Disclosure of Invention
The invention mainly aims to provide a smoke monitoring method based on an MEMS multichannel intelligent gas sensor and related equipment, which solve the technical problems that the traditional method mostly depends on simple threshold judgment, the dynamic change of smoke cannot be effectively identified, the monitoring result is inaccurate, and the trend of fire development is difficult to predict.
In order to achieve the above purpose, the invention provides a smoke monitoring method based on an MEMS multichannel intelligent gas sensor, which comprises the following steps:
the method comprises the steps that smoke in a target environment area is monitored in real time through a plurality of MEMS multichannel intelligent gas sensors, and smoke information is obtained;
extracting smoke characteristics of the smoke information through a wavelet transformation technology to obtain smoke dynamic characteristics, wherein the smoke characteristics comprise smoke movement characteristics and smoke morphological characteristics;
carrying out density statistics 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 smoke density distribution thermodynamic diagram;
Carrying out risk trend prediction based on the smoke concentration distribution thermodynamic diagram to obtain a risk prediction result;
and if the risk prediction result exceeds a preset risk range, early warning occurs.
Further, the method for monitoring smoke in real time through a plurality of MEMS multichannel intelligent gas sensors in a target environment area to obtain smoke information includes:
monitoring smoke in a target environment area in real time through a plurality of MEMS multichannel intelligent gas sensors to obtain preliminary smoke information, wherein the preliminary smoke information is provided with a plurality of sensors;
noise filtering is carried out on the preliminary smoke information to obtain noise-removed smoke information;
performing pattern recognition on the denoising smoke information to obtain a smoke pattern recognition result, wherein the smoke pattern recognition result comprises a smoke source type, a smoke concentration level and a smoke diffusion speed;
and fusing the pattern recognition results to obtain smoke information.
Further, the extracting the smoke characteristics of the smoke information by wavelet transformation technology to obtain smoke dynamic characteristics includes:
carrying out frequency component decomposition on the smoke information through a preset wavelet transformation technology to obtain smoke decomposition characteristics;
Extracting motion characteristics of the smoke decomposition characteristics to obtain smoke motion characteristics;
carrying out gradient calculation on smoke movement characteristics to obtain smoke gradient characteristics, wherein the smoke gradient characteristics comprise smoke flow speed characteristics and smoke diffusion direction characteristics;
performing track analysis on the smoke movement characteristics to obtain a smoke movement track;
carrying out diffusion characteristic identification and prediction on smoke based on the smoke movement track to obtain smoke diffusion characteristics;
extracting morphological characteristics of the smoke decomposition characteristics to obtain smoke boundary shape characteristics and smoke outline characteristics;
Performing time-frequency localization analysis based on the smoke boundary shape feature, the smoke outline feature and the smoke diffusion feature to obtain a smoke form change feature;
And carrying out feature fusion on the smoke movement features and the smoke form change features to obtain smoke dynamic features, wherein the smoke dynamic features comprise smoke movement speed change, smoke direction change frequency and smoke instantaneous acceleration.
Further, the calculating the density statistics of the dynamic characteristics of the smoke to obtain a smoke density distribution area includes:
normalizing the dynamic characteristics of the smoke to obtain normalized smoke data;
performing spatial rasterization 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;
Carrying out Gaussian kernel density calculation on the segmented time window smoke data to obtain time-varying smoke space density distribution;
smoothing the time-varying smoke space density distribution to obtain smoothed density data;
Detecting a density peak value of the smooth density data to obtain a density peak region, wherein the density peak region comprises a density peak value and a density distribution region corresponding to the density peak value;
Performing density similarity clustering analysis on the density distribution area based on the density peak value through a preset density clustering algorithm to obtain a similarity smoke density distribution area;
and acquiring a spatial distribution area of the target environment area, and mapping the similar smoke density distribution area based on the spatial distribution area to obtain a smoke density distribution area.
Further, the dynamically drawing the smoke density distribution area based on the smoke dynamic feature generates a dynamic smoke density distribution thermodynamic diagram, including:
Identifying a sparse region in the smoke density distribution region, and performing spatial interpolation calculation on the sparse region by adopting a preset Kriging interpolation method to obtain a full smoke density region;
Acquiring an unobserved region of the monitored region, and predicting the smoke density of the unobserved region based on the full smoke density region by an inverse distance weight algorithm to obtain the smoke density of the unobserved region;
carrying out region combination on the unobserved region corresponding to the smoke density of the unobserved region and the full smoke density region to obtain a combined smoke density region;
performing color mapping on the combined fog density areas by adopting different colors to generate a static fog concentration thermodynamic diagram;
generating a thermodynamic diagram of static smoke concentration by using a color gradient mapping technology to obtain a thermodynamic diagram of smoke concentration in each time period;
dynamically drawing the smoke concentration thermodynamic diagram of each time period based on the smoke dynamic characteristics based on an animation generation algorithm, and generating a dynamic smoke concentration distribution thermodynamic diagram or;
And carrying out time sequence combination and dynamic drawing on the smoke concentration thermodynamic diagrams of each time period based on the smoke dynamic characteristics through a frame animation technology, and generating a dynamic smoke concentration distribution thermodynamic diagram changing with time.
