A CDOM prediction method and related apparatus based on multi-source driving factors

Abstract

This application discloses a CDOM prediction method and related apparatus based on multiple driving factors, relating to the field of water environment management technology. The method includes: a trained CDOM inversion model obtained by training on measured CDOM concentration values ​​and remote sensing images corresponding to the measured CDOM concentration values ​​in time and space; long-sequence CDOM inversion data obtained from long-sequence remote sensing images; and a trained CDOM prediction model trained on the long-sequence CDOM inversion data and historical multiple driving factors. This CDOM prediction model can predict and output CDOM prediction data for the target water body's catchment area during the prediction period based on the multiple driving factors of the target water body's catchment area over historical periods. This application utilizes an LSTM model to learn the nonlinear coupling effect of multiple complex driving factors on CDOM, thus accurately predicting CDOM under the influence of multiple complex driving factors.

Description

Technical Field

[0001] This application relates to the field of water environment management technology, and in particular to a CDOM prediction method and related apparatus based on multi-source driving factors. Background Technology

[0002] With economic and social development, the scale of traditional industries, agriculture, animal husbandry, and urban activities has been continuously expanding, leading to a significant increase in the land-based input of colored dissolved organic matter (CDOM). CDOM mainly originates from the decomposition of terrestrial vegetation, the leaching of soil organic matter, and the discharge of industrial, agricultural, and domestic wastewater. It flows into rivers, lakes, and coastal areas through surface runoff and groundwater infiltration, significantly altering the optical properties and biogeochemical cycles of water bodies. Simultaneously, as an important carbon carrier, the spatiotemporal dynamics of CDOM have a profound impact on regional carbon budget assessments and global climate change research. Therefore, understanding the long-term evolution patterns and driving forces of CDOM, and achieving accurate prediction of CDOM spatiotemporal changes under different driving factors, provides crucial technical support for watershed pollution load early warning, dynamic assessment of carbon fluxes, and optimization of dual-carbon pathways.

[0003] Currently, monitoring research on colored dissolved organic matter (CDOM) in lakes mainly focuses on analyzing its spatiotemporal distribution characteristics and historical variation patterns through methods such as remote sensing inversion, in-situ sensor networks, or periodic sampling. For example, CDOM concentration is retrieved based on spectral characteristics, and terrestrial input fluxes are traced in conjunction with hydrological data. However, under the dual pressures of future climate change and human activities, forward-looking prediction methods for the dynamic evolution of lake CDOM remain insufficient. This limitation stems primarily from two aspects: First, the generation, migration, and transformation of CDOM involve complex biogeochemical processes (such as leaching of terrestrial organic matter, microbial degradation, and photochemical reactions), and its future behavior is affected by the nonlinear coupling of multiple factors such as changes in precipitation patterns, land use transformation, and carbon cycle disturbances. Existing mechanistic models have uncertainties in describing these complex physical processes. Second, the lack of long-term high-resolution monitoring data limits in-depth exploration of CDOM dynamic patterns, resulting in insufficient generalization ability of prediction models.

[0004] Therefore, it is particularly necessary to achieve accurate prediction of CDOM under the influence of multiple and complex driving factors, so as to provide key technical support for watershed pollution load early warning, dynamic assessment of carbon flux, and optimization of "dual carbon" pathways. Summary of the Invention

[0005] The purpose of this application is to provide a CDOM prediction method and related apparatus based on multi-source driving factors, which can accurately predict CDOM under the influence of multiple complex driving factors.

[0006] To achieve the above objectives, this application provides the following solution:

[0007] Firstly, this application provides a CDOM prediction method based on multi-source driving factors, comprising the following steps:

[0008] Obtain multi-source driving factors for the target water body's catchment area over historical periods; these multi-source driving factors include temperature data, rainfall data, wind speed data, runoff data, and pollution source data.

