Multi-sensor fusion smoke monitoring system and method based on 4G communication

Abstract

The invention relates to the technical field of information transmission, and provides a multi-sensor fusion smoke monitoring system and method based on 4G communication, and the method comprises the steps: obtaining temperature data, humidity data and smoke concentration data of an environment, carrying out the time sequence verification, obtaining a time sequence synchronization result, and carrying out the analysis; obtaining temperature, humidity and smoke concentration verification results, and performing different pretreatments to obtain smoke monitoring indexes; transmitting the smoke monitoring index to a central processing module for evaluation to obtain a smoke concentration evaluation result; the smoke concentration evaluation result is transmitted to a remote monitoring center by using the 4G communication module, and smoke monitoring and early warning are performed on the environment based on the smoke concentration evaluation result in the remote monitoring center. By adopting a sensor technology and a 4G communication system, full-process automation from data acquisition, verification, processing and final real-time monitoring and early warning is realized, and the problems of data transmission delay and information loss under the conditions of wide coverage range and large data volume are improved.

Description

Technical Field

[0001] The present application relates to the technical field of information transmission, and in particular to a multi-sensor fusion smoke monitoring system and method based on 4G communication. Background Art

[0002] With the rapid development of industrialization and urbanization, environmental pollution problems have become increasingly prominent, among which air quality monitoring has become an important task of public safety. As an important component of air pollution, smoke monitoring is not only related to fire prevention, but also directly affects people's quality of life and health. Therefore, it is particularly important to develop an effective smoke monitoring system. At present, multi-sensor fusion technology has been widely used in smoke monitoring, which improves the accuracy and reliability of monitoring by integrating multiple sensors.

[0003] Among the relevant technical means, sensors for multiple environmental parameters such as temperature, humidity and smoke concentration are usually combined. The data from these sensors are centrally processed to analyze and determine whether there is a potential danger caused by smoke in the environment. Through real-time data collection and central processing, the system can respond quickly and accurately locate the area where smoke occurs, while providing necessary early warning information. This technical layout makes smoke monitoring more comprehensive, ensuring high sensitivity and low false alarm rate.

[0004] Regarding the above technical solution, although the response speed and accuracy of the monitoring system can be greatly improved by integrating multiple sensors and using a central processing unit to process and analyze data in real time, there are problems of data transmission delay and information loss during data transmission and processing, especially when the coverage area is wide and the data volume is large. Summary of the invention

[0005] In order to improve the problems of data transmission delay and information loss when the coverage area is wide and the data volume is large, the present application provides a multi-sensor fusion smoke monitoring system and method based on 4G communication.

[0006] The present invention provides a multi-sensor fusion smoke monitoring method based on 4G communication, which is applied to a smoke monitoring system. The smoke monitoring system includes a central processing module and a 4G communication module. The smoke monitoring method includes: acquiring temperature data, humidity data and smoke concentration data of an environment, performing time series verification on the temperature data, the humidity data and the smoke concentration data to obtain a time series synchronization result; parsing the time series synchronization result to obtain a temperature verification result, a humidity verification result and a smoke concentration verification result, performing different preprocessing on the temperature verification result, the humidity verification result and the smoke concentration verification result respectively to obtain temperature processing data, humidity processing data and smoke concentration processing data; fusing the temperature processing data with the humidity processing data to obtain comprehensive environmental data, performing comprehensive analysis on the comprehensive environmental data and the smoke concentration verification result to obtain a smoke monitoring index; transmitting the smoke monitoring index to the central processing module, evaluating the smoke monitoring index based on a preset smoke evaluation model to obtain a smoke concentration evaluation result; using the 4G communication module to transmit the smoke concentration evaluation result to a remote monitoring center, and performing smoke monitoring and early warning on the environment based on the smoke concentration evaluation result at the remote monitoring center.

[0007] As a preferred solution, the step of performing timing verification on the temperature data, the humidity data and the smoke concentration data to obtain a timing synchronization result includes: collecting the temperature data of the environment through a temperature sensor, and recording the timestamp of the temperature collection, to obtain the temperature collection time and the temperature parameter corresponding to the temperature collection time; collecting the humidity data of the environment through a humidity sensor, and recording the timestamp of the humidity collection, to obtain the humidity collection time and the humidity parameter corresponding to the humidity collection time; collecting the smoke concentration data of the environment through a smoke sensor, and recording the timestamp of the smoke concentration collection, to obtain the smoke concentration collection time and the smoke concentration parameter corresponding to the smoke concentration collection time; and verifying the temperature collection time, the humidity collection time and the smoke concentration The acquisition time is time-series verified to obtain time-series difference data; the temperature parameter, the humidity parameter and the smoke concentration parameter are input into a preset time trend change model to perform time change deduction to obtain temperature deduction results, humidity deduction results and smoke concentration deduction results; the temperature data is time-series synchronized using the difference data and the temperature deduction result to obtain temperature synchronization results; the humidity data is time-series synchronized using the difference data and the humidity deduction result to obtain humidity synchronization results; the smoke concentration data is time-series synchronized using the difference data and the smoke concentration deduction result to obtain smoke concentration synchronization results; the temperature synchronization results, the humidity synchronization results and the smoke concentration synchronization results are comprehensively analyzed to obtain time-series synchronization results.