Further, the predicting risk trend based on the smoke concentration distribution thermodynamic diagram to obtain a risk prediction result includes:
extracting characteristics of smoke concentration change in a smoke concentration distribution thermodynamic diagram to obtain a dynamic change path of a high concentration region;
Analyzing the smoke diffusion direction based on the dynamic change path through a density gradient algorithm to obtain a smoke diffusion trend;
Obtaining a risk growth mode based on the smoke diffusion trend, wherein the risk growth mode comprises a rapid diffusion mode, a stable diffusion mode and a local aggregation mode;
inputting the risk growth mode into a preset risk prediction algorithm to perform risk calculation to obtain a risk growth speed;
Classifying the risk areas based on the risk growth speed to obtain risk areas with different risk levels;
Carrying out risk labeling on the risk areas with different risk levels by adopting different identifications to obtain a risk labeling thermodynamic diagram;
And carrying out risk trend prediction based on the risk labeling thermodynamic diagram to obtain a risk prediction result.
Further, by means of a density gradient algorithm, analysis of a smoke diffusion direction is performed based on the dynamic change path, so as to obtain a smoke diffusion trend, including:
The smoke concentration of the dynamic change path is acquired by a multipoint sampling technology, so that the spatial distribution concentration of the smoke is obtained;
Performing density gradient calculation on the spatial distribution concentration by a gradient calculation algorithm to obtain a concentration gradient field;
Carrying out directional analysis on the concentration gradient field through a directional analysis model to obtain a direction vector of smoke diffusion;
and carrying out diffusion trend analysis based on the direction vector to obtain a smoke diffusion trend, wherein the smoke diffusion trend comprises a smoke diffusion range, and the smoke diffusion range comprises a round shape, an oval shape and an irregular shape.
The invention also provides a smoke monitoring device based on the MEMS multichannel intelligent gas sensor, which comprises:
the monitoring module is used for monitoring smoke in real time in a target environment area through the MEMS multichannel intelligent gas sensors to obtain smoke information;
the extraction module is used for extracting smoke characteristics of the smoke information through a wavelet transformation technology to obtain smoke dynamic characteristics, wherein the smoke characteristics comprise smoke movement characteristics and smoke morphological characteristics;
the calculation module is used for carrying out density statistics calculation on the smoke dynamic characteristics to obtain a smoke density distribution area;
the drawing module is used for dynamically drawing the smoke density distribution area based on the smoke dynamic characteristics to generate a dynamic smoke density distribution thermodynamic diagram;
the prediction module is used for predicting the risk trend based on the smoke concentration distribution thermodynamic diagram to obtain a risk prediction result;
And the early warning module is used for early warning if the risk prediction result exceeds a preset risk range.
The invention also provides a computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of any of the methods described above when the computer program is executed.
The invention also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of any of the preceding claims.
The smoke monitoring method based on the MEMS multichannel intelligent gas sensor comprises the following steps of carrying out smoke real-time monitoring on a target environment area through a plurality of MEMS multichannel intelligent gas sensors to obtain smoke information, carrying out smoke feature extraction on the smoke information through a wavelet transformation technology to obtain smoke dynamic features, carrying out density statistics calculation on the smoke dynamic features to obtain a smoke density distribution area, carrying out dynamic drawing on the smoke density distribution area based on the smoke dynamic features to generate a dynamic smoke concentration distribution thermodynamic diagram, carrying out risk trend prediction based on the smoke concentration thermodynamic diagram to obtain a risk prediction result, and carrying out early warning if the risk prediction result exceeds a preset risk range.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic diagram illustrating steps of a smoke monitoring method based on a MEMS multichannel intelligent gas sensor according to an embodiment of the present invention;
the embodiment of the invention provides a smoke monitoring method based on an MEMS multichannel intelligent gas sensor, which comprises the following steps:
and S1, carrying out smoke real-time monitoring on a target environment area through a plurality of MEMS multichannel intelligent gas sensors to obtain smoke information.
Specifically, smoke is monitored in real time through a plurality of MEMS multichannel intelligent gas sensors in a target environment area, and smoke information is obtained.
The process involves the use of multiple MEMS multichannel intelligent gas sensors to monitor the smoke in real time in a target environmental area, ultimately obtaining smoke information.
And before monitoring starts, the MEMS multichannel intelligent gas sensor is firstly required to be subjected to parameter calibration so as to ensure that acquired data are accurate. This includes adjusting key parameters of the sensor's sensitivity, response time, etc., to ensure that they operate stably in a variety of environments.
The system comprises a monitoring task planning system, a monitoring system, a system and a monitoring system, wherein the monitoring system planning system is used for planning monitoring arrangement of the MEMS multichannel intelligent gas sensor based on the monitoring task planning system, the MEMS multichannel intelligent gas sensor is deployed according to the monitoring arrangement, smoke real-time monitoring is carried out on a target environment area through the MEMS multichannel intelligent gas sensor after parameter calibration in the deployment process, monitoring data of a single sensor are obtained, and a specific deployment position and a specific monitoring mode are formulated for each MEMS multichannel intelligent gas sensor according to the requirements of the monitoring task. The sensors are deployed according to the plans, and smoke conditions in the target environment area are monitored in real time after the deployment, and monitoring data of a group of single sensors are obtained through each monitoring.
And the data of the single sensor cannot comprehensively reflect the condition of the whole area because the target environment area is possibly larger. Therefore, the monitoring data of a plurality of sensors need to be fused to form a complete set of smoke information covering the target environment area.
And carrying out data processing on the smoke information covering the target environment area through a preset data processing algorithm to obtain processed smoke information, and taking the processed smoke information as smoke information. Finally, in order to improve the quality of the smoke information, so that the smoke information is more suitable for subsequent analysis work, the system processes the fused smoke information by using a preset data processing algorithm. The information thus processed is referred to as processed smoke information, and is also final smoke information. Such data processing may improve the accuracy of the information and help to better identify smoke characteristics and variations.
