Dynamic simulation method for lake and reservoir sewage sludge deposition process

Abstract

The invention relates to the technical field of sewage treatment, in particular to a dynamic simulation method for a lake and reservoir sewage sludge deposition process. The method comprises the following steps: collecting lake and reservoir hydrological station data, carrying out abnormal value filtering and missing value completion, generating a complete hydrological sequence, carrying out water flow state subdivision to obtain flow state hierarchical data, reconstructing a multi-layer hydrodynamic field, forming a hydrological dynamic field set, carrying out suspended matter analysis, and obtaining a hydrological dynamic field model; the method comprises the following steps of: extracting sewage sludge data, deconstructing transportation characteristics of the sewage sludge data, generating transportation dynamic data, performing water body stress tensor decomposition to obtain stress distribution data, performing substance transportation simulation on a complete hydrological sequence by using the stress distribution data, establishing a multi-scale sedimentation dynamic model, acquiring water body sedimentation change data in real time, comparing the water body sedimentation change data with the model, and if inconsistency exists, judging whether the water body sedimentation change data is consistent or not. If yes, the model is finely adjusted in real time, and an optimized multi-scale deposition dynamic model is generated. The more accurate dynamic simulation method for the lake and reservoir sewage and sludge deposition process is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of sewage treatment, and in particular to a method for kinetic simulation of lake sewage sludge sedimentation process. Background Art

[0002] As important water resources and ecosystems, lakes and reservoirs carry multiple functions such as water quality regulation, ecological protection and human activities. However, with the acceleration of urbanization and the frequency of human activities, the water environment of lakes and reservoirs is facing increasingly severe pollution problems, especially the increasingly serious siltation of sewage and sludge, which leads to a series of environmental problems such as water quality deterioration, biodiversity decline and eutrophication of water bodies. Traditional water quality monitoring and management methods are often difficult to timely and accurately reflect the dynamic changes of lakes and reservoirs, and lack effective prediction and regulation mechanisms. At present, the research on sewage and sludge in lakes and reservoirs is mostly focused on static analysis, lacking an in-depth understanding of hydrodynamics, and the relationship between water flow state and material transport is relatively weak. Many studies have failed to fully consider the impact of the hydrodynamic field, resulting in inaccurate simulation and prediction of siltation. In addition, the existing siltation models often cannot be updated in real time, and it is difficult to adapt to the rapid changes in the water environment, resulting in lagging and unscientific management measures. Summary of the invention

[0003] Based on this, it is necessary to provide a kinetic simulation method for the sedimentation process of lake sewage sludge to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for kinetic simulation of lake sewage sludge sedimentation process comprises the following steps:

[0005] Step S1: Collecting lake and reservoir hydrological station data; filtering the lake and reservoir hydrological station data for outliers to obtain cleaned hydrological data; completing missing value processing on the cleaned hydrological data to generate a complete hydrological sequence;

[0006] Step S2: performing water flow pattern segmentation on the complete hydrological sequence to obtain flow pattern layered data; performing multi-layer hydrodynamic field reconstruction on the flow pattern layered data to generate a hydrological dynamic field set;

[0007] Step S3: Perform suspended solids analysis on the complete hydrological sequence based on the hydrological dynamic field set to obtain sewage sludge data; perform transport characteristic deconstruction on the sewage sludge data to generate transport dynamics data;

[0008] Step S4: decomposing the water body stress tensor based on the transport dynamics data to obtain stress distribution data; simulating the material transport of the complete hydrological sequence according to the stress distribution data to generate a multi-scale sedimentation dynamics model;

[0009] Step S5: Collect real-time sedimentation change data of water bodies; compare the sedimentation change data of water bodies and the multi-scale sedimentation dynamics model. When the sedimentation changes are inconsistent, fine-tune the multi-scale sedimentation dynamics model in real time according to the real-time sedimentation change data of water bodies to generate an optimized multi-scale sedimentation dynamics model.

[0010] The present invention ensures the accuracy and reliability of data by collecting data from lake and reservoir hydrological stations and filtering outliers. The cleaned hydrological data generates a complete hydrological sequence by filling in missing values, which provides a solid foundation for subsequent analysis. The implementation of flow pattern dissection enables the formation of flow pattern stratified data, thereby enhancing the understanding and description of water flow characteristics. The reconstruction of multi-layer hydrodynamic fields enables the generation of hydrological dynamic field sets to reflect the hydrodynamic characteristics under different conditions, laying the foundation for subsequent suspended matter analysis. The suspended matter analysis based on the hydrological dynamic field set reveals the distribution of sewage sludge and its dynamic changes, and the subsequent deconstruction of transport characteristics enables the characteristics of sewage sludge data to be deeply analyzed, generating The resulting transport dynamics data provides a detailed basis for the decomposition of the water body stress tensor. The acquisition of stress distribution data further realizes the accurate simulation of water body material transport. The generated multi-scale sedimentation dynamics model provides a scientific explanation and prediction method for sedimentation phenomena. The real-time collected water body sedimentation change data is compared with the multi-scale sedimentation dynamics model, which can timely discover the inconsistency of sedimentation changes, thereby providing data support for the real-time fine-tuning of the model. The optimized multi-scale sedimentation dynamics model not only improves the adaptability and accuracy of the model, but also provides effective decision-making support for the management and governance of lakes and reservoirs, and overall promotes the scientific and systematic work of lake and reservoir water quality monitoring and environmental protection.

[0011] Preferably, step S1 comprises the following steps:

[0012] Step S11: Collecting lake and reservoir hydrological station data; reconstructing the lake and reservoir hydrological station data in time series to obtain continuous monitoring data;

[0013] Step S12: performing spatial interpolation calibration on the continuous monitoring data to generate gridded data; performing outlier detection and filtering on the gridded data to obtain cleaned hydrological data;

[0014] Step S13: Performing spatiotemporal constraint network mapping on the cleaned hydrological data to obtain a hydrological spatiotemporal matrix; performing local deviation analysis of adjacent sites based on the hydrological spatiotemporal matrix to obtain significantly deviated sites;

[0015] Step S14: Fill in the missing values ​​of the significantly deviated stations to generate a complete hydrological sequence.

[0016] The present invention provides continuous monitoring information through the collection and time series reconstruction of lake and reservoir hydrological station data. The grid data generated by spatial interpolation calibration enhances the spatial consistency and comprehensiveness of the data. Outlier detection and filtering ensure the accuracy and reliability of the data. The generation of cleaned hydrological data lays a solid foundation for subsequent analysis. The implementation of spatiotemporal constraint network mapping enables the formation of a hydrological spatiotemporal matrix that can better reflect the spatiotemporal characteristics of hydrological phenomena. The local deviation analysis of adjacent stations reveals significantly deviated stations, providing a basis for targeted data repair. The application of segmented missing value completion technology generates a complete hydrological sequence, providing high-quality data support for subsequent dynamic simulations, improving the accuracy and integrity of lake and reservoir hydrological data as a whole, providing a strong technical guarantee for the dynamic simulation of sewage sludge sedimentation processes, and promoting the scientific and systematic process of lake and reservoir management and governance.

[0017] Preferably, step S2 comprises the following steps:

[0018] Step S21: performing water flow identification on the complete hydrological sequence to obtain water level flow data; performing hydraulic feature extraction on the water level flow data to generate hydrodynamic data;

[0019] Step S22: performing boundary layer segmentation on the complete hydrological sequence based on the hydrodynamic data to obtain flow state stratification data;

[0020] Step S23: quantifying the turbulence intensity of the flow state stratification data to obtain turbulent field data; performing vortex structure decomposition on the turbulent field data to generate vortex field data;

[0021] Step S24: performing energy cascade decomposition on the vortex field data to obtain multi-scale field data; and reconstructing the hierarchical hydrodynamic field based on the multi-scale field data to generate a hydrological dynamic field set.

[0022] The present invention provides key water level flow data through water flow identification of complete hydrological sequences. The hydrodynamic data generated by hydraulic feature extraction provides a basis for flow state analysis. The implementation of boundary layer segmentation makes the flow state stratification data more accurate. The process of turbulence intensity quantification reveals the turbulent characteristics in the water body. The generated turbulent field data lays the foundation for subsequent structural analysis. The vortex structure splitting enables the formation of vortex field data to more finely reflect the flow characteristics. The application of energy cascade decomposition promotes the acquisition of multi-scale field data. The implementation of stratified hydrodynamic field reconstruction enables the hydrological dynamic field set to comprehensively present the hydrodynamic characteristics, which improves the understanding of the hydrological environment of lakes and reservoirs as a whole, provides rich data support and scientific basis for the dynamic simulation of sewage sludge sedimentation process, and promotes the optimization and implementation of relevant management and governance strategies.