[0008] As a preferred solution, the steps of performing different preprocessing on the temperature verification result, the humidity verification result and the smoke concentration verification result to obtain temperature processing data, humidity processing data and smoke concentration processing data respectively include: performing noise filtering on the temperature verification result to obtain noise-filtered data, and normalizing the noise-filtered data to obtain temperature processing data; performing multi-scale decomposition on the humidity verification result to extract humidity features of different scales, and performing feature selection on the humidity features after multi-scale decomposition to obtain humidity processing data; performing cluster analysis on the smoke concentration verification result to separate different types of smoke feature data, and using the cluster analysis results to reconstruct the different types of smoke feature data to obtain smoke processing data.

[0009] As a preferred scheme, the steps of performing noise filtering on the temperature verification result to obtain noise-filtered data, and normalizing the noise-filtered data to obtain temperature-processed data include: performing noise filtering on the temperature verification result by wavelet transform to obtain low-frequency components and high-frequency components, smoothing the low-frequency components by using a low-pass filter to obtain smoothed low-frequency components, performing noise reduction on the high-frequency components based on the smoothed low-frequency components to obtain noise-filtered temperature data, and normalizing the noise-filtered temperature data to obtain temperature-processed data.

[0010] As a preferred scheme, the step of performing multi-scale decomposition processing on the humidity verification result, extracting humidity features of different scales, performing feature selection on the humidity features after multi-scale decomposition, and obtaining humidity processing data includes: performing multi-scale decomposition processing on the humidity verification result by empirical mode decomposition to obtain intrinsic mode functions of different scales; screening the intrinsic mode functions according to energy entropy to obtain main intrinsic mode function components, performing frequency domain analysis on the main intrinsic mode function components by fast Fourier transform to extract frequency features; performing correlation analysis on the frequency features and the humidity verification result to obtain relevant features, performing feature selection on the relevant features to obtain humidity processing data.

[0011] As a preferred embodiment, the step of performing cluster analysis on the smoke concentration verification results to separate different types of smoke characteristic data, and using the cluster analysis results to reconstruct the different types of smoke characteristic data to obtain smoke processing data includes: performing preliminary cluster analysis on the smoke concentration verification results by K-means clustering to obtain a first cluster center; performing secondary clustering on the preliminary cluster center by iterative self-organizing data analysis to obtain a second cluster center; performing statistical analysis on the second cluster centers to obtain a feature vector of each of the second cluster centers; inputting the feature vector into a support vector machine classifier for classification and reconstruction to obtain classification and reconstruction results and different types of smoke characteristic data; and reconstructing the different types of smoke characteristic data by using the feature vector and the classification and reconstruction results to obtain smoke concentration processing data.

[0012] As a preferred scheme, the step of fusing the temperature processing data with the humidity processing data to obtain environmental comprehensive data, and comprehensively analyzing the environmental comprehensive data and the smoke concentration verification results to obtain smoke monitoring indicators includes: weighted fusion of the temperature processing data and the humidity processing data to obtain fused data, performing principal component analysis on the fused data to extract main components; performing correlation analysis on the main components and the smoke concentration verification results to find key indicators, calculating the key indicators by entropy weight method to obtain the weight of each indicator; adjusting the fused data using the weight to obtain environmental comprehensive data, and performing cluster analysis on the environmental comprehensive data and the smoke concentration verification results to obtain smoke monitoring indicators.

[0013] The present application provides a multi-sensor fusion smoke monitoring system based on 4G communication, the smoke monitoring system includes a central processing module and a 4G communication module, and also includes: an acquisition unit, used to acquire temperature data, humidity data and smoke concentration data of the environment, perform time sequence verification on the temperature data, the humidity data and the smoke concentration data, and obtain a time sequence synchronization result; a parsing unit, used to parse the time sequence synchronization result, obtain a temperature verification result, a humidity verification result and a smoke concentration verification result, respectively perform different preprocessing on the temperature verification result, the humidity verification result and the smoke concentration verification result, and obtain temperature processing data, humidity processing data The 4G communication module is used to transmit the smoke concentration evaluation result to the remote monitoring center, and the remote monitoring center performs smoke monitoring and early warning based on the smoke concentration evaluation result.

[0014] Compared with the prior art, this application has the following beneficial effects: timely response and high accuracy. The high accuracy and comprehensiveness of the data are ensured by the integrated multi-sensors, and the 4G communication technology provides fast and reliable data transmission capabilities, enabling the remote monitoring center to receive and process alarms in real time; at the same time, the application of data processing algorithms, including timing verification, data preprocessing, fusion and evaluation, not only improves the monitoring accuracy and the accuracy of early warning response, but also improves the efficiency and effect of environmental monitoring, especially in improving the real-time and accuracy of smoke monitoring, and improving the problems of data transmission delay and information loss in the case of wide coverage and large data volume. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0016] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0017] Figure 1 It is a flow chart of a multi-sensor fusion smoke monitoring method based on 4G communication provided by an embodiment of the present invention; Figure 2 It is a schematic block diagram of the structure of a multi-sensor fusion smoke monitoring system based on 4G communication provided by an embodiment of the present invention.

[0018] Description of reference numerals: 10. Multi-sensor fusion smoke monitoring system based on 4G communication; 11. Acquisition unit; 12. Parsing unit; 13. Analysis unit; 14. Evaluation unit; 15. Transmission unit. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0021] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0022] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0023] The technical solution of the present invention is further described below with reference to the accompanying drawings and through specific implementation methods.

