Short-term hydrological forecasting method and system for dam site section

Abstract

The invention discloses a short-term hydrological forecasting method and system for a dam site section, and relates to a data processing and hydrological forecasting technology, and the method comprises the steps: building a dam region terrain model in advance; video monitoring data of a historical dam area, rainfall data of a dam area coverage area and historical water level monitoring data of the dam area are obtained; searching a water level floating range corresponding to the water level monitoring data according to the historical water level monitoring data of the dam area; constructing a training data set, and training a short-term hydrological prediction model; performing prediction by utilizing the trained short-term hydrological prediction model according to the current video monitoring data and the water level monitoring data; and according to the predicted result, the current video monitoring data and the water level monitoring data, performing result simulation in a pre-established dam region terrain model. On one hand, the accuracy of short-term hydrological forecasting of the dam site section can be improved, and on the other hand, the auxiliary effect of short-term hydrological forecasting of the dam site section on users is improved.

Description

Technical Field

[0001] The present application relates to the fields of data processing and hydrological forecasting, and in particular to a short-term hydrological forecasting method and system for a dam site section. Background Art

[0002] Hydrological forecasting is of great significance in the fields of flood control, drought relief and rational use of water resources. Hydrological forecasting refers to the qualitative or quantitative prediction of the hydrological conditions of a water body, a region or a hydrological station in the future based on previous or current hydrological and meteorological data. In actual production and life, river basin hydrological forecasting can provide important decision-making support for river basin flood and drought disaster prevention, safe and economic operation of reservoir groups, scientific allocation of water resources and sustainable social development. At present, hydrological models are often used to achieve hydrological forecasting. The so-called hydrological model refers to an approximate scientific model that generalizes complex hydrological phenomena and processes using simulation methods.

[0003] On the one hand, the existing forecasting models have the problem of forecasting accuracy, which is affected by many uncertain factors, such as model input uncertainty and model structure uncertainty. On the other hand, the forecast results cannot be intuitively reflected, which reduces the auxiliary effect of the forecast results on users. Summary of the invention

[0004] The embodiment of the present application provides a method and system for short-term hydrological forecasting of a dam site section. By combining local monitoring video and water level data and constructing a dam area terrain model, on the one hand, the accuracy of short-term hydrological forecasting of the dam site section is improved, and on the other hand, the auxiliary effect of short-term hydrological forecasting of the dam site section on users is improved.

[0005] The present application embodiment provides a short-term hydrological forecasting method for a dam site section, comprising: Pre-establishing a dam area terrain model, and establishing a truncation surface according to the dam area terrain model based on the distribution of units behind the dam; Obtain historical video monitoring data of the dam area, rainfall data of the area covered by the dam area, and historical water level monitoring data of the dam area; Determining relevant video frames from the video monitoring data based on historical water level monitoring data of the dam area; Finding the water level floating range of the corresponding water level monitoring data in the relevant video frame; Using the water level monitoring data and the corresponding fluctuation range as a data group to construct a training data set; adding labels to the data groups in the training data set based on the rainfall data, so as to train a short-term hydrological prediction model using the labeled training data set; Perform forecasts using the trained short-term hydrological forecast model based on current video monitoring data and water level monitoring data; and, Based on the predicted results and the current video monitoring data and water level monitoring data, the results are simulated on the pre-established dam area terrain model and cross-section.

[0006] Optionally, establishing a truncation surface according to the dam area terrain model based on the distribution of units behind the dam includes: Obtain historical power generation data of the units behind the dam to determine the power generation utilization rate of each unit; The units with utilization rates higher than a preset threshold are determined at positions corresponding to the terrain model of the dam area, and a truncation surface is established.

[0007] Optionally, according to the historical water level monitoring data of the dam area, determining the relevant video frames from the video monitoring data includes: Determine the time stamp of any water level monitoring data; Determining, according to the time stamp, an associated video segment in the video monitoring data; From the associated video segments, several video frames are extracted at set intervals as related video frames.

