Smoke Monitoring Method and Related Equipment Based on MEMS Multi-Channel Intelligent Gas Sensors

Abstract

The present invention relates to a smoke monitoring method based on a MEMS multi-channel intelligent gas sensor, comprising the following steps: performing real-time smoke monitoring on a target environmental area through a plurality of the MEMS multi-channel intelligent gas sensors to obtain smoke information; extracting smoke characteristics from the smoke information through wavelet transform technology to obtain smoke dynamic characteristics; performing density statistical calculation on the smoke dynamic characteristics to obtain a smoke density distribution area; dynamically drawing the smoke density distribution area based on the smoke dynamic characteristics to generate a dynamic heat map of smoke concentration distribution; predicting a risk trend based on the heat map of smoke concentration distribution to obtain a risk prediction result; and if the risk prediction result exceeds a preset risk range, giving an alarm, thus solving the technical problem that most traditional methods rely on simple threshold judgment and cannot effectively identify the dynamic changes of smoke, resulting in inaccurate monitoring results and difficulty in predicting the development trend of a fire.

Description

Technical Field

[0001] The present invention relates to the technical field of smoke monitoring, and particularly to a smoke monitoring method and related equipment based on a MEMS multi-channel intelligent gas sensor. Background Art

[0002] Traditional smoke monitoring systems usually use a single sensor for monitoring. This method has certain limitations, such as limited monitoring range, low accuracy, weak anti-interference ability, etc. In addition, most of the data processing in traditional systems relies on simple threshold judgments and cannot effectively identify the dynamic changes of smoke, which results in inaccurate monitoring results and difficulty in predicting the development trend of fires. Therefore, in large buildings, public places or industrial environments, more efficient and reliable smoke monitoring methods are needed to improve the accuracy and timeliness of fire warnings. Summary of the Invention

[0003] The main object of the present invention is to provide a smoke monitoring method and related equipment based on a MEMS multi-channel intelligent gas sensor, which solves the technical problem that traditional methods mostly rely on simple threshold judgments and cannot effectively identify the dynamic changes of smoke, resulting in inaccurate monitoring results and difficulty in predicting the development trend of fires.

[0004] To achieve the above object, the present invention provides a smoke monitoring method based on a MEMS multi-channel intelligent gas sensor, including the following steps:

[0005] Real-time smoke monitoring of a target environmental area is performed by a plurality of the MEMS multi-channel intelligent gas sensors to obtain smoke information;

[0006] Wavelet transform technology is used to extract smoke characteristics from the smoke information to obtain smoke dynamic characteristics; wherein, the smoke characteristics include smoke movement characteristics and smoke morphology characteristics;

[0007] Density statistical calculation is performed on the smoke dynamic characteristics to obtain a smoke density distribution area;

[0008] Based on the smoke dynamic characteristics, the smoke density distribution area is dynamically drawn to generate a dynamic smoke concentration distribution heat map;

[0009] Risk trend prediction is performed based on the smoke concentration distribution heat map to obtain a risk prediction result;

[0010] If the risk prediction result exceeds a preset risk range, a warning is issued.

[0011] Further, the real-time monitoring of smoke in the target environmental area by a plurality of the MEMS multi-channel intelligent gas sensors to obtain smoke information includes:

[0012] Multiple of the MEMS multi-channel intelligent gas sensors are used to monitor the smoke in the target environmental area in real time, and preliminary smoke information is obtained; wherein, there are multiple pieces of the preliminary smoke information;

[0013] Noise filtering is performed on the preliminary smoke information to obtain denoised smoke information;

[0014] Pattern recognition is performed on the denoised smoke information to obtain a smoke pattern recognition result; wherein, the smoke pattern recognition result includes the type of smoke source, the level of smoke concentration, and the smoke diffusion speed;

[0015] The pattern recognition results are fused to obtain smoke information.

[0016] Furthermore, smoke feature extraction is performed on the smoke information through wavelet transform technology to obtain smoke dynamic features, including:

[0017] The frequency components of the smoke information are decomposed through a preset wavelet transform technology to obtain smoke decomposition features;

[0018] Motion feature extraction is performed on the smoke decomposition features to obtain smoke motion features;

[0019] Gradient calculation is performed on the smoke motion features to obtain smoke gradient features; wherein, the smoke gradient features include smoke flow speed features and smoke diffusion direction features;

[0020] Trajectory analysis is performed on the smoke motion features to obtain a smoke motion trajectory;

[0021] Based on the smoke motion trajectory, diffusion feature recognition and prediction of the smoke are performed to obtain smoke diffusion characteristics;

[0022] Morphological feature extraction is performed on the smoke decomposition features to obtain smoke boundary shape features and smoke contour features;

[0023] Time-frequency localization analysis is performed based on the smoke boundary shape features, smoke contour features, and smoke diffusion characteristics to obtain smoke morphological change features;

[0024] The smoke motion features and the smoke morphological change features are fused to obtain smoke dynamic features; wherein, the smoke dynamic features include smoke motion speed changes, smoke direction change frequencies, and smoke instantaneous accelerations.

[0025] Furthermore, density statistical calculation is performed on the smoke dynamic features to obtain a smoke density distribution area, including:

[0026] Normalization processing is performed on the smoke dynamic features to obtain normalized smoke data;

[0027] Spatially rasterize the normalized smoke data based on the target environmental area to obtain rasterized smoke data;

[0028] Perform time window segmentation on the rasterized smoke data to obtain segmented time window smoke data;

[0029] Perform Gaussian kernel density calculation on the segmented time window smoke data to obtain the spatial density distribution of smoke with time variation;

[0030] Perform smoothing processing on the spatial density distribution of smoke with time variation to obtain smoothed density data;

[0031] Perform density peak detection on the smoothed density data to obtain a density peak area; wherein, the density peak area includes a density peak value and a density distribution area corresponding to the density peak value;

[0032] Through a preset density clustering algorithm, perform density similarity clustering analysis on the density distribution area based on the density peak value to obtain a similar smoke density distribution area;

[0033] Obtain the spatial distribution area of the target environmental area, and perform mapping processing on the similar smoke density distribution area based on the spatial distribution area to obtain a smoke density distribution area.

[0034] Further, the dynamically drawing the smoke density distribution area based on the smoke dynamic characteristics to generate a dynamic smoke concentration distribution heat map includes:

[0035] Identify the sparse areas within the smoke density distribution area, and use a preset Kriging interpolation method to perform spatial interpolation calculation on the sparse areas to obtain a complemented smoke density area;

[0036] Obtain the unobserved areas of the monitoring area, and based on the complemented smoke density area, perform smoke density prediction on the unobserved areas through an inverse distance weighting algorithm to obtain the smoke density of the unobserved areas;

[0037] Merge the unobserved areas corresponding to the smoke density of the unobserved areas with the complemented smoke density area to obtain a merged fog density area;

[0038] Perform color mapping on the merged fog density area with different colors to generate a static smoke concentration heat map;

[0039] Generate a heat map for each time period by performing heat map generation on the static smoke concentration heat map through color gradient mapping technology;

[0040] Based on the animation generation algorithm, dynamically draw the smoke concentration heat map for each time period based on the dynamic characteristics of the smoke, and generate a dynamic smoke concentration distribution heat map, or;

[0041] Through frame animation technology, perform time series combination and dynamic drawing on the smoke concentration heat map for each time period based on the dynamic characteristics of the smoke, and generate a dynamic smoke concentration distribution heat map that changes over time.