[0023] Preferably, step S24 comprises the following steps:

[0024] Perform spectral decomposition on the vortex field data to obtain spectral distribution data;

[0025] Energy flux calibration is performed on vortex field data based on spectral distribution data to generate energy channel data;

[0026] The energy channel data is scale-screened according to a preset scale range to obtain multi-scale field data;

[0027] The multi-scale field data are vertically layered to obtain inter-layer structure data;

[0028] Construct laminar flow relationships based on interlayer structure data to generate laminar flow connection data;

[0029] Perform flow field splicing processing on the laminar flow connection data to obtain field intensity distribution data;

[0030] The field intensity distribution data are temporally and spatially aligned according to the multi-scale field data to generate a hydrological dynamic field ensemble.

[0031] The present invention provides detailed spectral distribution data through spectral decomposition processing of vortex field data. The energy channel data generated by energy flux calibration lays the foundation for subsequent energy analysis. The implementation of scale screening ensures the effectiveness and pertinence of multi-scale field data. The vertical stratification of water bodies makes the interlayer structure data more accurate. The process of laminar relationship construction reveals the flow characteristics inside the water body. The field intensity distribution data obtained by flow field splicing processing provides a clear image of the overall flow state. The application of time-space coordinate alignment ensures the accuracy and consistency of the hydrological dynamic field set, which enhances the dynamic understanding of the hydrological environment of lakes and reservoirs as a whole, provides a solid data foundation and scientific support for the accurate simulation of sewage sludge sedimentation process, and promotes the intelligent and systematic process of lake and reservoir management and water quality treatment.

[0032] Preferably, step S3 comprises the following steps:

[0033] Step S31: extracting suspended matter from the complete hydrological sequence to obtain suspended matter data; performing concentration identification on the suspended matter data to generate suspended matter concentration parameters;

[0034] Step S32: performing particle size spectrum analysis on the suspended matter concentration parameter to obtain particle characteristic data;

[0035] Step S33: performing flocculation dynamics analysis on the particle characteristic data based on the hydrological dynamic field set to obtain sewage sludge data; performing diffusion coefficient tensor decomposition on the sewage sludge data to obtain sludge diffusion characteristics;

[0036] Step S34: tracking the transport path of the sewage sludge data according to the sludge diffusion characteristics to generate migration path data; applying dynamic inversion to the sewage sludge data based on the migration path data to generate transport dynamic data.

[0037] The present invention provides accurate suspended matter data through suspended matter extraction of complete hydrological sequences. The suspended matter concentration parameters generated by concentration identification lay the foundation for subsequent analysis. The implementation of particle size spectrum analysis makes the acquisition of particle characteristic data more comprehensive. The flocculation dynamics analysis based on the hydrological dynamic field set reveals the characteristics of sewage sludge and provides important information for environmental governance. The process of diffusion coefficient tensor decomposition helps to deeply understand the diffusion characteristics of sludge. The generation of migration path data provides a clear trajectory for the transport process of sewage sludge. The application of dynamic inversion enables the generation of transport dynamics data to reflect the real sewage sludge dynamics, which improves the understanding of the sedimentation process of sewage sludge in lakes and reservoirs as a whole, provides a scientific basis and data support for water quality management and environmental protection, and promotes the intelligent and efficient process of lake and reservoir governance.

[0038] Preferably, step S34 includes the following steps:

[0039] Step S341: performing concentration gradient detection on sewage sludge data according to sludge diffusion characteristics to obtain gradient field data; performing flow direction vector calibration based on the gradient field data to generate sludge flow direction data;

[0040] Step S342: performing connectivity mapping on the sludge flow direction data to obtain connectivity path data; performing trajectory integration processing on the connectivity path data to generate migration path data;

[0041] Step S343: performing curvature analysis on the migration path data to obtain curvature distribution data; performing resistance simulation on the curvature distribution data to generate resistance field data;

[0042] Step S344: Apply dynamic inversion to the sewage sludge data based on the resistance field data to generate transport dynamic data.

[0043] The present invention generates gradient field data by detecting the concentration gradient of sewage sludge data through the sludge diffusion characteristics, which provides a basis for subsequent flow direction analysis. The implementation of flow direction vector calibration makes the sludge flow direction data clearer, which is helpful to understand the movement trend of sludge. Connectivity mapping reveals the flow path of sludge in water bodies. The migration path data generated by trajectory integral processing provides detailed trajectory information for dynamic simulation. The curvature analysis enables the formation of curvature distribution data, which provides an important reference for flow characteristics. The resistance field data generated by resistance simulation enables a deeper understanding of the resistance factors of sludge movement in water bodies. The application of dynamic inversion can accurately reflect the dynamic influence on sewage sludge during the flow process, which improves the accuracy and reliability of sewage sludge dynamic simulation as a whole, provides a scientific basis and technical support for lake management and water quality control, and promotes the efficient and intelligent process of environmental protection.

[0044] Preferably, step S344 includes the following steps:

[0045] Reconstruct the velocity field of the drag field data to generate velocity distribution;

[0046] The sludge settling evolution of sewage sludge data is analyzed based on the velocity distribution to generate sedimentation data;

[0047] Perform hydrodynamic calculations on sedimentation data based on velocity distribution to generate impact hydrodynamic parameters;

[0048] Separate the gravity term of the sedimentation data to obtain the gravity action data;

[0049] The buoyancy term is derived from the sedimentation data according to the gravity data to generate the buoyancy field data;

[0050] The impact hydrodynamic parameters, gravity data and buoyancy field data are integrated to generate transport force data.

[0051] The present invention generates velocity distribution through velocity field reconstruction of resistance field data, which provides key data for sludge settling evolution. The implementation of sludge settling evolution enables the generation of siltation and sedimentation data to accurately reflect the sedimentation process. The hydrodynamic calculation obtains impact hydrodynamic parameters based on the velocity distribution, which provides a quantitative basis for understanding hydrodynamic behavior. The separation of gravity terms ensures the accuracy of gravity data, which helps to deeply analyze the influence of gravity in the sedimentation process. The application of buoyancy term derivation generates buoyancy field data for siltation and sedimentation data. The process of force field integration combines impact hydrodynamic parameters, gravity data and buoyancy field data. The generated transport dynamic data comprehensively reflects the influence of various forces on sewage sludge in flow, which improves the accuracy and effectiveness of dynamic simulation of sewage sludge sedimentation process in lakes and reservoirs as a whole, and provides important technical support and scientific basis for water body management and environmental protection.

[0052] Preferably, step S4 comprises the following steps:

[0053] Step S41: performing shear stress decomposition on the transport force data to obtain stress component data;

[0054] Step S42: locating the principal stress direction of the stress component data to generate the principal stress direction; performing stress distribution analysis on the stress component data based on the principal stress direction to obtain stress distribution data;

[0055] Step S43: performing boundary layer pattern recognition on the stress distribution data to obtain interface characteristic data;

[0056] Step S44: Perform material transport simulation on the complete hydrological sequence based on the interface characteristic data to generate a multi-scale sedimentation dynamics model.

[0057] The present invention generates stress component data through shear stress decomposition of transport dynamic data, which provides a basis for subsequent stress analysis. The implementation of principal stress direction positioning makes the determination of the principal stress direction more accurate, which is helpful to understand the directionality of stress in flow. The stress distribution analysis based on the principal stress direction forms stress distribution data, which provides a quantitative basis for flow characteristics. The process of boundary layer pattern recognition reveals the characteristics of the water body interface, and the generation of interface characteristic data lays a foundation for material transport simulation. The material transport simulation of the complete hydrological sequence generates a multi-scale sedimentation dynamics model, which provides a systematic framework for the dynamic behavior of sewage sludge, and overall improves the understanding and prediction ability of the sedimentation process of sewage sludge in lakes and reservoirs, provides important technical support and scientific basis for environmental governance and water quality management, and promotes the intelligent and efficient management of lakes and reservoirs.

[0058] Preferably, step S44 includes the following steps:

[0059] Perform boundary condition calibration on interface characteristic data to obtain boundary condition parameters;

[0060] Carry out flow field mapping of the complete hydrological sequence according to boundary condition parameters to generate hydrological flow pattern distribution data;

[0061] Track particle movement based on hydrological flow distribution data to obtain sludge movement trajectory;

[0062] Reconstruct the sedimentation process based on the sludge movement trajectory to generate sedimentation process data;

[0063] Multi-scale model building is performed on the sedimentation process data to generate a multi-scale sedimentation dynamics model.