[0008] Optionally, searching for a water level floating range of corresponding water level monitoring data in the relevant video frame includes: Select one of the extracted multiple video frames as a base frame; In the base frame, multiple reference points are selected, and a coordinate system is established according to one of the reference points; and, Identify the boundary of the dam and the water area from each video frame, and perform line enhancement on the boundary of the identified water area; According to the remaining reference points, aligning the dam body in the coordinate system for other video frames; Adjusting the transparency within the boundary range of the water area in each aligned video frame to present the enhanced water boundary of each line in any video frame; The pixel fluctuation range of the corresponding water level monitoring data is determined according to the maximum value of the water area boundary after adjusting the transparency.

[0009] Optionally, the pixel fluctuation range corresponding to the water level monitoring data is determined according to the maximum value of the water area boundary after adjusting the transparency, including: According to the aligned dam body, several horizontal reference lines are determined based on the enhanced water boundaries of the presented lines; A horizontal reference line with the most maximum values ​​including the enhanced water area boundary is selected as the corresponding pixel fluctuation interval.

[0010] Optionally, determining the water level floating range corresponding to the water level monitoring data according to the maximum value of the water area boundary after adjusting the transparency also includes: Determine in advance the mapping relationship between the water level and each horizontal reference line based on the water level scale included in the video monitoring data; The water level floating range of the water level monitoring data is determined according to the mapping relationship and the determined pixel fluctuation interval corresponding to the water level monitoring data.

[0011] Optionally, based on the predicted results and the current video monitoring data and water level monitoring data, the result simulation is performed on the pre-established dam area terrain model and truncation surface, including: Pre-mapping each horizontal reference line to the dam area terrain model and controlling each horizontal reference line to be hidden; Presenting current water level monitoring data on the dam area terrain model; According to the water level prediction data and fluctuation range contained in the prediction results, the water level prediction data in the prediction results are dynamically presented based on the current water level monitoring data, and the horizontal reference line closest to the predicted fluctuation range is highlighted.

[0012] Optionally, based on the predicted results and the current video monitoring data and water level monitoring data, the result simulation on the pre-established dam area terrain model and truncation surface also includes: Marking the section surface based on each horizontal reference line; and, The water level prediction data in the prediction results are dynamically presented based on the current water level monitoring data, in synchronization with the dam area terrain model.

[0013] An embodiment of the present application also proposes a short-term hydrological forecasting system for a dam site section, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the steps of the short-term hydrological forecasting method for the dam site section as described above are implemented.

[0014] The forecasting method of the embodiment of the present application combines local monitoring video and water level data, and constructs a dam area terrain model to improve the accuracy of short-term hydrological forecast of the dam site section on the one hand, and on the other hand, improve the auxiliary effect of short-term hydrological forecast of the dam site section on the user.

[0015] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings: Figure 1 The following is a schematic diagram of the basic process of the short-term hydrological forecasting method for the dam site section of this embodiment. DETAILED DESCRIPTION

[0017] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0018] The present application embodiment proposes a short-term hydrological forecasting method for a dam site section, such as Figure 1 As shown, the following steps are included: In step S101, a dam area terrain model is pre-established, and a truncation surface is established according to the dam area terrain model based on the distribution of units behind the dam. In some examples, only the dam area terrain model and the truncation surface view can be presented in the same interface.

[0019] In step S102, the video monitoring data of the historical dam area, the rainfall data of the dam area coverage area, and the historical water level monitoring data of the dam area are obtained. In some embodiments, the power generation data of the unit behind the dam can also be obtained, and the power generation data of the unit and the rainfall data of the dam area coverage area can be used to add data tags.

[0020] In step S103, the relevant video frames are determined from the video monitoring data according to the historical water level monitoring data of the dam area. In some embodiments, the historical water level monitoring data of the dam area is discrete data and does not have continuity, while the video monitoring data has a certain continuity. The embodiment of the present application further determines the video frames of the corresponding time period from the video monitoring data based on the water level monitoring data, thereby judging the water level fluctuation of the corresponding time period.