[0024] Embodiment 1: like Figure 1 As shown, the multi-sensor fusion smoke monitoring method based on 4G communication provided in an embodiment of the present application is applied to a smoke monitoring system, and the smoke monitoring system includes a central processing module and a 4G communication module, including steps S100 to S500.

[0025] Step S100: Acquire temperature data, humidity data and smoke concentration data of the environment, perform time sequence verification on the temperature data, humidity data and smoke concentration data, and obtain a time sequence synchronization result.

[0026] In this step, environmental data is collected through a distributed sensor network. These sensors periodically send data to the data collection center; specifically, the data center uses a built-in timing verification algorithm to check the timestamp of each set of data to ensure the synchronization and timeliness of the data.

[0027] For example, assuming that the temperature data, humidity data, and smoke concentration data at a certain moment are sent by different sensors at the same time point, timing verification will confirm whether these data are actually recorded at the same predetermined time point to ensure data consistency and availability.

[0028] Step S200, analyzing the timing synchronization result to obtain temperature verification result, humidity verification result and smoke concentration verification result, respectively performing different preprocessing on the temperature verification result, humidity verification result and smoke concentration verification result to obtain temperature processing data, humidity processing data and smoke concentration processing data.

[0029] In this step, different preprocessing algorithms are applied to each validation result obtained to optimize the data quality; specifically, for the temperature data, it includes outlier removal and temperature smoothing, while the humidity data is calibrated and the smoke concentration data is subjected to sensitivity adjustment.

[0030] For example, if the temperature verification results show that the data points are in an abnormal temperature zone, these data will be excluded. If any deviations are detected in the humidity data, they will be corrected using historical data. The smoke data will be corrected through threshold settings to determine what level of smoke concentration is considered an alarm state.

[0031] Step S300: Fusing the temperature processing data with the humidity processing data to obtain comprehensive environmental data, and comprehensively analyzing the comprehensive environmental data and the smoke concentration verification result to obtain smoke monitoring indicators.

[0032] In this step, temperature and humidity data are combined through data fusion technology to form a comprehensive view of the environmental status; specifically, a weighted algorithm is used to optimize the impact of temperature and humidity data to improve the accuracy of smoke monitoring.

[0033] For example, the adjusted temperature and humidity data are used to calculate a comprehensive environmental index that affects smoke behavior, making smoke monitoring more accurate and timely.

[0034] Step S400: transmitting the smoke monitoring index to the central processing module, evaluating the smoke monitoring index based on a preset smoke evaluation model, and obtaining a smoke concentration evaluation result.

[0035] In this step, the data is sent to the central processing unit via wireless communication for further analysis; specifically, the smoke monitoring indicators are evaluated using a machine learning algorithm to predict the smoke development trend.

[0036] For example, a smoke concentration assessment model predicts future smoke events based on historical data, assessing their severity and duration.

[0037] Step S500: using the 4G communication module to transmit the smoke concentration assessment result to the remote monitoring center, and the remote monitoring center performs smoke monitoring and early warning on the environment based on the smoke concentration assessment result.

[0038] In this step, the smoke concentration assessment results are sent to the monitoring center in real time via the 4G communication network; specifically, the monitoring center uses these data to activate the early warning system and notify relevant personnel and agencies.

[0039] For example, if dangerous levels of smoke concentration are detected, the monitoring center will automatically trigger an alarm and send notifications to the public and emergency services via mobile apps, text messages and emails.

[0040] In this embodiment, the temperature data, humidity data and smoke concentration data of the environment are obtained through relevant sensors. These data are verified by time series to ensure the consistency of the information collected by each sensor in time, thereby improving the reliability of the data. Subsequently, the time series synchronization results are parsed to obtain the temperature verification results, humidity verification results and smoke concentration verification results respectively. This step ensures the accuracy of each data point. After that, the data is preprocessed, and the temperature verification results, humidity verification results and smoke concentration verification results are processed differently to obtain temperature processing data, humidity processing data and smoke concentration processing data. The temperature processing data and the humidity processing data are fused to generate comprehensive environmental data. This data fusion helps to evaluate the environmental state more comprehensively. The comprehensive environmental data and the smoke concentration verification results are comprehensively analyzed to formulate smoke monitoring indicators. These indicators will be transmitted to the central processing module, where the smoke monitoring indicators are evaluated based on the preset smoke evaluation model to obtain the smoke concentration evaluation results. Finally, the smoke concentration evaluation results are transmitted to the remote monitoring center using the 4G communication module, and the remote monitoring center will perform smoke monitoring and early warning of the environment based on these evaluation results.

[0041] Through the multi-sensor fusion method of this embodiment, environmental data can be obtained from multiple dimensions, which enhances the monitoring system's ability to identify potential hazards such as smoke and improves the accuracy and reliability of monitoring. Secondly, through the use of the 4G communication module, the rapid and real-time transmission of smoke monitoring data is ensured, which greatly improves the efficiency of data transmission and the response speed of remote monitoring, which is crucial for rapid response in emergency situations. In addition, timing verification and data fusion processing improve the integration of data and the depth of analysis, making the assessment of smoke concentration more accurate, thereby optimizing the triggering mechanism of the early warning system, reducing the risk of false alarms and missed alarms, and improving the problems of data transmission delays and information loss in cases of wide coverage and large data volumes.

