An artificial intelligence-based power plant safety monitoring system

Abstract

The application belongs to the technical field of power plant safety monitoring, and discloses a power plant safety monitoring system based on artificial intelligence; comprising the following modules: a data acquisition module, which is used for collecting environment data and equipment data of all monitoring areas in the power plant in real time. The application predicts the remaining time of fire occurrence through the setting of a fire occurrence remaining time prediction model and support scoring, and promotes the fire prediction from simple probability evaluation to accurate time prediction. The accurate prediction of the time interval from the current time to the time of fire occurrence makes the implementation of the prevention and control measures more targeted compared with the traditional fire warning system. On this basis, the support scoring is also established, and the state parameters such as the battery capacity of the robot, the remaining amount of fire extinguishing agent, historical reliability and arrival time are comprehensively considered to realize the optimal scheduling of the fire extinguishing robot. The intelligent scheduling method improves the utilization rate of the robot, and at the same time ensures that the prevention and control deployment is completed before the fire occurs.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power plant safety monitoring, and more particularly to a power plant safety monitoring system based on artificial intelligence. BACKGROUND

[0002] As a core component of national energy infrastructure, the safety production of power plants is related to social stability and economic development. With the expansion of power plant scale and the complexity of equipment, the traditional safety monitoring system has been difficult to meet the safety management needs of modern power plants.

[0003] The prior art lacks quantitative assessment and prediction capability for fire risk, and most systems can only achieve post-alarm, and cannot effectively warn before a fire occurs. When a fire is detected, the fire has often formed a scale, missing the best control opportunity. In terms of emergency resource scheduling, the existing system has serious deficiencies. Traditional fire-fighting robot scheduling mostly adopts fixed preplans or simple shortest path principles, lacking comprehensive consideration of robot state and environmental conditions. When multiple areas simultaneously issue warnings, the system cannot optimize scheduling according to the state parameters of the robot's battery power, fire extinguishing agent reserves, reliability, etc. In addition, the existing scheduling system usually ignores time constraints, which may cause the robot to arrive at the scene after the fire occurs, completely losing the significance of early warning and prevention. SUMMARY

[0004] The present application is proposed to solve the problems in the background art, and provides a power plant safety monitoring system based on artificial intelligence.

[0005] To achieve the above purpose, the present application provides the following technical scheme: a power plant safety monitoring system based on artificial intelligence, comprising the following modules:

[0006] A data acquisition module for real-time acquisition of environmental data and equipment data of all monitoring areas in the power plant;

[0007] A risk assessment module for generating environmental risk scores and equipment risk scores of all monitoring areas according to the collected environmental data and equipment data, generating a total risk score according to the environmental risk scores and equipment risk scores, and marking all monitoring areas as different levels of risk areas according to the total risk score;

[0008] A fire occurrence probability prediction module for obtaining historical fire occurrence probability data and constructing a fire occurrence probability prediction model to predict the fire occurrence probability of high-risk areas;

[0009] A fire occurrence remaining time prediction module for obtaining historical fire occurrence remaining time data and constructing a fire occurrence remaining time prediction model;

[0010] The evaluation scheduling module is configured to predict the fire occurrence remaining time of the unit area by using the fire occurrence remaining time prediction model, obtain the supportable fire extinguishing robots conforming to the fire occurrence remaining time, evaluate all the supportable fire extinguishing robots according to the support scores, and dispatch a proper number of fire extinguishing robots for support according to the support scores.

[0011] Further, the area in the power plant is divided into a plurality of monitoring areas with the same area, and monitoring points are arranged in the monitoring areas, and the monitoring points obtain environmental data and equipment data through monitoring equipment;

[0012] The environmental data includes combustible / toxic gas concentration, smoke concentration, environmental temperature, environmental humidity and dust concentration in the monitoring area;

[0013] The equipment data includes vibration spectrum, abnormal soundprint, surface temperature and current abnormality of the equipment in the monitoring area;

[0014] The combustible / toxic gas concentration is obtained by a gas sensor;

[0015] The smoke concentration is obtained by a smoke sensor;

[0016] The environmental temperature and humidity are obtained by a temperature and humidity sensor;

[0017] The dust concentration is obtained by a dust sensor;

[0018] The vibration spectrum is obtained by a vibration sensor;

[0019] The abnormal soundprint is obtained by an acoustic sensor, and the abnormal noise is identified by FFT analysis;

[0020] The surface temperature is obtained by an infrared thermal imager;

[0021] The current abnormality is obtained by a current sensor, and the deviation of the current from the rated value is calculated.

[0022] Further, the environmental data and equipment data are processed:

[0023] The combustible / toxic gas concentration is normalized:

[0024]

[0025] In the formula, is the normalized combustible / toxic gas concentration, is the measured combustible / toxic gas concentration, and are the lower limit and upper limit of the combustible / toxic gas concentration;

[0026] The smoke concentration is normalized:

[0027]

[0028] wherein, is the normalized smoke concentration, is the measured smoke concentration, and are the lower and upper limits of the smoke concentration;

[0029] The ambient temperature is normalized:

[0030]

[0031] wherein, is the normalized ambient temperature, is the measured ambient temperature, and are the lower and upper limits of the ambient temperature;

[0032] The ambient humidity is normalized:

[0033]

[0034] wherein, is the normalized ambient humidity, is the measured ambient humidity, and are the lower and upper limits of the ambient humidity;

[0035] The dust concentration is normalized:

[0036]

[0037] wherein, is the normalized dust concentration, is the measured dust concentration, and are the lower and upper limits of the dust concentration;

[0038] The vibration spectrum is normalized:

[0039]

[0040] wherein, is the normalized vibration spectrum, is the measured vibration spectrum, and are the lower and upper limits of the vibration spectrum;

[0041] The abnormal acoustic fingerprint is normalized:

[0042]

[0043] wherein, is the normalized abnormal acoustic print, is the measured abnormal acoustic print, and are the lower and upper limits of the abnormal acoustic print;

[0044] The surface temperature is normalized:

[0045]

[0046] wherein, is the normalized surface temperature, is the measured surface temperature, and are the lower and upper limits of the surface temperature;

[0047] The current abnormality is normalized:

[0048]

[0049] wherein, is the normalized current abnormality, is the measured current abnormality, and are the lower and upper limits of the current abnormality.