[0009] Multi-source driving factors are input into the trained CDOM prediction model, which outputs the CDOM prediction data of the target water body's catchment area during the prediction period. The trained CDOM prediction model is an LSTM model trained based on long-sequence CDOM inversion data and historical multi-source driving factors. The long-sequence CDOM inversion data is CDOM data obtained by inverting long-sequence remote sensing monitoring images through the trained CDOM inversion model. The trained CDOM inversion model is a linear regression model trained based on measured CDOM concentration values ​​and remote sensing monitoring images that correspond to the measured CDOM concentration values ​​in time and space.

[0010] Optionally, the method further includes the following steps:

[0011] Several detection points were selected in the catchment area of ​​the target water body, and the measured CDOM concentration value, corresponding detection time, and spatial coordinates of each detection point were obtained.

[0012] For any given detection point, based on the spatial coordinates of the detection point, the remote sensing water surface reflectance corresponding to the measured CDOM concentration value is obtained from remote sensing monitoring images near the detection time.

[0013] Based on the measured CDOM concentration values ​​and corresponding remote sensing water surface reflectance at each detection point, an inversion training sample set and an inversion verification sample set are constructed. Both the inversion training sample set and the inversion verification sample set include several inversion data pairs. Each inversion data pair includes the measured CDOM concentration value and corresponding remote sensing water surface reflectance at one detection point.

[0014] Based on the inversion training sample set, the remote sensing water surface reflectance is used as input and the corresponding CDOM concentration value is used as output to train several linear regression models. The linear regression models include stepwise linear regression model, backpropagation neural network and support vector machine model.

[0015] The linear regression models are evaluated based on the inversion validation sample set, and the linear regression model with the best inversion performance is selected as the trained CDOM inversion model.

[0016] Optionally, before obtaining the remote sensing water surface reflectance corresponding to the measured CDOM concentration value from remote sensing monitoring images near the detection time based on the spatial coordinates of the detection point, the following steps are also included:

[0017] Atmospheric correction was performed on all acquired remote sensing images to eliminate the effects of atmospheric absorption and scattering on spectral reflectance.

[0018] Optionally, the coefficient of determination and the relative root mean square error are selected to evaluate each linear regression model, and the linear regression model with the largest coefficient of determination and the smallest relative root mean square error among several linear regression models is selected as the trained CDOM inversion model.

[0019] Optionally, the method further includes the following steps:

[0020] Acquire long-series remote sensing images of the target water body's catchment area on various historical dates.

[0021] By using a trained CDOM inversion model, CDOM inversion data of the target water body's catchment area on various historical dates can be obtained from long-sequence remote sensing monitoring images.

[0022] For the CDOM inversion data of the target water body's catchment area on any historical date, historical multi-source driving factors for several dates prior to that date are obtained, and a sample of predicted data pairs is constructed.

[0023] Based on the historical prediction data of the target water body's catchment area for each date, a prediction training sample set and a prediction validation sample set are constructed.

[0024] Based on the predicted training sample set, historical multi-source driving factors are used as input, and CDOM inversion data is used as the target output to train the LSTM model, resulting in a trained LSTM model.

[0025] The trained LSTM model is evaluated based on the prediction validation sample set. If the prediction accuracy reaches the preset condition, the trained CDOM prediction model is obtained.

[0026] If the prediction accuracy does not meet the preset conditions, the hyperparameters of the LSTM model are adjusted using the Bayesian optimization algorithm, and the process jumps to the step "Based on the prediction training sample set, the historical multi-source driving factors are used as input, and the CDOM inversion data is used as the target output to train the LSTM model and obtain the trained LSTM model".

[0027] Optionally, the LSTM model structure includes an input layer, hidden layers, and an output layer; the hidden layer module consists of multiple neural units; each neural unit includes a forget gate, an input gate, and an output gate; the hyperparameters of the LSTM model include the number of hidden layers, the number of neurons, and the learning rate.

[0028] Secondly, this application provides a CDOM prediction system based on multi-source driving factors, including the following modules:

[0029] The multi-source driving factor acquisition module is used to acquire multi-source driving factors of the target water body's catchment area over historical periods; the multi-source driving factors include temperature data, rainfall data, wind speed data, runoff data, and pollution source data.

