Hydropower full-time-domain optimization scheduling system and method based on multi-source data fusion

Abstract

The invention discloses a hydropower full-time-domain optimization dispatching system and method based on multi-source data fusion, and relates to the technical field of hydropower dispatching. Meanwhile, the risk of the landslide hidden danger point to the hydropower station is analyzed by combining the distance between the landslide hidden danger point and the hydropower station, future rainfall abnormity is analyzed, and reservoir capacity data of the hydropower station is monitored by combining the current soil saturation, risk estimation of the landslide hidden danger point to the hydropower station and reservoir sediment abnormity risks caused by rainfall triggered landslide. According to the method, effective pre-judgment and risk assessment are carried out on the landslide hidden danger, the hydropower dispatching instruction is accurately triggered, the accuracy and controllability of risk early warning are improved, and the risk early warning efficiency is improved. And the refined and intelligent safety management requirements of the hydropower station are met.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water and electricity dispatching, in particular to a water and electricity full-time domain optimization dispatching system and method based on multi-source data fusion. BACKGROUND

[0002] At present, the landslide and sediment risk monitoring of the reservoir area of the hydropower station mostly depends on artificial inspection and geological survey, which is difficult to cover the vast and complex terrain of the reservoir area, and is prone to miss the hidden danger points with strong concealment and in the early deformation stage. Some scheduling decisions are made based on historical experience without considering the specific location, scale, current activity of the landslide and the actual defense capability of the hydropower station, which often leads to false positives or false negatives. The existing technology cannot form a closed-loop management from discovering landslide hazards to sediment risk prediction and then to water and electricity optimization dispatching, and it is difficult to meet the fine and intelligent dispatching needs of modern hydropower stations.

[0003] In order to solve the above problems, the present application provides a water and electricity full-time domain optimization dispatching system and method based on multi-source data fusion. SUMMARY

[0004] In view of the shortcomings of the prior art, the present application provides a water and electricity full-time domain optimization dispatching system and method based on multi-source data fusion. The present application effectively predicts and evaluates landslide hazards, accurately triggers water and electricity dispatching instructions, improves the accuracy and controllability of risk warning, and meets the fine and intelligent safety management needs of hydropower stations.

[0005] To achieve the above purpose, the present application provides a water and electricity full-time domain optimization dispatching method based on multi-source data fusion, which includes the following specific steps:

[0006] S1, generate a ground deformation rate map by synthetic aperture radar interferometry technology, screen deformation anomaly areas, collect images of the deformation anomaly areas by an unmanned aerial vehicle, and locate landslide hazard points;

[0007] S2, monitor water turbine generator set operation data and reservoir water level variation data, analyze water turbine generator set operation anomalies and reservoir water level variation anomalies, and analyze the risk of landslide hazard points to the hydropower station in combination with the distance between the landslide hazard points and the hydropower station;

[0008] S3, monitor rainfall data in a future set time period, analyze future rainfall anomalies, and estimate rainfall-triggered landslide-induced abnormal sediment risk to the hydropower station in combination with the current soil saturation and the risk of landslide hazard points to the hydropower station;

[0009] S4, monitor the reservoir capacity data of the hydropower station, obtain the sediment defense capability of the hydropower station, and determine whether to trigger the dispatching instructions of the hydropower station based on the sediment defense capability and the abnormal sediment risk into the reservoir.

[0010] Preferably, the S1 includes the following specific steps:

[0011] S11, collect multi-temporal synthetic aperture radar images of a set region and set monitoring time length, process the multi-temporal synthetic aperture radar images through time series InSAR technology, extract time series deformation information of each pixel point, generate a ground surface deformation rate map of the set region, and screen a region with a ground surface deformation rate exceeding a preset deformation threshold value as a deformation anomaly region;

[0012] S12, collect images of the deformation anomaly region through a UAV, generate a high-precision digital surface model, and automatically identify a landslide geomorphology boundary;

[0013] S13, obtain spatial similarity of the deformation anomaly region and the landslide geomorphology boundary based on an intersection over union, screen a region exceeding a spatial similarity threshold value as a landslide hidden danger point.