[0050] Further, the process of generating the environmental risk score and the equipment risk score of the entire monitoring area according to the collected environmental data and equipment data comprises:

[0051] The environmental risk score S:

[0052]

[0053] wherein, , , , and are weight coefficients, which are trained according to historical data;

[0054] The equipment risk score P:

[0055]

[0056] wherein, , , and are weight coefficients, which are trained according to historical data.

[0057] Further, the process of generating the total risk score according to the environmental risk score and the equipment risk score, and marking the entire monitoring area as a risk area of different levels according to the total risk score comprises:

[0058] Total risk score R:

[0059]

[0060] wherein and are weight coefficients, which are trained according to historical data;

[0061] Setting a suitable score threshold according to historical risk score data, which refers to a data set of total risk scores of the monitoring area in the past, comparing the total risk score of the monitoring area with the score threshold, and marking the monitoring area as a risk area of different levels;

[0062] When R≤X, the monitoring area is determined as a low-risk area;

[0063] When X≤R≤Y, the monitoring area is determined as a medium-risk area;

[0064] When R≥Y, the monitoring area is determined as a high-risk area.

[0065] Further, historical fire occurrence probability data is obtained and a fire occurrence probability prediction model is constructed, and the process of predicting the fire occurrence probability of the high-risk area by using the fire occurrence probability prediction model includes:

[0066] The fire occurrence probability refers to the probability of possible fire occurrence in the monitoring area;

[0067] The factors affecting the fire occurrence probability of the high-risk area include the environmental risk score, the equipment risk score, the historical fire frequency, the equipment aging index and the personnel activity density of the monitoring area;

[0068] The historical fire frequency refers to the number of past fires in the monitoring area, which comes from a historical database;

[0069] The equipment aging index refers to the service life of the equipment divided by the service life of the equipment, which comes from an equipment archive;

[0070] The personnel activity density is obtained by calculating the number of personnel in a unit time;

[0071] Setting a monitoring period, obtaining historical fire occurrence probability data of a single monitoring area in different monitoring periods, which includes the average environmental risk score, the average equipment risk score, the average historical fire frequency, the average equipment aging index and the average personnel activity density of the single monitoring area in different monitoring periods, and the historical fire occurrence probability of the single monitoring area in the monitoring period;

[0072] generate a fire occurrence probability prediction set according to the environmental risk score, the equipment risk score, the historical fire frequency, the equipment aging index, the personnel activity density and the corresponding historical fire occurrence probability of the corresponding monitoring area in the different historical fire occurrence probability data, and divide the fire occurrence probability prediction set into a first training set and a first test set;

[0073] construct a first convolutional neural network, take the environmental risk score, the equipment risk score, the historical fire frequency, the equipment aging index and the personnel activity density in the different historical fire occurrence probability data in the first training set as the input data of the first convolutional neural network, and take the corresponding historical fire occurrence probability in the first training set as the output data of the first convolutional neural network;

[0074] train the first convolutional neural network to obtain a first initial convolutional neural network, perform model verification on the first initial convolutional neural network by using the first test set, and output the first initial convolutional neural network with a first test error threshold value less than or equal to a preset first test error threshold value as a fire occurrence probability prediction model;

[0075] input the average environmental risk score, the average equipment risk score, the average historical fire frequency, the average equipment aging index and the average personnel activity density of each monitoring area in a monitoring period to the fire occurrence probability prediction model to obtain the predicted fire occurrence probability of each monitoring area;

[0076] set a suitable probability threshold according to historical probability data, the historical probability data refers to a data set of past monitoring area fire occurrence probability, compare the predicted fire occurrence probability of the monitoring area with the probability threshold value, when the predicted fire occurrence probability is greater than the probability threshold value, mark the monitoring area as a fire-prone area, remind the inspection personnel and the staff to evacuate quickly, and timely inform the relevant departments, set the evacuation route of the inspection personnel and the staff to the safety exit, and the evacuation route needs to be away from the fire-prone area.

[0077] Further, the process of obtaining historical fire occurrence remaining time data and constructing a fire occurrence remaining time prediction model includes:

[0078] divide the fire-prone area into a plurality of unit areas with the same area;

[0079] the fire occurrence remaining time refers to the time from the present moment to the ignition in the unit area;

[0080] the factors affecting the fire occurrence remaining time of the unit area include combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree and oxygen consumption rate;

[0081] the combustible gas concentration is measured in real time by a combustible gas sensor;

[0082] The temperature change rate refers to the temperature change per unit time, and is calculated based on historical data measured by a temperature sensor;

[0083] The equipment insulation aging index is derived from historical maintenance data;

[0084] The volatile material is detected by a volatile organic compound sensor, and the material pyrolysis degree is obtained by combining the two cases through multi-spectral imaging analysis of the color change of the material;

[0085] The oxygen consumption rate is obtained through time series analysis of oxygen sensor data;

[0086] Obtain historical fire occurrence remaining time data of a single unit area, wherein the historical fire occurrence remaining time data includes combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree, oxygen consumption rate of the single unit area, and historical fire occurrence remaining time of the single unit area;

[0087] According to the combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree, oxygen consumption rate and corresponding historical fire occurrence remaining time of the corresponding unit area in different historical fire occurrence remaining time data, a fire occurrence remaining time prediction set is generated, and is divided into a second training set and a second test set;

[0088] A second convolutional neural network is constructed, and the combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree and oxygen consumption rate in different historical fire occurrence remaining time data in the second training set are taken as input data of the second convolutional neural network, and the corresponding historical fire occurrence remaining time in the second training set is taken as output data of the second convolutional neural network;

[0089] The second convolutional neural network is trained to obtain a second initial convolutional neural network, and the second test set is used to verify the model of the second initial convolutional neural network, and the second initial convolutional neural network with a second test error threshold less than or equal to a preset second test error threshold is output as a fire occurrence remaining time prediction model.

