Reservoir capacity estimation method and system based on unmanned surveying vessel and airborne Lidar

Abstract

The invention provides a reservoir capacity estimation method and system based on an unmanned surveying vessel and airborne Lidar. The method comprises the following steps: planning a first measurement path and a second measurement path; the unmanned surveying vessel is used for completing measurement of reservoir underwater topographic data according to the first measurement path; meanwhile, an unmanned aerial vehicle with an airborne Lidar is used for completing measurement of water and shore terrain data of the reservoir according to a second measurement path; acquiring the interval duration of the measurement time; planning a third measurement path, and measuring the above-water topographic data and the shore topographic data of the reservoir again according to the third measurement path within the interval duration; preprocessing the obtained data: integrating measurement results of two times, and selecting reservoir overwater and shoreside terrain data with higher quality; and fusing the preprocessed reservoir underwater topographic data and the reservoir overwater and shoreside topographic data, constructing a reservoir three-dimensional model, and estimating the reservoir capacity according to the volume of the reservoir three-dimensional model. According to the invention, the efficiency and accuracy of reservoir capacity estimation can be improved.

Description

Technical Field

[0001] This application relates to the field of reservoir capacity measurement technology, specifically to a reservoir capacity estimation method and system based on an unmanned survey vessel and airborne LiDAR. Background Technology

[0002] Reservoir capacity measurement is a crucial component of water conservancy and hydropower engineering surveying and geographic information system (GIS) technology, playing a key role in the design, construction, and management of reservoirs. Traditional methods for reservoir capacity measurement mainly include three types: First, the water level-area method, which estimates capacity by measuring the water surface area at different water levels; this method is simple to operate but has low accuracy. Second, the cross-sectional method, which sets up several cross-sections, measures the water depth at each section, and then calculates the total volume; this method has higher accuracy but is time-consuming and labor-intensive. Third, the remote sensing image method, which acquires two-dimensional images of the water area through satellite or aircraft photography, then uses specialized software to convert them into three-dimensional topographic maps, using this as the basis for calculating the reservoir capacity; this method has a wide coverage area, but the accuracy of measurements in deep water areas and turbid water conditions is poor. In addition, some studies have attempted to use single sensors (such as a single UAV equipped with LiDAR equipment) for measurement, but due to factors such as flight altitude limitations and weather conditions, the measurement results still have certain limitations.

[0003] As can be seen from the above, existing measurement methods are either insufficiently accurate or complex to operate and costly, especially in reservoir environments with extreme climatic conditions or complex topographic features, where the effectiveness of traditional methods is even more limited. These problems not only affect the efficiency of measurement work but may also lead to unreliable measurement results, thereby affecting the scientific validity and effectiveness of reservoir management decisions. Therefore, there is an urgent need to provide an improved method for measuring and estimating reservoir capacity to achieve efficient and accurate estimation of reservoir capacity. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for estimating reservoir capacity based on an unmanned survey vessel and airborne LiDAR, which can effectively improve the efficiency and accuracy of reservoir capacity estimation.

[0005] The technical solution adopted in this invention is as follows:

[0006] In a first aspect, the present invention provides a method for estimating reservoir capacity based on an unmanned survey vessel and an airborne LiDAR, comprising:

[0007] Based on the reservoir's structural information and geographical information, a first measurement path and a first measurement speed, and a second measurement path and a second measurement speed are planned, so that the distance between the corresponding positions of the first measurement path and the second measurement path at the same time is greater than a preset distance.

[0008] The underwater topographic data of the reservoir is measured using an unmanned survey vessel following a first measurement path; the first measurement time for measuring the underwater topographic data of the reservoir is calculated according to a first measurement speed; simultaneously, the topographic data of the reservoir surface and shoreline is measured using an unmanned aerial vehicle equipped with an airborne LiDAR following a second measurement path; the second measurement time for measuring the topographic data of the reservoir surface and shoreline is calculated according to a second measurement speed; and the interval between the first measurement time and the second measurement time is calculated.

[0009] Based on reservoir construction information, reservoir geographic information and interval duration, a third measurement path and a third measurement speed are planned so that the distance between the corresponding positions of the third measurement path and the first measurement path at the same time is greater than the preset distance. Within the interval duration, a UAV equipped with airborne LiDAR is used to complete the measurement of the reservoir's water surface and shore topography data again according to the third measurement path.

[0010] The underwater topographic data of the reservoir and the surface and shore topographic data of the reservoir obtained by measurement are preprocessed. The preprocessing includes: performing similarity matching on the surface and shore topographic data of the reservoir at the same location obtained by two measurements, retaining one data point from a set of surface and shore topographic data with a similarity greater than a preset similarity, and selecting the surface and shore topographic data with higher data quality from the set of surface and shore topographic data with a self-similarity not greater than the preset similarity; fusing the preprocessed underwater topographic data and the surface and shore topographic data of the reservoir, constructing a three-dimensional model of the reservoir using the fused data, and estimating the reservoir capacity based on the volume of the three-dimensional model.

[0011] By adopting the above scheme, multi-source data for reservoir topographic measurement is obtained by combining the multibeam sonar of the unmanned survey vessel and airborne LiDAR, eliminating the error of a single measurement method and improving measurement accuracy. Considering that the measurement of reservoir topographic data on the water and shore is greatly affected by the environment, multiple measurements are conducted, further improving the accuracy and efficiency of the measurement. No personnel intervention is required during the entire measurement process, ensuring the safety of the personnel.

