Intelligent sewage detection method and system for municipal construction

Abstract

The invention relates to an intelligent sewage detection method and system for municipal construction, and the method comprises the following steps: carrying out the real-time spectrum scanning of sewage in a sewage pipe network in a target city through a preset spectrometer, and obtaining sewage spectrum scanning data; performing spectral feature extraction on the sewage spectral scanning data to obtain sewage spectral feature data; performing concentration detection analysis on the sewage based on the sewage spectral characteristic data to obtain a sewage concentration parameter; whether the sewage concentration parameter exceeds a preset sewage concentration range or not is judged, if yes, sewage classification is conducted on the sewage based on the sewage spectral feature data, and a sewage classification result is obtained; and a sewage management scheme is formulated based on the sewage concentration parameter and the sewage classification result. The technical problems that a traditional method is long in time consumption and high in cost, and real-time monitoring of the sewage quality is difficult to achieve are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage detection, and in particular to an intelligent sewage detection method and system for municipal construction. Background Art

[0002] With the acceleration of urbanization, the pressure on municipal infrastructure is increasing, especially in the field of sewage treatment, which is facing unprecedented challenges. Traditional sewage treatment methods mainly rely on manual sampling and laboratory analysis, which is not only time-consuming and costly, but also difficult to achieve real-time monitoring of sewage quality. With the development of social economy and the improvement of people's living standards, people's requirements for the quality of living environment are gradually increasing, and higher requirements are put forward for sewage discharge standards, which makes the traditional sewage treatment model unable to meet the needs of modern urban management.

[0003] In recent years, with the rapid development of information technology, the application of intelligent technology in various fields has gradually increased, providing new ideas for solving the problems existing in traditional sewage treatment. For example, by introducing advanced spectral analysis technology, it is possible to quickly and accurately detect sewage components, and then promptly discover and deal with abnormal water quality. However, the application of this technology is still in its infancy, and there are some urgent problems to be solved, such as how to ensure the accuracy of spectral data and how to efficiently extract useful information from massive data. If these problems are not effectively solved, they will directly affect the performance and reliability of the intelligent sewage detection system.

[0004] In addition, although intelligent sewage detection technology has obvious advantages, it is also necessary to consider the integration with the existing municipal management system in the actual application process. On the one hand, a complete data sharing mechanism needs to be established to ensure efficient collaboration between different departments. At the same time, the public's awareness and acceptance of new technologies is also one of the important factors affecting their promotion and use. Therefore, while promoting intelligent sewage detection technology, it is also necessary to do a good job in related publicity and education to enhance the awareness and support of environmental protection awareness among all sectors of society. Summary of the invention

[0005] The main purpose of the present invention is to provide an intelligent sewage detection method and system for municipal construction, which solves the technical problems that traditional methods are not only time-consuming and costly, but also difficult to achieve real-time monitoring of sewage quality.

[0006] To achieve the above object, the present invention provides an intelligent sewage detection method for municipal construction, comprising the following steps: Use a preset spectrometer to perform real-time spectral scanning on sewage in the sewage pipe network in the target city to obtain sewage spectral scanning data; Extracting spectral features from the sewage spectral scanning data to obtain sewage spectral feature data; Performing concentration detection and analysis of the sewage based on the sewage spectral characteristic data to obtain sewage concentration parameters; Determine whether the sewage concentration parameter exceeds a preset sewage concentration range, and if so, perform sewage classification of the sewage based on the sewage spectral characteristic data to obtain a sewage classification result; A sewage management plan is formulated based on the sewage concentration parameters and sewage classification results.

[0007] Furthermore, the use of a preset spectrometer to perform real-time spectral scanning on the sewage in the sewage pipe network in the target city to obtain sewage spectral scanning data includes: Use a preset spectrometer to perform continuous spectral scanning on the sewage in the sewage pipe network in the target city to obtain preliminary spectral data; Performing background subtraction processing on the preliminary spectral data using a preset background spectral model to obtain spectral data after background subtraction; Performing baseline correction on the spectral data after background subtraction to obtain corrected spectral data; The corrected spectral data is subjected to a first-order derivative transformation to obtain sewage spectral scanning data.

[0008] Furthermore, the extracting of spectral features from the sewage spectral scanning data to obtain sewage spectral feature data includes: The sewage spectrum scanning data is converted into a frequency domain by using a preset Fourier transform network model to obtain a frequency domain characteristic spectrum; Performing time series analysis on the sewage spectrum scanning data based on the frequency domain characteristic spectrum to obtain time series characteristics; The time series features are transformed into polar coordinates through a preset Gram angular field transformation network to obtain a GAF image; Extracting texture features of the GAF image to obtain image texture features; Using a preset continuous projection algorithm to extract feature bands from the image texture features to obtain band extraction features; The band extraction features are subjected to feature separation to obtain sewage spectral feature data.

[0009] Furthermore, the sewage concentration detection and analysis based on the sewage spectral characteristic data to obtain sewage concentration parameters includes: Performing spectrum analysis on the sewage spectrum characteristic data to obtain spectrum distribution data; Based on the spectrum distribution data, each component in the sewage is identified to obtain a component identification label and a characteristic frequency corresponding to the component identification label; wherein the component identification label includes ammonia nitrogen, COD and total phosphorus, and the characteristic frequency is a key indicator related to the component concentration; By using the component identification tag, the characteristic peaks of each component in the sewage are identified and detected based on the spectrum distribution data to obtain the absorption peak quantification parameters; wherein the absorption peak quantification parameters include the characteristic peak frequency and amplitude of the component; Calculate the preliminary concentration data of each of the sewage based on the characteristic peak frequency, amplitude and characteristic frequency by using a preset Lambert-Beer law formula; The preliminary concentration data are supplemented with concentration data integrity to obtain complete concentration estimation data; By using a Bayesian inference algorithm, a multi-dimensional concentration evaluation calculation is performed on the sewage based on the complete concentration estimation data to obtain sewage concentration parameters.

[0010] Furthermore, the sewage classification based on the sewage spectral characteristic data to obtain the sewage classification result includes: By using a preset multi-layer perceptron model, a time series analysis is performed on the components in the sewage based on the sewage spectral characteristic data to obtain component concentration change data; Obtaining environmental data corresponding to the surrounding environmental location near the sewage pipe network and the surrounding environmental location, and performing source analysis on the component concentration change data based on the environmental data by a preset parallel factor analysis method to obtain component source data; Combining the component source data and the sewage spectral characteristic data to obtain sewage combination data; The singular value decomposition algorithm is used to perform principal component analysis on the sewage combination data to obtain the sewage principal component feature vector; Performing kernel density estimation on the sewage principal component eigenvector to obtain sewage characteristic probability density distribution; wherein the sewage characteristic probability density distribution includes the probability density value corresponding to each eigenvalue, the value range of the eigenvalue and the kernel function bandwidth; The sewage is classified based on the probability density distribution of the sewage characteristics and the environmental data corresponding to the surrounding environment location to obtain a sewage classification result; wherein the sewage classification result includes domestic sewage, industrial sewage and agricultural sewage.

[0011] Furthermore, the sewage is classified based on the probability density distribution of the sewage characteristics and the environmental data corresponding to the surrounding environment location to obtain the sewage classification result, including: Using a preset restricted Boltzmann machine algorithm, potential pattern mining is performed on the sewage based on the probability density distribution of the sewage characteristics to obtain sewage potential pattern data; Performing multivariate time series analysis on the environmental data corresponding to the surrounding environment location to extract environmental change trend data; Using a long short-term memory network, the sewage potential pattern data is adjusted based on the environmental change trend data to obtain adjusted sewage potential pattern data; Performing matrix decomposition on the sewage spectral feature data to obtain a sewage spectral decomposition result; wherein the sewage spectral decomposition result includes a base matrix and a coefficient matrix, the base matrix is ​​used for the basic components of the sewage spectral feature, and the coefficient matrix is ​​used to represent the weight of each basic component in the sewage; The sewage is classified based on the sewage spectral decomposition result and the adjusted sewage potential pattern data to obtain a sewage classification result.

