An environmental online comprehensive monitoring method and system

Abstract

The application provides an environmental online comprehensive monitoring method and system, comprising the following steps: collecting noise, dust raising and water quality data in real time; pre-processing each data, removing abnormal, interference and impurity data, and performing smoothing, normalization and standardization processing; judging whether the data change degree of the next moment exceeds the set value based on the pre-processed data of the previous moment; if the set value is exceeded, performing spectrum, granularity and component analysis on the noise, dust raising and water quality data respectively, extracting characteristic information to form a matrix; according to the analysis result, using a corresponding model to analyze the characteristic information matrix to obtain an analysis result; if the analysis result is abnormal, obtaining abnormal data, using an abnormal analysis model to analyze, generating early warning information and sending the early warning information to a management platform. The application can comprehensively, dynamically and accurately monitor multiple environmental elements such as noise, dust raising and water quality, and improve the efficiency and accuracy of environmental monitoring.

Description

Technical Field

[0001] This invention relates to the field of environmental monitoring technology, and more specifically, to a method and system for online integrated environmental monitoring. Background Technology

[0002] In the field of environmental monitoring, traditional methods typically rely on manual sampling and laboratory analysis. This approach is not only time-consuming and labor-intensive but also fails to acquire environmental data in real time, resulting in a delayed response to environmental changes. With the development of sensor technology, sensor network-based environmental monitoring systems have emerged. These systems can collect environmental data in real time, but they often suffer from problems such as low data accuracy and difficulty in timely detection and handling of anomalies. For example, some monitoring systems, when collecting noise data, cannot effectively distinguish between noise sources, leading to mixed data and making it difficult to accurately analyze the causes of noise pollution. In dust monitoring, existing technologies struggle to accurately measure the particle size distribution of dust, failing to provide effective evidence for dust pollution control. Water quality monitoring faces similar challenges. Traditional water quality monitoring equipment often suffers from limitations such as high detection limits and long response times when analyzing water components, failing to meet the need for rapid and accurate monitoring of water quality changes.

[0003] In the process of implementing the embodiments of the present invention, the inventors have discovered that the prior art has at least the following problems or defects: the existing environmental monitoring methods cannot achieve comprehensive, real-time and accurate monitoring of multiple environmental elements such as noise, dust and water quality, and the data preprocessing is not perfect, which makes it difficult to detect and process abnormal data in a timely manner, and cannot provide timely and accurate information support for environmental management and decision-making. Summary of the Invention

[0004] This invention provides a method and system for online integrated environmental monitoring.

[0005] In a first aspect of the present invention, an online integrated environmental monitoring method is provided, comprising:

[0006] The noise data N obtained by the noise sensor, the dust data D obtained by the dust monitor, and the water quality data W obtained by the water quality monitoring equipment are collected in real time.

[0007] Each data point is preprocessed to remove outliers, interference, and impurities, and then smoothed, normalized, and standardized. Noisy data N is preprocessed to obtain preprocessed noisy data N. p Preprocessing the dust data D yields preprocessed dust data D. p Preprocessing the water quality data W yields preprocessed water quality data W. p ;

[0008] Based on the preprocessed noise data N from the previous moment p(t-1)Determine the preprocessed noise data N at the next time step. pt Is the degree of change greater than the set noise level value? N Based on the preprocessed dust data D from the previous moment p(t-1) Determine the preprocessed dust data D at the next moment. pt Is the degree of change greater than the set dust level value? D Based on the pretreated water quality data W from the previous moment p(t-1) Determine the pretreated water quality data W at the next moment. pt Is the degree of change greater than the set water quality level value? W If the values ​​exceed the set values, perform spectrum, particle size, and composition analysis on the noise, dust, and water quality data respectively, and extract feature information to form a matrix;

[0009] Based on the analysis results, the feature information matrix is ​​analyzed using the corresponding model to obtain the analysis results;

[0010] If the analysis results are abnormal, the abnormal data is obtained, the abnormal analysis model is used for analysis, an early warning message is generated and sent to the management platform.

[0011] Furthermore, the noise data N at the next time step pt Perform spectral analysis to determine the frequency distribution type T of the noise. N ,include:

[0012] The noise data N at the next moment pt Converted to the initial spectrum dataset S by a preset spectrum converter N ;

[0013] A spectrum analysis model based on historical noise data was used to analyze the initial spectrum dataset S. N Analysis yields the frequency distribution type T. N ;

[0014] The dust data D at the next moment pt Perform particle size analysis to determine the particle size distribution type T of the dust. D ,include:

[0015] The dust data at the next moment D pt Converted to the initial granularity dataset S by a preset granularity converter D ;

[0016] A particle size analysis model based on historical dust data was used to analyze the initial particle size dataset S. D Analysis was performed to obtain the particle size distribution type T. D The water quality data W at the next moment pt Perform component analysis to determine the component type T of the water quality. W,include:

[0017] The water quality data W at the next moment pt Converted to the initial component dataset S by the preset component converter W ;

[0018] A component analysis model based on historical water quality data was used to analyze the initial component dataset S. W Analysis was performed to obtain component type T. W .

[0019] Furthermore, the extraction process of the noise feature extraction model includes:

[0020] The noise data N at the next moment pt Data N is divided into multiple frequency bands according to frequency range. pt-s ;

[0021] According to frequency distribution type T N From data N in each frequency band pt-s Select those that match the corresponding frequency distribution type T N From the frequency band data, multiple key analytical frequency band data N were obtained. pt-k ;

[0022] The noise feature extractor is used to extract data N for each key analysis frequency band. pt-k The feature information is used to form multiple noise feature information matrices M. N .

[0023] Furthermore, the processing procedure of the noise analysis model includes:

[0024] According to frequency distribution type T N The noise feature information matrix M N The data is input into the corresponding noise information analysis unit;

[0025] Each noise information analysis unit determines the noise feature information matrix M based on its own preset noise feature information database. N Whether it conforms to the preset criteria of the noise feature information database; if it does not conform to the preset criteria, abnormal noise frequency band data N is obtained. a-f And obtain the noise analysis results R containing abnormal noise frequency band data information. N .

[0026] Furthermore, the specific process of extracting feature information to form a matrix is ​​as follows:

[0027] Based on the preprocessed noise data N from the previous moment p(t-1) Determine the preprocessed noise data N at the next time step. pt Is the degree of change greater than the set noise level value?N ;

[0028] If the noise data variation is greater than the set noise level value ∈ N Then for the noise data N at the next time step pt Perform spectral analysis to determine the frequency distribution type T of the noise. N The noise data N at the next time step is extracted using a noise feature extraction model. pt The feature information in the data is used to form a noise feature information matrix M. N ;

[0029] According to the frequency distribution type T of the noise N The noise feature information matrix M is analyzed using the corresponding information analysis unit in the noise analysis model. N The analysis yielded the noise analysis result R. N ;

[0030] When the noise analysis result R N When abnormal information exists, acquire abnormal noise data N for a set time period Δt starting from the next moment. a And a noise anomaly analysis model was used to analyze the abnormal noise data N. a The analysis yielded the noise anomaly analysis result R. Na ;

[0031] Based on the noise anomaly analysis results R Na Generate noise warning information E N Send various early warning information to the relevant management platform.

[0032] Furthermore, the analysis process of the noise anomaly analysis model includes:

[0033] Based on abnormal noise frequency band data information N a-f From the abnormal noise data N a Filter the data in the frequency range containing abnormal noise N a-f The noise data is used to obtain abnormal noise samples N. s-a ;

[0034] When abnormal noise samples N s-a The number exceeds the preset number n th Then, based on the frequency distribution type T N The corresponding preset noise anomaly information database is used for each abnormal noise sample N s-a Analysis was performed to determine the N of each anomalous noise sample. s-a First membership value μ 1N and the second membership value μ corresponding to the number of abnormal noise samples. 2N ;

[0035] The weighted average method is used to evaluate the first membership value μ.1N Second membership value μ 2N Process the data and calculate the average membership value μ. N The calculation formula is:

[0036] μ N =ω1μ 1N +ω2μ 2N

[0037] Where ω1 and ω2 are weighting coefficients, and ω1+ω2=1;

[0038] With average membership value μ N The corresponding level is designated as the abnormal level L. N According to frequency distribution type T N Abnormal level L N and each abnormal noise sample N s-a Noise anomaly analysis results R Na .

