Water conservancy scientific research project matching method and system based on big data
By conducting hyperspectral remote sensing data analysis on the water conservancy area, the scientific research theme portrait is determined, and based on the basic information of the scientific research team and the academic achievement information, the scientific research team is accurately matched, which solves the problem of inaccurate matching of scientific research projects and demands in the existing technology, and achieves efficient scientific research project procurement.
Patent Information
- Application Number
- CN202510361408.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-26
- Publication Date
- 2025-06-27
AI Technical Summary
In the existing technology, there is a certain degree of subjectivity that scientific research projects to be purchased cannot be accurately matched with scientific research needs.
By conducting hyperspectral remote sensing data analysis on the water conservancy area, potential risk factors and risk scores are determined, and the scientific research theme portrait is determined based on the basic information of the scientific research team and academic achievement information, and finally match the target scientific research team based on similarity.
It has achieved the analysis of potential risks in the water conservancy area from multiple dimensions and accurately matched the scientific research team, which has improved the accuracy of scientific research project procurement.
Smart Images
Figure CN120218864A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of big data procurement, and in particular relates to a matching method and system for water conservancy scientific research projects based on big data. Background Art
[0002] At present, there are serious pollution problems in water conservancy areas such as the ocean and wetlands. In order to explore the causes of pollution and formulate corresponding improvement measures, targeted scientific research activities are required.
[0003] In the prior art, scientific research projects are usually formulated and procured based on human experience. The determination of the above-mentioned scientific research projects and the selection of procurement objects are somewhat subjective, and it is impossible to accurately match the scientific research projects to be procured with scientific research needs. Summary of the Invention
[0004] The purpose of the present invention is to provide a matching method and system for water conservancy scientific research projects based on big data, so as to solve the technical problem that in the prior art, it is impossible to accurately match the scientific research projects to be procured with scientific research needs.
[0005] A matching method for water conservancy scientific research projects based on big data, the method includes: S1: Analyze the potential risk factors of the first water conservancy area to obtain a plurality of first potential risk factors and corresponding first risk scores; The first potential risk factors include two categories: absolute environmental assessment data and relative environmental assessment data; S2: Determine a first scientific research theme portrait according to the plurality of first risk scores and the plurality of first potential risk factors; The S2 includes the following sub-steps: S21: Determine the target potential risk factors from the first potential risk factors whose first risk scores meet the preset requirements; S22: Determine a first scientific research theme portrait according to the plurality of target potential risk factors; S3: Determine a plurality of first scientific research team portraits according to the first basic information and first research result information of a plurality of first scientific research teams; S4: Determine at least one target scientific research team for water conservancy scientific research project procurement according to the similarity between the first scientific research theme portrait and the plurality of first scientific research team portraits.
[0006] Preferably, the S1 includes the following sub-steps: S11: Obtain multiple groups of first hyperspectral remote sensing data of the first water conservancy area at a first time interval; S12: Analyze the most recent set of the first hyperspectral remote sensing data to obtain the first absolute environmental assessment data; wherein, the first absolute environmental assessment data includes water quality parameters, thermal pollution sources, and oil pollution source data; S13: Analyze multiple sets of the first hyperspectral remote sensing data to obtain the first relative environmental assessment data; wherein, the first relative environmental assessment data includes land coverage rate, heavy metal pollution rate, and biodiversity change rate data; S14: Obtain multiple first potential risk factors and corresponding first risk scores based on the first absolute environmental assessment data and the first relative environmental assessment data.
[0007] Preferably, S12 includes the following sub-steps: S121: Preprocess the most recent set of the first hyperspectral remote sensing data to obtain the first radiance map; S122: Calculate the first reflectance map based on the first radiance map; S123: Input the first reflectance map into the first absolute environmental assessment model to output the first absolute environmental assessment data.
[0008] Preferably, the reflectance ρ TOA in S122 is calculated by the following formula: ρ TOA =πL f d 2 / ESUN f cosθ; where L f is the first radiance value, d is the distance from the Earth to the Sun, ESUN f is the solar irradiance, and cosθ is the solar zenith angle.