[0014] Preferably, the S2 comprises the following specific steps:

[0015] S21, monitor water turbine generator set operation data, the water turbine generator set operation data comprising vibration frequency, vibration amplitude and temperature, obtain vibration frequency abnormal value based on the vibration frequency, obtain vibration amplitude abnormal value based on the vibration amplitude, obtain temperature abnormal value based on the temperature, and obtain water turbine generator set operation abnormal value based on weighted summation of the vibration frequency abnormal value, the vibration amplitude abnormal value and the temperature abnormal value;

[0016] S22, monitor reservoir water level variation data, the reservoir water level variation data comprising water level change rate, and obtain reservoir water level variation abnormal value based on a ratio of the water level change rate to a water level change rate safety threshold value;

[0017] S23, obtain hydropower station abnormal value based on weighted summation of the water turbine generator set operation abnormal value and the reservoir water level variation abnormal value;

[0018] S24, obtain potential landslide body volume, and calculate landslide potential influence value based on the potential landslide volume and the ground surface deformation rate, wherein the landslide potential influence value is obtained through a landslide potential influence value calculation formula, and the landslide potential influence value calculation formula is: , wherein, is the landslide potential influence value, is the potential landslide body volume, is a landslide body volume safety value, is a landslide body volume influence coefficient, is the ground surface deformation rate, is a preset deformation threshold value, is a deformation rate influence coefficient;

[0019] S25, obtain the distance between the landslide hidden danger point and the hydropower station, and obtain a distance influence value based on the distance, wherein the distance influence value is obtained through a distance influence value calculation formula, and the distance influence value calculation formula is: , wherein, is the distance influence value of the landslide hidden danger point and the hydropower station, is a distance attenuation coefficient, is the distance of the landslide hidden danger point and the hydropower station, is a maximum influence distance of the landslide;

[0020] S26, obtain a threat value of the landslide hidden danger point to the hydropower station based on the product of the landslide potential influence value, the hydropower station abnormal value and the distance influence value.

[0021] Preferably, the S3 comprises the following specific steps:

[0022] S31, monitor rainfall data in a future set time period, and obtain a future rainfall abnormal value based on the sum of the rainfall amount in the monitoring time period and the safe rainfall amount divided by the number of rainfall data;

[0023] S32, obtain a rainfall triggering influence coefficient based on weighted summation of the current soil saturation degree and the future rainfall abnormal value;

[0024] S33, obtain a rainfall triggered landslide caused reservoir sediment abnormal risk value based on the product of the threat value of the landslide hidden danger point to the hydropower station and the rainfall triggering influence coefficient plus the threat value of the landslide hidden danger point to the hydropower station.

[0025] Preferably, the S4 comprises the following specific steps:

[0026] S41, monitor the reservoir capacity data of the hydropower station, the reservoir capacity data of the hydropower station comprising a current remaining effective reservoir capacity, and obtain a sediment defense capability evaluation value of the hydropower station based on the ratio of the current remaining effective reservoir capacity to the designed effective reservoir capacity;

[0027] S42, obtain a sediment defense safety value based on the ratio of the sediment defense capability evaluation value to the reservoir sediment abnormal risk value, and determine whether to trigger the dispatching instruction of the hydropower station based on the comparison result of the sediment defense safety value and a sediment defense safety threshold value, if the sediment defense safety value is greater than the sediment defense safety threshold value, the dispatching instruction of the hydropower station is not triggered, and if the sediment defense safety value is less than or equal to the sediment defense safety threshold value, the dispatching instruction of the hydropower station is triggered.

[0028] The application also provides a hydropower full-time domain optimization scheduling system based on multi-source data fusion, comprising:

[0029] A landslide hidden danger point positioning module is configured to generate a ground surface deformation rate map through synthetic aperture radar interferometry, screen a deformation anomaly area, collect an image of the deformation anomaly area through a UAV, and locate a landslide hidden danger point.

[0030] The water power station risk analysis module is used for monitoring water turbine generator set operation data and reservoir water level variation data, analyzing water turbine generator set operation abnormity and reservoir water level variation abnormity, and analyzing the risk of landslide hidden danger points to the water power station in combination with the distance between the landslide hidden danger points and the water power station;

[0031] The reservoir sediment risk analysis module is used for monitoring rainfall data in a future set time period, analyzing future rainfall abnormity, and estimating the risk of reservoir sediment abnormity caused by rainfall triggering landslide in combination with the current soil saturation and the risk of landslide hidden danger points to the water power station.

[0032] The defense capability judgment module is used for monitoring reservoir capacity data of the water power station and acquiring the sediment defense capability of the water power station.

[0033] The dispatch triggering module is used for judging whether to trigger the dispatch instruction of the water power station based on the sediment defense capability and the reservoir sediment abnormity risk.

