Passive self-sensing active fire prevention method and system based on perfluorohexone microencapsulation

Abstract

The invention discloses a passive self-sensing active fire prevention method and system based on perfluorohexone microencapsulation, and relates to the technical field of fire safety, and the method comprises the steps: configuring a multi-element sensor to collect environment data; environment data preprocessing and feature extraction are carried out; an intelligent fire recognition algorithm is constructed, and corresponding grading strategies are formulated to regulate and control microcapsule fire extinguishing operation; the release of the microcapsules is intelligently regulated and controlled, and dynamic dosage control and self-adaptive microcapsule distribution optimization are carried out. According to the method, environmental parameters such as temperature, smoke, pressure and humidity are collected in real time by configuring a multi-element sensor, and dynamic analysis and grading response of fire risks are carried out in combination with an intelligent algorithm, so that intelligent regulation and control, dynamic dose control and self-adaptive distribution optimization of microcapsule release are realized; a multi-element sensor layout with precise monitoring and comprehensive coverage; efficient data preprocessing and feature extraction are achieved, and dynamic fire extinguishing agent release control and distribution optimization are completed based on an intelligent fire recognition algorithm of machine learning.

Description

Technical Field

[0001] The invention relates to the technical field of fire safety, and in particular to a passive self-sensing active fire prevention method and system based on perfluorohexanone microencapsulation. Background Art

[0002] With the widespread application of modern power and communication systems, distribution corridors and cabinets, as key infrastructure, have become the core places for power transmission and equipment operation. However, due to the large number of cables, switchgear and complex electrical environment inside, these areas are extremely vulnerable to fire risks. Once a fire occurs, it will not only cause equipment damage, but may also cause large-scale power outages or major safety accidents. Therefore, how to carry out fire protection efficiently and intelligently has become an important research direction in the industry.

[0003] Traditional fire protection systems mostly use smoke alarms, manual fire extinguishing equipment or automatic sprinkler systems. However, these methods have limited effects in confined spaces. For example, smoke alarms can only sound an alarm after a fire occurs, and cannot provide early warning of danger; manual fire extinguishing equipment relies on manual operation and is difficult to respond quickly when unattended; although the automatic sprinkler system can provide a certain degree of fire extinguishing capability, it will cause secondary damage to the equipment due to excessive water consumption, and it is difficult to accurately extinguish fires in complex pipe corridor environments. In recent years, perfluorohexanone fire extinguishing agents have gradually attracted attention due to their high-efficiency fire extinguishing performance and environmental protection characteristics. At the same time, the rise of microencapsulation technology has made it possible to encapsulate and accurately release fire extinguishing agents. In addition, with the development of artificial intelligence and Internet of Things technologies, the collection of environmental data by multi-sensors and combined with intelligent algorithm analysis can achieve accurate prediction and graded response to fire risks, providing technical support for the realization of automated and intelligent fire protection. The sensor layout of existing technologies is often concentrated in local areas, and cannot fully cover the key risk points inside the entire pipe corridor or cabinet. In addition, the use of a single type of sensor limits the comprehensive analysis of multidimensional parameters (such as temperature, humidity, smoke, and pressure) in complex environments. Many systems only have a single function, such as alarm or fire extinguishing, and lack the ability to intelligently identify fire risks and respond in a graded manner based on multivariate data analysis. In addition, existing systems rarely have the ability to dynamically control the amount of fire extinguishing agent released or optimize the distribution of microcapsules, making it difficult to strike a balance between resource utilization efficiency and fire extinguishing effect. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is that the existing technology based on microcapsule fire extinguishing agent has the problems of low monitoring accuracy, imprecise fire extinguishing response and lack of intelligent control.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation, comprising:

[0007] Configure multiple sensors to collect environmental data;

[0008] Perform environmental data preprocessing and feature extraction;

[0009] Build an intelligent fire identification algorithm and formulate a hierarchical strategy to regulate the microcapsule fire extinguishing operation;

[0010] Intelligently regulate microcapsule release, perform dynamic dosage control and adaptive microcapsule distribution optimization.

