Control method of spraying device for extinguishing oil fire on outer surface of armored car

Abstract

The invention relates to the technical field of armored car safety protection, in particular to a control method of a spraying device for extinguishing oil fire on the outer surface of an armored car, which comprises the following steps: step 1, on the basis of actual fire extinguishing requirements of the outer surface of the armored car, matching corresponding sensors for important areas on the outer surface of the armored car, acquiring outer surface temperature data and environment data of the armored car in real time, and generating a basic data acquisition table; and 2, on the basis of the basic data acquisition table, various data are preprocessed, the temperature change amplitude of the outer surface of the armored car is calculated in real time in cooperation with a sliding window method, an alarm is given out when the temperature change amplitude reaches a set value, and an abnormal temperature detection report is generated. By analyzing the oxygen concentration provided by the airflow and the influence on flame diffusion, the vehicle speed range beneficial for reducing the flame intensity is matched, so that the flame is prevented from quickly spreading under the combined action of the oxygen and the airflow at the initial stage of the flame, and the flame can be efficiently extinguished by foam within a short time.

Description

Technical Field

[0001] The invention relates to the technical field of armored vehicle safety protection control, and in particular to a control method for a spray device for extinguishing oil fires on the outer surface of an armored vehicle. Background Art

[0002] The products of petroleum after processing can be divided into two categories. One is the oil or raw materials used in industrial production, such as solvent oil used in the production of grease, rubber and paint; lubricating oil used as a lubricant on mechanical equipment; vaseline used as rust prevention and pharmaceutical use; paraffin used in the production of wax paper and insulating materials; asphalt used for paving, construction, and preservatives, as well as petroleum coke used in the production of electrodes and silicon carbide; the other is oily fuel. Commonly used oily fuels are mainly divided into four categories: petroleum, kerosene, diesel and heavy oil, which are extremely easy to obtain in daily life; therefore, the outer surface of the armored vehicle is also vulnerable to attack by oil substances, and because oil has strong adhesion and spread, when it catches fire, it may cause serious damage to some important parts of the armored vehicle body along the gaps between the armor.

[0003] At present, the extinguishing of oil fires on the outer surface of armored vehicles mainly relies on manual fire extinguishing equipment or automatic fire extinguishing devices. Manual fire extinguishing equipment requires personnel to respond quickly when a fire occurs, but in a high-pressure combat environment, personnel are often unable to deal with it in time. Automatic fire extinguishing devices rely more on the triggering of fire detectors and the spraying of foam fire extinguishing agents. However, affected by the complex external environment, the spraying of foam fire extinguishing agents may face problems such as accidental touch-off, unformed spraying, and poor extinguishing effect, which can easily lead to the escalation of the incident and cause more casualties.

[0004] Therefore, how to automatically control the spraying device on the armored vehicle to spray and extinguish oil burning materials is a technical problem that technicians need to solve at present. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a control method for a spray device for extinguishing oil fires on the outer surface of an armored vehicle.

[0006] In order to achieve the above-mentioned object, the present invention adopts the following technical scheme: a control method for a spray device for extinguishing oil fire on the outer surface of an armored vehicle, comprising the following steps: Step 1: Based on the actual fire extinguishing needs of the armored vehicle's outer surface, the vehicle's surface temperature data and environmental data are collected in real time through corresponding sensors to generate a basic data collection table; Step 2: Based on the basic data collection table, pre-process various types of data, calculate the temperature change amplitude of the outer surface of the armored vehicle, issue a warning according to the set threshold, estimate the coordinate range of the temperature abnormal area, analyze the relationship between various environmental factors in combination with the environmental parameters of the timestamp, and generate a current data analysis report; Step 3: Based on the current data analysis report, a flame propagation model is established to simulate the interaction between vehicle motion and ambient wind speed, the flame state is predicted through a regression model, and the vehicle speed is optimized through a simulated annealing algorithm to generate a recommended vehicle speed report; Step 4: Based on the recommended vehicle speed report, determine the basic flow rate of the sprayed foam, calculate the basic pressure to be applied according to the basic flow rate of the foam, and dynamically adjust the foam spray pressure according to the foam coverage area to generate a spray treatment report; Step 5: Based on the spray treatment report, follow-up monitoring is carried out, and data changes are displayed in the form of charts through regression algorithms and Tableau drawing tools to obtain a set of visual charts, and all data information and operation logs are backed up to generate an event processing record table.

[0007] Preferably, the environmental data includes wind speed, wind direction, vehicle speed, terrain, humidity and ambient temperature, wherein wind speed and wind direction are used to assess the risk of flame spread and provide a basis for determining the adjustment of the actual driving route of the armored vehicle; terrain is used to analyze the impact on the acceleration and deceleration capabilities of the armored vehicle and, together with the vehicle speed, reflects air mobility and oxygen supply; humidity and ambient temperature are used to determine the extent of the impact of the external environment on the flame spread rate.