[0064] The present invention generates boundary condition parameters through boundary condition calibration of interface characteristic data, which provides an accurate basis for subsequent flow field mapping. The implementation of flow field mapping forms hydrological flow state distribution data, which can comprehensively reflect the flow state in the water body. The process of particle motion tracking ensures the accurate capture of sludge movement trajectory, which provides important information for analyzing sludge behavior. The siltation process data generated by the reconstruction of the siltation process provides a detailed foundation for the construction of multi-scale models. The multi-scale model construction based on the siltation process data forms a multi-scale siltation dynamics model, which provides a systematic framework for the dynamic simulation of lake sewage sludge, improves the overall understanding and prediction ability of sewage sludge siltation process, provides important technical support and scientific basis for water body management and environmental protection, and promotes the efficient and intelligent process of lake and reservoir management.

[0065] Preferably, step S5 comprises the following steps:

[0066] Step S51: Perform multi-point dynamic monitoring of lake and reservoir water bodies to obtain multi-point monitoring water body data; perform feature hierarchical analysis on the multi-point monitoring water body data to generate hierarchical feature water body data;

[0067] Step S52: reconstructing the regional dynamic characteristics of the layered characteristic water body data to generate real-time siltation change data of the water body; dividing the real-time siltation change data of the water body into sections to obtain siltation section data;

[0068] Step S53: performing volume integration processing on the sedimentation section data to generate sedimentation quantitative data; performing sedimentation condition prediction based on a multi-scale sedimentation dynamics model to obtain a predicted sedimentation amount;

[0069] Step S54: performing deviation statistics on the siltation quantification data and the predicted siltation amount to obtain error distribution data;

[0070] Step S55: When the error distribution data is not zero, the multi-scale sedimentation dynamics model is subjected to parameter correction processing according to the sedimentation quantification data to obtain model correction parameters; the multi-scale sedimentation dynamics model is reconstructed based on the model correction parameters to generate an optimized multi-scale sedimentation dynamics model.

[0071] The present invention generates multi-point monitoring water body data through multi-point dynamic monitoring of lake and reservoir water bodies, which provides rich information for subsequent analysis. Feature stratification analysis ensures the formation of stratified feature water body data, which can describe the distribution characteristics of water bodies in detail. The implementation of regional dynamic characteristic reconstruction generates real-time sedimentation change data of water bodies, which provides a real-time basis for dynamic monitoring. The generation of sedimentation section data provides a clear spatial division for sedimentation analysis. The sedimentation quantification data formed by volume integral processing provides a quantitative basis for the sedimentation status. The sedimentation status prediction based on the multi-scale sedimentation dynamics model can effectively evaluate the future sedimentation amount. The deviation statistics ensure the generation of error distribution data, which provides an important reference for model correction. When the error distribution data is not zero, the parameter correction processing can optimize the accuracy of the model. The implementation of model reconstruction generates an optimized multi-scale sedimentation dynamics model, which improves the monitoring and prediction capabilities of the sewage sludge sedimentation process of lakes and reservoirs as a whole, provides a scientific basis and technical support for water body management and environmental protection, and promotes the efficient and intelligent process of lake and reservoir management. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 A schematic diagram of the steps of a method for kinetic simulation of lake and reservoir sewage sludge sedimentation process;

[0073] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0074] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0075] 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

[0076] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0077] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0078] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0079] To achieve this, please refer to Figures 1 to 3 A method for kinetic simulation of lake sewage sludge sedimentation process comprises the following steps:

[0080] Step S1: Collecting lake and reservoir hydrological station data; filtering the lake and reservoir hydrological station data for outliers to obtain cleaned hydrological data; completing missing value processing on the cleaned hydrological data to generate a complete hydrological sequence;

[0081] Step S2: performing water flow pattern segmentation on the complete hydrological sequence to obtain flow pattern layered data; performing multi-layer hydrodynamic field reconstruction on the flow pattern layered data to generate a hydrological dynamic field set;

[0082] Step S3: Perform suspended solids analysis on the complete hydrological sequence based on the hydrological dynamic field set to obtain sewage sludge data; perform transport characteristic deconstruction on the sewage sludge data to generate transport dynamics data;

[0083] Step S4: decomposing the water body stress tensor based on the transport dynamics data to obtain stress distribution data; simulating the material transport of the complete hydrological sequence according to the stress distribution data to generate a multi-scale sedimentation dynamics model;

[0084] Step S5: Collect real-time sedimentation change data of water bodies; compare the sedimentation change data of water bodies and the multi-scale sedimentation dynamics model. When the sedimentation changes are inconsistent, fine-tune the multi-scale sedimentation dynamics model in real time according to the real-time sedimentation change data of water bodies to generate an optimized multi-scale sedimentation dynamics model.

[0085] The present invention ensures the accuracy and reliability of data by collecting data from lake and reservoir hydrological stations and filtering outliers. The cleaned hydrological data generates a complete hydrological sequence by filling in missing values, which provides a solid foundation for subsequent analysis. The implementation of flow pattern dissection enables the formation of flow pattern stratified data, thereby enhancing the understanding and description of water flow characteristics. The reconstruction of multi-layer hydrodynamic fields enables the generation of hydrological dynamic field sets to reflect the hydrodynamic characteristics under different conditions, laying the foundation for subsequent suspended matter analysis. The suspended matter analysis based on the hydrological dynamic field set reveals the distribution of sewage sludge and its dynamic changes, and the subsequent deconstruction of transport characteristics enables the characteristics of sewage sludge data to be deeply analyzed, generating The resulting transport dynamics data provides a detailed basis for the decomposition of the water body stress tensor. The acquisition of stress distribution data further realizes the accurate simulation of water body material transport. The generated multi-scale sedimentation dynamics model provides a scientific explanation and prediction method for sedimentation phenomena. The real-time collected water body sedimentation change data is compared with the multi-scale sedimentation dynamics model, which can timely discover the inconsistency of sedimentation changes, thereby providing data support for the real-time fine-tuning of the model. The optimized multi-scale sedimentation dynamics model not only improves the adaptability and accuracy of the model, but also provides effective decision-making support for the management and governance of lakes and reservoirs, and overall promotes the scientific and systematic work of lake and reservoir water quality monitoring and environmental protection.

[0086] In an embodiment of the present invention, the method for kinetic simulation of lake sewage sludge sedimentation process comprises the following steps:

[0087] Step S1: Collecting lake and reservoir hydrological station data; filtering the lake and reservoir hydrological station data for outliers to obtain cleaned hydrological data; completing missing value processing on the cleaned hydrological data to generate a complete hydrological sequence;

[0088] In this embodiment, when collecting data from lake and reservoir hydrological sites, real-time data including water level, flow velocity, flow direction, turbidity and suspended particle concentration are collected through multi-parameter hydrological sensors installed in the lake and reservoir area. The sensor model is YSI EXO2 multi-parameter detector (YSI EXO2 is a multi-parameter sensing device suitable for water quality monitoring), and the data collection frequency is set to once per minute. After collection, the data interface module is used to connect to the data storage server for storage. After the collection is completed, the outlier filtering operation is performed through the anomaly detection algorithm based on the statistical threshold, and the outliers exceeding 3 times the standard deviation are marked as invalid data. The moving window method is used to replace the mean of the data before and after the marked point. After the outlier filtering is completed, the cleaned hydrological data is passed to the missing value completion module, and the missing points are completed by the space-time interpolation method. In the space-time interpolation method, the Lagrange interpolation algorithm is used in the time direction, and the neighborhood weighted calculation method based on Kriging (Kriging, a spatial interpolation method) is used to complete the space direction. After completion, complete hydrological sequence data is obtained.

[0089] Step S2: performing water flow pattern segmentation on the complete hydrological sequence to obtain flow pattern layered data; performing multi-layer hydrodynamic field reconstruction on the flow pattern layered data to generate a hydrological dynamic field set;

[0090] In this embodiment, after the complete hydrological sequence is imported, the water area is divided into several sub-areas according to the distribution of sampling points through the regional blocking method. Each sub-area is divided into layers with water depth as the vertical axis. The number of layers is determined according to the depth range of the water body, and the minimum layer height is 1 meter. The flow characteristic analysis module is used to take water level changes, velocity gradients and vorticity distribution as input parameters. The flow characteristics of each water layer are extracted by applying the segmentation algorithm based on flow tensor decomposition to obtain flow stratification data. The flow stratification data is passed to the hydrodynamic field reconstruction module. Through the multi-layer coupling algorithm, the three-dimensional finite volume method (FVM) is used to solve the flow state of each water layer, calculate the distribution characteristics of the velocity field, vorticity field and pressure field, generate the dynamic field data of each layer, and finally superimpose the dynamic field results of each layer to obtain the hydrological dynamic field set.