[0011] As a preferred solution of the passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation described in the present invention, wherein: the configuration of multi-sensors includes installing multiple sensors in the distribution corridor and cabinets, including temperature, smoke, pressure and humidity sensors;

[0012] Temperature sensors are placed in areas prone to heat; smoke sensors are placed on the top of the cabinet to monitor smoke diffusion; pressure sensors are installed inside the equipment cavity to identify overpressure; humidity sensors are placed in areas prone to condensation; the position of each sensor is calibrated through spatial coordinates to ensure the coverage and accuracy of monitoring; define sensor parameters, temperature sensor data T i , smoke sensor data S i , pressure sensor data P i , humidity sensor data H i , the spatial coordinates of the sensor layout point are expressed as (x i ,y i ,z i )express.

[0013] As a preferred solution of the passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation described in the present invention, wherein: the environmental data preprocessing and feature extraction include data collection and processing, all sensors collect data once per second to form a time series, and the data is sent to the central control unit in real time through a wireless transmission module;

[0014] The sensor collects data once per second:

[0015]

[0016] Where N is the total number of sensors;

[0017] The data is denoised to eliminate short-term abnormal fluctuations, and the sliding window technology is used to extract historical data and analyze the dynamic change trend of environmental parameters;

[0018] Data denoising, using the sliding mean filtering method to remove instantaneous fluctuations:

[0019]

[0020] Among them, n is the sliding window size; the fire feature database is established, and after data processing, a multidimensional environmental feature matrix is ​​formed, which contains the dynamic changes of temperature, smoke concentration, pressure and humidity, which is used for subsequent algorithm analysis and fire identification.

[0021] As a preferred solution of the passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation described in the present invention, wherein: the construction of an intelligent fire identification algorithm includes a machine learning model using a classification algorithm to identify the fire risk level;

[0022] Train a support vector machine model to identify fire risk signals:

[0023] y=f ( F; θ)

[0024] Where y represents the fire risk level, including low risk, medium risk, and high risk; f represents the classification model; θ represents the model parameters; when using a random forest or deep learning model instead of f, a similar formula can be used, and the parameters are adjusted to adapt to nonlinear data;

[0025] Anomaly detection model, anomaly detection based on unsupervised learning clustering algorithm:

[0026] D i =||X i -C k || 2

[0027] Among them, D i Represents data point X i With cluster center C k The Euclidean distance of , sets the abnormal threshold to δ;

[0028] If D i >δ, it is considered that the input features of the abnormal occurrence include the environmental parameters at the current time point and the data change trend within a certain historical range.

[0029] As a preferred solution of the passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation of the present invention, wherein: the formulation of a graded corresponding strategy to regulate the microcapsule fire extinguishing operation includes dividing the fire risk into low-risk alarm only, medium-risk local fire extinguishing and high-risk global fire extinguishing according to the training data annotation;

[0030] Anomaly detection uses unsupervised learning algorithms to cluster sensor data and dynamically identify anomalies. The degree of deviation of data points from normal categories is used to determine whether there is a potential fire risk.

[0031] Response strategy: The system combines the classification model and the results of anomaly detection to comprehensively judge the fire level. If the fire level is low risk, the system only issues an alarm; if it is medium risk, the local fire extinguishing unit near the fire source is activated; if it is high risk, the global fire extinguishing mechanism is triggered and the power supply is cut off.

[0032] As a preferred solution of the passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation described in the present invention, wherein: the dynamic dosage control includes modeling and formulating the intelligent regulation of microcapsule release;

[0033] Fixed-point release using sensor position P s = ( x s ,y s ,z s) and the fire source location P f = ( x f ,y f ,z f) The distance calculation formula is:

[0034] d s =||P s -P f ||

[0035] Activation of microcapsules meets the following conditions:

[0036] d s ≤R trigger

[0037] Among them, R trigger Indicates the trigger range; dynamic dose control, according to the fire scale S f Adjust the fire extinguishing agent release amount Q, the formula is expressed as:

[0038] Q=k·S f

[0039] Where k is the dose adjustment factor.

[0040] As a preferred solution of the passive self-sensing active fire protection method based on perfluorohexanone microencapsulation described in the present invention, wherein: the adaptive microcapsule distribution optimization includes temperature field modeling, and the temperature distribution T (x, y, z) is calculated based on sensor data interpolation:

[0041]

[0042] where wi is the sensor weight, inversely proportional to the distance

[0043]

[0044] Layout optimization is performed based on temperature field modeling: The optimization goal is to minimize the deviation between the amount of fire extinguishing agent released and the fire extinguishing efficiency. The formula is:

[0045]

[0046] Among them, Q j represents the actual release amount of the jth microcapsule, Q opt Indicates the ideal release amount.