[0008] Preferably, the specific steps of preprocessing various types of data are: Based on the data collected by various sensors, outliers are screened one by one through the box plot method, and corrections are made through interpolation or mean substitution to generate a preliminary data screening package; Based on the data, a preliminary screening package is performed, an exponentially weighted moving average is performed to reduce short-term fluctuations, and the data obtained by each sensor is synchronized in time by resampling to generate a data sorting package; Based on the data sorting package, normalization and standardization processing is performed to convert various data features into the same scale and generate a standard data package.

[0009] Preferably, the specific steps of estimating the spatial coordinate range of the temperature anomaly area are: Based on Kriging interpolation, the temperature data of the outer surface of the armored vehicle is interpolated, and the temperature values ​​of all locations in space are estimated in real time to obtain a three-dimensional temperature distribution map. The accuracy of the interpolation results is evaluated through cross-validation, and a statistical report on the temperature distribution of the outer surface of the armored vehicle is generated; Based on the statistical report of the temperature distribution on the outer surface of the armored vehicle, the temperature abnormality area in the space is identified by density clustering through the DBSCAN algorithm, the temperature abnormality area is divided into different clusters, and the spatial range of the temperature abnormality area is represented by the bounding box, convex hull and polygonal area in space to generate an abnormal temperature range table; Based on the abnormal temperature range table, the corresponding three-dimensional space coordinates are matched and periodically updated to generate an abnormal area coordinate marking table.

[0010] Preferably, the specific steps of analyzing the relationship between various environmental factors are: Model the temperature data based on the time series model, capture the time dependency of the data, and generate a temperature-time correlation table; Based on the temperature-time correlation table, the linear correlation between different variables is analyzed by Pearson correlation coefficient, a regression model is established, and regression analysis is performed on temperature and environmental factors according to historical data to generate a temperature-environment relationship analysis report; Based on the temperature-environment relationship analysis report, a multivariate time series model is constructed to quantify the relationship between the variable data, analyze the mutual influence of multiple time series data, and generate an external factor analysis report.

[0011] Preferably, the flame propagation model is established based on heat conduction, convection and radiation, wherein heat conduction is used to reflect the heat propagation from the fire source to nearby combustibles, described using Fourier's law, convection is used to reflect the process of heat transfer through air and smoke, described using the Navier-Stokes equation and the energy equation, and radiation is used to reflect the process of heat transfer through electromagnetic waves, described using the Stewart-Boltzmann law, and numerical simulation is performed using the FDS tool. Based on the simulated fire development and heat transfer process, the oxygen supply and heat flow caused by the airflow flowing around the vehicle are analyzed to improve flame propagation.

[0012] Preferably, according to flame state data under different historical experimental environments and vehicle speeds, the regression model is established and trained through machine learning to predict the flame state under different given conditions, and the vehicle speed is optimized through a simulated annealing algorithm based on the given environmental data to limit flame spread.

[0013] Preferably, the specific steps of generating the spray treatment report are: Based on the adjusted vehicle speed, the current wind speed is collected, combined with the basic formula of fluid mechanics, and the rheological model, Kelvin-Helmholtz instability model and fluid dynamics model parameters are adjusted according to the actual physical properties of the foam, and the foam property constant is determined by the density, viscosity, stability and nozzle characteristics of the foam to be sprayed; Based on the foam property constant and the current wind speed, the current spray foam basic flow rate is determined, and the required basic pressure is calculated in real time; Based on the basic pressure and in combination with the temperature variation trend collected by the sensor, the foam coverage area and coverage effect are determined, the pressure applied to the foam is dynamically adjusted, and a spray treatment report is generated.

[0014] Preferably, the specific steps of generating the visualization chart set are: Based on the data information collected in real time by various sensors, the data is sorted by timestamp within the set time window using the heap sort algorithm; Establish a mathematical model based on the regression algorithm, fit the sorted data, output the obtained fitting curve, and add it to the data set as a new column; Based on the Tableau drawing tool, the imported fitting results are matched with appropriate types of charts to display data changes and regression curves, and generate a visual chart set.