[0039] Furthermore, the obtained abnormal noise sample N s-a The screening process includes:

[0040] Based on abnormal noise frequency band data N a-f Feature information matrix M N-a-f Determine abnormal noise data N a The transformed feature information matrix M N-a Does the text contain characters with a similarity exceeding a set similarity threshold σ? N The feature information matrix;

[0041] If so, then the abnormal noise data N a N samples are labeled as anomalous noise. s-a .

[0042] Furthermore, the noise information analysis unit includes at least a traffic noise analysis unit, an industrial noise analysis unit, and a residential noise analysis unit.

[0043] The dust information analysis unit includes at least a construction dust analysis unit, a road dust analysis unit, and a stockpile dust analysis unit.

[0044] The water quality information analysis unit includes at least a heavy metal component analysis unit, an organic matter component analysis unit, and a microbial component analysis unit.

[0045] In a second aspect of the present invention, an online integrated environmental monitoring system is provided, comprising:

[0046] The system includes a data acquisition module, a data preprocessing module, an environmental analysis module, an anomaly analysis module, a data storage module, and an early warning module.

[0047] The aforementioned online integrated environmental monitoring system is used to implement the online integrated environmental monitoring method through the cooperation between various modules.

[0048] In a third aspect of the invention, an electronic device is provided, comprising: a memory, a processor, and an online integrated environmental monitoring program stored in the memory and executable on the processor, wherein the power equipment safety assessment program, when executed by the processor, implements the online integrated environmental monitoring method of the first aspect of the invention.

[0049] The above embodiments of the present invention have at least the following beneficial effects: The online integrated environmental monitoring method and system of the present invention can realize real-time and dynamic monitoring of multiple elements such as environmental noise, dust, and water quality. Through precise data preprocessing and judgment of the degree of change, abnormal fluctuations in environmental data can be detected in a timely manner. The system uses a specialized analysis model to conduct in-depth analysis of different types of environmental data, accurately extracting feature information and forming a feature information matrix, providing a solid foundation for subsequent anomaly analysis. After detecting abnormal data, the system can quickly activate the anomaly analysis model to conduct detailed analysis of the abnormal data, determine the anomaly level, generate corresponding early warning information, and promptly send it to the management platform so that relevant departments can quickly take measures to effectively address environmental problems.

[0050] Furthermore, this method and system can improve the efficiency and accuracy of environmental monitoring, reduce manual intervention, and lower monitoring costs. Through automated data acquisition, preprocessing, analysis, and early warning processes, it achieves intelligent and automated environmental monitoring, providing strong technical support for environmental management and decision-making, and helping to promptly identify and resolve environmental problems and protect the ecological environment. Attached Figure Description

[0051] The above and other objects, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description taken in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated in the drawings by way of example, not limitation, in which:

[0052] Figure 1 This is a flowchart illustrating an embodiment of the online integrated environmental monitoring method provided by the present invention.

[0053] Figure 2 This is a schematic diagram of the structure of an online integrated environmental monitoring system provided in an embodiment of the present invention;

[0054] Figure 3 A schematic diagram of the structure of an electronic device according to an embodiment of the present invention is shown. Detailed Implementation

[0055] The principles and spirit of the invention will now be described with reference to several exemplary embodiments. It should be understood that these embodiments are provided merely to enable those skilled in the art to better understand and implement the invention, and are not intended to limit the scope of the invention in any way. Rather, these embodiments are provided to make the invention more thorough and complete, and to fully convey the scope of the invention to those skilled in the art.

[0056] Those skilled in the art will understand that embodiments of the present invention can be implemented as a system, apparatus, device, method, or computer program product. Therefore, the present invention can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0057] It should be noted that the number of any elements in the accompanying drawings is for illustrative purposes only and not as a limitation, and any naming is for distinction only and has no limiting meaning.

[0058] The following is for reference. Figure 1 , Figure 1 This is a schematic flowchart of an online integrated environmental monitoring method provided in an embodiment of the present invention. Figure 1 As shown, an online integrated environmental monitoring method 100 includes:

[0059] Step 101: Real-time acquisition of noise data N from the noise sensor, dust data D from the dust monitor, and water quality data W from the water quality monitoring equipment.

[0060] Step 102: Preprocess the noise data N by removing abnormal noise values ​​and smoothing it to obtain preprocessed noise data N. p The dust data D is preprocessed to remove interference factors and normalize the data, resulting in preprocessed dust data D. p The water quality data W is preprocessed to remove impurities and standardize the data, resulting in preprocessed water quality data W. p .

[0061] Step 103, based on the preprocessed noise data N from the previous time step p(t-1) Determine the preprocessed noise data N at the next time step. pt Is the degree of change greater than the set noise level value? N Based on the preprocessed dust data D from the previous moment p(t-1) Determine the preprocessed dust data D at the next moment. pt Is the degree of change greater than the set dust level value? D Based on the pretreated water quality data W from the previous moment p(t-1) Determine the pretreated water quality data W at the next moment. ptIs the degree of change greater than the set water quality level value? W .

[0062] Step 104, if the degree of change in noise data is greater than the set noise level value ∈ N Then for the noise data N at the next time step pt Perform spectral analysis to determine the frequency distribution type T of the noise. N The noise data N at the next time step is extracted using a noise feature extraction model. pt The feature information in the data is used to form a noise feature information matrix M. N If the degree of change in dust pollution data exceeds the set dust pollution level value ∈ D Then, for the dust data D at the next moment... pt Perform particle size analysis to determine the particle size distribution type T of the dust. D The dust data D at the next moment was extracted using a dust feature extraction model. pt The characteristic information in the data is used to form a dust characteristic information matrix M. D If the degree of change in water quality data exceeds the set water quality level value ∈ W Then, for the water quality data W at the next moment... pt Perform component analysis to determine the component type T of the water quality. W The water quality data W at the next time step was extracted using a water quality feature extraction model. pt The characteristic information in the matrix is ​​used to form a water quality characteristic information matrix M. W .

[0063] Step 105, based on the noise frequency distribution type T N The noise feature information matrix M is analyzed using the corresponding information analysis unit in the noise analysis model. N The analysis yielded the noise analysis result R. N According to the particle size distribution type T of dust D The dust characteristic information matrix M is analyzed using the corresponding information analysis unit in the dust analysis model. D The analysis yielded the dust analysis result R. D According to the water quality composition type T W The water quality characteristic information matrix M is analyzed using the corresponding information analysis unit in the water quality analysis model. W The analysis yielded the water quality analysis result R. W .

[0064] Step 106, when the noise analysis result R N When abnormal information exists, acquire abnormal noise data N for a set time period Δt starting from the next moment. a And a noise anomaly analysis model was used to analyze the abnormal noise data N. aThe analysis yielded the noise anomaly analysis result R. Na When the dust analysis results R D When abnormal information is present, acquire abnormal dust data D for a set time period Δt starting from the next moment. a And a dust anomaly analysis model was used to analyze the abnormal dust data D. a Analysis was conducted, and the dust anomaly analysis result R was obtained. Da When the water quality analysis result R W When abnormal information is found, acquire abnormal water quality data W for a set time period Δt starting from the next moment. a And a water quality anomaly analysis model was used to analyze the abnormal water quality data W. a Analysis was performed, and the water quality anomaly analysis result R was obtained. Wa .

[0065] Based on the noise anomaly analysis results R Na Generate noise warning information E N According to the dust anomaly analysis results R Da Generate dust warning information E D According to the water quality anomaly analysis results R Wa Generate water quality early warning information E W And to send various early warning information to relevant management platforms.

[0066] It should be noted that this invention primarily involves the real-time acquisition of environmental data, specifically including noise data acquired by noise sensors, dust data acquired by dust monitors, and water quality data acquired by water quality monitoring equipment. Noise data refers to the sound intensity measured by specialized equipment, typically quantified in decibels (dB); dust data is the concentration of suspended particulate matter in the air measured by specific instruments, generally expressed in micrograms per cubic meter (μg / m³). 3 The data indicates that water quality data encompasses a variety of water quality parameters, such as pH, dissolved oxygen, and chemical oxygen demand. These parameters are acquired through water quality monitoring equipment to assess the quality of the water body.