[0042] Further, predicting the risk trend based on the smoke concentration distribution heat map to obtain a risk prediction result, including:

[0043] Extract the characteristics of the smoke concentration change in the smoke concentration distribution heat map to obtain the dynamic change path of the high-concentration area;

[0044] Through the density gradient algorithm, analyze the smoke diffusion direction based on the dynamic change path to obtain the smoke diffusion trend;

[0045] Obtain the risk growth mode based on the smoke diffusion trend; wherein, the risk growth mode includes a rapid diffusion mode, a stable diffusion mode, and a local aggregation mode;

[0046] Input the risk growth mode into a preset risk prediction algorithm for risk calculation to obtain the risk growth rate;

[0047] Classify the risk areas based on the risk growth rate to obtain risk areas with different risk levels;

[0048] Use different identifiers to label the risk areas with different risk levels to obtain a risk annotation heat map;

[0049] Predict the risk trend based on the risk annotation heat map to obtain the risk prediction result.

[0050] Further, through the density gradient algorithm, analyze the smoke diffusion direction based on the dynamic change path to obtain the smoke diffusion trend, including:

[0051] Collect the smoke concentration of the dynamic change path through multi-point sampling technology to obtain the spatial distribution concentration of the smoke;

[0052] Perform density gradient calculation on the spatial distribution concentration through the gradient calculation algorithm to obtain the concentration gradient field;

[0053] Perform directional analysis on the concentration gradient field through the directional analysis model to obtain the direction vector of smoke diffusion;

[0054] Diffusion trend analysis is performed based on the direction vector to obtain the smoke diffusion trend: among them, the smoke diffusion trend includes the smoke diffusion range, and the smoke diffusion range includes circular, elliptical and irregular shapes.

[0055] The present invention also provides a smoke monitoring device based on a MEMS multi-channel intelligent gas sensor, including:

[0056] A monitoring module for performing real-time smoke monitoring on a target environmental area through a plurality of the MEMS multi-channel intelligent gas sensors to obtain smoke information;

[0057] An extraction module for extracting smoke characteristics from the smoke information through wavelet transform technology to obtain smoke dynamic characteristics; among them, the smoke characteristics include smoke movement characteristics and smoke morphology characteristics;

[0058] A calculation module for performing density statistical calculation on the smoke dynamic characteristics to obtain a smoke density distribution area;

[0059] A drawing module for dynamically drawing the smoke density distribution area based on the smoke dynamic characteristics to generate a dynamic smoke concentration distribution heat map;

[0060] A prediction module for performing risk trend prediction based on the smoke concentration distribution heat map to obtain a risk prediction result;

[0061] An early warning module for giving an early warning if the risk prediction result exceeds a preset risk range.

[0062] The present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0063] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0064] The smoke monitoring method based on the MEMS multi-channel intelligent gas sensor provided by the present invention includes the following steps: performing real-time smoke monitoring on the target environmental area through a plurality of the MEMS multi-channel intelligent gas sensors to obtain smoke information; extracting smoke characteristics from the smoke information through wavelet transform technology to obtain smoke dynamic characteristics; performing density statistical calculation on the smoke dynamic characteristics to obtain a smoke density distribution area; dynamically drawing the smoke density distribution area based on the smoke dynamic characteristics to generate a dynamic heat map of smoke concentration distribution; predicting a risk trend based on the smoke concentration distribution heat map to obtain a risk prediction result; if the risk prediction result exceeds a preset risk range, an alarm is issued. Through the above technical solution, the technical problem that most traditional methods rely on simple threshold judgment and cannot effectively identify the dynamic changes of smoke, resulting in inaccurate monitoring results and difficulty in predicting the development trend of a fire, is solved. It greatly improves the ability to track the dynamic changes of smoke in real time, thereby realizing the accurate prediction of the development trend of a fire. Brief Description of the Drawings

[0065] Figure 1 is a schematic diagram of the steps of the smoke monitoring method based on the MEMS multi-channel intelligent gas sensor in an embodiment of the present invention;

[0066] Figure 2 is a structural block diagram of the smoke monitoring device based on the MEMS multi-channel intelligent gas sensor in an embodiment of the present invention;

[0067] Figure 3 is a schematic structural diagram of a computer device in an embodiment of the present invention.

[0068] The realization, functional features, and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiments

[0069] In order to make the object, technical solution, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0070] As Figure 1 shown, Figure 1 is a schematic diagram of the steps of the smoke monitoring method based on the MEMS multi-channel intelligent gas sensor in an embodiment of the present invention;

[0071] An embodiment of the present invention provides a smoke monitoring method based on the MEMS multi-channel intelligent gas sensor, including the following steps:

[0072] Step S1, use multiple of the MEMS multi-channel intelligent gas sensors to perform real-time monitoring of smoke in the target environmental area, and obtain smoke information.

[0073] Specifically, use multiple of the MEMS multi-channel intelligent gas sensors to perform real-time monitoring of smoke in the target environmental area, and obtain smoke information.

[0074] This process involves using multiple MEMS multi-channel intelligent gas sensors to perform real-time monitoring of smoke in the target environmental area, and finally obtaining smoke information.

[0075] Perform parameter calibration on the MEMS multi-channel intelligent gas sensors to obtain the MEMS multi-channel intelligent gas sensors after parameter calibration; Before the monitoring starts, it is first necessary to perform parameter calibration on the MEMS multi-channel intelligent gas sensors to ensure that the collected data is accurate. This includes adjusting key parameters such as the sensitivity and response time of the sensors to ensure that they can operate stably in various environments.

[0076] Based on the monitoring task plan, formulate the monitoring layout of the MEMS multi-channel intelligent gas sensors. The MEMS multi-channel intelligent gas sensors are deployed according to the monitoring layout. During the deployment process, use the MEMS multi-channel intelligent gas sensors after parameter calibration to perform real-time monitoring of smoke in the target environmental area, and obtain the monitoring data of a single sensor; According to the requirements of the monitoring task, the system will formulate specific deployment positions and monitoring modes for each MEMS multi-channel intelligent gas sensor. The sensors will be deployed according to these plans and start real-time monitoring of the smoke situation in the target environmental area after deployment. Each monitoring obtains a set of monitoring data of a single sensor.

[0077] Perform data fusion on the monitoring data of the single sensor to obtain smoke information covering the target environmental area; Since the target environmental area may be large, the data of a single sensor cannot comprehensively reflect the situation of the entire area. Therefore, it is necessary to fuse the monitoring data of multiple sensors to form a complete set of smoke information covering the target environmental area.

[0078] Perform data processing on the smoke information covering the target environmental area through a preset data processing algorithm to obtain the processed smoke information, and use the processed smoke information as the smoke information. Finally, in order to improve the quality of the smoke information and make it more suitable for subsequent analysis work, the system will use a preset data processing algorithm to process the fused smoke information. The information processed in this way is called the processed smoke information and is also the final smoke information. This data processing can improve the accuracy of the information and help better identify the characteristics and changes of the smoke.

[0079] In this way, the system can obtain various details of the smoke in the target area in real time, providing a basis for subsequent data analysis and processing. Such a design ensures the comprehensiveness and accuracy of the monitoring system, enabling the timely discovery of potential safety hazards, especially being of great significance in early fire warning.

[0080] Step S2: Extract the smoke characteristics from the smoke information through wavelet transform technology to obtain the dynamic smoke characteristics; wherein, the smoke characteristics include the smoke movement characteristics and the smoke shape characteristics.