In this way, the system can acquire various details of smoke in the target area in real time, and provide a basis for subsequent data analysis and processing. The design ensures the comprehensiveness and accuracy of the monitoring system, can discover potential safety hazards in time, and has important significance in early fire early warning.
And S2, extracting smoke characteristics of the smoke information through a wavelet transformation technology to obtain smoke dynamic characteristics, wherein the smoke characteristics comprise smoke movement characteristics and smoke morphological characteristics.
Specifically, the process involves feature extraction of the obtained smoke information by using wavelet transformation technology, and finally the smoke dynamic features, including the smoke motion features and the smoke morphological features, are obtained.
And after the smoke information is obtained, the system processes the information by using the wavelet transformation technology. Wavelet transformation is an effective signal processing method, which can capture local characteristics, especially time-frequency characteristics, in smoke information, so as to extract dynamic characteristics of smoke.
Wherein the smoke features include smoke movement features and smoke morphology features;
the smoke movement characteristics refer to dynamic properties such as moving speed, moving direction and diffusion condition of smoke;
the smoke morphological characteristics cover static attributes such as the shape, concentration distribution, boundary contour and the like of the smoke.
The dynamic characteristics of the smoke extracted by the wavelet transformation technology include, but are not limited to, information such as diffusion speed, diffusion direction, change of concentration distribution with time and the like of the smoke. This process ensures that key features extracted from the original smoke information accurately reflect the behavior pattern and trend of the smoke.
In this way, the system is able to extract important smoke dynamic features 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 timely identification and handling of potential security threats.
And step S3, carrying out density statistics calculation on the smoke dynamic characteristics to obtain a smoke density distribution area.
Specifically, the process involves performing a density statistical calculation on the extracted dynamic characteristics of the smoke, and finally obtaining a smoke density distribution area.
Firstly, according to the extracted smoke dynamic characteristics, the system can count the occurrence frequency and concentration of the smoke at different positions. This process typically involves analysis of a dynamic signature dataset of the smoke, by calculating the number of occurrences and concentration levels of smoke at each location, to obtain the density distribution of the smoke at each location.
The density statistics calculation is performed based on smoke motion features and smoke morphology features in the smoke dynamic features;
smoke movement characteristics, such as the speed and direction of smoke, help determine the change in position of smoke at different points in time;
smoke morphology features, such as shape and concentration profile of the smoke, are used to evaluate the density of the smoke at various locations.
By statistical analysis of these features, the system is able to identify areas of higher or lower smoke concentration, resulting in areas of smoke density distribution.
For example, by calculating the average concentration of smoke in a particular region over a period of time, it can be determined whether that region belongs to a high density distribution region.
The finally formed smoke density distribution area reflects the distribution condition of smoke in the whole monitoring area, is helpful for identifying the area with the most dense smoke, and provides important reference for subsequent risk assessment.
In this way, the system can extract the information of the smoke density distribution from the smoke dynamic characteristics, and provide basis for subsequent analysis and decision. Such a design helps to improve the accuracy and reliability of the monitoring system, ensuring timely identification and handling of potential security threats.
And S4, dynamically drawing the smoke density distribution area based on the smoke dynamic characteristics to generate a dynamic smoke density distribution thermodynamic diagram.
In particular, this process involves dynamically mapping the smoke concentration profile thermodynamic diagram using previously calculated smoke density profile data in combination with the smoke dynamic characteristics.
Based on the smoke dynamic characteristics, the smoke dynamic characteristics which are extracted and analyzed are referred to herein, and include, but are not limited to, information such as the speed, direction and change of smoke morphology.
The smoke density distribution area is dynamically drawn, and the step aims to update the visual representation of the smoke density distribution in real time by utilizing the data of the smoke density distribution area and combining the change condition of the smoke dynamic characteristics.
Generating a dynamic smoke concentration profile thermodynamic diagram by which the system generates a thermodynamic diagram reflecting the temporal and spatial variations of smoke concentration. The difference in color on the thermodynamic diagram represents the difference in smoke concentration, with a darker color generally representing a higher smoke concentration.
Specifically, the operation flow of the technical scheme is as follows:
the smoke dynamics are utilized by first determining the location and concentration profile of the smoke at different points in time based on previously obtained smoke dynamics such as the velocity, direction and time variation of the smoke.
Dynamically drawing a smoke density distribution area, namely dynamically updating a smoke density distribution diagram according to the data of the smoke density distribution area and combining the real-time change of the smoke dynamic characteristics. This means that over time the smoke density profile will be continually redrawn to reflect the latest smoke profile.
Generating a dynamic smoke concentration profile thermodynamic diagram by converting the data of the smoke concentration profile region into a color gradient. The thermodynamic diagram can intuitively show the concentration distribution of the smoke in the monitoring area along with the time.
By the method, the dynamic smoke concentration distribution thermodynamic diagram not only can display the distribution state of smoke at a certain moment, but also can show the trend of the smoke development along with time, and has important significance for monitoring the smoke diffusion condition and taking corresponding measures.
And S5, carrying out risk trend prediction based on the smoke concentration distribution thermodynamic diagram to obtain a risk prediction result.
Specifically, risk trend prediction is performed based on the smoke concentration distribution thermodynamic diagram, and a risk prediction result is obtained.
This process involves using a thermodynamic map of the smoke concentration profile to predict risk trends and ultimately yield risk prediction results.
Based on the smoke concentration profile thermodynamic diagram, which is referred to herein as the dynamic smoke concentration profile thermodynamic diagram that has been generated, the diagram is capable of visually exhibiting the concentration profile of smoke over time within a monitored area.
And (3) predicting risk trend, namely predicting the possibility and the influence range of smoke diffusion in a future period of time by adopting a proper prediction model and a proper technical means based on data in a smoke concentration distribution thermodynamic diagram.