[0047] As a preferred embodiment of the passive self-sensing active fire protection system based on perfluorohexanone microencapsulation of the present invention, wherein:

[0048] The sensor monitoring module is responsible for arranging various types of sensors in the distribution corridor and cabinets to achieve real-time monitoring and data collection of environmental parameters; the sensors are connected to the data processing module via wired or wireless means to ensure real-time data transmission;

[0049] Environmental data processing module, which processes the data collected by sensors, including denoising, feature extraction and generation of fire feature matrix; data transmission module, which sends data to the central processing module through the wireless transmission module;

[0050] The central processing module identifies fire levels and formulates graded response strategies based on fire risk analysis and intelligent algorithms; integrates classification models and anomaly detection models to analyze sensor data in real time and output response instructions;

[0051] The microcapsule release module intelligently controls the release of perfluorohexanone microcapsules according to the instructions of the decision-making center and optimizes the distribution and dosage of the fire extinguishing agent; the microcapsule device, the microcapsule shell is made of heat-sensitive or pressure-sensitive materials, and it ruptures under high temperature or overpressure conditions to release perfluorohexanone fire extinguishing agent.

[0052] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of a passive self-sensing active fire protection method based on perfluorohexanone microencapsulation.

[0053] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a passive self-sensing active fire protection method based on perfluorohexanone microencapsulation.

[0054] Beneficial effects of the present invention: The passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation provided by the present invention collects environmental parameters such as temperature, smoke, pressure, humidity, etc. in real time by configuring a multi-sensor, and combines an intelligent algorithm to perform dynamic analysis and graded response to fire risks, thereby realizing intelligent regulation, dynamic dosage control and adaptive distribution optimization of microcapsule release, precise monitoring and comprehensive coverage of the multi-sensor layout; realizing efficient data preprocessing and feature extraction, and completing dynamic fire extinguishing agent release control and distribution optimization through an intelligent fire recognition algorithm based on machine learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0056] Figure 1 An overall flow chart of a passive self-sensing active fire protection method based on perfluorohexanone microencapsulation provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0058] Example 1, reference Figure 1 , which is an embodiment of the present invention, provides a passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation, comprising:

[0059] S1: Configure multiple sensors to collect environmental data.

[0060] The configuration of multi-sensors includes installing multiple sensors in the distribution corridor and cabinets, including temperature, smoke, pressure and humidity sensors;

[0061] Temperature sensors, smoke sensors, pressure sensors and humidity sensors are used in distribution pipe corridors and cabinets. These sensors can monitor different environmental parameters in real time. The specific functions are as follows:

[0062] Temperature sensor: monitors temperature changes in the area and is used to detect high temperature hot spots caused by equipment overload or short circuit at an early stage. Layout location: installed in areas prone to heat such as cable connection points and switchgear to ensure sensitivity to local temperature anomalies.

[0063] Smoke sensor: monitors the changes in smoke concentration inside the cabinet and promptly detects the spread of smoke generated in the early stages of a fire. Layout: Placed on the top of the cabinet or in areas with concentrated air flow to ensure highly sensitive monitoring of smoke spread.

[0064] Pressure sensor: monitors the pressure inside the equipment cavity and identifies overpressure caused by fire or abnormal airflow. Location: installed inside the sealed cavity of the equipment, and can reflect the pressure changes that may be caused by fire.

[0065] Humidity sensor: monitors changes in ambient humidity and identifies potential short circuit hazards caused by moisture or condensation. Location: Placed in poorly ventilated areas or bottom areas of cabinets prone to water accumulation to ensure monitoring of sudden humidity increases.

[0066] The position of each sensor is calibrated by spatial coordinates to ensure the coverage and accuracy of monitoring; define sensor parameters, temperature sensor data T i , smoke sensor data S i , pressure sensor data P i , humidity sensor data H i , the spatial coordinates of the sensor layout point are expressed as (x i ,y i ,z i )express.