[0081] Specifically, this process involves using wavelet transform technology to extract the characteristics from the obtained smoke information, and finally obtaining the dynamic smoke characteristics, specifically including the smoke movement characteristics and the smoke shape characteristics.

[0082] Apply wavelet transform technology to the smoke information to extract the smoke characteristics and obtain the dynamic smoke characteristics; after obtaining the smoke information, the system uses wavelet transform technology to process this information. Wavelet transform is an effective signal processing method, which can capture the local characteristics in the smoke information, especially the time-frequency characteristics, so as to extract the dynamic characteristics of the smoke.

[0083] Wherein, the smoke characteristics include the smoke movement characteristics and the smoke shape characteristics;

[0084] The smoke movement characteristics refer to the dynamic attributes such as the moving speed, direction and diffusion situation of the smoke;

[0085] The smoke shape characteristics cover the static attributes such as the shape, concentration distribution and boundary contour of the smoke.

[0086] The dynamic smoke characteristics extracted through wavelet transform technology include but are not limited to information such as the diffusion speed, diffusion direction of the smoke, and the change of concentration distribution over time. This process ensures that the key characteristics extracted from the original smoke information can accurately reflect the behavior pattern and development trend of the smoke.

[0087] In this way, the system can extract important dynamic smoke characteristics from the smoke information, providing detailed data support for subsequent analysis and risk assessment. Such a design helps to improve the accuracy and reliability of the monitoring system, ensuring the timely identification and response to potential safety threats.

[0088] Step S3: Conduct density statistical calculation on the dynamic smoke characteristics to obtain the smoke density distribution area.

[0089] Specifically, this process involves conducting density statistical calculation on the extracted dynamic smoke characteristics, and finally obtaining the smoke density distribution area.

[0090] Perform density statistical calculation on the dynamic characteristics of the smoke to obtain the smoke density distribution area; First, based on the extracted dynamic characteristics of the smoke, the system will statistically analyze the appearance frequency and concentration of the smoke at different positions. This process usually involves analyzing the dynamic characteristic dataset of the smoke, and by calculating the number of times the smoke appears and the concentration level at each position, the density distribution of the smoke at each position is obtained.

[0091] The density statistical calculation is carried out based on the smoke movement characteristics and smoke morphology characteristics in the dynamic characteristics of the smoke;

[0092] Smoke movement characteristics, such as the speed and direction of the smoke, help determine the position change of the smoke at different time points;

[0093] Smoke morphology characteristics, such as the shape and concentration distribution of the smoke, are used to evaluate the density of the smoke at each position.

[0094] Through the statistical analysis of these characteristics, the system can identify the areas with higher or lower smoke concentration, thus obtaining the smoke density distribution area.

[0095] For example, by calculating the average concentration of the smoke in a specific area within a certain time period, it can be determined whether this area belongs to the high-density distribution area.

[0096] The finally formed smoke density distribution area reflects the distribution of the smoke in the entire monitoring area, helps identify the areas with the densest smoke, and provides an important reference for subsequent risk assessment.

[0097] In this way, the system can extract the information of the smoke density distribution from the dynamic characteristics of the smoke, providing a basis for subsequent analysis and decision-making. Such a design helps improve the accuracy and reliability of the monitoring system, ensuring the timely identification and response to potential security threats.

[0098] Step S4, based on the dynamic characteristics of the smoke, dynamically draw the smoke density distribution area to generate a dynamic heat map of the smoke concentration distribution.

[0099] Specifically, this process involves using the previously calculated smoke density distribution data, combined with the dynamic characteristics of the smoke, to dynamically draw the heat map of the smoke concentration distribution.

[0100] Based on the dynamic characteristics of the smoke: Here it refers to the dynamic characteristics of the smoke that have been extracted and analyzed, including but not limited to information such as the speed, direction of the smoke, and changes in the smoke morphology.

[0101] Dynamically draw the smoke density distribution area: This step aims to use the aforementioned smoke density distribution area data, combined with the changes in the dynamic characteristics of the smoke, to real-time update the visual representation of the smoke concentration distribution.

[0102] Generate a heat map of the dynamic smoke concentration distribution: Through the above steps, the system will generate a heat map that reflects the variation of smoke concentration over time and space. Different colors on the heat map represent different smoke concentrations, and generally, the darker the color, the higher the smoke concentration.

[0103] Specifically, the operation process of this technical solution is as follows:

[0104] Utilize the dynamic characteristics of smoke: First, based on the previously obtained dynamic characteristics of smoke, such as the speed, direction of the smoke, and its variation over time, determine the position and concentration distribution of the smoke at different time points.

[0105] Dynamically draw the smoke density distribution area: According to the data of the smoke density distribution area and in combination with the real-time changes in the dynamic characteristics of the smoke, dynamically update the smoke density distribution map. This means that over time, the smoke density distribution map will be continuously redrawn to reflect the latest smoke distribution.

[0106] Generate a heat map of the dynamic smoke concentration distribution: By converting the data of the smoke density distribution area into a color gradient, a heat map of the dynamic smoke concentration distribution is formed. This heat map can intuitively display the concentration distribution of the smoke over time in the monitored area.

[0107] In this way, the heat map of the dynamic smoke concentration distribution can not only show the distribution state of the smoke at a certain moment but also display its development trend over time, which is of great significance for monitoring the spread of smoke and taking corresponding measures.

[0108] Step S5: Based on the heat map of the smoke concentration distribution, conduct a risk trend prediction to obtain a risk prediction result.

[0109] Specifically, based on the heat map of the smoke concentration distribution, conduct a risk trend prediction to obtain a risk prediction result.

[0110] This process involves using the heat map of the smoke concentration distribution to predict the risk trend and finally obtaining the risk prediction result.

[0111] Based on the heat map of the smoke concentration distribution: Here, it refers to the generated heat map of the dynamic smoke concentration distribution, which can intuitively display the concentration distribution of the smoke over time in the monitored area.

[0112] Conduct a risk trend prediction: Based on the data in the heat map of the smoke concentration distribution, adopt appropriate prediction models and technical means to predict the possibility of smoke spread and its influence range in the future for a certain period of time.

[0113] Obtain risk prediction results: Integrate the data obtained from the above prediction process into an easy-to-understand form, namely risk prediction results. These results can be used to guide decision-makers to take corresponding preventive measures, such as evacuating the population in affected areas in advance, adjusting traffic routes, etc., so as to effectively reduce potential risks.

[0114] Specifically, the operation process of this technical solution is as follows:

[0115] Based on the heat map of the smoke concentration distribution, conduct risk trend prediction to obtain risk prediction results: Utilize the data in the heat map of the smoke concentration distribution, and through a preset prediction model and technical means, predict the possibility of future smoke diffusion and its influence range, and accordingly form risk prediction results to guide corresponding preventive measures.

[0116] In this way, the system can conduct risk trend prediction based on the heat map of the smoke concentration distribution, and accordingly form risk prediction results, providing a scientific basis for taking preventive measures.

[0117] Step S6, if the risk prediction result exceeds the preset risk range, an alarm is issued.

[0118] Specifically, if the risk prediction result exceeds the preset risk range, an alarm is issued.