[0012] Furthermore, formulating a sewage management plan based on the sewage concentration parameter and the sewage classification result includes: Constructing risk assessment indicators based on the sewage concentration parameters and sewage classification results to obtain a risk assessment indicator system; Calling a sewage pipe network image in a sewage pipe network in a database, analyzing the sewage pipe network based on the sewage pipe network image to obtain an analysis result; wherein the analysis result includes a topological structure of the sewage pipe and sewage flow data; Based on the topological structure and sewage flow data, graph theory analysis is performed to obtain the connectivity and flow distribution characteristics of the sewage pipe network; Based on the risk assessment indicator system, the connectivity of the sewage pipe network and the flow distribution characteristics, a sewage risk propagation simulation is performed on the sewage to obtain the sewage risk propagation path and impact range; Formulate treatment measures based on the sewage risk transmission path and impact range, and obtain a preliminary set of treatment measures; The preliminary treatment measure set is optimized according to the sewage classification result to obtain a sewage management plan.

[0013] The present invention also provides an intelligent sewage detection system for municipal construction, comprising: A scanning module is used to perform real-time spectral scanning of sewage in the sewage pipe network in the target city using a preset spectrometer to obtain sewage spectral scanning data; An extraction module, used for extracting spectral features from the sewage spectral scanning data to obtain sewage spectral feature data; An analysis module, used to detect and analyze the sewage concentration based on the sewage spectral characteristic data to obtain sewage concentration parameters; A classification module, used to determine whether the sewage concentration parameter exceeds a preset sewage concentration range, and if so, to classify the sewage based on the sewage spectral characteristic data to obtain a sewage classification result; A formulation module is used to formulate a sewage management plan based on the sewage concentration parameters and sewage classification results.

[0014] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0015] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0016] The intelligent sewage detection method for municipal construction provided by the present invention comprises the following steps: using a preset spectrometer to perform real-time spectral scanning on sewage in a sewage pipe network in a target city to obtain sewage spectral scanning data; performing spectral feature extraction on the sewage spectral scanning data to obtain sewage spectral feature data; performing concentration detection and analysis on the sewage based on the sewage spectral feature data to obtain sewage concentration parameters; judging whether the sewage concentration parameter exceeds a preset sewage concentration range, and if so, performing sewage classification on the sewage based on the sewage spectral feature data to obtain sewage classification results; formulating a sewage management plan based on the sewage concentration parameters and the sewage classification results. Through the above-mentioned technical means, the technical problems that the traditional method is not only time-consuming and costly, but also difficult to realize real-time monitoring of sewage water quality are solved, and it is realized that according to whether the sewage concentration parameter exceeds the preset range, the system can automatically classify the excessive sewage, and formulate corresponding management measures accordingly, thereby avoiding errors that may be caused by human judgment, and making the sewage treatment process more scientific and systematic. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the steps of an intelligent sewage detection method for municipal construction in one embodiment of the present invention; Figure 2 It is a structural block diagram of an intelligent sewage detection system for municipal construction in one embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0018] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] like Figure 1 As shown, Figure 1 It is a schematic diagram of the steps of an intelligent sewage detection method for municipal construction in one embodiment of the present invention; In one embodiment of the present invention, a smart sewage detection method for municipal construction is provided, comprising the following steps: Step S1, using a preset spectrometer to perform real-time spectral scanning on sewage in a sewage pipe network in a target city to obtain sewage spectral scanning data.

[0021] Specifically, in the intelligent sewage detection method of municipal construction, using a preset spectrometer to perform real-time spectral scanning of sewage in the sewage pipe network in the target city is a key step in achieving real-time monitoring and analysis of sewage. Specifically, spectrometers need to be installed at key locations in the city's sewage pipe network. The selection of these locations is usually based on a comprehensive consideration of the sewage flow path, flow changes, and potential pollution sources to ensure that the most important and sensitive parts of the entire pipe network system can be covered. Once the spectrometer is installed, it can be connected to the central control system through optical fiber or other transmission media, so that when the sewage flows through these key points, the spectrometer will automatically start and perform continuous spectral scanning of the passing sewage. The process of spectral scanning involves emitting a light source of a specific wavelength to illuminate the sewage sample, and then collecting the reflected or transmitted light signals, which contain the spectral information unique to various substances in the sewage. For example, near an industrial area in a city, there may be a large number of chemical plants, pharmaceutical plants and other enterprises, and the wastewater discharged from these enterprises may contain high concentrations of harmful substances, such as heavy metal ions or toxic organic compounds. At this time, the spectrometer installed here can monitor the presence and concentration changes of these specific substances in real time. Once an abnormality is found, an alarm can be triggered immediately and the relevant data can be transmitted to the environmental regulatory department so that appropriate treatment measures can be taken in time. This real-time monitoring method can not only help city managers respond quickly to sudden pollution incidents, but also track and observe the changing trends of sewage water quality over a long period of time, providing a scientific basis for formulating more effective sewage management and treatment strategies.

[0022] Step S2, extracting spectral features from the sewage spectral scanning data to obtain sewage spectral feature data.

[0023] Specifically, in the intelligent sewage detection method of municipal construction, extracting spectral features from the sewage spectral scanning data to obtain sewage spectral feature data is one of the core links in the entire process. This process follows the spectral scanning. When the spectrometer completes the scanning of the sewage sample, a large amount of raw spectral data will be generated. These data contain the reflection or absorption characteristics of all substances in the sewage. In order to extract useful feature information from these complex data, special algorithms and technical means are needed for processing. For example, mathematical filtering, Fourier transform and other methods can be used to remove noise and improve the quality of data; then, by comparing with the known standard spectral database, key parameters such as characteristic peak position, intensity and shape of specific pollutants in sewage are identified. These parameters can directly reflect the type and concentration of each component in sewage. Taking the sewage pipe network near the industrial area as an example, it is assumed that the sewage in this area often contains copper ions, a common heavy metal pollutant. In the process of spectral feature extraction, the system will first pre-process the collected spectral data to eliminate the influence of background interference and other non-target signals, and then use the standard spectral spectrum of copper ions as a reference to find matching characteristic peaks. Once a successful match is made, the presence of copper ions in the sewage can be determined and its concentration value can be further calculated. In this way, not only can the copper ion content in the sewage be accurately identified, but also a variety of other pollutants can be monitored simultaneously, such as heavy metal ions such as lead and cadmium, or organic compounds such as benzene and phenols. This multi-component simultaneous monitoring capability provides strong technical support for comprehensive assessment of sewage quality and timely detection of potential pollution problems.

[0024] Step S3, performing concentration detection and analysis on the sewage based on the sewage spectral characteristic data to obtain sewage concentration parameters.

[0025] Specifically, the concentration of the sewage is detected and analyzed based on the sewage spectral characteristic data to obtain the sewage concentration parameter, which is another important step in the intelligent sewage detection method. After the spectral feature extraction is completed, the next step is to use these characteristic data to determine the specific concentration of each component in the sewage. This process involves comparing the extracted spectral features with the spectral data of known standard substances. By calculating the similarity or difference between the two, the concentration level of specific substances in the sewage can be inferred. For example, in the monitoring of sewage pipe networks near industrial areas, when the system identifies the characteristic peak of copper ions, it will compare it with the copper ion standard spectral curve pre-stored in the database. By analyzing the position and intensity of the characteristic peaks and combining certain mathematical models, such as linear regression, multivariate statistical analysis, etc., the actual concentration value of copper ions in the sewage is finally calculated. In addition, for some complex sewage samples, there may be multiple pollutants at the same time. At this time, more advanced data processing techniques, such as principal component analysis (PCA) and partial least squares (PLS), are needed to achieve simultaneous quantitative analysis of multiple components. These methods can effectively eliminate the influence of mutual interference on the basis of processing a large amount of spectral data and ensure the accuracy of the concentration determination of each component. For example, in the sewage pipe network of the same industrial zone, in addition to copper ions, there may also be zinc, nickel and other heavy metals. By comprehensively analyzing the characteristic spectra of these metal ions, their respective concentration parameters can be obtained at one time, thus providing comprehensive and accurate data support for the subsequent sewage classification and management plan. In this way, urban environmental management departments can monitor and manage sewage discharge more efficiently, ensure water quality safety, and promote sustainable development.

