Control method and system of potassium ion aerosol fire extinguishing device based on big data analysis

Abstract

The invention discloses a control method and system for a potassium ion aerosol fire extinguishing device based on big data analysis, and the method comprises the steps: collecting a temperature signal and a smoke signal through a plurality of types of detection devices, carrying out the fusion processing, and carrying out the grouping analysis of the fusion signal through a data grouping method, and precisely obtaining a fire source position coordinate; and extracting temperature gradient data of a surrounding area based on the fire source position coordinates, and performing time sequence prediction by using a prediction model to determine a fire behavior change trend. If the trend exceeds a preset threshold value, the adjusting module is activated, a peak intensity index is obtained from the trend to generate a spraying intensity parameter to control the release amount of the potassium ion fire extinguishing core, a directional control instruction is generated according to the parameter to adjust the spraying angle and flow, and fire source area coverage distribution is achieved. Through dynamic adjustment fusion of a fire source positioning mechanism of signal fusion and grouping analysis and temperature gradient time sequence prediction, accurate and efficient fire extinguishing is ensured, the fire response speed is increased, fire extinguishing resource utilization is optimized, and the fire spreading risk is remarkably reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control of fire extinguishing devices, and particularly discloses a control method and system for a potassium ion gas mist fire extinguishing device based on big data analysis. BACKGROUND

[0002] Fire prevention is an important field to protect life and property safety, and its core lies in quickly and accurately detecting fire and effectively extinguishing flames to reduce casualties and economic losses. In modern complex scenarios such as cold storage and power distribution rooms, fire prevention systems need to balance efficiency, safety and environmental friendliness, and become a key pillar in the field of public safety. However, existing fire prevention methods often face limitations in complex environments, making it difficult to balance rapid response and accurate control, resulting in insufficient fire response efficiency or serious resource waste. These limitations mainly stem from the fragmentation of detection and execution, as well as the lack of dynamic adaptive control mechanisms, making it difficult to cope with the diversity and uncertainty of fire development.

[0003] The detection link of the current fire prevention system usually relies on a single type of sensor, such as temperature or smoke detectors, which are difficult to accurately locate the fire source in complex environments, especially in early fire, the temperature and smoke signals may be weak or mixed, leading to misjudgment or missed judgment. Even if the fire is detected, the existing system often adopts a uniform spraying strategy in the whole area, lacking directional control for the fire source area. This not only causes waste of fire extinguishing agent, but also may affect the safety of equipment in non-fire source areas due to excessive spraying range. For example, in a power distribution room, excessive fire extinguishing agent may deposit on precision instruments, causing equipment damage or subsequent cleaning difficulties.

[0004] The deeper technical difficulty lies in how to realize the dynamic cooperation of detection and fire extinguishing. In a fire scene, the speed and scale of fire development change rapidly, and a weak early fire may evolve into a large-scale spread within a few seconds. Existing systems usually use fixed fire extinguishing agent injection intensity, which is difficult to adjust in real time according to the change of fire, resulting in either insufficient fire extinguishing force and uncontrolled fire, or excessive injection, wasting resources and increasing environmental cleaning burden.

[0005] Therefore, how to achieve accurate positioning of the fire source and dynamic injection adjustment in complex scenarios has become a key problem in the field of fire prevention. SUMMARY

[0006] The present application provides a control method and system for a potassium ion gas mist fire extinguishing device based on big data analysis, aiming to solve at least one of the defects in the prior art.

[0007] One aspect of the present application relates to a control method for a potassium ion gas mist fire extinguishing device based on big data analysis, comprising the following steps: S100, collecting temperature signals and smoke signals by multiple types of detection devices and performing fusion processing, grouping and analyzing the fused temperature signals and smoke signals using a data grouping method to obtain fire source position coordinates, the fire source position coordinates representing spatial position information of a fire occurrence point; S200, extracting temperature gradient data of a surrounding area based on the fire source position coordinates, performing time series prediction on the temperature gradient data using a prediction model to determine a fire trend, wherein the temperature gradient data includes gradient distribution characteristics of temperature changes in the area; S300, if the fire trend exceeds a preset threshold, activating an adjustment module, obtaining a peak intensity indicator from the fire trend to obtain a spraying intensity parameter, the spraying intensity parameter being used to control the release amount of the potassium ion fire extinguishing core; S400, generating a directional control instruction for the spraying intensity parameter, adjusting the spraying angle and flow of the potassium ion gas mist fire extinguishing device through the directional control instruction to obtain fire source area coverage distribution.

[0008] Further, step S100 includes: S110, collecting temperature signals and smoke signals by infrared temperature sensors and smoke concentration sensors, and performing fusion processing on the temperature signals and smoke signals using a Kalman filter to obtain a fused signal; S120, determining whether the temperature component exceeds a preset threshold based on the fused signal, if the preset threshold is exceeded, determining a potential fire source area, dividing the fused signal into multiple signal groups through a data grouping method to obtain a spatial feature vector of each signal group; S130, classifying the signal groups using a support vector machine algorithm for the spatial feature vector, obtaining a fire candidate position set, and screening high-confidence position coordinates from the fire candidate position set; S140, integrating the high-confidence position coordinates by triangulation method to obtain fire source position coordinates, the fire source position coordinates representing spatial position information of a fire occurrence point.

[0009] Further, step S200 includes: S210, obtaining initial temperature data of a surrounding area based on the fire source position coordinates, and collecting the initial temperature data by a thermal imager to obtain a temperature distribution image; S220, determining whether the temperature difference of adjacent pixel points exceeds a preset threshold based on the temperature distribution image, if the preset threshold is exceeded, calculating the gradient direction and amplitude using a discrete gradient operator to obtain temperature gradient data, the temperature gradient data including gradient distribution characteristics of temperature changes in the area; S230, performing time series prediction on the temperature gradient data using a long short-term memory network to determine the gradient change pattern in a future time period and determine the fire trend.

[0010] Further, the step S300 comprises: S310, if the fire trend changes beyond the preset threshold, the peak intensity indicator is obtained from the fire trend change, the peak position and amplitude are determined by the peak detection algorithm, and the initial injection intensity parameter is obtained; S320, according to the initial injection intensity parameter, the proportional integral controller is used to adjust the release rate, the concentration gradient is extracted from the concentration distribution data of the potassium ion fire extinguishing core, and the adjusted injection intensity parameter is generated; S330, by the adjusted injection intensity parameter, the pulse width modulation signal generator outputs the control signal to determine the release amount distribution of the potassium ion fire extinguishing core; S340, if the release amount distribution meets the preset uniformity condition, the control signal is used to activate the adjustment module to obtain the injection intensity parameter.

