System and method for monitoring state of dry powder extinguishing system of underground substation

Abstract

The invention discloses an underground substation dry powder extinguishing system state monitoring system and method, belongs to the technical field of underground substation fire extinguishing, and solves the problem of how to improve the accuracy of real-time monitoring and state prediction of a dry powder extinguishing system. Through cooperative work of a sensor data acquisition module, an intelligent data fusion and analysis module, a self-adaptive alarm and response mechanism module and a remote monitoring and fault diagnosis module, multi-dimensional data, a system operation state and early warning information are uploaded to a cloud monitoring platform in real time; operation and maintenance personnel can carry out remote monitoring and fault diagnosis through the cloud monitoring platform, the safe operation condition of the transformer substation can be mastered in time, and comprehensive monitoring and efficient response to the fire extinguishing system in the underground transformer substation are achieved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of underground substation fire extinguishing, and relates to a state monitoring system and method for a dry powder fire extinguishing system of an underground substation. BACKGROUND

[0002] As an important power facility, the safe operation of an underground substation is crucial for ensuring power supply and social stability. However, due to the special geographical environment and complex layout of power equipment in underground substations, the electromagnetic environment is complex, the space is limited, and the fire risk is high. Equipment such as transformers, switchgear, and cables in the substation are in a long-term high-voltage and high-load operating state. Once a fault occurs, it may cause an electrical fire. In addition, the ventilation system of the underground substation is limited. Once a fire occurs, the fire spreads quickly, and it is difficult to evacuate and rescue quickly, further increasing the harm of the fire accident. In this context, the substation fire extinguishing system is particularly critical.

[0003] Dry powder fire extinguishing systems have become one of the common fire extinguishing methods for underground substations due to their rapid and effective fire extinguishing characteristics. Dry powder fire extinguishing systems can effectively extinguish different types of fires such as electrical fires by spraying dry powder extinguishing agents at the fire source. They have the advantages of rapid fire extinguishing and do not cause secondary damage to equipment. However, existing dry powder fire extinguishing systems usually rely on manual operation or automatic triggering, which may have a lagging response or difficulty in timely and comprehensive coverage of the fire source.

[0004] Prior art such as the invention patent with application publication number CN118430171A discloses an intelligent fire monitoring device for substations based on multi-modal fire feature fusion. It fuses multiple data through a multi-modal fire feature fusion module and uses an analysis and judgment module for comprehensive judgment to improve the accuracy of fire identification. However, this device only considers external environmental features such as images and sounds and does not consider the perception of internal factors such as the flowability of the dry powder fire extinguishing system and the internal factors of the fire extinguishing pipeline, making it difficult to achieve real-time monitoring and state prediction of the overall operation state of the fire extinguishing system.

[0005] Currently, research on multi-modal data for fire extinguishing systems mostly focuses on aspects such as the physical state monitoring of extinguishing agent storage and spray pipes, and has not yet formed real-time monitoring and state prediction of the overall operation state of the fire extinguishing system. At the same time, existing monitoring methods mostly rely on traditional temperature sensors, pressure sensors, and other equipment. These devices have poor reliability in extreme environments, and the real-time and accuracy of the monitoring data are often limited. For example, at the initial stage of a fire, existing monitoring systems may not be able to detect hidden fire signs in time, resulting in the fire extinguishing system not being able to start in the shortest time or the fire extinguishing system not being able to fully cover the entire fire source area. SUMMARY

[0006] The technical scheme of the present application is used to solve how to improve the accuracy of real-time monitoring and state prediction of the dry powder extinguishing system.

[0007] The present application solves the above technical problems through the following technical scheme:

[0008] A state monitoring system of a dry powder extinguishing system of an underground substation comprises:

[0009] A sensor data acquisition module is configured to acquire multi-dimensional data of the dry powder extinguishing system and monitor the running state of the dry powder extinguishing system in real time.

[0010] An intelligent data fusion and analysis module is configured to process the multi-dimensional data acquired by the sensor data acquisition module and predict the running state of the dry powder extinguishing system.

[0011] An adaptive alarm and response mechanism module is configured to analyze the multi-dimensional data, determine whether there is abnormal data, and respond.

[0012] A remote monitoring and fault diagnosis module is configured to upload the output results of the sensor data acquisition module, the intelligent data fusion and analysis module, and the adaptive alarm and response mechanism module to a cloud monitoring platform for remote monitoring and fault diagnosis.

[0013] Further, the sensor data acquisition module comprises a pressure sensor, a powder flowability sensor, a pipeline vibration sensor, a temperature sensor, a humidity sensor, an electrochemical sensor, and a non-dispersive infrared sensor.

[0014] Further, the multi-dimensional data comprises pressure, powder flowability, vibration, temperature, humidity, carbon monoxide concentration, carbon dioxide concentration, and oxygen concentration.

[0015] Further, the intelligent data fusion and analysis module comprises:

[0016] A state modeling unit is configured to model the state of the dry powder extinguishing system based on historical data and calculate a system state vector.

[0017] A state transition unit is configured to establish a state transition model of the dry powder extinguishing system, specifically by using a state transition matrix to estimate the state evolution of the dry powder extinguishing system at different time points.

[0018] An observation model unit is configured to establish an observation model of the dry powder extinguishing system, specifically by using an observation matrix to match the actual data obtained from the sensor with the state vector to obtain a more accurate state estimation, and the observation value of the sensor is described by an observation equation.

[0019] The state prediction and update unit predicts the state of the dry powder fire extinguishing system at the current time based on a Kalman filtering algorithm, and updates the observation data at the current time.