[0026] Step S4, determining whether the sewage concentration parameter exceeds a preset sewage concentration range, and if so, performing sewage classification on the sewage based on the sewage spectral characteristic data to obtain a sewage classification result.

[0027] Specifically, it is determined whether the sewage concentration parameter exceeds the preset sewage concentration range. If it exceeds, the sewage is classified based on the sewage spectral characteristic data to obtain the sewage classification result. This step is intended to ensure that sewage discharge meets environmental protection standards through intelligent analysis and provide a scientific basis for subsequent treatment. Specifically, after completing the detection of sewage concentration parameters, the system will compare these parameters with the preset safety threshold. The preset sewage concentration range here refers to the standard set according to the prescribed pollutant emission limit or the environmental carrying capacity of a specific area. When the monitored sewage concentration parameter is lower than or equal to the preset range, it indicates that the current sewage discharge is in a safe state and no additional measures are required; on the contrary, if the monitored concentration parameter exceeds the preset range, the system will automatically start further sewage classification procedures. For example, in the case of sewage pipe network monitoring near an industrial area, assuming that the system detects that the copper ion concentration in a sewage sample exceeds the preset safety threshold, the system will immediately trigger an alarm and call the previously extracted spectral characteristic data to perform a more detailed classification analysis of the sewage. This process involves a comprehensive assessment of the concentration, type and possible sources of all identified pollutants in the sewage, in order to determine which type of pollution the sewage belongs to, such as industrial wastewater, domestic sewage or mixed sewage. Based on these classification results, environmental regulatory authorities can take more targeted actions, such as notifying relevant companies to immediately stop discharging excessive sewage, sending professionals to the site to investigate the cause, or initiating emergency treatment measures to reduce the impact of pollution. Through such an intelligent monitoring and classification mechanism, not only can the problem of excessive sewage discharge be dealt with in a timely and effective manner, but it can also provide important support for the formulation of more scientific and reasonable sewage management, thereby ensuring the safety and health of the urban water environment.

[0028] Step S5, formulating a sewage management plan based on the sewage concentration parameters and the sewage classification results.

[0029] Specifically, formulating a sewage management plan based on the sewage concentration parameters and the sewage classification results is an important step in the intelligent sewage detection method to ensure water quality safety and rationally plan sewage treatment measures. After completing the detection and classification of sewage concentration parameters, the system will comprehensively consider the specific concentration of each pollutant in the sewage and its classification information, and formulate a specific management plan in combination with local environmental protection laws and regulations, sewage treatment capacity and environmental sensitivity. For example, in the sewage pipe network monitoring case near the aforementioned industrial zone, if the system detects that the copper ion concentration in a sewage sample is seriously exceeded, and through classification analysis, it is confirmed that the sewage mainly comes from a specific industrial enterprise, then the environmental regulatory department can take prompt action based on this information. On the one hand, a warning can be issued to the enterprise, requiring it to immediately stop the excessive discharge behavior and rectify it within a time limit; on the other hand, according to the specific nature and degree of pollution of the sewage, appropriate sewage treatment technology and facilities can be selected, such as adding chemical sedimentation tanks or biological treatment units, to ensure that the sewage meets safety standards before discharge. In addition, in response to long-standing pollution problems, a long-term environmental improvement plan can also be formulated, including strengthening source control, optimizing production processes, and improving sewage treatment efficiency, so as to achieve effective management and continuous improvement of sewage. In this way, not only can the current pollution problem be solved in a timely manner, but also a solid foundation can be laid for future environmental protection work.

[0030] In a specific embodiment, the use of a preset spectrometer to perform real-time spectral scanning on sewage in the sewage pipe network in the target city to obtain sewage spectral scanning data includes: Use a preset spectrometer to perform continuous spectral scanning on the sewage in the sewage pipe network in the target city to obtain preliminary spectral data; Performing background subtraction processing on the preliminary spectral data using a preset background spectral model to obtain spectral data after background subtraction; Performing baseline correction on the spectral data after background subtraction to obtain corrected spectral data; The corrected spectral data is subjected to a first-order derivative transformation to obtain sewage spectral scanning data.

[0031] Specifically, the use of a preset spectrometer to perform real-time spectral scanning of sewage in the sewage pipe network in the target city to obtain sewage spectral scanning data is a multi-step technical process aimed at improving the accuracy and reliability of sewage detection. First, the sewage in the sewage pipe network in the target city is continuously spectrally scanned using a preset spectrometer, which is a basic step to obtain high-quality spectral data. In this process, the spectrometer will be installed at key nodes of the sewage pipe network, such as industrial discharge outlets, residential drainage outlets and other places prone to pollution. The spectrometer emits light of a specific wavelength to illuminate the sewage samples passing through these nodes, and then collects the reflected or transmitted light signals, which contain the optical properties of various substances in the sewage. Since the flow of sewage is continuous, the spectrometer needs to be able to achieve continuous scanning to capture information about the changes in sewage components over time and space, and ensure the timeliness and representativeness of the monitoring results. Next, the preliminary spectral data is background-subtracted by a preset background spectral model to obtain spectral data after background subtraction. This is because in addition to pollutants, sewage samples may also contain other components such as water molecules and suspended particles, which will also scatter or absorb the incident light, thereby forming background signals in the spectral data. If not deducted, these background signals will interfere with the accurate identification of pollutant characteristics. To this end, it is necessary to establish a background spectrum model in advance, which is usually constructed based on the spectral characteristics of pure water or other known components. In actual operation, the system will compare the collected preliminary spectral data with the background spectrum model, and remove the influence of the background signal by subtraction operations, so as to obtain a clearer pollutant characteristic spectrum. Subsequently, the spectral data after background subtraction is baseline corrected to obtain corrected spectral data. Baseline correction refers to adjusting the spectral data to a unified baseline to eliminate baseline fluctuations caused by instrument drift, sample inhomogeneity, etc. Baseline fluctuations will cause the spectral peak to shift, affecting the accuracy of subsequent analysis. Commonly used baseline correction methods include polynomial fitting, iterative reweighted smoothing, etc. These methods can effectively straighten the baseline part in the spectral data, make the pollutant characteristic peak more prominent, and facilitate subsequent feature extraction and quantitative analysis. Finally, the corrected spectral data is subjected to a first-order derivative transformation to obtain sewage spectral scanning data. First-order derivative transformation is a commonly used signal processing technology that can amplify subtle changes in the spectrum, improve the resolution of characteristic peaks, and reduce noise interference. By calculating the first-order derivative of the spectral data, the originally flat spectral curve can be made sharper, making it easier to distinguish the characteristic peaks of different substances. For example, when monitoring copper ions in sewage pipes near industrial areas, even if the concentration of copper ions is low, its characteristic peak may overlap with the background signal of other substances and be difficult to distinguish.However, after the first-order derivative transformation, the slope of the characteristic peak of copper ions will change significantly, and the background signals of other substances will be more clearly separated, which provides more reliable data support for subsequent concentration detection and classification. Taking the monitoring of sewage pipe networks near industrial areas as an example, it is assumed that the sewage in this area often contains various heavy metal pollutants such as copper ions and lead ions. In practical applications, the spectrometer will continuously scan these sewage samples. After collecting preliminary spectral data, the background subtraction process is first used to remove the influence of background signals such as water molecules and suspended particles, and then the baseline correction is performed to ensure the stability and accuracy of the spectral characteristic peaks. Finally, the first-order derivative transformation is used to further enhance the resolution of the characteristic peaks. After this series of treatments, the system can accurately identify the presence of various heavy metal pollutants in sewage and determine their concentrations, providing a scientific basis for environmental regulatory authorities to take timely treatment measures. Through this intelligent monitoring method, not only can the occurrence of pollution accidents be effectively prevented, but also strong technical support can be provided for the long-term protection of urban water environment.