[0067] Specifically, noise sensors can be high-precision sound level meters with a wide measurement range, covering environments from extremely quiet (e.g., a library, approximately 30 dB) to very noisy (e.g., a rock concert, approximately 130 dB). Dust monitors employ advanced technologies, such as laser scattering, to accurately measure particulate matter concentrations of different sizes, ranging from extremely low concentrations of 0.001 μg / m³. 3 Up to 1000 μg / m 3Water quality monitoring equipment includes a variety of sensors, such as pH sensors to measure the acidity or alkalinity of water, with a measurement range typically between 0 and 14 and an accuracy of up to 0.01 pH units; conductivity sensors to assess the electrical conductivity of water; and dissolved oxygen sensors to measure the amount of dissolved oxygen in water. These parameters together depict the overall water quality.

[0068] Preferably, to ensure the accuracy and reliability of the data, the present invention preprocesses the collected data. For noisy data, outliers are identified and removed by setting reasonable thresholds. For example, if the value of a data point exceeds twice the value of the previous data point or is less than half of it, it is considered an outlier. Smoothing can be achieved using a moving average method, that is, calculating the average of several consecutive data points to replace the value of the current data point. When processing dust data, interference data caused by equipment failure or natural factors (such as strong winds or rainfall) needs to be excluded, and normalization processing is performed to standardize the data to the range of 0 to 1. For water quality data, after removing impurities caused by sensor failure or water sample pollution, standardization processing is performed to convert it into dimensionless standardized values. These preprocessing steps effectively improve data quality and provide a solid foundation for subsequent in-depth analysis and early warning.

[0069] In some embodiments, the preprocessed noise data N based on the previous time step p(t-1) Determine the preprocessed noise data N at the next time step. pt Is the degree of change greater than the set noise level value? N ,include:

[0070] Noise data is stored in chronological order.

[0071] Obtain the noise data value N from the previous moment. p(t-1) and the noise data value N at the next time step pt .

[0072] Calculate the noise data value N at the next time step. pt Noise data value N from the previous moment p(t-1) The degree of change between δ N δ, the degree of change N The calculation formula is:

[0073]

[0074] Where, N pt(i) N represents the preprocessed noise data at the next time step. pt The i-th data point in N p(t-1) (i) represents the preprocessed noise data N from the previous time step. p(t-1) The i-th data point in the dataset, where n is the total number of data points.

[0075] Determine the degree of change δ N Is it greater than the set noise level value? N .

[0076] Based on the preprocessed dust data D from the previous moment p(t-1) Determine the preprocessed dust data D at the next moment. pt Is the degree of change greater than the set dust level value? D ,include:

[0077] Dust data is stored in chronological order.

[0078] Get the dust data value D from the previous moment. p(t-1) Dust data value D at the next moment pt .

[0079] Calculate the dust data value D at the next moment. pt Compared with the dust data value D at the previous moment p(t-1) The degree of change between δ D δ, the degree of change D The calculation formula is:

[0080]

[0081] Among them, D pt (j) represents the preprocessed dust data D at the next moment. pt The j-th data point in D p(t-1) (j) represents the preprocessed dust data D from the previous moment. p(t-1) The j-th data point in the dataset, where m is the total number of data points.

[0082] Determine the degree of change δ D Is it greater than the set dust level value? D .

[0083] Based on the pretreated water quality data from the previous moment W p(t-1) Determine the pretreated water quality data W at the next moment. pt Is the degree of change greater than the set water quality level value? W ,include:

[0084] Water quality data is stored in chronological order.

[0085] Get the water quality data value W from the previous moment. p(t-1) and the water quality data value W at the next moment pt .

[0086] Calculate the water quality data value W at the next moment. pt Compared with the water quality data value W at the previous moment p(t-1) The degree of change between δ W δ, the degree of changeW The calculation formula is:

[0087]

[0088] Among them, W pt (k) represents the pre-processed water quality data W at the next time step. pt The k-th data point in the data, Preprocessing water quality data for a later time point W pt The mean, Preprocessing water quality data for a later time point W pt standard deviation; W p(t-1) (k) represents the pre-processed water quality data W from the previous moment. p(t-1) The k-th data point in the data, Preprocessed water quality data W from the previous moment p(t-1) The mean, Preprocessed water quality data W from the previous moment p(t-1) The standard deviation is denoted as l, where l is the total number of data points.

[0089] Determine the degree of change δ W Is it greater than the set water quality level value? W .

[0090] It should be noted that determining the degree of change in noise, dust, and water quality data is a key step in this invention. The degree of change value is a quantitative indicator that measures the change of data over time, used to determine whether the data has changed significantly. Specifically, the degree of change value for noise data is obtained by calculating the difference between noise data points at a later time and an earlier time and then standardizing it; the degree of change value for dust data is obtained by calculating the squared difference between dust data points at a later time and an earlier time and then standardizing it; and the degree of change value for water quality data is obtained by calculating the standardized difference between water quality data points at a later time and an earlier time. These methods for calculating the degree of change values ​​ensure sensitivity and accuracy in detecting data changes.

[0091] Specifically, for noise data, the formula for calculating the degree of change is as follows:

[0092]

[0093] Where, N i (t) represents the i-th data point in the preprocessed noise data at the next time step, where N is the data point. i (t-1) represents the i-th data point in the preprocessed noise data from the previous time step, where n is the total number of data points. For dust data, the formula for calculating the degree of change is:

[0094]

[0095] Among them, D i (t) and D i (t-1) represents the i-th data point in the preprocessed dust data at the next and previous time points, respectively. For water quality data, the formula for calculating the degree of change is...

[0096]

[0097] Among them, W i (t) represents the i-th data point in the preprocessed water quality data at the next time step. and σ represents the mean of the preprocessed water quality data at the next and previous time points, respectively. W (t-1) represents the standard deviation of the pre-processed water quality data at the previous time step.

[0098] Preferably, to improve the accuracy and practicality of the calculation of the degree of change value, the above calculation formula can be optimized. For example, when calculating the degree of change value of noise data, a weighting factor can be introduced to assign different weights to data points in different time periods to reflect the importance of data in different time periods. For dust data, the influence of environmental factors, such as wind speed and humidity, can be considered to correct the degree of change value. For water quality data, the calculation parameters of the degree of change value can be dynamically adjusted by combining historical water quality data and seasonal changes.

[0099] Furthermore, a threshold for the degree of change can be set. When the calculated degree of change exceeds the threshold, the system automatically triggers further analysis and early warning mechanisms to ensure timely detection and handling of abnormal changes in environmental data.

[0100] In some embodiments, the noise data N at the next time step pt Perform spectral analysis to determine the frequency distribution type T of the noise. N ,include:

[0101] The noise data N at the next moment pt Converted to the initial spectrum dataset S by a preset spectrum converter N .

[0102] A spectrum analysis model based on historical noise data was used to analyze the initial spectrum dataset S. N Analysis yields the frequency distribution type T. N .

[0103] The dust data D at the next moment pt Perform particle size analysis to determine the particle size distribution type T of the dust. D ,include:

[0104] The dust data at the next moment D pt Converted to the initial granularity dataset S by a preset granularity converter D .

[0105] A particle size analysis model based on historical dust data was used to analyze the initial particle size dataset S. D Analysis was performed to obtain the particle size distribution type T. D The water quality data W at the next moment. pt Perform component analysis to determine the component type T of the water quality. W ,include:

[0106] The water quality data W at the next moment pt Converted to the initial component dataset S by the preset component converter W .

[0107] A component analysis model based on historical water quality data was used to analyze the initial component dataset S. W Analysis was performed to obtain component type T. W .