[0090] Further, the process of predicting the fire occurrence remaining time of the unit area by using the fire occurrence remaining time prediction model and obtaining the supportable fire extinguishing robot in accordance with the fire occurrence remaining time comprises:

[0091] The supportable fire extinguishing robot refers to a fire extinguishing robot that can reach the unit area to be supported within the fire occurrence remaining time;

[0092] The unit area to be supported refers to a unit area where a fire is about to occur;

[0093] When it is needed to predict the fire occurrence remaining time of each unit area, the combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree and oxygen consumption rate of each unit area are input into the fire occurrence remaining time prediction model to obtain the predicted fire occurrence remaining time of each unit area;

[0094] The walking distance of the fire extinguishing robot from the unit area to be supported is divided by the walking speed of the fire extinguishing robot to obtain the arrival time of the fire extinguishing robot to the unit area to be supported, and the arrival time is compared with the predicted fire occurrence remaining time. The fire extinguishing robot with the arrival time less than the predicted fire occurrence remaining time is marked as a supportable fire extinguishing robot, and the fire extinguishing robot with the arrival time greater than the predicted fire occurrence remaining time is marked as an unsupportable fire extinguishing robot.

[0095] Further, the process of establishing a support score to evaluate all supportable fire extinguishing robots includes:

[0096] The support parameters of the supportable fire extinguishing robot are collected, and the support parameters include the battery capacity of the fire extinguishing robot, the extinguishing agent remaining amount, the historical failure index and the arrival time;

[0097] The battery capacity of the fire extinguishing robot is directly obtained from the battery management system of the fire extinguishing robot;

[0098] The extinguishing agent remaining amount of the fire extinguishing robot is obtained from the agent storage sensor of the fire extinguishing robot;

[0099] The historical failure index is calculated based on the maintenance record and failure history of the robot;

[0100] The support parameters are processed:

[0101] The battery remaining amount is normalized:

[0102]

[0103] In the formula, is the normalized battery remaining amount, is the measured battery remaining amount, is the maximum battery capacity;

[0104] The extinguishing agent remaining amount is normalized:

[0105]

[0106] In the formula, is the normalized extinguishing agent remaining amount, is the measured extinguishing agent remaining amount, is the maximum extinguishing agent amount;

[0107] The historical failure index is normalized as follows:

[0108]

[0109] wherein, is the normalized historical failure index, is the actual number of failures in a year, is the maximum allowed number of failures in a year;

[0110] The arrival time is normalized as follows:

[0111]

[0112] wherein, is the normalized arrival time, is the actual arrival time, is the maximum arrival time, i.e. the predicted remaining time to fire;

[0113] The support score E is:

[0114]

[0115] wherein, , , and are weight coefficients, which are trained according to historical data.

[0116] Further, the process of dispatching a proper number of fire-fighting robots for support according to the support score comprises:

[0117] The unit area in the high fire-prone area is about to have a fire, the remaining time to fire of the unit area is predicted by a remaining time to fire prediction model, the number of fire-fighting robots required is determined according to the predicted remaining time to fire, the longer the predicted remaining time to fire, the fewer the number of fire-fighting robots required, and the shorter the predicted remaining time to fire, the more the number of fire-fighting robots required;

[0118] Mark all fire extinguishing robots in the power plant as supportable fire extinguishing robots and non-supportable fire extinguishing robots according to the remaining time of fire occurrence, evaluate all supportable fire extinguishing robots through support scores, arrange the supportable fire extinguishing robots in order from high to low according to the support scores, select a proper number of supportable fire extinguishing robots according to the predicted remaining time of fire occurrence, mark the selected supportable fire extinguishing robots as quick support fire extinguishing robots, dispatch the quick support fire extinguishing robots into the unit area to prevent fire or extinguish fire, and mark the unselected supportable fire extinguishing robots and non-supportable fire extinguishing robots as backup fire extinguishing robots to replace the quick support fire extinguishing robots when the quick support fire extinguishing robots cannot continue to work or control the spread of fire outside the unit area.

[0119] The technical effects and advantages of the power plant safety monitoring system based on artificial intelligence are as follows:

[0120] (1) By setting the fire occurrence remaining time prediction model and the support score, the fire prediction is promoted from simple probability evaluation to accurate time prediction, the time interval from the current time to the fire of the unit area is accurately predicted through the key parameters such as the combustible gas concentration, the temperature change rate, the equipment insulation aging index, the material pyrolysis degree and the oxygen consumption rate, compared with the traditional fire warning system, the time dimension prediction makes the implementation of the prevention and control measures more targeted, on this basis, the support score is also established, the state parameters such as the battery capacity of the robot, the fire extinguishing agent remaining amount, the historical reliability and the arrival time are comprehensively considered to realize the optimal scheduling of the fire extinguishing robot, the intelligent scheduling method improves the utilization rate of the robot, and at the same time ensures that the prevention and control deployment is completed before the fire occurs.

[0121] (2) By setting the environmental risk score and the equipment risk score, compared with the traditional single parameter monitoring, the multi-source data fusion method improves the comprehensiveness of risk assessment, by setting the total risk score, the environmental risk score and the equipment risk score are weighted and fused to realize the accurate division of the risk level of the monitoring area, the hierarchical early warning mechanism enables the safety management personnel to quickly identify high-risk areas, preferentially configure protection resources, shorten the response time of accident prevention, and predict the probability of fire occurrence in the monitoring area through the fire occurrence probability prediction model, which can play a real-time prevention role. BRIEF DESCRIPTION OF DRAWINGS

[0122] Figure 1 The figure is a schematic diagram of the system of the application. DETAILED DESCRIPTION

[0123] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of the present application.

[0124] With reference to Figure 1 The power plant safety monitoring system based on artificial intelligence comprises the following modules.

[0125] A data acquisition module is configured to acquire environmental data and equipment data of all monitoring areas in the power plant in real time.