[0119] This process involves triggering an alarm mechanism when the risk prediction result exceeds the preset risk threshold. If the risk prediction result: Here, it refers to the risk prediction result obtained by analyzing the heat map of the smoke concentration distribution. Exceeds the preset risk range: It means that the risk prediction result shows that the possibility of future smoke diffusion and its influence range exceed the pre-set safety threshold. Then an alarm is issued: When the risk prediction result exceeds the preset risk range, the system will automatically trigger the alarm mechanism to notify relevant personnel to take necessary countermeasures in a timely manner. Specifically, the operation process of this technical solution is as follows: If the risk prediction result exceeds the preset risk range, an alarm is issued: When the risk prediction result indicates that the possibility of future smoke diffusion and its influence range exceed the pre-set safety threshold, the system will automatically activate the alarm mechanism, send an alarm signal to relevant departments and individuals, and remind them to take emergency measures to mitigate the possible danger. In this way, the system can promptly activate the alarm mechanism when the risk prediction result exceeds the safety threshold, ensuring that relevant personnel can respond quickly and take effective measures to reduce the impact of potential risks.

[0120] In a specific embodiment, the real-time monitoring of smoke in the target environmental area by multiple MEMS multi-channel intelligent gas sensors to obtain smoke information includes:

[0121] Real-time monitoring of smoke in the target environmental area is carried out through multiple MEMS multi-channel intelligent gas sensors, and preliminary smoke information is obtained; among them, there are multiple pieces of the preliminary smoke information;

[0122] Noise filtering is performed on the preliminary smoke information to obtain denoised smoke information;

[0123] Pattern recognition is performed on the denoised smoke information to obtain a smoke pattern recognition result; among them, the smoke pattern recognition result includes the smoke source type, the smoke concentration level, and the smoke diffusion speed;

[0124] The pattern recognition results are fused to obtain smoke information.

[0125] Specifically, multiple MEMS multi-channel intelligent gas sensors are used to monitor the smoke in the target environmental area in real time to obtain preliminary smoke information; among them, there are multiple pieces of the preliminary smoke information. When multiple MEMS multi-channel intelligent gas sensors are used to monitor the smoke in the target environmental area in real time, due to the large number of sensors, multiple pieces of preliminary smoke information will be obtained. The data collected by each sensor constitutes a piece of preliminary smoke information, and these information contain the basic characteristics of the smoke, such as the presence or absence of smoke, the concentration level, etc. Noise filtering is performed on the preliminary smoke information to obtain denoised smoke information; noise filtering processing is performed on the preliminary smoke information to remove the random noise and other interference signals generated during the sensor acquisition process, and relatively pure smoke information, that is, denoised smoke information, is obtained. Noise filtering can be achieved through digital signal processing techniques. For example, methods such as using low-pass filters and median filtering are used to reduce the influence of noise. Pattern recognition is performed on the denoised smoke information to obtain a smoke pattern recognition result; among them, the smoke pattern recognition result includes the smoke source type, the smoke concentration level, and the smoke diffusion speed; pattern recognition processing is performed on the denoised smoke information to identify key characteristics such as the smoke source type, concentration level, and diffusion speed by analyzing the characteristics of the smoke. Pattern recognition can be achieved using machine learning algorithms or specific pattern matching techniques. For example, methods such as support vector machine (SVM) and neural network are used. Fusion is performed on the pattern recognition result to obtain smoke information. Finally, information such as the smoke source type, smoke concentration level, and smoke diffusion speed obtained by pattern recognition is subjected to fusion processing, and the data of all sensors are comprehensively considered to obtain the final smoke information. Fusion processing can be achieved through data fusion algorithms. For example, methods such as Bayesian estimation and Kalman filtering are used to improve the accuracy and reliability of the information. In this way, the system can obtain real-time monitoring data of the smoke from multiple MEMS multi-channel intelligent gas sensors. After a series of processing, accurate and reliable smoke information is obtained, providing support for subsequent analysis and decision-making. This technical solution not only improves the accuracy and reliability of smoke monitoring but also can track the dynamic changes of the smoke in real time, thereby achieving accurate prediction of the development trend of the fire.

[0126] In a specific embodiment, the smoke feature extraction is performed on the smoke information through wavelet transform technology to obtain smoke dynamic features, including:

[0127] The frequency components of the smoke information are decomposed through a preset wavelet transform technology to obtain smoke decomposition features;

[0128] Motion feature extraction is performed on the smoke decomposition features to obtain smoke motion features;

[0129] Calculate the gradient of the smoke movement characteristics to obtain the smoke gradient characteristics; wherein, the smoke gradient characteristics include the smoke flow velocity characteristics and the smoke diffusion direction characteristics;

[0130] Perform trajectory analysis on the smoke movement characteristics to obtain the smoke movement trajectory;

[0131] Based on the smoke movement trajectory, identify and predict the diffusion characteristics of the smoke to obtain the smoke diffusion characteristics;

[0132] Extract the morphological characteristics of the smoke decomposition characteristics to obtain the smoke boundary shape characteristics and the smoke contour characteristics;

[0133] Based on the smoke boundary shape characteristics, the smoke contour characteristics and the smoke diffusion characteristics, perform time-frequency localization analysis to obtain the smoke morphological change characteristics;

[0134] Fuse the smoke movement characteristics and the smoke morphological change characteristics to obtain the smoke dynamic characteristics; wherein, the smoke dynamic characteristics include the smoke movement speed change, the smoke direction change frequency and the smoke instantaneous acceleration.

[0135] Specifically, the preset wavelet transform technology is used to decompose the frequency components of the smoke information to obtain smoke decomposition features; the preset wavelet transform technology is used to process the smoke information, decompose it into different frequency components, and obtain smoke decomposition features. Wavelet transform is a time-frequency analysis tool that can capture local features in smoke information, especially time-frequency characteristics, so as to extract the dynamic features of smoke. Through wavelet transform, the smoke information can be converted into coefficients at different scales and frequencies, and these coefficients reflect different features of the smoke. Extract the motion features of the smoke decomposition features to obtain smoke motion features; further analyze the smoke decomposition features to extract the motion features of the smoke, such as the moving speed and direction of the smoke. Motion feature extraction can be achieved using time series analysis or other signal processing techniques to identify the dynamic changes of the smoke. Calculate the gradient of the smoke motion features to obtain smoke gradient features; among them, the smoke gradient features include smoke flow speed features and smoke diffusion direction features; by calculating the gradient of the smoke motion features, smoke gradient features can be obtained, including smoke flow speed features and smoke diffusion direction features. Gradient calculation can help identify the change rate of smoke concentration over time and space, so as to infer the flow speed and diffusion direction of the smoke. Conduct trajectory analysis on the smoke motion features to obtain smoke motion trajectories; conduct trajectory analysis on the smoke motion features to obtain the motion trajectories of the smoke in space. Trajectory analysis can help identify the propagation path of the smoke in space, which is very important for predicting the diffusion range of the smoke. Based on the smoke motion trajectories, identify and predict the diffusion characteristics of the smoke to obtain smoke diffusion characteristics; according to the smoke motion trajectories, identify and predict the diffusion characteristics of the smoke, such as diffusion speed and diffusion range. Diffusion characteristic identification and prediction can be achieved using mathematical models and machine learning techniques to improve the accuracy of prediction. Extract the morphological features of the smoke decomposition features to obtain smoke boundary shape features and smoke contour features; extract the morphological features of the smoke decomposition features to obtain smoke boundary shape features and smoke contour features, which are used to describe the shape and boundary of the smoke. Morphological feature extraction can help identify the shape and edges of the smoke, which is crucial for understanding the morphological changes of the smoke. Based on the smoke boundary shape features, smoke contour features and smoke diffusion characteristics, conduct time-frequency localization analysis to obtain smoke morphological change features; based on the smoke boundary shape features, smoke contour features and smoke diffusion characteristics, conduct time-frequency localization analysis to obtain smoke morphological change features, which describe the change of the smoke shape over time. Time-frequency localization analysis can be achieved using wavelet transform or other related techniques to capture the details of the change of the smoke morphology over time. Fuse the smoke motion features and the smoke morphological change features to obtain smoke dynamic features; among them, the smoke dynamic features include smoke motion speed change, smoke direction change frequency and smoke instantaneous acceleration.Finally, the smoke motion characteristics and the smoke morphology change characteristics are fused to obtain the smoke dynamic characteristics, including key characteristics such as the change in smoke motion speed, the frequency of smoke direction change, and the instantaneous acceleration of the smoke. Feature fusion can be achieved through data fusion algorithms to improve the accuracy and reliability of the features. In this way, the system can use wavelet transform technology to extract the smoke dynamic characteristics from the smoke information, providing an important basis for subsequent risk assessment and early warning. This technical solution not only improves the accuracy and reliability of smoke monitoring but also can track the dynamic changes of smoke in real time, thereby achieving accurate prediction of the fire development trend.