[0034] The application further provides an electronic device, comprising a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes the water power full-time domain optimization dispatching method based on multi-source data fusion by calling the computer program stored in the memory.

[0035] The application further provides a computer readable storage medium storing instructions, which, when executed on a computer, cause the computer to execute the water power full-time domain optimization dispatching method based on multi-source data fusion.

[0036] The application has the following beneficial effects: the application generates a ground surface deformation rate map by using the synthetic aperture radar interferometry technology, screens a deformation abnormal area, collects images of the deformation abnormal area by using a UAV, locates landslide hidden danger points, monitors water turbine generator set operation data and reservoir water level variation data, analyzes water turbine generator set operation abnormity and reservoir water level variation abnormity, analyzes the risk of landslide hidden danger points to the water power station in combination with the distance between the landslide hidden danger points and the water power station, monitors rainfall data in a future set time period, analyzes future rainfall abnormity, estimates the risk of reservoir sediment abnormity caused by rainfall triggering landslide in combination with the current soil saturation and the risk of landslide hidden danger points to the water power station, monitors reservoir capacity data of the water power station, acquires the sediment defense capability of the water power station, and judges whether to trigger the dispatch instruction of the water power station based on the sediment defense capability and the reservoir sediment abnormity risk. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to make the technical solutions in the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description only aim to some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0038] Figure 1 Flowchart of the method for the full-time domain optimal scheduling of water and electricity based on multi-source data fusion of the present application;

[0039] Figure 2 Flowchart of S1 of the method for the full-time domain optimal scheduling of water and electricity based on multi-source data fusion of the present application;

[0040] Figure 3 Flowchart of S2 of the method for the full-time domain optimal scheduling of water and electricity based on multi-source data fusion of the present application;

[0041] Figure 4 Flowchart of S3 of the method for the full-time domain optimal scheduling of water and electricity based on multi-source data fusion of the present application;

[0042] Figure 5 Overall framework diagram of the system for the full-time domain optimal scheduling of water and electricity based on multi-source data fusion of the present application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those skilled in the art on the basis of the embodiments in the present application belong to the scope of protection of the present application.

[0044] Please refer to Figure 1 The embodiment of the present application provides a method for the full-time domain optimal scheduling of water and electricity based on multi-source data fusion, which comprises the following specific steps.

[0045] S1, generate a ground surface deformation rate map through synthetic aperture radar interferometry technology, screen a deformation anomaly area, collect images of the deformation anomaly area through a UAV, and locate landslide hidden danger points.

[0046] Please refer to Figure 2 In the embodiment, S1 comprises the following specific steps.

[0047] S11. Acquire multi-temporal synthetic aperture radar (SAR) images of the designated area for a set monitoring duration. In this embodiment, the monitoring duration can be set to one year to effectively separate seasonal deformation from long-term trend deformation. The multi-temporal synthetic aperture radar images are processed using time-series InSAR technology to extract the time-series deformation information of each pixel, i.e., to deduce the cumulative deformation sequence of each pixel in the image coverage area over time. Linear regression analysis is performed on the time-series deformation information of each point to calculate its annual average deformation rate. The deformation rates of all points are spatially interpolated or visualized to generate a surface deformation rate map of the designated area. Areas with surface deformation rates exceeding a preset deformation threshold are selected as deformation anomaly areas. In this embodiment, the preset deformation threshold is determined based on the geological background, vegetation cover, and InSAR data quality of the designated area, and can be ±10. mm / a (unit: millimeters per year), areas exceeding this threshold are considered to have significant deformation. On the surface deformation rate map of the set area, all pixels with deformation rates exceeding the preset threshold are automatically filtered out and aggregated into continuous polygonal regions, which are then marked as deformation anomaly areas.

[0048] S12. Collect images of the deformation anomaly area using drones, generate a high-precision digital surface model, and automatically identify the landslide landform boundary.