[0091] Step S3: Perform suspended solids analysis on the complete hydrological sequence based on the hydrological dynamic field set to obtain sewage sludge data; perform transport characteristic deconstruction on the sewage sludge data to generate transport dynamics data;

[0092] In this embodiment, the hydrological dynamic field set is used as input data, the suspended matter concentration characteristics in the dynamic field are extracted, and water samples of different water depths are collected using a stratified sampling device. The suspended particle concentration is optically analyzed using a HACH DR3900 spectrophotometer (HACH DR3900 is a water quality analyzer) to obtain concentration profile data. In combination with the flow velocity and vorticity characteristics in the hydrological dynamic field, the suspended matter transport characteristic model is used to reconstruct the transport path of the suspended particles. By recording the time distribution of the suspended particle movement at discrete points, sewage sludge data is generated, and the sewage sludge data is transferred to the transport characteristic deconstruction module. The transport equation based on the Euler method is used to simulate and analyze the movement trajectory and deposition rate of the sewage sludge, and the energy loss, velocity distribution and deposition characteristics of the suspended particle transport are obtained, and finally the transport dynamic data is generated.

[0093] Step S4: decomposing the water body stress tensor based on the transport dynamics data to obtain stress distribution data; simulating the material transport of the complete hydrological sequence according to the stress distribution data to generate a multi-scale sedimentation dynamics model;

[0094] In this embodiment, the transport dynamics data is input into the water body stress analysis module, and the three-dimensional stress distribution calculation tool is used to decompose the tensor field inside the water body based on the stress tensor decomposition method, and the principal stress direction and magnitude of the tensor are calculated to form a complete stress field distribution. The stress field distribution results are used as input, and combined with the flow state changes in the complete hydrological sequence, a transport model based on multi-scale deconstruction is used to simulate material transport. The transport model adopts a multi-layer grid division strategy to calculate the material deposition process at different spatial scales layer by layer, generate particle size gradient distribution and deposition rate distribution results, and finally superimpose the calculation results to obtain a multi-scale sedimentation dynamics model.

[0095] Step S5: Collect real-time sedimentation change data of water bodies; compare the sedimentation change data of water bodies and the multi-scale sedimentation dynamics model. When the sedimentation changes are inconsistent, fine-tune the multi-scale sedimentation dynamics model in real time according to the real-time sedimentation change data of water bodies to generate an optimized multi-scale sedimentation dynamics model.

[0096] In this embodiment, an ultrasonic depth sounder is installed in the lake water area, and an RDI Teledyne ADCP (Acoustic Doppler Current Profiler) is used to collect sedimentation change data in real time, including sedimentation thickness and water depth changes. The collected data is transmitted to the sedimentation change analysis module, and a comparison method based on differential analysis is used to perform a point-to-point comparison between the real-time data and the predicted data in the multi-scale sedimentation dynamics model. When it is found that the real-time data and the model output results are inconsistent, the difference feature quantity is extracted, and the dynamic parameter adjustment algorithm is used to fine-tune the key parameters of the dynamic model (such as sedimentation rate, migration path weight, etc.), and the adjusted model is verified to match the real-time data. When the matching degree reaches a preset threshold, the adjusted model is saved as an optimized multi-scale sedimentation dynamics model.

[0097] Preferably, step S1 comprises the following steps:

[0098] Step S11: Collecting lake and reservoir hydrological station data; reconstructing the lake and reservoir hydrological station data in time series to obtain continuous monitoring data;

[0099] Step S12: performing spatial interpolation calibration on the continuous monitoring data to generate gridded data; performing outlier detection and filtering on the gridded data to obtain cleaned hydrological data;

[0100] Step S13: Performing spatiotemporal constraint network mapping on the cleaned hydrological data to obtain a hydrological spatiotemporal matrix; performing local deviation analysis of adjacent sites based on the hydrological spatiotemporal matrix to obtain significantly deviated sites;

[0101] Step S14: Fill in the missing values ​​of the significantly deviated stations to generate a complete hydrological sequence.

[0102] In this embodiment, multiple hydrological monitoring stations are arranged in lakes and reservoirs, and multi-parameter water quality sensors with high-frequency sampling (such as YSI EXO2 multi-parameter water quality meter) are used to collect multi-dimensional hydrological data such as water level, water flow rate, water temperature and dissolved oxygen. The collection frequency is set to once an hour. After the data is collected, it is uploaded to the central data server through a wireless data transmission module (such as LoRa or NB-IoT module). In the central server, the collected hydrological station data are reconstructed in time series through a time series analysis tool (such as the Pandas library in Python). First, the timestamps are parsed and the time axis is aligned at the hourly level. Then, the missing data segments are filled with linear interpolation. After filling, the reconstructed time series is smoothed, and the sliding window method is used to perform a three-point moving average operation on the data. Finally, continuous monitoring data is generated, and the continuous monitoring data is imported into the Geographic Information System (GIS) platform. The inverse distance weighted method (IDW) is used The station data are spatially interpolated by weighting to generate grid data, and the interpolation resolution is set to 100 meters. The built-in geographic boundary constraint function of the GIS platform is used to limit the interpolation range to the boundary of the lake. The standard deviation of the data distribution is calculated point by point for the grid data generated by interpolation, and 2 times the standard deviation is set as the outlier detection threshold. The grid points exceeding this threshold are marked as outliers. The spatial neighborhood of the outliers is then detected by the KNN (K-Nearest Neighbors) algorithm, and the average value of the normal points in the neighborhood is used to replace the outliers to obtain cleaned hydrological data. The cleaned hydrological data is constructed as a spatiotemporal data cube, in which the time dimension is hourly and the spatial dimension is the station or grid position. The cleaned hydrological data is imported into the spatiotemporal constraint network model constructed by a machine learning framework (such as TensorFlow or PyTorch). The cube data is spatiotemporally mapped through the model to generate a hydrological spatiotemporal matrix. The dynamic time warping algorithm (DTW) is then used based on the matrix. The local time series deviation between adjacent stations is calculated by using the automatic time warping (AIM) algorithm. The deviation value is calculated for each pair of stations and compared with the set significance threshold. The stations with deviation values ​​exceeding the threshold are marked as significantly deviated stations. The missing values ​​of the stations marked as significantly deviated are filled in by segmentation. First, the time series of the deviated stations are segmented and divided into multiple data segments based on the time point when the station data is abnormal. Each segment of data is predicted and filled using the auto-regressive integrated moving average (ARIMA) model. The model training uses the historical data of the past 30 days. The missing values ​​are filled based on the trained model. The completed data is verified for continuity and stability by the time series smoothing algorithm again, and finally a complete hydrological series is generated.The complete hydrological series is spatially consistent with the gridded data to ensure that the matching degree of time and space data meets the analysis requirements.

[0103] Preferably, step S2 comprises the following steps:

[0104] Step S21: performing water flow identification on the complete hydrological sequence to obtain water level flow data; performing hydraulic feature extraction on the water level flow data to generate hydrodynamic data;

[0105] Step S22: performing boundary layer segmentation on the complete hydrological sequence based on the hydrodynamic data to obtain flow state stratification data;

[0106] Step S23: quantifying the turbulence intensity of the flow state stratification data to obtain turbulent field data; performing vortex structure decomposition on the turbulent field data to generate vortex field data;

[0107] Step S24: performing energy cascade decomposition on the vortex field data to obtain multi-scale field data; and reconstructing the hierarchical hydrodynamic field based on the multi-scale field data to generate a hydrological dynamic field set.

