Hydropower station safe operation prediction method and system taking environmental disasters into consideration

Abstract

Provided are a hydropower station safe operation prediction method and system taking environmental disasters into consideration. The method comprises: in view of terrain data, delimiting a terrain range affecting the safe operation of a current hydropower station, and determining a plurality of disaster risk sub-regions within the terrain range (S101); on the basis of historical weather data, determining a key risk time interval of the current hydropower station (S102); within the key risk time interval, collecting suspected disaster data of each disaster risk sub-region, and acquiring weather information within a specified future time period (S103); extracting suspected disaster feature information of the suspected disaster data of each disaster risk sub-region, and extracting weather feature information of the weather information (S104); according to a set order, combining the extracted suspected disaster feature information and weather feature information (S105); and inputting a feature sequence into a pre-trained prediction model, so as to obtain a prediction result of the safe operation of the hydropower station (S106).

Description

A method and system for predicting safe operation of a hydropower station considering environmental disasters Technical Field

[0001] The present application relates to the technical field of safe operation management of hydropower stations, and in particular to a method and system for predicting safe operation of hydropower stations taking environmental disasters into consideration. Background Art

[0002] Hydropower stations are important clean energy sources, and their safe and stable operation is crucial. Unlike thermal power plants, their safe and stable operation is easily affected by natural factors. For example, natural disasters such as heavy rain, landslides, and mudslides can cause the accumulation of silt, tree trunks, and other debris within the watershed. Although hydropower stations are generally equipped with trash racks at their entrances, the accumulation of debris there can also pose a safety hazard. Furthermore, rising silt levels in the waterway can breach the trash racks, and using silted water for power generation can cause stability issues such as excessive blade vibration.

[0003] Existing landslide monitoring systems, such as those based on Beidou satellite monitoring, remote sensing information, and lidar, use satellite data, remote sensing images, or radar signals to identify deformation in the target area. However, these methods are more suitable for large single landslides due to high costs, huge data processing volumes, and the influence of clouds and atmosphere. They cannot be applied to the safe operation prediction of hydropower stations. Human judgment requires continuous human monitoring and experienced operators, and unmanned operation is impossible in extreme weather. Summary of the Invention

[0004] The embodiments of the present application provide a method and system for predicting the safe operation of a hydropower station taking environmental disasters into consideration. A method for determining the flow rate and sediment content of the current basin in which the hydropower station operates is proposed, thereby enabling prediction of the safe operation of the hydropower station based on the flow rate and sediment content of the basin.

[0005] The present application embodiment proposes a method for predicting safe operation of a hydropower station taking into account environmental disasters, including:

[0006] Combined with topographic data, delineate the terrain range that affects the safe operation of the current hydropower station and identify multiple disaster risk sub-areas within the terrain range; and

[0007] Determine the key risk time intervals for the current hydropower station based on historical weather data;

[0008] During the key risk time interval, multiple disaster risk sub-areas are monitored to collect suspected disaster data for each disaster risk sub-area and obtain weather information for a specified future time period;

[0009] Extracting suspected disaster feature information of suspected disaster data of each disaster risk sub-region, and extracting weather feature information of weather information;

[0010] Combining the extracted suspected disaster feature information and weather feature information in a set order to obtain a feature sequence;

[0011] The feature sequence is input into a pre-trained prediction model to obtain a prediction result for the safe operation of the hydropower station, wherein the prediction model is used to predict the flow and sediment content of the current hydropower station basin in the future period.

[0012] Optionally, combined with terrain data, the terrain range that affects the safe operation of the current hydropower station can be delineated, including:

[0013] constructing a terrain model of the delineated terrain range based on the terrain data, and determining a basic watershed of the current hydropower station from the terrain data;

[0014] Taking the basic watershed as a reference, exchanging the height difference of the terrain data to determine an inverse model of the terrain model;

[0015] Determining, based on the inverse model, a plurality of suspected disaster basins connected to the basic basin, constructing a tree-like distribution map based on the plurality of suspected disaster basins, and determining a plurality of disaster risk sub-areas based on the tree-like distribution map;

[0016] According to the node relationship of the tree-like distribution diagram and the terrain model, a combination order of feature sequences is configured for each node.