[0021] In step S104, the water level floating range corresponding to the water level monitoring data is searched in the relevant video frames.

[0022] In step S105, a training data set is constructed based on the water level monitoring data and the corresponding fluctuation range as a data group.

[0023] In step S106, labels are added to the data groups in the training data set based on the rainfall data, so as to train a short-term hydrological prediction model using the labeled training data set. In some embodiments, the short-term hydrological prediction model can be a network model based on the Encoder+Decoder+LSTM architecture.

[0024] In step S107, prediction is performed using the trained short-term hydrological prediction model according to the current video monitoring data and water level monitoring data.

[0025] In step S108, the result simulation is performed on the pre-established dam area terrain model and cross-section according to the predicted result and the current video monitoring data and water level monitoring data.

[0026] The forecasting method of the embodiment of the present application combines local monitoring video and water level data, and constructs a dam area terrain model to improve the accuracy of short-term hydrological forecast of the dam site section on the one hand, and on the other hand, improve the auxiliary effect of short-term hydrological forecast of the dam site section on the user.

[0027] In some embodiments, establishing a truncation surface according to the dam area terrain model based on the distribution of units behind the dam includes: Obtain historical power generation data of the units behind the dam to determine the power generation utilization rate of each unit; The units with utilization rates higher than a preset threshold are determined at positions corresponding to the terrain model of the dam area, and a truncation surface is established. For example, in some examples, the water inlet area of ​​the unit in front of the dam site section is taken as a key area of ​​concern, and the units with high power generation utilization rates are determined based on the power generation utilization rates of each unit in the historical data, and the truncation surface is determined.

[0028] In some embodiments, according to the historical water level monitoring data of the dam area, determining the relevant video frames from the video monitoring data includes: Determine the time stamp of any water level monitoring data; According to the time stamp, the associated video segments in the video monitoring data are determined. For example, according to the generation time of any water level monitoring data, the video monitoring data is queried to obtain the associated video segments at the generation time. In some examples, the associated video segments may be video segments of a certain length before and after the generation time.

[0029] From the associated video segments, several video frames are extracted at set intervals as related video frames.

[0030] In some embodiments, searching for the water level floating range of the corresponding water level monitoring data in the relevant video frame includes: One frame is selected from the extracted multiple video frames as a base frame, and the base frame may be a frame randomly selected from the multiple video frames.

[0031] In the base frame, multiple reference points are selected, and a coordinate system is established based on one of the reference points. In some embodiments, the reference point may be a fixed mark or object on the dam or river bank, so that the dam in multiple video frames can be aligned based on the reference point.

[0032] The boundary of the dam body and the water area is identified from each video frame, and the line of the identified water area boundary is enhanced. In some embodiments, the boundary of the dam body and the water area can be determined based on an edge detection algorithm, and the line enhancement of the identified water area boundary can be performed by deepening the color value of the boundary pixel.

[0033] Based on the remaining reference points, the dam body is aligned in the coordinate system for other video frames.

[0034] The transparency within the boundary range of the water area in each aligned video frame is adjusted to present the enhanced water boundary of each line in any video frame. That is, after adjusting the transparency, the boundary of the water area in multiple video frames can be highlighted based on the dam body, and the boundary is also clearly retained after changing the transparency after the aforementioned line color value enhancement.

[0035] The pixel fluctuation range of the corresponding water level monitoring data is determined according to the maximum value of the water area boundary after adjusting the transparency.

[0036] In some embodiments, determining the pixel fluctuation interval corresponding to the water level monitoring data according to the maximum value of the water area boundary after adjusting the transparency includes: According to the aligned dam body, several horizontal reference lines are determined based on the enhanced water boundaries of the presented lines. In some embodiments, several horizontal reference lines can be determined based on the water level scale behind the dam. For example, several corresponding horizontal reference lines are established on the dam body image and the dam area terrain model at intervals of 3-5 cm. Specifically, several horizontal reference lines determined in the dam body image can be mapped based on the monitored pixel distance.