[0049] In this embodiment, for each deformation anomaly area, a UAV aerial survey route is designed to ensure that the overlap between the flight path and the lateral direction is greater than 75%. The UAV aerial survey task is executed to acquire high-resolution oblique images covering all deformation anomaly areas. ContextCapture is used to process the UAV images, generating a high-precision digital surface model (DSM) and a digital orthophoto map (DOM). The DSM meticulously depicts the surface undulations with centimeter-level precision. On the DSM and DOM, professionals perform visual interpretation or use deep learning algorithms to automatically identify landslide landform boundaries. The identified landforms mainly include:

[0050] Back slope: The steep boundary at the rear edge of a landslide;

[0051] Landslide perimeter cracks: including tension cracks at the rear edge and shear cracks on both sides;

[0052] Leading edge uplift (drum mound): The uplifted area formed by compression at the leading edge of the landslide body;

[0053] Landslide embankments: terraced areas within a landslide body;

[0054] S13, based on the intersection ratio, obtain the spatial similarity of the deformation anomaly area and the landslide geomorphic boundary. The spatial similarity refers to the consistency degree of the deformation anomaly area and the landslide geomorphic boundary in spatial position and shape. In the GIS platform, the two layers are superimposed, and the spatial similarity is obtained by calculating the intersection ratio of the deformation anomaly area and the landslide geomorphic boundary. The higher the intersection ratio value, the higher the overlap degree and the higher the similarity. The region exceeding the spatial similarity threshold value is screened. The spatial similarity threshold value in the embodiment can be set to 0.5 and set to the landslide hidden danger point.

[0055] In the specific implementation of the embodiment, when the spatial similarity of the deformation anomaly area and the landslide geomorphic boundary of a certain region exceeds the preset threshold value, it can be confirmed that the region is a landslide hidden danger point that has occurred deformation and has the characteristics of landslide shape. Once the hidden danger point loses stability, a large amount of sediment may enter the reservoir instantaneously, resulting in sudden reduction of reservoir capacity, wear and tear of the unit, and even failure of scheduling. Early identification and risk prediction can reduce sudden sediment entry into the reservoir, improve scheduling safety and controllability, dynamically adjust the flood discharge and sediment discharge window through sediment monitoring and prediction, for example, reduce the reservoir water level in advance, create sediment discharge conditions, actively reduce reservoir sedimentation, and prolong the service life of the reservoir before the activity of the landslide hidden danger point is enhanced and rainfall arrives.

[0056] S2, monitor the water turbine generator set operation data and the reservoir water level change data, analyze the water turbine generator set operation anomaly and the reservoir water level change anomaly, and analyze the risk of the landslide hidden danger point to the hydropower station in combination with the distance between the landslide hidden danger point and the hydropower station.

[0057] Please refer to Figure 3 In the embodiment, S2 includes the following specific steps:

[0058] S21, monitor the water turbine generator set operation data, the water turbine generator set operation data including vibration frequency, vibration amplitude and temperature, obtain vibration frequency abnormal value based on vibration frequency, obtain vibration amplitude abnormal value based on vibration amplitude, and obtain temperature abnormal value based on temperature. The vibration frequency abnormal value, the vibration amplitude abnormal value and the temperature abnormal value in the embodiment can be obtained through the abnormal value calculation formula. The abnormal value calculation formula is: , wherein, is the abnormal value of the i th operation data group, the operation data group including vibration frequency data, vibration amplitude data and temperature data, is the value of the i th operation data group at time t, is the safe operation value corresponding to the i th operation data group, the safe operation value including safe vibration frequency, safe vibration amplitude and safe temperature, the safe vibration frequency, the safe vibration amplitude and the safe temperature being obtained through the design and operation manual of the unit, is the monitoring time length of the operation data group, For time integration, the operation abnormal value of the hydro-generator unit is obtained by weighted summation based on the vibration frequency abnormal value, the vibration amplitude abnormal value and the temperature abnormal value;

[0059] S22, reservoir water level fluctuation data is monitored, the reservoir water level fluctuation data includes water level change rate per unit time, and the reservoir water level fluctuation abnormal value is obtained based on the ratio of the water level change rate to the water level change rate safety threshold value, the water level change rate safety threshold value in the embodiment is obtained by calculating the distribution of the water level change rate in the historical data, removing abnormal working conditions and taking the average value, which can be 0.3 m / h;

[0060] S23, the abnormal value of the hydropower station is obtained by weighted summation based on the operation abnormal value of the hydro-generator unit and the reservoir water level fluctuation abnormal value;