[0042] Embodiment 2: In step S100, the temperature data of the environment is collected by the temperature sensor, and the timestamp of the temperature collection is recorded to obtain the temperature collection time and the temperature parameters corresponding to the temperature collection time; the humidity data of the environment is collected by the humidity sensor, and the timestamp of the humidity collection is recorded to obtain the humidity collection time and the humidity parameters corresponding to the humidity collection time; the smoke concentration data of the environment is collected by the smoke sensor, and the timestamp of the smoke concentration collection is recorded to obtain the smoke concentration collection time and the smoke concentration parameters corresponding to the smoke concentration collection time.

[0043] Data collection is performed using high-precision sensors and synchronized timestamp recording methods; specifically, each sensor is equipped with a built-in clock or receives a time synchronization signal from a central system to ensure the accuracy and consistency of timestamps.

[0044] For example, temperature sensors are set up in different locations in industrial areas, and the temperature data with timestamps are sent to the data collection center in real time via wireless networks, thus ensuring the timeliness and accurate matching of the data.

[0045] The temperature collection time, humidity collection time and smoke concentration collection time are time-series verified to obtain time-series difference data.

[0046] The consistency of the timestamps of each data point is verified through time synchronization technology; specifically, the timestamps of temperature, humidity and smoke data are compared, the time difference is calculated, and the synchronization of data collection is evaluated.

[0047] For example, if the timestamps of temperature and humidity data differ by more than a preset threshold (e.g., 1 second), the system will mark these data as inconsistent in timing and require further processing or exclusion.

[0048] The temperature parameters, humidity parameters and smoke concentration parameters are input into the preset time trend change model to perform time change deduction to obtain temperature deduction results, humidity deduction results and smoke concentration deduction results.

[0049] Advanced data analysis models are used to predict trends and deduce changes in environmental data; specifically, machine learning models are used to analyze historical data and predict data trends in the future.

[0050] For example, temperature and humidity data can be used to predict trends over the next few hours using a time series analysis model, while smoke concentration data can be used to determine if there is an increased risk using a trend analysis model.

[0051] The temperature data is synchronized with the difference data and the temperature deduction result to obtain the temperature synchronization result; the humidity data is synchronized with the difference data and the humidity deduction result to obtain the humidity synchronization result; the smoke concentration data is synchronized with the difference data and the smoke concentration deduction result to obtain the smoke concentration synchronization result.

[0052] The algorithm is used to adjust the time deviation of the data to ensure the consistency of each data point; specifically, the difference adjustment algorithm is used to correct the time deviation of data collection so that all sensor data are aligned on the time axis.

[0053] For example, if there is a slight deviation between the timestamps of the temperature data and the humidity data, the data can be adjusted through interpolation or other mathematical methods to ensure that the synchronized data reflects the same point in time.

[0054] The temperature synchronization results, humidity synchronization results and smoke concentration synchronization results are comprehensively analyzed to obtain the time series synchronization results.

[0055] The consistency and reliability of data after synchronization are evaluated through comprehensive analysis methods; specifically, statistical analysis techniques are used to evaluate the quality of the synchronized data to ensure the accuracy of the data before the next step of processing.

[0056] For example, statistical methods such as standard deviation and variance analysis are used to ensure that the synchronized data have no statistically significant differences and are suitable for further data processing and analysis.

[0057] In this step, the high accuracy and time consistency of data collection are ensured by using sensor technology and timestamp synchronization. The time trend change model and difference adjustment algorithm are used to effectively deduce and synchronize the temperature, humidity and smoke concentration data, ensuring the accuracy and real-time performance of data analysis. In addition, the reliability of data synchronization results is further ensured through comprehensive analysis technology, providing a solid data foundation for subsequent environmental monitoring and early warning. Overall, this embodiment improves the efficiency and accuracy of the environmental monitoring system through technological innovation, enhances the ability to respond to complex environmental changes, and significantly improves the practicality of the system and the timeliness of early warning.

[0058] In step S200, the steps of performing different preprocessing on the temperature verification result, the humidity verification result and the smoke concentration verification result to obtain temperature processing data, humidity processing data and smoke concentration processing data respectively include: The temperature verification result is subjected to noise filtering to obtain noise-filtered data, and the noise-filtered data is normalized to obtain temperature-processed data.

[0059] The temperature data is filtered for noise by wavelet transform method; specifically, the temperature verification results are first decomposed into low-frequency and high-frequency components, the low-frequency components are smoothed by low-pass filter, and then the high-frequency components are denoised based on the smoothed low-frequency components to remove high-frequency noise. After noise filtering, the data is normalized so that different temperature data can be compared and analyzed under a unified standard.

[0060] For example, if the temperature data contains noise caused by sensor errors or environmental interference, these noises are smoothed and removed through wavelet transform processing to obtain more accurate temperature data, which is then normalized to ensure the consistency of data units in subsequent data analysis.

[0061] The humidity verification results are subjected to multi-scale decomposition processing to extract humidity features at different scales. Feature selection is performed on the humidity features after multi-scale decomposition to obtain humidity processing data.