[0012] Preferably, it also includes: during the measurement by the unmanned survey vessel, using water quality sensors and meteorological sensors installed on the unmanned survey vessel to collect real-time groundwater quality data and meteorological data of the reservoir;

[0013] The preprocessing of the measured underwater topographic data of the reservoir with the above-water and shore topographic data of the reservoir also includes:

[0014] The first return signal received during the process of obtaining underwater topographic data of a reservoir using an unmanned survey vessel is filtered and amplified. The filtered and amplified first return signal, reservoir groundwater quality data, and meteorological data are then input into a first quality assessment model, which outputs an assessment result indicating whether the quality of the first return signal is up to standard. First return signals with an assessment result indicating that the quality is not up to standard are discarded. The first quality assessment model is a neural network model, which is trained and generated using historical first return signals received during unmanned survey vessel measurements, historical reservoir groundwater quality data, and historical meteorological data, which are labeled with whether the quality is up to standard or not.

[0015] The second return signal received during the process of obtaining reservoir water surface and shore topographic data using unmanned survey vessels is filtered and amplified. The filtered and amplified second return signal and the meteorological data corresponding to the time of the received second return signal are input into a second quality assessment model, which outputs the assessment result of whether the signal quality is qualified or not, and the second return signal with the assessment result of unqualified signal quality is removed. The second quality assessment model adopts a neural network model, which is trained and generated by using historical second return signals received during the measurement process using unmanned survey vessels with quality marked as qualified or not, historical reservoir groundwater quality data and historical meteorological data.

[0016] By adopting the above scheme, considering the impact of water quality and meteorological data on pulse signal transmission and reception during reservoir underground topography measurement, the quality assessment threshold of received signals under different water quality and meteorological data conditions is adaptively adjusted to more accurately judge signal quality and eliminate unqualified return signals, thereby ensuring the reliability and accuracy of the final data.

[0017] Preferably, the step of using a UAV equipped with an airborne LiDAR to complete the measurement of the reservoir's surface and shoreline topographic data according to the second measurement path further includes:

[0018] This method addresses the process of using an airborne LiDAR-equipped UAV to measure the topographic data of a reservoir's surface and shoreline along a second measurement path. Real-time image acquisition of the reservoir along the path's forward direction is performed. Image analysis techniques are used to determine the topographic classification within the reservoir images. This classification includes water bodies, vegetation, bare soil, and buildings. Preset pulse parameters are matched to each determined topographic classification. Each topographic classification has preset pulse parameters, including pulse frequency and intensity. During the measurement of the reservoir's surface and shoreline topographic data using the airborne LiDAR-equipped UAV along the second measurement path, the matched pulse parameters are selected for pulse transmission, and the corresponding return signals are received.

[0019] By adopting the above scheme, considering that different terrain classifications have different absorption and reflection effects on airborne LiDAR pulse signals, the corresponding pulse parameters are automatically selected for transmission based on the terrain classification results, thereby improving the relevance and accuracy of signal reception and ensuring the accuracy of the final measurement results.

[0020] Preferably, it also includes: collecting reservoir environmental data; the reservoir environmental data includes water quality data and meteorological data;

[0021] The process of fusing preprocessed underwater topographic data with surface and shore topographic data includes: inputting reservoir environmental data, preprocessed underwater topographic data, and surface and shore topographic data into a multimodal data fusion model; and using the multimodal data fusion model to fuse the underwater topographic data with the surface and shore topographic data. The multimodal data fusion model employs a neural network model, which is generated by training on the reservoir capacity data corresponding to the calculated underwater topographic data and surface and shore topographic data corresponding to the reservoir capacity data with the smallest discrepancy between historical and actual reservoir capacity data, as well as historical underwater topographic data, historical surface and shore topographic data, and historical reservoir environmental data.

[0022] By adopting the above scheme, the influence of water quality and meteorological factors on the measurement results was considered. A multimodal data fusion model was used to fuse the underwater topographic data of the reservoir with the topographic data of the reservoir surface and shore, which further improved the quality of data fusion and enhanced the reliability and integrity of the data.

[0023] Preferred options also include:

[0024] The reservoir is divided into important and non-important areas based on its structural and geographical information.

[0025] For key areas of the reservoir, multiple unmanned survey vessels are used to record underwater topographic data of the reservoir in different time periods. Each unmanned survey vessel completes the measurement of the underwater topographic data of the reservoir according to the first measurement path. The preprocessing of the underwater topographic data of the reservoir and the topographic data of the reservoir surface and shoreline also includes: selecting the highest quality underwater topographic data of the reservoir from multiple sets of underwater topographic data corresponding to the same location in the first measurement path as the underwater topographic data of the reservoir at the current location.

[0026] By adopting the above scheme, and considering the important areas in the reservoir, multiple unmanned survey vessels are used for simultaneous measurement, which increases the diversity and redundancy of data collection, reduces the risk of measurement failure due to the failure of a single device, and selects the highest quality underwater topographic data of the reservoir as the data for the current location, thus ensuring the high quality and consistency of the final data.

[0027] Preferably, it also includes: calculating the reservoir capacity by using numerical integration or volume rendering to calculate the volume of the 3D reservoir model.

[0028] By adopting the above scheme, the accuracy and efficiency of reservoir capacity calculation are significantly improved by using numerical integration or volume rendering methods.

[0029] Preferably, the method further includes: during the process of measuring underwater topographic data of the reservoir using an unmanned survey vessel along the first measurement path, using a Kongsberg EM 2040 or Teledyne Reson T20-P multibeam sonar device to complete the measurement; and during the process of measuring surface and shore topographic data of the reservoir using an unmanned aerial vehicle equipped with an airborne LiDAR along the second measurement path, using a Riegl VUX-1HA LiDAR sensor to complete the measurement.

[0030] By adopting the above scheme, using high-precision multibeam sonar and radar, high-precision reservoir topographic data and detailed mapping of the water and shore topography are provided, effectively reducing measurement errors and improving the reliability and accuracy of the measurement.