[0015] Preferably, the data information and operation log are stored via a local hard disk and a cloud, and the operation log includes an IP address, operation type, operation time, operation object, operation result, and system response.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: 1. In the present invention, the oxygen concentration that the airflow can provide and the effect on the flame spread can be obtained through the flame propagation model and FDS tool when the armored vehicle is moving, and then the speed range that is conducive to reducing the flame intensity can be matched with the current environmental data in combination with the historical training data. Further, according to the current wind speed, foam properties, basic formulas of fluid mechanics and related models, the foam spraying flow rate and the pressure to be applied required to extinguish the flame can be determined. In summary, when the flame initially appears, the oxygen supply to the flame can be reduced in a short time by reducing the speed of the vehicle, so as to prevent the flame from spreading rapidly under the combined action of oxygen and airflow. At the same time, through the real-time calculated spraying flow rate and pressure, it can be ensured that the foam can resist the wind force in a formed posture and automatically cover the entire fire area, so that the flame can be extinguished efficiently in a short time, and the flame can be prevented from continuing to burn and blocking the driver's field of vision, affecting the driving of the armored vehicle, or causing damage to some important areas of the armored vehicle, which makes the armored vehicle unable to move normally.

[0017] 2. In the present invention, the temperature distribution map of the outer surface of the armored vehicle can be constructed based on the temperature data collected by the sensor through the spatial interpolation algorithm and the clustering algorithm, so that the fire area can be accurately estimated when the abnormal temperature is detected, and the spatial coordinates of the corresponding range can be marked, so as to facilitate the subsequent accurate processing of the fire area and reduce the probability of false alarms. At the same time, various environmental data can be quantitatively associated according to the time series model, so as to facilitate the subsequent analysis of the impact of the external environment on fire extinguishing.

[0018] 3. In the present invention, various data related to the flame can be displayed in appropriate charts through the Tableau drawing tool, so that the personnel inside the armored vehicle can understand the development degree of the flame and the treatment results, and thus can understand the overall status of the armored vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the main steps of the present invention.

[0020] Figure 2 It is a schematic diagram of the steps of estimating the spatial coordinate range of the temperature anomaly area of ​​the present invention.