[0136] In a specific embodiment, the density statistical calculation of the smoke dynamic characteristics to obtain the smoke density distribution area includes:

[0137] Normalize the smoke dynamic characteristics to obtain normalized smoke data;

[0138] Based on the target environmental area, perform spatial rasterization processing on the normalized smoke data to obtain rasterized smoke data;

[0139] Perform time window segmentation on the rasterized smoke data to obtain segmented time window smoke data;

[0140] Perform Gaussian kernel density calculation on the segmented time window smoke data to obtain the spatial density distribution of smoke with time change;

[0141] Perform smoothing processing on the spatial density distribution of smoke with time change to obtain smoothed density data;

[0142] Perform density peak detection on the smoothed density data to obtain the density peak area; wherein, the density peak area includes the density peak value and the density distribution area corresponding to the density peak value;

[0143] Through a preset density clustering algorithm, based on the density peak value, perform density similarity clustering analysis on the density distribution area to obtain a similar smoke density distribution area;

[0144] Obtain the spatial distribution area of the target environmental area, and perform mapping processing on the similar smoke density distribution area based on the spatial distribution area to obtain the smoke density distribution area.

[0145] Specifically, normalize the smoke dynamic characteristics to obtain normalized smoke data;

[0146] Normalize the dynamic characteristics of the smoke to eliminate the influence of the data dimension between different sensors and obtain normalized smoke data. The normalization process can be achieved by methods such as min-max normalization and Z-score standardization to ensure the comparison of data from all sensors on the same scale. Perform spatial rasterization on the normalized smoke data based on the target environmental area to obtain rasterized smoke data; perform spatial rasterization on the normalized smoke data according to the layout of the target environmental area, that is, divide the monitoring area into multiple grids to obtain rasterized smoke data. The spatial rasterization process can be implemented using Geographic Information System (GIS) technology to facilitate subsequent data analysis and visualization. Perform time window segmentation on the rasterized smoke data to obtain segmented time window smoke data; segment the rasterized smoke data at a certain time interval to obtain smoke data within different time windows, that is, segmented time window smoke data. The time window segmentation can be carried out based on a fixed time interval or an event-driven manner to adapt to different application scenarios. Perform Gaussian kernel density calculation on the segmented time window smoke data to obtain the spatial density distribution of time-varying smoke; perform Gaussian kernel density estimation on the segmented time window smoke data to reflect the density distribution of smoke at different time and spatial positions and obtain the spatial density distribution of time-varying smoke. The Gaussian kernel density calculation can be implemented using non-parametric probability density estimation methods to capture the details of the smoke density distribution. Perform smoothing processing on the spatial density distribution of time-varying smoke to obtain smoothed density data; perform smoothing processing on the spatial density distribution of time-varying smoke to remove the influence of noise and obtain smoother density data. The smoothing processing can be implemented using methods such as moving average and Gaussian smoothing to improve the quality of the density data. Perform density peak detection on the smoothed density data to obtain the density peak area; among them, the density peak area includes the density peak value and the density distribution area corresponding to the density peak value; perform density peak detection on the smoothed density data to find the area with the highest smoke density, that is, the density peak area, including the density peak value and its corresponding density distribution area. The density peak detection can be implemented using methods such as local maximum search and watershed algorithm to identify the peak area of the smoke density. Through a preset density clustering algorithm, perform density similarity clustering analysis on the density distribution area based on the density peak value to obtain a similar smoke density distribution area; use a preset density clustering algorithm to perform clustering analysis on the density distribution area according to the density peak value to obtain a similar smoke density distribution area. The density clustering algorithm can be implemented using methods such as DBSCAN and OPTICS to identify the aggregation area of the smoke density distribution. Obtain the spatial distribution area of the target environmental area, and perform mapping processing on the similar smoke density distribution area based on the spatial distribution area to obtain the smoke density distribution area. Obtain the spatial distribution area information of the target environmental area, and map the similar smoke density distribution area back to the actual spatial position to obtain the smoke density distribution area.The mapping process can be implemented using GIS technology to ensure that the smoke density distribution area matches the actual environment. In this way, starting from the dynamic characteristics of the smoke, the system can obtain the density distribution of the smoke in the target environmental area through a series of data processing steps, providing an important basis for subsequent risk assessment and early warning. This technical solution not only improves the accuracy and reliability of smoke monitoring but also enables real-time tracking of the dynamic changes of the smoke, thus achieving accurate prediction of the fire development trend.

[0147] In a specific embodiment, the dynamic drawing of the smoke density distribution area based on the dynamic characteristics of the smoke to generate a dynamic heat map of smoke concentration distribution includes:

[0148] Identify the sparse areas within the smoke density distribution area and use the preset Kriging interpolation method to perform spatial interpolation calculation on the sparse areas to obtain a complemented smoke density area;

[0149] Obtain the unobserved areas of the monitoring area, and based on the complemented smoke density area, predict the smoke density of the unobserved areas through the inverse distance weighting algorithm to obtain the smoke density of the unobserved areas;

[0150] Merge the unobserved areas corresponding to the smoke density of the unobserved areas with the complemented smoke density area to obtain a merged fog density area;

[0151] Use different colors to perform color mapping on the merged fog density area to generate a static heat map of smoke concentration;

[0152] Generate a heat map of smoke concentration for each time period by using color gradient mapping technology on the static heat map of smoke concentration;

[0153] Based on the animation generation algorithm, dynamically draw the heat map of smoke concentration for each time period based on the dynamic characteristics of the smoke to generate a dynamic heat map of smoke concentration distribution, or;

[0154] Through frame animation technology, perform time series combination and dynamic drawing on the heat map of smoke concentration for each time period based on the dynamic characteristics of the smoke to generate a dynamic heat map of smoke concentration distribution that changes over time.