[0108] In this embodiment, the complete hydrological sequence is extracted from the time series data of flow and water level by using flow identification algorithm and continuous flow velocity measurement technology, and the flow velocity and water level data are recorded by using electromagnetic current meter (Electromagnetic Current Meter) and differential pressure water level meter (Differential Pressure Water Level Meter), respectively. The collected data are imported into the flow calculation model, and the relationship between flow and water level is analyzed by control volume method to obtain water level flow data. The water level flow data is subjected to hydraulic feature extraction, and the dimension reduction method such as principal component analysis (PCA) is used to extract hydrodynamic characteristic variables, including flow velocity gradient, shear rate and pressure change, to generate hydrodynamic data. The boundary layer segmentation operation is performed based on the hydrodynamic data. First, direct numerical simulation (DNS) is used to extract the hydrodynamic characteristic variables, including flow velocity gradient, shear rate and pressure change. The hydrodynamic data is input into the boundary layer segmentation model by using the Particle Image Velocity Simulation (PIV) technology. The calculation conditions of the boundary layer thickness are set in the model, such as the position where the flow velocity reaches 99% of the free flow velocity. The velocity distribution characteristics of the water flow in different boundary layers are analyzed layer by layer. The flow state stratification data obtained by segmentation include laminar area, transition area and turbulent area. The velocity profile and turbulence intensity in each layer are calculated at the same time. The turbulence intensity of the flow state stratification data is quantified. The turbulent flow field characteristics in the water body are recorded by using a three-dimensional particle image velocimetry (PIV). The turbulence intensity is calculated by using data processing software through time series changes, including the ratio of the variance of the pulsating velocity to the average flow velocity. The turbulence intensity data is input into the vortex decomposition algorithm for vortex structure segmentation. The Q criterion and the λ2 criterion are used to mark the formation area of ​​the vortex structure. The vortex field data is further segmented to obtain the vortex field data. The vortex field data includes the intensity, direction and specific location distribution of the vortex core. The vortex field data is energy cascade decomposed and the Large Eddy Simulation (LES) is used. The energy transfer characteristics are extracted from the overall vorticity field by the Simulation method, and multi-scale field data are obtained through decomposition operations. The multi-scale field data include the different distributions of large vortices, medium vortices and small vortices. The stratified hydrodynamic field is reconstructed based on the multi-scale field data, and the multi-scale data are calculated in different regions using the hierarchical reconstruction algorithm. The dynamic characteristics of each scale are reconstructed into the stratified flow state to generate a hydrological dynamic field set. Finally, a complete data set containing multi-dimensional flow field characteristics and energy distribution characteristics is output. The reconstruction process sets the grid resolution to 50 meters to ensure that the spatial resolution of the hydrological dynamic field set matches the actual situation of the lake.

[0109] Preferably, step S24 comprises the following steps:

[0110] Perform spectral decomposition on the vortex field data to obtain spectral distribution data;

[0111] Energy flux calibration is performed on vortex field data based on spectral distribution data to generate energy channel data;

[0112] The energy channel data is scale-screened according to a preset scale range to obtain multi-scale field data;

[0113] The multi-scale field data are vertically layered to obtain inter-layer structure data;

[0114] Construct laminar flow relationships based on interlayer structure data to generate laminar flow connection data;

[0115] Perform flow field splicing processing on the laminar flow connection data to obtain field intensity distribution data;

[0116] The field intensity distribution data are temporally and spatially aligned according to the multi-scale field data to generate a hydrological dynamic field ensemble.

[0117] In this embodiment, the vortex field data is subjected to spectral decomposition processing, and the fast Fourier transform (FFT) algorithm is used to convert the vortex field data from the time domain to the frequency domain. The main energy distribution characteristics of the vortex field are extracted by spectrum analysis. The signal analyzer is used to collect and decompose the signal in the frequency range of 0.1 Hz to 10 Hz to obtain spectral distribution data. The spectral distribution data includes the specific corresponding relationship between frequency and energy. The data is stored in a multidimensional matrix format. The vortex field data is calibrated for energy flux based on the spectral distribution data. By introducing an energy flux model, the spectral distribution data is matched and calculated with the energy flux of the vortex field. A dynamometer is used to record the energy transfer rate in different frequency ranges of the vortex field. The calibration result is calibrated with the actual observed value to generate energy channel data. The energy channel data includes the energy transfer path and intensity at a specific frequency. The energy channel data is scaled according to a preset scale range. The energy channel data is graded using a data stratification algorithm, and the screening range is set to 0.The vortex scale is from 0.1m to 1m, and the invalid energy channels below the noise threshold are eliminated. The filtered data are reconstructed into a three-dimensional grid data format to obtain multi-scale field data. The multi-scale field data is used to characterize the energy distribution characteristics within different scale ranges. The multi-scale field data is vertically stratified by the water body. The ultrasonic Doppler current profiler (ADCP, Acoustic Doppler Current Profiler) is used to record the velocity gradient at different depths. The water body at different depths is divided by combining the stratification algorithm. The depth interval is set to every 2 meters to generate interlayer structure data. The interlayer structure data describes the velocity, vorticity and pressure distribution characteristics of each layer of the water body. The laminar relationship is constructed based on the interlayer structure data. The interlayer coupling analysis method is used to calculate the flow interaction relationship between different layers. The correlation matrix is ​​used to characterize the mutual influence intensity and direction between the layers. The matrix decomposition algorithm is used to extract the interlayer flow pattern and generate laminar connection data. The laminar connection data is saved in the form of a graph structure. Each node represents a single-layer fluid unit and the edge represents the interlayer flow. Relationship, flow field splicing processing is performed on laminar flow connection data, field splicing algorithm combined with spatial interpolation method is used to integrate the flow field data of each layer, and the splicing result is optimized by adjusting the node position and edge weight to ensure the continuity and consistency of the overall flow field. After the field intensity distribution data is generated, color images are used for visualization processing. Each pixel represents the intensity value of a flow field unit. The field intensity distribution data is subjected to spatiotemporal coordinate registration operation according to multi-scale field data. The field intensity distribution data is aligned in time and space coordinates through a three-dimensional spatiotemporal registration algorithm. The data is matched with the reference coordinates in the geographic information system (GIS) using grid coordinate transformation technology to generate a hydrological dynamic field set, which contains comprehensive flow field characteristics in time, space and energy dimensions. .

[0118] Preferably, step S3 comprises the following steps:

[0119] Step S31: extracting suspended matter from the complete hydrological sequence to obtain suspended matter data; performing concentration identification on the suspended matter data to generate suspended matter concentration parameters;

[0120] Step S32: performing particle size spectrum analysis on the suspended matter concentration parameter to obtain particle characteristic data;

[0121] Step S33: performing flocculation dynamics analysis on the particle characteristic data based on the hydrological dynamic field set to obtain sewage sludge data; performing diffusion coefficient tensor decomposition on the sewage sludge data to obtain sludge diffusion characteristics;

[0122] Step S34: tracking the transport path of the sewage sludge data according to the sludge diffusion characteristics to generate migration path data; applying dynamic inversion to the sewage sludge data based on the migration path data to generate transport dynamic data.

[0123] In this embodiment, when the suspended matter in the water body is extracted for the complete hydrological sequence, the suspended particles in the water body are separated and measured by using an optical turbidity sensor combined with a filter membrane sampling technology, the wavelength range of the optical sensor is set to 780nm to 1100nm, and a multi-point sampling device is used to perform synchronous sampling operations at different depths and areas of the water body. The sample is filtered and dried and then weighed to calculate the mass concentration of the suspended matter, and finally the suspended matter data is generated. When the suspended matter data is subjected to concentration identification, a particle concentration detector (such as a laser particle counter) is used to detect the sample, and the suspended matter mass concentration data is normalized with the water body volume to calculate the suspended matter concentration parameter. During the detection process, the sample flow rate needs to be controlled to be 0. .5L / min, the detection range is 1mg / L to 500mg / L. When the particle size spectrum is analyzed for the suspended matter concentration parameters, the particle size distribution of the suspended matter particles is measured by using a laser particle size analyzer. The suspended matter sample is diluted to a mass ratio of 5% and then injected into the detection cavity. The laser wavelength is set to 632.8nm. The dynamic light scattering technology is used to measure the particle characteristics with a particle size range of 0.01μm to 100μm. The particle characteristic data is analyzed and obtained. When the particle characteristic data is analyzed for flocculation dynamics based on the hydrodynamic field set, the flocculation model is introduced to simulate the coagulation behavior of the particles in the hydrodynamic field. The particle tracking algorithm is used to simulate the collision, adhesion and In the separation process, the flocculation efficiency is set to 70%, and the simulation results generate sewage sludge data. The sewage sludge data records the changes in particle mass in the form of time series. When the sewage sludge data is decomposed into a diffusion coefficient tensor, the sludge diffusion characteristics are analyzed by using a distributed computing model. The spatial frequency distribution characteristics are extracted based on Fourier transform. The main axis direction and diffusion coefficient of the diffusion tensor are calculated in combination with the sludge particle concentration gradient and the hydrodynamic field strength to generate sludge diffusion characteristic data. The sludge diffusion characteristic data contains diffusion speed, diffusion direction and corresponding time dimension information. When the sewage sludge data is transported and tracked according to the sludge diffusion characteristics, the Lagrangian tracking method is used in combination with the velocity field information of the hydrodynamic field. The migration path of sludge particles in the water body is calculated based on the information. The migration path data is generated by setting the initial coordinates and time step for numerical integration calculation. The migration path data is stored in the form of a path point sequence. Each point contains three-dimensional coordinates and time information. When dynamic inversion is applied to the sewage sludge data based on the migration path data, the inversion algorithm is used to restore and analyze the external force exerted on the sludge during migration. The spatial displacement and time relationship of the migration path data are adjusted and the step-by-step inversion is combined with the hydrodynamic field set to generate transport dynamics data. The transport dynamics data is stored in a matrix form, which contains multidimensional data characteristics of time, space and dynamic intensity. Finally, the transport dynamics information required for the hydrological dynamics model is obtained to complete the model construction.