[0009] Preferably, S13 includes the following sub-steps: S131: Preprocess multiple sets of the first hyperspectral remote sensing data respectively to obtain multiple sets of second hyperspectral remote sensing data; S132: Input multiple sets of the second hyperspectral remote sensing data into the relative environmental assessment model to obtain the first relative environmental assessment data.
[0010] Preferably, S3 includes the following sub-steps: S31: Obtain the first basic information of each first scientific research team; S32: Obtain the first research achievement information of each first scientific research team.
[0011] Preferably, S32 includes the following sub-steps: S321: Obtain multiple first scientific research project vectors according to the scientific research projects completed by each of the first scientific research teams; S322: Obtain multiple first academic achievement vectors according to each academic achievement information; S323: Perform weighted clustering on the multiple first scientific research project vectors and the multiple first academic achievement vectors to obtain the first scientific research team portrait.
[0012] This application also proposes a water conservancy scientific research project matching system based on big data, which is used to implement the water conservancy scientific research project matching method based on big data described in any one of the above.
[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: The water conservancy scientific research project matching method and system based on big data proposed in this application first collect hyperspectral remote sensing images of the first water conservancy area to be scientifically studied, and analyze the hyperspectral remote sensing images within a certain time interval to determine the potential risk factors and risk scores of the first water conservancy area from two dimensions of absolute data analysis and relative data analysis. Secondly, determine the target risk factors according to the potential risk factors and risk scores, and then determine the first scientific research theme portrait corresponding to the first water conservancy area. Then, use the basic information and academic achievement information of the scientific research teams to be pushed to determine multiple first scientific research team portraits. Finally, based on the similarity between the first scientific research theme portrait and the multiple first scientific research team portraits, determine the target scientific research team and conduct scientific research project procurement. Through the technical solution of this application, the potential risks of the water conservancy area can be analyzed from multiple dimensions, and a scientific research team with a matching degree meeting the preset requirements can be found for scientific research project procurement, which can effectively improve the accuracy of scientific research project procurement. Description of the Drawings
[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings described below are only exemplary, and for those of ordinary skill in the art, without creative efforts, other implementation drawings can be obtained according to the provided drawings.
[0015] Figure 1 It is the execution flowchart of a water conservancy scientific research project matching method based on big data of the present invention. Detailed Embodiments
[0016] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
[0017] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments, in which the schematic embodiments and descriptions are only used to explain the present invention, but not to limit the present invention.
[0018] A method and system for matching water conservancy scientific research projects based on big data of the present invention will be described in detail below.
[0019] This embodiment proposes a method for matching water conservancy scientific research projects based on big data, and the execution process is as Figure 1 shown, specifically including the following steps: S1: Analyze the potential risk factors of the first water conservancy area to obtain a plurality of first potential risk factors and corresponding first risk scores.
[0020] Common water conservancy areas may include areas such as the ocean and wetlands. Each type of common water conservancy area has corresponding functions. For example, the main functions of wetlands include water resource regulation, water quality purification, biodiversity maintenance, carbon sequestration, providing food and livelihoods, etc. The first water conservancy area may include the above common areas. In order to achieve precise procurement of water conservancy scientific research projects, it is necessary to first analyze the potential risk factors of the first water conservancy area in this step.
[0021] The S1 specifically includes the following sub-steps: S11: Obtain multiple groups of first hyperspectral remote sensing data of the first water conservancy area at a first time interval.
[0022] Assume that the first water conservancy area is a wetland area. By analyzing the hyperspectral remote sensing data of the first water conservancy area, the change trends in multiple aspects such as land cover, water quality parameters, and biodiversity of the first water conservancy area can be analyzed.
[0023] S12: Analyze the first absolute environmental assessment data through the most recent group of the first hyperspectral remote sensing data.
[0024] The first absolute pollution data refers to the environmental assessment data obtained by analyzing the most recent group of the first hyperspectral data of the first water conservancy area. The absolute assessment indicators for the first water conservancy areas such as wetlands may include water quality parameters, thermal pollution sources, oil pollution sources, etc. The above pollution situations can all be obtained by analyzing the first hyperspectral remote sensing data.
[0025] The specific steps of S12 are as follows: S121: Preprocess the most recent set of the first hyperspectral remote sensing data to obtain a first radiance map.