[0062] The humidity verification results are decomposed at multiple scales by the empirical mode decomposition (EMD) method, and the intrinsic mode functions (IMFs) of different scales are extracted. Specifically, the EMD method can decompose the humidity data into components at multiple scales, and analyze each component to extract useful humidity features. The most representative intrinsic mode function components are selected by energy entropy analysis, and further frequency domain analysis is performed to extract the frequency features in the humidity data.

[0063] For example, if the humidity data changes are very complex, the EMD method can decompose these changes into patterns at different levels, so that the system can optimize the analysis of humidity characteristics at different scales and select the most representative features for subsequent processing.

[0064] Cluster analysis is performed on the smoke concentration verification results to separate different types of smoke feature data. The cluster analysis results are used to reconstruct different types of smoke feature data to obtain smoke processing data.

[0065] The K-means clustering algorithm is used to perform a preliminary cluster analysis on the smoke concentration verification results to obtain the first cluster center, and then the smoke data is divided into different categories; specifically, the K-means algorithm groups all smoke concentration data according to their numerical characteristics and generates the cluster center of each group. Then, the iterative self-organizing data analysis method is used to perform secondary clustering on the first cluster center to obtain a more refined classification result.

[0066] For example, if the smoke concentration in the data exists in multiple different states (such as light smoke, moderate smoke, and heavy smoke), K-means clustering can be used to identify these different categories and perform further analysis and processing based on these classification results.

[0067] The steps of performing noise filtering on the temperature verification result to obtain noise-filtered data, and normalizing the noise-filtered data to obtain temperature-processed data include: The temperature verification result is subjected to noise filtering by wavelet transform to obtain low-frequency and high-frequency components. The low-frequency component is smoothed by a low-pass filter to obtain a smoothed low-frequency component. The high-frequency component is subjected to noise reduction based on the smoothed low-frequency component to obtain temperature data after noise filtering. The temperature data after noise filtering is normalized to obtain temperature processing data.

[0068] The temperature data is decomposed into low-frequency and high-frequency components through wavelet transform; specifically, the low-frequency component represents the trend of temperature change, while the high-frequency component contains noise. The low-frequency component is smoothed using a low-pass filter to make the trend part in the temperature data smoother and remove the high-frequency noise component. Then, based on the smoothed low-frequency component, the high-frequency component is denoised to remove interference signals, and finally the denoised temperature data is normalized to ensure that the data is further analyzed at a unified scale.

[0069] For example, if the temperature sensor is subject to external electromagnetic interference, the temperature data will show obvious fluctuations. After wavelet transform, the trend and noise of the temperature signal can be effectively distinguished, the noise part can be removed, and more stable and accurate temperature data can be obtained.

[0070] The steps of performing multi-scale decomposition processing on the humidity verification result, extracting humidity features at different scales, performing feature selection on the humidity features after multi-scale decomposition, and obtaining humidity processing data include: The humidity verification results are decomposed into multiple scales through empirical mode decomposition to obtain intrinsic mode functions of different scales.

[0071] The humidity verification results are decomposed into multiple scales by empirical mode decomposition (EMD) to obtain the intrinsic mode functions (IMFs) at multiple scales. Specifically, the EMD method can automatically decompose the humidity data into a series of components with different time scales, which represent the oscillation characteristics of the humidity data at different scales.

[0072] For example, humidity data contains seasonal changes and short-term fluctuations. EMD can be used to separate the different scale characteristics of long-term changes (such as seasonal humidity changes) and short-term fluctuations (such as daily fluctuations), which is helpful for subsequent analysis.

[0073] The intrinsic mode functions are screened according to energy entropy to obtain the main intrinsic mode function components, and the main intrinsic mode function components are analyzed in the frequency domain using fast Fourier transform to extract the frequency characteristics.

[0074] The energy entropy is used to analyze the energy distribution of each intrinsic mode function and select the modal components with larger energy. Next, the selected intrinsic mode functions are analyzed in the frequency domain using the fast Fourier transform (FFT) to extract the frequency characteristics of humidity changes. Frequency domain analysis can help identify periodic changes in humidity data, such as periodic humidity fluctuations or sudden events.

[0075] For example, the energy entropy method can ensure that the most important patterns in the humidity data are selected, and then FFT is applied to perform frequency analysis on it to extract relevant frequency features, further enhancing the analysis effect of the humidity data.

[0076] The frequency features and the humidity verification results are correlated and analyzed to obtain the relevant features. The relevant features are selected to obtain the humidity processing data.

[0077] The extracted humidity frequency features are correlated with the original humidity data to determine which frequency features are most closely related to humidity changes. Based on the correlation analysis results, the most influential features for humidity data analysis are selected and used as the final processing results of the humidity data.

[0078] For example, analyzing the correlation between humidity data and other environmental factors such as temperature and air pressure can help select features that are closely related to humidity changes and further optimize the processing of humidity data.

[0079] The steps of performing cluster analysis on the smoke concentration verification results to separate different types of smoke feature data, and reconstructing different types of smoke feature data using the cluster analysis results to obtain smoke processing data include: A preliminary cluster analysis of the smoke concentration verification results was performed using K-means clustering to obtain the first cluster center.

[0080] The smoke density verification data is preliminarily clustered using the K-means clustering algorithm. Specifically, the K-means algorithm divides the smoke density data into several categories based on their numerical characteristics and calculates the cluster center of each category. The cluster center represents the "typical" value in the data and can help identify the main patterns in the data.