[0108] It should be noted that the spectral analysis of noise data at a later time step in this invention, to determine the frequency distribution type of the noise, is for a deeper understanding of the noise's characteristics. Spectral analysis is an analytical method that converts a signal from the time domain to the frequency domain, allowing the identification of the distribution of different frequency components in the noise. The frequency distribution type refers to the energy distribution characteristics of noise in different frequency bands; for example, some noise may be mainly concentrated in the low-frequency band, while others may have significant energy in the high-frequency band. Similarly, particle size analysis of dust data is performed to determine the distribution of particles of different sizes in the dust; the particle size distribution type reflects the distribution characteristics of particle size in the dust. Composition analysis of water quality data is performed to determine the types and contents of different chemical components in the water, including but not limited to heavy metals, organic matter, and microorganisms.

[0109] Specifically, spectral analysis can be achieved using the Fast Fourier Transform (FFT) algorithm, converting time-domain noise signals into frequency-domain spectral data. For example, noise data can be divided into multiple frequency bands, such as low frequency (0-100Hz), mid-frequency (100-1000Hz), and high frequency (above 1000Hz), and the energy distribution of each frequency band can be calculated. Particle size analysis can be performed using a laser particle size analyzer, which determines the particle size distribution by measuring the scattering or diffraction of laser light by particles. For example, particle size can be classified into fine particles (PM2.5, particle size less than 2.5 micrometers), medium particles (PM10, particle size less than 10 micrometers), and coarse particles (particle size greater than 10 micrometers). Composition analysis can utilize techniques such as spectral analysis and chromatographic analysis to determine the content of various components in water quality. For example, water components can be classified into categories such as heavy metals (e.g., lead, mercury, cadmium), organic matter (e.g., benzene, phenol, pesticide residues), and microorganisms (e.g., E. coli, total bacterial count).

[0110] Preferably, to improve the accuracy and efficiency of the analysis, machine learning-based spectral analysis models, granularity analysis models, and component analysis models can be employed. These models can automatically identify and classify different frequency distribution types, granularity distribution types, and component types through learning and training on large amounts of historical data. For example, support vector machines (SVM) or neural network algorithms can be used to construct these analysis models. In practical applications, noise data, dust data, and water quality data from the next time step can be input into the corresponding models, and the models will automatically output analysis results, including frequency distribution type, granularity distribution type, and component type.

[0111] Furthermore, a confidence threshold can be set for the model. Only when the model's output reaches a certain confidence level is the analysis result considered reliable, thereby further improving the accuracy and reliability of the monitoring system.

[0112] In some embodiments, the noise feature extraction process of the noise feature extraction model includes:

[0113] The noise data N at the next moment pt Data N is divided into multiple frequency bands according to frequency range. pt-s .

[0114] According to frequency distribution type T N From data N in each frequency band pt-s Select those that match the corresponding frequency distribution type T N From the frequency band data, multiple key analytical frequency band data N were obtained. pt-k .

[0115] The noise feature extractor is used to extract data N for each key analysis frequency band. pt-k The feature information is used to form multiple noise feature information matrices M.N The extraction process of the dust feature extraction model includes:

[0116] The dust data at the next moment D pt Data D is divided into multiple granularity segments according to granularity range. pt-s .

[0117] According to particle size distribution type T D From the data D of each granularity segment pt-s Select from those that match the corresponding particle size distribution type T D From the granularity segment data, we obtained multiple key analytical granularity segment data D. pt-k .

[0118] Data D for each key analytical particle size range was extracted using a dust feature extractor. pt-k The feature information is used to form multiple dust feature information matrices M. D .

[0119] The extraction process of the water quality feature extraction model includes:

[0120] The water quality data W at the next moment pt The data is segmented into multiple component data W according to component category. pt-s .

[0121] According to component type T W From the component data W pt-s Select the components that match the corresponding component type T. W From the component data, we obtained multiple key analytical component data W. pt-k .

[0122] Data on key analytical components were extracted using a water quality characterizer. pt-k The feature information is used to form multiple water quality feature information matrices M. W .

[0123] It should be noted that the extraction processes of the noise feature extraction model, dust feature extraction model, and water quality feature extraction model mentioned in this invention are aimed at filtering representative and crucial feature information from a large amount of environmental data for subsequent analysis and processing. Feature extraction is a crucial step in data preprocessing. Through this process, raw data can be transformed into a more meaningful feature information matrix, thereby improving the efficiency and accuracy of data analysis. For example, the noise feature extraction model segments noise data according to frequency range and filters out frequency bands that conform to a specific frequency distribution type to form a noise feature information matrix; the dust feature extraction model segments dust data according to particle size range and filters out particle size segments that conform to a specific particle size distribution type to form a dust feature information matrix; and the water quality feature extraction model segments water quality data according to component categories and filters out component data that conforms to a specific component type to form a water quality feature information matrix.

[0124] Specifically, in noise feature extraction models, the frequency band division can be set according to the actual application scenario and monitoring requirements. For example, the frequency range can be divided into low-frequency band (0-100Hz), mid-frequency band (100-1000Hz), and high-frequency band (above 1000Hz). For each frequency band, a bandpass filter can be used to extract data within that frequency range. In dust feature extraction models, the particle size segment division can be determined based on the environmental impact and health risks of particulate matter. For example, particle size can be divided into fine particles (PM2.5, particle size less than 2.5 micrometers), medium particles (PM10, particle size less than 10 micrometers), and coarse particles (particle size greater than 10 micrometers). In water quality feature extraction models, the classification of component data can be set according to the standards and objectives of water quality monitoring. For example, components can be classified into categories such as heavy metals (e.g., lead, mercury, cadmium), organic matter (e.g., benzene, phenol, pesticide residues), and microorganisms (e.g., E. coli, total bacterial count).

[0125] Preferably, to further improve the accuracy and efficiency of feature extraction, advanced signal processing techniques and data analysis methods can be employed. For example, in noise feature extraction, wavelet transform can be used to replace traditional bandpass filters. Wavelet transform provides finer frequency and time resolution, thus extracting noise features more accurately. In dust feature extraction, image processing techniques can be combined to assist in particle size classification by analyzing the morphology and distribution of particles. In water quality feature extraction, chemometric methods, such as principal component analysis (PCA) or partial least squares (PLS), can be used to process complex component data and extract the most representative feature information. Furthermore, the parameters of the feature extraction model can be dynamically adjusted according to changes in the actual monitoring environment to adapt to different monitoring conditions and data characteristics.

[0126] In some embodiments, the processing steps of the noise analysis model include:

[0127] According to frequency distribution type T N The noise feature information matrix M N The information is input into the corresponding noise information analysis unit.

[0128] Each noise information analysis unit determines the noise feature information matrix M based on its own preset noise feature information database. N Whether it conforms to the preset criteria of the noise feature information database; if it does not conform to the preset criteria, abnormal noise frequency band data N is obtained. a-f And obtain the noise analysis results R containing abnormal noise frequency band data information. N .

[0129] The processing steps of the dust analysis model include:

[0130] According to particle size distribution type T D The dust feature information matrix M D The information is input into the corresponding dust information analysis unit.

[0131] Each dust information analysis unit determines the dust feature information matrix M based on its own preset dust feature information database. D Whether it meets the preset criteria of the dust characteristic information database; if it does not meet the preset criteria, abnormal dust particle size range data D is obtained. a-g And obtain the dust analysis results R containing abnormal dust particle size range data. D .

[0132] The processing steps of the water quality analysis model include:

[0133] According to component type T W The water quality characteristic information matrix M W Input the information into the corresponding water quality analysis unit.

[0134] Each water quality information analysis unit determines the water quality characteristic information matrix M based on its own preset water quality characteristic information database. W Whether it meets the preset standards of the water quality characteristic information database; if it does not meet the preset standards, abnormal water quality component data W is obtained. a-c And to obtain water quality analysis results R containing abnormal water quality component data. W .

[0135] It should be noted that the processing procedures of the noise analysis model, dust analysis model, and water quality analysis model in this invention aim to identify abnormal data that does not meet the standards by comparing the feature information matrix with a preset feature information database. The noise analysis model inputs the noise feature information matrix into the corresponding analysis unit and determines whether the data is abnormal based on the preset noise feature information database. The dust analysis model and water quality analysis model follow the same logic, analyzing dust and water quality data respectively. The preset feature information database contains standard settings for normal data, such as frequency range, particle size range, and component content range. These standards are used to determine whether the data conforms to the normal range.