[0032] In a specific embodiment, the extracting spectral features of the sewage spectral scanning data to obtain sewage spectral feature data includes: The sewage spectrum scanning data is converted into a frequency domain by using a preset Fourier transform network model to obtain a frequency domain characteristic spectrum; Performing time series analysis on the sewage spectrum scanning data based on the frequency domain characteristic spectrum to obtain time series characteristics; The time series features are transformed into polar coordinates through a preset Gram angular field transformation network to obtain a GAF image; Extracting texture features of the GAF image to obtain image texture features; Using a preset continuous projection algorithm to extract feature bands from the image texture features to obtain band extraction features; The band extraction features are subjected to feature separation to obtain sewage spectral feature data.

[0033] Specifically, the spectral feature extraction of the sewage spectral scanning data to obtain the sewage spectral feature data is a multi-level and multi-step technical process, which aims to mine representative feature information from complex spectral data to support subsequent concentration detection and classification analysis. First, the sewage spectral scanning data is converted into the frequency domain through a preset Fourier transform network model to obtain a frequency domain feature spectrum. This process converts the spectral data in the time domain into a frequency domain representation, so that different frequency components can be separated, which helps to identify the periodic and non-periodic features in the spectrum. Fourier transform is a classic mathematical tool that can decompose complex signals into a superposition of a series of sine waves. For spectral data, this means that the absorption or reflection characteristics of different substances can be converted into different frequency components, providing a basis for subsequent feature extraction. The sewage spectral scanning data is subjected to time series analysis based on the frequency domain feature spectrum to obtain time series features. This step focuses on the trend and law of spectral data changing over time. In the sewage monitoring scenario, since the sewage composition may be affected by various factors, such as periodic changes in industrial production activities, changes in weather conditions, etc., the spectral data will also show certain time series characteristics. By performing time series analysis on the frequency domain feature map, these time series features, such as periodic fluctuations, mutation points, etc., can be extracted. These features are of great significance for understanding the changing laws of sewage components and predicting future changing trends. Next, the time series features are transformed into polar coordinates through the preset Gram Angular Field Transformation Network to obtain a GAF image. Gramian Angular Field (GAF) is a method for converting time series data into images, which can preserve the dynamic relationship and structural information in the time series. In this process, the time series data is first converted into angle values, and then a matrix is ​​constructed by calculating the cosine similarity between the angle values. This matrix is ​​the GAF image. The GAF image not only retains the main features of the original time series, but also can intuitively show the dynamic change process of the time series, which is very suitable for subsequent image processing and analysis. Texture feature extraction is performed on the GAF image to obtain image texture features. Texture feature refers to the regularity of pixel distribution in an image, which is an important feature in image recognition and classification. In GAF images, the spectral characteristics of different substances will appear as different texture patterns. By applying texture analysis methods, such as gray-level co-occurrence matrix (GLCM), local binary pattern (LBP), etc., texture features reflecting the characteristics of different substances can be extracted from GAF images. These features can not only reflect the static properties of the substance, but also reflect the dynamic characteristics of the substance over time, providing rich information for subsequent feature separation. The image texture features are extracted by using a preset continuous projection algorithm to obtain band extraction features.Continuum Removal (CR) is a commonly used spectral feature extraction method. It removes background interference and highlights the absorption features in the spectrum by projecting the spectral curve onto an idealized continuous background. In this step, the continuous projection algorithm is applied to the texture features of the GAF image, with the aim of selecting the most representative band features from a large number of texture features. These band features usually correspond to the absorption or reflection peaks of specific pollutants in sewage, and play a key role in identifying the types and concentrations of pollutants. Finally, the band extraction features are feature separated to obtain sewage spectral feature data. Feature separation refers to further refining the extracted band features and dividing them into multiple independent feature subsets, each of which corresponds to one or several specific pollutants in sewage. This process can be achieved through cluster analysis, principal component analysis and other methods. The final sewage spectral feature data not only contains the type information of the pollutants, but also contains the concentration information of each pollutant, providing accurate data support for subsequent concentration detection and classification. Taking the sewage pipe network monitoring near the industrial area as an example, it is assumed that the sewage in this area contains a variety of heavy metal pollutants such as copper ions and lead ions. In practical applications, the system first performs Fourier transform on the collected sewage spectral scanning data, converts the time domain data into frequency domain feature maps, and separates the frequency components of different substances. Then, through the time series analysis of the frequency domain feature maps, it can be found that the concentration changes of copper ions and lead ions are closely related to the periodic changes of industrial production activities. Then, the time series features are converted into image form through GAF transformation, and the texture feature extraction method is used to identify the unique texture patterns of copper ions and lead ions from the GAF image. Next, the continuous projection algorithm is used to extract the most representative band features from these texture features, which can accurately reflect the absorption characteristics of copper ions and lead ions. Finally, through feature separation, the system can divide these band features into feature subsets corresponding to copper ions and lead ions, thereby realizing the accurate identification and concentration determination of these two heavy metal pollutants. The application of this series of technical means not only improves the accuracy and efficiency of pollutant detection, but also provides a scientific basis for environmental regulatory authorities to take timely governance measures.

[0034] In a specific embodiment, the concentration detection and analysis of the sewage based on the sewage spectral characteristic data to obtain the sewage concentration parameter includes: Performing spectrum analysis on the sewage spectrum characteristic data to obtain spectrum distribution data; Based on the spectrum distribution data, each component in the sewage is identified to obtain a component identification label and a characteristic frequency corresponding to the component identification label; wherein the component identification label includes ammonia nitrogen, COD and total phosphorus, and the characteristic frequency is a key indicator related to the component concentration; By using the component identification tag, the characteristic peaks of each component in the sewage are identified and detected based on the spectrum distribution data to obtain the absorption peak quantification parameters; wherein the absorption peak quantification parameters include the characteristic peak frequency and amplitude of the component; Calculate the preliminary concentration data of each of the sewage based on the characteristic peak frequency, amplitude and characteristic frequency by using a preset Lambert-Beer law formula; The preliminary concentration data are supplemented with concentration data integrity to obtain complete concentration estimation data; By using a Bayesian inference algorithm, a multi-dimensional concentration evaluation calculation is performed on the sewage based on the complete concentration estimation data to obtain sewage concentration parameters.