[0031] Secondly, the present invention provides a reservoir capacity estimation system based on an unmanned survey vessel and airborne LiDAR, comprising:

[0032] The measurement parameter planning module is used to plan a first measurement path and a first measurement speed, and a second measurement path and a second measurement speed based on the reservoir structure information and reservoir geographic information, so that the distance between the corresponding positions of the first measurement path and the second measurement path at the same time is greater than a preset distance.

[0033] The measurement data acquisition module is used to complete the measurement of underwater topographic data of the reservoir using an unmanned survey vessel according to a first measurement path; calculate the first measurement time for completing the underwater topographic data measurement of the reservoir according to a first measurement speed; simultaneously, use an unmanned aerial vehicle equipped with an airborne LiDAR to complete the measurement of the topographic data of the reservoir surface and shore according to a second measurement path; calculate the second measurement time for measuring the topographic data of the reservoir surface and shore according to a second measurement speed; and calculate the interval between the first measurement time and the second measurement time.

[0034] The redundant measurement data acquisition module is used to plan a third measurement path and a third measurement speed based on reservoir building information, reservoir geographic information and interval duration, so that the distance between the corresponding positions of the third measurement path and the first measurement path at the same time is greater than a preset distance, and within the interval duration, the UAV with airborne LiDAR is used to complete the measurement of the reservoir water and shore topography data again according to the third measurement path.

[0035] The reservoir capacity estimation module is used to preprocess the underwater topographic data and the surface and shore topographic data of the reservoir obtained by measurement. The preprocessing includes: performing similarity matching on the surface and shore topographic data of the reservoir obtained from two measurements at the same location; retaining one data point from a set of surface and shore topographic data with a similarity greater than a preset similarity; selecting the surface and shore topographic data with higher data quality from the set of surface and shore topographic data with a self-similarity not greater than the preset similarity; fusing the preprocessed underwater topographic data and the surface and shore topographic data; constructing a three-dimensional model of the reservoir using the fused data; and estimating the reservoir capacity based on the volume of the three-dimensional model.

[0036] By adopting the above scheme, multimodal measurement data is acquired through the collaborative operation of unmanned survey vessels and airborne LiDAR, eliminating the errors present in single measurement methods. By combining multiple measurements and data preprocessing, high-quality data is retained for fusion, ensuring that the final constructed three-dimensional model is more accurate and reliable, thereby achieving the accuracy of reservoir capacity calculation.

[0037] Thirdly, the present invention provides a computer-readable storage medium comprising a stored computer program, wherein the computer program, when running, controls the device containing the computer-readable storage medium to perform the method described above.

[0038] Fourthly, the present invention provides a computer device, the computer device including a memory, a processor and a program stored in the memory and executable thereon, the program being executed by the processor to perform the steps of the method described above.

[0039] The reservoir capacity estimation method and system based on unmanned survey vessels and airborne LiDAR provided by this invention have the following advantages:

[0040] 1. By using unmanned survey vessels and airborne LiDAR to conduct measurements, multi-source data on the reservoir and its shoreline topography are obtained, improving measurement accuracy. At the same time, considering actual environmental factors and measurement process duration, multiple water measurements are conducted to increase data redundancy and ensure the accuracy and efficiency of the measurement data. No human intervention is required throughout the entire measurement process, improving efficiency and ensuring the personal safety of personnel.

[0041] 2. Considering the impact of environmental factors on the quality of measurement data, and combining the adaptability of neural network algorithms, the quality assessment threshold of received signals under different water quality and meteorological data conditions is adjusted according to environmental data to eliminate measurement data that does not meet the quality requirements, thereby ensuring the reliability and accuracy of the measurement data;

[0042] 3. Considering the influence of different terrain types on the absorption and reflection of transmitted pulse signals, the appropriate pulse parameters are automatically selected for transmission based on the terrain classification results, improving the relevance and accuracy of signal reception, thereby ensuring the accuracy of the final measurement data. Attached Figure Description

[0043] Figure 1 This is a flowchart of a reservoir capacity estimation method based on an unmanned survey vessel and airborne LiDAR in a specific embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of the reservoir capacity estimation system based on an unmanned survey vessel and airborne LiDAR in a specific embodiment of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0046] like Figure 1 As shown in the figure, this application discloses a method for estimating reservoir capacity based on an unmanned survey vessel and airborne LiDAR, the specific steps of which include:

[0047] S1. Based on the reservoir structure information and reservoir geographic information, plan the measurement parameters, including: planning the first measurement path and the first measurement speed, the second measurement path and the second measurement speed, so that the distance between the corresponding positions of the first measurement path and the second measurement path at the same time is greater than the preset distance;

[0048] To comprehensively measure the reservoir's capacity, an unmanned survey vessel equipped with multibeam sonar was used for underwater topographic scanning, while a drone equipped with onboard LiDAR was used for detailed mapping of the above-water portion and shoreline area. Key components on the unmanned survey vessel include a high-performance microprocessor as the central controller, an industrial-grade GPS module for accurate positioning, and a multibeam sonar for deep-water detection. Different brands and models of multibeam sonar can be selected based on budget and project requirements, such as the Kongsberg EM 2040 or Teledyne Reson T20-P. For the drone, in addition to the basic flight platform, a high-resolution LiDAR sensor, a three-axis stabilized gimbal, and sufficient battery life are required. In this embodiment, the Riegl VUX-1HA LiDAR sensor was selected due to its long-range measurement capabilities and high data density, making it ideal for high-speed scanning of large, complex terrains. Several LiDAR options are available for the drone, including the Velodyne PuckLITE and Hesai Pandar series, in addition to the Riegl VUX-1HA.