[0017] Optionally, taking the basic watershed as a reference, exchanging the height difference of the terrain data to determine the inverse model of the terrain model includes:

[0018] Obtaining altitude information of the basic watershed;

[0019] Taking the low altitude interval in the altitude information as a reference benchmark, and subtracting the altitude information of the basic watershed from the reference benchmark to determine the simulated watershed trend in the inverse model; and

[0020] For the terrain data in the terrain data, which has an angle with the basic watershed plane greater than a preset angle threshold on any side of the basic watershed, the actual watershed-topography relationship is retained and no exchange processing is performed to determine the inverse model of the terrain model.

[0021] Optionally, taking the basic watershed as a reference, exchanging the height difference of the terrain data to determine the inverse model of the terrain model further includes:

[0022] Using the reference benchmark, exchanging the height difference of the terrain data, wherein the lower the altitude of the basin in the basic basin is, the higher the altitude of the basin in the inverse model is after the exchange;

[0023] constructing the inverse model based on the replaced basic watershed and the retained watershed-topographic relationship;

[0024] According to the inverse model, multiple suspected disaster basins connected to the basic basin are determined to include:

[0025] In a simulation environment, performing flow simulation based on the inverse model to obtain a reverse flow simulation result;

[0026] According to the flow simulation results, tributaries with flow rates greater than a specified flow threshold are selected as suspected disaster basins.

[0027] Optionally, according to the node relationship of the tree-like distribution diagram and the terrain model, the combination order of configuring the feature sequence for each node includes:

[0028] According to the flow simulation results and the positional relationship between each node and the current hydropower station in the tree-like distribution diagram, a combination order of feature sequences is configured for each node, wherein the larger the value of the flow simulation result of any tributary node position and the closer the node position is to the current hydropower station, the higher the order.

[0029] Optionally, suspected disaster data collected for each disaster risk sub-area includes:

[0030] For each suspected disaster basin, data is collected during the same period, where the data collection period for each suspected disaster basin is the same; and

[0031] For any suspected disaster basin, data are collected during the corresponding collection period according to the flow direction of the tributaries of the suspected disaster basin and the time sequence of the collection period to obtain the suspected disaster data of each disaster risk sub-area.

[0032] Optionally, the prediction model is pre-trained based on historical weather data and disaster data of the main stream and tributaries of each river basin, and is configured with corresponding global weights for any geological condition so as to adapt to the current hydropower station basin environment based on the global weights.

[0033] An embodiment of the present application also proposes a hydropower station safe operation prediction system taking into account environmental disasters. The system includes a processor and a memory, and a computer program is stored on the memory. When the computer program is executed by the processor, it implements the steps of the hydropower station safe operation prediction method taking into account environmental disasters as described above.

[0034] The method of the present application proposes a method for determining the flow and sediment content of the current basin, thereby realizing the prediction of the safe operation of the hydropower station based on the flow and sediment content of the basin.

[0035] 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

[0036] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0037] FIG1 is a schematic diagram of the basic process of the method for predicting safe operation of a hydropower station according to this embodiment. DETAILED DESCRIPTION

[0038] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although 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. Rather, 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.

[0039] This embodiment of the present application proposes a method for predicting safe operation of a hydropower station taking environmental disasters into consideration, as shown in FIG1 , including the following steps:

[0040] In step S101, the terrain data is combined to delineate the terrain range that could affect the safe operation of the current hydropower station, and multiple disaster risk sub-regions within the terrain range are identified. In some specific examples, the terrain data can be used to identify the range that could potentially affect the watershed, while other ranges are not processed to avoid introducing unnecessary noise and thus reduce the amount of subsequent data processing.

[0041] In some embodiments, in combination with terrain data, delineating the terrain range that affects the safe operation of the current hydropower station includes:

[0042] Based on the terrain data, a terrain model of the defined terrain range is constructed, and the basic watershed of the current hydropower station is determined from the terrain data. In some specific examples, the determined basic watershed may include at least the main stream and may also include, for example, tributaries. In the embodiments of the present application, not only tributaries are used as suspected disaster watersheds, but the suspected disaster watershed is determined using the suspected disaster watershed determination method designed in the present application. For details, please refer to the subsequent embodiments.