[0026] The preprocessing includes atmospheric correction and radiometric correction.
[0027] In the atmospheric correction step, the dark pixel method is used to remove the influence of factors such as atmospheric scattering and absorption on the signal to obtain the true surface reflectance.
[0028] In the radiometric correction step, using the calibration parameters provided by the sensor, the digital number DN in the first hyperspectral remote sensing data is converted into an actual radiance map.
[0029] S122: Calculate a first reflectance map based on the first radiance map.
[0030] In this step, it is necessary to convert the first radiance map calculated in S121 into a TOA reflectance map. The formula for the reflectance ρ TOA is as follows: ρ TOA = πL f d 2 / ESUN f cosθ; where L f is the first radiance value, d is the distance from the Earth to the Sun, ESUN f is the solar irradiance, and cosθ is the solar zenith angle.
[0031] In the first reflectance map, the reflectance values corresponding to each coordinate point in the first hyperspectral remote sensing data are recorded.
[0032] S123: Input the first reflectance map into the first absolute environmental assessment model to output the first absolute environmental assessment data.
[0033] In the first absolute environmental assessment model, the first reflectance map can be analyzed to obtain the first absolute environmental assessment data composed of water quality risk and oil pollution risk. Among them, when problems such as eutrophication occur in the water body, specific types of pollutants may change the color or transparency of the water, thus affecting the reflectance; when an oil pollution leak occurs in the water body, an oil film may form on the water surface, which appears as patches in the remote sensing image, thus affecting the reflectance.
[0034] The first absolute environmental assessment model is obtained by training a convolutional neural network. During the training process, historical data of other water conservancy regions is used as sample data. The input data is a reflectivity map, and the output data is the absolute environmental assessment data and the corresponding risk scores.
[0035] S13: Analyze multiple groups of the first hyperspectral remote sensing data to obtain the first relative environmental assessment data.
[0036] The first relative environmental assessment data refers to the environmental assessment data obtained by comparing and analyzing multiple groups of the first hyperspectral remote sensing data within a certain time interval. The relative environmental assessment indicators for water conservancy regions such as wetlands may include land coverage rate, heavy metal pollution rate, and biodiversity change rate. The above pollution situations can all be obtained by comparing and analyzing multiple groups of the first hyperspectral remote sensing data.
[0037] The S13 includes the following sub-steps: S131: Preprocess multiple groups of the first hyperspectral remote sensing data respectively to obtain multiple groups of second hyperspectral remote sensing data.
[0038] The preprocessing includes steps such as radiometric correction, atmospheric correction, geometric correction, shadow correction, and noise removal.
[0039] S132: Input multiple groups of the second hyperspectral remote sensing data into the relative environmental assessment model to obtain the first relative environmental assessment data.
[0040] The relative environmental assessment model is obtained by training an LSTM convolutional neural network model. During the training process, historical data of other water conservancy regions is used as sample data. The input data is multiple groups of hyperspectral sensing data, and the output data is the relative environmental assessment data and the corresponding risk scores.
[0041] In the relative environmental assessment model, by analyzing multiple groups of hyperspectral remote sensing data in the time series, the objective laws between the changes in the remote sensing data and the land coverage rate, heavy metal pollution rate, and biodiversity change rate can be obtained.
[0042] The first relative environmental assessment data includes the land cover change rate, plant morphological change rate, and biodiversity change rate.
[0043] S14: Obtain multiple first potential risk factors and the corresponding first risk scores according to the first absolute environmental assessment data and the first relative environmental assessment data.
[0044] S2: Determine the first scientific research topic portrait according to multiple first risk scores and multiple first potential risk factors.
[0045] In the above S1, multiple first potential risk factors have been analyzed for the first water conservancy area, including two aspects: first absolute environmental assessment data and first relative environmental assessment data. Generally, the higher the risk value of a potential risk factor, the higher the research necessity and urgency. Therefore, in this step, multiple potential risk factors to be analyzed need to be obtained through comprehensive evaluation, and then multiple first scientific research theme portraits are determined.
[0046] The above S2 includes the following sub-steps: S21: Determine the target potential risk factors whose first risk scores meet the preset requirements.