[0020] Further, the adaptive alarm and response mechanism module comprises:

[0021] The abnormal data judgment unit is configured to analyze the multi-dimensional data to determine whether there is abnormal data, and calculate a comprehensive abnormality degree value.

[0022] The abnormal type and severity judgment unit is configured to control the system to make a response according to the comprehensive abnormality degree value, determine the abnormal type and severity, and send corresponding early warning information.

[0023] Further, the abnormal data judgment unit determines whether there is abnormal data by judging whether the temperature, carbon monoxide concentration, carbon dioxide concentration, oxygen concentration, pressure and vibration are abnormal based on abnormal data judgment indexes.

[0024] When the temperature T2 is in the range of [T a ,T b ], the temperature T2 is in the normal range, otherwise the temperature T2 is abnormal data, wherein T a is a first temperature threshold, and T b is a second temperature threshold.

[0025] When the carbon monoxide concentration C CO is in the range of [C1,C2], the carbon monoxide concentration C CO is in the normal range, otherwise the carbon monoxide concentration C CO is abnormal data, wherein C1 is a first carbon monoxide concentration threshold, and C2 is a second carbon monoxide concentration threshold.

[0026] When the carbon dioxide concentration C CO2 is in the range of [C4,C5], the carbon dioxide concentration C CO2 is in the normal range, otherwise the carbon dioxide concentration C CO2 is abnormal data, wherein C4 is a first carbon dioxide concentration threshold, and C5 is a second carbon dioxide concentration threshold.

[0027] When the oxygen concentration C O2 is in the range of [C7,C8], the oxygen concentration C O2 is in the normal range, otherwise the oxygen concentration C O2 is abnormal data, wherein C7 is a first oxygen concentration threshold, and C8 is a second oxygen concentration threshold.

[0028] When the pressure P2 is in the range of [P a ,P bWhen [condition], pressure P2 is within the normal range; otherwise, pressure P2 is abnormal. Where P... a P is the first pressure threshold. b This is the second pressure threshold;

[0029] When vibration V2∈[V a V b When the vibration V2 is within the normal range, it is considered normal; otherwise, the vibration V2 is considered abnormal. a V is the first vibration threshold. b This is the second vibration threshold.

[0030] Furthermore, the calculation of the comprehensive anomaly degree value in the abnormal data judgment unit specifically involves:

[0031] The degree of temperature anomaly Q is calculated using the following logic. T :

[0032]

[0033] Among them, T c This is the third temperature threshold;

[0034] The degree of carbon monoxide concentration anomaly Q is calculated using the following logic. CO :

[0035]

[0036] Wherein, C3 is the third carbon monoxide concentration threshold;

[0037] The degree of carbon dioxide concentration anomaly Q is calculated using the following logic. CO2 :

[0038]

[0039] C6 is the third carbon dioxide concentration threshold;

[0040] The degree of oxygen concentration anomaly Q is calculated using the following logic. O2 :

[0041]

[0042] C9 is the third oxygen concentration threshold.

[0043] The degree of pressure anomaly Q is calculated using the following logic. P :

[0044]

[0045] Among them, P c The third pressure threshold, P dis a fourth pressure threshold value; the vibration abnormality degree Q is calculated by using the following logic V :

[0046]

[0047] wherein, V c is a third vibration threshold value;

[0048] The comprehensive abnormality degree value Q is calculated by using the following logic expression:

[0049]

[0050] wherein, Q represents the comprehensive abnormality degree value, Q i represents the abnormality degree value of the i-th index, i [1, n], n represents the total number of indexes, w i represents the weight corresponding to Q i .

[0051] Further, the abnormality type and severity judging unit is specifically:

[0052] When the comprehensive abnormality degree value Q [Q1, Q2], it is judged as a slight abnormality, wherein Q1 is a first abnormality threshold value, and Q2 is a second abnormality threshold value; at this time, the state monitoring system automatically records abnormal data and time, issues a yellow pre-warning, and reminds the operation and maintenance personnel to pay attention to the substation equipment;

[0053] When the comprehensive abnormality degree value Q [Q2, Q3], it is judged as a serious abnormality, wherein Q3 is a third abnormality threshold value; at this time, the state monitoring system issues an orange alarm, and notifies the on-site operator to check the equipment state, and determines whether to need to carry out maintenance or take preventive measures;

[0054] When the comprehensive abnormality degree value Q [Q3, Q4], it is judged as an extremely dangerous abnormality, wherein Q4 is a fourth abnormality threshold value; at this time, the state monitoring system immediately issues a red emergency alarm, and immediately alarms the emergency personnel and the operation and maintenance personnel; automatically starts the fire extinguishing system to spray dry powder; the substation equipment is automatically powered off, and the power supply with the high-risk area is cut off to prevent the fire from spreading.

[0055] The application also provides a state monitoring method of a dry powder fire extinguishing system of an underground substation, comprising the following steps:

[0056] S1, collecting multi-dimensional data of the dry powder fire extinguishing system, and monitoring the running state of the dry powder fire extinguishing system in real time;

[0057] S2, performing data processing on the multi-dimensional data collected in S1, and predicting the running state of the dry powder fire extinguishing system;

[0058] S3, data analysis is carried out on the multidimensional data to determine whether there is abnormal data and to respond;

[0059] S4, the output results of S1-S3 are uploaded to the cloud monitoring platform for remote monitoring and fault diagnosis.