[0035] Specifically, the concentration detection and analysis of the sewage based on the sewage spectral characteristic data to obtain the sewage concentration parameter is a multi-step technical process, which aims to accurately identify and quantify various components in the sewage from complex spectral characteristic data. This process first requires spectral analysis of the sewage spectral characteristic data to obtain spectral distribution data. Spectral analysis is to convert spectral data into frequency domain representation by mathematical methods, thereby revealing the distribution of different frequency components. In the sewage monitoring scenario, different pollutants have different absorption or reflection characteristics at specific frequencies, so spectrum analysis can preliminarily identify the types of pollutants that may exist in the sewage and their relative contents. For example, in the monitoring of sewage pipe networks near industrial areas, spectrum analysis can help identify the characteristic frequencies of common pollutants such as ammonia nitrogen, chemical oxygen demand (COD), and total phosphorus. Based on the spectral distribution data, the components in the sewage are identified to obtain component identification labels and characteristic frequencies corresponding to the component identification labels. Component identification is to determine the specific types of pollutants in the sewage by comparing the characteristic frequencies in the spectral distribution data based on the known pollutant spectral characteristic library. During this process, the system will match the spectrum distribution data with the spectral characteristics of the standard pollutants, thereby assigning a component identification label to each identified pollutant. These labels not only identify the type of pollutant, but also contain key indicators related to the concentration of the component-characteristic frequency. Taking ammonia nitrogen as an example, it will have obvious absorption peaks within a specific frequency range, and the frequencies of these absorption peaks are the characteristic frequencies of ammonia nitrogen. Similarly, COD and total phosphorus also have their own unique characteristic frequencies, and this information is crucial for subsequent concentration detection. Through the component identification label, the characteristic peaks of each component in the sewage are identified and detected based on the spectrum distribution data to obtain the absorption peak quantification parameters. Characteristic peak identification refers to accurately finding the position corresponding to the characteristic frequency of the pollutant from the spectrum distribution data, that is, the position of the characteristic peak. Characteristic peak detection is to further measure the amplitude of these characteristic peaks to quantify the concentration information of the pollutant. The absorption peak quantification parameters include the characteristic peak frequency and amplitude of the component, which are the basis for calculating the concentration of the pollutant. For example, when monitoring the ammonia nitrogen concentration in the sewage network in an industrial area, the system will identify the ammonia nitrogen composition label, search for the characteristic peak corresponding to its characteristic frequency in the spectrum distribution data, and then measure the amplitude of the characteristic peak to obtain the quantitative parameters of the ammonia nitrogen absorption peak. The preliminary concentration data of each sewage is calculated based on the characteristic peak frequency, amplitude and characteristic frequency using the preset Lambert-Beer law formula. The Lambert-Beer law is a basic principle commonly used in spectral analysis, which describes the linear relationship between absorbance and solution concentration when light passes through a solution. In this step, the system will use the Lambert-Beer law formula ,in represents the absorbance (i.e. the amplitude of the characteristic peak), is the molar absorptivity, is the solution concentration, is the optical path length. The preliminary concentration data of each component in the sewage can be calculated by combining the known characteristic peak frequency, amplitude and characteristic frequency with the experimentally measured molar absorption coefficient and optical path length. The preliminary concentration data is supplemented with concentration data integrity to obtain complete concentration estimation data. The preliminary concentration data may be missing or inaccurate due to various reasons (such as instrument error, background interference, etc.), so it is necessary to use data completion methods to improve the integrity and accuracy of the data. Commonly used completion methods include interpolation and regression analysis. For example, when monitoring the COD concentration in the sewage network in an industrial area, if the data of a certain time period in the preliminary concentration data is missing, the data of the adjacent time periods can be linearly interpolated to fill the missing values, thereby obtaining complete time series concentration data. Using the Bayesian inference algorithm, the sewage is evaluated and calculated in a multi-dimensional concentration based on the complete concentration estimation data to obtain sewage concentration parameters. Bayesian inference is a method based on probability theory that can estimate unknown parameters by combining prior knowledge and observation data. In this step, the system uses the Bayesian inference algorithm, combined with the historical concentration data (prior knowledge) of each component in the sewage and the current complete concentration estimation data (observation data), to perform multi-dimensional evaluation and calculation of the concentrations of ammonia nitrogen, COD, total phosphorus and other components in the sewage. Through Bayesian inference, the true concentration of each component can be estimated more accurately, and the corresponding confidence interval can be given, thereby providing reliable decision-making support for environmental regulatory departments. Taking the sewage pipe network monitoring near the industrial area as an example, it is assumed that the system detects that the concentrations of ammonia nitrogen, COD, and total phosphorus in a sewage sample have increased. First, the system will perform spectral analysis on the collected sewage spectral feature data to identify the characteristic frequencies of ammonia nitrogen, COD, and total phosphorus. Then, through characteristic peak identification and detection, the absorption peak quantitative parameters of these components are obtained, including characteristic peak frequency and amplitude. Next, the preliminary concentration data is calculated using the Lambert-Beer law formula, and the preliminary concentration data is supplemented with integrity to ensure the continuity and accuracy of the data. Finally, the complete concentration estimation data is evaluated and calculated in multiple dimensions by the Bayesian inference algorithm to obtain the final sewage concentration parameters. The application of this series of technical means not only improves the accuracy and reliability of pollutant concentration detection, but also provides a scientific basis for environmental regulatory departments to take timely control measures, helping to ensure the safety and health of the urban water environment.

[0036] In a specific embodiment, the sewage classification based on the sewage spectral characteristic data to obtain the sewage classification result includes: By using a preset multi-layer perceptron model, a time series analysis is performed on the components in the sewage based on the sewage spectral characteristic data to obtain component concentration change data; Obtaining environmental data corresponding to the surrounding environmental location near the sewage pipe network and the surrounding environmental location, and performing source analysis on the component concentration change data based on the environmental data by a preset parallel factor analysis method to obtain component source data; Combining the component source data and the sewage spectral characteristic data to obtain sewage combination data; The singular value decomposition algorithm is used to perform principal component analysis on the sewage combination data to obtain the sewage principal component feature vector; Performing kernel density estimation on the sewage principal component eigenvector to obtain sewage characteristic probability density distribution; wherein the sewage characteristic probability density distribution includes the probability density value corresponding to each eigenvalue, the value range of the eigenvalue and the kernel function bandwidth; The sewage is classified based on the probability density distribution of the sewage characteristics and the environmental data corresponding to the surrounding environment location to obtain a sewage classification result; wherein the sewage classification result includes domestic sewage, industrial sewage and agricultural sewage.

