A method for identifying and blocking power pre-alarm in case of fire

Abstract

The present application relates to the technical field of fire warning, in particular to a kind of identification fire occurrence power early warning blocking method;Based on the power early warning signal obtained, continuously receive the electrical parameter data on the sensor deployed in each power grid node, detect the abnormal fluctuation of electrical parameter data, determine the abnormal type of electrical parameter data, set the early warning parameter set of electrical parameter data;Determine that the abnormal rule that early warning parameter set meets in preset time period, set the blocking threshold of each power grid node using the power grid node corresponding to abnormal rule;Adjacent area analysis is carried out to the power grid node to which blocking threshold is set, and hierarchical blocking mode is set;The power grid node corresponding to hierarchical blocking mode is connected, a blocking path is formed, and the blocking processing after power grid node early warning is completed;The timeliness of power early warning blocking response and fire warning efficiency are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fire warning, in particular to a method for identifying fire occurrence and power warning and blocking. BACKGROUND

[0002] Traditional power fire warning systems are usually based on fixed threshold or single parameter monitoring, which has problems of response lag, high false alarm rate and inability to accurately locate the fire source. Due to the complex topology of power grid and strong correlation of equipment, a single node anomaly may trigger a chain reaction, and the traditional method lacks dynamic analysis capability for abnormal propagation path, resulting in extensive blocking strategy and easy large-scale power outage or insufficient protection of key areas.

[0003] For example, Chinese Patent Publication No. CN112330910A discloses an information processing method, device, storage medium and processor for fire warning. The method includes: a fire supervision platform obtains first warning information sent by a fire warning platform, wherein the first warning information is uploaded to the fire warning platform by a fire warning device; the fire supervision platform obtains a review message corresponding to the first warning information, wherein the review message is obtained by the fire warning platform reviewing the first warning information; and the fire supervision platform determines whether to retain the first warning information based on the review message.

[0004] For example, Chinese Patent Publication No. CN112991658A discloses a fire warning method and an uninterruptible power supply. The method includes the following steps performed at a preset period: adjusting the working state of the uninterruptible power supply in a preset manner under the current working condition of the uninterruptible power supply; generating a first cumulative factor according to the actual temperature parameter in the cabinet before and after the adjustment of the working state of the uninterruptible power supply and the pre-stored reference temperature parameter under the same working condition; generating a fire warning index according to the first cumulative factor, and outputting an alarm signal when the fire warning index reaches a preset value. The present application actively adjusts the working state of the uninterruptible power supply, and generates a first cumulative factor by comparing the actual temperature variation parameter before and after the adjustment with the pre-stored reference temperature variation parameter, and generates a fire warning index according to the first cumulative factor to perform fire warning.

[0005] In the prior art, fire warning information is reviewed to process fire information, and the temperature is used as a reference factor to identify the temperature change of electronic components in the current fire scene, to realize the temperature change rate of electronic components in the fire warning. However, the prior art is biased towards processing single data anomalies, which requires the collection of current and other parameters of related equipment in historical data for comprehensive identification, and the use of abnormal composite method for topology hierarchical perception of the node to realize comprehensive control of multiple nodes, complete abnormal propagation and blocking setting of multiple power grid nodes in the fire warning scene, and finally reduce the persistent damage caused by equipment operation in the fire. SUMMARY

[0006] To solve the above technical problems, the technical scheme adopted by the present application is: a fire occurrence power early warning blocking method, comprising: S1, based on the acquired power early warning signal, continuously receiving the electrical parameter data uploaded on the sensors deployed at each power grid node. The electrical parameter data includes voltage, current and temperature.

[0007] S2, detecting abnormal fluctuations of the electrical parameter data, determining the abnormal type of the electrical parameter data, and performing data measurement on the source location of the power early warning signal according to the abnormal type, and setting a set of early warning parameters of the electrical parameter data.

[0008] S3, determining whether the set of early warning parameters conforms to the abnormal rule in the preset time period, and setting the blocking threshold of each power grid node using the power grid node corresponding to the abnormal rule.

[0009] S4, analyzing the adjacent area of the power grid node with the set blocking threshold, verifying whether the state identifiers of the nearest power grid nodes corresponding to the current power grid node are consistent, and if so, setting a hierarchical blocking mode according to the number of consistency; if not, traversing the blocking thresholds of each power grid node in multiple time intervals, analyzing the abnormal propagation path under the hierarchical blocking mode; based on the analysis result of the abnormal propagation path, setting the hierarchical blocking mode.

[0010] S5, connecting the power grid nodes corresponding to the hierarchical blocking mode to form a blocking path, and completing the blocking processing of the power grid nodes after early warning.

[0011] The beneficial effects of the present application are: first, the present application continuously collects voltage, current and temperature parameters, calculates the parameter change rate in combination with the device working state, identifies abnormal fluctuations by comparing the sliding average value with the historical data, determines the abnormal type under different trigger conditions by using context flag weighting, and matches the early warning parameters of abnormal data in the form of current-voltage-temperature multi-parameter correlation through instantaneous change rate and target value, in combination with time classification analysis of abnormal type, to improve the matching accuracy of power early warning under fire.