[0155] Specifically, identify the sparse regions within the smoke density distribution area and use the preset Kriging interpolation method to perform spatial interpolation calculations on the sparse regions to obtain a complemented smoke density area; identify the sparse regions with less or no monitoring data within the smoke density distribution area. The Kriging Interpolation method is used, which is a spatial interpolation method based on statistical principles and is used to perform spatial interpolation calculations on the sparse regions to fill in the missing data and obtain a complemented smoke density area. The Kriging interpolation method can estimate the values of unknown points based on the data of surrounding known points, taking into account the spatial correlation of the data to improve the accuracy of interpolation. Obtain the unobserved areas of the monitoring area and, through the inverse distance weighting algorithm, predict the smoke density of the unobserved areas based on the complemented smoke density area to obtain the smoke density of the unobserved areas; obtain the unobserved areas within the monitoring area that have not been covered by sensors. The inverse distance weighting algorithm (IDW) is used, which is a commonly used interpolation method. Based on the data in the complemented smoke density area, it predicts the smoke density of the unobserved areas to obtain the smoke density of the unobserved areas. The inverse distance weighting algorithm calculates the estimated value of the unknown point based on the distance and its value of the surrounding known points, and the points closer have a greater contribution. Merge the unobserved areas corresponding to the smoke density of the unobserved areas with the complemented smoke density area to obtain a merged fog density area; merge the unobserved areas corresponding to the smoke density of the unobserved areas with the complemented smoke density area to form a complete smoke density distribution area, that is, the merged fog density area. The area merge ensures the integrity and consistency of the smoke density distribution data for the entire monitoring area. Use different colors to perform color mapping on the merged fog density area to generate a static smoke concentration heat map; use color mapping technology to color the merged fog density area with different colors according to different values of the smoke density to generate a static smoke concentration heat map. Color mapping usually uses a chromatogram, such as a heat map chromatogram, to represent areas with different smoke densities, and the shade of the color represents the level of the smoke concentration. Generate a smoke concentration heat map for each time period through the heat map generation of the static smoke concentration heat map using color gradient mapping technology; use color gradient mapping technology to further optimize the visual effect of the static smoke concentration heat map to ensure natural color transition and obtain a smoke concentration heat map for each time period. Color gradient mapping can enhance the readability of the heat map and enable observers to intuitively see the change trend of the smoke concentration.

[0156] Based on the animation generation algorithm, dynamically draw the smoke concentration heat map for each time period based on the dynamic characteristics of the smoke to generate a dynamic smoke concentration distribution heat map, or; use the animation generation algorithm to dynamically draw the smoke concentration heat map for each time period according to the dynamic characteristics of the smoke to generate a dynamic smoke concentration distribution heat map. The animation generation algorithm can automatically create a continuous sequence of images based on time series data to show the change of smoke concentration over time. Through the frame animation technology, perform time series combination and dynamic drawing on the smoke concentration heat map for each time period based on the dynamic characteristics of the smoke to generate a dynamic smoke concentration distribution heat map that changes over time. Utilize the frame animation technology to combine the smoke concentration heat maps of each time period in chronological order and perform dynamic drawing to generate a dynamic smoke concentration distribution heat map that changes over time. The frame animation technology can play a series of static images in chronological order to form a continuous dynamic effect, so as to more intuitively observe the trend of smoke diffusion. In this way, the system can start from the smoke density distribution area, through a series of data processing and visualization steps, and finally generate a dynamic smoke concentration distribution heat map that changes over time, providing an important visual reference for subsequent risk assessment and early warning. This technical solution not only improves the visualization degree of smoke monitoring, but also helps decision-makers quickly understand the situation of smoke diffusion.

[0157] In a specific embodiment, the risk trend prediction based on the smoke concentration distribution heat map to obtain a risk prediction result includes:

[0158] Extract the characteristics of the change in smoke concentration in the smoke concentration distribution heat map to obtain the dynamic change path of the high-concentration area;

[0159] Through the density gradient algorithm, analyze the smoke diffusion direction based on the dynamic change path to obtain the smoke diffusion trend;

[0160] Obtain the risk growth mode based on the smoke diffusion trend; wherein, the risk growth mode includes a rapid diffusion mode, a stable diffusion mode, and a local aggregation mode;

[0161] Input the risk growth mode into a preset risk prediction algorithm for risk calculation to obtain the risk growth rate;

[0162] Classify the risk areas based on the risk growth rate to obtain risk areas with different risk levels;

[0163] Use different identifiers to perform risk annotation on the risk areas with different risk levels to obtain a risk annotation heat map;

[0164] Based on the risk annotation heat map, perform risk trend prediction to obtain a risk prediction result.

[0165] Specifically, extract the characteristics of the smoke concentration changes in the smoke concentration distribution heat map to obtain the dynamic change path of the high-concentration area; analyze the smoke concentration changes in the smoke concentration distribution heat map, especially in the high-concentration area. Extract the trajectories of these high-concentration areas over time, that is, the dynamic change path, which helps to understand the behavior pattern of smoke diffusion. Through the density gradient algorithm, analyze the smoke diffusion direction based on the dynamic change path to obtain the smoke diffusion trend; apply the density gradient algorithm, which is a mathematical method used to quantify the change rate and direction of smoke concentration. Based on the previously extracted dynamic change path of the high-concentration area, analyze the main direction of smoke diffusion to obtain the smoke diffusion trend. Obtain the risk growth pattern based on the smoke diffusion trend; among them, the risk growth pattern includes the rapid diffusion pattern, the stable diffusion pattern, and the local aggregation pattern; according to the smoke diffusion trend, identify three main risk growth patterns: Rapid diffusion pattern: The smoke rapidly diffuses in multiple directions, indicating a high risk of fire or smoke diffusion. Stable diffusion pattern: The smoke diffuses slowly and evenly, which may indicate a stable smoke source, but continuous monitoring is still required. Local aggregation pattern: The smoke accumulates in certain areas without spreading outward, which may be due to physical obstacles or other factors causing the smoke to accumulate at specific locations. Input the risk growth pattern into a preset risk prediction algorithm for risk calculation to obtain the risk growth rate; input the identified risk growth pattern into a pre-set risk prediction algorithm, which will consider the impact of various factors (such as wind speed, terrain, etc.) on smoke diffusion. Calculate the speed of smoke diffusion and the risk growth rate it may cause through the risk prediction algorithm. Classify the risk areas based on the risk growth rate to obtain risk areas with different risk levels; according to the speed of the risk growth rate, divide the monitoring area into areas with different risk levels. These areas can include low-risk areas, medium-risk areas, and high-risk areas, etc., in order to facilitate the adoption of corresponding preventive measures. Use different markings to label the risk areas with different risk levels to obtain a risk-labeled heat map; use different colors or symbols to mark the areas with different risk levels to generate a risk-labeled heat map. The risk-labeled heat map can clearly show which areas are in a high-risk state, facilitating observers to quickly understand the risk distribution of smoke diffusion. Based on the risk-labeled heat map, predict the risk trend to obtain the risk prediction result. Combine the risk-labeled heat map and further analyze the risk trend that may be caused by future smoke diffusion. The risk prediction result includes how the smoke diffusion may develop and which areas' risks will increase or decrease, which is crucial for formulating an emergency response plan. In this way, the system can start from the smoke concentration distribution heat map, through a series of data processing and analysis steps, and finally generate the risk prediction result, providing an important reference basis for subsequent risk management and emergency response.This technical solution not only improves the accuracy of risk prediction but also helps decision-makers take effective measures in a timely manner to reduce potential hazards.

[0166] In a specific embodiment, through the density gradient algorithm, based on the dynamic change path, the analysis of the smoke diffusion direction is carried out to obtain the smoke diffusion trend, including:

[0167] The smoke concentration is collected for the dynamic change path through the multi-point sampling technique to obtain the spatial distribution concentration of the smoke;

[0168] The density gradient calculation of the spatial distribution concentration is carried out through the gradient calculation algorithm to obtain the concentration gradient field;

[0169] The directional analysis of the concentration gradient field is carried out through the directional analysis model to obtain the direction vector of the smoke diffusion;

[0170] Based on the direction vector, the diffusion trend analysis is carried out to obtain the smoke diffusion trend: wherein, the smoke diffusion trend includes the smoke diffusion range, and the smoke diffusion range includes circular, elliptical and irregular shapes.