[0030] The CDOM prediction module is used to input multi-source driving factors into the trained CDOM prediction model and output the CDOM prediction data of the target water body's catchment area during the prediction period. The trained CDOM prediction model is an LSTM model trained based on long-sequence CDOM inversion data and historical multi-source driving factors. The long-sequence CDOM inversion data is CDOM data obtained by inverting long-sequence remote sensing monitoring images through the trained CDOM inversion model. The trained CDOM inversion model is a linear regression model trained based on measured CDOM concentration values ​​and remote sensing monitoring images that correspond to the measured CDOM concentration values ​​in time and space.

[0031] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the CDOM prediction method based on multi-source driving factors described above.

[0032] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the CDOM prediction method based on multi-source driving factors described above.

[0033] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the CDOM prediction method based on multi-source driving factors described above.

[0034] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0035] This application provides a CDOM prediction method and related apparatus based on multi-source driving factors. In this method, a trained CDOM inversion model is obtained by training based on measured CDOM concentration values ​​and remote sensing monitoring images that correspond to the measured CDOM concentration values ​​in time and space. A long-sequence CDOM inversion data is obtained by inverting CDOM data from long-sequence remote sensing monitoring images. Based on the long-sequence CDOM inversion data and historical multi-source driving factors, a trained CDOM prediction model is obtained. The CDOM prediction model can predict and output the CDOM prediction data of the target water body's catchment area in the prediction period according to the multi-source driving factors of the target water body's catchment area in historical time periods. This application utilizes historical temperature, rainfall, wind speed, runoff, and pollution source data, employing an LSTM model to learn the nonlinear coupling effects of multiple complex driving factors on the CDOM (Carbon Diffusion Mapping). This enables accurate prediction of the CDOM under the influence of these factors. Furthermore, since the CDOM prediction model is trained on long-sequence CDOM inversion data, it allows for in-depth analysis of the dynamic patterns of the CDOM, improving the generalization ability of the CDOM prediction model. This provides crucial technical support for watershed pollution load early warning, dynamic assessment of carbon flux, and optimization of "dual carbon" pathways. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart of a CDOM prediction method based on multi-source driving factors provided in an embodiment of this application.

[0038] Figure 2 This is a flowchart illustrating the construction and training of a CDOM inversion model in a CDOM prediction method based on multi-source driving factors, provided in an embodiment of this application.

[0039] Figure 3 This is a schematic diagram of the Chagan Lake area in a CDOM prediction method based on multi-source driving factors provided in an embodiment of this application.

[0040] Figure 4 This is a scatter plot of three linear regression models, SVR, SLR, and BP, trained in a CDOM prediction method based on multi-source driving factors provided in an embodiment of this application.

[0041] Figure 5This is a flowchart illustrating the construction and training of a CDOM prediction model in a CDOM prediction method based on multi-source driving factors, provided in an embodiment of this application.

[0042] Figure 6 This is a schematic diagram of the LSTM-based CDOM prediction model in a CDOM prediction method based on multiple driving factors provided in an embodiment of this application.

[0043] Figure 7 This is a schematic diagram of the functional modules of a CDOM prediction system based on multi-source driving factors, provided for another embodiment of this application.

[0044] Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0045] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] This application provides a CDOM prediction method based on multiple driving factors. In one exemplary embodiment, such as... Figure 1 As shown, it includes the following steps:

[0048] A1. Obtain multi-source driving factors for the target water body's catchment area over historical periods; these factors include temperature data, rainfall data, wind speed data, runoff data, and pollution source data. Pollution source data includes agricultural fertilization, livestock and poultry farming, and wastewater treatment plant discharge.