[0012] Second, the present application generates a list of adjacent nodes based on the power grid topology, realizes dynamic comparison of blocking thresholds by using a sub-type inheritance mechanism, optimizes the blocking range by using blocking coverage calculation and secondary blocking mode, quantifies the relative position allocation and path identification of each power grid node in the power blocking scene, and completes the power early warning of the multi-node common abnormal scene to realize the accuracy of power early warning.

[0013] Thirdly, the application sets weights for the blocking path nodes, acquires frequencies under multiple areas through area embedding, detects the number of weight changes through a sliding window, improves the accuracy of risk trend prediction in multiple areas, and completes the processing of power early warning trend changes. BRIEF DESCRIPTION OF DRAWINGS

[0014] The application will be further described below in combination with the drawings and examples.

[0015] Figure 1 It is a flowchart of a fire occurrence power early warning blocking method.

[0016] Figure 2 It is a flowchart of step S2 of the fire occurrence power early warning blocking method.

[0017] Figure 3 It is a flowchart of step S3 of the fire occurrence power early warning blocking method.

[0018] Figure 4 It is a flowchart of step S4 of the fire occurrence power early warning blocking method. DETAILED DESCRIPTION

[0019] The embodiments of the application will be described in detail below. The embodiments described below are exemplary and are only used to explain the application and cannot be understood as a limitation of the application. If the specific technology or conditions are not specified in the embodiments, the technology or conditions described in the literature in the art or according to the product manual are used.

[0020] Reference Figure 1 A fire occurrence power early warning blocking method, comprising: S1, based on the acquired power early warning signal, continuously receiving the electrical parameter data on the sensors deployed at each power grid node. The electrical parameter data includes voltage, current and temperature.

[0021] S2, the electrical parameter data is subjected to abnormal fluctuation detection, the abnormal type of the electrical parameter data is determined, the source position of the power early warning signal is subjected to data metering according to the abnormal type, and a set of early warning parameters of the electrical parameter data is set.

[0022] S3, judging whether the early warning parameter set meets the abnormal rule in the preset time period, setting the blocking threshold of each power grid node by using the power grid node corresponding to the abnormal rule.

[0023] S4, the power grid node setting the blocking threshold value is analyzed in the adjacent area, and the state identifier of the nearest power grid node corresponding to the current power grid node is verified. If it is consistent, the hierarchical blocking mode is set according to the consistent number; if it is not consistent, the blocking threshold value of each power grid node in multiple time intervals is traversed, and the abnormal propagation path under the hierarchical blocking mode is analyzed; based on the analysis result of the abnormal propagation path, the hierarchical blocking mode is set.

[0024] S5, the power grid node corresponding to the hierarchical blocking mode is connected to form a blocking path, and the blocking processing after the power grid node warning is completed.

[0025] The electrical parameter data such as current, voltage, power, leakage current, temperature and other parameters, and the related data of each power grid node in the fire are collected by temperature sensor, voltmeter and ammeter and other equipment, so that the power parameter data received by the power grid node can be quickly judged to affect the area of the fire, and the power warning is sent in the first time after confirmation, and the related power supply is automatically switched, so as to prevent the fire from expanding and reduce secondary damage.

[0026] Preferably, the power warning signal can be a signal sent when any part of the electrical parameter data is identified as abnormal, or a warning signal sent when a fire is identified. The warning signal can be used to identify the possible warning form and position, and the related data can be continuously checked as the main analysis of the electrical parameter data.

[0027] Preferably, the equipment represented by the power grid node includes but is not limited to power generation equipment, transformer, circuit breaker, disconnecting switch, transmission line, power distribution cabinet and other equipment. These devices will configure corresponding current, voltage and temperature sensors to obtain the relative state of each node in the current power grid, so as to explain how to block in the fire scene.

[0028] When receiving the electrical parameter data, the implementation of step S1 further includes: according to the uninterrupted working state of each power grid node, the change rate of the electrical parameter data of each power grid node is extracted, and compared with the preset warning threshold value in turn. When the preset warning threshold value is met, the electrical parameter data corresponding to each power grid node is output with the trigger condition of the current power warning signal.

[0029] Preferably, the trigger condition is divided into electrical parameter anomaly and fire confirmation signal, and when any one of the electrical parameter data is identified, the power grid node at the corresponding position is monitored. As for the fire confirmation signal, it is the fire identified by the sensor configured on the power grid node, or the smoke and other contents conforming to the characteristics of the fire, at this time, in addition to the temperature, current and voltage sensors, the smoke alarm and camera equipment are also included to collect the working conditions of the equipment at the position configured on each power grid node, and then the smoke alarm and camera equipment transmit the part of the fire confirmation signal to the electrical parameter anomaly transmitted by the temperature, current and voltage sensors as the basis for monitoring the data of each power grid node; as long as there is data that meets the current trigger condition, the electrical parameter data of each power grid node is processed to identify the related characteristics when the fire occurs, and how to block power according to the corresponding characteristics when the fire occurs is judged.