[0037] Specifically, the sewage classification based on the sewage spectral characteristic data to obtain the sewage classification result is a complex and sophisticated technical process, which aims to accurately identify the source and type of sewage through multi-step analysis and processing. This process first requires a preset multi-layer perceptron model to perform a time series analysis on the components in the sewage based on the sewage spectral characteristic data to obtain component concentration change data. Multilayer Perceptron (MLP) is a feedforward neural network that can learn the nonlinear relationship between input data and output data. In the sewage monitoring scenario, the MLP model can be used to analyze the change trend of sewage spectral characteristic data over time, thereby revealing the change law of the concentration of each component in the sewage. For example, in the monitoring of sewage pipe networks near industrial areas, the MLP model is used to perform a time series analysis on sewage spectral characteristic data, and the changes in the concentration of components such as ammonia nitrogen, COD, and total phosphorus in different time periods can be found. These changes may be caused by periodic changes in industrial production activities. Obtain the surrounding environment location near the sewage pipe network and the environmental data corresponding to the surrounding environment location, and perform source analysis on the component concentration change data based on the environmental data through the preset parallel factor analysis method to obtain the component source data. Environmental data include meteorological data, topography, land use type, etc., which can provide external influencing factors of sewage composition changes. Parallel Factor Analysis (PARAFAC) is a multidimensional data analysis method that can extract potential factors from complex multidimensional data sets. In this step, the PARAFAC method is used to analyze the relationship between sewage component concentration change data and environmental data, so as to identify the possible sources of sewage components. For example, through PARAFAC analysis, it can be found that the increase in ammonia nitrogen concentration over a period of time is related to the production activities of a nearby factory, while the change in COD concentration has a strong correlation with rainfall. The component source data and the sewage spectral feature data are combined to obtain sewage combination data. This process is to integrate the component source data with the sewage spectral feature data to form a new multidimensional data set. The sewage combination data not only contains the concentration information of each component in the sewage, but also contains the possible source information of these components, which provides a rich data basis for the subsequent principal component analysis. For example, in the monitoring of sewage pipe networks near industrial areas, sewage combination data can include ammonia nitrogen concentration, source factory number, production activity time and other information. The principal component analysis of sewage combination data was performed using the singular value decomposition algorithm to obtain the sewage principal component eigenvector. Singular value decomposition (SVD) is a commonly used matrix decomposition method that can decompose a matrix into the product of three matrices to extract the main components in the data.In this step, the SVD algorithm is used to perform principal component analysis on the sewage combination data, with the aim of extracting the most important eigenvectors from the multidimensional data. These principal component eigenvectors can represent the main change trends and structural information of the sewage combination data, and provide important feature support for subsequent classification. For example, through principal component analysis, the two most important eigenvectors, the ammonia nitrogen concentration in the sewage and the source factory number, can be extracted. The sewage principal component eigenvector is subjected to kernel density estimation to obtain the sewage characteristic probability density distribution. Kernel density estimation (KDE) is a non-parametric probability density estimation method that can estimate the probability density function of an unknown distribution based on sample data. In this step, the KDE method is used to estimate the probability density distribution of the sewage principal component eigenvector, thereby revealing the distribution law of the sewage characteristic value. The sewage characteristic probability density distribution includes the probability density value corresponding to each eigenvalue, the value range of the eigenvalue and the kernel function bandwidth. For example, through kernel density estimation, the probability density distribution of ammonia nitrogen concentration under different source factory numbers can be obtained, thereby finding that the ammonia nitrogen concentration in the sewage discharged by some factories is higher. Based on the probability density distribution of the sewage characteristics and the environmental data corresponding to the surrounding environment, the sewage is classified to obtain the sewage classification result. The sewage classification results include domestic sewage, industrial sewage and agricultural sewage. In this process, the system will conduct a multi-dimensional comprehensive evaluation of the sewage based on the probability density distribution of the sewage characteristics and the environmental data to determine the specific type of sewage. For example, in the monitoring of the sewage network near the industrial area, if the characteristic values ​​of the ammonia nitrogen concentration and the source factory number of a sewage sample fall within a certain probability density interval, and the interval is mainly caused by industrial activities, then the system will classify the sewage sample as industrial sewage. Similarly, if the characteristic value of a sewage sample is mainly related to the living activities in the residential area, it will be classified as domestic sewage; if it is related to farmland irrigation activities, it will be classified as agricultural sewage. Taking the monitoring of the sewage network near the industrial area as an example, assuming that the system detects a high ammonia nitrogen concentration in a sewage sample, the time series analysis is performed through the multi-layer perceptron model, and it is found that the concentration change is consistent with the production activity cycle of a factory. Next, we obtained environmental data around the factory, including meteorological data, land use type, etc., and used the parallel factor analysis method to analyze the source of the component concentration change data, confirming that the main source of ammonia nitrogen was the factory. Then, we combined the component source data with the sewage spectral feature data to form sewage combination data. We used the singular value decomposition algorithm to perform principal component analysis on the sewage combination data, and extracted two main eigenvectors: ammonia nitrogen concentration and source factory number. We performed kernel density estimation on these eigenvectors to obtain the sewage characteristic probability density distribution, and found that the eigenvalues ​​of the sewage sample were mainly concentrated in the probability density interval caused by industrial activities. Finally, based on the sewage characteristic probability density distribution and environmental data, the system classified the sewage sample into industrial sewage.The application of this series of technical means not only improves the accuracy and reliability of sewage classification, but also provides scientific decision-making support for environmental regulatory departments, helping to achieve effective management and treatment of different types of sewage.

[0038] In a specific embodiment, the sewage is classified based on the probability density distribution of the sewage characteristics and the environmental data corresponding to the surrounding environment location to obtain the sewage classification result, including: Using a preset restricted Boltzmann machine algorithm, potential pattern mining is performed on the sewage based on the probability density distribution of the sewage characteristics to obtain sewage potential pattern data; Performing multivariate time series analysis on the environmental data corresponding to the surrounding environment location to extract environmental change trend data; Using a long short-term memory network, the sewage potential pattern data is adjusted based on the environmental change trend data to obtain adjusted sewage potential pattern data; Performing matrix decomposition on the sewage spectral feature data to obtain a sewage spectral decomposition result; wherein the sewage spectral decomposition result includes a base matrix and a coefficient matrix, the base matrix is ​​used for the basic components of the sewage spectral feature, and the coefficient matrix is ​​used to represent the weight of each basic component in the sewage; The sewage is classified based on the sewage spectral decomposition result and the adjusted sewage potential pattern data to obtain a sewage classification result.

[0039] Specifically, the classification of the sewage based on the probability density distribution of the sewage characteristics and the environmental data corresponding to the surrounding environment location to obtain the sewage classification result is a complex process of multi-step and multi-technology integration, which aims to accurately identify the source and type of sewage by deeply mining and analyzing data. First, the preset restricted Boltzmann machine algorithm is used to mine the potential pattern of the sewage based on the probability density distribution of the sewage characteristics to obtain the sewage potential pattern data. Restricted Boltzmann Machine (RBM) is an unsupervised learning algorithm that can learn potential feature representations from complex data. In the sewage monitoring scenario, the RBM algorithm can identify the potential patterns of different components in the sewage by learning the probability density distribution of sewage characteristics. These patterns reflect the distribution characteristics of sewage components in different time and space. For example, in the monitoring of sewage pipe networks near industrial areas, the RBM algorithm can mine the potential patterns of components such as ammonia nitrogen, COD, and total phosphorus in the emission cycle of a specific factory. Multivariate time series analysis is performed on the environmental data corresponding to the surrounding environment location to extract environmental change trend data. Environmental data include meteorological data, topography, land use type, etc., which can provide external influencing factors of sewage composition changes. Multivariate time series analysis is a statistical method used to analyze the trends and interrelationships of multiple variables over time. In this step, by performing multivariate time series analysis on environmental data, the changing trends of environmental factors such as temperature, humidity, and rainfall can be extracted. These trend data are helpful to understand the external driving factors of sewage composition changes. For example, through multivariate time series analysis, it can be found that the increase in rainfall is often accompanied by an increase in COD concentration, which may be due to rainwater washing ground pollutants into the sewage network. Using a long short-term memory network, the sewage potential pattern data is adjusted based on the environmental change trend data to obtain the adjusted sewage potential pattern data. Long Short-Term Memory (LSTM) is a special recurrent neural network that can effectively handle long dependency problems in time series data. In this step, the LSTM network is used to adjust the sewage potential pattern data in combination with the environmental change trend data, with the aim of eliminating the interference of environmental factors on the changes in sewage composition and making the potential pattern more accurately reflect the intrinsic characteristics of sewage composition. For example, the effect of rainfall on COD concentration changes can be adjusted through the LSTM network, so as to more accurately identify the contribution of industrial emissions to COD concentration. The sewage spectral feature data is subjected to matrix decomposition to obtain the sewage spectral decomposition result. Matrix decomposition is a mathematical method used to decompose a matrix into the product of two or more matrices to extract the basic components in the data. In sewage monitoring, matrix decomposition can decompose complex sewage spectral feature data into a combination of basic components.Specifically, the sewage spectrum decomposition results include a base matrix and a coefficient matrix. The base matrix is ​​used to represent the basic components of the sewage spectrum characteristics, and the coefficient matrix is ​​used to represent the weight of each basic component in the sewage. For example, the sewage spectrum characteristic data can be decomposed into several main basic components, such as ammonia nitrogen, COD, total phosphorus, etc. through matrix decomposition. The weight of each basic component reflects its relative content in the sewage. Based on the sewage spectrum decomposition results and the adjusted sewage potential pattern data, the sewage is classified to obtain the sewage classification results. This process combines the sewage spectrum decomposition results with the adjusted sewage potential pattern data to determine the specific type of sewage through comprehensive analysis. For example, in the sewage pipe network monitoring near the industrial area, if the spectral characteristics of a sewage sample are mainly composed of ammonia nitrogen and COD, and its potential pattern is highly consistent with the discharge cycle of a factory, then the system will classify the sewage sample as industrial sewage. Similarly, if the spectral characteristics of a sewage sample are mainly composed of domestic pollutants, and its potential pattern is consistent with the life activity cycle of the residential area, it will be classified as domestic sewage; if there are more components related to farmland irrigation activities, it will be classified as agricultural sewage. Taking the monitoring of sewage pipe networks near industrial areas as an example, assuming that the system detects that ammonia nitrogen and COD concentrations in a sewage sample are high, the restricted Boltzmann machine algorithm is used to mine the potential pattern of the sewage characteristic probability density distribution, and it is found that the potential pattern of the sewage sample is consistent with the emission cycle of a certain factory. Next, a multivariate time series analysis is performed on the environmental data around the factory to extract the changing trends of environmental factors such as rainfall and temperature. Then, the long short-term memory network is used to process the environmental change trend data, adjust the sewage potential pattern data, and eliminate the influence of environmental factors. Next, the sewage spectral feature data is matrix decomposed to obtain the basis matrix and coefficient matrix, and it is found that ammonia nitrogen and COD are the main basic components. Finally, based on the sewage spectral decomposition results and the adjusted sewage potential pattern data, the system classifies the sewage sample into industrial sewage. The application of this series of technical means not only improves the accuracy and reliability of sewage classification, but also provides scientific decision-making support for environmental regulatory departments, which helps to achieve effective management and governance of different types of sewage.