[0021] Figure 3 It is a schematic diagram of the steps of visualizing a graph set of the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0023] See also Figure 1 -Attached Figure 3 The present invention provides a technical solution: a control method for a spray device for extinguishing oil fire on the outer surface of an armored vehicle, comprising the following steps: Step 1: Based on the actual fire extinguishing needs of the outer surface of the armored vehicle, the vehicle surface temperature data and environmental data are collected in real time through the corresponding sensors. Various sensors distributed in different areas of the outer surface of the armored vehicle obtain different types of data parameters such as the armored vehicle surface temperature, environmental temperature, humidity and wind speed in real time, generate basic data collection tables, and then transmit them through the sensor network. The subsequent host performs centralized processing and analysis to gain an in-depth understanding of the key factors such as the operating status of the armored vehicle under different environmental conditions, the heating and heat dissipation of the vehicle surface, and the impact of external meteorological factors on the performance of the armored vehicle. This can further help optimize the safety of the vehicle and promptly warn of potential faults or abnormal conditions to ensure that the armored vehicle can always maintain the best condition in a complex and changing environment; Step 2: Based on the basic data collection table, pre-process various types of data, calculate the temperature change amplitude of the outer surface of the armored vehicle, and issue a warning according to the set threshold. The rationality of the data can be improved by pre-processing various types of acquired data, and the massive data collected by the temperature sensor can be sorted by classic sorting algorithms such as quick sort, merge sort or heap sort algorithm to effectively reduce data redundancy. In the sorting process, by selecting a suitable time window, the valid data within a specific time range can be effectively screened out, thereby avoiding the impact of a large amount of irrelevant data in the historical data on the system performance. At the same time, the time range of the data can be dynamically adjusted through the preset sliding window to ensure that only the latest data in the current period is paid attention to, avoiding the calculation of too much useless information. Then, with another preset sliding window with a shorter period, the temperature data can be compared and calculated one by one. In this process, abnormal temperature fluctuations can be identified by comparing the temperature data in continuous time periods. For example, when the amplitude of the temperature change continues to exceed a certain threshold, it may mean that a fire has occurred somewhere on the outer surface of the armored vehicle. At this time, the system will automatically trigger an alarm to remind the armored vehicle driver or the command center to pay attention to the potential fire risk and estimate the coordinate range of the abnormal temperature area. Furthermore, the temperature data of each area on the outer surface of the armored vehicle can be converted into a three-dimensional temperature distribution map through the spatial interpolation algorithm and the clustering algorithm. Among them, the spatial interpolation algorithm interpolates the surrounding area in space through the known temperature data points to obtain a continuous temperature field, thereby making up for the gaps between temperature sensors and providing a more accurate temperature distribution. The clustering algorithm can cluster the temperature data according to different areas to identify the temperature anomalies in a specific area. In summary, the temperature distribution map can not only intuitively present the temperature changes in each area, but also assist in determining the location and scope of the fire. The relationship between various environmental factors is analyzed in combination with the environmental parameters of the timestamp to generate a current data analysis report. On this basis, the timestamp can be used to associate all temperature data collected at a specific time point with other environmental data. By combining relevant models, the temperature, humidity and other data in the current environment can be comprehensively analyzed to evaluate the mutual influence of various environmental factors. For example, changes in wind speed may aggravate the spread of fire, and increased humidity may affect the burning state of the flame. Then, the integrated analysis of environmental data can more reliably assist in subsequent fire prediction. Step 3: Based on the current data analysis report, a flame propagation model is established. According to the specific area where the flame occurs and the current integrated environmental factors, a flame propagation model is established from three aspects: heat conduction, convection and radiation. Heat conduction refers to the transfer of heat from the flame source to the surroundings through direct contact. Since metals usually have high thermal conductivity, heat will be quickly transferred to other parts of the vehicle body. Convection refers to the flow of hot air rising around the fire source and being replaced by the surrounding cooler air. Therefore, the air flow inside and outside the car, the ventilation system and the temperature gradient inside the car will affect the spread speed and range of the flame. Radiation refers to the flame itself releasing a large amount of thermal radiation, which in turn increases the temperature of other areas without direct contact. In summary, by quantitatively modeling heat conduction, convection and radiation, the flame propagation path and speed can be relatively completely predicted. By quantitatively modeling these physical processes, the interaction between vehicle motion and ambient wind speed can be simulated. In addition, combined with computational fluid dynamics technology, airflow and heat The amount distribution can be used to accurately calculate the oxygen concentration and heat flow trend, and further reveal the flame spread path and possible impact area, so as to preliminarily understand the possible consequences of the flame if it is not handled in time, such as the flame continues to burn and block the driver's field of vision, or continues to burn in the key area of ​​the armored vehicle, causing the area to affect normal use due to high temperature, etc. The flame state is predicted by the regression model. Furthermore, a regression model can be trained by a large amount of historical experimental data, and the regression model can predict the trend and direction of the flame under the current environmental conditions. The vehicle speed is optimized by the simulated annealing algorithm to generate a recommended speed report. At the same time, the flame state at different speeds can be inferred by the simulated annealing algorithm to obtain the appropriate vehicle speed to reduce the oxygen concentration and the flame spread speed, so as to achieve the purpose of reducing the flame intensity, and then assist the subsequent fire extinguishing operation. The speed range also needs to consider the actual adhesion and molding effect of the foam sprayed by the foam fire extinguishing agent to prevent the foam from being affected by the external wind speed and unable to maintain its own shape normally; Step 4: Based on the recommended vehicle speed report, determine the basic flow rate of the sprayed foam. After adjusting the vehicle speed to reduce the flame intensity and spread rate, calculate the foam property constant after combining the physical properties of the foam with the characteristics of the sprinkler device using the basic fluid mechanics formula and related models. Then, based on the fact that the foam spraying speed must be greater than the actual wind speed, the foam spraying speed can be preliminarily determined based on the wind speed data actually collected, and the basic pressure to be applied can be calculated based on the basic foam flow rate. After further determining the spraying speed, the foam property constant and the spraying speed can be used to calculate the foam pressure to be applied. At the same time, the foam injection pressure is dynamically adjusted according to the foam coverage area to generate a sprinkler treatment report. Since the role of foam is not only to isolate oxygen, but also to effectively absorb heat and quickly reduce the flame temperature, in actual applications, it is necessary to make real-time adjustments based on the design redundancy principle and the changes in the on-site fire to maintain sufficient coverage area, thereby minimizing the continued spread of the fire. Step 5: Follow up monitoring based on the spray treatment report. The temperature sensor can be used to monitor the flame status in real time. The data can then be collated and analyzed through regression algorithms to reveal the laws and trends of flame development. The data changes are displayed in the form of charts through regression algorithms and Tableau drawing tools (an existing Tableau visualization data analysis software). A set of visualization charts is obtained. The collated data can be further displayed in the form of charts through Tableau visualization tools, thereby helping the personnel inside the armored vehicle to more clearly understand the dynamic changes of the flame. At the same time, when the flame shows a tendency to reignite, the operator can also It can detect fire in time and take corresponding fire-fighting measures before the fire develops further, back up all data information and operation logs, and generate event processing record tables. In addition, the data collected by sensors can not only provide support for the current fire-fighting process, but also serve as historical data for subsequent model training. Through the continuous accumulation of historical data, the relevant machine learning models and algorithms can be continuously optimized, thereby improving the overall fire-fighting efficiency and system response speed. At the same time, for situations where the flame cannot be effectively extinguished, historical records and data backup can also be used to troubleshoot and analyze problems in the future, thereby optimizing more accurate flame identification and fire-fighting plans, further reducing the risk of re-ignition and improving fire emergency response capabilities.