[0060] Further, the S2 comprises:

[0061] S21, modeling the state of the dry powder extinguishing system, modeling the state of the dry powder extinguishing system based on historical data, calculating the system state vector;

[0062] S22, a state transition model of the dry powder extinguishing system is established, specifically, the state evolution of the dry powder extinguishing system at different time points is estimated through a state transition matrix;

[0063] S23, an observation model of the dry powder extinguishing system is established, specifically, the actual data obtained from the sensor is matched with the state vector to obtain more accurate state estimation, and the observation value of the sensor is described by an observation equation

[0064] S24, the state of the dry powder extinguishing system at the current time is predicted based on the Kalman filtering algorithm, and the observation data at the current time is updated.

[0065] Further, the S3 comprises:

[0066] S31, data analysis is carried out on the multidimensional data to determine whether there is abnormal data, and the comprehensive abnormality degree value is calculated;

[0067] S32, the control system responds according to the comprehensive abnormality degree value, judges the abnormal type and the abnormal severity, and sends corresponding warning information.

[0068] The advantages of the present application are:

[0069] (1) Through the collaborative work of the sensor data acquisition module, the intelligent data fusion and analysis module, the self-adaptive alarm and response mechanism module and the remote monitoring and fault diagnosis module, the multidimensional data, the system running state and the warning information are uploaded to the cloud monitoring platform in real time, the operation and maintenance personnel can carry out remote monitoring and fault diagnosis through the cloud monitoring platform, the safety operation of the substation can be grasped in time, and the comprehensive monitoring and efficient response of the extinguishing system in the underground substation are realized.

[0070] (2) The sensor data acquisition module provided by the present application not only uses traditional temperature and pressure sensors, but also combines the characteristics of dry powder extinguishing systems, considers internal influencing factors such as extinguishing agent fluidity and pipeline vibration, and comprehensively predicts the running state of the extinguishing system based on multiple dimensional sensor data, as well as the identification and early warning of abnormal sensor data, thereby improving the accuracy of real-time monitoring and state prediction of the system, eliminating the errors of a single sensor, and overcoming the problem of false positives caused by data distortion of traditional temperature sensors and pressure sensors under electromagnetic interference, thereby enhancing the overall understanding and early warning capability of the system state. Predicting hidden faults such as dry powder caking and pipeline rupture.

[0071] (3) The intelligent data fusion and analysis module provided by the present application constructs a multi-dimensional state vector representation system, realizes state space modeling and prediction of the extinguishing system through a dynamic evolution model combining state modeling, state transition and observation model, and realizes noise adaptive filtering based on Kalman filtering. Through this multi-source heterogeneous data fusion architecture, the system realizes multi-dimensional coupling analysis of historical operation mode and real-time sensing information, and has high-precision state tracking capability with time and space continuity.

[0072] (4) In the present application, all monitoring data, analysis results and fault information can be uploaded to the cloud in real time, and the operator can perform remote monitoring and fault diagnosis through the cloud platform. The chart and report functions provided by the platform enable the staff to view the system state in real time and respond in a timely manner. At the same time, the cloud platform can also predict potential risks and faults, optimize operation and scheduling, reduce manual intervention, and improve the intelligent management level of the system.

[0073] (5) The present application introduces powder flowability sensors, pipeline vibration sensors and humidity sensors, etc., which can comprehensively monitor each link in the extinguishing system, especially the flowability of dry powder, the spraying coverage range and whether the pipeline has faults and other important factors. The application of these sensors not only improves the comprehensiveness of monitoring, but also enhances the reliability of the system, which can timely discover potential problems in the system.

[0074] (6) The present application establishes multiple types of abnormal data and severity classification, and adopts a strategy based on comprehensive abnormality degree value to divide response levels, and responds differently according to the type and severity of the abnormality, from warning notification of slight abnormality to automatic start of the extinguishing system in case of serious abnormality. Such design improves the intelligence and accuracy of the monitoring system, avoids frequent false alarms and failure to respond in time, and improves the safety of the system. BRIEF DESCRIPTION OF DRAWINGS

[0075] Figure 1 is an execution flow diagram of the underground substation dry powder extinguishing system state monitoring system of embodiment one of the present application. DETAILED DESCRIPTION

[0076] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0077] The technical solutions of the present application will be further described below in conjunction with the drawings in the specification and specific embodiments:

[0078] Embodiment one

[0079] The underground substation dry powder fire extinguishing system state monitoring method proposed by the present application will be described in detail below in conjunction with specific embodiments. The equipment and methods used in this embodiment are examples and are not limited thereto. In actual application, the related equipment, sensors, algorithms and response mechanisms should be properly adjusted and optimized according to the needs.

[0080] As shown in Figure 1 , specifically, a kind of underground substation dry powder fire extinguishing system state monitoring system based on multi-source perception and intelligent prediction is disclosed, including sensor data acquisition module, intelligent data fusion and analysis module, adaptive alarm and response mechanism module and remote monitoring and fault diagnosis module.

[0081] The sensor data acquisition module is used to acquire multi-dimensional data of the dry powder fire extinguishing system and monitor the running state of the dry powder fire extinguishing system in real time.

[0082] The sensor data acquisition module includes pressure sensor, powder flowability sensor, pipeline vibration sensor, temperature sensor, humidity sensor, electrochemical sensor and non-dispersive infrared (NDIR) sensor; a plurality of sensor data are collected in real time, and the data are transmitted to the cloud monitoring platform in real time through wireless mode.

[0083] In this embodiment, a plurality of types of sensors are arranged in the underground substation to monitor the running state of the fire extinguishing system in real time, and the specific configuration is as follows:

[0084] Specifically, the pressure sensor is arranged at the key position of the fire extinguishing pipeline, including the pipeline inlet end, branch pipeline, pipeline close to electrical equipment, pipeline end, pump station connection and powder tank, for monitoring the air pressure change in the fire extinguishing pipeline in real time. The air pressure change in the pipeline reflects the flow state of the fire extinguishing agent and whether there is abnormal situation such as blockage and damage in the pipeline.