[0030] Preferably, the above-mentioned preset warning threshold value is based on the average value of the upper limit value of the temperature, voltage and current of each power grid node under normal operation in the historical data as the preset warning threshold value of the current power grid node, to indicate that any one of the current, voltage or temperature of the current power grid node is abnormal.

[0031] In an embodiment of the present application, in step S2, mainly for detecting the abnormal fluctuation existing in the electrical parameter data, to obtain the abnormal type of the current electrical parameter, such as short circuit, overload and the like, and analyze the time sequence of the received electrical parameter data, in combination with the correlation of multiple abnormal types under fire, the abnormal type is extracted in the form of correlation analysis and data measurement of the current identified abnormal data.

[0032] As shown in Figure 2 The implementation of step S2 includes: using the time sequence of the electrical parameter data, the sliding average value of the dimension of the electrical parameter data, comparing the electrical parameter data with the historical data, taking the current dimension as the influencing factor, and determining the parameter threshold value of different influencing factors before and after the fire occurs; the above-mentioned dimension represents that the data of the current collected temperature, voltage and current is divided into multiple dimensions, and the abnormality is identified for each dimension according to the parameter threshold value. The above-mentioned parameter threshold value is represented as the average value of the historical data to judge the sliding average value of the current electrical parameter data under the same dimension.

[0033] The context mark of the current influencing factor is obtained, and the weight of the influencing factor is judged based on the context mark; the context mark identifies the state of the corresponding equipment when the power warning signal is sent, such as power consumption peak period, equipment start-stop event and the like, which represents the state of the current equipment under the warning condition, which is taken as the basis for judging the current electrical parameter anomaly.

[0034] Based on the weight of the influencing factor and the parameter threshold, the abnormal type of the current influencing factor is identified. The identified abnormal type includes current overload, harmonic, flicker, etc. representing the data abnormal type of the device at the power grid node.

[0035] Preferably, the context flag is set according to the relevant value of the historical data, such as determining the peak power consumption period, if the current collected current exceeds 80% of the average value of the peak power consumption period in the historical data, the corresponding influencing factor flag is set to peak period or non-peak period, and 1 or 0 is used to represent its relative situation, and the device start-stop time length is used to indicate the length of the device start-stop time, and these data are used to normalize or other forms to mark the current collected electrical parameters.

[0036] Therefore, the implementation mode of determining the weight of the influencing factor based on the context flag further includes: obtaining the number of influencing factors of the current power grid node, and summarizing the context flags in the number of influencing factors, and setting the weight of the current influencing factor based on the average characteristic value of the summarized context flags; when setting the weight of the current influencing factor, the context flags appearing in a single dimension are summarized, such as identifying the peak period, device start time, current fluctuation, etc. through current collection, and then the average value of the preset weight of each context flag in the database is obtained, and then the weight of the current influencing factor is obtained. As for the number of influencing factors, it represents the dimensions that are gradually processed for the current identified abnormal dimensions, and whether the current and voltage exist abnormally in these dimensions is identified through the abnormal type of the influencing factor.

[0037] Preferably, the implementation mode of identifying the abnormal type of the current influencing factor includes: clustering the electrical parameter data based on the parameter threshold of the influencing factor, and obtaining a plurality of clustered abnormal types. The abnormal types are sorted according to the weight of the influencing factor, and the order of the abnormal types under different influencing factors is obtained. For example, the short-circuit cluster feature: current surge (> 300% rated value), temperature rapid rise (> 80℃ / s); overload cluster feature: current continuous high (120%-150% rated value), temperature slow rise (< 10℃ / s); after clustering the currently identified data, different abnormal types are obtained, and further, in the high weight scene such as peak period, the overload related abnormal type is determined first, and then the identified instantaneous fluctuation or harmonic type is marked.

[0038] After clustering the data of similar patterns, the clustered values are directly compared with the contents represented by the short-circuit cluster feature, the overload cluster feature, etc. to determine the abnormal type. The abnormal type determined in this step is used to indicate the obvious property problem existing in the current, voltage and temperature. The power reference data and the historical data can be further compared to obtain the obvious abnormal type of the current from the database and other data sources.

[0039] When setting the early warning parameter set of the power early warning signal, according to the source position of the power early warning signal, such as a specific sensor, a device node or a regional power grid, the corresponding power reference data is further compared with historical data to preliminarily judge the current abnormal range; and data in the abnormal range is outputted.

[0040] Therefore, the implementation manner of setting the early warning parameter set of the power early warning signal further includes: performing data measurement on the electrical parameter data according to the co-occurrence frequency of the abnormal type at the power grid node, and sequentially analyzing the current, voltage and temperature of the electrical parameter data according to the occurrence time of the measured electrical parameter data; the current abnormal type, the voltage abnormal type and the temperature abnormal type in the region where the power grid node is located are acquired respectively, and the corresponding data is stored in the early warning parameter set according to the processing mode of the current abnormal type, the voltage abnormal type and the temperature abnormal type.