[0024] See also Figure 1 Environmental data include wind speed, wind direction, vehicle speed, terrain, humidity and ambient temperature. Wind speed and direction are used to assess the risk of flame spread and provide a basis for the adjustment of the actual driving route of the armored vehicle. Changes in wind speed and direction can directly affect the path and speed of flame propagation. Therefore, when a fire occurs, real-time monitoring of wind speed and direction can effectively predict the trend of flame spread. In some special cases, the driving route of the armored vehicle can be adjusted to prevent the fire from spreading to key areas, thereby ensuring the safety of important facilities and personnel. Terrain is used to analyze the impact on the acceleration and deceleration capabilities of the armored vehicle, and together with the vehicle speed, it reflects the air flow and oxygen supply. At the same time, the undulation of the terrain will also have an impact on The driving speed and speed-changing ability of the armored vehicle have an impact, which in turn can affect the flow of heat and the supply of oxygen. Since the spread speed of the flame is also closely related to the heat transfer and oxygen supply, the change of terrain may indirectly affect the spread of the fire by changing the speed of the armored vehicle to a certain extent. Humidity and ambient temperature are used to judge the influence of the external environment on the spread speed of the flame. In addition, in a high temperature and dry environment, the flame is more likely to obtain sufficient oxygen and heat and spread faster. Therefore, the actual humidity value in the air can inhibit the spread of the flame to a certain extent. In summary, when a fire occurs, it is necessary to comprehensively consider the main factors such as wind speed, wind direction, terrain, humidity and ambient temperature in order to conduct a full analysis and judgment.

[0025] See also Figure 1, the specific steps for preprocessing various types of data are: Based on the data collected by various sensors, outliers are screened one by one through the box plot method. Since the data acquired by sensors may mutate due to hardware failure, signal noise or environmental changes in actual applications, abnormal data can be automatically identified and eliminated through statistical threshold judgment or machine learning algorithm, and corrected through interpolation or mean substitution to generate a preliminary data screening package. Then, interpolation or mean substitution can be used to complete the data to fill in the missing data, maintain the integrity of the data, ensure that subsequent analysis and decision-making are not affected by missing data, and achieve the purpose of improving the reliability and credibility of the data; Based on the preliminary screening set of data, exponential weighted moving average is performed to reduce short-term fluctuations. Exponential weighted moving average can reduce short-term fluctuations and noise of data, provide more stable signals, highlight the trend of changes, and synchronize the data obtained by each sensor through resampling to generate a data sorting package. In a multi-sensor system, the sampling frequencies of different sensors may be different, which may lead to inconsistent timestamps of data and affect the synchronization and comparability of data. Therefore, resampling can be used to align signals with inconsistent sampling frequencies to ensure that the data of each sensor is at the same timestamp; Based on the data sorting package, normalization and standardization processing is performed to convert various data features into the same scale and generate a standard data packet. Through normalization and standardization processing, the data can be used for corresponding algorithms or models.

[0026] See also Figure 2 ,The specific steps for estimating the spatial coordinate range of the temperature anomaly area are: Based on Kriging interpolation, the temperature data of the outer surface of the armored vehicle is interpolated to estimate the temperature values ​​of all locations in space in real time. Kriging interpolation can be used to calculate the distance and temperature difference between the measurement points based on the known temperature data, and the semivariogram function is used for spatial interpolation to obtain the estimated temperature value of each target point and obtain a three-dimensional temperature distribution map. Then, a three-dimensional temperature distribution map can be constructed based on the estimated temperature to more intuitively understand the temperature distribution of the outer surface of the armored vehicle. The accuracy of the interpolation results is evaluated through cross-validation, and a statistical report on the temperature distribution of the outer surface of the armored vehicle is generated. Based on the statistical report of the temperature distribution on the outer surface of the armored vehicle, the DBSCAN algorithm is used to identify the temperature anomaly area in space by density clustering. The DBSCAN clustering algorithm can effectively identify multiple clusters in the data and divide the temperature anomaly area into different clusters. Since each cluster is composed of a group of spatially adjacent data points with similar characteristics, the DBSCAN algorithm can be used to find spatially abnormal temperatures from a large amount of data during temperature monitoring, and the spatial range of the temperature anomaly area is represented by the spatial bounding box, convex hull and polygonal area to generate an abnormal temperature range table. Then, the DBSCAN clustering algorithm can generate cluster information to realize the digital positioning of the temperature anomaly area, thereby providing information support for subsequent temperature monitoring and safety warning. Among them, DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is a density-based clustering algorithm. Unlike partitioning and hierarchical clustering methods, it can define a cluster as the maximum set of density-connected points, divide areas with sufficiently high density into clusters, and find clusters of arbitrary shapes in the spatial database of noise.

