CN119761265A - Dynamic estimation method for parameters of drainage basin hydrologic model based on digital twin technology - Google Patents
Dynamic estimation method for parameters of drainage basin hydrologic model based on digital twin technology Download PDFInfo
- Publication number
- CN119761265A CN119761265A CN202510258551.6A CN202510258551A CN119761265A CN 119761265 A CN119761265 A CN 119761265A CN 202510258551 A CN202510258551 A CN 202510258551A CN 119761265 A CN119761265 A CN 119761265A
- Authority
- CN
- China
- Prior art keywords
- water level
- flash
- data
- flash flood
- risk
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
本发明公开了基于数字孪生技术的流域水文模型参数动态估计方法，涉及流域管理技术领域，包括以下步骤：通过流域内的多模态数据和地形、遥感数据，对流域的水文、气象、地理特征进行全面建模；基于地形信息和气象数据，使用空间聚类算法对流域进行分区，找出山洪易发区域；对于每个山洪易发区，利用所采集的多模态数据提取反映区域山洪暴发潜在风险的关键特征。本发明通过融合多模态数据和地理特征，构建精确的流域模型，采用空间聚类识别山洪易发区，并通过特征工程和机器学习预测风险。结合时空敏感度智能调整，系统能实时响应风险变化，提升预测精度和响应速度，为流域管理者提供精准决策支持，有效预防和减轻山洪灾害的影响。
The present invention discloses a method for dynamic estimation of parameters of a watershed hydrological model based on digital twin technology, which relates to the technical field of watershed management, and includes the following steps: comprehensively modeling the hydrological, meteorological, and geographical characteristics of the watershed through multimodal data, terrain, and remote sensing data within the watershed; partitioning the watershed based on terrain information and meteorological data using a spatial clustering algorithm to find areas prone to flash floods; for each area prone to flash floods, using the collected multimodal data to extract key features that reflect the potential risk of flash floods in the region. The present invention constructs an accurate watershed model by fusing multimodal data and geographical features, uses spatial clustering to identify areas prone to flash floods, and predicts risks through feature engineering and machine learning. Combined with intelligent adjustment of spatiotemporal sensitivity, the system can respond to risk changes in real time, improve prediction accuracy and response speed, provide accurate decision-making support for watershed managers, and effectively prevent and mitigate the impact of flash floods.
Description
Technical Field
The invention relates to the technical field of drainage basin management, in particular to a dynamic estimation method for drainage basin hydrologic model parameters based on a digital twin technology.
Background
The dynamic estimation of parameters of the hydrologic model of the river basin is based on a digital twin technology, namely, key parameters of the hydrologic model are dynamically estimated and optimized by combining a real-time feedback and self-adaptive correction mechanism of the digital twin model through real-time acquisition of multi-source heterogeneous data (such as water level, flow, rainfall, temperature and the like) in the river basin. In the process, firstly, space-time alignment and dynamic fusion are carried out on heterogeneous data by utilizing a multi-mode data fusion technology to generate unified hydrologic dynamic data representation, and then, model parameters are adjusted by adopting a method based on an adaptive correction algorithm through difference calculation of real-time data and historical data, so that the change of hydrologic states in a river basin is accurately reflected. In addition, the Bayesian dynamic network or the countermeasure learning algorithm is combined, the model core parameters are further optimized, and the influence of the model core parameters on the future hydrologic state is predicted, so that the precision and the adaptability of the hydrologic model in the drainage basin water resource scheduling optimization are improved. Finally, the model can realize automatic dynamic adjustment, and ensures efficient response and accurate prediction of complex hydrologic conditions.
The prior art has the following defects:
In adaptive zoning, dynamic weight allocation of a zone typically depends on hydrologic characteristics (such as rainfall intensity, flow fluctuation amplitude, etc.). However, certain areas (such as mountain floods) may not be weighted in time due to terrain complexity or sudden local extreme climatic conditions. For example, a small basin suddenly experiences a strong torrential flood during rainfall, and the model fails to assign sufficient weight to the small basin. This can lead to insufficient water level and flow estimation in the small basin, missing early warning, eventually leading to small-scale mountain torrent disasters and even affecting the safety of downstream areas.
The above information disclosed in the background section is only for enhancement of understanding of the background of the disclosure and therefore it may include information that does not form the prior art that is already known to a person of ordinary skill in the art.