[0047] The preset requirements can be that the first risk scores rank among the top N or the first risk scores are greater than the preset value.
[0048] S22: Determine the first scientific research theme portrait based on multiple target potential risk factors.
[0049] In this step, semantic analysis is performed on multiple target potential risk factors, and multiple theme tags are extracted from multiple dimensions such as causes, risk content, and expected consequences. The first scientific research theme portrait is determined based on multiple theme tags.
[0050] S3: Determine multiple first scientific research team portraits based on the first basic information and first research result information of multiple first scientific research teams.
[0051] In the above S1 and S2, the potential risk analysis of the first water conservancy area has been completed, and the first scientific research theme portrait for characterizing the target potential risk factors has been obtained. In order to find a scientific research team that matches the first scientific research theme portrait to complete the procurement of scientific research projects, multiple first scientific research team portraits need to be determined for multiple first scientific research teams in this step.
[0052] The above S3 includes the following sub-steps: S31: Obtain the first basic information of each first scientific research team.
[0053] Among them, each first scientific research team can be composed of scientific research personnel from units or institutions such as universities and research institutes.
[0054] The first basic information mainly includes information such as the gender, educational background, and research direction of the main scientific research personnel in the first scientific research team.
[0055] S32: Obtain the first research result information of each first scientific research team.
[0056] The first research result information mainly includes scientific research projects such as the topics and projects completed by the first scientific research team. In addition, the first research result information also includes academic achievement information such as the monographs, papers, and patents published by all scientific research personnel.
[0057] S32 includes the following sub-steps: S321: Obtain a plurality of first scientific research project vectors according to the scientific research projects completed by each of the first scientific research teams.
[0058] The first scientific research project vector is obtained after semantic analysis of the scientific research project, and a plurality of keywords are extracted according to the semantic analysis result to form the first scientific research project vector.
[0059] S322: Obtain a plurality of first academic achievement vectors according to each academic achievement information.
[0060] The first academic achievement vector is obtained after semantic analysis of the academic achievements of the scientific research personnel, and a plurality of keywords are extracted according to the semantic analysis result to form the first academic achievement vector.
[0061] S323: Perform weighted clustering on the plurality of first scientific research project vectors and the plurality of first academic achievement vectors to obtain a portrait of the first scientific research team.
[0062] In this step, when performing weighted clustering on the plurality of first scientific research project vectors and the plurality of first academic achievement vectors, K-MEANS can be used as the weighted clustering algorithm. A higher weight value can be assigned to the first scientific research project vector, and a lower weight value can be assigned to the first academic achievement vector. Thus, a plurality of theme tags representing the main research directions of the first scientific research team can be obtained, and then the portrait of the first scientific research team can be determined according to the plurality of theme tags.
[0063] S4: Determine at least one target scientific research team for water conservancy scientific research project procurement according to the similarity between the first scientific research theme portrait and the plurality of first scientific research team portraits.
[0064] Compare the first scientific research theme portrait determined in S2 with the plurality of first scientific research team portraits determined in S3, and thus determine the scientific research team corresponding to the first scientific research team portrait whose similarity meets the preset requirements as the target scientific research team, so as to send a scientific research project procurement application to the determined target scientific research team subsequently.
[0065] This application also proposes a water conservancy scientific research project matching system based on big data for executing any one of the above-mentioned water conservancy scientific research project matching methods.
[0066] A method and system for matching water conservancy scientific research projects based on big data proposed in this application first collects hyperspectral remote sensing images of a first water conservancy area to be scientifically researched, analyzes the hyperspectral remote sensing images within a certain time interval to determine potential risk factors and risk scores of the first water conservancy area from two dimensions of absolute data analysis and relative data analysis. Secondly, target risk factors are determined according to the potential risk factors and risk scores, and then a first scientific research theme portrait corresponding to the first water conservancy area is determined. Then, using the basic information and academic achievement information of the scientific research teams to be pushed, multiple first scientific research team portraits are determined. Finally, based on the similarity between the first scientific research theme portrait and multiple first scientific research team portraits, a target scientific research team is determined and a scientific research project is purchased. Through the technical solution of this application, the potential risks of the water conservancy area can be analyzed from multiple dimensions, and a scientific research team with a matching degree meeting the preset requirements can be found for the purchase of scientific research projects, which can effectively improve the accuracy of scientific research project procurement.