[0136] Specifically, the preset noise feature information database in the noise analysis model may include a standard set frequency range [f min f max ] and the set frequency variation range of the element and its surrounding elements [Δf min , Δf max For example, for traffic noise, the standard set frequency range could be [50Hz, 1000Hz], while the frequency variation range could be [10Hz, 50Hz]. The preset dust characteristic information database in the dust analysis model can include a standard set particle size range [dmin, dmax] and a set particle size variation range [Δdmin, Δdmax] between the element and surrounding elements. For example, for construction dust, the standard set particle size range could be [1μm, 100μm], and the particle size variation range could be [5μm, 20μm]. The preset water quality characteristic information database in the water quality analysis model can include a standard set component content range [cmin, cmax] and a set component content variation range [Δcmin, Δcmax] between the element and surrounding elements. For example, for heavy metal components, the standard set component content range could be [0.01mg / L, 0.1mg / L], and the component content variation range could be [0.005mg / L, 0.02mg / L].

[0137] Preferably, to improve the accuracy of the analysis, more refined analytical methods can be employed. For example, in noise analysis, machine learning algorithms, such as decision trees or random forests, can be introduced to automatically identify abnormal noise frequency bands. These algorithms can train models based on historical data, thereby more accurately determining whether the data is abnormal. In dust analysis, meteorological data, such as wind speed and direction, can be combined to help determine the source and scope of dust pollution. In water quality analysis, multivariate statistical analysis methods, such as cluster analysis, can be used to identify abnormal changes in water quality components. Furthermore, the standards of the preset feature information database can be dynamically adjusted according to changes in the actual monitoring environment to adapt to different monitoring conditions and data characteristics.

[0138] In some embodiments, the setting criteria for the preset noise feature information database include a noise feature information matrix M. N The standard frequency range of each element in [f] min f max ], the set frequency variation range of the element and its surrounding elements [Δf min , Δf max ].

[0139] The criteria for setting the preset dust feature information database include the dust feature information matrix M. D The standard granularity range for each element in [g] min g max ], The range of granularity variation between the element and its surrounding elements [Δg] min , Δg max ].

[0140] The criteria for setting the preset water quality characteristic information database include the water quality characteristic information matrix M. W The standard set content range of each element in [c] min c max ], the range of variation in the content of elements and surrounding elements [Δc] min , Δc max ].

[0141] It should be noted that the analysis process of the noise anomaly analysis model, dust anomaly analysis model, and water quality anomaly analysis model in this invention aims to determine the anomaly level and form the final anomaly analysis result by screening anomaly samples and calculating their membership values. Membership value is a concept in fuzzy mathematics used to represent the degree to which a sample belongs to a certain category. In this invention, the first membership value represents the similarity between the anomaly sample and the preset anomaly information database, and the second membership value represents the weight of the number of anomaly samples. Calculating the average membership value using a weighted average method can more accurately assess the severity of the anomaly.

[0142] Specifically, in the noise anomaly analysis model, the selection of abnormal noise samples is based on the abnormal noise frequency band data. Noise data containing feature information matrices with a similarity exceeding a set similarity threshold is selected from the abnormal noise data. For example, setting the similarity threshold to 0.8 means that noise data is only marked as an abnormal noise sample when the similarity of the feature information matrices exceeds 0.8. The dust anomaly analysis model and the water quality anomaly analysis model follow the same logic, selecting abnormal samples for dust and water quality data respectively. When calculating membership values, the weighting coefficients can be adjusted according to the actual application scenario. For example, for noise anomaly analysis, the weighting coefficients can be set to α1 = 0.6 and α2 = 0.4, indicating that the first membership value has a larger weight, while the second membership value has a smaller weight.

[0143] Preferably, to improve the accuracy and reliability of anomaly analysis, more advanced data analysis methods can be employed. For example, in noise anomaly analysis, deep learning algorithms, such as convolutional neural networks (CNNs), can be introduced to automatically extract features from anomalous noise samples and calculate membership values. These algorithms can automatically identify patterns in anomalous noise by learning from large amounts of historical data. In dust anomaly analysis, geographic information system (GIS) data can be combined to analyze the spatial distribution characteristics of dust anomaly samples, thereby more accurately determining the anomaly level. In water quality anomaly analysis, time series analysis methods, such as autoregressive moving average (ARMA) models, can be used to predict the changing trends of water quality anomalies. Furthermore, similarity thresholds and weighting coefficients can be dynamically adjusted according to changes in the actual monitoring environment to adapt to different monitoring conditions and data characteristics.

[0144] In some embodiments, the analysis process of the noise anomaly analysis model includes:

[0145] Based on abnormal noise frequency band data information N a-f From the abnormal noise data N a Filter the data in the frequency range containing abnormal noise N a-f The noise data is used to obtain abnormal noise samples N. s-a .

[0146] When abnormal noise samples N s-a The number exceeds the preset number n th Then, based on the frequency distribution type T N The corresponding preset noise anomaly information database is used for each abnormal noise sample N s-a Analysis was performed to determine the N of each anomalous noise sample. s-a First membership value μ 1N and the second membership value μ corresponding to the number of abnormal noise samples. 2N .

[0147] The weighted average method is used to evaluate the first membership value μ. 1N Second membership value μ 2N Process the data and calculate the average membership value μ. N The calculation formula is:

[0148] μ N =ω1μ 1N +ω2μ 2N

[0149] Where ω1 and ω2 are weighting coefficients, and ω1+ω2=1.

[0150] With average membership value μ N The corresponding level is designated as the abnormal level L. N According to frequency distribution type TN Abnormal level L N and each abnormal noise sample N s-a Noise anomaly analysis results R Na .

[0151] The analysis process of the dust anomaly analysis model includes:

[0152] Based on abnormal dust particle size data D a-G From abnormal dust data D a Filtering data D containing abnormal dust particle size a-g Dust data to obtain abnormal dust samples D s-a .

[0153] When abnormal dust sample D s-a The quantity exceeds the preset quantity m th Then, based on the granularity distribution type T D The corresponding preset dust anomaly information database is used for each abnormal dust sample D s-a Analysis was conducted to determine the D values ​​of each abnormal dust sample. s-a First membership value μ 1D and the second membership value μ corresponding to the number of abnormal dust samples 2D .

[0154] The weighted average method is used to evaluate the first membership value μ. 1d Second membership value μ 2D Process the data and calculate the average membership value μ. D The calculation formula is:

[0155] μ D =ω3μ 1D +ω4μ 2D

[0156] Where ω3 and ω4 are weighting coefficients, and ω3+ω4=1.

[0157] With average membership value μ D The corresponding level is designated as the abnormal level L. D According to particle size distribution type T D Abnormal level L D and various abnormal dust samples D s-a The results of the dust anomaly analysis are as follows: Da .

[0158] The analysis process of the water quality anomaly analysis model includes:

[0159] Based on abnormal water quality composition data W a-c From abnormal water quality data W a Screening data containing abnormal water quality components W a-cWater quality data, to obtain abnormal water quality samples W s-a .

[0160] When abnormal water quality sample W s-a The quantity exceeds the preset quantity. th Then, based on the component type T W The corresponding preset water quality anomaly information database is used for each abnormal water quality sample W. s-a Analysis was conducted to determine the W values ​​of each abnormal water quality sample. s-a First membership value μ 1W and the second membership value μ corresponding to the number of abnormal water quality samples 2W .

[0161] The weighted average method is used to evaluate the first membership value μ. 1W Second membership value μ 2W Process the data and calculate the average membership value μ. W The calculation formula is:

[0162] μ W =ω5μ 1W +ω6μ 2W

[0163] Wherein, ω5 and ω6 are weighting coefficients, and ω5+ω6=1.

[0164] With average membership value μ W The corresponding level is designated as the abnormal level L. W According to component type T W Abnormal level L W and various abnormal water quality samples W s-a The results of the water quality anomaly analysis are as follows: Wa .

[0165] It should be noted that the analysis process of the noise anomaly analysis model, dust anomaly analysis model, and water quality anomaly analysis model in this invention aims to determine the anomaly level and form the final anomaly analysis result by screening anomaly samples and calculating their membership values. Membership value is a concept in fuzzy mathematics used to represent the degree to which a sample belongs to a certain category. In this invention, the first membership value represents the similarity between the anomaly sample and the preset anomaly information database, and the second membership value represents the weight of the number of anomaly samples. Calculating the average membership value using a weighted average method can more accurately assess the severity of the anomaly.