[0081] For example, if the trend of smoke concentration changes is complex, the K-means algorithm can automatically classify these changes into several categories, such as low concentration, medium concentration, and high concentration, thereby helping to identify different types of smoke environments.

[0082] The preliminary cluster centers are clustered again using iterative self-organizing data analysis to obtain the second cluster centers.

[0083] The iterative self-organizing data analysis (ISODATA) method is used to optimize the initial clustering results obtained by the K-means algorithm, further refine the cluster center, and obtain a more accurate second cluster center. The ISODATA method adjusts the number of clusters according to the distribution of the data and optimizes each cluster center in an iterative manner to ensure the accuracy of the clustering results.

[0084] For example, if the smoke concentration changes slightly in some areas, the ISODATA method can automatically subdivide the clustering results of these areas to obtain a more accurate smoke concentration distribution.

[0085] Statistical analysis is performed on the second cluster centers to obtain the characteristic vector of each second cluster center.

[0086] The smoke concentration data of the second cluster center is statistically analyzed to calculate the eigenvectors of each cluster center. These eigenvectors contain the key features of the smoke concentration data for subsequent classification and reconstruction.

[0087] For example, by statistically analyzing the smoke data of the second cluster center, the typical smoke characteristics of each area (such as concentration variation range, peak time, etc.) can be obtained. These characteristics will help to more accurately identify and process different types of smoke.

[0088] The feature vector is input into the support vector machine classifier for classification and reconstruction, and the classification and reconstruction results and different types of smoke feature data are obtained.

[0089] The feature vector of the smoke data is input into the support vector machine (SVM) classifier, and the SVM classifies the smoke data into different types of categories based on the feature vector. The support vector machine classifier can learn how to distinguish different types of smoke feature data through the feature vector of the training set.

[0090] For example, the support vector machine classifier can identify different types of smoke concentrations (such as heavy smoke and light smoke) and accurately classify them to generate more detailed classification results.

[0091] The characteristic vectors and classification reconstruction results are used to reconstruct different types of smoke feature data to obtain smoke concentration processing data.

[0092] According to the classification results, different types of smoke feature data are reconstructed in combination with feature vectors to generate processed data that conforms to the smoke concentration verification results. Through classification and reconstruction, different types of smoke concentration data can be effectively integrated and optimized.

[0093] For example, for high-concentration smoke and low-concentration smoke, the system can reconstruct them according to the classification results of the feature vector so as to accurately reflect the actual conditions of different smoke in subsequent monitoring and early warning.

[0094] In step S300, weighted fusion is performed on the temperature processing data and the humidity processing data to obtain fused data, and principal component analysis is performed on the fused data to extract the main components.

[0095] The temperature processing data and humidity processing data are weighted and integrated through the weighted fusion algorithm; specifically, the weight is assigned according to the importance of each data using the weighted average method, and the fused data will be analyzed as comprehensive environmental data. Subsequently, the principal component analysis (PCA) method is used to reduce the dimension of the fused data and extract the most representative components for subsequent monitoring indicator calculations.

[0096] For example, in the weighted fusion process, if humidity has a greater impact on smoke concentration, the weight of the humidity data will increase accordingly. In principal component analysis, the system will analyze the covariance matrix of the fused data and extract the principal component most relevant to the ambient smoke concentration, thereby effectively reducing the computational complexity.

[0097] The main components are correlated with the smoke concentration verification results to find out the key indicators, and the entropy weight method is used to calculate the key indicators to obtain the weight of each indicator.

[0098] The relationship between environmental factors such as temperature and humidity and smoke concentration is evaluated through correlation analysis. Specifically, the Pearson correlation coefficient or other correlation indicators are used to quantify the mutual influence between environmental factors and smoke concentration. Subsequently, the entropy weight method is used to calculate the weight of each indicator to determine the contribution of each environmental factor to smoke monitoring.

[0099] For example, assuming that humidity and temperature have a greater impact on smoke concentration in a certain area, the entropy weight method will calculate the entropy value of each indicator and give a corresponding weight to guide subsequent data processing and analysis.

[0100] The fused data are adjusted using weights to obtain comprehensive environmental data, and cluster analysis is performed on the comprehensive environmental data and the smoke concentration verification results to obtain smoke monitoring indicators.

[0101] The weighted fusion data is then used to construct comprehensive environmental data. Specifically, the comprehensive environmental index is calculated by combining multidimensional data such as temperature, humidity, and smoke concentration, which serves as the basis for comprehensive assessment of environmental conditions. Then, cluster analysis methods (such as K-means or DBSCAN) are used to classify the comprehensive environmental data and smoke concentration verification results to obtain the final smoke monitoring index.

[0102] For example, in cluster analysis, if the comprehensive environmental data and smoke concentration verification results show obvious regularity, the system will automatically classify the data and generate different monitoring levels (such as low concentration, medium concentration and high concentration) to provide a basis for decision-making.

[0103] Embodiment 3: like Figure 2As shown, the present application also provides a multi-sensor fusion smoke monitoring system 10 based on 4G communication. The smoke monitoring system includes a central processing module and a 4G communication module, and also includes an acquisition unit 11, a parsing unit 12, an analysis unit 13, an evaluation unit 14 and a transmission unit 15.