[0049] A2. Input the multi-source driving factors into the trained CDOM prediction model and output the CDOM prediction data of the target water body's catchment area during the prediction period. The trained CDOM prediction model is an LSTM model trained based on long-sequence CDOM inversion data and historical multi-source driving factors. The long-sequence CDOM inversion data is CDOM data obtained by inverting long-sequence remote sensing monitoring images through the trained CDOM inversion model. The trained CDOM inversion model is a linear regression model trained based on measured CDOM concentration values ​​and remote sensing monitoring images that correspond to the measured CDOM concentration values ​​in time and space.

[0050] In one exemplary embodiment, the method further includes a process of building and training a CDOM inversion model, such as... Figure 2 As shown, it includes the following steps:

[0051] B1. Select several monitoring points in the catchment area of ​​the target water body, and measure the measured CDOM concentration value, corresponding monitoring time, and spatial coordinates of each monitoring point. Specifically, such as... Figure 3 As shown, Chagan Lake was selected as the research object in this example. Detection points were set at intervals of 500-1000m, and each point was 1 kilometer away from the shore. Three samplings were conducted between May and November 2021 (covering spring, summer, and autumn), with 60 samples collected each time, resulting in a total of 180 sets of field-measured CDOMs, which met the requirements for establishing the CDOM inversion model.

[0052] B2. For any detection point, based on the spatial coordinates of the detection point, obtain the remote sensing water surface reflectance corresponding to the measured CDOM concentration value from remote sensing monitoring images near the detection time. Specifically, based on the detection time and spatial coordinate information, download OLCI and MODIS remote sensing data covering the corresponding time period (±3 hours) of Chagan Lake. Before extracting the reflectance data, to reduce external influences, this embodiment of the method further includes the following steps:

[0053] C1. Atmospheric correction was performed on all acquired remote sensing images to eliminate the influence of atmospheric absorption and scattering on spectral reflectance. In this example, SeaDAS software was used to perform atmospheric correction on both types of data to obtain the true remote sensing water surface reflectance after eliminating the effects of atmospheric absorption and scattering. Finally, 180 sets of matching data pairs were obtained between the measured CDOM concentration values ​​and the corresponding spatiotemporal OLCI and MODIS image remote sensing water surface reflectance.

[0054] B3. Based on the measured CDOM concentration values ​​and corresponding remote sensing water surface reflectance at each detection point, construct an inversion training sample set and an inversion validation sample set. Both the inversion training sample set and the inversion validation sample set include several inversion data pairs. Each inversion data pair includes the measured CDOM concentration value and corresponding remote sensing water surface reflectance at one detection point. 70% (126 sets) of the 180 "remote sensing reflectance-ground CDOM concentration" data sets are used as the inversion training sample set, and 30% (54 sets) are used as the inversion validation sample set.

[0055] B4. Based on the inverted training sample set, using remotely sensed water surface reflectance as input and the corresponding CDOM concentration value as output, several linear regression models were trained. These linear regression models included stepwise linear regression, backpropagation neural network, and support vector machine models. Specifically, the obtained blue, green, red, and near-infrared band remotely sensed reflectance were used as input parameters, and the measured CDOM concentration value was used as the target parameter. CDOM was fitted using stepwise linear regression (SLR), backpropagation neural network (BP), and support vector machine (SVR) models, respectively. Figure 4 As shown, three CDOM inversion models were obtained.

[0056] B5. Evaluate each linear regression model based on the inversion validation sample set, and select the linear regression model with the best inversion performance as the trained CDOM inversion model. This is done by comparing the model evaluation parameters (R²). 2 And rRMSE), found that the SVR model showed the best performance in both training and validation evaluation metrics (training set: R). 2 =0.88, rRMSE=23%, Validation set: R 2 =0.86, rRMSE=24%. Therefore, SVR was selected as the final CDOM inversion model.

[0057] Specifically, in step B5, the coefficient of determination and the relative root mean square error are selected to evaluate each linear regression model. The linear regression model with the largest coefficient of determination and the smallest relative root mean square error among several linear regression models is selected as the trained CDOM inversion model.