[0166] Specifically, in the noise anomaly analysis model, the selection of abnormal noise samples is based on the abnormal noise frequency band data. Noise data containing feature information matrices with a similarity exceeding a set similarity threshold is selected from the abnormal noise data. For example, setting the similarity threshold to 0.8 means that noise data is only marked as an abnormal noise sample when the similarity of the feature information matrices exceeds 0.8. The dust anomaly analysis model and the water quality anomaly analysis model follow the same logic, selecting abnormal samples for dust and water quality data respectively. When calculating membership values, the weighting coefficients can be adjusted according to the actual application scenario. For example, for noise anomaly analysis, the weighting coefficients can be set to α1 = 0.6 and α2 = 0.4, indicating that the first membership value has a larger weight, while the second membership value has a smaller weight.

[0167] Preferably, to improve the accuracy and reliability of anomaly analysis, more advanced data analysis methods can be employed. For example, in noise anomaly analysis, deep learning algorithms, such as convolutional neural networks (CNNs), can be introduced to automatically extract features from anomalous noise samples and calculate membership values. These algorithms can automatically identify patterns in anomalous noise by learning from large amounts of historical data. In dust anomaly analysis, geographic information system (GIS) data can be combined to analyze the spatial distribution characteristics of dust anomaly samples, thereby more accurately determining the anomaly level. In water quality anomaly analysis, time series analysis methods, such as autoregressive moving average (ARMA) models, can be used to predict the changing trends of water quality anomalies. Furthermore, similarity thresholds and weighting coefficients can be dynamically adjusted according to changes in the actual monitoring environment to adapt to different monitoring conditions and data characteristics.

[0168] In some embodiments, the abnormal noise sample N is obtained s-a The screening process includes:

[0169] Based on abnormal noise frequency band data N a-f Feature information matrix M N-a-f Determine abnormal noise data N a The transformed feature information matrix M N-a Does the text contain characters with a similarity exceeding a set similarity threshold σ? N The feature information matrix.

[0170] If so, then the abnormal noise data N a N samples are labeled as anomalous noise. s-a .

[0171] The abnormal dust sample D was obtained s-a The screening process includes:

[0172] Based on abnormal dust particle size range data D a-g Feature information matrix M D-a-gDetermine abnormal dust data D a The transformed feature information matrix M D-a Does the text contain characters with a similarity exceeding a set similarity threshold σ? D The feature information matrix.

[0173] If so, then the abnormal dust data D a Sample D marked as abnormal dust s-a .

[0174] The abnormal water quality sample W was obtained s-a The screening process includes:

[0175] Based on abnormal water quality composition data W a-c Feature information matrix M W-a-c Determine abnormal water quality data W a The transformed feature information matrix M W-a Does the text contain characters with a similarity exceeding a set similarity threshold σ? W The feature information matrix.

[0176] If so, then the abnormal water quality data W a Water sample W marked as abnormal s-a .

[0177] It should be noted that the screening process for abnormal noise samples, abnormal dust samples, and abnormal water quality samples mentioned in this invention is designed to identify representative and critical samples from a large amount of abnormal data for further analysis and processing. The screening process is based on the similarity of feature information matrices. By comparing the feature information matrix transformed from abnormal data with the feature information matrices of abnormal frequency segments, granularity segments, or component data, it determines whether a sample is marked as abnormal. This process ensures that only data highly similar to known abnormal patterns are selected, improving the accuracy and efficiency of anomaly detection.

[0178] Specifically, similarity is a key parameter in the screening process for anomalous noise samples. For example, a similarity threshold of 0.75 can be set, meaning that noise data will only be labeled as an anomalous noise sample if the similarity between the feature information matrix converted from the anomalous noise data and the feature information matrix of the anomalous noise frequency band data exceeds 0.75. The similarity threshold also applies to dust and water quality data and can be adjusted according to the specific application scenario and data characteristics. For example, a similarity threshold of 0.8 can be set for dust data, while a similarity threshold of 0.7 can be set for water quality data. These thresholds are based on the analysis of historical data and experimental verification to ensure the effectiveness of the screening process.

[0179] Preferably, to further improve the accuracy and adaptability of the screening process, a method of dynamically adjusting the similarity threshold can be adopted. For example, the similarity threshold can be adjusted in real time according to data fluctuations and environmental changes. When data fluctuations are large or environmental changes are frequent, the similarity threshold can be appropriately lowered to improve detection sensitivity; conversely, when data is stable or environmental changes are small, the similarity threshold can be appropriately increased to reduce false alarms.

[0180] Furthermore, other auxiliary information, such as timestamps and geographical locations, can be combined to comprehensively determine the anomalies in the data. For example, when analyzing dust data, combining wind direction and speed information can more accurately determine the source and scope of dust impact, thereby improving the accuracy of anomaly sample screening.

[0181] In some embodiments, the noise information analysis unit includes at least a traffic noise analysis unit, an industrial noise analysis unit, and a residential noise analysis unit.

[0182] The dust information analysis unit includes at least a construction dust analysis unit, a road dust analysis unit, and a stockpile dust analysis unit.

[0183] The water quality information analysis unit includes at least a heavy metal component analysis unit, an organic matter component analysis unit, and a microbial component analysis unit.

[0184] It should be noted that the noise information analysis unit, dust information analysis unit, and water quality information analysis unit in this invention are modules specifically designed for analyzing different types of environmental data. These analysis units employ different analysis methods and models based on the source and characteristics of the data to provide more accurate and detailed analysis results. For example, the traffic noise analysis unit is specifically used to analyze noise generated by vehicles, the industrial noise analysis unit is used to analyze noise generated by factory equipment operation, and the residential noise analysis unit is used to analyze noise in daily life. Similarly, the construction dust analysis unit, road dust analysis unit, and stockpile dust analysis unit are used to analyze dust data from different sources. The water quality information analysis unit includes a heavy metal component analysis unit, an organic matter component analysis unit, and a microbial component analysis unit, used to analyze different chemical and biological components in water.

[0185] Specifically, within the noise information analysis unit, the traffic noise analysis unit can be set with specific frequency ranges and time patterns. For example, traffic noise is typically more pronounced in the low-frequency range (50Hz-250Hz) and is more concentrated during peak hours on weekdays (such as 7:00 AM to 9:00 AM and 5:00 PM to 7:00 PM). The industrial noise analysis unit can focus on a broader frequency range and is related to factory operating hours. The residential noise analysis unit can focus more on noise levels at night and on weekends. For dust information analysis units, the construction dust analysis unit can set parameters based on the activity time at the construction site and the effectiveness of dust control measures. The road dust analysis unit can be related to traffic flow and road maintenance conditions. The stockpile dust analysis unit needs to consider the type of material and the stockpiling method. In the water quality information analysis unit, the heavy metal composition analysis unit can set detection limits and alarm thresholds for specific heavy metal elements, the organic matter composition analysis unit can focus on specific organic pollutants, and the microbial composition analysis unit can detect bacteria and viruses in the water.

[0186] Preferably, to improve the accuracy and practicality of the analysis, these analysis units can be further optimized. For example, the traffic noise analysis unit can combine real-time traffic flow data and meteorological conditions to adjust analysis parameters for more accurate prediction and analysis of the impact of traffic noise. The industrial noise analysis unit can be combined with factory production plans and equipment maintenance records to identify the sources of noise anomalies. The residential noise analysis unit can utilize community feedback and resident activity patterns to optimize the analysis. For the dust analysis unit, satellite imagery and Geographic Information System (GIS) data can be combined to assess the spread and impact area of ​​dust. In the water quality analysis unit, the analysis model can be calibrated by comparing online monitoring data and laboratory analysis results to improve detection accuracy. Furthermore, machine learning algorithms, such as Support Vector Machines (SVM) or neural networks, can be introduced to automatically identify and classify different types of environmental data, thereby improving the efficiency and reliability of the analysis.