[0041] When processing the current abnormality, if it is determined to be overload, flexible power-off is triggered, such as gradually reducing the load within 30 seconds; if it is short circuit, the power supply is immediately cut off. When processing the voltage abnormality, if the voltage is overvoltage, that is, > 110% rated voltage, the voltage stabilizer is started; if the voltage is under voltage, that is, < 90% average voltage value, the standby power supply is switched. As for the temperature abnormality, if the temperature exceeds the threshold value, such as the current set temperature is 60℃, the heat dissipation system or current limiting protection is started.

[0042] The early warning threshold value set for the current abnormality is 1.5 times the rated current, and the short circuit threshold value is 3 times the rated current, to judge the abnormal situation of the current value and the threshold value when the abnormal situation is judged. The voltage abnormality sets the voltage fluctuation range to ±10%, and the response time reaches 50ms, to trigger the voltage abnormality processing. The temperature abnormality can set the temperature threshold value, taking the maximum working temperature allowed by the device at the location of the power grid node as the threshold value, for example, the conductor temperature > 70℃ triggers the early warning, to set the early warning threshold value for the identified abnormality, and store the early warning threshold value associated with the electrical parameter data in the early warning parameter set.

[0043] The numerical range described above is a schematic representation of the current power grid node processing, and the specific setting of the numerical range needs to be adjusted according to the device type of the power grid node to realize the monitoring and processing of multiple devices on the power grid node.

[0044] In an embodiment of the present application, when explaining the abnormal rules of the current early warning parameter set, the compliance degree of each power grid node is calculated based on historical data and real-time load, and then the blocking threshold value set by each power grid node is defined by using the compliance degree of each power grid node.

[0045] Step S3 identifies whether there is a sudden increase and sudden drop phenomenon of voltage and current in the instantaneous change rate in the early warning parameter set when checking the abnormal rule, and when the instantaneous change rate changes in a single time period, the abnormal rule matching is performed to obtain the similar situation of the historical data and the current power grid node, and the current data situation under the abnormal rule is output, such as overload, short circuit and the like, and then the source of the location of the power grid node is marked to indicate the maximum allowed power-off duration and the minimum safe current of the current power grid node as the blocking threshold of the current power grid node.

[0046] As shown in Figure 3 The implementation mode of step S3 further includes: based on the early warning parameter set in the preset time period, extracting the early warning threshold of the early warning parameter set, and comparing the early warning threshold with the real-time electrical parameter data to obtain the instantaneous change rate of the electrical parameter data under different early warning thresholds; at this time, by using the early warning threshold set under different abnormal types, the electrical parameter data is divided into multiple data sets, and the data change rate of each data set in the preset time period is checked, that is, the instantaneous change rate value of temperature, current and voltage in the preset time period, and the length of the preset time period can be set to 5 minutes, which is used to continuously monitor the working condition of each node in the power grid, if there is a part of data continuously processing short circuit, overload and the like, the related node will be blocked, and the maximum allowed power-off duration and the minimum safe current of each power grid node are extracted to control the blocking of the power grid node.

[0047] The time points corresponding to the instantaneous change rate of the electrical parameter data are classified into events, and the abnormal type after event classification is matched with the abnormal rule.

[0048] Based on the result of the abnormal rule matching, the maximum allowed power-off duration and the minimum safe current of the current power grid node are extracted as the blocking threshold of the current power grid node.

[0049] Preferably, when classifying events, the abnormal types in the electrical parameter data are arranged in the order of time points to explain that in the scenario of current fire or fire warning, the abnormal types that can appear together in a short time period, and the abnormal rule matching is performed according to the order of these abnormal types and the time period to explain whether the current current, voltage and temperature related abnormality belongs to the fire scenario.

[0050] That is, when the abnormal rule matching is performed, the implementation mode includes: converting the instantaneous change rate and time point of the electrical parameter data under the abnormal type into a vector representation, and matching the vector with a preset rule event, using vector cross multiplication to calculate the matching degree, if the matching degree is 0, it is judged as complete matching, and the time points after event classification are matched one by one to complete the abnormal rule matching result of the electrical parameter data under different early warning thresholds.

[0051] The preset rule event represents a set of data collection of abnormal type combinations of multiple currents, voltages and temperatures in a fire scene or a scene where a fire is about to occur, and the data collection represents a composite description of the data collection in the fire scene. The preset rule event sets multiple data collections according to the type of the power grid node to complete whether the current power grid node has a corresponding abnormal condition under multiple abnormal type combinations.

[0052] After the abnormal rule matching is completed and the maximum allowed power-off duration and the minimum safe current of the current power grid node are extracted, the corresponding abnormal type is called for the current electrical parameter data of the current, voltage and temperature, respectively, and the abnormal type after calling is combined to complete the setting of the blocking threshold and the like.

[0053] That is, the implementation manner of the abnormal rule matching further includes: based on the preset rule event, performing abnormal rule matching on the electrical parameter data, calling the abnormal type of the current, obtaining a target current value, and matching the target current value with the instantaneous change rate of the current abnormal type and the warning threshold. In a case where the current value is not the target current value and the matching degree is less than a preset threshold, the abnormal type of the current is regarded as the result of the current abnormal rule matching.