[0058] In one exemplary embodiment, the method further includes a process of building and training a CDOM prediction model, such as... Figure 5 As shown, it includes the following steps:

[0059] D1. Acquire long-series remote sensing images of the target water body's catchment area for various historical dates. Download long-series remote sensing images covering Chagan Lake from 2014 to 2024. Specifically, acquire long-series OLCI and MODIS images for this period, one image every 2-3 days.

[0060] D2. Using the trained CDOM inversion model, obtain the CDOM inversion data of the target water body's catchment area for each historical date based on long-sequence remote sensing images. The trained CDOM inversion model obtained through steps B1 to B5 can directly invert the CDOM concentration data of Chagan Lake for the past ten years based on long-sequence remote sensing images. Since multiple pixels of the remote sensing images cover Chagan Lake, the average CDOM value of all pixels covering the water body of Chagan Lake is taken as the lake's CDOM concentration value for that day.

[0061] D3. For the CDOM inversion data of the target water body's catchment area on any historical date, obtain the historical multi-source driving factors for several dates prior to that date, and construct a sample of predicted data pairs. Specifically, collect and organize natural climate data such as temperature, rainfall, wind speed, and runoff in the Chagan Lake basin from 2014 to 2024, as well as human activity data from surrounding counties and cities, including agricultural fertilizer application, livestock farming, and sewage treatment plant discharge data.

[0062] In an improved embodiment, to reduce the complexity of the LSTM model, strong correlations are selected from the above multi-source driving factors through correlation analysis. The historical multi-source driving factors obtained in step D3 are screened using mutual information to determine the optimal historical multi-source driving factors as the model's input parameters. Specifically, this involves calculating the mutual information value I(X1, X2, etc.) between the CDOM concentration value Y and each driving factor (such as temperature X1, rainfall X2, etc.) using the following formula. i ;Y):

[0063]

[0064] Where Y is the CDOM concentration value, X i Let I(X,y) be the i-th driving factor, p(x,y) be the joint probability density function of X and Y, p(x) be the probability density function of X, and p(y) be the probability density function of Y. According to I(X... i If Y)>0.3, filter out the driving parameters that are closely related to the dynamic changes of CDOM and remove irrelevant or redundant parameters.

[0065] D4. Based on the historical prediction data of the target water body's catchment area on various dates, construct a prediction training sample set and a prediction validation sample set.

[0066] D5. Based on the predicted training sample set, historical multi-source driving factors are used as input, and CDOM inversion data is used as the target output to train the LSTM model, resulting in a trained LSTM model. Specifically, the LSTM model structure includes an input layer, hidden layers, and an output layer; the hidden layer module consists of multiple neural units; each neural unit includes a forget gate, an input gate, and an output gate; the hyperparameters of the LSTM model include the number of hidden layers, the number of neurons, and the learning rate. The hyperparameter settings are as follows: number of hidden layers: 1–3 layers, number of neurons: 32–256, learning rate: 0.0001–0.01.

[0067] D6. Evaluate the trained LSTM model based on the prediction validation sample set. If the prediction accuracy does not meet the preset conditions, proceed to step D7. If the prediction accuracy meets the preset conditions, the trained CDOM prediction model is obtained.

[0068] D7. Use the Bayesian optimization algorithm to adjust the hyperparameters of the LSTM model, and then jump to step D5.

[0069] Specifically, with R 2 As the primary metric for measuring model accuracy, rRMSE is used as a secondary metric. If R... 2 If the hyperparameters are less than 0.85 and the rRMSE is greater than 25%, the Bayesian optimization algorithm is used to adjust the hyperparameters of the LSTM model, and the process jumps to step D5 to continue optimizing and adjusting the model until the final CDOM prediction model is obtained.

[0070] Specifically, in this embodiment, such as Figure 6 As shown, inputting the multi-source driving factors (rainfall, temperature, wind speed, sewage treatment plant discharge, and agricultural fertilizer application) from the past three days into a pre-trained LSTM-based CDOM prediction model can output the predicted CDOM concentration of Chagan Lake for the next day. A month may contain multiple remote sensing datasets; CDOM concentration prediction is performed for each dataset, and the average of all predictions for the month is taken as the final CDOM concentration prediction for that month. It can be observed that the LSTM-based CDOM prediction model trained in this application achieves good simulation accuracy, with a training period R... 2 The accuracy is 0.88 and the rRMSE is 12.6%, which meets the accuracy requirements and can fully capture the dynamic changes of the CDOM in Chagan Lake.