[0011] Further, the step S400 comprises: S410, receiving the injection intensity parameter, if the injection intensity parameter exceeds the preset flow threshold, the peak flow indicator is obtained from the injection intensity parameter, the peak position and amplitude are determined by the peak detection algorithm, and the initial directional control instruction is obtained; S420, according to the initial directional control instruction, the proportional integral controller is used to adjust the injection angle parameter, the angle deviation is extracted from the sensor data of the potassium ion gas fire extinguishing device, and the adjusted directional control instruction is generated; S430, by the adjusted directional control instruction, the servo motor module outputs the angle control signal to determine the injection angle distribution and flow distribution parameter; S440, if the injection angle distribution meets the preset coverage uniformity condition, the pulse width modulation signal generator is activated according to the flow distribution parameter to obtain the fire source area coverage distribution.

[0012] Another aspect of the present application relates to a control system of a potassium ion gas fire extinguishing device based on big data analysis, which is used to execute the control method of the potassium ion gas fire extinguishing device based on big data analysis described above, comprising: The fire source position coordinate acquisition module is used to collect temperature signals and smoke signals by a plurality of types of detection devices and perform fusion processing, and the grouped analysis is performed on the fused temperature signals and smoke signals by using a data grouping method to obtain fire source position coordinates, which represent the spatial position information of the fire occurrence point; The fire trend change determination module is used to extract temperature gradient data of the surrounding area based on the fire source position coordinates, and the time series prediction is performed on the temperature gradient data by using a prediction model to determine the fire trend change, wherein the temperature gradient data includes the gradient distribution characteristics of the temperature change in the area; The spray intensity parameter acquisition module is used to activate the adjustment module if the fire change trend exceeds the preset threshold, obtain the peak intensity index from the fire change trend, and obtain the spray intensity parameter. The spray intensity parameter is used to control the release amount of potassium ion fire extinguishing core. The fire source area coverage distribution acquisition module is used to generate directional control commands based on the spray intensity parameters. By adjusting the spray angle and flow rate of the potassium ion aerosol fire extinguishing device through the directional control commands, the fire source area coverage distribution can be obtained.

[0013] The beneficial effects achieved by this invention are as follows: This invention discloses a control method and system for a potassium ion aerosol fire extinguishing device based on big data analysis. It addresses the challenges of uncertain fire source location and drastic fire intensity changes leading to delayed fire extinguishing response and resource waste during fires. This problem involves multiple challenges, including inaccurate signal acquisition, location deviation, inaccurate trend prediction, and incompatible spray control. The method involves collecting and fusing temperature and smoke signals using various detection devices. A data grouping method is employed to analyze the fused signals to accurately obtain the fire source location coordinates. Based on these coordinates, temperature gradient data of the surrounding area is extracted, and a predictive model is used to predict the fire intensity trend. If the trend exceeds a preset threshold, an adjustment module is activated. Peak intensity indicators are obtained from the trend to generate spray intensity parameters that control the release rate of the potassium ion extinguishing core. Directional control commands are generated based on these parameters to adjust the spray angle and flow rate, achieving comprehensive coverage of the fire source area. The control method and system of the potassium ion aerosol fire extinguishing device based on big data analysis disclosed in this invention ensures accurate and efficient fire extinguishing by integrating a fire source location mechanism based on signal fusion and group analysis with dynamic adjustment based on temperature gradient time-series prediction. This improves fire response speed, optimizes the utilization of fire extinguishing resources, and significantly reduces the risk of fire spread. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating an embodiment of a control method for a potassium ion aerosol fire extinguishing device based on big data analysis according to the present invention. Figure 2 This is a functional block diagram of an embodiment of the control system of a potassium ion aerosol fire extinguishing device based on big data analysis according to the present invention.

[0015] Explanation of icon numbers: 10. Module for obtaining the coordinates of the fire source location; 20. Module for determining the trend of fire change; 30. Module for obtaining the spray intensity parameters; 40. Module for obtaining the coverage distribution of the fire source area. Detailed Implementation

[0016] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0017] As Figure 1 The first embodiment of the present application proposes a control method of potassium ion aerosol fire extinguishing device based on big data analysis, including the following steps: Step S100, collect temperature signals and smoke signals by multiple types of detection devices and perform fusion processing, use data grouping method to analyze the grouped temperature signals and smoke signals, obtain fire source position coordinates, and the fire source position coordinates represent the spatial position information of the fire occurrence point.

[0018] By deploying complementary types of detection devices (such as temperature sensors and smoke detectors), temperature signals (reflecting heat intensity) and smoke signals (reflecting combustion product concentration) in the protected area are collected synchronously; multi-source data fusion algorithm (such as D-S evidence theory, weighted average method) is used to perform time and space alignment and feature fusion on the two types of signals, eliminate the detection blind area of single sensor (such as high temperature without smoke or smoke with low temperature scene); based on the spatial topology structure of the protected area (such as according to building partition, grid unit), the fused signals are grouped, the confidence (fire probability) and spatial distribution characteristics of the signals in each group are calculated; finally, the group with the highest confidence is selected as the fire source core area, and the three-dimensional spatial coordinates of the fire occurrence point are output through the spatial coordinate weighted calculation of the sensors in the area, which provides accurate position reference for subsequent fire analysis and directional fire extinguishing.

[0019] Step S200, extract temperature gradient data of the surrounding area based on the fire source position coordinates, use prediction model to perform time series prediction on the temperature gradient data, and determine the fire trend, wherein the temperature gradient data includes the gradient distribution characteristics of the temperature change in the area.

[0020] The fire source position coordinates output in step S100 are used as a space reference to demarcate a peripheral analysis region with a radius of 10-15 m; real-time temperature data of different space points in the region are collected by the distributed temperature sensor, the temperature difference value and the space distance ratio between adjacent points are calculated, and temperature gradient data (reflecting the strength and direction distribution characteristics of temperature change in the region) containing the radial (outward from the fire source) and tangential (perpendicular to the radial) directions are generated; the temperature gradient data are sorted into a feature sequence according to a time sequence (such as 10 seconds / interval) and input into a preset time series prediction model (such as LSTM (Long Short-Term Memory) and ARIMA (Autoregressive Integrated Moving Average Model)); the time series prediction model learns the historical gradient change rule and outputs gradient prediction values in the future 30 seconds-5 minutes, in combination with gradient direction entropy, spread speed and other indicators, to determine whether the fire is “fast spreading”, “directional spread” or “stable weakening”, and finally determine the fire trend, providing a dynamic basis for subsequent injection intensity adjustment.

[0021] In step S300, if the fire trend exceeds a preset threshold, the adjustment module is activated, the peak intensity indicator is obtained from the fire trend, and the injection intensity parameter is obtained. The injection intensity parameter is used to control the release amount of the potassium ion fire extinguishing core.

[0022] The fire trend (such as spread speed and intensity increase) output in step S200 is used as a judgment basis and compared with a system preset safety threshold (divided into two levels of “warning threshold” and “extinguishing threshold”); if the fire trend exceeds the extinguishing threshold, the adjustment module (including the pressure control unit and the flow calculation unit) is automatically activated; the core indicators such as “peak temperature gradient”, “30-second intensity increase peak value” and “maximum heat release rate of the fire source area” are extracted from the fire trend data and integrated into a peak intensity indicator (quantifying the parameter of the most intense state of the fire); through a preset “peak intensity-injection intensity” mapping algorithm (such as linear fitting and piecewise function), the peak intensity indicator is converted into a specific injection intensity parameter (including driving pressure, flow rate and injection time), and finally the release amount of the potassium ion fire extinguishing core is controlled through the parameter to ensure that the release amount is accurately matched with the intensity of the fire, which meets the fire extinguishing demand and avoids waste of the medicament.