[0054] After the abnormal type of the current is called, the current abnormal rule matching is directed to the voltage abnormal rule matching. Based on the result of the current abnormal rule matching, a current threshold lower limit and a current threshold upper limit are obtained. In a case where the current power grid node has the abnormal type of the voltage, it is judged that the instantaneous change rate of the voltage is greater than the voltage drop threshold and the time difference when the current reaches the current threshold upper limit. If the condition is met, a composite abnormality is triggered, the abnormal type of the current and the abnormal type of the voltage are combined, and the result of the current and voltage abnormal rule matching is regarded as the result of the current and voltage abnormal rule matching. Otherwise, it is regarded as only the result of the current abnormal rule matching.

[0055] The result of the current abnormal rule matching is directed to the temperature abnormal rule matching. In a case where the current power grid node has the abnormal type of the temperature, it is judged that the current reaches the current threshold lower limit of the temperature abnormal type, and the corresponding temperature abnormal type and the result of the current abnormal rule matching are combined. The multiple combination data to which the result of the current abnormal rule matching is directed are regarded as the result of the output abnormal rule matching.

[0056] Based on the obtained composite form of the abnormal rule matching, the maximum allowed power-off duration and the minimum safe current corresponding to the current scene are extracted from the database.

[0057] Preferably, the lower limit of the current threshold value and the upper limit of the current threshold value represent the abnormal range of the current under the current abnormal type, and the value represents the threshold value of the abnormal setting of the current; the lower limit of the current threshold value is used to prevent false positives and avoid false composite abnormalities caused by temporary current fluctuations to reduce false blockages; the upper limit of the current threshold value is used to define a serious abnormality to ensure that high-risk events are immediately handled to avoid fire spread.

[0058] Preferably, the target current value obtained when calling the abnormal type of the current is used to represent the target value under a single abnormal type, such as overload, which is usually set to 1.2-1.5 times the rated current, short circuit, which is usually set to more than 3 times the rated current, and harmonic, which is described by a total harmonic distortion greater than 5%, etc. The target current value is used to indicate that the current value is in the main judgment value of the corresponding abnormal type, and then the subsequent description of the current value is not in the target current value, which is used to indicate that the current value is already in the abnormal range of the current target current value. For example, when judging overload, the current value is not in the range of less than 1.2 times the rated current. As for the matching degree, the form of vector cross product is used. When the matching degree is less than the preset threshold value, it indicates that the current is abnormal and meets the current abnormal type. The preset threshold value can be based on the average value of the historical data matching the current abnormal rule as the preset threshold value at this time.

[0059] Preferably, when the abnormal rule matching result of the current is directed to the abnormal type of the voltage, the main judgment is whether the instantaneous change ratio of the voltage is too large or too small. When the voltage is greater than the voltage drop threshold value, it indicates that the current voltage is also in synchronous abnormality in addition to the current abnormality, such as the voltage drop threshold value, which can be set to 20%, to indicate the change in the unit time when the voltage changes. If the time difference between the voltage and the current is less than a certain value, it indicates that the two are close to the concurrent time of the same abnormality, which meets the composite abnormality. For example, if the interval between the occurrence of the two is less than 10s, it is considered that the voltage abnormal type and the current abnormal type recognized at the present time belong to concurrent events under the fire scene, which need to be merged and analyzed to query the blocking mode required by the current scene.

[0060] Preferably, the voltage drop threshold value is based on the average value of the voltage drop of the current power grid node in the historical data to set the threshold value for judging the voltage drop at this time.

[0061] Preferably, the average value of the time interval at which the voltage and the voltage in the historical data meet the instantaneous change rate of the voltage greater than the voltage drop threshold value and the current reaches the upper limit of the current threshold value is used as the standard for the current judgment. When the time interval is less than the average value, it is considered to meet the composite abnormality.

[0062] Preferably, when the abnormal rule matching result of the current is directed to the abnormal type of the temperature, it is mainly judged that the temperature has met the corresponding abnormal type scene when the current starts to appear abnormally, and the abnormal type of the current and the abnormal type of the temperature at this time are combined to serve as the basis for subsequent search for blocking mode, and then the combined data of the three abnormal types of current, voltage and temperature after the abnormal rule matching is taken as the output abnormal rule matching result, and the maximum allowed power-off time and the minimum safe current that can be selected in the current multi-abnormal type scene are queried from the database according to the result, so as to complete the control and processing of each power grid node.

[0063] The abnormal rule obtained in step S3 is relative to the abnormal type, and the abnormal rule emphasizes the context flag and threshold of the combination of multiple abnormal types, judges whether the risk existing in the composite scene is a fire risk, and describes the current, voltage and temperature in the current electrical parameter data in the form of structured threshold, time sequence mode and vector combination, and explains the data form in the state of impending or ongoing fire, so that the subsequent power grid node can play a role in fire warning and blocking according to the current obtained data.