[0067] It should be noted that three-dimensional spatial modeling: In order to ensure that the spatial position corresponding to the data of each sensor is highly accurate, the system uses spatial coordinate calibration technology to establish a three-dimensional model of the distribution corridor and the inside of the cabinet. Each sensor is calibrated in the model, and the coordinates are represented by (x, y, z). For example, the temperature sensor is located near the cable connector, and the coordinates are calibrated as (x1, y1, z1). Sensor spacing design, based on the monitoring range and sensitivity, determines the optimal spacing between sensors. For example, the temperature sensor spacing is designed to be 50 cm, and the smoke sensor spacing is 80 cm to maximize monitoring coverage. Calibration of spatial calibration, the system calibrates and calibrates the actual installation position of the sensor through laser scanning or structured light measurement technology. Ensure that the installation position of the sensor is consistent with the three-dimensional model, and the error is controlled within ±2 cm.

[0068] S2: Perform environmental data preprocessing and feature extraction.

[0069] Environmental data preprocessing and feature extraction include data collection and processing. All sensors collect data once a second to form a time series. The data is sent to the central control unit in real time through the wireless transmission module. The sensor collects data once a second:

[0070]

[0071] Where N is the total number of sensors;

[0072] The data is denoised to eliminate short-term abnormal fluctuations, and the sliding window technology is used to extract historical data and analyze the dynamic change trends of environmental parameters.

[0073] Capture the dynamic change trend of environmental parameters and use sliding window technology to extract historical data features. Sliding window is a time series processing method that intercepts data in a continuous time period by setting a fixed-size time window to analyze parameter changes in the short term. This method can capture the fluctuation pattern of different parameters and provide multi-dimensional dynamic information data denoising for subsequent fire risk assessment. The sliding mean filtering method is used to remove instantaneous fluctuations:

[0074]

[0075] Among them, n is the sliding window size; the fire feature database is established, and after data processing, a multidimensional environmental feature matrix is ​​formed, which contains the dynamic changes of temperature, smoke concentration, pressure and humidity, which is used for subsequent algorithm analysis and fire identification.

[0076] It should be noted that the processed sensor data is integrated to form a fire feature database, including the dynamic change information of temperature, smoke concentration, pressure and humidity at each time point. By organizing these data into a multi-dimensional environmental feature matrix, the changing trend of each parameter over time can be more intuitively represented, providing input for the intelligent fire identification algorithm. For example, a typical feature matrix form: rows represent the sampled data at each time point; columns represent the parameter values ​​of temperature, smoke, pressure and humidity; data units represent the sampling results of a certain parameter at a specific time point. The establishment of this feature matrix provides a high-quality data foundation for the dynamic analysis of fire risks and significantly improves the recognition accuracy of subsequent algorithms.

[0077] S3: Build an intelligent fire identification algorithm and formulate a hierarchical strategy to regulate the microcapsule fire extinguishing operation.

[0078] The construction of the intelligent fire identification algorithm includes the machine learning model using a classification algorithm to identify the fire risk level.

[0079] Train a support vector machine model to identify fire risk signals:

[0080] y=f( F; θ)

[0081] Among them, y represents the fire risk level, including low risk, medium risk, and high risk; f represents the classification model; θ represents the model parameters; when using a random forest or deep learning model instead of f, it can be expressed by a similar formula, and the parameters are adjusted to adapt to nonlinear data.

[0082] The system uses support vector machine as the core classification algorithm, and learns the relationship between fire risk characteristics and environmental parameters through training on historical fire data. The input features include time series data of temperature, smoke concentration, pressure and humidity, and the dynamic feature matrix generated by processing these data. The model outputs the fire risk level, which is divided into:

[0083] Low risk: The environmental anomaly is minor, no fire extinguishing is required, only an alarm is required. Medium risk: The environmental anomaly is obvious, there is a potential fire source, and local fire extinguishing is required. High risk: The environmental parameters are beyond the safe range, the fire has entered the rapid expansion stage, and comprehensive fire extinguishing and linkage with other safety measures are required.

[0084] When using random forests or deep learning models (such as multi-layer perceptrons or long short-term memory networks LSTM) instead of SVM, the parameters are adjusted to adapt to complex nonlinear features. Random forests improve the accuracy of judgment by feature splitting, and deep learning models capture the dynamic changes of fire occurrence through time series features.

[0085] Anomaly detection model, anomaly detection based on unsupervised learning clustering algorithm:

[0086] D i =||X i -C k || 2

[0087] Among them, D i Represents data point X i With cluster center C k The Euclidean distance of , sets the abnormal threshold to δ;

[0088] If D i >δ, it is considered that the input features of the abnormal occurrence include the environmental parameters at the current time point and the data change trend within a certain historical range.