[0049] To complete the measurement, it is necessary to plan the measurement paths and speeds of the unmanned survey vessel and drone in advance based on the reservoir's architectural and geographical information. Specifically, based on the reservoir's location, area, structure, and information on obstacles or structures within the reservoir recorded in the architectural and geographical information, a measurement route is designed for the unmanned survey vessel to ensure that the measurement range covers the entire underground water area of ​​the reservoir; this is denoted as the first measurement path. To ensure effective transmission and reception of signals from the multibeam sonar equipment at each path location, a corresponding measurement speed is designed. This speed can be determined based on the propagation speed of the transmitted pulse signal and the depth of the underground water area of ​​the reservoir, or by the most commonly used speed recorded in historical measurements; this is denoted as the first measurement speed.

[0050] To improve measurement efficiency, the plan is to simultaneously conduct measurements on the reservoir's surface and shoreline topography using an unmanned survey vessel (USV) equipped with an airborne LiDAR. Considering potential signal interference between adjacent USVs and USVs during measurements, the planned USV measurement path, covering the reservoir's surface and shoreline areas, ensures that the distance between the corresponding positions on the first and second measurement paths at any given moment is greater than a preset distance, denoted as the second measurement path. This preset distance can be determined based on the interference range of different frequency signals emitted by the airborne LiDAR and multibeam sonar equipment. To ensure effective transmission and reception of signals from the multibeam sonar equipment at each path location, a corresponding measurement speed is designed. This speed can be determined based on the signal propagation speed and a preset flight altitude, or by the most frequently used speed recorded in historical measurements, and is denoted as the second measurement speed.

[0051] Furthermore, considering the potential hazards of underground obstacles in the reservoir, and the greater impact of variable environmental data on underwater topographic surveying, the first measurement speed of the planned unmanned survey vessel is lower than the second measurement speed of the UAV.

[0052] S2. Use an unmanned survey vessel to complete the measurement of underwater topographic data of the reservoir according to the first measurement path; calculate the first measurement time for completing the measurement of underwater topographic data of the reservoir according to the first measurement speed; at the same time, use an unmanned aerial vehicle equipped with an airborne LiDAR to complete the measurement of topographic data of the reservoir surface and shore according to the second measurement path; calculate the second measurement time for measuring topographic data of the reservoir surface and shore according to the second measurement speed; calculate the interval between the first measurement time and the second measurement time.

[0053] Specifically, staff control the unmanned survey vessel and drone to send a start command. Upon receiving the command, the unmanned vessel's built-in program automatically adjusts its attitude to move smoothly along the first measurement path and controls its sonar device to periodically emit pulsed sound waves in all directions, capturing the echoes generated when it encounters obstacles, thus completing the measurement of the reservoir's underwater topographic data. Simultaneously, the drone, after taking off upon receiving the start command, moves along the second measurement path. The LiDAR continuously emits laser pulses towards the ground; each successful hit generates a reflected signal that is transmitted back to the receiver, completing the measurement of the reservoir's surface and shore topographic data.

[0054] In addition, during the entire measurement process, the first measurement time for completing the underwater topographic data measurement of the reservoir is calculated according to the first measurement speed, and the second measurement time for measuring the topographic data of the reservoir surface and shore is calculated according to the second measurement speed; the interval between the first measurement time and the second measurement time is calculated and combined with the interval.

[0055] S3. Based on the reservoir structure information, reservoir geographic information and interval duration, plan the third measurement path and the third measurement speed so that the distance between the corresponding positions of the third measurement path and the first measurement path at the same time is greater than the preset distance, and within the interval duration, use a UAV with airborne LiDAR to complete the measurement of the reservoir's water surface and shore topography data again according to the third measurement path.

[0056] Considering the impact of variable environmental data on underwater topographic surveying of reservoirs, the first measurement speed of the planned unmanned survey vessel is lower than the second measurement speed of the UAV. The interval between measurement completions allows the UAV to acquire topographic data of the reservoir surface and shoreline again, thus achieving redundant acquisition of topographic data of the reservoir surface and shoreline.

[0057] Specifically, based on reservoir structure information, reservoir geographic information, and interval duration, a third measurement path and a third measurement speed for the UAV are planned. This ensures that the distance between the positions corresponding to the third measurement path and the planned first measurement path at the same instant is greater than a preset distance, and that the third path avoids signal interference. Within the interval duration, the UAV, equipped with an onboard LiDAR, re-measures the reservoir's surface and shoreline topographic data according to the third measurement path. To guarantee successful UAV measurements, the third measurement speed is kept below a UAV measurement speed threshold.

[0058] S4. Preprocess the underwater topographic data of the reservoir and the topographic data of the reservoir surface and shore obtained by measurement. The preprocessing includes: performing similarity matching on the topographic data of the reservoir surface and shore at the same location obtained by two measurements, retaining one data from a set of topographic data of the reservoir surface and shore with a similarity greater than a preset similarity, and selecting the topographic data of the reservoir surface and shore with higher data quality from the topographic data of the reservoir surface and shore with a similarity not greater than the preset similarity.

[0059] Specifically, the preprocessing includes filtering and amplification, noise reduction, redundant data screening, and measurement data integration processing to align the spatial locations corresponding to the planned paths. Among these, for reservoir water surface and shoreline topographic data, some data is redundant. Similarity matching is performed on the reservoir water surface and shoreline topographic data at the same location obtained from two measurements (e.g., the same location in the second and third paths). One data point from the set of reservoir water surface and shoreline topographic data with a similarity greater than a preset similarity is retained. From the set of reservoir water surface and shoreline topographic data with a self-similarity not greater than the preset similarity, the data with higher quality is selected.

[0060] The preset similarity can be set manually, such as 95% or higher. The quality of the measurement data during the measurement process can be determined based on the received signal quality indicators, including signal-to-noise ratio, signal strength, etc. One indicator can be selected as the evaluation standard or the evaluation can be based on the weighted result of several quality indicator parameters.