[0023] In step S400, a directional control instruction is generated for the injection intensity parameter, the injection angle and flow of the potassium ion gas mist fire extinguishing device are adjusted through the directional control instruction, and the fire source area coverage distribution is obtained.

[0024] Based on the spray intensity parameters (including driving pressure and flow rate) output in step S300, and combined with the fire source location coordinates and spatial topology information (such as the layout of equipment in the protected area and the installation position of nozzles) from step S100, directional control commands are generated through spatial geometric algorithms and flow distribution models. The directional control commands include two types of core parameters: spray angle parameters (horizontal rotation angle and vertical elevation angle) and flow adjustment parameters (flow rate ratio of each nozzle). The directional control commands are transmitted to the actuators (angle adjustment motor and electromagnetic flow valve) of the potassium ion aerosol fire extinguishing device to adjust the spray direction of the nozzles and the agent flow rate of each nozzle in real time. The aerosol diffusion range is monitored in real time through infrared imaging or pressure feedback, ultimately forming a fire source area coverage distribution that covers the core area of ​​the fire source (coverage rate ≥95%) and the transition area (coverage rate ≥85%), ensuring that the fire extinguishing aerosol acts efficiently on the fire area and shortens the fire extinguishing time.

[0025] Furthermore, the control method for the potassium ion aerosol fire extinguishing device based on big data analysis provided in this embodiment includes step S100: Step S110: Collect temperature signals and smoke signals using an infrared temperature sensor and a smoke concentration sensor, and use a Kalman filter to fuse the temperature signals and smoke signals to obtain a fused signal.

[0026] The state update process of the Kalman filter for the temperature signal is described by the following formula: (1) In formula (1), Indicates the first The temperature signal after Kalman filtering at any given time. Indicates the first The predicted temperature value at any time, The Kalman gain representing the temperature signal. Indicates the first The original temperature signal measured by the infrared temperature sensor at any time. The control logic of formula (1) is "based on the predicted value, adaptively adjust according to the reliability of the measurement deviation (Kalman gain), balance the error between prediction and measurement, and output a more stable temperature signal".

[0027] The state update process of the Kalman filter for the smoke signal is described by the following formula: (2) In formula (2), Indicates the first The smoke concentration signal after Kalman filtering at any given time. Indicates the first Predicted smoke concentration values ​​at all times. The Kalman gain representing the smoke signal. represents the original smoke signal measured by the smoke concentration sensor at the The control logic of formula (2) is "balancing the error between prediction and measurement by Kalman gain, and adaptively adjusting the output to make the smoke concentration signal smoother and more stable".

[0028] The process of weighting and fusing the temperature signal and smoke signal after Kalman filtering is described by the following formula: (3) In formula (3), represents the fusion signal output at the represents the fusion weight coefficient of the temperature signal, represents the temperature signal after Kalman filtering at the represents the fusion weight coefficient of the smoke signal, represents the filtered smoke concentration signal at the The control logic of formula (3) is "integrating the two types of signals into a comprehensive output by adjusting the contribution proportion of temperature and smoke signals through weight", and the weight can be flexibly adjusted according to the actual scene priority of temperature / smoke attention.

[0029] By collecting temperature signals and smoke signals through infrared temperature sensors and smoke concentration sensors, the warehouse environment can be monitored in real time, for example, the sensors are deployed on the warehouse ceiling and collect data every second, the temperature signal shows that the local area rises to 50 degrees Celsius, and the smoke signal concentration reaches 0.5 milligrams per cubic meter. This multi-sensor collection method improves data accuracy, avoids false judgments by single sensors, and thus improves the effect of early fire warning.

[0030] The Kalman filter is used to fuse and process the temperature signal and the smoke signal to obtain the fusion signal. The filter fuses multi-source data through prediction and update steps, for example, the initial state of the temperature signal has large noise, the filter estimates the true temperature as 45 degrees Celsius, and the noise of the fused signal is reduced by 20%. This helps to reduce false positives caused by environmental interference and enhances signal reliability.

[0031] Step S120, according to the fusion signal, judge whether the temperature component exceeds the preset threshold, if it exceeds the preset threshold, determine the potential fire source area, divide the fusion signal into multiple signal groups by data grouping method, and obtain the spatial feature vector of each signal group.

[0032] The judgment standard of the potential fire source area is defined by the following formula: (4) In formula (4), ​​This indicates the result of the potential ignition source area detection. A value of 1 indicates that a potential ignition source area has been detected, while a value of 0 indicates that no ignition source has been detected. This indicates the detected temperature component value. This indicates the preset temperature threshold.

[0033] The spatial feature vector of a signal group is obtained by calculating the average spatial location of all signals within the group using the following formula: (5) In formula (5), Indicates the first Spatial feature vectors of a signal group Indicates the first The number of signals contained in a signal group. Indicates the first In the signal group, the first The spatial coordinate vector of a signal.

[0034] The temperature component of the fused signal is determined to be above a preset threshold. If it is above the threshold, a potential fire source area is identified. For example, if the preset threshold is 40 degrees Celsius and the temperature component in the fused signal reaches 55 degrees Celsius, an alarm is triggered, and the northwest corner of the warehouse is designated as a potential area. This threshold determination mechanism can quickly respond to potential risks, shorten response time, and improve fire extinguishing efficiency.

[0035] The fused signal is divided into multiple signal groups by a data grouping method, and the spatial feature vector of each signal group is obtained. For example, the signal is grouped according to a spatial grid, with each group containing 10 data points. The feature vector includes an average temperature of 52 degrees Celsius and a smoke peak of 0.6 milligrams per cubic meter. This grouping method is convenient for processing large-scale data and supports the accuracy of subsequent classification.

[0036] Step S130: Use the support vector machine algorithm to classify the signal group based on the spatial feature vector to obtain a set of candidate fire source locations, and filter the high-confidence location coordinates from the set of candidate fire source locations.

[0037] The mathematical representation of the set of candidate fire source locations is described by the following formula: (6) In formula (6), This represents the set of candidate locations for fire sources. Indicates the first Spatial coordinates of a location, This represents the corresponding feature vector. This indicates the category score for that position. This represents the total number of detected locations. When the classification score is greater than zero, the location is included in the fire source candidate location set.

[0038] The confidence degree of the position is comprehensively evaluated by a sigmoid function and feature intensity normalization using the following formula: (7) In formula (7), represents the position confidence score, represents the distance of the position to the classification hyperplane, represents the feature intensity value of the position, represents the maximum feature intensity. The control logic of formula (7) is “reflecting the classification reliability with the classification boundary distance, reflecting the feature significance with the feature intensity normalization, and fusing the two through the Sigmoid function to obtain the position confidence”. The higher the score, the more reliable the position.