[0085] The powder flow sensor is arranged at the distribution end of the fire extinguishing pipeline, and is used for monitoring the flow and distribution of the dry powder fire extinguishing agent in real time, and detecting whether the dry powder can smoothly pass through the pipeline to reach the fire source, so as to ensure the actual fire extinguishing effect.

[0086] The pipeline vibration sensor is arranged at the pipeline connection point, and is used for monitoring whether the pipeline has abnormal vibration in real time, and predicting that the pipeline is broken or has other mechanical faults if the pipeline has abnormal vibration.

[0087] The temperature sensor is arranged at multiple positions in and outside the substation, such as the spraying area of the fire extinguishing system and the equipment area in the substation, and is used for detecting the temperature change of the dry powder fire extinguishing system, especially the temperature change of the substation environment, and identifying potential fire sources and fire signs.

[0088] The humidity sensor is arranged at the fire extinguisher nozzle area, the electrical equipment room and the transformer room, and is used for detecting the humidity change of the environment around the fire extinguishing system. The abnormal change of the humidity will affect the flow of the fire extinguishing agent (such as the caking of the fire extinguishing agent powder, affecting the fire extinguishing effect) and the accuracy of the alarm.

[0089] The electrochemical sensor is arranged at the nozzle area of the fire extinguishing system, the electrical equipment room and the core area of the substation, and is used for detecting the concentration of carbon monoxide and oxygen. In the embodiment, the electrochemical sensor detects the concentration of the gas by using a chemical reaction, and is composed of an electrolytic cell and an electrode. When the electrochemical reaction occurs, the gas molecules will chemically react with the electrode, and generate a change in current or voltage, which is then converted into a signal of the gas concentration. The electrochemical sensor can accurately monitor the carbon monoxide gas and reduce the interference of other gases. The electrochemical sensor has low power consumption, is suitable for the underground substation environment, and has fast response speed, so that the gas leakage can be found in time.

[0090] The non-dispersive infrared (NDIR) sensor is arranged near the fire extinguisher nozzle and the key equipment area, and near the ventilation opening and the air conditioning equipment, and is used for detecting the concentration of carbon dioxide. The non-dispersive infrared (NDIR) sensor detects the concentration of the gas by using the absorption characteristics of the gas to the specific wavelength of infrared light. When the infrared light passes through the gas to be detected, the gas will absorb the infrared wavelength related to the molecular vibration or rotation of the gas. The sensor detects the change in the intensity of the infrared light after passing through the gas, and calculates the concentration of the gas by the relationship between the absorbance and the concentration of the gas. The non-dispersive infrared (NDIR) sensor has the advantages of corrosion resistance, no need of contact and long-term stability, and is also suitable for the underground substation environment.

[0091] The above sensors transmit data to the cloud monitoring platform in real time through a wireless network, so as to ensure the immediacy and integrity of the data.

[0092] The multi-dimensional data includes pressure, powder flowability, vibration, temperature, humidity, carbon monoxide concentration, carbon dioxide concentration and oxygen concentration.

[0093] The intelligent data fusion and analysis module is used for data processing on the multi-dimensional data collected by the sensor data collection module on a cloud monitoring platform, and predicting the running state of the dry powder fire extinguishing system.

[0094] In the embodiment, a plurality of types of sensor data are comprehensively analyzed to improve the accuracy and real-time performance of the state monitoring system, and accurately reflect the actual running state of the fire extinguishing system; through multi-dimensional data fusion, errors caused by a single type of sensor are eliminated, reliable real-time state estimation is provided, and potential abnormalities or faults of the fire extinguishing system are discovered in time. Specifically, the intelligent data fusion and analysis module comprises:

[0095] A state modeling unit models the state of the dry powder fire extinguishing system based on historical data, and uses the following logic to represent the system state vector x at time k k :

[0096]

[0097] Wherein, P1 represents historical pipeline pressure, L1 represents historical powder flowability, V1 represents historical pipeline vibration, T1 represents historical temperature, D1 represents historical humidity, T represents a transposition operation, a represents a differential order, β represents a differential order, represents a fractional differential of the pipeline pressure P, represents a fractional differential of the powder flowability L.

[0098] In the embodiment, since pressure directly drives fluid flow, and there is a nonlinear relationship between pressure and flow, the pipeline pressure has a greater impact on the flow; since the pipeline pressure will experience relatively severe fluctuations and rapid response, the differential order is large, the change of the powder flowability is usually slow and continuous, and its impact on the flow is less sensitive than that of the pressure, so a smaller differential order is used, and therefore a = 0.5 and β = 0.3 are taken.

[0099] A state transition unit is used to establish a state transition model of the dry powder fire extinguishing system, specifically to estimate the state evolution of the dry powder fire extinguishing system at different time points through a state transition matrix. In the embodiment, a linear relationship is used to represent the state change of the dry powder fire extinguishing system over time, and the following logic is used to represent the system state transition model:

[0100] x k = Ax k-1 + Bu k + w k

[0101] Wherein, xk Let A be the system state vector at time k; A is the state transition matrix, representing the system dynamics from time k-1 to time k, and x k-1 B represents the system state vector at time k-1; B is the input matrix, representing the degree of influence of the fire extinguishing signal on the system state changes, determining how the control input affects the various state variables of the fire extinguishing system. k For control input, it represents the fire extinguishing system activation signal, when u k =0 indicates that the system is in a closed state, u k =1 indicates that the system has started and begun working; w k The process noise is represented by w, which follows a zero-mean Gaussian distribution. k ~N(0,Q), where Q is the covariance matrix of the process noise.