[0043] Taking the basic watershed as a reference, exchanging the height difference of the terrain data to determine an inverse model of the terrain model;

[0044] According to the inverse model, multiple suspected disaster basins connected to the basic basin are determined, and a tree-like distribution map is constructed based on the multiple suspected disaster basins, so as to determine multiple disaster risk sub-areas based on the tree-like distribution map. In the embodiment of the present application, based on the inverse model, multiple suspected disaster basins connected to the basic basin are determined. Compared with human experience judgment, the use of the inverse model can identify more suspected disaster basins that are directly or indirectly connected to the current basic basin. The suspected disaster basins in this application can be the current tributaries or valley areas. There is no river water or less river water in normal weather, but in extreme weather, such as heavy rain, the basins that form tributaries can be identified based on the inverse model of the embodiment of the present application.

[0045] Based on the node relationships of the tree-like distribution diagram and the terrain model, a combination order of feature sequences is configured for each node. The tree-like distribution diagram constructed in the embodiment of the present application also includes tributary node location information and tributary scale information of each suspected disaster basin. Feature sequences can be established based on this information in subsequent identification to perform identification.

[0046] In step S102, the key risk time interval for the current hydropower station is determined based on historical weather data. In some examples, the key risk time interval can be determined based on the weather data at the hydropower station's location. For example, the rainy season in southern China falls in July and August, and these periods are the key risk time intervals. In some examples, the key time intervals can also be updated based on weather forecast information for the next week or two weeks.

[0047] In step S103, during the key risk time interval, multiple disaster risk sub-areas are monitored to collect suspected disaster data of each disaster risk sub-area and obtain weather information within a specified future time period.

[0048] In step S104, suspected disaster feature information of suspected disaster data of each disaster risk sub-region is extracted, and weather feature information of weather information is extracted.

[0049] In step S105 , the extracted suspected disaster feature information and weather feature information are combined in a set order to obtain a feature sequence.

[0050] In step S106, the feature sequence is input into a pre-trained prediction model to obtain a prediction result for the safe operation of the hydropower station, wherein the prediction model is used to predict the flow and sediment content of the current hydropower station basin in the future period. In some examples, the extracted weather feature information can be input into the prediction model with the suspected disaster feature information in a fixed order combination. In some embodiments, the pre-trained prediction model is pre-trained based on historical weather data and disaster data of the main stream and tributaries of each basin, and is configured with corresponding global weights for any geological conditions to adapt to the current hydropower station basin environment based on the global weights. In a specific implementation, a sub-model can be trained based on historical data (adding corresponding labels) and flow simulation (to expand the sample). For each suspected disaster basin in the current hydropower station basin, a sub-model weight can be configured in combination with the geological conditions and the specifications of the suspected disaster basin, so that the current hydropower station basin contains multiple adapted sub-models to constitute the prediction model of the current hydropower station basin, thereby predicting the sediment content in the basin in the future period through the prediction model. In specific implementation, the risk level of safe operation of a hydropower station can be graded based on the sediment content, so that the risk level of safe operation of the current hydropower station can be determined according to the prediction results.

[0051] The method of the present application proposes a method for determining the flow and sediment content of the current basin, thereby realizing the prediction of the safe operation of the hydropower station based on the flow and sediment content of the basin.

[0052] In some embodiments, taking the basic watershed as a reference, exchanging the height difference of the terrain data to determine the inverse model of the terrain model includes:

[0053] The altitude information of the basic watershed is obtained, for example, the altitude information of the riverbed of the basic watershed may be obtained.

[0054] Taking the low altitude interval in the altitude information as a reference benchmark, and subtracting the altitude information of the basic watershed from the reference benchmark to determine the simulated watershed trend in the inverse model; and

[0055] For the terrain data in the terrain data, any side of the basic watershed, and the terrain data with an angle greater than a preset angle threshold with the basic watershed plane, the actual watershed terrain relationship is retained, and no exchange processing is performed to determine the inverse model of the terrain model. In some specific examples, the inverse model can be constructed based on 3D terrain data, or it can be implemented in combination with 3D modeling software and fluid simulation software. In some specific examples, for example, for parts with mountains on both sides of the watershed, similar parts are not easy to introduce sediment in extreme weather during actual operation. In the embodiment of the present application, the actual watershed terrain relationship of this part is retained, and the corresponding terrain relationship is also retained for other similar parts, so that the suspected disaster watershed that may introduce sediment is included in the inverse model.