[0039] The signal groups are classified by the support vector machine algorithm for the spatial feature vector to obtain a fire source candidate position set. For example, the support vector machine algorithm classifies 3 groups as high-risk after training, and the candidate positions include coordinates (10, 15, 2) and (12, 18, 3). This machine learning classification can extract patterns from noise, reduce human intervention, and improve the positioning accuracy to 85%.

[0040] High-confidence position coordinates are selected from the fire source candidate position set. Assuming that the confidence threshold is set to 0.8, only the position coordinate (10, 15, 2) is retained because its position confidence score is 0.9. This screening process filters low-reliability items, optimizes computing resources, and ensures the reliability of the final position.

[0041] In step S140, the high-confidence position coordinates are integrated by the triangulation method to obtain fire source position coordinates, which represent the spatial position information of the fire occurrence point.

[0042] The following formula is used to integrate multiple high-confidence position coordinates by weighted average to obtain the final position of the fire source: (8) In formula (8), represents the final determined fire source position coordinates, represents the number of high-confidence position coordinate points participating in triangulation, represents the confidence weight of the th position coordinate point, represents the spatial coordinate vector of the th high-confidence position coordinate point. The control logic of formula (8) is “letting the position coordinates with higher confidence occupy a larger proportion in the final result, integrating multiple candidate positions by weighted average, and improving the reliability of fire source positioning”.

[0043] By integrating high-confidence location coordinates using triangulation, the location coordinates of the fire source can be obtained. For example, by using data from three sensors to calculate the intersection point, the fire source is found to be at (11, 16, 2.5), which represents the spatial location information of the fire location. This method integrates multi-view data, improving positioning accuracy to the centimeter level, which helps fire brigades to intervene accurately and reduce property damage.

[0044] Furthermore, the control method for the potassium ion aerosol fire extinguishing device based on big data analysis provided in this embodiment includes step S200: Step S210: Obtain the initial temperature data of the surrounding area based on the coordinates of the fire source location, and obtain the temperature distribution image by collecting the initial temperature data through a thermal imager.

[0045] The physical process by which a thermal imager converts temperature information into an image signal is described by the following formula: (9) In formula (9), Indicates the position of the thermal imager at the pixel. The intensity of infrared radiation collected at that location. Indicates the emissivity of the target surface. This represents the Stefan Boltzmann constant. This indicates the actual temperature corresponding to that pixel. The response function of the thermal imager is represented. The control logic of formula (9) is to "first calculate the target radiation according to the thermal radiation law, and then combine the equipment response characteristics to convert the temperature into a radiation intensity signal that the thermal imager can collect".

[0046] The system obtains initial temperature data of the surrounding area based on the coordinates of the fire source location. For example, in a warehouse environment, if the coordinates of the fire source location are (11, 16, 2.5), the system collects data on the area within 5 meters around that point. The system uses a thermal imager to scan once per second to obtain a temperature distribution image. This temperature distribution image displays the temperature distribution in pixel form, such as a pixel value of 60 degrees Celsius in the center area and 30 degrees Celsius at the edge. This helps to visualize the spread of heat sources and improve the accuracy of fire monitoring.

[0047] Thermal imagers capture temperature information through infrared radiation, avoiding the limitations of traditional sensors in the presence of smoke interference, thus enabling contactless monitoring in high-shelf areas of warehouses and providing a safer early warning effect.

[0048] Step S220: Determine whether the temperature difference between adjacent pixels exceeds a preset threshold based on the temperature distribution image. If it exceeds the preset threshold, use the discrete gradient operator to calculate the gradient direction and magnitude to obtain temperature gradient data. The temperature gradient data includes the gradient distribution characteristics of temperature changes within the region. The triggering condition for gradient calculation is defined by the following formula: (10) In formula (10), represents the temperature value at the pixel point, represents the temperature value at the adjacent pixel point, represents the temperature difference between the adjacent pixel points, represents a preset temperature difference threshold, and the gradient calculation is triggered when the temperature difference exceeds the threshold.

[0049] The gradient amplitude of each point in the temperature field is calculated by the following formula: (11) In formula (11), represents the amplitude of the temperature gradient, represents the partial derivative of the temperature in the direction, represents the partial derivative of the temperature in the direction. The gradient direction is calculated by the following formula:

[0050] (12) In formula (12), represents the direction angle of the temperature gradient, represents the discrete gradient component in the direction, represents the discrete gradient component in the direction, The function is used to calculate the angle between the gradient vector and the positive direction of the axis, and determine the main direction of temperature change. The temperature distribution image can reveal the uneven distribution of heat, support subsequent analysis to reduce misjudgment. According to the temperature distribution image, whether the temperature difference between adjacent pixel points exceeds the preset threshold is judged, for example, the threshold is set to 10 degrees Celsius, if a pixel is 55 degrees Celsius and its adjacent pixel is 40 degrees Celsius in the temperature distribution image, the difference is 15 degrees Celsius which exceeds the threshold, then further processing is triggered, which verifies the heat anomaly from multiple aspects, such as combining historical fusion signals, to avoid false alarms caused by single pixel noise. This judgment mechanism scans the image grid, for example, in the northwest corner area of the warehouse, analyzes 100 pairs of pixels, finds that 20 pairs of difference exceed the standard, which indicates that the fire may spread, which helps to quickly identify the risk boundary and improve response efficiency. From the time dimension, if the difference value continuously increases in consecutive frame images, it can support the judgment of fire intensification, and optimize the allocation of fire extinguishing resources.

[0051]

[0052]

[0053] ​​​​​​​If the temperature exceeds a preset threshold, the discrete gradient operator is used to calculate the gradient direction and magnitude to obtain temperature gradient data. For example, when the Sobel operator is used to process an image, the gradient magnitude is calculated to be 8 degrees Celsius per pixel, and the direction is pointing towards the center of the heat source. This describes the gradient distribution characteristics of temperature change. For example, in a warehouse scene, the gradient data shows a steep change that extends eastward from the fire source.

[0054] The discrete gradient operator extracts edge features through difference operations; for example, regions with an amplitude greater than 5 are marked as high gradient areas. This supports fire propagation path analysis from a spatial perspective, reducing computational complexity and improving positioning accuracy. Combined with historical grouping methods, gradient data can be fused with smoke signals to form a consistent heat diffusion pattern, resulting in more reliable trend prediction.

[0055] Step S230: Based on the temperature gradient data, perform time-series prediction through a long short-term memory network to determine the gradient change pattern in the future time period and determine the trend of fire change. The following formula can be used to predict the development trend of a fire: (13) In formula (13), Indicates future time The predicted value of the fire's changing trend, This represents the gradient influence coefficient. Indicates the number of gradient sampling points. Indicates the first The weight of each sampling point Indicates the first At time 1 point The gradient rate of change, This represents the trend baseline offset. The control logic of formula (13) is "to use the weighted information of the current fire gradient as the core, and to predict the subsequent development direction of the fire by adjusting the coefficients and offsets".