[0126] A risk assessment module is configured to generate environmental risk scores and equipment risk scores of all monitoring areas according to the acquired environmental data and equipment data, generate a total risk score according to the environmental risk scores and the equipment risk scores, and mark all monitoring areas as risk areas of different levels according to the total risk score.

[0127] A fire occurrence probability prediction module is configured to acquire historical fire occurrence probability data and construct a fire occurrence probability prediction model, and predict the fire occurrence probability of a high-risk area by using the fire occurrence probability prediction model.

[0128] A fire occurrence remaining time prediction module is configured to acquire historical fire occurrence remaining time data and construct a fire occurrence remaining time prediction model.

[0129] An evaluation and scheduling module is configured to predict the fire occurrence remaining time of a unit area by using the fire occurrence remaining time prediction model, obtain supportable fire extinguishing robots that meet the fire occurrence remaining time, evaluate all supportable fire extinguishing robots by establishing a support score, and schedule a proper number of fire extinguishing robots for support according to the support score.

[0130] It should be further explained that in the specific implementation process, the area in the power plant is divided into a plurality of monitoring areas with the same area, and monitoring points are arranged in the monitoring areas. The monitoring points obtain environmental data and equipment data through monitoring equipment.

[0131] The environmental data comprises combustible / toxic gas concentration, smoke concentration, environmental temperature, environmental humidity and dust concentration in the monitoring area.

[0132] The equipment data comprises vibration spectrum, abnormal voiceprint, surface temperature and current abnormality of equipment in the monitoring area.

[0133] The combustible / toxic gas concentration is obtained by a gas sensor, and reflects the explosion or poisoning risk.

[0134] The smoke concentration is obtained by a smoke sensor, reflecting the early signs of fire;

[0135] The ambient temperature and ambient humidity are obtained by a temperature and humidity sensor, high temperature can cause equipment overheating, and high humidity can cause electrical short circuit;

[0136] The dust concentration is obtained by a dust sensor, dust accumulation can cause explosion or fire;

[0137] The vibration spectrum is obtained by a vibration sensor, reflecting mechanical failure of the equipment;

[0138] The abnormal soundprint is obtained by an acoustic sensor, abnormal noise is identified by FFT analysis, reflecting internal failure of the equipment;

[0139] The surface temperature is obtained by an infrared thermal imager, reflecting equipment overheating;

[0140] The current abnormality is obtained by a current sensor, calculating the deviation of current from the rated value, reflecting electrical failure.

[0141] It needs to be further explained that in the specific implementation process, the environmental data and equipment data are processed:

[0142] The combustible / toxic gas concentration is normalized:

[0143]

[0144] In the formula, is the normalized combustible / toxic gas concentration, is the measured combustible / toxic gas concentration, and are the lower limit and upper limit of the combustible / toxic gas concentration, specifically 0 and 1000 respectively;

[0145] The smoke concentration is normalized:

[0146]

[0147] In the formula, is the normalized smoke concentration, is the measured smoke concentration, and are the lower limit and upper limit of the smoke concentration, specifically 0 and 1000 respectively;

[0148] The ambient temperature is normalized:

[0149]

[0150] In the formula, is the normalized ambient temperature, is the measured ambient temperature, and are the lower and upper limits of the ambient temperature, specifically 0 and 100, respectively;

[0151] The ambient humidity is normalized as follows:

[0152]

[0153] wherein, is the normalized ambient humidity, is the measured ambient humidity, and are the lower and upper limits of the ambient humidity, specifically 0 and 100, respectively;

[0154] The dust concentration is normalized as follows:

[0155]

[0156] wherein, is the normalized dust concentration, is the measured dust concentration, and are the lower and upper limits of the dust concentration, specifically 0 and 500, respectively;

[0157] The vibration spectrum is normalized as follows:

[0158]

[0159] wherein, is the normalized vibration spectrum, is the measured vibration spectrum, and are the lower and upper limits of the vibration spectrum, specifically 0 and 10, respectively;

[0160] The abnormal voiceprint is normalized as follows:

[0161]

[0162] wherein, is the normalized abnormal voiceprint, is the measured abnormal voiceprint, and are the lower and upper limits of the abnormal voiceprint, specifically 0 and 120, respectively;

[0163] The surface temperature is normalized as follows:

[0164]

[0165] wherein, is the normalized surface temperature, the measured surface temperature, and are the lower limit and upper limit of the surface temperature, specifically 0 and 200, respectively;

[0166] The current abnormality is normalized:

[0167]

[0168] In the formula, is the normalized current abnormality, is the measured current abnormality, and are the lower limit and upper limit of the current abnormality, specifically 0 and 50, respectively.

[0169] It should be further explained that, in the specific implementation process, the process of generating the environmental risk score and the equipment risk score of all monitoring areas according to the collected environmental data and equipment data includes:

[0170] Environmental risk score S:

[0171]

[0172] In the formula, , , , and are weight coefficients, which are trained according to historical data and are set to 0.3, 0.2, 0.2, 0.15 and 0.15, respectively;

[0173] If the environmental data in a certain monitoring area is: 500, 500, 50, 10, 300, then the environmental risk score S of the monitoring area is 0.575;

[0174] Equipment risk score P:

[0175]

[0176] In the formula, , , and are weight coefficients, which are trained according to historical data and are set to 0.3, 0.3, 0.2 and 0.2, respectively;

[0177] If the equipment data of a certain monitoring area is: F is 2, V is 30, t is 100, and I is 10, then the equipment risk score P of the equipment in the monitoring area is 0.275.

[0178] It needs to be further explained that, in the specific implementation process, the process of generating the total risk score according to the environmental risk score and the equipment risk score, and marking all monitoring areas as different levels of risk areas according to the total risk score includes:

[0179] The total risk score R is:

[0180]

[0181] In the formula and are weight coefficients, which are trained according to historical data and are set to 0.6 and 0.4 respectively;

[0182] If the environmental risk score S of a monitoring area is 0.575 and the equipment risk score P is 0.275, then the total risk score R of the monitoring area is 0.455;

[0183] According to the historical risk score data, which refers to the data set of the total risk score of the monitoring area in the past, the total risk score of the monitoring area is compared with the score threshold, and the monitoring area is marked as different levels of risk areas;

[0184] When R≤X, the monitoring area is determined as a low-risk area;

[0185] When X≤R≤Y, the monitoring area is determined as a medium-risk area;

[0186] When R≥Y, the monitoring area is determined as a high-risk area;

[0187] If X is 0.4 and Y is 0.6, then the monitoring area with a total risk score R of 0.455 is determined as a medium-risk area.