[0089] The formulation of a hierarchical corresponding strategy to regulate the microcapsule fire extinguishing operation includes classifying the fire risk into low risk alarm only, medium risk local fire extinguishing and high risk global fire extinguishing according to the training data annotation.

[0090] Anomaly detection uses unsupervised learning algorithms to cluster sensor data and dynamically identify anomalies; the degree of deviation of data points from normal categories is used to determine whether there is a potential fire risk.

[0091] Formulate a response strategy, and the system combines the classification model and the results of anomaly detection to comprehensively judge the fire level. If the fire level is low risk, the system only issues an alarm; if it is medium risk, it activates the local fire extinguishing unit near the fire source; if it is high risk, it triggers the global fire extinguishing mechanism and cuts off the power supply.

[0092] It should be noted that the unsupervised learning system detects anomalies through unsupervised learning methods without manual labeling. Environmental data is divided into normal and abnormal categories, and the distance between the data points of each category and the cluster center is calculated. If the Euclidean distance of the data point exceeds the set threshold, it is judged as abnormal. By dynamically adjusting the threshold, it adapts to different environmental conditions and enhances robustness. Abnormal data is marked as potential risk and further combined with the classification model to determine the fire level.

[0093] S4: Intelligently regulate microcapsule release, perform dynamic dosage control and adaptive microcapsule distribution optimization.

[0094] The dynamic dosage control includes modeling and formulating the intelligent regulation of microcapsule release.

[0095] Fixed-point release using sensor position P s =(x s ,y s ,z s ) and the fire source location P f =(x f ,y f ,z f ) distance calculation formula:

[0096] d s =||P s -P f ||

[0097] Activation of microcapsules meets the following conditions:

[0098] d s ≤R trigger

[0099] Among them, R trigger Indicates the trigger range; dynamic dose control, according to the fire scale S f Adjust the fire extinguishing agent release amount Q, the formula is expressed as:

[0100] Q=k·S f

[0101] Where k is the dose adjustment factor.

[0102] The adaptive microcapsule distribution optimization includes: performing temperature field modeling, temperature distribution T ( x,y,z) Interpolation calculation based on sensor data:

[0103]

[0104] where w i is the sensor weight, inversely proportional to the distance

[0105]

[0106] Layout optimization is performed based on temperature field modeling: The optimization goal is to minimize the deviation between the amount of fire extinguishing agent released and the fire extinguishing efficiency. The formula is:

[0107]

[0108] Among them, Q j represents the actual release amount of the jth microcapsule, Q opt Indicates the ideal release amount.

[0109] Example 2 is an embodiment of the present invention, which provides a passive self-sensing active fire protection method based on perfluorohexanone microencapsulation. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0110] First, test preparation: various types of sensors are arranged in the environment of the simulated power distribution corridor, including temperature sensors, smoke sensors, pressure sensors and humidity sensors, covering the key risk points of the entire experimental area. The specific layout is as follows:

[0111] Temperature sensors are placed in high heat prone areas such as cable connection points and around switchgear.

[0112] Smoke sensors are placed on the top of the tunnel to detect smoke diffusion.

[0113] The pressure sensor is installed inside the closed equipment cavity to monitor the pressure changes caused by the fire.

[0114] Humidity sensors are placed in low ventilation areas to detect sudden increases in humidity or condensation.

[0115] The spatial position of each sensor is calibrated by coordinates to ensure full coverage of the monitoring range. The sensor collects data at a frequency of 1Hz and transmits it to the central control unit via a wireless network.

[0116] Implementation process: The sensor records environmental parameters (temperature, smoke concentration, pressure, humidity) once per second. To ensure data quality, the sliding mean filter method is used to denoise the data and eliminate short-term fluctuations.

[0117] Integrate multivariate data into a dynamic feature matrix for fire risk analysis. Build a support vector machine (SVM) classification model and train the fire recognition algorithm in combination with the feature matrix. The model divides fire risks into three levels: low, medium, and high based on the labeled data input before the experiment. At the same time, use the unsupervised learning K-means clustering algorithm to detect anomalies in real-time data and dynamically identify potential fire risks.

[0118] According to the fire level output by the algorithm, the system performs the corresponding response:

[0119] Low risk: Sound the alarm.