[0056] In some embodiments, the step of replacing the height difference of the terrain data with the basic watershed as a reference to determine the inverse model of the terrain model further includes:

[0057] Using the reference benchmark, exchanging the height difference of the terrain data, wherein the lower the altitude of the basin in the basic basin is, the higher the altitude of the basin in the inverse model is after the exchange;

[0058] constructing the inverse model based on the replaced basic watershed and the retained watershed-topographic relationship;

[0059] According to the inverse model, multiple suspected disaster basins connected to the basic basin are determined to include:

[0060] In a simulation environment, performing flow simulation based on the inverse model to obtain a reverse flow simulation result;

[0061] Based on the flow simulation results, tributaries with flow rates greater than a specified flow threshold are selected as suspected disaster basins. In some specific examples, flow simulation is performed based on the inverse model, allowing flow to be reversed to possible water development areas based on the current hydropower station basin. Correspondingly, during the rainy season and extreme weather conditions, these water development areas are possible water sources and flow into the current hydropower station basin. This design can identify more suspected disaster-prone tributaries.

[0062] In some embodiments, according to the node relationship of the tree-like distribution diagram and the terrain model, the combination order of configuring the feature sequence for each node includes:

[0063] According to the flow simulation results and the positional relationship between each node and the current hydropower station in the tree-like distribution diagram, a combination order of feature sequences is configured for each node, wherein the larger the value of the flow simulation result of any tributary node position and the closer the node position is to the current hydropower station, the higher the order.

[0064] In some specific examples, the combination order of feature sequences is configured for each node, so that the feature sequences can be identified based on the sub-models and their configured weights. In some specific examples, the combination order is considered. After configuring the weights, the prediction model as a whole can tend to the tributary nodes with a higher combination order, so that the prediction model focuses on key tributaries.

[0065] In some embodiments, collecting suspected disaster data for each disaster risk sub-area includes:

[0066] For each suspected disaster basin, data is collected during the same period, where the data collection period for each suspected disaster basin is the same; and

[0067] For any suspected disaster basin, during the corresponding collection period, collection is carried out according to the flow direction of the tributaries of the suspected disaster basin and the time sequence of the collection period to obtain suspected disaster data of each disaster risk sub-area. The specific suspected disaster data collected may include image data, video data, etc.

[0068] In the embodiments of this application, simultaneous data collection is not performed. Instead, data is collected for each suspected disaster basin at the same time and in chronological order, thereby obtaining disaster characteristics that indicate dynamic development trends for the suspected disaster basin. In other embodiments, for example, a data collection period can be set based on the update cycle of a weather forecast, thereby ensuring that the weather characteristics and the data collection period have the same update cycle.

[0069] In some embodiments, extracting suspected disaster characteristic information of suspected disaster data of each disaster risk sub-area includes: for the suspected disaster data collected in any disaster risk sub-area, extracting tributary features in the image data or video data, and splicing them in the time sequence of collection, thereby forming suspected disaster characteristic information of any disaster risk sub-area. In this way, part of the recognition sensitivity can be discarded, thereby focusing on the overall flow and sediment development trend during the collection period.

[0070] The feature sequence is then input into a pre-trained prediction model to obtain prediction results for the safe operation of the hydropower station, thereby predicting the flow and sediment content in the current hydropower station basin in the future period.

[0071] The method for predicting the safe operation of a hydropower station taking environmental disasters into consideration in the embodiment of the present application utilizes a prediction model to achieve matching with the current hydropower station basin, and identifies potential safety hazards of the hydropower station caused by environmental disasters, especially extreme weather such as heavy rain and mudslides, thereby assisting hydropower station operation managers in making judgments and improving the reliability of the safe operation of the hydropower station.

[0072] An embodiment of the present application also proposes a hydropower station safe operation prediction system taking into account environmental disasters. The system includes a processor and a memory, and a computer program is stored on the memory. When the computer program is executed by the processor, it implements the steps of the hydropower station safe operation prediction method taking into account environmental disasters as described above.

[0073] Furthermore, 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., solutions that intersect various embodiments), adaptations, or changes. The present invention is not limited to the examples described in this specification or during the practice of this application, which examples are to be construed as non-exclusive.