[0187] The above-described embodiments of the present invention have the following beneficial effects: The online integrated environmental monitoring method and system of the present invention can realize real-time and dynamic monitoring of multiple elements such as environmental noise, dust, and water quality. Through precise data preprocessing, including removing outliers, interference factors, and impurity data, and performing smoothing, normalization, and standardization, the accuracy and reliability of the data can be ensured. The system can determine whether the degree of change in the data at the next moment exceeds a set value based on the preprocessed data at the previous moment, thereby timely detecting abnormal fluctuations in environmental data. For data exceeding the set value, the system can further perform spectrum, granularity, and component analysis, extract feature information to form a matrix, and provide a basis for in-depth analysis. Based on the analysis results, the feature information matrix can be analyzed using a corresponding model to obtain accurate analysis results. If the analysis results show anomalies, the system can acquire abnormal data, analyze it using an anomaly analysis model, generate early warning information, and send it to the management platform so that relevant departments can quickly take measures to effectively address environmental problems.

[0188] Furthermore, this method and system can improve the efficiency and accuracy of environmental monitoring, reduce manual intervention, and lower monitoring costs. Through automated data acquisition, preprocessing, analysis, and early warning processes, intelligent and automated environmental monitoring is achieved. For example, the noise information analysis unit includes units for traffic, industrial, and residential noise analysis; the dust information analysis unit includes units for construction, road, and storage yard dust analysis; and the water quality information analysis unit includes units for heavy metals, organic matter, and microbial composition analysis. These specialized analysis units can perform precise analysis on environmental data from different sources and types, providing more detailed environmental information. This comprehensive monitoring method and system can provide timely and accurate information support for environmental management and decision-making, helping to promptly identify and resolve environmental problems and protect the ecological environment.

[0189] like Figure 2 As shown in some embodiments, an online integrated environmental monitoring system 200 includes:

[0190] The system includes a data acquisition module 201, a data preprocessing module 202, an environmental analysis module 203, an anomaly analysis module 204, a data storage module 205, and an early warning module 206.

[0191] The aforementioned online integrated environmental monitoring system is used to implement the online integrated environmental monitoring method through the cooperation between various modules.

[0192] It is understandable that the modules recorded in this online integrated environmental monitoring system 200 are similar to those in the reference system. Figure 1The steps described in the online integrated environmental monitoring method correspond to each other. Therefore, the operation, characteristics, and beneficial effects described above for the online integrated environmental monitoring method also apply to the online integrated environmental monitoring system 200 and its included modules, and will not be repeated here.

[0193] The following is for reference. Figure 3 The diagram illustrates a structural schematic of an electronic device 300 suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 3 The terminal device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention.

[0194] like Figure 3 As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0195] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0196] Furthermore, the storage medium in the embodiments of this application stores program instructions capable of implementing all the above methods. These program instructions can be stored in the storage medium in the form of a software product, including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, or terminal devices such as computers, servers, mobile phones, and tablets.

[0197] The above description is merely a selection of preferred embodiments of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention as described in the embodiments is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A method for online integrated environmental monitoring, characterized in that, include: Real-time acquisition of noise data from noise sensors Dust data acquired by dust monitoring instruments Water quality data obtained from water quality monitoring equipment ; noise data Preprocessing is performed to remove abnormal noise values ​​and smooth the data, resulting in preprocessed noise data. Dust data Preprocessing is performed to remove interference factors from the data and then normalize the data to obtain preprocessed dust data. ; water quality data Pretreatment is performed to remove impurities and standardize the data to obtain pretreated water quality data. ; Based on the preprocessed noise data from the previous moment Determine the preprocessed noise data at the next time step. Is the degree of change greater than the set noise level value? Based on the preprocessed dust data from the previous moment To determine the preprocessed dust data at the next moment. Is the degree of change greater than the set dust level value? Based on the pre-treated water quality data from the previous moment To determine the pre-treated water quality data at the next moment. Is the degree of change greater than the set water quality level value? ; If the noise data changes more than the set noise level value Then, for the noise data at the next time step Perform spectral analysis to determine the frequency distribution type of the noise. Noise data at the next time step is extracted using a noise feature extraction model. The feature information in the data is used to form a noise feature information matrix. If the change in dust data exceeds the set dust level value... Then the dust data at the next moment Perform particle size analysis to determine the particle size distribution type of dust. The dust data at the next moment was extracted using a dust feature extraction model. The characteristic information in the data is used to form a dust characteristic information matrix. If the change in water quality data exceeds the set water quality level value; Then the water quality data at the next moment Perform component analysis to determine the types of water components. Water quality data at the next time step was extracted using a water quality feature extraction model. The characteristic information in the matrix is ​​used to form a water quality characteristic information matrix. ; Based on the frequency distribution type of noise The noise feature information matrix is ​​analyzed using the corresponding information analysis unit in the noise analysis model. The analysis was performed to obtain the noise analysis results. According to the particle size distribution type of dust The dust characteristic information matrix is ​​analyzed using the corresponding information analysis unit in the dust analysis model. Analysis was conducted to obtain dust analysis results. According to the composition type of water quality The water quality characteristic information matrix is ​​analyzed using the corresponding information analysis unit in the water quality analysis model. The analysis was performed, and the water quality analysis results were obtained. ; When noise analysis results When abnormal information is found, retrieve the set time period starting from the next moment. Abnormal noise data And a noise anomaly analysis model was used to analyze the abnormal noise data. Analysis was performed to obtain the noise anomaly analysis results. When dust analysis results When abnormal information is found, retrieve the set time period starting from the next moment. Abnormal dust data And a dust anomaly analysis model was used to analyze abnormal dust data. Analysis was conducted to obtain the results of the dust anomaly analysis. When water quality analysis results When abnormal information is found, retrieve the set time period starting from the next moment. Abnormal water quality data And a water quality anomaly analysis model was used to analyze the abnormal water quality data. Analysis was conducted to obtain the results of the water quality anomaly analysis. ; Based on the noise anomaly analysis results Generate noise warning information Based on the results of the dust anomaly analysis Generate dust warning information Based on the results of the water quality anomaly analysis Generate water quality early warning information And to send various early warning information to relevant management platforms.

2. The online integrated environmental monitoring method according to claim 1, characterized in that, The preprocessed noise data based on the previous time step Determine the preprocessed noise data at the next time step. Is the degree of change greater than the set noise level value? ,include: Noise data is stored in chronological order; Obtain the noise data value from the previous moment. and noise data values ​​at the next time step ; Calculate the noise data value at the next time step. Noise data value compared to the previous moment The degree of change between degree of change The calculation formula is: in, This indicates the preprocessed noise data for the next time step. The first in Data points, This indicates the preprocessed noise data from the previous moment. The first in Data points, This represents the total number of data points. Determine the degree of change Is it greater than the set noise level value? ; Based on the preprocessed dust data from the previous moment To determine the preprocessed dust data at the next moment. Is the degree of change greater than the set dust level value? ,include: Dust data is stored in chronological order; Get the dust data value from the previous moment. Dust data value at the next moment ; Calculate the dust data value at the next moment. Compared with the dust data value at the previous moment The degree of change between degree of change The calculation formula is: in, This indicates the preprocessing of dust data at the next moment. The first in Data points, This indicates the preprocessed dust data from the previous moment. The first in Data points, This represents the total number of data points. Determine the degree of change Is it greater than the set dust level value? ; Based on the pretreated water quality data from the previous moment To determine the pre-treated water quality data at the next moment. Is the degree of change greater than the set water quality level value? ,include: Water quality data is stored in chronological order; Get the water quality data value from the previous moment. and the water quality data value at the next moment ; Calculate the water quality data value at the next moment. Compared with the water quality data value at the previous moment The degree of change between degree of change The calculation formula is: in, This indicates the preprocessed water quality data at the next moment. The first in Data points, Preprocessing water quality data for later stages The mean, Preprocessing water quality data for later stages Standard deviation; This indicates the pre-processed water quality data from the previous moment. The first in Data points, Preprocessing water quality data from the previous moment The mean, Preprocessing water quality data from the previous moment standard deviation This represents the total number of data points. Determine the degree of change Is it greater than the set water quality level value? .