[0188] It needs to be further explained that, in the specific implementation process, the process of obtaining historical fire occurrence probability data and constructing a fire occurrence probability prediction model, and predicting the fire occurrence probability of the high-risk area using the fire occurrence probability prediction model includes:

[0189] The fire occurrence probability refers to the probability of fire occurring in the monitoring area;

[0190] The factors affecting the fire occurrence probability of the high-risk area include the environmental risk score, the equipment risk score, the historical fire frequency, the equipment aging index, and the personnel activity density of the monitoring area;

[0191] The historical fire frequency refers to the number of past fires in the monitoring area, which comes from a historical database;

[0192] The equipment aging index refers to the equipment service life divided by the equipment life, from the equipment archives;

[0193] The personnel activity density is obtained by calculating the number of personnel in a unit time;

[0194] A monitoring period is set, and historical fire occurrence probability data of a single monitoring area in different monitoring periods is obtained, wherein the historical fire occurrence probability data includes average environmental risk scores, equipment risk scores, historical fire frequencies, equipment aging indexes, personnel activity densities of the single monitoring area in different monitoring periods, and historical fire occurrence probabilities of the single monitoring area in the monitoring periods;

[0195] A fire occurrence probability prediction set is generated according to the environmental risk scores, equipment risk scores, historical fire frequencies, equipment aging indexes, personnel activity densities, and corresponding historical fire occurrence probabilities in different historical fire occurrence probability data corresponding to the monitoring area, and the fire occurrence probability prediction set is divided into a first training set and a first test set;

[0196] A first convolutional neural network is constructed, the environmental risk scores, equipment risk scores, historical fire frequencies, equipment aging indexes, and personnel activity densities in different historical fire occurrence probability data in the first training set are taken as input data of the first convolutional neural network, and the corresponding historical fire occurrence probabilities in the first training set are taken as output data of the first convolutional neural network;

[0197] The first convolutional neural network is trained to obtain a first initial convolutional neural network, the first initial convolutional neural network is subjected to model verification by using the first test set, and the first initial convolutional neural network with a first test error threshold value less than or equal to a preset first test error threshold value is output as a fire occurrence probability prediction model;

[0198] Every time a monitoring period is reached, the average environmental risk scores, equipment risk scores, historical fire frequencies, equipment aging indexes, and personnel activity densities of each monitoring area in the monitoring period are input into the fire occurrence probability prediction model to obtain a predicted fire occurrence probability of each monitoring area;

[0199] A suitable probability threshold value is set according to historical probability data, the historical probability data refers to a data set of past monitoring area fire occurrence probabilities, the predicted fire occurrence probability of the monitoring area is compared with the probability threshold value, when the predicted fire occurrence probability is greater than the probability threshold value, the monitoring area is marked as a fire-prone area, and inspection personnel and workers need to be reminded to evacuate quickly, and relevant departments are informed in time, an evacuation route of the inspection personnel and workers to a safety exit is set, and the evacuation route needs to be away from the fire-prone area;

[0200] In the embodiments of the present application, the predicted fire occurrence probability of all monitoring areas is derived by a fire occurrence probability prediction model, and the predicted fire occurrence probability is related to the environmental risk score, the equipment risk score, the historical fire frequency, the equipment aging index and the personnel activity density;

[0201] The high and low of the environmental risk score directly affect the size of the predicted fire occurrence probability. The higher the environmental risk score is, the better the fire condition given by the environment is, and the greater the predicted fire occurrence probability is. Therefore, the environmental risk score is positively correlated with the fire occurrence probability.

[0202] The high and low of the equipment risk score directly affect the size of the predicted fire occurrence probability. The higher the equipment risk score is, the better the fire condition given by the equipment is, and the greater the predicted fire occurrence probability is. Therefore, the equipment risk score is positively correlated with the fire occurrence probability.

[0203] The more and less of the historical fire frequency directly affect the size of the predicted fire occurrence probability. The more the historical fire frequency is, the easier it is to catch fire. Therefore, the historical fire frequency is positively correlated with the fire occurrence probability.

[0204] The high and low of the equipment aging index directly affect the size of the predicted fire occurrence probability. The higher the equipment aging index is, the worse the condition of the equipment is, and the greater the predicted fire occurrence probability is. Therefore, the equipment aging index is positively correlated with the fire occurrence probability.

[0205] The large and small of the personnel activity density directly affect the size of the predicted fire occurrence probability. The greater the personnel activity density is, the greater the fire risk caused by human error is, and the greater the predicted fire occurrence probability is. Therefore, the personnel activity density is positively correlated with the fire occurrence probability.

[0206] It should be further explained that, in the specific implementation process, the process of obtaining the historical fire occurrence remaining time data and constructing the fire occurrence remaining time prediction model includes:

[0207] The fire high-incidence area is evenly divided into a plurality of unit areas with the same area;

[0208] The fire occurrence remaining time refers to the time from the present moment to the ignition in the unit area;

[0209] The factors affecting the fire occurrence remaining time of the unit area include the combustible gas concentration, the temperature change rate, the equipment insulation aging index, the material pyrolysis degree and the oxygen consumption rate;

[0210] The combustible gas concentration is measured in real time by a combustible gas sensor;

[0211] The temperature change rate refers to the temperature change per unit time, which is measured by a temperature sensor and calculated based on historical data;

[0212] The equipment insulation aging index is derived from historical maintenance data;

[0213] The degree of material pyrolysis is obtained by detecting volatile organic compounds using a volatile organic compound sensor, analyzing material color changes using multispectral imaging, and combining both results.