[0171] Specifically, when implementing the content in the above-mentioned claim book, the first thing to clarify is the analysis process of the smoke diffusion direction, which includes several key steps: the acquisition of the dynamic change path, the determination of the smoke spatial distribution concentration, the calculation of the density gradient field, and the final analysis of the smoke diffusion trend. Initially, the density gradient algorithm is used to collect the dynamic change path of the smoke. This process is completed through the multi-point sampling technique. The multi-point sampling technique can collect smoke concentration data at different points in space. The selection of these sampling points is usually based on the possible diffusion paths of the smoke, the initial position of the smoke source, and various influencing factors of the environment. By collecting the concentration data at these sampling points, a preliminary understanding of the smoke distribution in space can be obtained. These data not only reflect the static distribution of the smoke at a certain moment but also can show the dynamic diffusion process of the smoke through the analysis of the time series. Next, the gradient calculation algorithm is used to process the obtained spatial distribution concentration data. The core of the gradient calculation is to construct a concentration gradient field through the spatial change of the concentration. Specifically, the gradient calculation algorithm will determine how the smoke diffuses in space according to the concentration difference between the sampling points. By calculating these differences, a density gradient field can be constructed, which reflects the diffusion speed, direction, and intensity of the smoke in space. In other words, the density gradient field is a vector field of smoke diffusion, and the vector at each point in the field represents the concentration change rate and direction of the smoke at that point. After obtaining the concentration gradient field, the further work is to perform a directional analysis on the gradient field through a directional analysis model. The purpose of the directional analysis is to determine the main direction of the smoke diffusion, and this step is very important for understanding and predicting the smoke diffusion path. The directional analysis model will calculate the main diffusion directions of the smoke at different positions according to the vector information in the gradient field. These direction vectors not only show the diffusion direction of the smoke but also can provide information about the diffusion speed, thus helping to predict the future diffusion trend. Once the directional analysis is completed, a more detailed diffusion trend analysis can be carried out based on the direction vectors. The diffusion trend of the smoke is obtained through the comprehensive analysis of the direction vectors, which includes analyzing the diffusion range of the smoke. The shape of the diffusion range can be regular, such as circular or elliptical, or irregular, and the specific shape depends on the initial state of the smoke source, environmental conditions, and other factors that may affect the smoke diffusion. By comprehensively considering these factors, a more accurate prediction of the smoke diffusion range can be obtained, thus providing guidance for actual response and handling. In practical applications, the above steps can help us better understand the smoke diffusion process. Especially in cases involving smoke diffusion such as fires and industrial accidents, through these analysis steps, the diffusion direction and range of the smoke can be effectively predicted, winning time for taking corresponding measures. These steps can also be applied to air pollution monitoring, smoke control, and other environmental management fields.Through the density gradient algorithm and the directional analysis model, combined with real-time data collection and processing, accurate prediction and monitoring of smoke diffusion can be achieved, providing strong guarantees for environmental safety and public health.

[0172] The above describes the smoke monitoring method based on the MEMS multi-channel intelligent gas sensor in the embodiments of the present invention. Next, the smoke monitoring device based on the MEMS multi-channel intelligent gas sensor in the embodiments of the present invention will be described. Please refer to Figure 2 One embodiment of the smoke monitoring device based on the MEMS multi-channel intelligent gas sensor in the embodiments of the present invention includes:

[0173] A monitoring module 21, configured to perform real-time monitoring of smoke in a target environmental area through the multiple MEMS multi-channel intelligent gas sensors to obtain smoke information;

[0174] An extraction module 22, configured to extract smoke characteristics from the smoke information through wavelet transform technology to obtain smoke dynamic characteristics; wherein, the smoke characteristics include smoke movement characteristics and smoke morphology characteristics;

[0175] A calculation module 23, configured to perform density statistical calculation on the smoke dynamic characteristics to obtain a smoke density distribution area;

[0176] A drawing module 24, configured to dynamically draw the smoke density distribution area based on the smoke dynamic characteristics to generate a dynamic smoke concentration distribution heat map;

[0177] A prediction module 25, configured to perform risk trend prediction based on the smoke concentration distribution heat map to obtain a risk prediction result;

[0178] An early warning module 26, configured to give an early warning if the risk prediction result exceeds a preset risk range.

[0179] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to that described in the above method embodiment, and details will not be repeated here.

[0180] Refer to Figure 3 In addition, an embodiment of the present invention also provides a computer device, and its internal structure can be as Figure 3As shown in the figure. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0181] Those skilled in the art can understand that Figure 3 the structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0182] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0183] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0184] It should be noted that in this text, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, apparatus, article or method comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such a process, apparatus, article or method. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of additional identical elements in the process, apparatus, article or method comprising such element.

[0185] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A smoke monitoring method based on MEMS multi-channel intelligent gas sensor, characterized in that: The following steps are involved: The target environment area is monitored for smoke in real time by using a plurality of the MEMS multi-channel intelligent gas sensors to obtain smoke information; Extracting smoke features from the smoke information by wavelet transform technology to obtain smoke dynamic features; wherein the smoke features include smoke motion features and smoke morphology features; Performing density statistics calculation on the dynamic characteristics of the smoke to obtain a smoke density distribution area; Dynamically draw the smoke density distribution area based on the dynamic characteristics of the smoke to generate a dynamic smoke concentration distribution heat map; Perform risk trend prediction based on the smoke concentration distribution heat map to obtain a risk prediction result; If the risk prediction result exceeds the preset risk range, an early warning is issued; The smoke information is subjected to smoke feature extraction by wavelet transform technology to obtain smoke dynamic features, including: Decomposing the smoke information by frequency components through a preset wavelet transform technique to obtain smoke decomposition features; Extracting motion features from the smoke decomposition features to obtain smoke motion features; Performing gradient calculation on the smoke motion characteristics to obtain smoke gradient characteristics; wherein the smoke gradient characteristics include smoke flow velocity characteristics and smoke diffusion direction characteristics; Performing trajectory analysis on the smoke movement characteristics to obtain a smoke movement trajectory; Based on the smoke movement trajectory, the diffusion characteristics of the smoke are identified and predicted to obtain the smoke diffusion characteristics; Performing morphological feature extraction on the smoke decomposition features to obtain smoke boundary shape features and smoke contour features; Based on the smoke boundary shape characteristics, smoke contour characteristics and smoke diffusion characteristics, time-frequency localization analysis is performed to obtain smoke morphology change characteristics; The smoke motion feature and the smoke morphology change feature are integrated to obtain the smoke dynamic feature; wherein the smoke dynamic feature includes the change of smoke motion speed, the frequency of smoke direction change and the instantaneous acceleration of smoke; The performing density statistics calculation on the dynamic characteristics of the smoke to obtain the smoke density distribution area includes: Normalizing the dynamic characteristics of the smoke to obtain normalized smoke data; Performing spatial rasterization processing on the normalized smoke data based on the target environment area to obtain rasterized smoke data; Performing time window segmentation on the rasterized smoke data to obtain segmented time window smoke data; Performing Gaussian kernel density calculation on the segmented time window smoke data to obtain time-varying smoke spatial density distribution; Smoothing the time-varying smoke spatial density distribution to obtain smoothed density data; Perform density peak detection on the smoothed density data to obtain a density peak area; wherein the density peak area includes the density peak and the density distribution area corresponding to the density peak; By using a preset density clustering algorithm, density similarity clustering analysis is performed on the density distribution area based on the density peak value to obtain a similar smoke density distribution area; The spatial distribution area of ​​the target environment area is acquired, and the similar smoke density distribution area is mapped based on the spatial distribution area to obtain the smoke density distribution area.