And (3) obtaining a risk prediction result, namely integrating the data obtained in the prediction process into a form which is easy to understand, namely the risk prediction result. These results can be used to guide decision makers to take corresponding precautions, such as evacuating the population in the affected area in advance, adjusting the traffic route, etc., thereby effectively reducing the potential risk.
Specifically, the operation flow of the technical scheme is as follows:
And predicting the risk trend based on the smoke concentration distribution thermodynamic diagram to obtain a risk prediction result, namely predicting the possibility of future smoke diffusion and the influence range thereof by using data in the smoke concentration distribution thermodynamic diagram through a preset prediction model and a technical means, and forming the risk prediction result according to the possibility of future smoke diffusion so as to guide corresponding preventive measures.
By the method, the system can predict the risk trend according to the smoke concentration distribution thermodynamic diagram, and form a risk prediction result according to the risk trend prediction result, so that scientific basis is provided for taking preventive measures.
And S6, if the risk prediction result exceeds a preset risk range, early warning occurs.
Specifically, if the risk prediction result exceeds a preset risk range, early warning occurs.
This process involves triggering an early warning mechanism when the risk prediction exceeds a preset risk threshold. The risk prediction result refers to a risk prediction result obtained by analyzing a smoke concentration distribution thermodynamic diagram. Exceeding the preset risk range means that the risk prediction result shows that the future smoke diffusion possibility and the influence range thereof exceed the preset safety threshold. And early warning occurs, namely when the risk prediction result exceeds a preset risk range, the system automatically triggers an early warning mechanism to inform related personnel to take necessary countermeasures in time. Specifically, the technical scheme comprises the following steps that if the risk prediction result exceeds a preset risk range, early warning occurs, when the risk prediction result shows that the possibility of future smoke diffusion and the influence range thereof exceed a preset safety threshold, the system automatically starts an early warning mechanism to send early warning signals to related departments and individuals to remind the departments and individuals to take emergency measures to reduce possible danger. By the method, the system can timely start the early warning mechanism when the risk prediction result exceeds the safety threshold value, ensure that related personnel can respond quickly, and take effective measures to reduce the influence of potential risks.
In a specific embodiment, the monitoring the smoke in real time through the plurality of MEMS multichannel intelligent gas sensors to the target environment area to obtain smoke information includes:
monitoring smoke in a target environment area in real time through a plurality of MEMS multichannel intelligent gas sensors to obtain preliminary smoke information, wherein the preliminary smoke information is provided with a plurality of sensors;
noise filtering is carried out on the preliminary smoke information to obtain noise-removed smoke information;
performing pattern recognition on the denoising smoke information to obtain a smoke pattern recognition result, wherein the smoke pattern recognition result comprises a smoke source type, a smoke concentration level and a smoke diffusion speed;
and fusing the pattern recognition results to obtain smoke information.
Specifically, smoke in a target environment area is monitored in real time through a plurality of MEMS multichannel intelligent gas sensors to obtain preliminary smoke information, wherein the number of the preliminary smoke information is multiple, and the smoke in the target environment area is monitored in real time through the MEMS multichannel intelligent gas sensors. The data collected by each sensor constitutes a preliminary smoke message that contains essential characteristics of the smoke, such as the presence or absence of smoke, concentration levels, etc. Noise filtering is carried out on the preliminary smoke information, random noise and other interference signals generated in the acquisition process of the sensor are removed, and purer smoke information, namely the noise-removed smoke information, is obtained. Noise filtering may be implemented by digital signal processing techniques, such as using low pass filters, median filtering, and the like to reduce the effects of noise. The method comprises the steps of carrying out pattern recognition on the denoising smoke information to obtain a smoke pattern recognition result, wherein the smoke pattern recognition result comprises a smoke source type, a smoke concentration level, a smoke diffusion speed, carrying out pattern recognition processing on the denoising smoke information, and recognizing key characteristics such as the smoke source type, the smoke concentration level, the smoke diffusion speed and the like by analyzing the characteristics of the smoke. Pattern recognition may be implemented using machine learning algorithms or specific pattern matching techniques, such as Support Vector Machines (SVMs), neural networks, and the like. And fusing the pattern recognition results to obtain smoke information. And finally, carrying out fusion processing on the information such as the smoke source type, the smoke concentration level, the smoke diffusion speed and the like obtained by the pattern recognition, comprehensively considering the data of all the sensors, and obtaining final smoke information. The fusion processing can be realized through a data fusion algorithm, such as Bayesian estimation, kalman filtering and other methods, so as to improve the accuracy and reliability of the information. By the mode, the system can acquire the real-time monitoring data of the smoke from the MEMS multichannel intelligent gas sensors, and accurate and reliable smoke information is obtained after a series of processing, so that support is provided for subsequent analysis and decision. The technical scheme not only improves the accuracy and reliability of smoke monitoring, but also can track the dynamic change of smoke in real time, thereby realizing the accurate prediction of the development trend of fire.
In a specific embodiment, the extracting the smoke characteristics of the smoke information by using a wavelet transformation technology to obtain smoke dynamic characteristics includes:
carrying out frequency component decomposition on the smoke information through a preset wavelet transformation technology to obtain smoke decomposition characteristics;
Extracting motion characteristics of the smoke decomposition characteristics to obtain smoke motion characteristics;
carrying out gradient calculation on smoke movement characteristics to obtain smoke gradient characteristics, wherein the smoke gradient characteristics comprise smoke flow speed characteristics and smoke diffusion direction characteristics;
performing track analysis on the smoke movement characteristics to obtain a smoke movement track;
carrying out diffusion characteristic identification and prediction on smoke based on the smoke movement track to obtain smoke diffusion characteristics;
extracting morphological characteristics of the smoke decomposition characteristics to obtain smoke boundary shape characteristics and smoke outline characteristics;
Performing time-frequency localization analysis based on the smoke boundary shape feature, the smoke outline feature and the smoke diffusion feature to obtain a smoke form change feature;
And carrying out feature fusion on the smoke movement features and the smoke form change features to obtain smoke dynamic features, wherein the smoke dynamic features comprise smoke movement speed change, smoke direction change frequency and smoke instantaneous acceleration.