Disclosure of Invention
The invention aims to provide a dynamic estimation method for parameters of a river basin hydrological model based on a digital twin technology, which establishes an accurate river basin model by integrating multi-modal data and geographic features and identifies a mountain torrent prone area by adopting a spatial clustering technology. The key risk features are extracted through the feature engineering technology, prediction is carried out by utilizing machine learning, and the intelligent adjustment of space-time sensitivity is combined, so that the system can dynamically respond to the regional risk change according to real-time data and the torrent outbreak prediction result, the accuracy and response speed of torrent risk prediction are greatly improved, the basin manager is ensured to obtain more accurate and dynamic decision support in time, and the influence of torrent disasters is effectively prevented and lightened, so that the problem in the background technology is solved.
In order to achieve the purpose, the invention provides the following technical scheme that the method for dynamically estimating the parameters of the drainage basin hydrologic model based on the digital twin technology comprises the following steps:
comprehensively modeling hydrologic, meteorological and geographic features of the river basin through multi-mode data, topography and remote sensing data in the river basin, wherein the multi-mode data comprises real-time hydrologic data and meteorological data;
Based on the topographic information and the meteorological data, partitioning the river basin by using a spatial clustering algorithm, and finding out a mountain torrent prone area;
For each mountain torrent prone area, extracting key features reflecting the potential risk of mountain torrent in the area by utilizing the acquired multi-mode data, carrying out deep analysis and processing on the extracted features by a feature engineering technology, inputting the processed features as feature vectors into a machine learning model which is trained and put into use, and predicting the potential mountain torrent;
When a potential flood burst risk exists in the flood burst area, the time-space sensitivity threshold of the flood burst area is intelligently adjusted by combining the real-time burst prediction result of the flood burst area and the characteristics of the flood burst prediction result along with time and space changes, and the time-space sensitivity is dynamically improved in the potential flood burst area so as to enhance the response capability to the risk change, and ensure that a river basin manager can obtain more accurate and timely decision support to cope with the potential threat brought by the flood burst.
Preferably, the step of comprehensively modeling hydrologic, meteorological and geographic features of the river basin through multi-mode data, topography and remote sensing data in the river basin comprises the following steps:
Collecting and integrating data from different sources, including real-time hydrologic data, meteorological data, topographic information and remote sensing image data;
preprocessing the collected data, including denoising, standardization and space-time alignment, to ensure the quality and consistency of the data;
carrying out geographical partitioning on the river basin by a space analysis method, and dividing the river basin into a plurality of sub-areas with similar hydrological characteristics;
and analyzing and extracting features of various data by adopting a statistical method, constructing a multidimensional model capable of reflecting hydrologic, meteorological and geographic features in a flow field, and realizing comprehensive modeling of hydrologic, meteorological and geographic features by a multi-mode data fusion technology.
Preferably, the specific steps of using a spatial clustering algorithm to partition the basin based on the topographic information and the meteorological data include:
collecting and integrating the topography and meteorological data in the flow domain, and converting the topography and meteorological data into characteristic vectors suitable for processing by a clustering algorithm;
selecting a clustering algorithm to perform clustering analysis, revealing different areas in a streaming domain through a clustering result, and identifying areas highly related to the mountain torrent, namely mountain torrent easy-to-develop areas;
And (3) displaying the clustering result through a visualization tool, and comprehensively integrating hydrology, weather and geographic features by combining a multi-mode data fusion technology to construct an omnibearing drainage basin model.
Preferably, for each mountain torrent prone region, key features reflecting the potential risk of mountain torrent in the region are extracted by utilizing the acquired multi-mode data, wherein the extracted features comprise rising and falling amplitude of water level and frequency and amplitude of flow change in the flow domain, the rising and falling amplitude of water level and the frequency and amplitude of flow change in the flow domain are subjected to deep analysis and processing by a characteristic engineering technology in the monitoring period, a water level fluctuation reference value and a flow fluctuation reference value are respectively generated, the water level fluctuation reference value and the flow fluctuation reference value are used as feature vectors to be input into a machine learning model which is trained and put into use, and the potential mountain torrent is predicted by a mountain torrent risk index generated by the machine learning model.