[0214] The oxygen consumption rate was obtained through time-series analysis of oxygen sensor data.

[0215] Acquire historical fire time remaining data for a single unit area, including combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree, oxygen consumption rate, and historical fire time remaining for that single unit area.

[0216] Based on the combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree, oxygen consumption rate and corresponding historical fire remaining time data of the corresponding unit area, a fire remaining time prediction set is generated and divided into a second training set and a second test set.

[0217] A second convolutional neural network is constructed. The combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree and oxygen consumption rate in the remaining time data of different historical fires in the second training set are used as the input data of the second convolutional neural network, and the remaining time of the corresponding historical fires in the second training set is used as the output data of the second convolutional neural network.

[0218] The second convolutional neural network is trained to obtain the second initial convolutional neural network. The second initial convolutional neural network is validated using the second test set. The second initial convolutional neural network whose output is less than or equal to the preset second test error threshold is used as the fire remaining time prediction model.

[0219] It should be further explained that, in the specific implementation process, the process of predicting the remaining time of fire in a unit area using a fire remaining time prediction model and obtaining a fire-fighting robot that matches the remaining time of fire includes:

[0220] The fire-fighting robot mentioned above refers to a fire-fighting robot that can reach the area of ​​the unit to be supported within the remaining time of a fire.

[0221] The unit area awaiting support refers to the unit area where a fire is about to occur;

[0222] When it is needed to predict the fire occurrence remaining time of each unit area, the combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree and oxygen consumption rate of each unit area are input into the fire occurrence remaining time prediction model to obtain the predicted fire occurrence remaining time of each unit area;

[0223] The walking distance of the fire extinguishing robot from the unit area to be supported is divided by the walking speed of the fire extinguishing robot to obtain the arrival time of the fire extinguishing robot to the unit area to be supported, and the arrival time is compared with the predicted fire occurrence remaining time. The fire extinguishing robot with the arrival time less than the predicted fire occurrence remaining time is marked as a supportable fire extinguishing robot, and the fire extinguishing robot with the arrival time greater than the predicted fire occurrence remaining time is marked as an unsupportable fire extinguishing robot.

[0224] It needs to be further explained that, in the specific implementation process, the process of establishing the support score to evaluate all supportable fire extinguishing robots includes:

[0225] The support parameters of the supportable fire extinguishing robot are collected, and the support parameters include the battery capacity of the fire extinguishing robot, the extinguishing agent remaining amount, the historical failure index and the arrival time;

[0226] The battery capacity of the fire extinguishing robot is directly obtained from the battery management system of the fire extinguishing robot;

[0227] The extinguishing agent remaining amount of the fire extinguishing robot is obtained from the agent storage sensor of the fire extinguishing robot;

[0228] The historical failure index is calculated based on the maintenance record and failure history of the robot;

[0229] The support parameters are processed:

[0230] The battery remaining amount is normalized:

[0231]

[0232] In the formula, is the normalized battery remaining amount, is the measured battery remaining amount, is the maximum battery capacity, specifically 100;

[0233] The extinguishing agent remaining amount is normalized:

[0234]

[0235] In the formula, is the normalized extinguishing agent remaining amount, is the measured extinguishing agent remaining amount, is the maximum amount of extinguishing agent, specifically 100;

[0236] The historical failure index is normalized as follows:

[0237]

[0238] wherein, is the normalized historical failure index, is the actual number of failures in a year, is the maximum allowed number of failures in a year, specifically 10;

[0239] The arrival time is normalized as follows:

[0240]

[0241] wherein, is the normalized arrival time, is the actual arrival time, is the maximum arrival time, i.e., the predicted remaining time to fire, and is set to 10 minutes;

[0242] The support score E is:

[0243]

[0244] wherein, , , and are weight coefficients, which are trained according to historical data and are set to 0.2, 0.3, 0.2 and 0.3, respectively;

[0245] If the support parameter of a fire-fighting robot is: is 80, is 80, is 2, is 2, then the support score E of the fire-fighting robot is 0.8.

[0246] It needs to be further explained that, in the specific implementation process, the process of dispatching a proper number of fire-fighting robots for support according to the support score includes:

[0247] The unit area in the high-fire area is about to have a fire, the remaining time to fire of the unit area is predicted by the remaining time to fire prediction model, the number of fire-fighting robots required is determined according to the predicted remaining time to fire, the longer the predicted remaining time to fire, the fewer the number of fire-fighting robots required, and the shorter the predicted remaining time to fire, the more the number of fire-fighting robots required;

[0248] According to the fire occurrence remaining time, all the fire extinguishing robots in the power plant are marked as supportable fire extinguishing robots and non-supportable fire extinguishing robots, all the supportable fire extinguishing robots are evaluated through support scores, the supportable fire extinguishing robots are arranged in order from high to low according to the support scores, a proper number of supportable fire extinguishing robots are selected according to the predicted fire occurrence remaining time, the supportable fire extinguishing robots are marked as quick support fire extinguishing robots, the quick support fire extinguishing robots are dispatched into the unit area to carry out the fire prevention or fire extinguishing work, the unselected supportable fire extinguishing robots and the non-supportable fire extinguishing robots are used as backup fire extinguishing robots to replace the quick support fire extinguishing robots when the quick support fire extinguishing robots cannot continue to work or to control the fire spread outside the unit area.

[0249] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited to this, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0250] Finally, the above merely provides the preferred embodiments of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be covered in the protection scope of the present application.

Claims

1. A power plant safety monitoring system based on artificial intelligence, characterized in that, Includes the following modules: The data acquisition module is used to collect environmental and equipment data from all monitored areas within the power plant in real time. The risk assessment module is used to generate environmental risk scores and equipment risk scores for all monitored areas based on the collected environmental and equipment data, generate a total risk score based on the environmental and equipment risk scores, and mark all monitored areas as risk areas of different levels based on the total risk score. The fire probability prediction module is used to acquire historical fire probability data and build a fire probability prediction model, and use the fire probability prediction model to predict the fire probability in high-risk areas. The fire remaining time prediction module is used to acquire historical fire remaining time data and build a fire remaining time prediction model. The evaluation and scheduling module is used to predict the remaining time of fire in a unit area using a fire remaining time prediction model, and to obtain the available fire-fighting robots that meet the remaining time of fire. It establishes a support score to evaluate all available fire-fighting robots, and schedules an appropriate number of fire-fighting robots to provide support based on the support score.