The method comprises the steps of carrying out frequency component decomposition on smoke information through a preset wavelet transformation technology to obtain smoke decomposition characteristics, and carrying out processing on the smoke information through the preset wavelet transformation technology to decompose the smoke information into different frequency components to obtain the smoke decomposition characteristics. The wavelet transformation is a time-frequency analysis tool which can capture local features in smoke information, particularly time-frequency characteristics, so as to extract dynamic features of smoke. Through wavelet transformation, the smoke information can be converted into coefficients at different scales and frequencies, which reflect different characteristics of the smoke. And further analyzing the smoke decomposition characteristics to extract the motion characteristics of the smoke, such as the moving speed, the moving direction and the like of the smoke. Motion feature extraction may be implemented using time series analysis or other signal processing techniques to identify dynamic changes in smoke. The smoke gradient characteristics comprise smoke flow speed characteristics and smoke diffusion direction characteristics, and the smoke gradient characteristics comprise the smoke flow speed characteristics and the smoke diffusion direction characteristics can be obtained by carrying out gradient calculation on the smoke movement characteristics. Gradient calculations can help identify the rate of change of smoke concentration over time and space, and thus infer the flow rate and direction of diffusion of smoke. And carrying out track analysis on the smoke motion characteristics to obtain a motion track of the smoke in space. Trajectory analysis can help identify the propagation path of smoke in space, which is important for predicting the extent of smoke spread. And identifying and predicting the diffusion characteristics of the smoke, such as diffusion speed, diffusion range and the like, according to the smoke motion trail. The diffuse feature recognition and prediction may be implemented using mathematical models and machine learning techniques to improve the accuracy of the prediction. The smoke boundary shape feature and the smoke outline feature can be obtained by extracting the morphological feature of the smoke decomposition feature, and are used for describing the shape and the boundary of smoke. Morphology feature extraction can help identify the shape and edges of smoke, which is critical to understanding the morphology changes of smoke. And carrying out time-frequency localized analysis based on the smoke boundary shape characteristics, the smoke outline characteristics and the smoke diffusion characteristics to obtain smoke shape change characteristics, and carrying out time-frequency localized analysis based on the smoke boundary shape characteristics, the smoke outline characteristics and the smoke diffusion characteristics to obtain smoke shape change characteristics and describe the change condition of the smoke shape along with time. Time-frequency localization analysis may be implemented using wavelet transforms or other related techniques to capture details of changes in smoke morphology over time. And carrying out feature fusion on the smoke movement features and the smoke form change features to obtain smoke dynamic features, wherein the smoke dynamic features comprise smoke movement speed change, smoke direction change frequency and smoke instantaneous acceleration. And finally, fusing the smoke movement characteristics and the smoke form change characteristics to obtain smoke dynamic characteristics, wherein the smoke dynamic characteristics comprise key characteristics such as smoke movement speed change, smoke direction change frequency, smoke instantaneous acceleration and the like. Feature fusion can be realized through a data fusion algorithm so as to improve the accuracy and reliability of features. In this way, the system can extract the dynamic characteristics of the smoke from the smoke information by using wavelet transformation technology, and provides important basis for subsequent risk assessment and early warning. The technical scheme not only improves the accuracy and reliability of smoke monitoring, but also can track the dynamic change of smoke in real time, thereby realizing the accurate prediction of the development trend of fire.
In a specific embodiment, the calculating the density statistics of the dynamic characteristics of the smoke to obtain a smoke density distribution area includes:
normalizing the dynamic characteristics of the smoke to obtain normalized smoke data;
performing spatial rasterization 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;
Carrying out Gaussian kernel density calculation on the segmented time window smoke data to obtain time-varying smoke space density distribution;
smoothing the time-varying smoke space density distribution to obtain smoothed density data;
Detecting a density peak value of the smooth density data to obtain a density peak region, wherein the density peak region comprises a density peak value and a density distribution region corresponding to the density peak value;
Performing density similarity clustering analysis on the density distribution area based on the density peak value through a preset density clustering algorithm to obtain a similarity smoke density distribution area;
and acquiring a spatial distribution area of the target environment area, and mapping the similar smoke density distribution area based on the spatial distribution area to obtain a smoke density distribution area.