[0074] 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.

[0075] 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 should also be deemed to fall within the protection scope of the present invention.

Claims

1. A method for predicting safe operation of a hydropower station taking environmental disasters into consideration, characterized in that: include: Combined with terrain data, the terrain range that affects the safe operation of the current hydropower station is delineated, and multiple disaster risk sub-areas within the terrain range are identified; as well as Determine the key risk time interval of the current hydropower station based on historical weather data; During the key risk time interval, multiple disaster risk sub-areas are monitored to collect suspected disaster data of each disaster risk sub-area and obtain weather information within a specified future period; Extract suspected disaster characteristic information of suspected disaster data of each disaster risk sub-area, and extract weather characteristic information of weather information; Combining the extracted suspected disaster feature information and weather feature information in a set order to obtain a feature sequence; Input the feature sequence into a pre-trained prediction model to obtain a prediction result of the safe operation of the hydropower station, wherein the prediction model is used to predict the flow and sediment content of the current hydropower station basin in the future period; Among them, combined with terrain data, the terrain range that affects the safe operation of the current hydropower station is delineated, including: Constructing a terrain model of the defined terrain range according to the terrain data, and determining a basic watershed of the current hydropower station from the terrain data; Taking the basic watershed as a reference, exchanging the height difference of the terrain data, and determining an inverse model of the terrain model; Determine a plurality of suspected disaster basins connected to the basic basin according to the inverse model, construct a tree-like distribution map based on the plurality of suspected disaster basins, and determine a plurality of disaster risk sub-areas based on the tree-like distribution map; According to the node relationship of the tree-like distribution diagram and the terrain model, a combination order of the feature sequence is configured for each node; Taking the basic watershed as a reference, exchanging the height difference of the terrain data, and determining the inverse model of the terrain model, including: Obtaining the altitude information of the basic watershed; Taking the low altitude interval in the altitude information as a reference benchmark, and subtracting the altitude information of the basic watershed from the reference benchmark to determine the simulated watershed trend in the inverse model; For terrain data in which the angle between any side of the basic watershed in the terrain data and the basic watershed plane is greater than a preset angle threshold, retain the actual watershed terrain relationship and do not perform the exchange process, so as to determine the inverse model of the terrain model; Using the reference datum, replacing the height difference of the terrain data, wherein the lower the altitude of the basin in the basic basin is, the higher the altitude of the basin in the inverse model is after the replacement; Based on the replaced basic watershed and the retained watershed terrain relationship, construct the inverse model; According to the inverse model, multiple suspected disaster basins connected to the basic basin are determined, including: In a simulation environment, performing flow simulation based on the inverse model to obtain a flow simulation result of reverse flow; According to the flow simulation results, a tributary whose flow is greater than a specified flow threshold is selected as a suspected disaster basin; According to the node relationship of the tree-like distribution diagram and the terrain model, a combination order of feature sequences is configured for each node, including: According to the flow simulation results and the positional relationship between each node and the current hydropower station in the tree-like distribution diagram, a combination order of feature sequences is configured for each node, wherein the larger the value of the flow simulation result of any tributary node position and the closer the node position is to the current hydropower station, the higher the order.

2. The method for predicting safe operation of a hydropower station taking environmental disasters into consideration according to claim 1, characterized in that: The suspected disaster data collected for each disaster risk sub-area include: For each suspected disaster basin, data is collected during the same period, where the data collection period for each suspected disaster basin is the same; and For any suspected disaster basin, data are collected during the corresponding collection period according to the flow direction of the tributaries of the suspected disaster basin and the time sequence of the collection period to obtain the suspected disaster data of each disaster risk sub-area.

3. The method for predicting safe operation of a hydropower station taking environmental disasters into consideration according to claim 1, characterized in that: The pre-trained prediction model is pre-trained based on historical weather data and disaster data of the main stream and tributaries of each river basin, and is configured with corresponding global weights for any geological condition so as to adapt to the current hydropower station basin environment based on the global weights.

4. A hydropower station safe operation prediction system considering environmental disasters, characterized in that: The system includes 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 method for predicting safe operation of a hydropower station considering environmental disasters as described in any one of claims 1 to 3 are implemented.

Images