Preferably, the machine learning model which is completed through training and is put into use is used for comparing and analyzing the mountain torrent risk index generated when the potential mountain torrent is predicted with a preset mountain torrent risk index reference threshold value, and the potential mountain torrent is predicted, and the specific steps are as follows:
If the mountain torrent risk index is smaller than or equal to the mountain torrent risk index reference threshold, a normal state signal is generated, and the state of the mountain torrent is stable and does not exist.
Preferably, when a potential torrent outbreak risk exists in the torrent prone region, the time-space sensitivity threshold of the torrent prone region is intelligently adjusted by combining the real-time outbreak prediction result of the torrent prone region and the characteristics of the torrent prone region along with time and space changes, and the specific steps are as follows:
Firstly, calculating a space-time sensitivity threshold of a mountain torrent prone region, wherein the space-time sensitivity threshold is dynamically adjusted according to a mountain torrent risk index and a mountain torrent risk index reference threshold, and the space-time sensitivity threshold is calculated according to the following formula: Wherein, the method comprises the steps of, wherein, As a threshold value for the temporal-spatial sensitivity,And is a mountain torrent risk index which represents the mountain torrent risk of the current river basin,A threshold value is referenced for the mountain torrent risk index,The time change rate of the mountain torrent risk index, namely the change speed of the mountain torrent risk index along with time, reflects the change trend of the mountain torrent risk,AndRespectively representing the contribution weights of the current torrent risk index and the torrent risk index reference threshold value to the time space sensitivity threshold value for the preset weighting coefficient,An influence factor representing time variation, and weighing the influence degree of the time variation in the river basin on risk response;
After the time-space sensitivity threshold is adjusted, re-evaluating the mountain torrent outbreak risk of the mountain torrent easy-to-develop area, determining the risk level according to the new time-space sensitivity threshold, and intelligently adjusting the time-space sensitivity threshold of the mountain torrent easy-to-develop area when the potential mountain torrent outbreak risk exists in the mountain torrent easy-to-develop area, wherein the dynamic adjustment expression of the new time-space sensitivity threshold is as follows: Wherein: is a time-space response adjustment factor used for controlling the dynamic magnification of a time-space sensitivity threshold value when the flood burst risk is changed, And in order to adjust the space-time sensitivity threshold, dynamically improving the influence based on the deviation between the torrent risk index and the torrent risk index reference threshold, and enhancing the response capability to torrent outbreak risk.
Preferably, for each mountain torrent plot, the specific steps of generating a water level fluctuation reference value by carrying out deep analysis and processing on the rising and falling amplitude of the water level in the flow area through the characteristic engineering technology in the monitoring period are as follows:
For each time window Introducing a water level change amplification factor to identify the intensity of water level change in each time window, wherein the calculation expression of the water level change amplification factor is as follows: Wherein: represents a water level change amplification factor for quantifying the fluctuation intensity of the water level in the time window, Is the water level value corresponding to the j-th time stamp,Is a parameter for controlling the sensitivity of the fluctuation amplitude of the water level, controls the weighting degree of the change,Is a time windowThe number of data points within the set,Is the water level value at the previous time point;
After the calculation of the water level fluctuation amplification in each time window is completed, the water level fluctuation reference value is further generated through the local abnormality measurement For measuring the deviation of the water level variation in each time window relative to the overall trend, and measuring the local abnormalityThe definition is as follows: Wherein: Is a predicted water level value obtained using a locally weighted regression model, Is a parameter controlling sensitivity of the anomaly measure;
And (3) obtaining a final water level fluctuation reference value by combining and weighting the abnormal measurement of each time window with the water level change amplification factor, wherein the calculation expression of the water level fluctuation reference value is as follows: Wherein: Is a reference value for the fluctuation of the water level, Is a weight factor for weighting the fluctuation situation of each time window, i represents the ith time window during monitoring, i.e. the basin water level data is divided into a plurality of time windows during monitoring, m represents the total number of divided time windows during the whole monitoring, i.e. the whole monitoring period is divided into m time windows.