[0124] Preferably, step S34 includes the following steps:

[0125] Step S341: performing concentration gradient detection on sewage sludge data according to sludge diffusion characteristics to obtain gradient field data; performing flow direction vector calibration based on the gradient field data to generate sludge flow direction data;

[0126] Step S342: performing connectivity mapping on the sludge flow direction data to obtain connectivity path data; performing trajectory integration processing on the connectivity path data to generate migration path data;

[0127] Step S343: performing curvature analysis on the migration path data to obtain curvature distribution data; performing resistance simulation on the curvature distribution data to generate resistance field data;

[0128] Step S344: Apply dynamic inversion to the sewage sludge data based on the resistance field data to generate transport dynamic data.

[0129] In this embodiment, when the concentration gradient of sewage sludge data is detected according to the sludge diffusion characteristics, the two-dimensional grid division technology is used to divide the entire study area into cells of 100 meters × 100 meters. The sludge particle concentration data is counted in each cell, and the concentration gradient between each grid is calculated based on the concentration distribution. The concentration difference and the spatial distance between the centers of adjacent grids are ratioed using the gradient calculation formula to generate gradient field data. The gradient field data is stored in the form of vectors, and each vector records the grid coordinates of the starting point, the grid coordinates of the end point, and the gradient size. When the flow direction vector is calibrated based on the gradient field data, the direction of all vectors in the gradient field data is normalized to a unit vector using a flow field analysis tool, and a statistical method is used to calibrate the flow direction vector. The vector directions of adjacent grids are spatially interpolated to generate a continuous flow direction vector field. The density of the flow direction vector field is set to 100 vectors per square kilometer. The sludge flow direction data is finally output. The sludge flow direction data is stored in a three-column record table, including the vector starting point coordinates, end point coordinates and vector direction angle. When the sludge flow direction data is connected, the Dijkstra shortest path algorithm is used to calculate the path connectivity between any two points. The flow direction vector data is used to determine the consistency of the flow direction on each path, and the paths with good connectivity are output in the form of connected path data. The connected path data is stored as a three-dimensional list. Each path records the point sequence, total length and direction consistency coefficient. The connected path data is subjected to trajectory integration. When performing curvature analysis on the migration path data, the path curvature is calculated based on the three-dimensional path coordinate points by the arc length method. The sampling interval is set to 5 meters using the data analysis software to generate curvature distribution data. The curvature distribution data is stored in the corresponding relationship between the path number and the curvature value. The path curvature value is used to reflect the curvature degree and shape change of the path. When performing resistance simulation on the curvature distribution data, the resistance model is established in combination with the fluid mechanics parameters. The path curvature value and flow velocity are used to calculate the hydrodynamic resistance on the path. The fluid viscosity is set to 0.001 kg / (m·s), and the resistance value is matched with the spatial coordinates of the path one by one to generate resistance field data. The resistance field data is stored in the form of a three-dimensional tensor, including the three-dimensional coordinates of the path points and the corresponding resistance values. When dynamic inversion is applied to the sewage sludge data based on the resistance field data, the inversion algorithm is used in combination with the sewage dynamic field model to infer the force conditions of the sludge particles. By dynamically adjusting the resistance and flow velocity parameters, the historical trajectory of the particle movement is gradually reconstructed to generate transport dynamic data. The transport dynamic data records the power intensity, direction and spatial position at each moment in a time series, and finally forms a complete description of the sludge dynamic characteristics.

[0130] Preferably, step S344 includes the following steps:

[0131] Reconstruct the velocity field of the drag field data to generate velocity distribution;

[0132] The sludge settling evolution of sewage sludge data is analyzed based on the velocity distribution to generate sedimentation data;

[0133] Perform hydrodynamic calculations on sedimentation data based on velocity distribution to generate impact hydrodynamic parameters;

[0134] Separate the gravity term of the sedimentation data to obtain the gravity action data;

[0135] The buoyancy term is derived from the sedimentation data according to the gravity data to generate the buoyancy field data;

[0136] The impact hydrodynamic parameters, gravity data and buoyancy field data are integrated to generate transport force data.

[0137] In this embodiment, when reconstructing the velocity field of the resistance field data, the fluid mechanics analysis tool is used to calculate the resistance field data point by point, and the velocity value of each point is obtained in a grid distribution manner. The grid size is set to 50 meters × 50 meters. The interpolation algorithm is used to calculate the velocity value of the uncovered area. The velocity value is expressed in vector form, including the velocity magnitude and direction. The velocity distribution is displayed through a two-dimensional vector field graph. When the sewage sludge data is subjected to sludge settling evolution based on the velocity distribution, a particle dynamics model is established, and the input particle density is 1.5g / cm 3 The particle sedimentation simulation software is used for numerical calculation, and the sedimentation distribution data is output every 30 minutes. The sedimentation distribution data is stored in the form of a three-dimensional matrix, in which each matrix element records the sedimentation thickness value at a specific location. When the hydrodynamic calculation of the siltation sedimentation data is performed according to the velocity distribution, the hydrodynamic analysis tool based on the Lagrangian method is used to input the sedimentation thickness and velocity field data to calculate the hydrodynamic impact force of the siltation area. The water viscosity is set to 0.001 kg / (m·s) and the time step is 10 seconds. The generated impact hydrodynamic parameters include the hydrodynamic intensity and direction of each grid point. The impact parameters are recorded in CSV format, and each line contains the coordinate point position and the corresponding hydrodynamic vector information. When the gravity term is separated from the siltation sedimentation data, the gravity effect in the sedimentation data is extracted by static decomposition technology. The gravity value of each point is calculated using a high-precision terrain model (DEM) combined with the siltation data, and the particle density is set to 1.5 g / cm 3 and the acceleration due to gravity is 9.8 m / s 2 The generated gravity data records the gravity size and distribution of each point. When deriving the buoyancy term of the sedimentation data based on the gravity data, the fluid density is 1.0g / cm based on the Archimedean principle.3 The buoyancy of each point is calculated based on the assumption that the buoyancy value is obtained through point-by-point calculation and recorded as buoyancy field data. The buoyancy field data is represented in the form of a vector field, including the buoyancy size and direction of each point. When the impact hydrodynamic parameters, gravity action data and buoyancy field data are integrated, the force field integration analysis tool is used to superimpose the three force field data point by point according to the position to obtain the comprehensive force field data. The consistency of the vector direction and size is ensured during the superposition calculation process. The output of each data point includes the comprehensive force size and direction vector.

[0138] Preferably, step S4 comprises the following steps:

[0139] Step S41: performing shear stress decomposition on the transport force data to obtain stress component data;

[0140] Step S42: locating the principal stress direction of the stress component data to generate the principal stress direction; performing stress distribution analysis on the stress component data based on the principal stress direction to obtain stress distribution data;

[0141] Step S43: performing boundary layer pattern recognition on the stress distribution data to obtain interface characteristic data;

[0142] Step S44: Perform material transport simulation on the complete hydrological sequence based on the interface characteristic data to generate a multi-scale sedimentation dynamics model.