[0061] S24, the potential landslide volume is obtained, in the embodiment, the landslide boundary is identified through the texture, tone, vegetation coverage and other characteristics of the image, the range of the landslide body is determined in combination with the digital elevation model, the volume calculation tool in the GIS software is used, the elevation difference between the landslide body surface and the original ground is compared, the volume of the landslide body is calculated, and the landslide potential influence value is calculated based on the potential landslide volume and the ground deformation rate, wherein the landslide potential influence value is obtained by the landslide potential influence value calculation formula, and the landslide potential influence value calculation formula is: , wherein, is the landslide potential influence value, is the potential landslide volume, is the landslide volume safety value, the landslide volume safety value in the embodiment is obtained by collecting the median of the typical landslide volume or all landslide volumes in the region, is the landslide volume influence coefficient, is the ground deformation rate, is the preset deformation threshold value, is the deformation rate influence coefficient, in the embodiment, detailed information of landslides occurred in the research area is collected, the volume, deformation rate, actual damage degree and other data of each landslide case are extracted and arranged into a structured database, the correlation between the volume, deformation rate and actual damage degree is analyzed, statistical software is used for multiple regression analysis, the initial value of the landslide volume influence coefficient is set to 0.5, the initial value of the deformation rate influence coefficient is set to 1, the rationality of the initial parameters is verified by using historical data, and the parameter values are adjusted according to the verification result, so that the model prediction result is consistent with the actual situation;

[0062] S25, the distance between the landslide hidden danger point and the hydropower station is obtained, and the distance influence value is obtained based on the distance, wherein the distance influence value is obtained by the distance influence value calculation formula, and the distance influence value calculation formula is: , wherein, is the distance influence value of the landslide hidden danger point and the hydropower station, As the distance attenuation coefficient, the embodiment collects distance-impact degree data of historical landslides, performs exponential function fitting on the data, extracts the distance attenuation coefficient from the fitting result, and checks the goodness of fitting, As the distance between the landslide hidden danger point and the hydropower station, As the maximum impact distance of the landslide, the embodiment establishes a three-dimensional geological model of the landslide body, inputs rock and soil mechanics parameters, sets boundary conditions of the landslide, simulates the landslide movement process, determines the impact range boundary, and extracts the maximum impact distance from the simulation result;

[0063] S26, obtaining a threat value of the landslide hidden danger point to the hydropower station based on the product of the landslide potential impact value, the hydropower station abnormal value and the distance impact value.

[0064] The scale and damage range of the landslide directly affect the safety of the hydropower station. For example, a large landslide may destroy the slope of the water diversion channel, causing the water turbine unit to shut down or be damaged, affecting power generation efficiency, while a close-range landslide may directly cause structural damage, and a long-range landslide may indirectly affect the operation stability of the hydropower station through seepage, foundation settlement, etc. The embodiment combines potential landslide risk, hydropower station operation data and spatial position to comprehensively evaluate the risk, thereby improving the accuracy of risk early warning.

[0065] S3, monitoring rainfall data in a future set time period, analyzing future rainfall anomalies, combining current soil saturation, and estimating the risk of rainfall triggering landslide-induced reservoir sedimentation anomalies of the landslide hidden danger point to the hydropower station;

[0066] Please refer to Figure 4 In the embodiment, S3 includes the following specific steps:

[0067] S31, monitoring rainfall data in a future set time period, obtaining a future rainfall anomaly value based on the sum of the ratio of rainfall amount to safe rainfall amount in the monitoring time period divided by the number of rainfall data, and the safe rainfall amount of the embodiment can be 100 mm of rainfall within 3 hours;

[0068] S32, obtaining soil saturation through a soil moisture monitoring station, and obtaining a rainfall triggering impact coefficient based on weighted summation of the current soil saturation and the future rainfall anomaly value;

[0069] S33, obtaining a rainfall triggering landslide-induced reservoir sedimentation anomaly risk value based on the product of the threat value of the landslide hidden danger point to the hydropower station and the rainfall triggering impact coefficient plus the threat value of the landslide hidden danger point to the hydropower station, future rainfall anomalies directly affect runoff intensity and catchment speed, and soil saturation directly affects shear strength and landslide sensitivity. When the soil moisture content is close to saturation, the friction between soil particles decreases, which is easy to trigger landslides or collapses, resulting in a large amount of sediment entrained by the landslide body into the reservoir. The embodiment predicts the chain risk scenario triggered by geological disasters based on the threat of the landslide hidden danger point and the rainfall triggering impact.