[0056] Based on temperature gradient data, a long short-term memory network is used for time-series prediction to determine the gradient change pattern in the future time period and to determine the trend of fire change. For example, if the network is input with the gradient sequence of the past 10 seconds, it predicts that the amplitude will increase from 8 to 12 in the next 5 minutes. The gradient change pattern shows accelerated spread, which indicates that the fire is tending to worsen.

[0057] Long Short-Term Memory (LSTM) networks utilize gating mechanisms to handle temporal dependencies, such as forget gates to filter noise and update gates to incorporate new data. In continuous warehouse monitoring, this logical progression from the past to the future ensures a prediction accuracy of up to 80%.

[0058] Preferably, the control method for the potassium ion aerosol fire extinguishing device based on big data analysis provided in this embodiment includes step S300: Step S310: If the fire change trend exceeds the preset threshold, the peak intensity index is obtained from the fire change trend, and the peak position and amplitude are determined by the peak detection algorithm to obtain the initial spray intensity parameters.

[0059] Peak intensity index is derived using the following formula: (14) In formula (14), Indicates peak intensity index, This indicates the total length of the fire data sequence. Indicates the first The fire intensity value at each moment is used to find the local maximum value as the peak intensity by comparing adjacent data points. The control logic of formula (14) is to "first locate the local peak points in the fire sequence, and then take the maximum value among these peaks as the representative of the peak intensity".

[0060] The corresponding spray intensity is calculated based on the peak detection results using the following formula: (15) In formula (15), This represents the initial injection intensity parameter, with units of Pa·m³ / s; Indicates the system calibration coefficient. Indicates the detected peak amplitude. This indicates the distance to the peak location, in meters (m). The nozzle diameter parameter is represented in meters (m). The control logic of formula (15) is to combine the peak characteristics with the system / nozzle parameters to calibrate and correct the initial injection intensity to obtain the final injection intensity.

[0061] If the fire's trend exceeds a preset threshold, for example, if the prediction model in the warehouse shows a trend of 15 degrees Celsius per minute, the peak intensity index is obtained from the trend data. A peak detection algorithm, such as local maximum search, determines the peak location at the 8th minute of the trend curve, with an amplitude of 20 degrees Celsius. This allows the initial spray intensity parameter to be set to a medium level. This helps to accurately capture the fire's peak intensity, avoid over-responding to small fluctuations, and improve fire suppression efficiency.

[0062] Peak detection algorithms analyze trend sequences to identify the heat source diffusion points corresponding to the peak locations in the warehouse shelving area. The amplitude reflects the fire intensity. This method supports decision-making from a time-series perspective, reducing resource waste. For example, in the fire trend curve of the northwest corner of the warehouse, a peak amplitude of 18 degrees Celsius was detected at the 10th minute, supporting the adjustment of the initial parameters to a spray intensity of 5 units per second, resulting in faster fire control.

[0063] Step S320, according to the initial injection intensity parameter, the proportional-integral controller is used to adjust the release rate, and the concentration gradient is extracted from the concentration distribution data of the potassium ion fire extinguishing core to generate the adjusted injection intensity parameter.

[0064] The adjusted injection intensity parameter is obtained by the following formula: (16) In formula (16), denotes the adjusted injection intensity parameter, denotes the initial injection intensity parameter, denotes the proportional controller gain coefficient, denotes the integral controller gain coefficient, denotes the concentration error signal, denotes the integral variable. The control logic of formula (16) is to combine the proportional response of the current error with the integral compensation of the cumulative error to achieve accurate and static-free adjustment of the injection intensity.

[0065] The adjusted release rate is obtained by the following formula: (17) In formula (17), denotes the adjusted release rate, with the unit of kg / s (kilogram / second); denotes the basic release rate, with the unit of kg / s (kilogram / second); denotes the rate adjustment coefficient, denotes the potassium ion fire extinguishing core concentration, with the unit of kg / m³ (kilogram / cubic meter); denotes the time variable, with the unit of s (second). The control logic of formula (17) is to dynamically correct the basic release rate according to the speed of change of the potassium ion concentration, so that the release rate adapts to the real-time change of the concentration.

[0066] According to the initial injection intensity parameter, the proportional-integral controller is used to adjust the release rate, for example, when the parameter is 5 units per second, the controller calculates the deviation and integrates the historical error, and optimizes the rate to 6 units per second, while extracting the concentration gradient from the concentration distribution data of the potassium ion fire extinguishing core, such as the center concentration gradually changing from 80% to the edge 50%, to generate the adjusted injection intensity parameter of 7 units per second. This ensures uniform release and suppresses fire rekindling.

[0067] The proportional-integral controller quickly responds to the deviation through the proportional term and eliminates the steady-state error through the integral term. In the warehouse environment, features are extracted from the concentration gradient data, such as a gradient value of 2% per meter, which supports parameter adjustment and enhances the consistency of fire extinguishing agent distribution. The steep gradient area in the concentration distribution corresponds to the fire source periphery, and the adjusted parameter stabilizes the release rate, resulting in a reduced risk of secondary ignition.

[0068] Step S330, drive the pulse width modulation signal generator to output the control signal by adjusting the jet intensity parameter, determine the release amount distribution of the potassium ion fire extinguishing core.

[0069] The mechanism of the pulse width modulation signal generator generating the control signal according to the input parameter is described by the following formula: (18) In formula (18), represents the output control signal of the pulse width modulation signal, represents the signal amplitude, represents the sinusoidal modulation wave, represents the modulation frequency, represents the phase angle, represents the step function, represents the input intensity signal, represents the trigger threshold. The control logic of formula (18) is "based on sinusoidal wave modulation, compare the input intensity with the threshold, and conditionally output the PWM control signal", which realizes the control effect of "triggering the modulation signal only when reaching the threshold".

[0070] The release amount distribution of the potassium ion fire extinguishing core in space is described by the following formula, which presents Gaussian distribution characteristics: (19) In formula (19), represents the release amount of the potassium ion fire extinguishing core at the coordinate , represents the maximum release amount, and represent the release center coordinates, represents the release diffusion parameter. The control logic of formula (19) is "to simulate the release rule of the potassium ion fire extinguishing core from the center to the periphery, and to quantify the release amount at each position in space with Gaussian distribution".

[0071] Drive the pulse width modulation signal generator to output the control signal by adjusting the jet intensity parameter, for example, parameter 7 units per second generates a pulse signal with a duty cycle of 70%, to determine the release amount distribution of the potassium ion fire extinguishing core as 10 grams per square meter in the center area and 5 grams per square meter at the edge. This realizes precise control and optimizes fire extinguishing coverage.

[0072] The pulse width modulation signal generator controls the release valve opening degree according to the parameter modulation signal width, ensures uniform distribution in the high shelf area of the warehouse, and avoids local excess or deficiency. In an embodiment, the release amount distribution display shows a center peak of 12 grams after the signal output, which supports fire source suppression and brings an effect of shortening the overall fire extinguishing response time by 20%.