[0037] The horizontal reference line with the most maximum values ​​that contains the enhanced water area boundary is selected as the corresponding pixel fluctuation interval. In some examples, for the mixed image after changing the transparency within the boundary range of the water area, the horizontal reference line with the most maximum value points falling on the horizontal reference line is considered to be the maximum value fluctuation interval of the corresponding time range. In this way, the local maximum value noise is also filtered out, thereby improving the recognition accuracy.

[0038] In some embodiments, determining the water level floating range corresponding to the water level monitoring data according to the maximum value of the water area boundary after adjusting the transparency also includes: Determine in advance the mapping relationship between the water level and each horizontal reference line based on the water level scale included in the video monitoring data; The water level floating range of the water level monitoring data is determined according to the mapping relationship and the pixel fluctuation interval of the corresponding water level monitoring data. That is, in this example, the relationship with the horizontal reference line is determined according to the position of the water level scale in the video monitoring, so that the fluctuation range of the water level monitoring data can be obtained.

[0039] In some embodiments, based on the predicted results and the current video monitoring data and water level monitoring data, the result simulation is performed on the pre-established dam area terrain model and the truncation surface, including: Each horizontal reference line is mapped to the dam area terrain model in advance and each horizontal reference line is controlled to be hidden. In this example, each determined horizontal reference line is further mapped to the dam area terrain model, and a clean dam area terrain model can be provided by controlling hiding.

[0040] Current water level monitoring data is presented on the dam area terrain model.

[0041] According to the water level forecast data and fluctuation range contained in the forecast results, the water level forecast data in the forecast results are dynamically presented based on the current water level monitoring data, and the horizontal reference line closest to the forecast fluctuation range is highlighted. In this way, the water level forecast data and fluctuation range can be highlighted intuitively on the dam area terrain model, thereby improving the auxiliary effect for the staff.

[0042] In some embodiments, based on the predicted results and the current video monitoring data and water level monitoring data, the result simulation on the pre-established dam area terrain model and the truncation surface further includes: Marking the section surface based on each horizontal reference line; and, The water level prediction data in the prediction results are dynamically presented based on the current water level monitoring data in synchronization with the dam area terrain model. The data display efficiency can be further improved by truncating the surface, thereby assisting in improving the hydrological regulation effect of the dam area.

[0043] An embodiment of the present application also proposes a short-term hydrological forecasting system for a dam site section, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the steps of the short-term hydrological forecasting method for the dam site section as described above are implemented.

[0044] In addition, although exemplary embodiments have been described herein, the scope includes any and all embodiments based on the present disclosure with equivalent elements, modifications, omissions, combinations (e.g., various embodiments intersecting schemes), adaptations or changes. It is not limited to the examples described in this specification or during the implementation of this application, and its examples will be interpreted as non-exclusive.

[0045] The above description is intended to be illustrative rather than restrictive. For example, the above examples (or one or more of them) can be used in combination with each other. For example, those of ordinary skill in the art may use other embodiments when reading the above description.

[0046] The above embodiments are merely exemplary embodiments of the present disclosure. Those skilled in the art may make various modifications or equivalent substitutions to the present invention within the essence and protection scope of the present disclosure, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present invention.

Claims

1. A short-term hydrological forecasting method for a dam site section, characterized in that: include: Pre-establishing a dam area terrain model, and establishing a truncation surface according to the dam area terrain model based on the distribution of units behind the dam; Obtain historical video monitoring data of the dam area, rainfall data of the area covered by the dam area, and historical water level monitoring data of the dam area; Determining relevant video frames from the video monitoring data based on historical water level monitoring data of the dam area; Finding the water level floating range of the corresponding water level monitoring data in the relevant video frame; Using the water level monitoring data and the corresponding fluctuation range as a data group to construct a training data set; adding labels to the data groups in the training data set based on the rainfall data, so as to train a short-term hydrological prediction model using the labeled training data set; Perform forecasts using the trained short-term hydrological forecast model based on current video monitoring data and water level monitoring data; as well as, Based on the predicted results and the current video monitoring data and water level monitoring data, the results are simulated on the pre-established dam area terrain model and cross-section.