[0102] The observation model unit is used to establish the observation model of the dry powder fire extinguishing system. Specifically, it uses an observation matrix to match the actual data obtained from the sensors with the state vector to obtain a more accurate state estimate. The sensor observations are described by observation equations, which are represented by the following logic:

[0103] z k =Hx k +v k

[0104] Among them, z k This represents the observation data at time k. P2 represents pressure, output from a pressure sensor; L2 represents powder flowability, output from a powder flowability sensor; V2 represents vibration, output from a pipe vibration sensor; T2 represents temperature, output from a temperature sensor; D2 represents humidity, output from a humidity sensor; H represents the observation matrix, describing the relationship between the observed data and the state vector; v k This represents the observation noise, which follows a zero-mean Gaussian distribution, v k ~N(0,R), where R is the covariance matrix of the observation noise.

[0105] The state prediction and update unit, based on the Kalman filter algorithm, predicts the current state of the dry powder fire extinguishing system and updates the current observation data. Specifically:

[0106] In this embodiment, the Kalman filter algorithm includes two stages: prediction and update. In the prediction stage, based on the state estimate of the previous time step, the current state and the error covariance matrix are predicted, using the following logical representation:

[0107]

[0108] P k|k-1 =AP k-1|k-1 AT +Q

[0109] wherein, represents the state prediction of the current time based on the state estimation of the last time, P k|k-1 represents the covariance matrix of the state estimation error at the prediction time k, P k-1|k-1 represents the estimation error covariance matrix of the last time, represents the state estimation of the last time.

[0110] In the update stage, new observation data z k , the system state is updated, specifically, the state estimation and error covariance matrix are calculated and updated using Kalman gain, and the following logic is used to represent:

[0111] K k = P k|k-1 H T (HP k|k-1 H T +R) -1

[0112]

[0113] P k =(I-K k H)P k|k-1

[0114] wherein, is an identity matrix, represents the Kalman gain, and R represents the observation noise matrix, represents the final state estimation at the k time, represents the updated state estimation error covariance matrix.

[0115] Based on the Kalman filtering algorithm, the real-time data and historical data of the sensor are combined to predict the observation data of the dry powder fire extinguishing system, providing a basic basis for the adaptive alarm and response mechanism module, so that the state monitoring system can identify potential faults or abnormal conditions and alarm in time. Further, the state monitoring system can also predict the flow condition of the fire extinguishing agent and whether there is an obstacle in the pipeline, helping to achieve efficient operation of the system.

[0116] The adaptive alarm and response mechanism module is used for data analysis of multi-dimensional data to determine whether there is abnormal data and respond, comprising:

[0117] An abnormal data determination unit is configured to analyze multi-dimensional data to determine whether there is abnormal data and calculate a comprehensive abnormality degree value.

[0118] The abnormal data refers to data points that do not conform to the expected pattern or threshold under normal operating conditions, which reflect equipment failure, environmental changes, sensor failure or potential fire hazards. In this embodiment, the judgment of abnormal data is based on the degree of deviation of monitoring data from normal behavior pattern, as well as the comparison of historical data and the analysis of prediction model. In this embodiment, the carbon monoxide concentration and oxygen concentration are monitored by the electrochemical sensor of the sensor data acquisition module, and the carbon dioxide concentration is monitored by the non-dispersive infrared (NDIR) sensor, combined with the output of pressure sensor, pipeline vibration sensor and temperature sensor, based on the abnormal data judgment index, whether the temperature, carbon monoxide concentration, carbon dioxide concentration, oxygen concentration, pressure and vibration exist abnormal data, when the above index exceeds the normal range, it is considered that there is abnormal data, specifically:

[0119] When temperature T2∈[T a ,T b ], the temperature T2 is in the normal range, otherwise the temperature T2 is abnormal data, wherein T a is the first temperature threshold, T b is the second temperature threshold;

[0120] When carbon monoxide concentration C CO ∈[C1,C2], the carbon monoxide concentration C CO is in the normal range, otherwise the carbon monoxide concentration C CO is abnormal data, wherein C1 is the first carbon monoxide concentration threshold, C2 is the second carbon monoxide concentration threshold;

[0121] When carbon dioxide concentration C CO2 ∈[C4,C5], the carbon dioxide concentration C CO2 is in the normal range, otherwise the carbon dioxide concentration C CO2 is abnormal data, wherein C4 is the first carbon dioxide concentration threshold, C5 is the second carbon dioxide concentration threshold;

[0122] When oxygen concentration C O2 ∈[C7,C8], the oxygen concentration C O2 is in the normal range, otherwise the oxygen concentration C O2 is abnormal data, wherein C7 is the first oxygen concentration threshold, C8 is the second oxygen concentration threshold;

[0123] When pressure P2∈[P a ,P b ], the pressure P2 is in the normal range, otherwise the pressure P2 is abnormal data, wherein P a is the first pressure threshold, P b is the second pressure threshold;

[0124] When vibration V2∈[Va V b ]time, the vibration V2 is in the normal range, otherwise the vibration V2 is abnormal data, wherein V a is the first vibration threshold value, V b is the second vibration threshold value.

[0125] In the embodiment, the range of each index of the multi-dimensional data can be shown in Table 1 as follows:

[0126] Table 1 Abnormal data judgment index table

[0127]

[0128] As shown in Table 1 above, when the temperature T2 satisfies 0℃≤T2≤40℃, the temperature T2 is in the normal range;

[0129] When the carbon monoxide concentration C CO satisfies 0ppm≤C CO ≤10ppm, the carbon monoxide concentration C CO is in the normal range;

[0130] When the carbon dioxide concentration C CO2 satisfies 400ppm≤C CO2 ≤1000ppm, the carbon dioxide concentration C CO2 is in the normal range;

[0131] When the oxygen concentration C O2 satisfies 19.5%≤C O2 ≤21%, the oxygen concentration C O2 is in the normal range;

[0132] When the pressure P2 satisfies 0.5MPa≤P2≤2MPa, the pressure P2 is in the normal range;

[0133] When the vibration V2 satisfies 0.1mm / s≤V2≤0.5mm / s, the vibration V2 is in the normal range.