Preferably, for each mountain torrent plot, the specific steps of generating a flow variation reference value by carrying out deep analysis and processing on the frequency and amplitude of the flow variation in the flow area through the characteristic engineering technology in the monitoring period are as follows:
High-order moment analysis is carried out on flow data, complex modes and abnormal fluctuation of flow change are revealed, and skewness in high-order moment is used Kurtosis ofTo capture the asymmetry and extreme fluctuation characteristics of the flow distribution, skewnessKurtosis ofThe specific calculation formula of (2) is as follows: Wherein: For the flow rate at the time of the k-th, Is the average value of the flow rate,N is the total sample number of the data;
The complexity of flow fluctuation is estimated through fractal dimension, the fractal dimension is used for describing the self-similarity and the complexity of a flow curve, the flow fluctuation detail is measured, the fractal dimension of the flow curve is calculated through the following formula, and the calculated expression is: Wherein: Representing the dimension of the fractal dimension, Expressed in scaleThe coverage of the lower flow curve, i.e. the level of detail of the flow change,Is a scale parameter;
Degree of combined deflection Kurtosis degreeFractal dimensionCalculating a flow variation reference value, the calculated expression being: Wherein: and (3) representing a flow variation reference value, and comprehensively measuring the variation characteristics of the flow by combining the skewness, kurtosis and fractal dimension.
In the technical scheme, the invention has the technical effects and advantages that:
According to the invention, by integrating multi-mode data and geographic features, an accurate river basin model is established, and a mountain torrent prone area is identified by adopting a spatial clustering technology. The key risk features are extracted through the feature engineering technology, prediction is carried out by utilizing machine learning, and the intelligent adjustment of space-time sensitivity is combined, so that the system can dynamically respond to the regional risk change according to real-time data and the torrent outbreak prediction result, the accuracy and response speed of torrent risk prediction are greatly improved, the river basin manager is ensured to obtain more accurate and dynamic decision support in time, and the influence of torrent disasters is effectively prevented and lightened.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments described in the present application, and other drawings may be obtained according to these drawings for those skilled in the art.
FIG. 1 is a flow chart of a method for dynamically estimating parameters of a basin hydrological model based on a digital twin technology.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments may be embodied in many different forms and should not be construed as limited to the examples set forth herein, but rather, the example embodiments are provided so that this disclosure will be more thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art.
The invention provides a dynamic estimation method of parameters of a basin hydrological model based on a digital twin technology as shown in fig. 1, which comprises the following steps:
Comprehensively modeling hydrologic, meteorological and geographic features of the river basin through multi-mode data (such as rainfall, water level, flow, air temperature, soil humidity and the like) and topography and remote sensing data in the river basin, wherein the multi-mode data comprises real-time hydrologic data and meteorological data;
The steps of comprehensively modeling hydrologic, meteorological and geographic features of the river basin through multi-mode data (such as rainfall, water level, flow, air temperature, soil humidity and the like) and terrain and remote sensing data in the river basin comprise the following steps:
first, data from different sources, including real-time hydrologic data (e.g., rainfall, flow, water level, etc.), meteorological data (e.g., temperature, humidity, wind speed, etc.), and topographical information (e.g., grade, altitude, river basin morphology, etc.), and remote sensing image data (e.g., land use type, vegetation coverage, etc.), are collected and integrated.
Next, the data is pre-processed, including denoising, normalization, and spatio-temporal alignment, to ensure quality and consistency of the data.
Then, the river basin is geographically partitioned by a space analysis method (such as GIS technology), and the river basin is divided into a plurality of subareas with similar hydrological characteristics.
On the basis, various data are analyzed and feature extracted by adopting a statistical method, a multi-dimensional model capable of reflecting hydrologic, meteorological and geographic features in a flow field is constructed, and comprehensive modeling of hydrologic, meteorological and geographic features is realized by a multi-mode data fusion technology.
The modeling process provides a scientific basis for subsequent risk assessment, disaster prediction and emergency response.
Based on topographic information (such as gradient, river basin shape and the like) and meteorological data (such as precipitation, temperature change and the like), a spatial clustering algorithm (such as K-means and DBSCAN) is used for partitioning the river basin, and a mountain torrent prone area is found;
Based on topographic information (e.g., slope, drainage basin morphology, etc.) and meteorological data (e.g., precipitation, temperature changes, etc.), the specific steps of using spatial clustering algorithms (e.g., K-means, DBSCAN) to partition the drainage basin include:
First, the terrain and weather data in the domain are collected and integrated, which are converted into feature vectors suitable for processing by a clustering algorithm. For example, data such as gradient, altitude, river basin shape, precipitation, temperature, etc. are used as the characteristic inputs.