[0064] In an embodiment of the application, when the hierarchical blocking mode is set, the state and relative situation of the power grid node are obtained by checking the identification of each power grid node, the state identification represents the abnormal type and blocking threshold of the corresponding node, and the equipment and position represented by each power grid node, and these data are taken as the state identification of each power grid node.

[0065] In the case of inconsistency, it is necessary to judge the propagation path related to the abnormal type, such as Figure 4 As shown in the figure, the implementation mode of step S4 includes: analyzing the adjacent groups of power grid nodes to form an adjacent node list by the power grid topology structure; and each row of data in the adjacent node list represents a group of adjacent power grid nodes.

[0066] The abnormal type of each group of power grid nodes in the adjacent node list is identified, if any abnormal type of the current power grid node belongs to the sub-type of the adjacent power grid node, the number of abnormal types of the current power grid node belonging to the sub-type is identified from the abnormal types of the adjacent power grid node, the abnormal types of the current power grid node and the adjacent power grid node are connected, and the blocking thresholds of the adjacent power grid node and the current power grid node are compared, if the blocking thresholds are consistent, the blocking mode of the current power grid node and the adjacent power grid node is regarded as the same hierarchical blocking mode.

[0067] If any abnormal type of the current power grid node does not belong to the sub-type of the adjacent power grid node, the abnormal type of the current power grid node is compared with the adjacent power grid node, the current power grid node is regarded as a terminal node, and the blocking threshold of the current power grid node is selected to select a hierarchical blocking mode. At this time, when the hierarchical blocking mode is set, the current, voltage and other data are blocked based on the hierarchical blocking mode of the plurality of current power grid nodes to prevent the occurrence of fire and further damage.

[0068] As for setting the hierarchical blocking mode of the same level, if the blocking threshold is inconsistent, the blocking threshold of the adjacent power grid node is set to the hierarchical blocking mode, and the current node is the secondary blocking mode of the hierarchical blocking mode of the adjacent power grid node.

[0069] Preferably, when judging whether the state identifier is consistent, each group of adjacent power grid nodes is compared one by one. If there are more than 50% of the abnormal labels in the adjacent power grid nodes and the blocking threshold is the same in multiple time periods, it is determined that they are consistent, and the blocking mode is set for the current power grid node and the adjacent power grid node. For example, the blocking mode can be divided into three levels. The first level of blocking expands the blocked power grid node to the adjacent plurality of power grid nodes of the adjacent power grid node. The second level of blocking takes the current power grid node and the adjacent power grid node as the blocking content. The third level of blocking only blocks the current power grid node, and then the blocking threshold queried by the database is used to block in real time.

[0070] Preferably, if there is an inconsistent scene, it is judged whether the adjacent power grid node and the current power grid node are in the parent-child type relationship. Then when the blocking threshold is consistent with the adjacent node, the blocking area is merged. If it is inconsistent, the current power grid node is regarded as a secondary blocking mode to merge the propagation mode of various abnormal types.

[0071] Only for the nodes that do not belong to the sub-type, the time interval is traversed, the abnormal type of the current power grid node is triggered, the reverse tracing is performed, the power grid nodes having topological connection relationship with the current power grid node are identified, and then the blocking threshold of each power grid node is viewed in the reverse tracing manner. According to the value range of the blocking threshold, the hierarchical blocking mode is set in turn, and the blocking level of each power grid node is set.

[0072] As for the setting mode, the state identifier of the current power grid node is taken as a retrieval condition, the hierarchical blocking mode that meets the conditions of the abnormal type, the blocking threshold and other conditions of the current power grid node is viewed from the database in a composite retrieval manner, and then the hierarchical blocking mode is set one by one.

[0073] When the hierarchical blocking mode is output, the hierarchical blocking mode is output one by one in the order of each hierarchical level according to the hierarchical level when the hierarchical blocking mode is set.

[0074] After the hierarchical blocking mode is set, the hierarchical blocking mode set for the plurality of power grid nodes in the adjacent node list is also needed to be verified step by step, that is, the implementation mode of step S4 further comprises: verifying the hierarchical blocking mode of each power grid node, identifying the blocking coverage of each hierarchy for the hierarchical blocking mode, starting from any one power grid node where the blocking starts, identifying other plurality of power grid nodes connected with the current blocked power grid node, and sequentially outputting the control command corresponding to the hierarchical blocking mode by using the blocking coverage in the hierarchical blocking mode.

[0075] At this time, the ratio of the number of abnormal power grid nodes covered in each hierarchy to the total number of abnormal nodes is taken as the basis for sorting, and the blocking mode in each hierarchy is output from large to small to complete the blocking processing of the current for abnormal conditions.

[0076] In an embodiment of the present application, when forming the blocking path, a plurality of power grid nodes need to be connected, and the number of times each power grid node is marked in the blocking path is captured to complete the multi-region identification of power blocking after the power warning blocking.

[0077] Therefore, the implementation mode of step S5 comprises: data determination on the power grid nodes on the blocking path, setting the count weight of the power grid nodes according to the time when each power grid node appears in the blocking path and the hierarchical blocking mode, region embedding correlation of the power grid nodes, taking any one power grid node on the current blocking path as the starting point, judging the change frequency of the count weight of each power grid node in a plurality of time periods, and if the change frequency exceeds a preset frequency threshold, marking the current blocking path and outputting it as power grid warning information.