[0134] In the embodiment, when any one of the temperature T2, the carbon monoxide concentration C CO , the carbon dioxide concentration C CO2 , the oxygen concentration C O2 , the pressure P2 and the vibration V2 has abnormal data, the multi-dimensional data is considered to have abnormal data. First, the abnormal degree value of each index is calculated respectively, and the temperature abnormal degree Q T is calculated by using the following logic:

[0135]

[0136] Wherein, T cThe third temperature threshold is shown in Table 1 above. c 50℃ is acceptable.

[0137] The degree of carbon monoxide concentration anomaly Q is calculated using the following logic. cO :

[0138]

[0139] C3 is the threshold concentration of the third carbon monoxide, and as shown in Table 1 above, C3 can be 30 ppm.

[0140] The degree of carbon dioxide concentration anomaly Q is calculated using the following logic. CO2 :

[0141]

[0142] C6 is the third carbon dioxide concentration threshold, and as shown in Table 1 above, C6 can be 5000 ppm.

[0143] The degree of oxygen concentration anomaly Q is calculated using the following logic. O2 :

[0144]

[0145] C9 is the third oxygen concentration threshold, and as shown in Table 1 above, C9 can be 19%.

[0146] The degree of pressure anomaly Q is calculated using the following logic. P :

[0147]

[0148] Among them, P c P is the third pressure threshold. d The fourth pressure threshold is shown in Table 1 above. c 0.3 MPa can be taken, P d 3MPa is acceptable.

[0149] The degree of vibration anomaly Q is calculated using the following logic. v :

[0150]

[0151] Among them, V c The third vibration threshold is shown in Table 1 above. c 1 mm / s is acceptable.

[0152] Secondly, the overall anomaly level value Q is calculated using the following logical representation:

[0153]

[0154] wherein Q represents a comprehensive abnormality degree value, Q i represents an abnormality degree value of the i-th index, i∈[1,n], n represents the total number of indexes, w i represents the weight corresponding to Q i ; in this embodiment, i=1,...,6, n takes 6, then Q i represents Q T , Q CO , Q CO2 , Q O2 , Q P and Q V , respectively.

[0155] In this embodiment, considering that the underground substation is relatively closed and the ventilation condition is limited. Once a fire occurs, smoke and harmful gases are difficult to spread quickly, and oxygen is consumed faster, so that oxygen concentration monitoring becomes important, but its weight is still slightly lower compared with temperature, carbon monoxide, etc.; and the pressure and vibration are mainly considered from the perspective of equipment safety and potential failure to cause fire, so their weights can be slightly lower than the indexes directly related to the fire burning process. This embodiment provides a set of corresponding weight values of each index as follows:

[0156] The corresponding weight w1 of the temperature abnormality degree Q T takes 0.25, the corresponding weight w2 of the carbon monoxide concentration abnormality degree Q CO takes 0.25, the corresponding weight w3 of the carbon dioxide concentration abnormality degree Q CO2 takes 0.15, the corresponding weight w4 of the oxygen concentration abnormality degree Q O2 takes 0.15, the corresponding weight w5 of the pressure abnormality degree Q P takes 0.1, and the corresponding weight w6 of the vibration abnormality degree Q V takes 0.1.

[0157] An abnormality type and severity judgment unit is configured to control the system to make a response according to the comprehensive abnormality degree value, judge the abnormality type and severity, and send corresponding warning information, specifically:

[0158] When the comprehensive abnormality degree value Q∈[Q1,Q2], it is judged as a slight abnormality, wherein Q1 is a first abnormality threshold, and Q2 is a second abnormality threshold; at this time, the state monitoring system automatically records the abnormal data and time, issues a yellow warning, and reminds the operation and maintenance personnel to pay attention to the substation equipment.

[0159] When the comprehensive abnormality degree value Q∈[Q2,Q3], it is judged as a serious abnormality, wherein Q3 is a third abnormality threshold; at this time, the state monitoring system issues an orange alarm and notifies the on-site operator to check the equipment state and determine whether maintenance or preventive measures are needed.

[0160] When the comprehensive abnormality degree value Q is in the range of [Q3, Q4], it is judged as an extremely dangerous abnormality, wherein Q4 is a fourth abnormality threshold value; at this time, the state monitoring system immediately issues a red emergency alarm, immediately alarms the emergency personnel and the operation and maintenance personnel; the automatic fire extinguishing system is started, and dry powder spraying is carried out; the substation equipment is automatically powered off, and the power supply to the high-risk area is cut off to prevent the spread of fire.

[0161] In the embodiment, the abnormality type is judged according to the comprehensive abnormality degree value as shown in Table 2:

[0162] Table 2 Abnormality type table

[0163] Abnormality type Mild abnormality Severe abnormality Extremely dangerous abnormality Integrated abnormality degree value (Q) 0＜Q≤0.3 0.3＜Q≤0.7 Q＞0.7

[0164] Further, the threshold range of the abnormality data judgment index and the abnormality type judgment threshold in the adaptive alarm and response mechanism module can be adjusted according to the actual situation to meet different fire fighting needs.