[0073] Step S340: If the release amount distribution meets the preset uniformity condition, then the adjustment module is activated according to the control signal to obtain the injection intensity parameters.

[0074] The uniformity index of release distribution is defined by the following formula: (20) In formula (20), An index indicating the uniformity of the release distribution. Indicates the total number of measurement points. Indicates the first Release amount at each location, This represents the average amount released at all locations. This indicates the preset uniformity threshold condition. The control logic of formula (20) is to "measure the uniformity of the release distribution by the percentage of the deviation between the release amount and the mean value".

[0075] The injection intensity parameter is obtained by the following formula: (twenty one) In formula (21), Indicates the injection intensity parameter. Represents the system gain coefficient. Indicates the base injection pressure. This represents the adjustment coefficient. This represents the correction factor of the adjustment module. The control logic of formula (21) is "first, the base pressure is adapted by the gain, and then the injection intensity is dynamically adjusted in combination with the correction factor to obtain the injection intensity that meets the current requirements".

[0076] If the release distribution meets the preset uniformity conditions, such as a distribution variance of less than 2 grams per square meter, the adjustment module is activated according to the control signal, such as fine-tuning the valve opening, to obtain a spray intensity parameter of 8 units per second. This logical progression from distribution verification to final adjustment ensures system robustness.

[0077] The adjustment module refines response signals and, combined with uniformity checks, prevents distribution deviations from causing fire suppression blind spots in warehouse scenarios. For example, once the conditions are met, the final parameters ensure a spray coverage rate of 95%, supporting rapid fire suppression, reducing property damage, and improving the reliability of safety warnings.

[0078] Furthermore, the control method for the potassium ion aerosol fire extinguishing device based on big data analysis provided in this embodiment includes step S400: Step S410: Receive the injection intensity parameter. If the injection intensity parameter exceeds the preset flow threshold, obtain the peak flow index from the injection intensity parameter, determine the peak position and amplitude through the peak detection algorithm, and obtain the initial directional control command.

[0079] The judgment condition of the peak detection processing is defined by the following formula: (22) In formula (22), represents the injection intensity parameter, represents the preset flow threshold value, represents the threshold judgment result, and outputs 1 when the injection intensity parameter exceeds the preset flow threshold value, and otherwise outputs 0, for judging whether the subsequent peak detection processing needs to be performed.

[0080] The position and amplitude information obtained by the peak detection are converted into the initial directional control instruction by the following formula: (23) In formula (23), represents the initial directional control instruction, represents the peak amplitude, represents the peak position angle, represents the peak position parameter, and represent the control coefficients. The control logic of formula (23) is to first perform an “amplitude-angle correlation operation” (i.e. , which embodies the coupling relationship between amplitude and position) on the peak amplitude and the peak position angle ; then weight the peak position parameter ; and finally adjust the contribution weights of the two items by the control coefficients , , and add them to obtain the initial directional control instruction (the essence is to generate the basic instruction of directional control by combining the “amplitude + position information of the peak”).

[0081] After receiving the injection intensity parameter, if it exceeds the preset flow threshold value such as 10 units per second, the peak flow index is obtained from the parameter, the peak position is determined to be the 5th second of the sequence and the amplitude is 12 units per second by the peak detection algorithm such as the sliding window method, so as to obtain the initial directional control instruction pointing to the warehouse shelf area.

[0082] In the warehouse environment, the peak detection algorithm scans the parameter sequence, identifies the peak position corresponding to the heat source concentration point, and the amplitude reflects the flow intensity. This method extracts key indicators from sequence data to support instruction generation.

[0083] In step S420, the proportional integral controller is used to adjust the injection angle parameter according to the initial directional control instruction, the angle deviation is extracted from the sensor data of the potassium ion gas mist fire extinguishing device, and the adjusted directional control instruction is generated.

[0084] The angle deviation is obtained by the following formula: (24) In formula (24), denotes the angle deviation at time , denotes the target angle corresponding to the initial directional control instruction at time , denotes the actual spray angle extracted from the sensor data at time .

[0085] The adjusted spray angle parameter is obtained by the following formula: (25) In formula (25), denotes the spray angle parameter generated by the proportional-integral controller at time , denotes the proportional gain coefficient, denotes the integral gain coefficient, denotes the input angle deviation at time , denotes the integral variable; the control logic of formula (25) is "to realize accurate and static-free adjustment of the spray angle by combining the immediate response of the current angle deviation with the compensation of the accumulated deviation".

[0086] The adjusted directional control instruction is obtained by the following formula: (26) In formula (26), denotes the adjusted directional control instruction generated at time , denotes the angle value corresponding to the initial directional control instruction, denotes the spray angle adjustment amount output by the controller at time . The control logic of formula (26) is "to dynamically update the directional control instruction by combining the real-time angle adjustment amount on the basis of the initial directional instruction".

[0087] According to the initial directional control instruction, the proportional-integral controller is used to adjust the spray angle parameter, the angle deviation such as 3 degrees deviation is extracted from the sensor data of the potassium ion gas fire extinguishing device, and the adjusted directional control instruction is set to 45 degrees inclination. For example, the proportional-integral controller processes the instruction deviation, combines the sensor data such as real-time angle reading, and optimizes the parameter to 50 degrees to ensure the accuracy of the instruction.

[0088] In step S430, the adjusted directional control instruction is used to drive the servo motor module to output an angle control signal, and the spray angle distribution and flow distribution parameters are determined; The linear relationship between the adjusted directional control command and the servo motor angular output is described by the following equation: (27) In equation (27), represents the spray angle output of the servo motor, represents the angular control gain coefficient, represents the adjusted directional control command voltage, represents the initial reference angle offset.

[0089] The flow distribution parameters at different angular positions are calculated by the following equation: (28) In equation (28), represents the flow distribution value of the th spray point, represents the system maximum flow output, represents the angular parameter of the th spray point, represents the flow distribution coefficient of the th spray point.

[0090] The overall spray angle distribution characteristics are determined by the following equation: (29) In equation (29), represents the spray angle distribution parameter, represents the number of sampling points of the angular distribution, represents the weight factor of the th sampling point, represents the angular value corresponding to the th sampling point.

[0091] By adjusting the directional control command, the servo motor module outputs an angular control signal, such as a pulse sequence control motor to rotate, determines the spray angle distribution to be 60 degrees covering the center area, and the flow distribution parameter to be 4 units per second at the edge area. For example, the servo motor responds to the command adjustment device orientation to form an angular distribution in the warehouse shelf area that gradually changes from the center outward, with the distribution parameter specifying a center flow of 6 units per second, supporting coverage planning.

[0092] Step S440, if the spray angle distribution meets the preset coverage uniformity condition, activate the pulse width modulation signal generator according to the flow distribution parameter to obtain the fire source area coverage distribution.