Then, a suitable clustering algorithm (such as K-means, DBSCAN, etc.) is selected for cluster analysis. The K-means algorithm divides the river basin into K sub-areas through iteration, the characteristics of each sub-area are similar, the DBSCAN identifies the mountain torrent prone area through density, and the high risk area with dense rainfall and special topography characteristics can be automatically detected. The clustering results will reveal different regions within the flow domain and specifically identify those regions highly correlated to the onset of mountain floods, i.e., mountain floods prone regions.
Finally, the clustering result is displayed through a visualization tool (such as GIS software), the multi-mode data fusion technology is combined, hydrologic, meteorological and geographic features are integrated comprehensively, an omnibearing drainage basin model is built, and accurate data support is provided for subsequent prediction and decision.
The process classifies regions with similar geographic and meteorological features in the flow domain through spatial clustering, accurately identifies the mountain torrent easy-occurrence region and provides support for disaster early warning and resource scheduling.
For each mountain torrent prone area, extracting key features reflecting the potential risk of mountain torrent in the area by utilizing the acquired multi-mode data, carrying out deep analysis and processing on the extracted features by a feature engineering technology, inputting the processed features as feature vectors into a machine learning model which is trained and put into use, and predicting the potential mountain torrent;
and for each mountain torrent prone region, extracting key features reflecting the potential risk of mountain torrent in the region by utilizing the acquired multi-mode data, wherein the extracted features comprise rising and falling amplitude of the water level and the frequency and amplitude of flow change in the flow domain, carrying out deep analysis and processing on the rising and falling amplitude of the water level and the frequency and amplitude of flow change in the flow domain by a characteristic engineering technology in a monitoring period, respectively generating a water level fluctuation reference value and a flow fluctuation reference value, inputting the water level fluctuation reference value and the flow fluctuation reference value as feature vectors into a machine learning model which is trained and put into use, and predicting the potential mountain torrent by using the mountain torrent risk index generated by the machine learning model.
The machine learning model after training is a process of learning and optimizing historical data through a machine learning algorithm, and the model automatically identifies nonlinear or complex relations between input features (such as water level fluctuation, flow fluctuation and the like) and mountain torrents and floods from the data in the process. The core purpose of the training process is to find potential modes in the data, and gradually optimize the parameters of the model through the technologies of back propagation (in a neural network), gradient descent (in a regression model) and the like, so that the prediction of the input characteristics is more accurate.
Taking a neural network as an example, in the training process, original hydrologic data (including historical water level data, flow data, rainfall and the like) is input into the network. Through repeated forward propagation and backward propagation, the network continuously adjusts the weight and bias in the network until the model can accurately capture the complex relationship between the characteristics such as water level fluctuation, flow variation and the like and the torrential flood outbreak. After training is completed, the model is tested through a verification set (namely data which is not used for training), and the accuracy and the robustness of the model in practical application are evaluated, so that the model can effectively cope with different hydrologic and meteorological conditions.
The trained model not only can generate a prediction result according to historical data, but also can rapidly respond to input real-time data to predict the possibility of mountain torrent outbreak. This enables the model to be used in real-time applications as a real-time early warning system. In this step, the model is not just a theoretical tool, but is proved by practice to have the capability of continuously and effectively predicting in a complex and dynamic environment. The trained model is continuously optimized and adjusted by using a back propagation algorithm and different evaluation indexes (such as accuracy, recall rate, F1 value and the like), and finally, the precise prediction of the torrent risk is realized.
The machine learning model put into use means that the model has completed training and has been applied in the actual environment, and can receive new data in real time and generate corresponding prediction results. This stage is critical in the prediction of torrent outbreaks because it ensures that the machine learning model can process real-time data in real scenes and provides a practically effective support for torrent monitoring and early warning.
For machine learning models for mountain torrent risk prediction, the adoption means seamless engagement of the model in data flow, model calculation and decision support systems. The machine learning model inputs the multi-mode data such as water level, flow, rainfall and the like into the trained model through real-time monitoring, and a prediction result about the torrent risk is generated. For example, the model calculates the "mountain torrent risk index" of the area according to the input water level fluctuation reference value and the flow fluctuation reference value by using the patterns learned in the previous training, and provides the early warning information to the river basin manager based on the mountain torrent risk index.