3. The online integrated environmental monitoring method according to claim 2, characterized in that, The noise data for the next time moment Perform spectral analysis to determine the frequency distribution type of the noise. ,include: Noise data at the next moment Converted into an initial spectrum dataset by a preset spectrum converter ; A spectrum analysis model based on historical noise data was used to analyze the initial spectrum dataset. Analysis was performed to obtain the frequency distribution type. ; The dust data at the next moment Perform particle size analysis to determine the particle size distribution type of dust. ,include: Dust data at the next moment Converted to initial granularity dataset by a preset granularity converter ; A granularity analysis model based on historical dust data was used to analyze the initial granularity dataset. Analysis was performed to obtain the particle size distribution type. The water quality data at the next moment. Perform component analysis to determine the types of water components. ,include: Water quality data at the next moment Converted into an initial component dataset via a preset component converter ; A component analysis model based on historical water quality data was used to analyze the initial component dataset. Analysis was performed to obtain the component types. .

4. The online integrated environmental monitoring method according to claim 2, characterized in that, The noise feature extraction model includes the following extraction process: Noise data at the next moment Data is divided into multiple frequency bands according to frequency range. ; According to frequency distribution type From data in each frequency band Select those that match the corresponding frequency distribution type From the frequency band data, multiple key analytical frequency band data were obtained. ; Data for each key analysis frequency band was extracted using a noise feature extractor. The feature information is used to form multiple noise feature information matrices. The extraction process of the dust feature extraction model includes: Dust data at the next moment Data is divided into multiple granularity segments according to granularity range. ; According to particle size distribution type From data at various granularity levels Select those that match the corresponding particle size distribution type. The granularity data was used to obtain multiple key analytical granularity data segments. ; Data for each key analytical particle size range was extracted using a dust feature extractor. The characteristic information is used to form multiple dust characteristic information matrices. ; The extraction process of the water quality feature extraction model includes: Water quality data at the next moment The data is segmented into multiple component data according to component category. ; According to ingredient type From the data of each component Select the components that match the corresponding ingredient type From the component data, we obtained data on several key analytical components. ; Data on key analytical components were extracted using a water quality characterizer. The characteristic information is used to form multiple water quality characteristic information matrices. .

5. The online integrated environmental monitoring method according to claim 4, characterized in that, The processing steps of the noise analysis model include: According to frequency distribution type The noise feature information matrix The data is input into the corresponding noise information analysis unit. Each noise information analysis unit determines the noise feature information matrix based on its own preset noise feature information database. Whether it conforms to the preset criteria of the noise feature information database; if it does not conform to the preset criteria, abnormal noise frequency band data is obtained. And obtain noise analysis results containing abnormal noise frequency band data information. ; The processing steps of the dust analysis model include: According to particle size distribution type The dust characteristic information matrix Input into the corresponding dust information analysis unit; Each dust information analysis unit determines the dust feature information matrix based on its own preset dust feature information database. Whether it meets the preset criteria of the dust characteristic information database; if it does not meet the preset criteria, abnormal dust particle size range data is obtained. And obtain dust analysis results containing data on abnormal dust particle size ranges. ; The processing steps of the water quality analysis model include: According to ingredient type The water quality characteristic information matrix Input into the corresponding water quality information analysis unit; Each water quality information analysis unit determines the water quality characteristic information matrix based on its own preset water quality characteristic information database. Whether it meets the preset standards of the water quality characteristic information database; if it does not meet the preset standards, abnormal water quality component data is obtained. And to obtain water quality analysis results containing data information on abnormal water quality components. .

6. The online integrated environmental monitoring method according to claim 5, characterized in that, The criteria for setting the preset noise feature information database include a noise feature information matrix. Standard frequency ranges for each element The set frequency variation range of the element and its surrounding elements ; The criteria for setting the preset dust feature information database include a dust feature information matrix. The standard granularity range for each element in the standard settings The range of granularity variation between elements and surrounding elements ; The criteria for setting the preset water quality characteristic information database include a water quality characteristic information matrix. The standard set content range of each element in the composition The range of variation in the content of elements and surrounding elements. .

7. The online integrated environmental monitoring method according to claim 5, characterized in that, The analysis process of the noise analysis model includes: Based on abnormal noise frequency band data information From abnormal noise data Filtering data in frequency ranges containing abnormal noise The noise data was used to obtain abnormal noise samples. ; When abnormal noise samples The quantity exceeds the preset quantity Then, based on the frequency distribution type The corresponding preset noise anomaly information database for each abnormal noise sample Analysis was conducted to identify each abnormal noise sample. First membership value and the second membership value corresponding to the number of abnormal noise samples. ; The weighted average method is used to evaluate the first membership value. Second membership value Process the data and calculate the average membership value. The calculation formula is: in, and These are the weighting coefficients, and ; Average membership value The corresponding level is designated as the abnormal level. According to frequency distribution type Abnormal level and various abnormal noise samples Noise anomaly analysis results ; The analysis process of the dust anomaly analysis model includes: Based on abnormal dust particle size data From abnormal dust data Filtering data containing abnormal dust particle size ranges Dust data to obtain abnormal dust samples. ; When abnormal dust samples The quantity exceeds the preset quantity Then, based on the granularity distribution type The corresponding preset dust anomaly information database is used for each abnormal dust sample. Analysis was conducted to identify each abnormal dust sample. First membership value and the second membership value corresponding to the number of abnormal dust samples. ; The weighted average method is used to evaluate the first membership value. Second membership value Process the data and calculate the average membership value. The calculation formula is: in, and These are the weighting coefficients, and ; Average membership value The corresponding level is designated as the abnormal level. According to the particle size distribution type Abnormal level and various abnormal dust samples Dust anomaly analysis results ; The analysis process of the water quality anomaly analysis model includes: Based on abnormal water quality composition data From abnormal water quality data Screening data containing abnormal water quality components Water quality data to obtain abnormal water quality samples. ; When abnormal water quality samples The quantity exceeds the preset quantity Then, based on the component type The corresponding preset water quality anomaly information database is used for each abnormal water quality sample. Analysis was conducted to identify the abnormal water quality samples. First membership value and the second membership value corresponding to the number of abnormal water quality samples. ; The weighted average method is used to evaluate the first membership value. Second membership value Process the data and calculate the average membership value. The calculation formula is: in, and These are the weighting coefficients, and ; Average membership value The corresponding level is designated as the abnormal level. According to the type of ingredients Abnormal level and various abnormal water quality samples Results of water quality anomaly analysis .

8. The online integrated environmental monitoring method according to claim 7, characterized in that, The obtained abnormal noise sample The screening process includes: Based on abnormal noise frequency band data Feature information matrix Identify abnormal noise data Transformed feature information matrix Does the content contain elements with a similarity exceeding the set similarity threshold? The feature information matrix; If so, then abnormal noise data Samples marked as anomalous noise ; The abnormal dust sample obtained The screening process includes: Based on abnormal dust particle size range data Feature information matrix Determine abnormal dust data Transformed feature information matrix Does the content contain elements with a similarity exceeding the set similarity threshold? The feature information matrix; If so, then the abnormal dust data will be recorded. Samples marked as abnormal dust ; The abnormal water quality sample obtained The screening process includes: Based on abnormal water quality composition data Feature information matrix To identify abnormal water quality data Transformed feature information matrix Does the content contain elements with a similarity exceeding the set similarity threshold? The feature information matrix; If so, then the abnormal water quality data will be recorded. Water samples marked as abnormal .

9. The online integrated environmental monitoring method according to claim 5, characterized in that, The noise information analysis unit includes at least a traffic noise analysis unit, an industrial noise analysis unit, and a residential noise analysis unit. The dust information analysis unit includes at least a construction dust analysis unit, a road dust analysis unit, and a stockpile dust analysis unit; The water quality information analysis unit includes at least a heavy metal component analysis unit, an organic matter component analysis unit, and a microbial component analysis unit.

10. An online integrated environmental monitoring system, characterized in that, The online environmental integrated monitoring system includes a data acquisition module, a data preprocessing module, an environmental analysis module, an anomaly analysis module, a data storage module, and an early warning module; the online environmental integrated monitoring system is used to implement the online environmental integrated monitoring method as described in any one of claims 1-9 through the cooperation between the modules.

Images