[0143] In this embodiment, when the transport force data is subjected to shear stress decomposition, a multidimensional stress analysis tool is used to calculate the transport force data point by point, and the transport force vector of each point is decomposed into tangential and normal components. The tangential component is extracted as the main component of the shear stress. A three-dimensional grid division method is used in the analysis process, and the size of each grid unit is 50 meters × 50 meters × 5 meters. The input data includes the magnitude and direction of the transport force at each point. The output stress component data is saved in the form of a three-dimensional matrix containing shear stress values. Each element of the matrix records the shear stress magnitude of the corresponding grid point. When the principal stress direction of the stress component data is located, a tensor calculation is performed on the shear stress component based on the principal stress direction algorithm. The principal stress direction of each point is obtained by using a tensor decomposition method. The principal stress direction is represented in the form of a three-dimensional vector, including the spatial angle information of the direction. During the analysis process, the angle resolution is set to 1 degree and a layer-by-layer analysis method is used. The layered data is statistically averaged to ensure the overall direction consistency. The generated stress principal direction data is stored in CSV format. When stress component data is used for stress distribution analysis, stress distribution statistics tools are used to calculate the spatial distribution characteristics of stress point by point. When boundary layer pattern recognition is performed on stress distribution data, boundary features are extracted using a machine learning classification model based on a boundary layer recognition algorithm. The model input includes stress distribution data and training samples of known boundary layer patterns. The classification model is a multi-classifier based on support vector machine (SVM). After training, the global stress distribution data is classified. The output interface characteristic data includes boundary layer pattern categories and characteristic parameters (such as boundary layer thickness and stress gradient value). When material transport is simulated for a complete hydrological sequence based on interface characteristic data, a multi-scale hydrodynamic simulation tool is used to input interface characteristic data and complete hydrological sequence data. The hydrological sequence data includes lake and reservoir water level changes, flow velocity changes, and particle distribution information. During the simulation process, a three-dimensional finite difference method is used for numerical solution. The time step is set to 1 hour, and the spatial resolution is consistent with the interface characteristic data. The output simulation results include material transport paths, velocity fields, and sedimentation thickness distribution.

[0144] Preferably, step S44 includes the following steps:

[0145] Perform boundary condition calibration on interface characteristic data to obtain boundary condition parameters;

[0146] Carry out flow field mapping of the complete hydrological sequence according to boundary condition parameters to generate hydrological flow pattern distribution data;

[0147] Track particle movement based on hydrological flow distribution data to obtain sludge movement trajectory;

[0148] Reconstruct the sedimentation process based on the sludge movement trajectory to generate sedimentation process data;

[0149] Multi-scale model building is performed on the sedimentation process data to generate a multi-scale sedimentation dynamics model.

[0150] In this embodiment, when the interface characteristic data is calibrated for boundary conditions, a three-dimensional boundary condition calibration tool is used to parse and process characteristic parameters such as boundary layer thickness and stress gradient in the interface characteristic data. During the parsing, the boundary type is determined based on the boundary layer pattern classification result, including laminar boundary, turbulent boundary and other categories, and the calibration is performed in combination with actual hydrological measurement data such as velocity field and water level change data. During the calibration process, the gradient descent method is used to optimize the boundary parameters to ensure the global consistency of the calculation results. The boundary condition parameters finally obtained are stored in JSON format, including boundary type, boundary position and time variation characteristics. When the flow field mapping of the complete hydrological sequence is performed according to the boundary condition parameters, the flow field calculation module is used to map the hydrological sequence. The velocity, direction and time change information in the data are interpolated point by point. The interpolation method adopts a bilinear interpolation algorithm to ensure the accurate match of boundary conditions and flow field data. The flow field model used in the calculation is a three-dimensional unsteady-state model. The spatial resolution of the grid division is 25 meters × 25 meters × 1 meter, and the time resolution is 1 hour. The mapping result is output as hydrological flow distribution data, which is saved in a three-dimensional vector file format and contains the size and direction information of the velocity vector. When tracking particle motion based on hydrological flow distribution data, the Lagrangian particle motion model is used to simulate and calculate the particle trajectory. The input parameters in the model include particle size distribution (1 micron to 100 microns), density range (1.05 to 2.65 grams per cubic centimeter), initial position and hydrological flow distribution data. The tracking algorithm uses the fourth-order Runge-Kutta method to iteratively calculate the particle motion trajectory with a time step of 10 minutes. The sedimentation velocity, rising velocity and lateral diffusion characteristics of the particles in the fluid are considered in the calculation process. When reconstructing the sedimentation process of the sludge movement trajectory, the trajectory data is spatially reconstructed using three-dimensional grid reconstruction technology. The reconstruction process determines the sedimentation thickness distribution by calculating the residence time and deposition probability of the particles in the bottom area. The reconstruction grid used is consistent with the flow field grid, with a spatial resolution of 25 meters × 25 meters × 1 meter. CUDA acceleration is used for data processing to improve computing efficiency. The reconstruction result is output as sedimentation process data in the form of a three-dimensional grid. The grid data file contains the siltation thickness, sedimentary particle size distribution and time variation information of the grid unit. When constructing a multi-scale model for the siltation process data, based on the hierarchical multi-scale model method, the siltation data is first decomposed and processed according to the time scale and spatial scale. The time scale is divided into three categories: short-term (1 day), medium-term (1 month) and long-term (1 year). The spatial scale is divided into three categories: local (less than 1 square kilometer), regional (1 to 10 square kilometers) and global (greater than 10 square kilometers). The model construction process uses a multi-level linear regression algorithm to perform correlation analysis on data of different scales. The generated multi-scale siltation dynamics model includes the changes in siltation thickness, material transport paths and sedimentation rate distribution at each scale.

[0151] Preferably, step S5 comprises the following steps:

[0152] Step S51: Perform multi-point dynamic monitoring of lake and reservoir water bodies to obtain multi-point monitoring water body data; perform feature hierarchical analysis on the multi-point monitoring water body data to generate hierarchical feature water body data;

[0153] Step S52: reconstructing the regional dynamic characteristics of the layered characteristic water body data to generate real-time siltation change data of the water body; dividing the real-time siltation change data of the water body into sections to obtain siltation section data;

[0154] Step S53: performing volume integration processing on the sedimentation section data to generate sedimentation quantitative data; performing sedimentation condition prediction based on a multi-scale sedimentation dynamics model to obtain a predicted sedimentation amount;

[0155] Step S54: performing deviation statistics on the siltation quantification data and the predicted siltation amount to obtain error distribution data;

[0156] Step S55: When the error distribution data is not zero, the multi-scale sedimentation dynamics model is subjected to parameter correction processing according to the sedimentation quantification data to obtain model correction parameters; the multi-scale sedimentation dynamics model is reconstructed based on the model correction parameters to generate an optimized multi-scale sedimentation dynamics model.

[0157] In this embodiment, when the lake and reservoir water bodies are dynamically monitored at multiple points, an automated water quality monitoring buoy system is deployed at different locations of the lake and reservoir. The monitoring points are reasonably distributed according to the size of the lake and reservoir area and the terrain characteristics. Each monitoring point is equipped with a multi-parameter water quality sensor. The monitoring parameters include temperature, pH value, dissolved oxygen, suspended solids concentration, etc. The monitoring data is sent to the central data processing server through a wireless transmission module at intervals of 5 minutes. During processing, the original monitoring data is time-series corrected and missing values ​​are filled in to generate complete multi-point monitoring water body data. When the multi-point monitoring water body data is subjected to feature stratification analysis, a stratified analysis algorithm is used to vertically stratify the water body according to features such as water depth, temperature gradient and suspended solids concentration. The stratification process is based on Based on the hierarchical clustering method, the position of the stratified interface is determined by analyzing the data change rules of different depths of each monitoring point. The key characteristic parameters of each layer of water body data are extracted respectively, including the average temperature, average suspended solids concentration and layer thickness within the stratified range. When reconstructing the regional dynamic characteristics of the stratified characteristic water body data, the spatial interpolation method and time-weighted smoothing algorithm are used based on the time series and spatial distribution characteristics of the stratified characteristic data to reconstruct the dynamic changes of the water body stratification of the entire lake reservoir. During the reconstruction, the grid resolution is set to 50 meters × 50 meters, and the dynamic water quality characteristic data of each grid unit is output. Then, the changes in suspended solids concentration of each grid unit are analyzed according to the reconstruction results, the real-time sedimentation change trend is extracted, and the real-time sedimentation change data of the water body is generated. The siltation area is divided into sections based on terrain elevation data. When the siltation section data is processed for volume integration, a three-dimensional section model is used. The section boundary is determined by surface fitting of the section shape. Each section is integrated and calculated in combination with terrain data and real-time siltation thickness data. The calculation results include the siltation volume and particle size distribution of each section. The generated siltation quantification data is stored in JSON format. At the same time, a multi-scale siltation dynamics model is called to input real-time siltation change data of the water body. A prediction algorithm based on principal component analysis is used to predict the siltation status at different time scales in the future and generate a predicted siltation amount. When performing deviation statistics on the siltation quantification data and the predicted siltation amount, a double-sample t-test method is used to compare the quantitative data and the predicted data. Statistical deviation analysis was performed. During the analysis, the two sets of data were first tested for normality to ensure the statistical basis of the deviation analysis. The deviation values ​​at each time and in each area were then calculated and error distribution data was generated. When the error distribution data was not zero, the multi-scale sedimentation dynamics model was corrected according to the sedimentation quantification data. First, the main sources of deviation of the model parameters were identified by analyzing the error distribution data. Then, the Bayesian optimization method was used to adjust the key parameters of the model, such as the sedimentation rate coefficient and the particle settling velocity. After the parameter correction was completed, the model was rerun to verify the correction effect. Finally, the multi-scale sedimentation dynamics model was reconstructed based on the model correction parameters. The reconstruction process globally optimized the model parameters and spatial distribution characteristics.