[0027] Based on the abnormal temperature range table, the corresponding three-dimensional space coordinates are matched and regularly refreshed to generate an abnormal area coordinate marking table. The temperature abnormality area, i.e., the range coordinates of each cluster, can be updated through real-time refreshing, thereby accurately determining the coordinates and area of ​​the fire area.

[0028] See also Figure 1 ,The specific steps to analyze the relationship between various environmental factors are: Model the temperature data based on the time series model, capture the time dependency of the data, generate a temperature-time association table, and associate time with the corresponding temperature according to the time series model; Based on the temperature-time association table, the linear correlation between different variables is analyzed by the Pearson correlation coefficient. The linear relationship between environmental factors such as humidity, wind direction and air pressure and temperature is calculated by the Pearson correlation coefficient. The correlation between environmental factors and temperature changes is quantified, and a regression model is established. Based on historical data, regression analysis is performed on temperature and environmental factors to generate a temperature-environment relationship analysis report. Then, regression analysis can be performed based on the selected environmental factors with high correlation, and the regression coefficient can be obtained by fitting the regression model. The regression coefficient can further reflect the degree of influence of each environmental factor on temperature changes. Among them, the Pearson correlation coefficient is a statistical indicator used to measure the degree of linear correlation between two variables. Its value is between -1 and 1, and it is mainly used to analyze the linear relationship between two continuous variables.

[0029] Based on the temperature-environment relationship analysis report, a multivariate time series model is constructed to quantify the relationship between each variable data. Through the vector autoregression model and dynamic causal analysis, multiple environmental factors that change over time can be analyzed, and the mutual influence and action relationship between different environmental factors can be quantified. Among them, by constructing a vector autoregression model, the dynamic changes of various environmental factors and their dependence on time series can be described, the influence of multiple time series data on each other can be analyzed, and an external factor analysis report can be generated. Then, through dynamic causal analysis, the causal relationship can be revealed from the time series data, and the direct impact of a certain environmental factor on another environmental factor can be determined. In summary, the purpose of fully revealing the complex interactions between multiple environmental factors can be achieved, and accurate quantitative results can be provided to complete the integration of diversified environmental factors.

[0030] See also Figure 1 The flame propagation model is established based on heat conduction, convection and radiation, in which heat conduction is used to reflect the heat propagation from the fire source to the nearby combustibles, and is described by Fourier's law. Convection is used to reflect the process of heat transfer through air and smoke, and is described by the Navier-Stokes equation and the energy equation. Radiation is used to reflect the process of heat transfer through electromagnetic waves, and is described by the Stewart-Boltzmann law. The flame propagation model established by heat conduction, convection and radiation is numerically simulated using the FDS tool. According to the simulated fire development and heat transfer process, the oxygen supply and heat flow caused by the airflow around the vehicle are analyzed to improve the flame propagation. The FDS tool is used to simulate the changes in the current flame propagation model caused by the changes in oxygen concentration and heat flow after the airflow acts on it, so as to further improve the accuracy of the flame propagation model.

[0031] It should be noted that the Navier-Stokes equation (NS equation for short) is a partial differential equation that describes the motion of viscous fluid and is used to simulate the weather fluid mechanics system; the FDS tool is a fire dynamics simulation tool and a model of computational fluid dynamics (CFD) that simulates the energy of fire driving fluid flow.

[0032] See also Figure 1According to the flame state data under different historical experimental environments and vehicle speeds, a regression model is established and trained through machine learning. Machine learning establishes and trains a suitable regression model based on historical experimental data to predict the flame state under different given conditions, and then predicts the current flame development trend in combination with the current integrated environmental data. The vehicle speed is optimized based on the given environmental data through the simulated annealing algorithm to limit the flame spread, and the airflow velocity after the vehicle speed is changed by the simulated annealing algorithm, so that the integrated environmental data can change accordingly. Further, the regression model can predict the flame intensity and flame spread speed under different vehicle speeds. In summary, through multiple simulations, the appropriate vehicle speed to reduce the flame intensity and spread range can be obtained to reduce the difficulty of subsequent foam fire extinguishing.