[0120] Medium risk: Activate microcapsule fire extinguishing units near the fire source.

[0121] High risk: trigger global microcapsule release, cut off power supply, and trigger the alarm system.

[0122] Based on the location of the fire source, the spatial distance between the microcapsule and the fire source is calculated, and the release of the fire extinguishing agent within the trigger range is activated.

[0123] Dynamically control the release dose and adjust the number of microcapsules according to the scale of the fire to ensure resource utilization efficiency and fire extinguishing effect.

[0124] Table 1 Test results record

[0125]

[0126]

[0127] It can be clearly seen from Table 1 that the passive self-sensing active fire protection system based on perfluorohexanone microencapsulation has no false triggering in a normal environment, indicating that the monitoring and algorithm have good stability. In the simulation of low-heat hazards and initial smoke accumulation scenarios, the system can quickly identify the risks in the fire and trigger local fire extinguishing operations in time (the reaction time is 1.5 seconds and 2 seconds respectively), and the released fire extinguishing dose is controlled within the minimum range.

[0128] For the expansion of enclosed fire and full fire scenarios, the system can judge as high risk with high accuracy and quickly release fire extinguishing agent (reaction time is 1 second and 0.8 seconds respectively), reflecting the rapid response capability. In the humidity surge scenario, the system did not misjudge it as high risk due to non-fire factors, and only released an appropriate amount of fire extinguishing agent (10g), further proving the robustness of the algorithm.

[0129] Embodiment 3 is an embodiment of the present invention, which provides a passive self-sensing active fire protection system based on perfluorohexanone microencapsulation, comprising:

[0130] The sensor monitoring module is responsible for arranging various types of sensors in the distribution corridors and cabinets to achieve real-time monitoring and data collection of environmental parameters; the sensors are connected to the data processing module via wired or wireless means to ensure real-time data transmission.

[0131] Environmental data processing module, which processes the data collected by the sensor, including denoising, feature extraction and generation of fire feature matrix; data transmission module, which sends the data to the central processing module through the wireless transmission module.

[0132] The central processing module identifies fire levels and formulates graded response strategies based on fire risk analysis and intelligent algorithms; integrates classification models and anomaly detection models to analyze sensor data in real time and output response instructions.

[0133] The microcapsule release module intelligently controls the release of perfluorohexanone microcapsules according to the instructions of the decision-making center and optimizes the distribution and dosage of the fire extinguishing agent; the microcapsule device, the microcapsule shell is made of heat-sensitive or pressure-sensitive materials, and it ruptures under high temperature or overpressure conditions to release perfluorohexanone fire extinguishing agent.

[0134] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0135] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0136] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0137] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0138] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation, characterized in that: include: Configure multiple sensors to collect environmental data; Perform environmental data preprocessing and feature extraction; Construct an intelligent fire identification algorithm and formulate a hierarchical corresponding strategy to regulate the microcapsule fire extinguishing operation; Intelligently regulate microcapsule release, perform dynamic dosage control and adaptive microcapsule distribution optimization.

2. The passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation as claimed in claim 1, characterized in that: The configuration of multi-sensors includes installing multiple sensors in the distribution corridor and cabinets, including temperature, smoke, pressure and humidity sensors; Temperature sensors are placed in areas prone to heat; smoke sensors are placed on the top of the cabinet to monitor smoke diffusion; pressure sensors are installed inside the equipment cavity to identify overpressure; humidity sensors are placed in areas prone to condensation; the position of each sensor is calibrated through spatial coordinates to ensure the coverage and accuracy of monitoring; define sensor parameters, temperature sensor data T i , smoke sensor data S i , pressure sensor data P i , humidity sensor data H i , the spatial coordinates of the sensor layout point are expressed as (x i ,y i ,z i express.

3. The passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation as claimed in claim 2, characterized in that: The environmental data preprocessing and feature extraction includes data collection and processing, where all sensors collect data once per second to form a time series, and the data is sent to the central control unit in real time through a wireless transmission module; The sensor collects data once per second: Where N is the total number of sensors; The data is denoised to eliminate short-term abnormal fluctuations, and the sliding window technology is used to extract historical data and analyze the dynamic change trend of environmental parameters; Data denoising, using the sliding mean filtering method to remove instantaneous fluctuations: Among them, n is the sliding window size; the fire feature database is established, and after data processing, a multidimensional environmental feature matrix is ​​formed, which contains the dynamic changes of temperature, smoke concentration, pressure and humidity, which is used for subsequent algorithm analysis and fire identification.