2. The short-term hydrological forecasting method for a dam site section as claimed in claim 1, characterized in that: Establishing the truncation surface based on the distribution of units behind the dam and the terrain model of the dam area includes: Obtain historical power generation data of the units behind the dam to determine the power generation utilization rate of each unit; The units with utilization rates higher than a preset threshold are determined at positions corresponding to the terrain model of the dam area, and a truncation surface is established.

3. The short-term hydrological forecasting method for a dam site section according to claim 1, characterized in that: According to the historical water level monitoring data of the dam area, the relevant video frames are determined from the video monitoring data including: Determine the time stamp of any water level monitoring data; Determining, according to the time stamp, an associated video segment in the video monitoring data; From the associated video segments, several video frames are extracted at set intervals as related video frames.

4. The short-term hydrological forecasting method for a dam site section as claimed in claim 3, characterized in that: Finding the water level floating range of the corresponding water level monitoring data in the relevant video frame includes: Select one of the extracted multiple video frames as a base frame; In the base frame, multiple reference points are selected, and a coordinate system is established according to one of the reference points; and, Identify the boundary of the dam and the water area from each video frame, and perform line enhancement on the boundary of the identified water area; According to the remaining reference points, aligning the dam body in the coordinate system for other video frames; Adjusting the transparency within the boundary range of the water area in each aligned video frame to present the enhanced water boundary of each line in any video frame; The pixel fluctuation range of the corresponding water level monitoring data is determined according to the maximum value of the water area boundary after adjusting the transparency.

5. The short-term hydrological forecasting method for a dam site section as claimed in claim 4, characterized in that: The pixel fluctuation range of the corresponding water level monitoring data is determined according to the maximum value of the water area boundary after adjusting the transparency, including: According to the aligned dam body, several horizontal reference lines are determined based on the enhanced water boundaries of the presented lines; A horizontal reference line with the most maximum values ​​including the enhanced water area boundary is selected as the corresponding pixel fluctuation interval.

6. The short-term hydrological forecasting method for a dam site section as claimed in claim 5, characterized in that: The water level floating range corresponding to the water level monitoring data determined according to the maximum value of the water area boundary after adjusting the transparency also includes: Determine in advance the mapping relationship between the water level and each horizontal reference line based on the water level scale included in the video monitoring data; The water level floating range of the water level monitoring data is determined according to the mapping relationship and the determined pixel fluctuation interval corresponding to the water level monitoring data.

7. The short-term hydrological forecasting method for a dam site section as claimed in claim 6, characterized in that: Based on the predicted results and the current video monitoring data and water level monitoring data, the results are simulated on the pre-established dam area terrain model and cross-section, including: Pre-mapping each horizontal reference line to the dam area terrain model and controlling each horizontal reference line to be hidden; Presenting current water level monitoring data on the dam area terrain model; According to the water level prediction data and fluctuation range contained in the prediction results, the water level prediction data in the prediction results are dynamically presented based on the current water level monitoring data, and the horizontal reference line closest to the predicted fluctuation range is highlighted.

8. The short-term hydrological forecasting method for a dam site section as claimed in claim 7, characterized in that: Based on the predicted results and the current video monitoring data and water level monitoring data, the result simulation is carried out on the pre-established dam area terrain model and cross-section, including: Marking the section surface based on each horizontal reference line; and, The water level prediction data in the prediction results are dynamically presented based on the current water level monitoring data, in synchronization with the dam area terrain model.

9. A short-term hydrological forecasting system for a dam site section, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the short-term hydrological forecasting method for a dam site section as described in any one of claims 1 to 8 are implemented.

Images