2. The power plant safety monitoring system based on artificial intelligence according to claim 1, characterized in that, The area within the power plant is divided into multiple monitoring zones of equal size. Monitoring points are set up within each monitoring zone, and environmental and equipment data are obtained from the monitoring points through monitoring equipment. The environmental data includes the concentration of combustible / toxic gases, smoke concentration, ambient temperature, ambient humidity, and dust concentration within the monitoring area; The equipment data includes the vibration spectrum, abnormal acoustic signature, surface temperature, and current anomaly of the equipment within the monitoring area. The concentration of flammable / toxic gases is obtained through gas sensors; Smoke concentration is obtained through a smoke sensor; Ambient temperature and humidity are obtained through temperature and humidity sensors; Dust concentration is obtained through a dust sensor; The vibration spectrum is obtained through a vibration sensor; Abnormal voiceprints are obtained through acoustic sensors, and abnormal noise is identified through FFT analysis; Surface temperature was obtained using an infrared thermal imager; The current anomaly is obtained through a current sensor, and the deviation of the current from the rated value is calculated.

3. The power plant safety monitoring system based on artificial intelligence according to claim 2, characterized in that, Processing environmental and equipment data: Normalize the concentrations of flammable / toxic gases: In the formula, To normalize the concentration of flammable / toxic gases, This represents the measured concentration of flammable / toxic gases. and These are the lower and upper limits for the concentration of flammable / toxic gases; Normalize the smoke concentration: In the formula, To normalize the smoke concentration, This represents the measured smoke concentration. and These are the lower and upper limits of smoke concentration; The ambient temperature was normalized. In the formula, To normalize the ambient temperature, The actual ambient temperature. and These are the lower and upper limits of the ambient temperature; Normalize the ambient humidity: In the formula, To normalize ambient humidity, The measured ambient humidity and These are the lower and upper limits of ambient humidity. The dust concentration was normalized. In the formula, To normalize dust concentration, This represents the measured dust concentration. and These are the lower and upper limits of dust concentration; The vibration spectrum was normalized. In the formula, For the normalized vibration spectrum, This is the measured vibration spectrum. and These represent the lower and upper limits of the vibration spectrum; Abnormal voiceprints are normalized: In the formula, To normalize abnormal voiceprints, These are abnormal voiceprints measured in practice. and These are the lower and upper limits for abnormal voiceprints; The surface temperature was normalized. In the formula, To normalize the surface temperature, This is the measured surface temperature. and These are the lower and upper limits of the surface temperature; Normalize the current anomaly: In the formula, To normalize the current anomaly degree, This represents the measured current anomaly. and These represent the lower and upper limits of the current anomaly.

4. The power plant safety monitoring system based on artificial intelligence according to claim 3, characterized in that, The process of generating environmental risk scores and equipment risk scores for the entire monitoring area based on the collected environmental and equipment data includes: Environmental Risk Score: In the formula, , , , and These are weighting coefficients, obtained through training based on historical data; Equipment risk score P: In the formula, , , and These are weighting coefficients, obtained through training based on historical data.

5. The power plant safety monitoring system based on artificial intelligence according to claim 4, characterized in that, The process of generating a total risk score based on environmental risk scores and equipment risk scores, and then marking all monitored areas as risk zones of different levels based on the total risk score, includes: Overall risk score R: In the formula and These are weighting coefficients, obtained through training based on historical data; An appropriate scoring threshold is set based on historical risk scoring data, which refers to the set of data on the total risk score of the monitored area in the past. The total risk score of the monitored area is compared with the scoring threshold, and the monitored area is marked as a risk area of ​​different levels. When R≤X, the monitoring area is determined to be a low-risk area; When X≤R≤Y, the monitoring area is determined to be a medium-risk area; When R ≥ Y, the monitoring area is determined to be a high-risk area.

6. The power plant safety monitoring system based on artificial intelligence according to claim 5, characterized in that, The process of acquiring historical fire probability data and constructing a fire probability prediction model, and then using that model to predict the probability of fires in high-risk areas, includes: The probability of fire occurrence refers to the probability that a fire may occur within the monitored area; Factors influencing the probability of fires in high-risk areas include: environmental risk score of the monitored area, equipment risk score, historical fire frequency, equipment aging index, and personnel activity density. Historical fire frequency refers to the number of fires that have occurred in the monitored area in the past, and is derived from a historical database. The equipment aging index is calculated by dividing the equipment's service life by its lifespan, and is derived from equipment records. Personnel activity density is obtained by calculating the number of people per unit time. Set a monitoring period and obtain historical fire occurrence probability data for a single monitoring area in different monitoring periods. The historical fire occurrence probability data includes the average environmental risk score, equipment risk score, historical fire frequency, equipment aging index, personnel activity density, and historical fire occurrence probability of the single monitoring area in different monitoring periods. Based on the environmental risk score, equipment risk score, historical fire frequency, equipment aging index, personnel activity density and corresponding historical fire probability of the monitoring area in different historical fire probability data, a fire probability prediction set is generated and divided into the first training set and the first test set. Construct a first convolutional neural network, using environmental risk score, equipment risk score, historical fire frequency, equipment aging index and personnel activity density from different historical fire occurrence probability data in the first training set as the input data of the first convolutional neural network, and using the corresponding historical fire occurrence probability in the first training set as the output data of the first convolutional neural network. The first convolutional neural network is trained to obtain the first initial convolutional neural network. The first initial convolutional neural network is validated using the first test set. The first initial convolutional neural network, whose output is less than or equal to the preset first test error threshold, is used as the fire occurrence probability prediction model. At each monitoring cycle, the average environmental risk score, equipment risk score, historical fire frequency, equipment aging index and personnel activity density of each monitoring area within the monitoring cycle are input into the fire probability prediction model to obtain the predicted fire probability of each monitoring area. An appropriate probability threshold is set based on historical probability data, which refers to the data set of past fire occurrence probabilities in the monitored area. The predicted fire occurrence probability of the monitored area is compared with the probability threshold. When the predicted fire occurrence probability is greater than the probability threshold, the monitored area is marked as a high-risk fire area. Patrol personnel and staff need to be reminded to evacuate quickly and relevant departments should be notified in a timely manner. Evacuation routes for patrol personnel and staff to safety exits should be set up, and the evacuation routes should be far away from the high-risk fire area.