[0093] The condition for activating the pulse width modulation signal generator is defined by the following equation: (30) In formula (30), denotes the coverage uniformity index of the spray angle distribution, denotes the total number of spray points, denotes the angle value of the th spray point, denotes the average value of all spray angles, denotes the preset coverage uniformity threshold condition. The control logic of formula (30) is "to determine whether the uniformity condition of activating the PWM signal generator is met by the proportion of standard points in the deviation of spray angle from the average value".

[0094] The duty cycle of the pulse width modulation signal is obtained by the following formula: (31) In formula (31), denotes the duty cycle of the th pulse width modulation signal, denotes the on-time of the th signal, in seconds; denotes the pulse period, in seconds; denotes the spray medium density, in kilograms per cubic meter; denotes the maximum flow value of the spray, in cubic meters per second; denotes the efficiency coefficient of the th injector, denotes the spray flow demand, in cubic meters per second. The control logic of formula (31) is to determine the duty cycle by "deriving the on-time from the spray parameters → calculating the period proportion".

[0095] If the spray angle distribution meets the preset coverage uniformity condition, such as the variance being less than 1 degree, the pulse width modulation signal generator is activated according to the flow distribution parameters to generate a signal with a duty cycle of 60%, resulting in a fire source area coverage distribution that is densely covered in the center and gradually weakens at the edges. For example, the pulse width modulation signal generator based on the parameter modulation signal ensures that the distribution expands from the center 8 units to the periphery at the warehouse fire point, supporting the integrity of the area.

[0096] See Figure 2This invention provides a control system for a potassium ion aerosol fire extinguishing device based on big data analysis, used to execute the aforementioned control method for the potassium ion aerosol fire extinguishing device based on big data analysis. The system includes a fire source location coordinate acquisition module 10, a fire intensity trend determination module 20, a spray intensity parameter acquisition module 30, and a fire source area coverage distribution acquisition module 40. The fire source location coordinate acquisition module 10 collects temperature and smoke signals through various types of detection devices and performs fusion processing. It then uses a data grouping method to group and analyze the fused temperature and smoke signals to obtain the fire source location coordinates, which represent the spatial location information of the fire occurrence point. The fire intensity trend determination module 20... The system is used to extract temperature gradient data of the surrounding area based on the coordinates of the fire source location, and to use a prediction model to perform time-series prediction on the temperature gradient data to determine the fire change trend. The temperature gradient data includes the gradient distribution characteristics of temperature changes within the area. The spray intensity parameter acquisition module 30 is used to activate the adjustment module if the fire change trend exceeds a preset threshold, and to obtain the peak intensity index from the fire change trend to obtain the spray intensity parameter. The spray intensity parameter is used to control the release amount of the potassium ion extinguishing core. The fire source area coverage distribution acquisition module 40 is used to generate directional control commands based on the spray intensity parameters, and to adjust the spray angle and flow rate of the potassium ion aerosol extinguishing device through the directional control commands to obtain the fire source area coverage distribution.

[0097] This embodiment provides a control method and system for a potassium ion aerosol fire extinguishing device based on big data analysis. Compared with existing technologies, this method collects and fuses temperature and smoke signals using various types of detection devices. It employs a data grouping method to analyze the fused signals to accurately obtain the fire source location coordinates. Based on these coordinates, it extracts temperature gradient data from the surrounding area and uses a predictive model to predict the fire's trend over time. If the trend exceeds a preset threshold, the adjustment module is activated. Peak intensity indicators are obtained from the trend to generate spray intensity parameters that control the potassium ion extinguishing core release. Directional control commands are generated based on these parameters to adjust the spray angle and flow rate, achieving comprehensive coverage of the fire source area. This embodiment, through the fusion and grouping analysis of the fire source location mechanism and the dynamic adjustment of temperature gradient time-series prediction, ensures precise and efficient fire extinguishing, improving fire response speed, optimizing fire extinguishing resource utilization, and significantly reducing the risk of fire spread.

[0098] While the preferred embodiments of the application have been described, additional variations and modifications can be made to these embodiments by those skilled in the art once they have the benefit of the present disclosure without departing from the spirit and scope of the application. Accordingly, it is intended that the appended claims include all such variations and modifications as fall within the scope of the present application. It is apparent that those skilled in the art can modify and adapt the present application in various ways without departing from the spirit and scope of the present application. It is therefore intended that the present application encompass all such modifications and variations as fall within the scope of the claims and their equivalents.

Claims

1. A control method for a potassium ion aerosol fire extinguishing device based on big data analysis, characterized in that, Includes the following steps: S100. Temperature signals and smoke signals are collected and fused using various types of detection devices. The fused temperature signals and smoke signals are then analyzed by grouping data using a data grouping method to obtain the coordinates of the fire source location. The coordinates of the fire source location represent the spatial location information of the fire occurrence point. S200. Based on the coordinates of the fire source location, extract the temperature gradient data of the surrounding area, use a prediction model to perform time-series prediction on the temperature gradient data, and determine the fire change trend, wherein the temperature gradient data includes the gradient distribution characteristics of temperature changes within the area. S300. If the fire change trend exceeds a preset threshold, the adjustment module is activated to obtain the peak intensity index from the fire change trend and obtain the spray intensity parameter. The spray intensity parameter is used to control the release amount of the potassium ion extinguishing core. S400: Generate directional control commands based on the spray intensity parameters, and adjust the spray angle and flow rate of the potassium ion aerosol fire extinguishing device through the directional control commands to obtain the coverage distribution of the fire source area.

2. The control method for the potassium ion aerosol fire extinguishing device based on big data analysis according to claim 1, characterized in that, Step S100 includes: S110. Temperature signals and smoke signals are acquired through an infrared temperature sensor and a smoke concentration sensor. The temperature signals and smoke signals are fused using a Kalman filter to obtain a fused signal. S120. Determine whether the temperature component exceeds a preset threshold based on the fused signal. If it exceeds the preset threshold, determine the potential fire source area. Divide the fused signal into multiple signal groups using a data grouping method to obtain the spatial feature vector of each signal group. S130. The support vector machine algorithm is used to classify the signal group based on the spatial feature vector to obtain a set of candidate fire source locations, and high-confidence location coordinates are selected from the set of candidate fire source locations. S140. The high-confidence location coordinates are integrated using triangulation to obtain the fire source location coordinates, which represent the spatial location information of the fire occurrence point.

3. The control method for the potassium ion aerosol fire extinguishing device based on big data analysis according to claim 1, characterized in that, Step S200 includes: S210. Obtain initial temperature data of the surrounding area based on the coordinates of the fire source location, and obtain a temperature distribution image by acquiring the initial temperature data through a thermal imager. S220. Based on the temperature distribution image, determine whether the temperature difference between adjacent pixels exceeds a preset threshold. If it exceeds the preset threshold, use a discrete gradient operator to calculate the gradient direction and magnitude to obtain temperature gradient data. The temperature gradient data includes the gradient distribution characteristics of temperature changes within the region. S230. Based on the temperature gradient data, perform time-series prediction using a long short-term memory network to determine the gradient change pattern in the future time period and identify the fire change trend.