[0070] The embodiment obtains the weight coefficients of the vibration frequency abnormal value, the vibration amplitude abnormal value, the temperature abnormal value, the water turbine generator set operation abnormal value, the reservoir water level variation abnormal value, the current soil saturation and the future rainfall abnormal value through the coefficient of variation method. The specific steps of the coefficient of variation method are as follows: calculating the average value and the standard deviation corresponding to each index, then taking the quotient of the standard deviation of each index and the average value to obtain the coefficient of variation of the index, which represents the discrete degree of the index data, then adding the coefficients of variation of each index, and finally obtaining the weight coefficient of each index by calculating the proportion of each coefficient of variation in the total.

[0071] S4, monitoring the reservoir capacity data of the hydropower station, obtaining the sediment defense capability of the hydropower station, and judging whether to trigger the dispatching instruction of the hydropower station based on the sediment defense capability and the abnormal risk of the incoming reservoir sediment.

[0072] In the embodiment, S4 includes the following specific steps:

[0073] S41, monitoring the reservoir capacity data of the hydropower station, the reservoir capacity data of the hydropower station including the current remaining effective capacity, the current remaining effective capacity in the embodiment being obtained by the difference between the total capacity (to the original design normal water level) under the current state and the dead storage under the current state, that is, obtaining the remaining reservoir capacity that can be used for sediment accumulation, and obtaining the sediment defense capability evaluation value of the hydropower station based on the ratio of the current remaining effective capacity to the design effective capacity;

[0074] S42, obtaining the sediment defense safety value based on the ratio of the sediment defense capability evaluation value to the abnormal risk value of the incoming reservoir sediment, and judging whether to trigger the dispatching instruction of the hydropower station based on the comparison result of the sediment defense safety value and the sediment defense safety threshold value. If the sediment defense safety value is greater than the sediment defense safety threshold value, the dispatching instruction of the hydropower station is not triggered. If the sediment defense safety value is less than or equal to the sediment defense safety threshold value, the dispatching instruction of the hydropower station is triggered. For example, the system can automatically generate and execute the optimized dispatching strategy of pre-discharge reservoir cleaning and sediment discharge.

[0075] The sediment defense safety threshold value of the embodiment can be 1.2. The specific obtaining steps are as follows: collecting historical data of past years, including hydrology, sediment, meteorology and reservoir operation records. In the historical data, the dangerous events of serious accumulation, blockage of sediment discharge facilities and difficult dispatching due to sediment problems are distinguished from the safe events of smooth reservoir operation and non-protruding sediment problems. The sediment defense safety value sequence of each set period in history is calculated, the sediment defense safety values corresponding to all safe events are counted, and the 25% quantile of the sediment defense safety values is taken as the sediment defense safety threshold value. When the sediment defense safety value is higher than this value, it can be considered that the hydropower station is relatively safe.

[0076] Please refer to Figure 5The embodiment of the present application also provides a hydropower full-time domain optimization scheduling system based on multi-source data fusion, comprising: a landslide hidden danger point positioning module, which is used for generating a ground surface deformation rate graph through an interferometric synthetic aperture radar technology, screening a deformation anomaly area, collecting an image of the deformation anomaly area through an unmanned aerial vehicle, and positioning a landslide hidden danger point;

[0077] A hydropower station risk analysis module is used for monitoring water turbine generator set operation data and reservoir water level variation data, analyzing water turbine generator set operation abnormalities and reservoir water level variation abnormalities, and analyzing the risk of the landslide hidden danger point to the hydropower station in combination with the distance between the landslide hidden danger point and the hydropower station;

[0078] A reservoir sediment risk analysis module is used for monitoring rainfall data in a future set time period, analyzing future rainfall abnormalities, and estimating the risk of rainfall triggering landslide caused reservoir sediment abnormalities in combination with current soil saturation and the risk of the landslide hidden danger point to the hydropower station.

[0079] A defense capability judgment module is used for monitoring reservoir capacity data of the hydropower station and acquiring the sediment defense capability of the hydropower station.

[0080] A scheduling trigger module is used for judging whether to trigger a scheduling instruction of the hydropower station based on the sediment defense capability and the reservoir sediment abnormal risk.

[0081] The embodiment of the present application also provides an electronic device, comprising a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes the computer program stored in the memory to perform the hydropower full-time domain optimization scheduling method based on multi-source data fusion.

[0082] The electronic device can have great differences due to different configurations or performances, and can comprise one or more processors (Central Processing Units, CPUs) and one or more memories, wherein the memory stores at least one computer program, the computer program is loaded and executed by the processor to realize the hydropower full-time domain optimization scheduling method based on multi-source data fusion provided by the method embodiment. The electronic device can also comprise other components for realizing device functions, for example, the electronic device can also have a wired or wireless network interface and an input and output interface and other components to input and output data. This embodiment will not be described here.