[0104] The acquisition unit 11 is mainly used to acquire the temperature data, humidity data and smoke concentration data of the environment, perform time sequence verification on the temperature data, humidity data and smoke concentration data, and obtain a time sequence synchronization result.

[0105] The analysis unit 12 is mainly used to analyze the timing synchronization results to obtain temperature verification results, humidity verification results and smoke concentration verification results, and perform different preprocessing on the temperature verification results, humidity verification results and smoke concentration verification results to obtain temperature processing data, humidity processing data and smoke concentration processing data.

[0106] The analysis unit 13 is mainly used to fuse the temperature processing data with the humidity processing data to obtain the environmental comprehensive data, and to perform comprehensive analysis on the environmental comprehensive data and the smoke concentration verification result to obtain the smoke monitoring index.

[0107] The evaluation unit 14 is mainly used to transmit the smoke monitoring index to the central processing module, evaluate the smoke monitoring index based on a preset smoke evaluation model, and obtain a smoke concentration evaluation result.

[0108] The transmission unit 15 is mainly used to transmit the smoke concentration evaluation result to the remote monitoring center using the 4G communication module, and the remote monitoring center performs smoke monitoring and early warning on the environment based on the smoke concentration evaluation result.

[0109] In this embodiment, by integrating 4G communication technology and multi-sensor data fusion method, this smoke monitoring system can realize efficient and real-time environmental monitoring and data processing. Different units of the system work together to optimize the entire process from data collection to final warning. Specifically, the acquisition unit 11 collects key environmental parameters such as temperature, humidity and smoke concentration through a precisely synchronized multi-sensor network to ensure the timeliness and accuracy of the obtained data. Subsequently, the parsing unit 12 conducts in-depth analysis of these data, extracts key verification results, and enhances the availability and signal-to-noise ratio of the data through preprocessing. The analysis unit 13 further fuses the temperature and humidity data through an algorithm, combines the smoke concentration verification results, and comprehensively generates smoke monitoring indicators, which can more comprehensively reflect the current environmental status. With the support of the central processing module, the evaluation unit 14 uses a preset evaluation model to analyze the monitoring indicators and generates smoke concentration evaluation results, which reflect the potential smoke risk level. Finally, the transmission unit 15 uses the high speed and stability characteristics of the 4G communication module to ensure that the evaluation results are transmitted to the remote monitoring center in a timely and accurate manner to achieve rapid response and early warning. This system not only improves the accuracy and timeliness of smoke monitoring, but also greatly enhances the automation and intelligence level of environmental monitoring, providing strong technical support for public safety.

[0110] It should be noted that, technical personnel in the relevant technical field can clearly understand that, for the convenience and conciseness of description, the specific working process of the system and each unit described above can refer to the corresponding process in the aforementioned multi-sensor fusion smoke monitoring method based on 4G communication embodiment, and will not be repeated here.

[0111] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-sensor fusion smoke monitoring method based on 4G communication, applied to a smoke monitoring system, wherein the smoke monitoring system includes a central processing module and a 4G communication module, characterized in that: The smoke monitoring method comprises: Acquire temperature data, humidity data and smoke concentration data of the environment, perform time sequence verification on the temperature data, humidity data and smoke concentration data, and obtain a time sequence synchronization result; The timing synchronization result is parsed to obtain a temperature verification result, a humidity verification result, and a smoke concentration verification result, and different preprocessing is performed on the temperature verification result, the humidity verification result, and the smoke concentration verification result to obtain temperature processing data, humidity processing data, and smoke concentration processing data; The temperature processing data and the humidity processing data are fused to obtain comprehensive environmental data, and the comprehensive environmental data and the smoke concentration verification result are comprehensively analyzed to obtain a smoke monitoring index; The smoke monitoring index is transmitted to the central processing module, and the smoke monitoring index is evaluated based on a preset smoke evaluation model to obtain a smoke concentration evaluation result; The 4G communication module is used to transmit the smoke concentration assessment result to a remote monitoring center, and the remote monitoring center performs smoke monitoring and early warning on the environment based on the smoke concentration assessment result.

2. The multi-sensor fusion smoke monitoring method based on 4G communication according to claim 1 is characterized in that: The step of performing time series verification on the temperature data, the humidity data, and the smoke concentration data to obtain a time series synchronization result includes: The temperature data of the environment is collected by the temperature sensor, and the timestamp of the temperature collection is recorded to obtain the temperature collection time and the temperature parameter corresponding to the temperature collection time; the humidity data of the environment is collected by the humidity sensor, and the timestamp of the humidity collection is recorded to obtain the humidity collection time and the humidity parameter corresponding to the humidity collection time; the smoke concentration data of the environment is collected by the smoke sensor, and the timestamp of the smoke concentration collection is recorded to obtain the smoke concentration collection time and the smoke concentration parameter corresponding to the smoke concentration collection time; Performing time series verification on the temperature collection time, the humidity collection time, and the smoke concentration collection time to obtain time series difference data; Input the temperature parameter, the humidity parameter and the smoke concentration parameter into a preset time trend change model to perform time change deduction to obtain a temperature deduction result, a humidity deduction result and a smoke concentration deduction result; The temperature data is synchronized with the difference data and the temperature deduction result to obtain a temperature synchronization result; the humidity data is synchronized with the difference data and the humidity deduction result to obtain a humidity synchronization result; the smoke concentration data is synchronized with the difference data and the smoke concentration deduction result to obtain a smoke concentration synchronization result; The temperature synchronization result, the humidity synchronization result and the smoke concentration synchronization result are comprehensively analyzed to obtain a timing synchronization result.