[0078] Preferably, step S5 records the first blocking time and the latest blocking time of the power grid node when blocking, checks the difference in time dimension of the corresponding power grid node in two or more times, and then checks the identification of the current power grid node belonging to the first blocking, the second blocking or the third blocking. At this time, the weight of each power grid node is calculated.

[0079] For example, the weights of the first, second and third blocking are set to 2.0, 1.0 and 0.5 respectively to illustrate the verification degree in the current blocking processing, and then the weight is set by using the blocking time difference of the power grid node in a plurality of times. ; wherein, count weight, initial weight, a value set based on the identification of the current power grid node being in the first blocking, the second blocking or the third blocking, represents the decay coefficient, wherein the decay coefficient is taken as 0.05, for weakening the difference between the first blocking time and the recent blocking time in a shorter time period, to strengthen the smoothness of the body blocking mode processing, and to retain the relative weight of each power grid node under long-term risk; represents the time period difference between the first blocking time and the recent blocking time, and when calculating the value, the current time difference is processed in a standardized form to eliminate the dimension when calculating the weight; represents the exponential constant. The count weight is used to identify whether there is a time decay form of the blocking mode of the current power grid node in different time periods, to facilitate subsequent traversal of the related situation of multiple power grid nodes appearing abnormal in multiple time periods on the blocking path, and to facilitate subsequent prediction of the device configuration in the current power grid when a fire risk occurs.

[0080] Preferably, when the region embedding association is performed, multiple power grid nodes are clustered according to the region where they are located to identify the location of the region where the current blocking path is located, and then whether the region where the current power grid node is located is a high-frequency occurrence region is described by comparing whether the count weight changes in multiple time periods. If the count weight changes, it means that the corresponding region still has a corresponding power blocking event, and a mark needs to be set on the related blocking path. For example, the number of changes of a node in three consecutive windows is greater than 2, triggering a regional level warning. In this way, sudden blocking fluctuations caused by short circuits, lightning strikes and the like can be quickly identified, and the change trend in the corresponding time period can be described in the form of a time sequence to achieve preventive maintenance and complete adaptive optimization of the power grid node blocking.

[0081] Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application, which are still covered by the protection scope of the present application.

Claims

1. A method for identifying a fire occurrence power pre-warning blocking, characterized in that, Comprise: S1, based on the obtained power warning signal, continuously receive the electrical parameter data on the sensor deployed in each power grid node, the electrical parameter data including voltage, current and temperature; S2, the abnormal fluctuation detection of electrical parameter data, the abnormal type of electrical parameter data is determined, and the data of the source position of the power warning signal is measured, and the warning parameter set of electrical parameter data is set; S3, judging whether the warning parameter set meets the abnormal rule in the preset time period, setting the blocking threshold of each power grid node by using the abnormal rule corresponding to the power grid node; S4, the adjacent area analysis of the power grid node with set blocking threshold, verifying whether the state identifier of the nearest power grid node corresponding to the current power grid node is consistent, if consistent, setting the hierarchical blocking mode according to the number of consistency; If not consistent, traverse the blocking threshold of each power grid node in multiple time intervals, analyze the abnormal propagation path under the hierarchical blocking mode; Based on the analysis result of abnormal propagation path, set the hierarchical blocking mode; S5, connecting the power grid node corresponding to the hierarchical blocking mode to form a blocking path, completing the blocking processing after the power grid node warning; The implementation mode of step S5 comprises: The data of the power grid node on the blocking path is determined, the time of each power grid node in the blocking path and the hierarchical blocking mode are set, the counting weight of the power grid node is set, the power grid node is embedded and associated in the region, any power grid node on the current blocking path is taken as the starting point, the change frequency of the counting weight of each power grid node in multiple time periods is judged, if the change frequency exceeds the preset frequency threshold, the current blocking path is marked, and the power grid warning information is output; ; wherein, represents a count weight, represents an initial weight, represents a decay coefficient, represents a time period difference between a first blocking time and a recent blocking time, represents an exponential constant.

2. The method of claim 1, wherein the method comprises: When receiving the electrical parameter data, the implementation mode of step S1 further comprises: According to the uninterrupted working state of each power grid node, the change rate of the electrical parameter data of each power grid node is extracted, and the change rate is compared with the preset warning threshold in turn, when the preset warning threshold is met, the trigger condition of the current power warning signal is outputted, and the electrical parameter data corresponding to each power grid node is outputted.

3. The method of claim 1, wherein the method further comprises: detecting a fire; and blocking power supply to the electronic device. The implementation mode of step S2 comprises: Using the time sequence of electrical parameter data, the sliding average value of the dimension of electrical parameter data is used to compare electrical parameter data with historical data, taking the current dimension as the influencing factor to determine the parameter threshold of different influencing factors before and after the fire occurs; The context mark of the current influencing factor is obtained, and the weight of the influencing factor is judged based on the context mark; Based on the weight of the influencing factor and the parameter threshold, the abnormal type of the current influencing factor is identified.