4. The passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation as claimed in claim 3, characterized in that: The constructing of the intelligent fire identification algorithm includes the machine learning model using a classification algorithm to identify the fire risk level; Train a support vector machine model to identify fire risk signals: y = f(F; θ Where y represents the fire risk level, including low risk, medium risk, and high risk; f represents the classification model; θ represents the model parameters; when using a random forest or deep learning model instead of f, a similar formula can be used, and the parameters are adjusted to adapt to nonlinear data; Anomaly detection model, anomaly detection based on unsupervised learning clustering algorithm: D i =||X i -C k || 2 Among them, D i Represents data point X i With cluster center C k The Euclidean distance of , sets the abnormal threshold to δ; If D i >δ, it is considered that the input features of the abnormal occurrence include the environmental parameters at the current time point and the data change trend within a certain historical range.

5. The passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation as claimed in claim 4, characterized in that: The formulation of a hierarchical corresponding strategy to regulate the microcapsule fire extinguishing operation includes classifying the fire risk into low risk alarm only, medium risk local fire extinguishing and high risk global fire extinguishing according to the training data annotation; Anomaly detection, using unsupervised learning algorithms to cluster sensor data and dynamically identify anomalies; The degree of deviation of a data point from the normal category is used to determine whether there is a potential fire risk; Response strategy, the system combines the classification model and the results of anomaly detection to comprehensively judge the fire level. If the fire level is low risk, the system only issues an alarm; If the risk is medium, local fire extinguishing units near the fire source are activated; if the risk is high, the global fire extinguishing mechanism is triggered and the power supply is cut off.

6. The passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation as claimed in claim 5, characterized in that: The dynamic dosage control includes modeling and formulating the release of intelligently regulated microcapsules; Fixed-point release using sensor position P s =(x s ,y s ,z s ) and the fire source location P f =(x f ,y f ,z f ) distance calculation formula: d s =||P s -P f || Activation of microcapsules meets the following conditions: d s ≤R trigger Among them, R trigger Indicates the trigger range; dynamic dose control, according to the fire scale S f Adjust the fire extinguishing agent release amount Q, the formula is expressed as: Q=k·S f Where k is the dose adjustment factor.

7. The passive self-sensing active fire prevention method based on perfluorohexanone microencapsulation as claimed in claim 6, characterized in that: The adaptive microcapsule distribution optimization includes temperature field modeling, and the temperature distribution T(x, y, z) is calculated based on sensor data interpolation: where w i is the sensor weight, inversely proportional to the distance Layout optimization is performed based on temperature field modeling: The optimization goal is to minimize the deviation between the amount of fire extinguishing agent released and the fire extinguishing efficiency. The formula is: Among them, Q j represents the actual release amount of the jth microcapsule, Q opt Indicates the ideal release amount.

8. A system using the passive self-sensing active fire protection method based on perfluorohexanone microencapsulation as claimed in any one of claims 1 to 7, characterized in that: The sensor monitoring module is responsible for arranging various types of sensors in the distribution corridor and cabinets to achieve real-time monitoring and data collection of environmental parameters; the sensors are connected to the data processing module via wired or wireless means to ensure real-time data transmission; Environmental data processing module, which processes the data collected by sensors, including denoising, feature extraction and generation of fire feature matrix; Data transmission module, data is sent to the central processing module through the wireless transmission module; The central processing module identifies fire levels and formulates graded response strategies based on fire risk analysis and intelligent algorithms; integrates classification models and anomaly detection models to analyze sensor data in real time and output response instructions; The microcapsule release module intelligently controls the release of perfluorohexanone microcapsules according to the instructions of the decision-making center and optimizes the distribution and dosage of the fire extinguishing agent; the microcapsule device, the microcapsule shell is made of heat-sensitive or pressure-sensitive materials, and it ruptures under high temperature or overpressure conditions to release perfluorohexanone fire extinguishing agent.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the passive self-induced active fire protection method based on perfluorohexanone microencapsulation according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the passive self-induced active fire protection method based on perfluorohexanone microencapsulation according to any one of claims 1 to 7 are implemented.

Images