[0061] S5. The preprocessed underwater topographic data of the reservoir is fused with the topographic data of the reservoir surface and shore. The fused data is used to construct a three-dimensional model of the reservoir, and the reservoir capacity is estimated based on the volume of the three-dimensional model.

[0062] Specifically, data fusion algorithms, such as weighted averaging, Kalman filtering, or optimization theory algorithms, can be used to preprocess the underwater topographic data and the surface and shore topographic data of the reservoir. This includes: extracting key features from the preprocessed underwater and surface / shore topographic data; using feature matching algorithms or spatial relationships based on the planned route to identify corresponding points or similar areas in the two sets of data that are vertically aligned; weighted fusion of the identified key features; converting the fused data into point cloud coordinates using a specific algorithm (such as a 3D reconstruction algorithm); constructing a 3D model of the reservoir using the generated point cloud coordinates; and calculating the reservoir capacity by numerical integration or volume rendering.

[0063] In a specific embodiment, considering the impact of different water quality and meteorological data on reservoir underground topography measurements or meteorological data on reservoir surface and shoreline topography measurements, the threshold for judging the received signal quality is adaptively adjusted under poor water quality or severe weather conditions to effectively remove noise and outliers; the method further includes:

[0064] When conducting measurements on an unmanned survey vessel, water quality sensors and meteorological sensors installed on the unmanned survey vessel are used to collect real-time groundwater quality data and meteorological data of the reservoir; when conducting measurements on a drone equipped with lidar, meteorological sensors installed on the drone are used to collect real-time meteorological data.

[0065] The preprocessing of the measured underwater topographic data of the reservoir with the above-water and shore topographic data of the reservoir also includes:

[0066] The first return signal received during the process of obtaining underwater topographic data of the reservoir using an unmanned survey vessel is filtered and amplified. The filtered and amplified first return signal, the reservoir groundwater quality data and meteorological data corresponding to the time of the first return signal are input into the first quality assessment model. The model outputs the assessment result of whether the quality of the first return signal is qualified or not, and the first return signal with the assessment result of unqualified quality is removed.

[0067] The first quality assessment model is a neural network model, which is generated by training on historical groundwater quality data and meteorological data received during the measurement process using an unmanned measuring vessel, which are labeled with whether the quality is qualified or not. The first quality assessment threshold in the first quality assessment model is adaptively adjusted according to the groundwater quality data and meteorological data.

[0068] The second return signal received during the process of obtaining reservoir water surface and shore topographic data using unmanned survey vessels is filtered and amplified; the filtered and amplified second return signal and the meteorological data corresponding to the time of the received second return signal are input into the second quality assessment model, and the assessment result of whether the signal quality is qualified or not is output, and the second return signal with the assessment result of unqualified signal quality is removed;

[0069] The second quality assessment model is a neural network model, which is generated by training on historical data of groundwater quality in reservoirs and historical meteorological data, including the second return signal received during the measurement process using an unmanned measuring vessel and the data labeled as qualified or unqualified. The second quality assessment threshold in the second quality assessment model is adaptively adjusted according to the meteorological data.

[0070] In a specific embodiment, considering the impact of different topographic classifications on reservoir underground topographic measurements or meteorological data on reservoir surface and shore topographic measurements, the measurement signal transmission strategy is adjusted to adapt to different reservoir topographic and reflection conditions, thereby improving signal reception quality. The method further includes: the measurement of reservoir surface and shore topographic data using a UAV equipped with an airborne LiDAR along a second measurement path further includes:

[0071] For the process of using an UAV equipped with an airborne LiDAR to complete the measurement of the topographic data of the reservoir's surface and shoreline along the second measurement path, a camera device also mounted on the UAV is used to collect real-time image information of the reservoir in the area along the direction of the second measurement path.

[0072] Based on image analysis technology, the topographic classification of the reservoir image information is determined. This classification includes: water bodies, vegetation, bare soil, and buildings. Preset pulse parameters are matched according to the determined topographic classification. Each topographic classification has preset pulse parameters, including pulse frequency and intensity. The specific pulse parameters matched for each topographic classification can be determined based on the quality of the returned signals received under different pulse frequencies and intensities under historical conditions of the same topographic type. The pulse frequency and intensity corresponding to the strongest returned signal quality are selected, or the selection is based on expert experience. During the measurement of the reservoir's surface and shoreline topographic data using a UAV equipped with an airborne LiDAR along a second measurement path, the matched pulse parameters are selected for pulse transmission, and the corresponding returned signals are received.

[0073] In addition, the use of unmanned survey vessels to complete the measurement of underwater topographic data of the reservoir according to the first measurement path also includes:

[0074] Based on reservoir structure information and historical underwater image data, the measurement terrain classification of different areas in the reservoir's underground water area is determined in advance. When the unmanned measuring vessel travels to a specific area in the reservoir's underground water area according to the first measurement path, the preset pulse parameters are matched according to the measurement terrain classification of that specific area. During the process of measuring the underwater terrain data of the reservoir using the unmanned measuring vessel according to the first measurement path, the matched pulse parameters are selected to transmit pulse signals and receive corresponding return signals.

[0075] In a specific embodiment, considering the impact of different environmental data on the accuracy of multi-source measurement data during the data fusion process, the fusion weights of different multi-source measurement data are adjusted accordingly to achieve more accurate data fusion; the method further includes: collecting reservoir environmental data; the reservoir environmental data includes water quality data and meteorological data;

[0076] The process of fusing the preprocessed underwater topographic data of the reservoir with the above-water and shoreline topographic data includes:

[0077] Instead of directly using the weighted fusion method in the above embodiments, a multimodal data fusion model is selected to dynamically fuse underwater topographic data and above-water and shore topographic data of the reservoir.