[0158] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0159] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for kinetic simulation of lake sewage sludge sedimentation process, characterized in that: The following steps are involved: Step S1: Collecting lake and reservoir hydrological station data; Filter outliers from lake and reservoir hydrological station data to obtain cleaned hydrological data; Perform missing value completion processing on the cleaned hydrological data to generate a complete hydrological series; Step S2: performing water flow pattern segmentation on the complete hydrological sequence to obtain flow pattern layered data; performing multi-layer hydrodynamic field reconstruction on the flow pattern layered data to generate a hydrological dynamic field set; Step S3: Perform suspended solids analysis on the complete hydrological sequence based on the hydrological dynamic field set to obtain sewage sludge data; perform transport characteristic deconstruction on the sewage sludge data to generate transport dynamics data; Step S4: decomposing the water body stress tensor based on the transport dynamics data to obtain stress distribution data; Material transport simulations of the complete hydrological sequence were performed based on stress distribution data to generate a multiscale sedimentation dynamics model; Step S5: Collect real-time sedimentation change data of water bodies; compare the sedimentation change data of water bodies and the multi-scale sedimentation dynamics model. When the sedimentation changes are inconsistent, fine-tune the multi-scale sedimentation dynamics model in real time according to the real-time sedimentation change data of water bodies to generate an optimized multi-scale sedimentation dynamics model.

2. The method for kinetic simulation of lake sewage sludge sedimentation process according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Collecting lake and reservoir hydrological station data; reconstructing the lake and reservoir hydrological station data in time series to obtain continuous monitoring data; Step S12: performing spatial interpolation calibration on the continuous monitoring data to generate gridded data; performing outlier detection and filtering on the gridded data to obtain cleaned hydrological data; Step S13: Performing spatiotemporal constraint network mapping on the cleaned hydrological data to obtain a hydrological spatiotemporal matrix; performing local deviation analysis of adjacent sites based on the hydrological spatiotemporal matrix to obtain significantly deviated sites; Step S14: Fill in the missing values ​​of the significantly deviated stations to generate a complete hydrological sequence.

3. The method for kinetic simulation of lake sewage sludge sedimentation process according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: performing water flow identification on the complete hydrological sequence to obtain water level flow data; performing hydraulic feature extraction on the water level flow data to generate hydrodynamic data; Step S22: performing boundary layer segmentation on the complete hydrological sequence based on the hydrodynamic data to obtain flow state stratification data; Step S23: quantifying the turbulence intensity of the flow state stratification data to obtain turbulent field data; performing vortex structure decomposition on the turbulent field data to generate vortex field data; Step S24: performing energy cascade decomposition on the vortex field data to obtain multi-scale field data; and reconstructing the hierarchical hydrodynamic field based on the multi-scale field data to generate a hydrological dynamic field set.

4. The method for kinetic simulation of lake sewage sludge sedimentation process according to claim 3 is characterized in that: Step S24 includes the following steps: Perform spectral decomposition on the vortex field data to obtain spectral distribution data; Energy flux calibration is performed on vortex field data based on spectral distribution data to generate energy channel data; The energy channel data is scale-screened according to a preset scale range to obtain multi-scale field data; The multi-scale field data are vertically layered to obtain inter-layer structure data; Construct laminar flow relationships based on interlayer structure data to generate laminar flow connection data; Perform flow field splicing processing on the laminar flow connection data to obtain field intensity distribution data; The field intensity distribution data are temporally and spatially aligned according to the multi-scale field data to generate a hydrological dynamic field ensemble.

5. The method for kinetic simulation of lake sewage sludge sedimentation process according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: extracting suspended matter from the complete hydrological sequence to obtain suspended matter data; performing concentration identification on the suspended matter data to generate suspended matter concentration parameters; Step S32: performing particle size spectrum analysis on the suspended matter concentration parameter to obtain particle characteristic data; Step S33: performing flocculation dynamics analysis on the particle characteristic data based on the hydrological dynamic field set to obtain sewage sludge data; performing diffusion coefficient tensor decomposition on the sewage sludge data to obtain sludge diffusion characteristics; Step S34: tracking the transport path of the sewage sludge data according to the sludge diffusion characteristics to generate migration path data; applying dynamic inversion to the sewage sludge data based on the migration path data to generate transport dynamic data.

6. The method for kinetic simulation of lake sewage sludge sedimentation process according to claim 5 is characterized in that: Step S34 includes the following steps: Step S341: performing concentration gradient detection on sewage sludge data according to sludge diffusion characteristics to obtain gradient field data; performing flow direction vector calibration based on the gradient field data to generate sludge flow direction data; Step S342: performing connectivity mapping on the sludge flow direction data to obtain connectivity path data; performing trajectory integration processing on the connectivity path data to generate migration path data; Step S343: performing curvature analysis on the migration path data to obtain curvature distribution data; performing resistance simulation on the curvature distribution data to generate resistance field data; Step S344: Apply dynamic inversion to the sewage sludge data based on the resistance field data to generate transport dynamic data.

7. The method for kinetic simulation of lake sewage sludge sedimentation process according to claim 6 is characterized in that: Step S344 includes the following steps: Reconstruct the velocity field of the drag field data to generate velocity distribution; The sludge settling evolution of sewage sludge data is analyzed based on the velocity distribution to generate sedimentation data; Perform hydrodynamic calculations on sedimentation data based on velocity distribution to generate impact hydrodynamic parameters; Separate the gravity term of the sedimentation data to obtain the gravity action data; The buoyancy term is derived from the sedimentation data according to the gravity data to generate the buoyancy field data; The impact hydrodynamic parameters, gravity data and buoyancy field data are integrated to generate transport force data.

8. The method for kinetic simulation of lake sewage sludge sedimentation process according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: performing shear stress decomposition on the transport force data to obtain stress component data; Step S42: locating the principal stress direction of the stress component data to generate the principal stress direction; performing stress distribution analysis on the stress component data based on the principal stress direction to obtain stress distribution data; Step S43: performing boundary layer pattern recognition on the stress distribution data to obtain interface characteristic data; Step S44: Perform material transport simulation on the complete hydrological sequence based on the interface characteristic data to generate a multi-scale sedimentation dynamics model.

9. The method for kinetic simulation of lake and reservoir sewage sludge sedimentation process according to claim 8, characterized in that: Step S44 includes the following steps: Perform boundary condition calibration on interface characteristic data to obtain boundary condition parameters; Carry out flow field mapping of the complete hydrological sequence according to boundary condition parameters to generate hydrological flow pattern distribution data; Track particle movement based on hydrological flow distribution data to obtain sludge movement trajectory; Reconstruct the sedimentation process based on the sludge movement trajectory to generate sedimentation process data; Multi-scale model building is performed on the sedimentation process data to generate a multi-scale sedimentation dynamics model.

10. The method for kinetic simulation of lake sewage sludge sedimentation process according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: Perform multi-point dynamic monitoring of lake and reservoir water bodies to obtain multi-point monitoring water body data; perform feature hierarchical analysis on the multi-point monitoring water body data to generate hierarchical feature water body data; Step S52: reconstructing the regional dynamic characteristics of the layered characteristic water body data to generate real-time siltation change data of the water body; dividing the real-time siltation change data of the water body into sections to obtain siltation section data; Step S53: performing volume integration processing on the sedimentation section data to generate sedimentation quantitative data; performing sedimentation condition prediction based on a multi-scale sedimentation dynamics model to obtain a predicted sedimentation amount; Step S54: performing deviation statistics on the siltation quantification data and the predicted siltation amount to obtain error distribution data; Step S55: When the error distribution data is not zero, the multi-scale sedimentation dynamics model is subjected to parameter correction processing according to the sedimentation quantification data to obtain model correction parameters; the multi-scale sedimentation dynamics model is reconstructed based on the model correction parameters to generate an optimized multi-scale sedimentation dynamics model.

Images