4. The control method for the potassium ion aerosol fire extinguishing device based on big data analysis according to claim 1, characterized in that, Step S300 includes: S310. If the fire change trend exceeds the preset threshold, the peak intensity index is obtained from the fire change trend, and the peak position and amplitude are determined by the peak detection algorithm to obtain the initial spray intensity parameters. Peak intensity index is derived using the following formula: in, Indicates peak intensity index, This indicates the total length of the fire data sequence. Indicates the first The fire intensity value at each moment is used to find the local maximum value as the peak intensity by comparing adjacent data points; The corresponding spray intensity is calculated based on the peak detection results using the following formula: in, This represents the initial injection intensity parameter. Indicates the system calibration coefficient. Indicates the detected peak amplitude. Indicates the distance to the peak position. Indicates the nozzle diameter parameter; S320. Based on the initial spray intensity parameters, the release rate is adjusted using a proportional-integral controller, and the concentration gradient is extracted from the concentration distribution data of the potassium ion extinguishing core to generate the adjusted spray intensity parameters. S330: By adjusting the spray intensity parameters, the pulse width modulation signal generator is driven to output a control signal to determine the release distribution of the potassium ion extinguishing core. S340. If the release amount distribution meets the preset uniformity condition, the adjustment module is activated according to the control signal to obtain the injection intensity parameter.

5. The control method for the potassium ion aerosol fire extinguishing device based on big data analysis according to claim 4, characterized in that, In step S320, the adjusted injection intensity parameter is obtained using the following formula: in, This indicates the adjusted injection intensity parameter. This represents the initial injection intensity parameter. This represents the gain coefficient of the proportional controller. This represents the gain coefficient of the integral controller. This indicates the concentration error signal. Represents the integral variable. Represents a time variable; The adjusted release rate is obtained using the following formula: in, This indicates the adjusted release rate. Indicates the base release rate. Indicates the rate adjustment coefficient. This indicates the concentration of potassium ion fire extinguishing core.

6. The control method for the potassium ion aerosol fire extinguishing device based on big data analysis according to claim 5, characterized in that, In step S330, the mechanism by which the pulse width modulation signal generator generates control signals based on input parameters is described by the following formula: in, This represents the output control signal for the pulse width modulation signal. Indicates the signal amplitude. Represents a sinusoidal modulated wave. Indicates the modulation frequency. Indicates the phase angle. Represents the step function. Indicates the input intensity signal. Indicates the trigger threshold; The following formula describes the distribution of potassium ion release from the fire extinguishing core in space, exhibiting a Gaussian distribution characteristic: in, This indicates that the potassium ion fire extinguishing core is located at coordinates. Release amount at the location, Indicates the maximum release amount. and Indicates the release of the center coordinates. This indicates the release diffusion parameters.

7. The control method for the potassium ion aerosol fire extinguishing device based on big data analysis according to claim 6, characterized in that, In step S340, the uniformity index of the release distribution is defined by the following formula: in, An index indicating the uniformity of the release distribution. Indicates the total number of measurement points. Indicates the first Release amount at each location, This represents the average amount released at all locations. This indicates a preset uniformity threshold condition; The injection intensity parameter is obtained using the following formula: in, Indicates the injection intensity parameter. Represents the system gain coefficient. Indicates the base injection pressure. This represents the adjustment coefficient. This represents the correction factor of the adjustment module.

8. The control method for the potassium ion aerosol fire extinguishing device based on big data analysis according to claim 1, characterized in that, Step S400 includes: S410. Receive the injection intensity parameter. If the injection intensity parameter exceeds a preset flow threshold, obtain the peak flow index from the injection intensity parameter, determine the peak position and amplitude through a peak detection algorithm, and obtain an initial directional control command. The judgment criteria for peak detection processing are defined by the following formula: in, Indicates the injection intensity parameter. This indicates the preset traffic threshold. This indicates the threshold judgment result. When the injection intensity parameter exceeds the preset flow rate threshold, it outputs 1; otherwise, it outputs 0. This is used to determine whether subsequent peak detection processing is required. The position and amplitude information obtained from peak detection is converted into initial orientation control commands using the following formula: in, This indicates the initial orientation control command. Indicates peak amplitude. Indicates the angle of the peak position. Indicates the peak position parameter. and Indicates the control coefficient; S420. According to the initial directional control command, the spray angle parameter is adjusted using a proportional-integral controller, the angle deviation is extracted from the sensor data of the potassium ion aerosol fire extinguishing device, and the adjusted directional control command is generated. S430: By adjusting the directional control command, the servo motor module is driven to output an angle control signal to determine the injection angle distribution and flow distribution parameters. S440. If the spray angle distribution meets the preset coverage uniformity condition, then the pulse width modulation signal generator is activated according to the flow distribution parameter to obtain the fire source area coverage distribution.

9. The control method for the potassium ion aerosol fire extinguishing device based on big data analysis according to claim 8, characterized in that, In step S420, the angle deviation is obtained using the following formula: in, Indicates time angular deviation, Indicates time The target angle corresponding to the initial orientation control command. Indicates time The actual injection angle extracted from sensor data; The adjusted injection angle parameters are obtained using the following formula: in, Indicates the proportional-integral controller in time The generated jet angle parameters, Represents the proportional gain coefficient. Represents the integral gain coefficient. Indicates time Input angle deviation, Represents the integral variable; The adjusted directional control command is derived using the following formula: in, Indicates time The generated adjusted directional control command, This represents the angle value corresponding to the initial orientation control command. Indicates time The amount of spray angle adjustment output by the controller.

10. A control system for a potassium ion aerosol fire extinguishing device based on big data analysis, used to execute the control method for the potassium ion aerosol fire extinguishing device based on big data analysis as described in any one of claims 1 to 9, characterized in that, include: The fire source location coordinate acquisition module (10) is used to collect temperature signals and smoke signals through various types of detection devices and perform fusion processing. The fused temperature signals and smoke signals are grouped and analyzed using a data grouping method to obtain the fire source location coordinates. The fire source location coordinates represent the spatial location information of the fire occurrence point. The fire change trend determination module (20) is used to extract temperature gradient data of the surrounding area based on the coordinates of the fire source location, use a prediction model to perform time-series prediction on the temperature gradient data, and determine the fire change trend, wherein the temperature gradient data includes the gradient distribution characteristics of temperature changes in the area. The spray intensity parameter acquisition module (30) is used to activate the adjustment module if the fire change trend exceeds a preset threshold, obtain the peak intensity index from the fire change trend, and obtain the spray intensity parameter. The spray intensity parameter is used to control the release amount of potassium ion fire extinguishing core. The fire source area coverage distribution acquisition module (40) is used to generate directional control commands for the spray intensity parameters, and adjust the spray angle and flow rate of the potassium ion aerosol fire extinguishing device through the directional control commands to obtain the fire source area coverage distribution.

Images