[0078] Environmental data of the reservoir, preprocessed underwater topographic data of the reservoir, and topographic data of the reservoir surface and shore are input into a multimodal data fusion model. The multimodal data fusion model is used to fuse the underwater topographic data of the reservoir and the topographic data of the reservoir surface and shore. The multimodal data fusion model adopts a neural network model, which is generated by training on the underwater topographic data of the reservoir corresponding to the reservoir capacity data with the smallest difference between historical and actual reservoir capacity data, the fused data of the underwater topographic data of the reservoir surface and shore, historical underwater topographic data of the reservoir, historical topographic data of the reservoir surface and shore, and historical environmental data.

[0079] In addition, after obtaining the reservoir capacity, the obtained reservoir capacity is compared with the actual reservoir capacity to determine whether the difference between the two is greater than a preset difference. If it is greater than the preset difference, incremental learning optimization is performed on the multimodal data fusion model. The underwater topographic data of the reservoir and the topographic data of the reservoir surface and shore are fused again using the optimized multimodal data fusion model. The obtained reservoir capacity and the actual reservoir capacity are calculated repeatedly, and the difference between the two is determined whether it is greater than the preset difference. The optimization of the multimodal data fusion model is stopped when the difference is greater than the preset difference.

[0080] In a specific embodiment, to further ensure the accuracy of the measurement data, the redundancy of the measurement data can be increased to filter out higher-quality measurement data and improve the accuracy of the final reservoir capacity. The method also includes:

[0081] The reservoir is divided into important and non-important areas based on its architectural and geographical information; the important areas can be predefined.

[0082] For key areas of the reservoir, multiple unmanned survey vessels were used to record underwater topographic data of the reservoir by each unmanned survey vessel according to the first survey path at different time periods; that is, the measurement data completed by different unmanned survey vessels in key areas of the reservoir were recorded at different time periods.

[0083] The preprocessing of the underwater topographic data of the reservoir obtained by measurement and the topographic data of the reservoir surface and shoreline also includes: selecting the highest quality underwater topographic data of the reservoir from multiple sets of underwater topographic data corresponding to the same location in the first measurement path as the underwater topographic data of the reservoir at the current location.

[0084] In one specific embodiment, the reservoir capacity is estimated by calculating the volume of a 3D reservoir model using numerical integration or volume rendering.

[0085] In one specific embodiment, during the process of measuring underwater topographic data of the reservoir using an unmanned survey vessel along the first measurement path, a multibeam sonar device of the Kongsberg EM 2040 or Teledyne Reson T20-P model is selected to complete the measurement; during the process of measuring surface and shore topographic data of the reservoir using an unmanned aerial vehicle equipped with an airborne LiDAR along the second measurement path, a Riegl VUX-1HA LiDAR sensor is selected to complete the measurement.

[0086] like Figure 2 As shown, a reservoir capacity estimation system based on an unmanned survey vessel and airborne LiDAR specifically includes:

[0087] The measurement parameter planning module 101 is used to plan a first measurement path and a first measurement speed, a second measurement path and a second measurement speed based on the reservoir structure information and reservoir geographic information, so that the distance between the corresponding positions of the first measurement path and the second measurement path at the same time is greater than a preset distance.

[0088] The measurement data acquisition module 102 is used to complete the measurement of underwater topographic data of the reservoir using an unmanned survey vessel according to a first measurement path; calculate the first measurement time for completing the underwater topographic data measurement of the reservoir according to a first measurement speed; simultaneously, use an unmanned aerial vehicle equipped with an airborne LiDAR to complete the measurement of the topographic data of the reservoir surface and shoreline according to a second measurement path; calculate the second measurement time for the topographic data of the reservoir surface and shoreline according to a second measurement speed; and calculate the interval between the first measurement time and the second measurement time.

[0089] The measurement data redundancy acquisition module 103 is used to plan a third measurement path and a third measurement speed based on reservoir building information, reservoir geographic information and interval duration, so that the distance between the corresponding positions of the third measurement path and the first measurement path at the same time is greater than a preset distance, and within the interval duration, the UAV with airborne LiDAR is used to complete the measurement of the reservoir water and shore topography data again according to the third measurement path.

[0090] The reservoir capacity estimation module 104 is used to preprocess the underwater topographic data and the surface and shore topographic data of the reservoir obtained by measurement. The preprocessing includes: performing similarity matching on the surface and shore topographic data of the reservoir obtained from two measurements at the same location; retaining one data point from a set of surface and shore topographic data with a similarity greater than a preset similarity; selecting the surface and shore topographic data with higher data quality from the surface and shore topographic data with a self-similarity not greater than the preset similarity; fusing the preprocessed underwater topographic data and the surface and shore topographic data; constructing a three-dimensional model of the reservoir using the fused data; and estimating the reservoir capacity based on the volume of the three-dimensional model.

[0091] In one specific embodiment, the system further includes a measurement parameter acquisition module 105, which is used to collect real-time groundwater quality data and meteorological data of the reservoir using water quality sensors and meteorological sensors installed on the unmanned measurement vessel during measurement.

[0092] The reservoir capacity estimation module 104 in the system is also used to filter and amplify the first return signal received during the process of obtaining underwater topographic data of the reservoir using unmanned survey vessels; input the filtered and amplified first return signal, reservoir groundwater quality data and meteorological data into the first quality assessment model, output the assessment result of whether the quality of the first return signal is qualified or not, and discard the first return signal with the assessment result of unqualified quality; the first quality assessment model adopts a neural network model, which is generated by training through historical first return signals received during the measurement process using unmanned survey vessels with quality marked as qualified or not.