[0040] In a specific embodiment, the formulating of a sewage management plan based on the sewage concentration parameter and the sewage classification result includes: Constructing risk assessment indicators based on the sewage concentration parameters and sewage classification results to obtain a risk assessment indicator system; Calling a sewage pipe network image in a sewage pipe network in a database, analyzing the sewage pipe network based on the sewage pipe network image to obtain an analysis result; wherein the analysis result includes a topological structure of the sewage pipe and sewage flow data; Based on the topological structure and sewage flow data, graph theory analysis is performed to obtain the connectivity and flow distribution characteristics of the sewage pipe network; Based on the risk assessment indicator system, the connectivity of the sewage pipe network and the flow distribution characteristics, a sewage risk propagation simulation is performed on the sewage to obtain the sewage risk propagation path and impact range; Formulate treatment measures based on the sewage risk transmission path and impact range, and obtain a preliminary set of treatment measures; The preliminary treatment measure set is optimized according to the sewage classification result to obtain a sewage management plan.

[0041] Specifically, the formulation of the sewage management plan based on the sewage concentration parameters and the sewage classification results is a comprehensive process of multi-step and multi-technical integration, which aims to formulate effective sewage management measures through comprehensive risk assessment and scientific analysis methods. This process first requires the construction of risk assessment indicators based on the sewage concentration parameters and the sewage classification results to obtain a risk assessment indicator system. The risk assessment indicator system is the basis for assessing sewage risks, which covers multiple dimensions such as the concentration of each component in the sewage, the classification type of sewage, and the discharge location. For example, in the monitoring of the sewage network near the industrial area, the system will construct a risk assessment system including indicators such as pollutant concentration, discharge location, and discharge time based on the concentration parameters of components such as ammonia nitrogen, COD, and total phosphorus, and the result that the sewage is classified as industrial sewage. These indicators can fully reflect the risk level of sewage and provide a basis for subsequent analysis and management. The sewage network image in the sewage network in the database is called, and the sewage network is analyzed based on the sewage network image to obtain the analysis result. This process involves calling pre-stored sewage network image data, which usually includes information such as the layout diagram of the sewage network, pipe diameter, and connection points. Through image processing and analysis technology, the topological structure and sewage flow data of the sewage pipe network can be extracted. For example, the system can extract the connection relationship and diameter size of the pipes from the sewage pipe network image through methods such as image segmentation and edge detection, thereby constructing a topological structure diagram of the sewage pipe network. At the same time, combined with the data of the flow sensor, the flow distribution of each pipe can be obtained. Based on the topological structure and sewage flow data, graph analysis is performed to obtain the connectivity and flow distribution characteristics of the sewage pipe network. Graph analysis is a mathematical method used to study the structure and properties of a graph. In sewage pipe network management, graph analysis can reveal the connectivity of the sewage pipe network, that is, which pipes are connected and which are isolated. In addition, by analyzing the flow data, the flow of sewage in the pipe network can be understood, including the main flow direction, flow peak and other characteristics. For example, through graph analysis, it can be found that a certain pipe is the main channel connecting multiple communities, its flow is large, and the flow peak is obvious during the morning and evening peak hours. Based on the risk assessment indicator system, the connectivity and flow distribution characteristics of the sewage pipe network, the sewage risk propagation simulation is performed on the sewage to obtain the sewage risk propagation path and impact range. Sewage risk propagation simulation is to predict the propagation path and impact range of sewage in the pipe network through computer simulation technology. This process requires the establishment of a sewage propagation model in combination with the risk assessment indicator system, the connectivity and flow distribution characteristics of the sewage pipe network. For example, assuming that the sewage discharged by a factory contains a high concentration of ammonia nitrogen, the propagation path of the sewage in the pipe network and the areas that may be affected can be predicted through risk propagation simulation. This information helps to take preventive measures in advance to avoid the spread of pollution. Treatment measures are formulated for the sewage risk propagation path and impact range, and a preliminary set of treatment measures is obtained.The set of treatment measures includes a series of prevention and control measures for sewage risks, such as strengthening monitoring, adding treatment facilities, adjusting discharge time, etc. For example, if the simulation results show that a certain pipeline is the main transmission path, monitoring equipment can be added at the key nodes of the pipeline to monitor water quality changes in real time; if a certain area is greatly affected, sewage treatment facilities can be added to improve treatment capacity. According to the sewage classification results, the preliminary treatment measures are optimized to obtain a sewage management plan. This process requires the optimization of preliminary treatment measures in combination with the sewage classification results to ensure the pertinence and effectiveness of the measures. For example, if sewage is classified as industrial sewage, relevant enterprises can be required to improve production processes and reduce pollutant emissions; if sewage is classified as domestic sewage, the management of sewage discharge in residential areas can be strengthened and the coverage of sewage treatment facilities can be increased. Through these optimization measures, sewage risks can be better controlled and the water environment can be protected. Taking the monitoring of sewage pipe networks near industrial areas as an example, it is assumed that the system detects that the concentrations of ammonia nitrogen and COD in sewage discharged by a factory are high and classifies it as industrial sewage. First, the system will build a risk assessment indicator system based on these concentration parameters and classification results, including indicators such as pollutant concentration, discharge location, and discharge time. Then, the sewage network image in the database was called to extract the topological structure and flow data of the sewage network. Through graph theory analysis, it was found that the sewage discharged by the factory mainly flows to the downstream area through a trunk pipeline, and the flow is large during peak hours in the morning and evening. Based on this information, the system simulates the spread of sewage risks and predicts the spread path and impact range of sewage. According to the simulation results, a series of treatment measures were initially formulated, such as adding monitoring equipment at key nodes of the trunk pipeline and requiring the factory to adjust the discharge time. Finally, according to the result that the sewage was classified as industrial sewage, the preliminary treatment measures were optimized, requiring the factory to improve the production process and reduce pollutant emissions. Through this series of comprehensive measures, the system can effectively control sewage risks and protect the safety and health of the urban water environment.

[0042] The above describes the intelligent sewage detection method for municipal construction in the embodiment of the present invention. The following describes the intelligent sewage detection system for municipal construction in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, an intelligent sewage detection system for municipal construction includes: The scanning module 21 is used to perform real-time spectral scanning of sewage in the sewage pipe network in the target city using a preset spectrometer to obtain sewage spectral scanning data; An extraction module 22 is used to extract spectral features from the sewage spectral scanning data to obtain sewage spectral feature data; An analysis module 23 is used to detect and analyze the concentration of the sewage based on the sewage spectral characteristic data to obtain sewage concentration parameters; The classification module 24 is used to determine whether the sewage concentration parameter exceeds a preset sewage concentration range. If it exceeds, the sewage is classified based on the sewage spectral characteristic data to obtain a sewage classification result; The formulation module 25 is used to formulate a sewage management plan based on the sewage concentration parameters and the sewage classification results.