7. The power plant safety monitoring system based on artificial intelligence according to claim 6, characterized in that, The process of obtaining historical fire time remaining data and building a fire time remaining prediction model includes: The fire-prone area was evenly divided into multiple unit areas of equal size. The remaining time for the fire refers to the time from this moment until the fire starts within the unit area; Factors affecting the remaining time of a fire in a unit area include the concentration of combustible gas, the rate of temperature change, the aging index of equipment insulation, the degree of material pyrolysis, and the oxygen consumption rate. The concentration of combustible gas is measured in real time by a combustible gas sensor; The rate of temperature change refers to the change in temperature per unit time. It is calculated by measuring temperature data through a temperature sensor and based on historical data. The equipment insulation aging index is derived from historical maintenance data; The degree of material pyrolysis is obtained by detecting volatile organic compounds using a volatile organic compound sensor, analyzing material color changes using multispectral imaging, and combining both results. The oxygen consumption rate was obtained through time-series analysis of oxygen sensor data. Acquire historical fire time remaining data for a single unit area, including combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree, oxygen consumption rate, and historical fire time remaining for that single unit area. Based on the combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree, oxygen consumption rate and corresponding historical fire remaining time data of the corresponding unit area, a fire remaining time prediction set is generated and divided into a second training set and a second test set. A second convolutional neural network is constructed. The combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree and oxygen consumption rate in the remaining time data of different historical fires in the second training set are used as the input data of the second convolutional neural network, and the remaining time of the corresponding historical fires in the second training set is used as the output data of the second convolutional neural network. The second convolutional neural network is trained to obtain the second initial convolutional neural network. The second initial convolutional neural network is validated using the second test set. The second initial convolutional neural network whose output is less than or equal to the preset second test error threshold is used as the fire remaining time prediction model.

8. The power plant safety monitoring system based on artificial intelligence according to claim 7, characterized in that, The process of predicting the remaining time of fire in a unit area using a fire time-of-flight prediction model and obtaining a fire-fighting robot that matches the remaining time of fire includes: The fire-fighting robot mentioned above refers to a fire-fighting robot that can reach the area of ​​the unit to be supported within the remaining time of a fire. The unit area awaiting support refers to the unit area where a fire is about to occur; When it is necessary to predict the remaining time of fire in each unit area, the real-time combustible gas concentration, temperature change rate, equipment insulation aging index, material pyrolysis degree and oxygen consumption rate of each unit area are input into the remaining time of fire prediction model to obtain the predicted remaining time of fire in each unit area. All firefighting robots in the power plant were analyzed. The arrival time of the firefighting robot to the area to be supported was obtained by dividing the walking distance of the firefighting robot from the walking speed of the firefighting robot to the area to be supported. The arrival time was compared with the predicted remaining time of the fire. Firefighting robots whose arrival time was less than the predicted remaining time of the fire were marked as firefighting robots that could be supported, and firefighting robots whose arrival time was greater than the predicted remaining time of the fire were marked as firefighting robots that could not be supported.

9. The power plant safety monitoring system based on artificial intelligence according to claim 8, characterized in that, The process of establishing a support rating system to evaluate all available firefighting robots includes: Collect support parameters that can support the firefighting robot, including the firefighting robot's battery level, remaining extinguishing agent, historical failure index, and arrival time; The battery power of the firefighting robot can be obtained directly from its battery management system. The remaining amount of extinguishing agent in the fire-fighting robot is obtained from the agent storage sensor of the fire-fighting robot; The historical failure index is calculated based on the robot's maintenance records and failure history. Processing support parameters: Normalize the remaining battery capacity: In the formula, To normalize the battery capacity, This represents the actual remaining battery capacity. Maximum battery capacity; The remaining extinguishing agent was normalized: In the formula, This is the normalized residual amount of extinguishing agent. This represents the measured residual amount of extinguishing agent. This is the maximum amount of extinguishing agent; The historical failure index was normalized: In the formula, To normalize the historical failure index, This represents the actual number of failures within a year. The maximum number of failures allowed within one year; The arrival time is normalized: In the formula, To normalize arrival time, This is the actual arrival time. The maximum arrival time is the estimated remaining time before the fire occurs. Support rating E is: In the formula, , , and These are weighting coefficients, obtained through training based on historical data.

10. The power plant safety monitoring system based on artificial intelligence according to claim 9, characterized in that, The process of dispatching an appropriate number of firefighting robots to provide support based on the support score includes: A fire is about to break out in a unit area within a high-risk fire zone. The remaining time of the fire in the unit area is predicted using a fire time prediction model. The number of fire-fighting robots required is determined based on the predicted remaining time. The longer the predicted remaining time, the fewer fire-fighting robots are needed. The shorter the predicted remaining time, the more intense the fire and the more fire-fighting robots are needed. Based on the remaining time before the fire occurs, all firefighting robots in the power plant are marked as either supportable or non-supportable. All supportable firefighting robots are evaluated using a support score and ranked from highest to lowest. Based on the predicted remaining time before the fire occurs, an appropriate number of supportable firefighting robots are selected and marked as rapid support firefighting robots. These rapid support firefighting robots are dispatched to the unit area to carry out fire prevention or firefighting work. The unselected supportable and non-supportable firefighting robots serve as backup firefighting robots, which will take over the role of rapid support firefighting robots after they become unable to continue working, or control the spread of fire outside the unit area.

Images