[0033] See also Figure 1 , the specific steps for generating the spray treatment report are: Based on the adjusted vehicle speed, the current wind speed is collected, combined with the basic formula of fluid mechanics, and the rheological model, Kelvin-Helmholtz instability model and fluid dynamics model parameters are adjusted according to the actual physical properties of the foam. The foam property constant is determined by the density, viscosity, stability and nozzle characteristics of the foam to be sprayed. By determining the foam property constant, the interference of the foam type and nozzle type on the pressure required to spray the foam can be reduced; among them, the Kelvin-Helmholtz instability model refers to the dynamic instability caused by the shear of the horizontal wind speed at the interface between two fluids with different densities, referred to as KH instability; Based on the foam property constant and the current wind speed, the current spray foam basic flow rate is determined. Since the foam basic flow rate must be greater than the actual wind speed, the foam basic flow rate can be preliminarily estimated by the actual wind speed, and the required basic pressure can be calculated in real time. Then, the required basic pressure can be calculated by the foam property constant and the estimated foam basic flow rate. Based on the basic pressure and the temperature change trend collected by the sensor, the foam coverage area and coverage effect are determined. The foam coverage area and coverage effect are determined by the temperature change actually detected by the sensor, and the pressure applied to the foam is dynamically adjusted to generate a spray treatment report. The applied pressure can then be adjusted according to actual needs and design redundancy to ensure that the foam can quickly extinguish the flame in a short time.

[0034] See also Figure 3 ,The specific steps for generating a visualization chart set are: Based on the data information collected in real time by various sensors, the heap sort algorithm is used to sort the data by timestamp within the set time window, and the data is sorted in chronological order for subsequent retrieval and use; Establish a mathematical model based on the regression algorithm, fit the sorted data, output the obtained fitting curve, add it to the data set as a new column, and obtain the data change curve through its output results by inputting the data into the regression model; Based on the Tableau drawing tool, the imported fitting results are matched with appropriate types of charts to display data changes and regression curves, and a set of visual charts is generated. According to the results output by the regression model, that is, the data change curve, the data changes can be intuitively reflected by appropriate visual charts through the Tableau drawing tool.

[0035] See also Figure 1 , data information and operation logs are stored on the local hard disk and the cloud. Through the local hard disk and the cloud, appropriate channels can be selected for backup records according to needs in experiments or actual emergencies, and then the data parameters related to the flame and all operation information can be used as historical data to enrich the subsequent model and algorithm training content, and can be retrieved in the future to find the root cause of the problem. The operation log includes IP address, operation type, operation time, operation object, operation result and system response. Among them, the IP address can assist in tracing the source of the operation, the operation type is to record the specific operation behavior performed by the user or administrator, the operation time is used to track the specific time when the operation occurs, and help analyze the timeliness of the operation. The operation object is specific data and steps, and the operation result is used to determine whether the operation meets expectations. The system response is return information, error code, response time, etc.

[0036] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for controlling a spray device for extinguishing oil fire on the outer surface of an armored vehicle, characterized in that: The following steps are involved: Step 1: Collect vehicle surface temperature data and environmental data in real time through corresponding sensors to generate a basic data collection table; Step 2: Based on the basic data collection table, pre-process various types of data, calculate the temperature change amplitude of the outer surface of the armored vehicle, issue a warning according to the set threshold, estimate the coordinate range of the temperature abnormal area, analyze the relationship between various environmental factors in combination with the environmental parameters of the timestamp, and generate a current data analysis report; Step 3: Based on the current data analysis report, a flame propagation model is established to simulate the interaction between vehicle motion and ambient wind speed, the flame state is predicted through a regression model, and the vehicle speed is optimized through a simulated annealing algorithm to generate a recommended vehicle speed report; Step 4: Based on the recommended vehicle speed report, determine the basic flow rate of the sprayed foam, calculate the basic pressure to be applied according to the basic flow rate of the foam, and dynamically adjust the foam spray pressure according to the foam coverage area to generate a spray treatment report; Step 5: Based on the spray treatment report, the data changes are displayed in the form of charts through regression algorithms and Tableau drawing tools to obtain a visual chart set, and all data information and operation logs are backed up to generate an event processing record table.

2. A control method for a spray device for extinguishing oil fire on the outer surface of an armored vehicle according to claim 1, characterized in that: The environmental data include wind speed, wind direction, vehicle speed, terrain, humidity and ambient temperature. The wind speed and wind direction are used to assess the risk of flame spread and provide a basis for adjusting the actual driving route of the armored vehicle. The terrain is used to analyze the impact on the acceleration and deceleration capabilities of the armored vehicle, and together with the vehicle speed, reflects the air flow and oxygen supply. Humidity and ambient temperature are used to determine the extent of the impact of the external environment on the flame spread speed.

3. The control method of a spray device for extinguishing oil fire on the outer surface of an armored vehicle according to claim 1 is characterized in that: In step 2, the specific steps of preprocessing various types of data are: Based on the data collected by various sensors, outliers are screened one by one through the box plot method, and corrections are made through interpolation or mean substitution to generate a preliminary data screening package; Based on the data, a preliminary screening package is performed, an exponentially weighted moving average is performed, and the data obtained by each sensor is synchronized in time by resampling to generate a data sorting package; Based on the data sorting package, normalization and standardization processing is performed to convert various data features into the same scale and generate a standard data package.