Specifically, the dynamic characteristics of the smoke are normalized to obtain normalized smoke data;
And carrying out normalization processing on the dynamic characteristics of the smoke so as to eliminate the influence of data dimension among different sensors and obtain normalized smoke data. Normalization can be achieved by minimum and maximum normalization, Z-score normalization, etc. to ensure that all sensors' data are compared at the same scale. And performing spatial rasterization on the normalized smoke data according to the layout of the target environment area, namely dividing the monitoring area into a plurality of grids to obtain the rasterized smoke data. The spatial rasterization process may be implemented using Geographic Information System (GIS) technology to facilitate subsequent data analysis and visualization. And dividing the rasterized smoke data according to a certain time interval to obtain smoke data in different time windows, namely, the smoke data of the segmented time window. The time window segmentation may be performed based on a fixed time interval or event driven manner to accommodate different application scenarios. And estimating the Gaussian kernel density of the segmented time window smoke data to reflect the density distribution of the smoke at different time and space positions, so as to obtain the time-varying smoke space density distribution. Gaussian kernel density calculations may be implemented using non-parametric probability density estimation methods to capture details of the smoke density distribution. Smoothing the time-varying smoke space density distribution to obtain smoothed density data; smoothing the spatial density distribution of the time-varying smoke to remove noise effects and obtain smoother density data. The smoothing process may be implemented using moving average, gaussian smoothing, etc. methods to improve the quality of the density data. And detecting the smooth density data to find out the area with the highest smoke density, namely the density peak area comprising the density peak and the density distribution area corresponding to the density peak. density peak detection may be achieved using local maxima searching, watershed algorithms, etc. to identify peak areas of smoke density. And carrying out density similar clustering analysis on the density distribution areas based on the density peak values through a preset density clustering algorithm to obtain the density distribution areas of the similar smoke, and carrying out clustering analysis on the density distribution areas according to the density peak values by using the preset density clustering algorithm to obtain the density distribution areas of the similar smoke. The density clustering algorithm may be implemented using DBSCAN, OPTICS, or the like, to identify the aggregate region of the smoke density distribution. And acquiring a spatial distribution area of the target environment area, and mapping the similar smoke density distribution area based on the spatial distribution area to obtain a smoke density distribution area. And acquiring the space distribution area information of the target environment area, and mapping the similarity smoke density distribution area back to the actual space position to obtain the smoke density distribution area. The mapping process may be implemented using GIS technology to ensure that the smoke density distribution area matches the actual environment. In this way, the system can finally obtain the density distribution condition of the smoke in the target environment area through a series of data processing steps from the smoke dynamic characteristics, and an important basis is provided for subsequent risk assessment and early warning. The technical scheme not only improves the accuracy and reliability of smoke monitoring, but also can track the dynamic change of smoke in real time, thereby realizing the accurate prediction of the development trend of fire.
In a specific embodiment, the dynamically drawing the smoke density distribution area based on the smoke dynamic feature generates a dynamic smoke density distribution thermodynamic diagram, including:
Identifying a sparse region in the smoke density distribution region, and performing spatial interpolation calculation on the sparse region by adopting a preset Kriging interpolation method to obtain a full smoke density region;
Acquiring an unobserved region of the monitored region, and predicting the smoke density of the unobserved region based on the full smoke density region by an inverse distance weight algorithm to obtain the smoke density of the unobserved region;
carrying out region combination on the unobserved region corresponding to the smoke density of the unobserved region and the full smoke density region to obtain a combined smoke density region;
performing color mapping on the combined fog density areas by adopting different colors to generate a static fog concentration thermodynamic diagram;
generating a thermodynamic diagram of static smoke concentration by using a color gradient mapping technology to obtain a thermodynamic diagram of smoke concentration in each time period;
dynamically drawing the smoke concentration thermodynamic diagram of each time period based on the smoke dynamic characteristics based on an animation generation algorithm, and generating a dynamic smoke concentration distribution thermodynamic diagram or;
And carrying out time sequence combination and dynamic drawing on the smoke concentration thermodynamic diagrams of each time period based on the smoke dynamic characteristics through a frame animation technology, and generating a dynamic smoke concentration distribution thermodynamic diagram changing with time.
Specifically, a sparse region in the smoke density distribution region is identified, a preset Kriging interpolation method is adopted to conduct spatial interpolation calculation on the sparse region to obtain a full smoke density region, and a sparse region with less or no monitoring data in the smoke density distribution region is identified. A Kriging interpolation method (Kriging Interpolation) is utilized, which is a spatial interpolation method based on a statistical principle and is used for carrying out spatial interpolation calculation on a sparse region so as to fill a data missing part and obtain a full smoke density region. The kriging interpolation method can estimate the value of an unknown point according to the data of surrounding known points, and the accuracy of interpolation is improved by taking the spatial correlation of the data into consideration. And obtaining the unobserved area of the monitoring area, predicting the smoke density of the unobserved area based on the full smoke density area by an inverse distance weight algorithm to obtain the smoke density of the unobserved area, and obtaining the unobserved area which is not covered by the sensor in the monitoring area. And (3) utilizing an inverse distance weighting algorithm (INVERSE DISTANCE WEIGHTING, IDW), which is a common interpolation method, and carrying out smoke density prediction on the unobserved area based on the data in the full smoke density area to obtain the smoke density of the unobserved area. The inverse distance weighting algorithm calculates an estimated value of an unknown point from the distance of surrounding known points and the value thereof, the closer the distance, the greater the contribution of the point to it. And combining the unobserved region corresponding to the smoke density of the unobserved region with the full smoke density region to form a complete smoke density distribution region, namely a combined smoke density region. The region merging ensures that the smoke density distribution data is complete and consistent throughout the monitored region. And coloring the combined fog density region by different colors according to different values of the fog density by using a color mapping technology to generate a static fog density thermodynamic diagram. Color mapping typically uses a chromatogram, such as a thermodynamic diagram, to represent regions of different smoke density, with the shades representing the levels of smoke concentration. The color gradient mapping technology is utilized to further optimize the visual effect of the static smoke concentration thermodynamic diagram, ensure the nature of color transition and obtain the smoke concentration thermodynamic diagram of each time period. The color gradient mapping can enhance the readability of the thermodynamic diagram, enabling an observer to intuitively see the trend of the smoke concentration.