[0067] The above are only the preferred embodiments of the present invention. Therefore, all equivalent changes or modifications made according to the structures, features and principles described in the scope of this invention patent application are included in the scope of this invention patent application.
Claims
1. A water conservancy scientific research project matching method based on big data, characterized in that: The method includes: S1: Analyze the potential risk factors of the first water conservancy area to obtain a plurality of first potential risk factors and corresponding first risk scores; The first potential risk factor includes two types: absolute environmental assessment data and relative environmental assessment data; S2: determining a first scientific research topic profile according to a plurality of the first risk scores and a plurality of the first potential risk factors; The S2 comprises the following sub-steps: S21: determining the first potential risk factor whose first risk score meets a preset requirement as a target potential risk factor; S22: Determine a first scientific research subject profile based on a plurality of potential risk factors of the target; S3: Determine multiple first scientific research team portraits based on the first basic information and the first research achievement information of the multiple first scientific research teams; S4: Based on the similarity between the first scientific research theme portrait and the portraits of multiple first scientific research teams, determine at least one target scientific research team for water conservancy scientific research project procurement.
2. According to the big data-based water conservancy scientific research project matching method of claim 1, it is characterized in that: The S1 comprises the following sub-steps: S11: Acquire multiple groups of first hyperspectral remote sensing data of the first water conservancy area at first time intervals; S12: Analyze and obtain first absolute environmental assessment data through a group of the first hyperspectral remote sensing data that is most recent in time; wherein the first absolute environmental assessment data includes water quality parameters, thermal pollution sources, and oil pollution source data; S13: Analyze and obtain first relative environmental assessment data through multiple groups of the first hyperspectral remote sensing data; wherein the first relative environmental assessment data includes land coverage rate, heavy metal pollution rate and biodiversity change rate data; S14: Obtain a plurality of first potential risk factors and corresponding first risk scores according to the first absolute environment assessment data and the first relative environment assessment data.
3. A water conservancy scientific research project matching method based on big data according to claim 2, characterized in that: The S12 comprises the following sub-steps: S121: preprocessing a group of the first hyperspectral remote sensing data that is most recent in time to obtain a first radiation brightness map; S122: Calculating a first reflectivity map according to the first radiance map; S123: Inputting the first reflectivity map into a first absolute environment assessment model to output first absolute environment assessment data.
4. A water conservancy scientific research project matching method based on big data according to claim 3, characterized in that: The reflectivity ρ in S122 TOA The calculation formula is as follows: ρ TOA =πL f d 2 / ESUN f cosθ; where L f is the first radiance value, d is the distance from the Earth to the Sun, ESUN f is the solar irradiance and cosθ is the solar zenith angle.
5. A water conservancy scientific research project matching method based on big data according to claim 4, characterized in that: The S13 comprises the following sub-steps: S131: Preprocessing the multiple groups of the first hyperspectral remote sensing data respectively to obtain multiple groups of second hyperspectral remote sensing data; S132: Inputting multiple groups of the second hyperspectral remote sensing data into a relative environmental assessment model to obtain the first relative environmental assessment data.
6. A water conservancy scientific research project matching method based on big data according to claim 5, characterized in that: The S3 comprises the following sub-steps: S31: Obtaining first basic information of each of the first scientific research teams; S32: Obtain the first research achievement information of each of the first scientific research teams.
7. A water conservancy scientific research project matching method based on big data according to claim 6, characterized in that: The S32 includes the following sub-steps: S321: Obtain multiple first scientific research project vectors according to the scientific research projects completed by each of the first scientific research teams; S322: Obtain multiple first academic achievement vectors according to each academic achievement information; S323: Perform weighted clustering on multiple first scientific research project vectors and multiple first academic achievement vectors to obtain a portrait of the first scientific research team.
8. A water conservancy scientific research project matching system based on big data, used to implement a water conservancy scientific research project matching method based on big data as described in any one of claims 1 to 7.