[0071] Based on the same inventive concept, this application also provides a system for implementing the CDOM prediction method based on multi-source driving factors described above. The solution provided by this system is similar to the implementation described in the above method. In an exemplary embodiment, such as... Figure 7 As shown, a CDOM prediction system based on multi-source driving factors is provided, including the following modules:

[0072] The multi-source driving factor acquisition module is used to acquire multi-source driving factors of the target water body's catchment area over historical periods; the multi-source driving factors include temperature data, rainfall data, wind speed data, runoff data, and pollution source data.

[0073] The CDOM prediction module is used to input multi-source driving factors into the trained CDOM prediction model and output the CDOM prediction data of the target water body's catchment area during the prediction period. The trained CDOM prediction model is an LSTM model trained based on long-sequence CDOM inversion data and historical multi-source driving factors. The long-sequence CDOM inversion data is CDOM data obtained by inverting long-sequence remote sensing monitoring images through the trained CDOM inversion model. The trained CDOM inversion model is a linear regression model trained based on measured CDOM concentration values ​​and remote sensing monitoring images that correspond to the measured CDOM concentration values ​​in time and space.

[0074] certainly, Figure 7 The architecture shown is merely exemplary; it can be omitted as needed when implementing different functionalities. Figure 7 One or at least two components of the system shown.

[0075] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 8 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it can implement the CDOM prediction method based on multi-source driving factors provided in the aforementioned embodiment.

[0076] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0077] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0078] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0079] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0080] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0081] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0082] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0083] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0084] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for predicting CDOM based on multi-source driving factors, characterized in that, The method comprises the following steps: obtaining multi-source driving factors of a target water body catchment area in a historical period; the multi-source driving factors include air temperature data, rainfall data, wind speed data, runoff data and pollution source data; inputting the multi-source driving factors of the target water body catchment area in the historical period into a trained CDOM prediction model to output CDOM prediction data of the target water body catchment area in a prediction period; the trained CDOM prediction model is an LSTM model trained based on long-sequence CDOM inversion data and historical multi-source driving factors; the long-sequence CDOM inversion data are CDOM inversion data of the target water body catchment area on each date in the history obtained by a trained CDOM inversion model according to long-sequence remote sensing monitoring images; for the CDOM inversion data of the target water body catchment area on any date in the history, the corresponding historical multi-source driving factors are the multi-source driving factors of several dates before the date; the trained CDOM inversion model is a linear regression model trained based on measured CDOM concentration values and remote sensing monitoring images corresponding in space and time to the measured CDOM concentration values.

2. The multi-source driving factor based CDOM prediction method of claim 1, wherein, The method further comprises the following steps: selecting several detection points in the target water body catchment area to measure the measured CDOM concentration values, corresponding detection times and point space coordinates of the detection points; for any detection point, acquiring remote sensing water surface reflectance corresponding to the measured CDOM concentration value in a remote sensing monitoring image at a close detection time according to the point space coordinates of the detection point; constructing an inversion training sample set and an inversion verification sample set according to the measured CDOM concentration values and corresponding remote sensing water surface reflectance of each detection point; the inversion training sample set and the inversion verification sample set each comprise several inversion data pair samples; the inversion data pair sample comprises the measured CDOM concentration value and the corresponding remote sensing water surface reflectance of a detection point; based on the inversion training sample set, the remote sensing water surface reflectance is taken as input and the corresponding CDOM concentration value is taken as output to train several linear regression models respectively; the linear regression model includes a stepwise linear regression model, a back propagation neural network and a support vector machine model; based on the inversion verification sample set, each linear regression model is evaluated, and the linear regression model with the optimal inversion effect is taken as the trained CDOM inversion model.