This process involves a "prediction-feedback-optimization" loop. Meanwhile, in practical application, a river basin manager can feed back the prediction result of the model, and the feedback data can continuously supplement a training data set, so that the model is further optimized. As real-time data is continuously input, the model gradually improves the accuracy of its predictions. The feedback learning not only enables the model to keep high efficiency, but also can cope with uncertainty caused by climate, topography and other changes, and ensures long-term effectiveness of the model.
In practical use, the investment of the machine learning model also means that the machine learning model can be integrated into an automatic early warning system to work together with other monitoring devices (such as weather radar, remote sensing satellites and the like). For example, when the water level of a mountain torrent easy-to-send area exceeds a certain threshold, the model can automatically trigger early warning, and risk warning information is sent to related personnel through short messages, mails, APP pushing and other modes. This allows for a more rapid and efficient early warning and emergency response of the flood burst.
Furthermore, the adoption of machine learning models also emphasizes their suitability for real-time data. The model can adapt to various changes under various different climatic conditions and topography conditions to generate accurate predictions. This capability makes the machine learning model an indispensable tool in disaster prevention and reduction of flood peak, not just a static analysis tool, but a dynamic, continuously optimized decision support system.
The magnitude of the rise and fall of the water level within the flow area is an important indicator for measuring the potential risk of flood peak. In general, when the water level variation amplitude is large, this means that the precipitation amount suddenly increases or the water in the basin rapidly gathers, resulting in a rapid rise in the water level. Such drastic changes often occur under extreme weather conditions, especially in short periods of heavy rain or concentrated precipitation, in which case the drainage capacity of the basin may be inadequate and mountain floods are extremely likely to occur. In addition, the rapid drop of the water level may reflect the rapid loss of the water flow after the mountain floods are exploded, and the hydrologic condition in the river basin is changed severely, which indicates that the water resources in the river basin have large fluctuation in a short time, and the mountain floods may be initiated. Thus, a region with a large water level fluctuation range generally means a region with a higher risk of flood peak, while a region with a smaller water level fluctuation range may mean a region with a lower risk of flood peak. By monitoring the trend of the water level change, the potential flood peak risk can be effectively predicted and estimated.
For each mountain torrent plot, in the monitoring period, the rising and falling amplitude of the water level in the flow area is subjected to deep analysis and processing by the characteristic engineering technology, and the specific steps for generating the water level fluctuation reference value are as follows:
Firstly, the water level fluctuation sequence needs to be processed in a segmentation mode according to the time stamp, and the trend of water level change in each time period is calculated. Specifically, the change characteristics of the water level in each short time window are analyzed by calculation of the local slope. For each time window (I.e., a selected set of watershed water level data over a time horizon), a water level change amplification factor is introduced to identify the severity of the water level change over each small time window. The formula is:
Wherein: Indicating the water level change amplification factor, Is the water level value corresponding to the j-th time stamp,Is a parameter for controlling the sensitivity of the fluctuation amplitude of the water level, controls the weighting degree of the change,Is a time windowThe number of data points within the set,Is the water level value at the previous point in time. Water level change amplification factorFor quantifying the intensity of the fluctuation of the water level in a short time window;
The effect of this step is to highlight local sharp fluctuations by the adaptive amplification factor, preserving the nonlinear fluctuation characteristics of the data.
After the calculation of the water level fluctuation amplification in each time window is completed, the water level fluctuation reference value is further generated through the local abnormality measurementFor measuring the deviation of the water level variation in each time window relative to the overall trend, especially the sudden event of rapid rise or fall, local abnormality measurementThe definition is as follows:
Wherein: is a predicted water level value obtained using a locally weighted regression model such as gaussian weighted regression, Is a parameter controlling sensitivity of the anomaly measure;
By this formula, the degree of abnormality of the water level fluctuation can be quantified, considering whether the variation of the water level in a certain time window deviates significantly from the overall trend. Finally, the water level fluctuation reference value is obtained by combining and weighting the abnormal measurement of each time window with the water level change amplification factor, and the expression of the water level fluctuation reference value is as follows: Wherein: Is a reference value for the fluctuation of the water level, Is a weight factor for weighting fluctuation conditions of the respective time windows, i represents an i-th time window during monitoring, i.e., the basin water level data is divided into a plurality of time windows during monitoring, m represents the total number of divided time windows during the entire monitoring, i.e., the entire monitoring period is divided into m time windows;
the function of this step is to comprehensively evaluate the intensity and degree of abnormality of the water level fluctuation in different time windows, thereby obtaining an integral water level fluctuation reference value with early warning capability. The water level fluctuation reference value can accurately reflect the severe fluctuation of the water level in the flow field and is further used for evaluating the potential risk of torrential flood outbreak.