2. The smoke monitoring method based on MEMS multi-channel intelligent gas sensor according to claim 1 is characterized in that: The method of monitoring smoke in the target environment area in real time by using a plurality of MEMS multi-channel intelligent gas sensors to obtain smoke information includes: The smoke in the target environment area is monitored in real time by a plurality of the MEMS multi-channel intelligent gas sensors to obtain preliminary smoke information; wherein the preliminary smoke information has a plurality of; Performing noise filtering on the preliminary smoke information to obtain denoised smoke information; Performing pattern recognition on the denoised smoke information to obtain a smoke pattern recognition result; wherein the smoke pattern recognition result includes smoke source type, smoke concentration level, and smoke diffusion speed; The pattern recognition results are fused to obtain smoke information.

3. The smoke monitoring method based on MEMS multi-channel intelligent gas sensor according to claim 1 is characterized in that: The dynamically drawing the smoke density distribution area based on the dynamic characteristics of the smoke to generate a dynamic smoke concentration distribution heat map includes: Identify the sparse area in the smoke density distribution area, and use a preset Kriging interpolation method to perform spatial interpolation calculation on the sparse area to obtain a completed smoke density area; Obtaining an unobserved area of ​​the monitoring area, and predicting the smoke density of the unobserved area based on the completed smoke density area by using an inverse distance weighted algorithm to obtain the smoke density of the unobserved area; Merging the unobserved area corresponding to the smoke density of the unobserved area with the completed smoke density area to obtain a merged fog density area; Using different colors to perform color mapping on the combined fog density area to generate a static smoke concentration heat map; The static smoke concentration heat map is generated by color gradient mapping technology to obtain the smoke concentration heat map for each time period; Based on the animation generation algorithm, dynamically draw the smoke concentration heat map of each time period based on the dynamic characteristics of the smoke to generate a dynamic smoke concentration distribution heat map, or; By using frame animation technology, the smoke concentration heat map of each time period is combined and dynamically drawn in time series based on the dynamic characteristics of the smoke, so as to generate a dynamic smoke concentration distribution heat map that changes with time.

4. The smoke monitoring method based on MEMS multi-channel intelligent gas sensor according to claim 1 is characterized in that: The risk trend prediction is performed based on the smoke concentration distribution heat map to obtain a risk prediction result, including: Extract features of smoke concentration changes in the smoke concentration distribution thermal map to obtain the dynamic change path of high-concentration areas; By using a density gradient algorithm, the smoke diffusion direction is analyzed based on the dynamically changing path to obtain the smoke diffusion trend; A risk growth pattern is obtained based on the smoke diffusion trend; wherein the risk growth pattern includes a rapid diffusion pattern, a stable diffusion pattern and a local aggregation pattern; Inputting the risk growth model into a preset risk prediction algorithm to perform risk calculation to obtain a risk growth rate; Classifying risk areas based on the risk growth rate to obtain risk areas of different risk levels; Using different identifiers to label the risk areas of different risk levels, and obtaining a risk labeling heat map; A risk trend prediction is performed based on the risk annotation heat map to obtain a risk prediction result.

5. The smoke monitoring method based on MEMS multi-channel intelligent gas sensor according to claim 4 is characterized in that: The density gradient algorithm is used to analyze the smoke diffusion direction based on the dynamic change path to obtain the smoke diffusion trend, including: The smoke concentration is collected on the dynamically changing path through multi-point sampling technology to obtain the spatial distribution concentration of the smoke; The density gradient of the spatial distribution concentration is calculated by the gradient calculation algorithm to obtain the concentration gradient field; Directional analysis of the concentration gradient field is performed through a directional analysis model to obtain the direction vector of smoke diffusion; A diffusion trend analysis is performed based on the direction vector to obtain a smoke diffusion trend: wherein the smoke diffusion trend includes a smoke diffusion range, and the smoke diffusion range includes a circular, elliptical and irregular shape.

6. A smoke monitoring device based on MEMS multi-channel intelligent gas sensor, characterized in that: The steps for implementing the method according to any one of claims 1 to 5 include: A monitoring module, used to monitor smoke in a target environment area in real time through a plurality of the MEMS multi-channel intelligent gas sensors to obtain smoke information; An extraction module is used to extract smoke features from the smoke information by wavelet transform technology to obtain smoke dynamic features; wherein the smoke features include smoke motion features and smoke morphology features; A calculation module, used for performing density statistics calculation on the dynamic characteristics of the smoke to obtain a smoke density distribution area; A drawing module, used for dynamically drawing the smoke density distribution area based on the dynamic characteristics of the smoke, and generating a dynamic smoke concentration distribution heat map; A prediction module, used to predict risk trends based on the smoke concentration distribution thermal map to obtain risk prediction results; An early warning module is used to issue an early warning if the risk prediction result exceeds a preset risk range; The smoke information is subjected to smoke feature extraction by wavelet transform technology to obtain smoke dynamic features, including: Decomposing the smoke information by frequency components through a preset wavelet transform technique to obtain smoke decomposition features; Extracting motion features from the smoke decomposition features to obtain smoke motion features; Performing gradient calculation on the smoke motion characteristics to obtain smoke gradient characteristics; wherein the smoke gradient characteristics include smoke flow velocity characteristics and smoke diffusion direction characteristics; Performing trajectory analysis on the smoke movement characteristics to obtain a smoke movement trajectory; Based on the smoke movement trajectory, the diffusion characteristics of the smoke are identified and predicted to obtain the smoke diffusion characteristics; Performing morphological feature extraction on the smoke decomposition features to obtain smoke boundary shape features and smoke contour features; Based on the smoke boundary shape characteristics, smoke contour characteristics and smoke diffusion characteristics, time-frequency localization analysis is performed to obtain smoke morphology change characteristics; The smoke motion feature and the smoke morphology change feature are integrated to obtain the smoke dynamic feature; wherein the smoke dynamic feature includes the change of smoke motion speed, the frequency of smoke direction change and the instantaneous acceleration of smoke; The performing density statistics calculation on the dynamic characteristics of the smoke to obtain the smoke density distribution area includes: Normalizing the dynamic characteristics of the smoke to obtain normalized smoke data; Performing spatial rasterization processing on the normalized smoke data based on the target environment area to obtain rasterized smoke data; Performing time window segmentation on the rasterized smoke data to obtain segmented time window smoke data; Performing Gaussian kernel density calculation on the segmented time window smoke data to obtain time-varying smoke spatial density distribution; Smoothing the time-varying smoke spatial density distribution to obtain smoothed density data; Perform density peak detection on the smoothed density data to obtain a density peak area; wherein the density peak area includes the density peak and the density distribution area corresponding to the density peak; By using a preset density clustering algorithm, density similarity clustering analysis is performed on the density distribution area based on the density peak value to obtain a similar smoke density distribution area; The spatial distribution area of ​​the target environment area is acquired, and the similar smoke density distribution area is mapped based on the spatial distribution area to obtain the smoke density distribution area.

7. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. 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 method according to any one of claims 1 to 5 are implemented.

Images