[0093] In one specific embodiment, the measurement data acquisition module 102 in the system is further configured to: acquire reservoir image information in real time along the direction of the second measurement path during the measurement process of reservoir topographic data using a drone equipped with an airborne LiDAR along a second measurement path; determine the measurement terrain classification in the reservoir image information based on image analysis technology; the measurement terrain classification includes: water body, vegetation, bare soil, and buildings; match preset pulse parameters according to the determined measurement terrain classification; each measurement terrain classification has preset matching pulse parameters, including: pulse frequency and intensity; during the measurement of reservoir topographic data using a drone equipped with an airborne LiDAR along the second measurement path, select the matching pulse parameters for pulse transmission and receive the corresponding return signal.

[0094] Furthermore: the second return signal received during the process of obtaining reservoir water surface and shoreline topographic data using unmanned survey vessels is filtered and amplified; the filtered and amplified second return signal and the meteorological data corresponding to the time of the received second return signal are input into the second quality assessment model, and the assessment result of whether the signal quality is qualified or not is output, and the second return signal with the assessment result of unqualified signal quality is removed; the second quality assessment model adopts a neural network model, which is trained and generated by using historical second return signals received during the measurement process using unmanned survey vessels with quality marked as qualified or not, historical reservoir groundwater quality data and historical meteorological data.

[0095] In one specific embodiment, the reservoir capacity estimation module 104 in the system is further used to input the reservoir's environmental data, preprocessed underwater topographic data, and above-water and shore-side topographic data into a multimodal data fusion model, and to perform data fusion of the underwater topographic data and above-water and shore-side topographic data using the multimodal data fusion model; wherein, the multimodal data fusion model adopts a neural network model, and is generated by training the underwater topographic data and above-water and shore-side topographic data corresponding to the reservoir capacity data with the smallest difference between historical and actual reservoir capacity data, historical underwater topographic data, historical above-water and shore-side topographic data, and historical environmental data.

[0096] In one specific embodiment, the measurement data acquisition module 102 in the system is further used to divide the reservoir into important areas and non-important areas based on the reservoir structure information and reservoir geographic information; for the important areas of the reservoir, multiple unmanned survey vessels are used to record the underwater topographic data of the reservoir completed by each unmanned survey vessel according to the first measurement path in different time periods; the reservoir capacity estimation module 104 in the system is further used to select the highest quality underwater topographic data of the reservoir from multiple sets of underwater topographic data corresponding to the same location in the first measurement path as the underwater topographic data of the reservoir at the current location.

[0097] This application also discloses a computer-readable storage medium.

[0098] Specifically, the computer-readable storage medium stores a computer program that can be loaded by a processor and executed, such as the reservoir capacity estimation method based on the unmanned survey vessel and airborne LiDAR described above. The computer-readable storage medium includes, for example, various media that can store program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0099] This application also discloses a computer device.

[0100] Specifically, the computer device includes a memory and a processor. The memory stores a computer program that can be loaded and executed by the processor to estimate the reservoir capacity based on the unmanned survey vessel and airborne LiDAR.

[0101] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for estimating reservoir capacity based on unmanned survey vessels and airborne LiDAR, characterized in that, include: Based on the reservoir's structural information and geographical information, a first measurement path and a first measurement speed, and a second measurement path and a second measurement speed are planned, so that the distance between the corresponding positions of the first measurement path and the second measurement path at the same time is greater than a preset distance. The underwater topographic data of the reservoir is measured using an unmanned survey vessel following a first measurement path; the first measurement time for measuring the underwater topographic data of the reservoir is calculated according to a first measurement speed; simultaneously, the topographic data of the reservoir surface and shoreline is measured using an unmanned aerial vehicle equipped with an airborne LiDAR following a second measurement path; the second measurement time for measuring the topographic data of the reservoir surface and shoreline is calculated according to a second measurement speed; and the interval between the first measurement time and the second measurement time is calculated. Based on reservoir construction information, reservoir geographic information and interval duration, a third measurement path and a third measurement speed are planned so that the distance between the corresponding positions of the third measurement path and the first measurement path at the same time is greater than the preset distance. Within the interval duration, a UAV equipped with airborne LiDAR is used to complete the measurement of the reservoir's water surface and shore topography data again according to the third measurement path. The underwater topographic data of the reservoir and the surface and shore topographic data of the reservoir obtained by measurement are preprocessed. The preprocessing includes: performing similarity matching on the surface and shore topographic data of the reservoir at the same location obtained by two measurements, retaining one data point from a set of surface and shore topographic data with a similarity greater than a preset similarity, and selecting the surface and shore topographic data with higher data quality from the set of surface and shore topographic data with a self-similarity not greater than the preset similarity; fusing the preprocessed underwater topographic data and the surface and shore topographic data of the reservoir, constructing a three-dimensional model of the reservoir using the fused data, and estimating the reservoir capacity based on the volume of the three-dimensional model.

2. The method for estimating reservoir capacity based on an unmanned survey vessel and airborne LiDAR as described in claim 1, characterized in that, Also includes: During the measurement by the unmanned survey vessel, water quality sensors and meteorological sensors installed on the unmanned survey vessel are used to collect real-time groundwater quality data and meteorological data of the reservoir. The preprocessing of the measured underwater topographic data of the reservoir with the above-water and shore topographic data of the reservoir also includes: The first return signal received during the process of obtaining underwater topographic data of a reservoir using an unmanned survey vessel is filtered and amplified. The filtered and amplified first return signal, reservoir groundwater quality data, and meteorological data are input into the first quality assessment model, which outputs the assessment result of whether the first return signal is qualified or not, and discards the first return signal with the assessment result of unqualified quality; the first quality assessment model adopts a neural network model, which is trained and generated by using the first return signal received during the measurement process of the unmanned measuring vessel with the quality marked as qualified or not. The second return signal received during the process of obtaining reservoir water surface and shore topographic data using unmanned survey vessels is filtered and amplified. The filtered and amplified second return signal and the meteorological data corresponding to the time of the received second return signal are input into a second quality assessment model, which outputs the assessment result of whether the signal quality is qualified or not, and the second return signal with the assessment result of unqualified signal quality is removed. The second quality assessment model adopts a neural network model, which is trained and generated by using historical second return signals received during the measurement process using unmanned survey vessels with quality marked as qualified or not, historical reservoir groundwater quality data and historical meteorological data.