[0083] The embodiment of the present application also provides a computer readable storage medium, which stores instructions, when a computer program runs on a computer device, so that the computer device executes the hydropower full-time domain optimization scheduling method based on multi-source data fusion.

[0084] For example, the computer readable storage medium can be read-only memory (ROM), random access memory (RAM), compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0085] The above-described embodiments can be implemented in part or in whole through software, hardware, firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of one or more computer programs that are loaded in and executed by a computer. The computer program can be stored in a computer readable medium, such as a memory, a storage device, or a data storage medium such as a compact disc read-only memory (CD-ROM), a floppy disk, a USB flash drive, or a hard disk. The computer program can be distributed over the Internet or over a network connected to the Internet.

[0086] In the description of the present specification, the description referring to the terms "one embodiment", "an example", "a specific example", and the like means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present specification, and the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example, and the specific features, structures, materials, or characteristics described can be incorporated in any one or more embodiments or examples in a suitable manner.

[0087] The above-disclosed preferred embodiments of the present application are only used to help explain the present application, and the preferred embodiments do not describe all the details, nor limit the present application to the specific embodiments. Obviously, many modifications and variations can be made to the present application according to the contents of the present specification. The present specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can well understand and utilize the present application. The present application is limited only by the claims and their full scope and equivalents.

Claims

1. A method for full-time domain optimal scheduling of hydropower based on multi-source data fusion, characterized in that, The method comprises the following specific steps: S1, generating a ground surface deformation rate map through synthetic aperture radar interferometry technology, screening a deformation anomaly area, collecting images of the deformation anomaly area through a UAV, and locating a landslide hidden danger point; S2, monitoring water turbine generator set operation data and reservoir water level variation data, analyzing water turbine generator set operation anomalies and reservoir water level variation anomalies, and analyzing the risk of the landslide hidden danger point to the hydropower station in combination with the distance between the landslide hidden danger point and the hydropower station; S3, monitoring rainfall data in a future set time period, analyzing future rainfall anomalies, combining the current soil saturation degree, and estimating the risk of the landslide hidden danger point to the hydropower station to predict the abnormal risk of the incoming sediment caused by rainfall triggering landslides; S4, monitoring the reservoir capacity data of the hydropower station, obtaining the sediment defense capability of the hydropower station, and judging whether the dispatching instruction of the hydropower station is triggered based on the sediment defense capability and the abnormal risk of the incoming sediment.

2. The multi-source data fusion based hydro-thermal full-time domain optimal scheduling method according to claim 1, characterized in that, The S1 comprises the following specific steps: S11, collecting multi-temporal synthetic aperture radar images of a set area for a set monitoring time period, processing the multi-temporal synthetic aperture radar images, extracting time sequence deformation information of each pixel point, generating a ground surface deformation rate map of the set area, screening an area with a ground surface deformation rate exceeding a preset deformation threshold, and setting the area as a deformation anomaly area; S12, collecting images of the deformation anomaly area through a UAV, generating a high-precision digital surface model, and automatically identifying a landslide topographic boundary; S13, obtaining the spatial similarity of the deformation anomaly area and the landslide topographic boundary based on the intersection over union, screening an area exceeding a spatial similarity threshold, and setting the area as a landslide hidden danger point.

3. The water and electricity full-time domain optimization scheduling method based on multi-source data fusion according to claim 2, characterized in that, The S2 comprises the following specific steps: S21, monitoring water turbine generator set operation data, the water turbine generator set operation data comprising vibration frequency, vibration amplitude and temperature, obtaining vibration frequency abnormal values based on the vibration frequency, obtaining vibration amplitude abnormal values based on the vibration amplitude, obtaining temperature abnormal values based on the temperature, and obtaining water turbine generator set operation abnormal values by weighted summation based on the vibration frequency abnormal values, the vibration amplitude abnormal values and the temperature abnormal values; S22, monitoring reservoir water level variation data, the reservoir water level variation data comprising water level change rate, and obtaining reservoir water level variation abnormal values based on the ratio of the water level change rate to a water level change rate safety threshold; S23, obtaining water power station abnormal values by weighted summation based on the water turbine generator set operation abnormal values and the reservoir water level variation abnormal values; S24, obtain the potential landslide volume, and calculate the landslide potential influence value based on the potential landslide volume and the surface deformation rate, wherein the landslide potential influence value is obtained through a landslide potential influence value calculation formula, and the landslide potential influence value calculation formula is: , wherein, is the landslide potential influence value, is the potential landslide volume, is the landslide volume safety value, is the landslide volume influence coefficient, is the surface deformation rate, is the preset deformation threshold, is the deformation rate influence coefficient. S25, acquire the distance between the landslide hidden danger point and the hydropower station, and acquire the distance influence value based on the distance, wherein the distance influence value is acquired through a distance influence value calculation formula, and the distance influence value calculation formula is: , wherein, is the distance influence value of the landslide hidden danger point and the hydropower station, is a distance attenuation coefficient, is the distance of the landslide hidden danger point and the hydropower station, is the maximum influence distance of the landslide. S26, obtaining the threat value of the landslide hidden danger point to the hydropower station based on the product of the landslide potential influence value, the water power station abnormal value and the distance influence value.