[0165] The remote monitoring and fault diagnosis module is used for uploading the output results of the sensor data acquisition module, the intelligent data fusion and analysis module and the adaptive alarm and response mechanism module to a cloud monitoring platform for remote monitoring and fault diagnosis.

[0166] In the embodiment, the multi-dimensional data collected by the sensor data acquisition module, the system running state predicted by the intelligent data fusion and analysis module and the early warning information sent by the adaptive alarm and response mechanism module are uploaded to the cloud monitoring platform in real time, and the cloud monitoring platform outputs charts and reports to display the real-time state of the system. The operation and maintenance personnel can perform remote monitoring and fault diagnosis through the cloud monitoring platform, and can timely master the safe operation of the substation.

[0167] Further, the embodiment can also help the operation and maintenance personnel to predict future risks by providing analysis and trend prediction of historical data, and to remotely debug or update the system by adjusting the system parameters through the cloud monitoring platform; if the system fails, the cloud monitoring platform diagnoses the fault type and provides corresponding repair suggestions.

[0168] The application also provides a dry powder fire extinguishing system state monitoring method for an underground substation, comprising the following steps:

[0169] S1, multi-dimensional data of the dry powder fire extinguishing system is collected to monitor the running state of the dry powder fire extinguishing system in real time.

[0170] The sensor data acquisition module comprises a pressure sensor, a powder flowability sensor, a pipeline vibration sensor, a temperature sensor, a humidity sensor, an electrochemical sensor and a non-dispersive infrared (NDIR) sensor; a plurality of sensor data are collected in real time, and the data are transmitted to the cloud monitoring platform in real time through a wireless mode.

[0171] The multi-dimensional data includes pressure, powder fluidity, vibration, temperature, humidity, carbon monoxide concentration, carbon dioxide concentration and oxygen concentration.

[0172] S2, the multi-dimensional data collected by S1 is processed to predict the running state of the dry powder fire extinguishing system; S2 further includes the following steps:

[0173] S21, the state of the dry powder fire extinguishing system is modeled based on historical data, and a system state vector is calculated.

[0174] S22, a state transition model of the dry powder fire extinguishing system is established, specifically, a state transition matrix is used to estimate the state evolution of the dry powder fire extinguishing system at different time points.

[0175] S23, an observation model of the dry powder fire extinguishing system is established, specifically, an observation matrix is used to match the actual data obtained from the sensor with the state vector to obtain more accurate state estimation, and the observation value of the sensor is described by an observation equation.

[0176] S24, the state of the dry powder fire extinguishing system at the current time is predicted based on the Kalman filtering algorithm, and the observation data at the current time is updated.

[0177] S3, the multi-dimensional data is analyzed to determine whether there is abnormal data and to respond; S3 includes the following steps:

[0178] S31, the multi-dimensional data is analyzed to determine whether there is abnormal data, and a comprehensive abnormality degree value is calculated.

[0179] S32, the control system responds according to the comprehensive abnormality degree value, judges the abnormal type and the abnormal severity, and sends corresponding warning information.

[0180] S4, the output results of S1-S3 are uploaded to a cloud monitoring platform for remote monitoring and fault diagnosis.

[0181] The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A status monitoring system for a dry powder fire extinguishing system in an underground substation, characterized in that, include: The sensor data acquisition module is used to collect multi-dimensional data of the dry powder fire extinguishing system and monitor the operating status of the dry powder fire extinguishing system in real time. The intelligent data fusion and analysis module is used to process the multi-dimensional data collected by the sensor data acquisition module and predict the operating status of the dry powder fire extinguishing system. The adaptive alarm and response mechanism module is used to perform data analysis on multi-dimensional data, determine whether there is abnormal data, and respond accordingly. The remote monitoring and fault diagnosis module is used to upload the output results of the sensor data acquisition module, intelligent data fusion and analysis module, and adaptive alarm and response mechanism module to the cloud monitoring platform for remote monitoring and fault diagnosis.

2. The status monitoring system for a dry powder fire extinguishing system in an underground substation according to claim 1, characterized in that, The sensor data acquisition module includes a pressure sensor, a powder flowability sensor, a pipeline vibration sensor, a temperature sensor, a humidity sensor, an electrochemical sensor, and a non-dispersive infrared sensor. The sensor data acquisition module collects data from multiple sensors in real time and transmits the data to a cloud monitoring platform wirelessly.

3. The status monitoring system for a dry powder fire extinguishing system in an underground substation according to claim 1, characterized in that, The multi-dimensional data includes pressure, powder flowability, vibration, temperature, humidity, carbon monoxide concentration, carbon dioxide concentration, and oxygen concentration.

4. The status monitoring system for a dry powder fire extinguishing system in an underground substation according to claim 1, characterized in that, The intelligent data fusion and analysis module includes: The state modeling unit models the state of the dry powder fire extinguishing system based on historical data and calculates the system state vector. The state transition unit is used to establish the state transition model of the dry powder fire extinguishing system. Specifically, it estimates the state evolution of the dry powder fire extinguishing system at different time points through the state transition matrix. The observation model unit is used to establish the observation model of the dry powder fire extinguishing system. Specifically, it uses an observation matrix to match the actual data obtained from the sensors with the state vector to obtain a more accurate state estimate. The sensor observations are described by the observation equation. The state prediction and update unit uses the Kalman filter algorithm to predict the current state of the dry powder fire extinguishing system and update the current observation data.

5. The status monitoring system for a dry powder fire extinguishing system in an underground substation according to claim 1, characterized in that, The adaptive alarm and response mechanism module includes: The abnormal data judgment unit is used to perform data analysis on multi-dimensional data to determine whether there is abnormal data and to calculate the comprehensive abnormality value. The anomaly type and severity judgment unit is used to control the system to respond based on the comprehensive anomaly degree value, judge the anomaly type and severity, and send corresponding early warning information.