4. The method of claim 3, wherein the method further comprises: The implementation mode of judging the weight of the influencing factor based on the context mark further comprises: The number of influencing factors of the current power grid node is obtained, the context mark is summarized based on the number of influencing factors, and the average characteristic value of the summarized context mark is used to set the weight of the current influencing factor.

5. The method for identifying and blocking power supply in case of fire according to claim 1, characterized in that, The implementation mode of setting the warning parameter set of the power warning signal further comprises: The co-occurrence frequency of the abnormal types at the power grid nodes is used to meter the electrical parameter data, and the current, voltage and temperature of the electrical parameter data are analyzed in sequence according to the occurrence time of the metered electrical parameter data; the current abnormal type, voltage abnormal type and temperature abnormal type in the region where the power grid node is located are obtained respectively, and the corresponding data is stored in the early warning parameter set according to the processing mode of the current abnormal type, voltage abnormal type and temperature abnormal type.

6. The method for identifying and blocking power supply in case of fire according to claim 1, characterized in that, The implementation mode of step S3 further includes: Based on the early warning parameter set in the preset time period, the early warning threshold of the early warning parameter set is extracted, and the early warning threshold is compared with the real-time electrical parameter data to obtain the instantaneous change rate of the electrical parameter data under different early warning thresholds; The time point corresponding to the instantaneous change rate of the electrical parameter data is used for event classification, and the abnormal type after event classification is used for abnormal rule matching; Based on the result of abnormal rule matching, the maximum allowed power-off time and the minimum safe current of the current power grid node are extracted as the blocking threshold of the current power grid node.

7. The method of claim 6, wherein the method further comprises: determining whether the power supply is in a fire occurrence area; and if the power supply is in the fire occurrence area, blocking the power supply. The implementation mode of abnormal rule matching further includes: ​ Based on the pre-set preset rule event, the electrical parameter data is subjected to abnormal rule matching, the abnormal type of the current is called to obtain a target current value, and the target current value is matched with the instantaneous change rate and the early warning threshold of the current abnormal type, in the case that the current value is not the target current value and the matching degree is less than the preset threshold, the abnormal type of the current is regarded as the result of current abnormal rule matching; The current abnormal rule matching is directed to voltage abnormal rule matching, based on the result of current abnormal rule matching, the current threshold lower limit and the current threshold upper limit are obtained, in the case that there is a voltage abnormal type at the current power grid node, the time difference between the voltage instantaneous change rate greater than the voltage drop threshold and the current reaching the current threshold upper limit is judged, if it is satisfied, a composite abnormality is triggered, the abnormal type of the current and the abnormal type of the voltage are combined, and it is regarded as the result of current and voltage abnormal rule matching; otherwise, it is regarded as only the result of current abnormal rule matching; The result of current abnormal rule matching is directed to temperature abnormal rule matching, in the case that there is a temperature abnormal type at the current power grid node, the temperature abnormal type of the current reaching the current threshold lower limit is judged, and the corresponding temperature abnormal type and the result of current abnormal rule matching are combined, the multiple combination data pointed by the result of current abnormal rule matching is regarded as the output result of abnormal rule matching.

8. The method for identifying and blocking power supply in case of fire according to claim 1, characterized in that, The implementation mode of step S4 includes: The adjacent groups of power grid nodes are analyzed, and the power grid nodes are combined into an adjacent node list according to the power grid topology; each row of data in the adjacent node list represents a group of adjacent power grid nodes; Identify the abnormal type of each set of power grid nodes in the adjacent node list. If any abnormal type of the current power grid node belongs to the sub-type of the adjacent power grid node, identify the number of abnormal types of the current power grid node that belong to the sub-type, starting from the abnormal type of the adjacent power grid node, connect the abnormal types of the current power grid node and the adjacent power grid node, and compare the blocking threshold values of the adjacent power grid node and the current power grid node. If the blocking threshold values are consistent, the blocking mode of the current power grid node and the adjacent power grid node is considered to be the same level of hierarchical blocking mode. When setting the hierarchical blocking mode of the same level, if the blocking threshold values are inconsistent, set the hierarchical blocking mode with the blocking threshold value of the adjacent power grid node, and let the current node be the secondary blocking mode of the hierarchical blocking mode of the adjacent power grid node. If any abnormal type of the current power grid node does not belong to the sub-type of the adjacent power grid node, compare the abnormal types of the current power grid node and the adjacent power grid node, let the current power grid node be the terminal node, and select the hierarchical blocking mode with the blocking threshold value of the current power grid node.

9. The method of claim 8, wherein the method further comprises: detecting a fire; and blocking power supply to the electronic device. The implementation of step S4 also includes: Inspect the hierarchical blocking mode of each power grid node, identify the blocking coverage under each hierarchy for the hierarchical blocking mode, start from any power grid node where the blocking starts, identify other multiple power grid nodes connected to the current blocked power grid node, and output the control command corresponding to the hierarchical blocking mode in turn using the blocking coverage under the hierarchical blocking mode.

Images