4. The water and electricity full-time domain optimization scheduling method based on multi-source data fusion according to claim 3, characterized in that, The S3 comprises the following specific steps: S31, monitoring rainfall data in a future set time period, obtaining future rainfall anomaly values based on the sum of the ratio of the rainfall amount to the safe rainfall amount in the monitoring time period divided by the number of rainfall data; S32, obtaining rainfall triggering influence coefficients by weighted summation based on the current soil saturation degree and the future rainfall anomaly values; S33, obtaining the abnormal risk value of the incoming sediment caused by rainfall triggering landslides based on the product of the threat value of the landslide hidden danger point to the hydropower station and the rainfall triggering influence coefficient plus the threat value of the landslide hidden danger point to the hydropower station.

5. The multi-source data fusion based hydro-thermal full-time domain optimal scheduling method according to claim 4, characterized in that, The S4 comprises the following specific steps: S41, monitor the reservoir capacity data of the hydropower station, the reservoir capacity data of the hydropower station including the current remaining effective capacity, and the sediment defense capability evaluation value of the hydropower station is obtained based on the ratio of the current remaining effective capacity to the designed effective capacity; S42, the sediment defense safety value is obtained based on the ratio of the sediment defense capability evaluation value to the abnormal risk value of the incoming reservoir sediment, and whether the dispatching instruction of the hydropower station is triggered is judged based on the comparison result of the sediment defense safety value and the sediment defense safety threshold value, if the sediment defense safety value is greater than the sediment defense safety threshold value, the dispatching instruction of the hydropower station is not triggered, if the sediment defense safety value is less than or equal to the sediment defense safety threshold value, the dispatching instruction of the hydropower station is triggered.

6. A water and electricity full-time domain optimization scheduling system based on multi-source data fusion, used for realizing the water and electricity full-time domain optimization scheduling method based on multi-source data fusion according to any one of claims 1-5, characterized in that, Comprise: The landslide hazard point positioning module is used for generating the ground surface deformation rate map through the synthetic aperture radar interferometry technology, screening the deformation anomaly area, collecting the image of the deformation anomaly area through the unmanned aerial vehicle, and positioning the landslide hazard point; The hydropower station risk analysis module is used for monitoring the operation data of the hydroelectric generator set and the reservoir water level variation data, analyzing the operation anomaly of the hydroelectric generator set and the reservoir water level variation anomaly, and analyzing the risk of the landslide hazard point to the hydropower station in combination with the distance between the landslide hazard point and the hydropower station; The incoming sediment risk analysis module is used for monitoring the rainfall data of the future set time period, analyzing the future rainfall anomaly, and combining the current soil saturation degree, the risk of the landslide hazard point to the hydropower station, and the risk estimation of the incoming sediment caused by the rainfall triggering landslide; The defense capability judgment module is used for monitoring the reservoir capacity data of the hydropower station, and obtaining the sediment defense capability of the hydropower station; The dispatching triggering module is used for judging whether the dispatching instruction of the hydropower station is triggered based on the sediment defense capability and the incoming sediment abnormal risk.

7. An electronic device, comprising: Comprise: A memory and a processor, wherein the memory stores a computer program that can be called by the processor, and the processor executes the method of claim 1-5 by calling the computer program stored in the memory.

8. A computer-readable storage medium, characterized in that, The instructions are stored in the computer, when the instructions are run on the computer, the computer executes the method of claim 1-5 based on the multi-source data fusion of the hydropower full-time domain optimization scheduling.

Images