conditions respectively; x j : Actual measured characteristic parameter values; μ j,F ,σ j,F Correction characteristic parameter x j Mean and standard deviation under fault conditions; μ j,N ,σ j,N Correction characteristic parameter x j Mean and standard deviation under normal conditions; exp: the natural exponential function, i.e., e (·) ; The calculation of the weighted log-likelihood function includes determining: Log-likelihood function under fault conditions: Log-likelihood function under normal conditions: in: lnL(F i |X i X was observed under fault conditions. i The log-likelihood value; X was observed under normal conditions. i The log-likelihood value; w j Correction characteristic parameter x j The weights; Among them, the likelihood ratio is calculated and the posterior failure probability P(F) is calculated using Bayes' theorem. i |X i Specifically, this includes: Calculate the likelihood ratio Λ: in: Λ: Likelihood ratio, which represents the ratio of the probability of observing data under fault conditions to the probability of observing data under normal conditions.

7. The modular rapid detection method for distributed collaborative operation of a prefabricated substation as described in claim 6, characterized in that, Calculate the posterior failure probability P(F) i |X i ): in: P(F i |X i Given observation data X i Then, the posterior probability of distributed module i failing; P(F i ): The prior probability of distributed module i failing; The prior probability of distributed module i operating normally.

8. A distributed, collaborative, modular, rapid detection device for prefabricated substations, characterized in that, The device employs a distributed, collaborative, modular, rapid detection method for prefabricated substations as described in any one of claims 1-7, wherein: The probability assessment module is used to determine whether there is a possibility of failure based on the collected operational data; the signaling initiation module is used to initiate abnormal signaling when there is a possibility of failure. The signaling determination module is used to confirm whether its own correction characteristic parameters are filled in the corresponding position in the abnormal signaling; The fault determination module is used to summarize all the obtained correction feature parameters, calculate the posterior fault probability, and determine that a fault exists when the posterior fault probability exceeds a set threshold.

Images