And dynamically drawing the smoke concentration thermodynamic diagram of each time period based on the smoke dynamic characteristics to generate a dynamic smoke concentration distribution thermodynamic diagram, or dynamically drawing the smoke concentration thermodynamic diagram of each time period according to the smoke dynamic characteristics to generate a dynamic smoke concentration distribution thermodynamic diagram by using an animation generation algorithm. The animation generation algorithm may automatically create a continuous sequence of images from the time series data, exhibiting a change in smoke concentration over time. And carrying out time sequence combination and dynamic drawing on the smoke concentration thermodynamic diagrams of each time period based on the smoke dynamic characteristics through a frame animation technology, and generating a dynamic smoke concentration distribution thermodynamic diagram changing with time. The smoke concentration thermodynamic diagrams of each time period are combined in time sequence by utilizing a frame animation technology, and are dynamically drawn to generate a dynamic smoke concentration distribution thermodynamic diagram which changes with time. The frame animation technique enables a series of still images to be played in time sequence 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, and finally generate a dynamic smoke density distribution thermodynamic diagram which changes with time through a series of data processing and visualization steps, so as to provide important visual references for subsequent risk assessment and early warning. The technical scheme not only improves the visualization degree of smoke monitoring, but also helps a decision maker to quickly understand the situation of smoke diffusion.
In a specific embodiment, the predicting risk trend based on the smoke concentration profile thermodynamic diagram to obtain a risk prediction result includes:
extracting characteristics of smoke concentration change in a smoke concentration distribution thermodynamic diagram to obtain a dynamic change path of a high concentration region;
Analyzing the smoke diffusion direction based on the dynamic change path through a density gradient algorithm to obtain a smoke diffusion trend;
Obtaining a risk growth mode based on the smoke diffusion trend, wherein the risk growth mode comprises a rapid diffusion mode, a stable diffusion mode and a local aggregation mode;
inputting the risk growth mode into a preset risk prediction algorithm to perform risk calculation to obtain a risk growth speed;
Classifying the risk areas based on the risk growth speed to obtain risk areas with different risk levels;
Carrying out risk labeling on the risk areas with different risk levels by adopting different identifications to obtain a risk labeling thermodynamic diagram;
And carrying out risk trend prediction based on the risk labeling thermodynamic diagram to obtain a risk prediction result.
Specifically, the method comprises the steps of extracting characteristics of smoke concentration changes in a smoke concentration distribution thermodynamic diagram to obtain a dynamic change path of a high concentration region, and analyzing the smoke concentration change condition in the smoke concentration distribution thermodynamic diagram, in particular to the high concentration region. Extracting the time-varying trajectories, i.e. the dynamic paths, of these high concentration regions helps to understand the behavior pattern of smoke diffusion. And (3) analyzing the smoke diffusion direction based on the dynamic change path through a density gradient algorithm to obtain a smoke diffusion trend, and applying the density gradient algorithm, which is a mathematical method for quantifying the change rate and direction of the smoke concentration. Based on the dynamic change path of the high concentration area extracted before, the main direction of smoke diffusion is analyzed, so that the smoke diffusion trend is obtained. The risk growth mode is obtained based on the smoke diffusion trend, wherein the risk growth mode comprises a rapid diffusion mode, a stable diffusion mode and a local aggregation mode, and three main risk growth modes are identified according to the smoke diffusion trend, namely the rapid diffusion mode is that smoke diffuses rapidly to multiple directions, and the situation that a high fire or smoke diffusion risk exists is indicated. Stable diffusion mode-a slow and uniform rate of smoke diffusion, may indicate that the smoke source is stable, but still requires continuous monitoring. Local accumulation mode-smoke accumulates in certain areas without spreading out, possibly due to physical obstructions or other factors causing the smoke to accumulate in specific locations. The risk growth mode is input into a preset risk prediction algorithm to perform risk calculation to obtain a risk growth speed, and the identified risk growth mode is input into the preset risk prediction algorithm, wherein the algorithm considers the influence of various factors (such as wind speed, topography and the like) on smoke diffusion. The speed of smoke diffusion and the possible risk growth speed thereof are calculated through a risk prediction algorithm. And classifying the risk areas based on the risk growth speed to obtain risk areas with different risk levels, and dividing the monitoring area into areas with different risk levels according to the speed of the risk growth speed. These areas may include low risk areas, medium risk areas, high risk areas, etc. to facilitate taking corresponding precautions. Carrying out risk labeling on the risk areas with different risk levels by adopting different identifications to obtain a risk labeling thermodynamic diagram; and marking the areas with different risk levels by using different colors or symbols to generate a risk labeling thermodynamic diagram. The risk labeling thermodynamic diagram can clearly display which areas are in a high risk state, so that an observer can quickly understand the risk distribution of smoke diffusion. And carrying out risk trend prediction based on the risk labeling thermodynamic diagram to obtain a risk prediction result. And further analyzing risk trends possibly caused by future smoke diffusion in combination with risk labeling thermodynamic diagrams. The risk prediction results include how smoke diffusion may develop and which areas the risk will increase or decrease, which is critical to the planning of emergency response. In this way, the system can finally generate a risk prediction result through a series of data processing and analysis steps from the smoke concentration distribution thermodynamic diagram, and an important reference basis is provided for subsequent risk management and emergency response. the technical scheme not only improves the accuracy of risk prediction, but also helps a decision maker to take effective measures in time to reduce potential harm.