3. The multi-source driving factor based CDOM prediction method of claim 2, wherein, Before acquiring the remote sensing water surface reflectance corresponding to the measured CDOM concentration value in the remote sensing monitoring image at the close detection time according to the point space coordinates of the detection point, the method further comprises the following steps: atmospheric correction is performed on all acquired remote sensing monitoring images respectively to eliminate the influence of atmospheric absorption and scattering on spectral reflectance.

4. The multi-source driving factor based CDOM prediction method of claim 2, wherein, determination coefficients and relative root mean square errors are selected to evaluate each linear regression model, and the linear regression model with the maximum determination coefficient and the minimum relative root mean square error in the several linear regression models is taken as the trained CDOM inversion model.

5. The multi-source driving factor based CDOM prediction method of claim 1, wherein, The method further comprises the following steps: obtaining long-sequence remote sensing monitoring images of the target water body catchment area on each date in the history; CDOM inversion data of the target water body catchment area on each date in history is obtained through the trained CDOM inversion model according to long-sequence remote sensing monitoring images of the target water body catchment area on each date in history; For the CDOM inversion data of the target water body catchment area on any date in history, historical multi-source driving factors of several dates before the date are obtained, and a prediction data pair sample is constructed; According to the prediction data pair sample of the target water body catchment area on each date in history, a prediction training sample set and a prediction verification sample set are constructed; Based on the prediction training sample set, the historical multi-source driving factors are taken as input, and the CDOM inversion data is taken as target output, and the LSTM model is trained to obtain a trained LSTM model; Based on the prediction verification sample set, the trained LSTM model is evaluated, and if the prediction accuracy reaches a preset condition, a trained CDOM prediction model is obtained. If the prediction accuracy does not reach the preset condition, the hyperparameters of the LSTM model are adjusted using a Bayesian optimization algorithm, and the step of "based on the prediction training sample set, taking the historical multi-source driving factors as input and the CDOM inversion data as target output, training the LSTM model to obtain a trained LSTM model" is jumped to.

6. The multi-source driving factor based CDOM prediction method of claim 5, wherein, The LSTM model structure includes an input layer, a hidden layer and an output layer; the hidden layer module is composed of a plurality of neural units; each neural unit includes a forgetting gate, an input gate and an output gate; the hyperparameters of the LSTM model include the number of hidden layers, the number of neurons and the learning rate.

7. A multi-source driving factor based CDOM prediction system, comprising: It comprises: A multi-source driving factor acquisition module for acquiring multi-source driving factors of a target water body catchment area in a historical period; The multi-source driving factors include temperature data, rainfall data, wind speed data, runoff data and pollution source data; A CDOM prediction module for inputting the multi-source driving factors of the target water body catchment area in the historical period into a trained CDOM prediction model to output CDOM prediction data of the target water body catchment area in a prediction period; the trained CDOM prediction model is a LSTM model trained based on long-sequence CDOM inversion data and historical multi-source driving factors; the long-sequence CDOM inversion data is CDOM inversion data of the target water body catchment area on each date in history obtained by a trained CDOM inversion model according to long-sequence remote sensing monitoring images; for the CDOM inversion data of the target water body catchment area on any date in history, the corresponding historical multi-source driving factors are multi-source driving factors of several dates before the date; The trained CDOM inversion model is a linear regression model trained based on measured CDOM concentration values and remote sensing monitoring images corresponding in space and time to the measured CDOM concentration values.

8. A computer device comprising: Memory, processor and computer program stored on the memory and executable on the processor, characterized in that the processor executes the computer program to implement the multi-source driving factor based CDOM prediction method of any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program, which is executed by a processor, implements the method for predicting CDOM based on multi-source driving factors according to any one of claims 1-6.

10. A computer program product comprising a computer program, characterized in that, The computer program, which is executed by a processor, implements the method for predicting CDOM based on multi-source driving factors according to any one of claims 1-6.

Images