3. The method for estimating reservoir capacity based on an unmanned survey vessel and airborne LiDAR as described in claim 1, characterized in that, The method of using a drone equipped with airborne LiDAR to complete the measurement of reservoir topographic data on the water surface and shoreline according to the second measurement path also includes: This method addresses the process of using an airborne LiDAR-equipped UAV to measure the topographic data of a reservoir's surface and shoreline along a second measurement path. Real-time image acquisition of the reservoir along the path's forward direction is performed. Image analysis techniques are used to determine the topographic classification within the reservoir images. This classification includes water bodies, vegetation, bare soil, and buildings. Preset pulse parameters are matched to each determined topographic classification. Each topographic classification has preset pulse parameters, including pulse frequency and intensity. During the measurement of the reservoir's surface and shoreline topographic data using the airborne LiDAR-equipped UAV along the second measurement path, the matched pulse parameters are selected for pulse transmission, and the corresponding return signals are received.

4. The method for estimating reservoir capacity based on an unmanned survey vessel and airborne LiDAR as described in claim 1, characterized in that, Also includes: Collect reservoir environmental data; the reservoir environmental data includes water quality data and meteorological data; The process of fusing preprocessed underwater topographic data with surface and shore topographic data includes: inputting reservoir environmental data, preprocessed underwater topographic data, and surface and shore topographic data into a multimodal data fusion model; and using the multimodal data fusion model to fuse the underwater topographic data with the surface and shore topographic data. The multimodal data fusion model employs a neural network model, which is generated by training on the reservoir capacity data corresponding to the calculated underwater topographic data and surface and shore topographic data corresponding to the reservoir capacity data with the smallest discrepancy between historical and actual reservoir capacity data, as well as historical underwater topographic data, historical surface and shore topographic data, and historical reservoir environmental data.

5. The method for estimating reservoir capacity based on an unmanned survey vessel and airborne LiDAR as described in claim 1, characterized in that, Also includes: The reservoir is divided into important and non-important areas based on its structural and geographical information. For key areas of the reservoir, multiple unmanned survey vessels are used to record underwater topographic data of the reservoir in different time periods. Each unmanned survey vessel completes the measurement of the underwater topographic data of the reservoir according to the first measurement path. The preprocessing of the underwater topographic data of the reservoir and the topographic data of the reservoir surface and shoreline also includes: selecting the highest quality underwater topographic data of the reservoir from multiple sets of underwater topographic data corresponding to the same location in the first measurement path as the underwater topographic data of the reservoir at the current location.

6. The method for estimating reservoir capacity based on an unmanned survey vessel and airborne LiDAR as described in claim 1, characterized in that, Also includes: The reservoir capacity is estimated by calculating the volume of the 3D model of the reservoir using numerical integration or volume rendering.

7. The method for estimating reservoir capacity based on an unmanned survey vessel and airborne LiDAR as described in claim 1, characterized in that, It also includes: during the process of using an unmanned survey vessel to complete the measurement of underwater topographic data of the reservoir according to the first measurement path, the Kongsberg EM 2040 or Teledyne Reson T20-P multibeam sonar device is used to complete the measurement; during the process of using an unmanned aerial vehicle with an airborne LiDAR to complete the measurement of topographic data of the reservoir above water and on the shore according to the second measurement path, the Riegl VUX-1HA LiDAR sensor is used to complete the measurement.

8. A reservoir capacity estimation system based on an unmanned survey vessel and airborne LiDAR, characterized in that, include: The measurement parameter planning module is used to plan a first measurement path and a first measurement speed, and a second measurement path and a second measurement speed based on the reservoir structure information and reservoir geographic information, so that the distance between the corresponding positions of the first measurement path and the second measurement path at the same time is greater than a preset distance. The measurement data acquisition module is used to complete the measurement of underwater topographic data of the reservoir using an unmanned survey vessel according to a first measurement path; calculate the first measurement time for completing the underwater topographic data measurement of the reservoir according to a first measurement speed; simultaneously, use an unmanned aerial vehicle equipped with an airborne LiDAR to complete the measurement of the topographic data of the reservoir surface and shore according to a second measurement path; calculate the second measurement time for measuring the topographic data of the reservoir surface and shore according to a second measurement speed; and calculate the interval between the first measurement time and the second measurement time. The redundant measurement data acquisition module is used to plan a third measurement path and a third measurement speed based on reservoir building information, reservoir geographic information and interval duration, so that the distance between the corresponding positions of the third measurement path and the first measurement path at the same time is greater than a preset distance, and within the interval duration, the UAV with airborne LiDAR is used to complete the measurement of the reservoir water and shore topography data again according to the third measurement path. The reservoir capacity estimation module is used to preprocess the underwater topographic data and the surface and shore topographic data of the reservoir obtained by measurement. The preprocessing includes: performing similarity matching on the surface and shore topographic data of the reservoir obtained from two measurements at the same location; retaining one data point from a set of surface and shore topographic data with a similarity greater than a preset similarity; selecting the surface and shore topographic data with higher data quality from the set of surface and shore topographic data with a self-similarity not greater than the preset similarity; fusing the preprocessed underwater topographic data and the surface and shore topographic data; constructing a three-dimensional model of the reservoir using the fused data; and estimating the reservoir capacity based on the volume of the three-dimensional model.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the method as described in any one of claims 1 to 7.

10. A computer device, characterized in that, The computer device includes a memory, a processor, and a program stored in and executable on the memory, the program being executed by the processor to implement the steps of the method as described in any one of claims 1 to 7.

Images