In a specific embodiment, analyzing the smoke diffusion direction based on the dynamic change path through a density gradient algorithm to obtain a smoke diffusion trend, including:
The smoke concentration of the dynamic change path is acquired by a multipoint sampling technology, so that the spatial distribution concentration of the smoke is obtained;
Performing density gradient calculation on the spatial distribution concentration by a gradient calculation algorithm to obtain a concentration gradient field;
Carrying out directional analysis on the concentration gradient field through a directional analysis model to obtain a direction vector of smoke diffusion;
and carrying out diffusion trend analysis based on the direction vector to obtain a smoke diffusion trend, wherein the smoke diffusion trend comprises a smoke diffusion range, and the smoke diffusion range comprises a round shape, an oval shape and an irregular shape.
In particular, in carrying out the above claims, it is first of all necessary to define the analysis of the direction of smoke diffusion, which includes several key steps, namely the acquisition of the dynamic path of variation, the determination of the spatial distribution concentration of smoke, the calculation of the density gradient field and the final analysis of the trend of smoke diffusion. Initially, the dynamic path of smoke is acquired by a density gradient algorithm. This is done by a multi-point sampling technique. The multi-point sampling technique is capable of collecting smoke concentration data at different points in space. The choice of these sampling points is typically based on the path along which the smoke may spread, the initial location of the smoke source, and environmental factors. By collecting concentration data from these sampling points, a preliminary knowledge of the spatial distribution of smoke can be obtained. These data reflect not only the static distribution of the smoke at a certain moment, but also the dynamic diffusion process of the smoke through analysis of the time series. Next, the obtained spatially distributed concentration data is processed using a gradient calculation algorithm. The core of gradient computation is to construct a concentration gradient field by spatial variation of concentration. In particular, the gradient calculation algorithm will determine how smoke spreads in space based on the concentration differences between the sampling points. By calculating these differences, a density gradient field can be constructed that reflects the diffusion rate, direction and intensity of smoke in space. In other words, the density gradient field is a vector field of smoke diffusion, the vector of each point in the field representing the rate and direction of change of the concentration of smoke at that point. After the concentration gradient field is obtained, a further effort is made to conduct a directional analysis of the gradient field by means of a directional analysis model. The purpose of the directional analysis is to determine the main direction of smoke diffusion, a step which is important for understanding and predicting the diffusion path of smoke. The directional analysis model calculates the main diffusion directions of the smoke at different positions according to vector information in the gradient field. These direction vectors not only show the direction of the smoke diffusion, but also provide information about the diffusion rate, thereby helping to predict future diffusion trends. Once the directionality analysis is complete, a more detailed diffusion trend analysis can be performed based on the direction vector. The tendency of smoke to spread is obtained by comprehensive analysis of the direction vector, which includes analysis of the spread range of smoke. The shape of the diffusion region may be regular, such as circular or elliptical, or irregular, depending on the initial state of the smoke source, the environmental conditions, and other factors that may affect smoke diffusion. By comprehensively considering the factors, a more accurate smoke diffusion range prediction can be obtained, so that guidance is provided for actual coping and processing. In practical application, the steps can help us to better understand the smoke diffusion process, particularly in the case of fire, industrial accidents and the like involving smoke diffusion, the diffusion direction and range of the smoke can be effectively predicted through the analysis steps, and time is won for taking corresponding measures. These steps may also be applied in the field of air pollution monitoring, smoke control and other environmental management. Through a density gradient algorithm and a directivity analysis model, by combining real-time data acquisition and processing, the accurate prediction and monitoring of smoke diffusion can be realized, and powerful guarantee is provided for environmental safety and public health.
The smoke monitoring method based on the MEMS multichannel intelligent gas sensor in the embodiment of the present invention is described above, and the smoke monitoring device based on the MEMS multichannel intelligent gas sensor in the embodiment of the present invention is described below, referring to fig. 2, one embodiment of the smoke monitoring device based on the MEMS multichannel intelligent gas sensor in the embodiment of the present invention includes:
The monitoring module 21 is configured to monitor smoke in real time through a plurality of the MEMS multichannel intelligent gas sensors in a target environment area to obtain smoke information;
an extraction module 22, configured to extract smoke features from the smoke information by using a wavelet transform technology, so as to obtain smoke dynamic features, where the smoke features include a smoke motion feature and a smoke morphological feature;
the calculating module 23 is configured to perform density statistics calculation on the dynamic smoke features to obtain a smoke density distribution area;
a mapping module 24, configured to dynamically map the smoke density distribution area based on the smoke dynamic feature, and generate a dynamic smoke density distribution thermodynamic diagram;
the prediction module 25 is configured to perform risk trend prediction based on the smoke concentration distribution thermodynamic diagram, so as to obtain a risk prediction result;
and the early warning module 26 is configured to perform early warning if the risk prediction result exceeds a preset risk range.
In this embodiment, for specific implementation of each unit in the above system embodiment, please refer to the description in the above method embodiment, and no further description is given here.
Referring to fig. 3, a computer device is further provided in an embodiment of the present invention, and the internal structure of the computer device may be as shown in fig. 3. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected by a system bus. Wherein the computer is configured 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, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. 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 for communicating with an external terminal through a network connection. Which computer program, when being executed by a processor, carries out the above-mentioned method.
It will be appreciated by those skilled in the art that the architecture shown in fig. 3 is merely a block diagram of a portion of the architecture in connection with the present inventive arrangements and is not intended to limit the computer devices to which the present inventive arrangements are applicable.
An embodiment of the present invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the above method. It is understood that the computer readable storage medium in this embodiment may be a volatile readable storage medium or a nonvolatile readable storage medium.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium provided by the present invention and used in embodiments may include non-volatile and/or volatile memory. The nonvolatile 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 external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one does not exclude the presence of other like elements in a process, apparatus, article, or method that comprises the element.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes using the descriptions and drawings of the present invention or direct or indirect application in other related technical fields are included in the scope of the present invention.