4. The control method of a spray device for extinguishing oil fire on the outer surface of an armored vehicle according to claim 1 is characterized in that: In step 2, the specific steps of estimating the coordinate range of the temperature anomaly area are: Based on Kriging interpolation, the temperature data of the outer surface of the armored vehicle is interpolated, and the temperature values ​​of all locations in space are estimated in real time to obtain a three-dimensional temperature distribution map. The accuracy of the interpolation results is evaluated through cross-validation, and a statistical report on the temperature distribution of the outer surface of the armored vehicle is generated; Based on the statistical report of the temperature distribution on the outer surface of the armored vehicle, the temperature abnormality area in the space is identified by density clustering through the DBSCAN algorithm, the temperature abnormality area is divided into different clusters, and the spatial range of the temperature abnormality area is represented by the bounding box, convex hull and polygonal area in space to generate an abnormal temperature range table; Based on the abnormal temperature range table, the corresponding three-dimensional space coordinates are matched and periodically updated to generate an abnormal area coordinate marking table.

5. The control method of a spray device for extinguishing oil fire on the outer surface of an armored vehicle according to claim 1 is characterized in that: In the step 2, the specific steps of analyzing the relationship between various environmental factors in combination with the environmental parameters of the timestamp are: Model the temperature data based on the time series model, capture the time dependency of the data, and generate a temperature-time correlation table; Based on the temperature-time correlation table, the linear correlation between different variables is analyzed by Pearson correlation coefficient, a regression model is established, and regression analysis is performed on temperature and environmental factors according to historical data to generate a temperature-environment relationship analysis report; Based on the temperature-environment relationship analysis report, a multivariate time series model is constructed to quantify the relationship between the variable data, analyze the mutual influence of multiple time series data, and generate an external factor analysis report.

6. The control method of a spray device for extinguishing oil fire on the outer surface of an armored vehicle according to claim 1 is characterized in that: In the step 3, the flame propagation model is established based on heat conduction, convection and radiation; Among them, the heat conduction is used to reflect the heat transfer from the fire source to the nearby combustibles, which is described by Fourier's law; the convection is used to reflect the process of heat transfer through air and smoke, which is described by the Navier-Stokes equation and the energy equation; the radiation is used to reflect the process of heat transfer through electromagnetic waves, which is described by the Stewart-Boltzmann law and numerically simulated by the FDS tool; Based on the simulated fire development and heat transfer process, the oxygen supply and heat flow caused by the airflow around the vehicle are analyzed to improve the flame propagation.

7. The control method of a spray device for extinguishing oil fire on the outer surface of an armored vehicle according to claim 1 is characterized in that: In the step three, predicting the flame state through the regression model includes: establishing and training the regression model through machine learning based on the flame state data under different historical experimental environments and vehicle speeds, predicting the flame state under different given conditions, and optimizing the vehicle speed through a simulated annealing algorithm based on the given environmental data to limit the flame spread.

8. The control method of a spray device for extinguishing oil fire on the outer surface of an armored vehicle according to claim 1 is characterized in that: In step 4, the specific steps of generating the spray treatment report are: Based on the adjusted vehicle speed, the current wind speed is collected, combined with the basic formula of fluid mechanics, and the rheological model, Kelvin-Helmholtz instability model and fluid dynamics model parameters are adjusted according to the actual physical properties of the foam, and the foam property constant is determined by the density, viscosity, stability and nozzle characteristics of the foam to be sprayed; Based on the foam property constant and the current wind speed, the current spray foam basic flow rate is determined, and the required basic pressure is calculated in real time; Based on the basic pressure and in combination with the temperature variation trend collected by the sensor, the foam coverage area and coverage effect are determined, the pressure applied to the foam is dynamically adjusted, and a spray treatment report is generated.

9. The control method of a spray device for extinguishing oil fire on the outer surface of an armored vehicle according to claim 1, characterized in that: In step 5, the specific steps of generating the visualization chart set are: Based on the data information collected in real time by various sensors, the data is sorted by timestamp within the set time window using the heap sort algorithm; Establish a mathematical model based on the regression algorithm, fit the sorted data, output the obtained fitting curve, and add it to the data set as a new column; Based on the Tableau drawing tool, the imported fitting results are matched with appropriate types of charts to display data changes and regression curves, and generate a visual chart set.

10. The control method of a spray device for extinguishing oil fire on the outer surface of an armored vehicle according to claim 1, characterized in that: In the step five, backing up all data information and operation logs includes: the data information and operation logs are stored on the local hard disk and in the cloud, and the operation logs include the IP address, operation type, operation time, operation object, operation result and system response.

Images