[0043] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the above method embodiment, which will not be repeated here.

[0044] Reference Figure 3 The present invention also provides a computer device in an embodiment, wherein the internal structure of the computer device can be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0045] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0046] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0047] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0048] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0049] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An intelligent sewage detection method for municipal construction, characterized in that: The following steps are involved: Use a preset spectrometer to perform real-time spectral scanning on sewage in the sewage pipe network in the target city to obtain sewage spectral scanning data; Extracting spectral features from the sewage spectral scanning data to obtain sewage spectral feature data; Performing concentration detection and analysis of the sewage based on the sewage spectral characteristic data to obtain sewage concentration parameters; Determine whether the sewage concentration parameter exceeds a preset sewage concentration range, and if so, perform sewage classification of the sewage based on the sewage spectral characteristic data to obtain a sewage classification result; A sewage management plan is formulated based on the sewage concentration parameters and sewage classification results.

2. The intelligent sewage detection method for municipal construction according to claim 1 is characterized in that: The method of using a preset spectrometer to perform real-time spectral scanning on sewage in the sewage pipe network in the target city to obtain sewage spectral scanning data includes: Use a preset spectrometer to perform continuous spectral scanning on the sewage in the sewage pipe network in the target city to obtain preliminary spectral data; Performing background subtraction processing on the preliminary spectral data using a preset background spectral model to obtain spectral data after background subtraction; Performing baseline correction on the spectral data after background subtraction to obtain corrected spectral data; The corrected spectral data is subjected to a first-order derivative transformation to obtain sewage spectral scanning data.

3. The intelligent sewage detection method for municipal construction according to claim 1 is characterized in that: The extracting of spectral features from the sewage spectral scanning data to obtain sewage spectral feature data includes: The sewage spectrum scanning data is converted into a frequency domain by using a preset Fourier transform network model to obtain a frequency domain characteristic spectrum; Performing time series analysis on the sewage spectrum scanning data based on the frequency domain characteristic spectrum to obtain time series characteristics; The time series features are transformed into polar coordinates through a preset Gram angular field transformation network to obtain a GAF image; Extracting texture features of the GAF image to obtain image texture features; Using a preset continuous projection algorithm to extract feature bands from the image texture features to obtain band extraction features; The band extraction features are subjected to feature separation to obtain sewage spectral feature data.

4. The intelligent sewage detection method for municipal construction according to claim 1 is characterized in that: The sewage concentration detection and analysis based on the sewage spectral characteristic data to obtain sewage concentration parameters includes: Performing spectrum analysis on the sewage spectrum characteristic data to obtain spectrum distribution data; Based on the spectrum distribution data, each component in the sewage is identified to obtain a component identification label and a characteristic frequency corresponding to the component identification label; wherein the component identification label includes ammonia nitrogen, COD and total phosphorus, and the characteristic frequency is a key indicator related to the component concentration; By using the component identification tag, the characteristic peaks of each component in the sewage are identified and detected based on the spectrum distribution data to obtain the absorption peak quantification parameters; wherein the absorption peak quantification parameters include the characteristic peak frequency and amplitude of the component; Calculate the preliminary concentration data of each of the sewage based on the characteristic peak frequency, amplitude and characteristic frequency by using a preset Lambert-Beer law formula; The preliminary concentration data are supplemented with concentration data integrity to obtain complete concentration estimation data; By using a Bayesian inference algorithm, a multi-dimensional concentration evaluation calculation is performed on the sewage based on the complete concentration estimation data to obtain sewage concentration parameters.

5. The intelligent sewage detection method for municipal construction according to claim 1 is characterized in that: The sewage classification based on the sewage spectral characteristic data to obtain the sewage classification result includes: By using a preset multi-layer perceptron model, a time series analysis is performed on the components in the sewage based on the sewage spectral characteristic data to obtain component concentration change data; Obtaining environmental data corresponding to the surrounding environmental location near the sewage pipe network and the surrounding environmental location, and performing source analysis on the component concentration change data based on the environmental data by a preset parallel factor analysis method to obtain component source data; Combining the component source data and the sewage spectral characteristic data to obtain sewage combination data; The singular value decomposition algorithm is used to perform principal component analysis on the sewage combination data to obtain the sewage principal component feature vector; Performing kernel density estimation on the sewage principal component eigenvector to obtain sewage characteristic probability density distribution; wherein the sewage characteristic probability density distribution includes the probability density value corresponding to each eigenvalue, the value range of the eigenvalue and the kernel function bandwidth; The sewage is classified based on the probability density distribution of the sewage characteristics and the environmental data corresponding to the surrounding environment location to obtain a sewage classification result; wherein the sewage classification result includes domestic sewage, industrial sewage and agricultural sewage.

6. The intelligent sewage detection method for municipal construction according to claim 5 is characterized in that: The classifying the sewage based on the probability density distribution of the sewage characteristics and the environmental data corresponding to the surrounding environment position to obtain the sewage classification result includes: Using a preset restricted Boltzmann machine algorithm, potential pattern mining is performed on the sewage based on the probability density distribution of the sewage characteristics to obtain sewage potential pattern data; Performing multivariate time series analysis on the environmental data corresponding to the surrounding environment location to extract environmental change trend data; Using a long short-term memory network, the sewage potential pattern data is adjusted based on the environmental change trend data to obtain adjusted sewage potential pattern data; Performing matrix decomposition on the sewage spectral feature data to obtain a sewage spectral decomposition result; wherein the sewage spectral decomposition result includes a base matrix and a coefficient matrix, the base matrix is ​​used for the basic components of the sewage spectral feature, and the coefficient matrix is ​​used to represent the weight of each basic component in the sewage; The sewage is classified based on the sewage spectral decomposition result and the adjusted sewage potential pattern data to obtain a sewage classification result.

7. The intelligent sewage detection method for municipal construction according to claim 1 is characterized in that: The formulating of a sewage management plan based on the sewage concentration parameter and the sewage classification result includes: Constructing risk assessment indicators based on the sewage concentration parameters and sewage classification results to obtain a risk assessment indicator system; Calling a sewage pipe network image in a sewage pipe network in a database, analyzing the sewage pipe network based on the sewage pipe network image to obtain an analysis result; wherein the analysis result includes a topological structure of the sewage pipe and sewage flow data; Based on the topological structure and sewage flow data, graph theory analysis is performed to obtain the connectivity and flow distribution characteristics of the sewage pipe network; Based on the risk assessment indicator system, the connectivity of the sewage pipe network and the flow distribution characteristics, a sewage risk propagation simulation is performed on the sewage to obtain the sewage risk propagation path and impact range; Formulate treatment measures based on the sewage risk transmission path and impact range, and obtain a preliminary set of treatment measures; The preliminary treatment measure set is optimized according to the sewage classification result to obtain a sewage management plan.

8. An intelligent sewage detection system for municipal construction, characterized in that: include: A scanning module is used to perform real-time spectral scanning of sewage in the sewage pipe network in the target city using a preset spectrometer to obtain sewage spectral scanning data; An extraction module, used for extracting spectral features from the sewage spectral scanning data to obtain sewage spectral feature data; An analysis module, used to detect and analyze the sewage concentration based on the sewage spectral characteristic data to obtain sewage concentration parameters; A classification module, used to determine whether the sewage concentration parameter exceeds a preset sewage concentration range, and if so, to classify the sewage based on the sewage spectral characteristic data to obtain a sewage classification result; A formulation module is used to formulate a sewage management plan based on the sewage concentration parameters and sewage classification results.

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

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images