6. The status monitoring system for a dry powder fire extinguishing system in an underground substation according to claim 5, characterized in that, The abnormal data judgment unit determines whether abnormal data exists by: judging whether abnormal data exists in temperature, carbon monoxide concentration, carbon dioxide concentration, oxygen concentration, pressure, and vibration based on the abnormal data judgment index. When temperature T2∈[T a ,T b When T2 is within the normal range, otherwise T2 is considered abnormal. a T is the first temperature threshold. b This is the second temperature threshold; When the carbon monoxide concentration C CO When C ∈ [C1, C2], the carbon monoxide concentration C CO If it is within the normal range, otherwise the carbon monoxide concentration C CO These are abnormal data, where C1 is the first carbon monoxide concentration threshold and C2 is the second carbon monoxide concentration threshold. When carbon dioxide concentration C CO2 When ∈[C4,C5], the carbon dioxide concentration C CO2 If it is within the normal range, otherwise the carbon dioxide concentration C CO2 These are abnormal data, where C4 is the first carbon dioxide concentration threshold and C5 is the second carbon dioxide concentration threshold. When oxygen concentration C O2 When ∈[C7,C8], the oxygen concentration C O2 If it is within the normal range, otherwise the oxygen concentration C O2 These are abnormal data, where C7 is the first oxygen concentration threshold and C8 is the second oxygen concentration threshold. When pressure P2∈[P a ,P b When [condition], pressure P2 is within the normal range; otherwise, pressure P2 is abnormal. Where P... a P is the first pressure threshold. b This is the second pressure threshold; When vibration V2∈[V a V b When the vibration V2 is within the normal range, it is considered normal; otherwise, the vibration V2 is considered abnormal. a V is the first vibration threshold. b This is the second vibration threshold.

7. A status monitoring system for a dry powder fire extinguishing system in an underground substation according to claim 6, characterized in that, The calculation of the comprehensive anomaly degree value in the abnormal data judgment unit is specifically as follows: The degree of temperature anomaly Q is calculated using the following logic. T : Among them, T c This is the third temperature threshold; The degree of carbon monoxide concentration anomaly Q is calculated using the following logic. CO : Wherein, C3 is the third carbon monoxide concentration threshold; The degree of carbon dioxide concentration anomaly Q is calculated using the following logic. CO2 : C6 is the third carbon dioxide concentration threshold; The degree of oxygen concentration anomaly Q is calculated using the following logic. O2 : C9 is the third oxygen concentration threshold. The degree of pressure anomaly Q is calculated using the following logic. P : Among them, P c The third pressure threshold, P d The fourth pressure threshold is used; the degree of vibration anomaly Q is calculated using the following logic. V : Among them, V c The third vibration threshold; The overall anomaly level value Q is calculated using the following logical representation: Where Q represents the overall anomaly level value, Q i Let w represent the degree of abnormality of the i-th indicator, where i ∈ [1, n], and n represents the total number of indicators. i Indicates the corresponding Q i The weight.

8. A status monitoring system for a dry powder fire extinguishing system in an underground substation according to claim 7, characterized in that, The specific anomaly type and severity judgment unit is as follows: When the comprehensive anomaly severity value Q∈[Q1,Q2], it is judged as a minor anomaly, where Q1 is the first anomaly threshold and Q2 is the second anomaly threshold; at this time, the status monitoring system automatically records the abnormal data and time, issues a yellow warning, and reminds the operation and maintenance personnel to pay attention to the substation equipment; When the comprehensive anomaly level value Q∈[Q2,Q3], it is judged as a serious anomaly, where Q3 is the third anomaly threshold; at this time, the status monitoring system issues an orange alarm and notifies the on-site operators to check the equipment status and determine whether maintenance or preventive measures are required. When the comprehensive anomaly level value Q∈[Q3,Q4], it is judged as an extremely dangerous anomaly, where Q4 is the fourth anomaly threshold; at this time, the status monitoring system immediately issues a red emergency alarm and immediately alerts emergency personnel and maintenance personnel; the fire extinguishing system is automatically activated to carry out dry powder spraying; the substation equipment is automatically de-energized to cut off the power supply to the high-risk area and prevent the fire from spreading.

9. A method for monitoring the status of a dry powder fire extinguishing system in an underground substation, characterized in that, Includes the following steps: S1 collects multi-dimensional data from the dry powder fire extinguishing system and monitors the operating status of the dry powder fire extinguishing system in real time. S2 processes the multi-dimensional data collected in S1 to predict the operating status of the dry powder fire extinguishing system. S3 performs data analysis on multi-dimensional data to determine if there is any abnormal data and respond accordingly. S4 uploads the output results of S1 to S3 to the cloud monitoring platform for remote monitoring and fault diagnosis.

10. A method for monitoring the status of a dry powder fire extinguishing system in an underground substation according to claim 9, characterized in that, S2 includes: S21, Model the state of the dry powder fire extinguishing system, model the state of the dry powder fire extinguishing system based on historical data, and calculate the system state vector; S22, Establish a state transition model for the dry powder fire extinguishing system, specifically by estimating the state evolution of the dry powder fire extinguishing system at different time points using the state transition matrix; S23. Establish an observation model for the dry powder fire extinguishing system. Specifically, this involves using an observation matrix to match the actual data obtained from the sensors with the state vector to obtain a more accurate state estimate. The sensor observations are described by observation equations. S24. Based on the Kalman filter algorithm, the current state of the dry powder fire extinguishing system is predicted, and the current observation data is updated.

Images