3. The multi-sensor fusion smoke monitoring method based on 4G communication according to claim 1 is characterized in that: The step of performing different preprocessing on the temperature verification result, the humidity verification result and the smoke concentration verification result to obtain temperature processing data, humidity processing data and smoke concentration processing data comprises: Performing noise filtering processing on the temperature verification result to obtain noise-filtered data, and normalizing the noise-filtered data to obtain temperature-processed data; Performing multi-scale decomposition processing on the humidity verification result, extracting humidity features at different scales, performing feature selection on the humidity features after the multi-scale decomposition, and obtaining humidity processing data; Cluster analysis is performed on the smoke concentration verification result to separate different types of smoke characteristic data, and the cluster analysis result is used to reconstruct the different types of smoke characteristic data to obtain smoke processing data.

4. The multi-sensor fusion smoke monitoring method based on 4G communication according to claim 3 is characterized in that: The step of performing noise filtering on the temperature verification result to obtain noise-filtered data, and normalizing the noise-filtered data to obtain temperature-processed data comprises: The temperature verification result is subjected to noise filtering through wavelet transform to obtain a low-frequency component and a high-frequency component, the low-frequency component is smoothed by a low-pass filter to obtain a smoothed low-frequency component, the high-frequency component is subjected to noise reduction processing based on the smoothed low-frequency component to obtain temperature data after noise filtering, and the temperature data after noise filtering is normalized to obtain temperature processing data.

5. The multi-sensor fusion smoke monitoring method based on 4G communication according to claim 3 is characterized in that: The step of performing multi-scale decomposition processing on the humidity verification result, extracting humidity features at different scales, performing feature selection on the humidity features after the multi-scale decomposition, and obtaining humidity processing data includes: Performing multi-scale decomposition processing on the humidity verification result through empirical mode decomposition to obtain intrinsic mode functions of different scales; The intrinsic mode function is screened according to energy entropy to obtain main intrinsic mode function components, and the main intrinsic mode function components are analyzed in the frequency domain by fast Fourier transform to extract frequency characteristics; A correlation analysis is performed on the frequency feature and the humidity verification result to obtain relevant features, and feature selection is performed on the relevant features to obtain humidity processing data.

6. The multi-sensor fusion smoke monitoring method based on 4G communication according to claim 3 is characterized in that: The step of performing cluster analysis on the smoke concentration verification result to separate different types of smoke feature data, and reconstructing the different types of smoke feature data using the cluster analysis result to obtain smoke processing data includes: Performing a preliminary cluster analysis on the smoke concentration verification result by K-means clustering to obtain the first cluster center; Performing secondary clustering on the preliminary cluster centers by using iterative self-organizing data analysis to obtain second cluster centers; Performing statistical analysis on the second cluster centers to obtain a feature vector of each of the second cluster centers; Inputting the feature vector into a support vector machine classifier for classification and reconstruction to obtain classification and reconstruction results and different types of smoke feature data; The characteristic data of the different types of smoke are reconstructed using the characteristic vector and the classification reconstruction result to obtain smoke concentration processed data.

7. The multi-sensor fusion smoke monitoring method based on 4G communication according to claim 1 is characterized in that: The step of fusing the temperature processing data with the humidity processing data to obtain comprehensive environmental data, and comprehensively analyzing the comprehensive environmental data and the smoke concentration verification result to obtain the smoke monitoring index includes: Performing weighted fusion on the temperature processing data and the humidity processing data to obtain fused data, and performing principal component analysis on the fused data to extract main components; Perform correlation analysis on the main components and the smoke concentration verification results to find out key indicators, calculate the key indicators using entropy weight method, and obtain the weight of each indicator; The fused data is adjusted using the weight to obtain comprehensive environmental data, and the comprehensive environmental data is clustered and analyzed with the smoke concentration verification result to obtain a smoke monitoring index.

8. A multi-sensor fusion smoke monitoring system based on 4G communication, characterized in that: The smoke monitoring system includes a central processing module and a 4G communication module, and also includes: An acquisition unit, used to acquire temperature data, humidity data and smoke concentration data of the environment, perform time sequence verification on the temperature data, humidity data and smoke concentration data, and obtain a time sequence synchronization result; An analysis unit is used to analyze the timing synchronization result to obtain a temperature verification result, a humidity verification result and a smoke concentration verification result, and to perform different preprocessing on the temperature verification result, the humidity verification result and the smoke concentration verification result to obtain temperature processing data, humidity processing data and smoke concentration processing data; An analysis unit, used for fusing the temperature processing data with the humidity processing data to obtain comprehensive environmental data, and performing comprehensive analysis on the comprehensive environmental data and the smoke concentration verification result to obtain a smoke monitoring index; An evaluation unit, used for transmitting the smoke monitoring index to the central processing module, evaluating the smoke monitoring index based on a preset smoke evaluation model, and obtaining a smoke concentration evaluation result; A transmission unit is used to transmit the smoke concentration evaluation result to a remote monitoring center using the 4G communication module, and the remote monitoring center performs smoke monitoring and early warning on the environment based on the smoke concentration evaluation result.

Images