According to the water level fluctuation reference value, for each mountain torrent plot, in the monitoring period, the larger the expression value of the water level fluctuation reference value generated by carrying out deep analysis and processing on the rising and falling amplitude of the water level in the flow area through the characteristic engineering technology is, the larger the risk of potential mountain torrent in the mountain torrent plot is indicated. The water level fluctuation reference value is obtained by performing depth analysis on the rising and falling amplitude of the water level in the flow area, and comprehensively reflects the degree of drastic change of the water level in a short time. When the water level rises or falls rapidly, this usually means a sharp increase in precipitation or a sharp pooling of water in the basin, such a sharp water level fluctuation being a sign of a torrential flood. In mountain floods prone areas, an increase in the water level fluctuation reference typically indicates a rapid change in the hydrologic conditions within the basin, possibly exceeding the drainage capacity of the basin, resulting in an increased risk of mountain floods. Therefore, the larger the expression value of the water level fluctuation reference value is, the more severe the water level fluctuation is, and the higher the possibility of the flood is, otherwise, if the water level fluctuation reference value is smaller, the water level change is stable, and the potential risk of the flood is lower.
For each torrential flood plot, a greater frequency and magnitude of flow changes within the stream generally indicates that the plot is at a higher risk of torrential flood outbreaks. The reasons for frequent and large flow changes are often related to sudden strong precipitation, mountain terrain features or other factors (such as soil saturation, river blockage, etc.). In mountain torrents in areas where there is a tendency for precipitation to be high, the water flow is likely to quickly sink into streams or channels, resulting in a rapid rise in flow. This rapid flow change reflects the speed and intensity of the response of the basin to precipitation. Frequent and severe flow fluctuations often indicate poor drainage in this area, and the accumulation and rapid flow of water may exceed the capacity of natural or artificial drainage facilities, thereby creating conditions for torrential flood outbreaks. Therefore, the high frequency and the high amplitude of the flow change are an important index of the occurrence of the torrential flood in the torrential flood prone area.
For each mountain torrent plot, the specific steps of generating a flow variation reference value by carrying out deep analysis and processing on the frequency and the amplitude of the flow variation in the flow domain through the characteristic engineering technology in the monitoring period are as follows:
high-order moment analysis is performed on the flow data to reveal complex patterns and abnormal fluctuations in flow changes. High-order moment analysis focuses on the asymmetry and kurtosis (i.e., the sharpness of the fluctuation) of the flow fluctuations. In this step, the skewness is used Kurtosis ofTo capture the asymmetry and extreme fluctuation characteristics of the flow distribution. The specific formula is as follows: Wherein: For the flow rate at the time of the k-th, Is the average value of the flow rate,N is the total sample number of the data;
Skewness is used to quantify the degree of deviation of a flow change (i.e., whether the change tends to rise or fall), and kurtosis is used to quantify the sharpness of a flow fluctuation (i.e., whether there is a severe fluctuation or extreme change). High skewness and kurtosis values often represent severe fluctuations in flow, indicating a potential risk of torrential flood outbreaks;
the method is mainly used for capturing asymmetry and abnormal fluctuation information of flow fluctuation and providing basic nonlinear characteristics for flow change index calculation in the next step. This helps reveal the complexity of the flow variation, especially in the context of abrupt or abnormal fluctuations.
The complexity of the flow fluctuations is assessed by the fractal dimension. Fractal dimension is used to describe the self-similarity and complexity of flow curves, and is a method for measuring flow fluctuation details. The fractal dimension of the flow curve can be calculated by the following formula: Wherein: Representing the dimension of the fractal dimension, Expressed in scaleThe coverage of the lower flow curve, i.e. the level of detail of the flow change,Is a scale parameter. In particular the number of the elements,Is the number of divisions of the time series of the flow at different scales, along withThe diameter of the pipe is gradually reduced,Changes in (c) reflect the complexity of the flow fluctuations. If the fractal dimension is high, it is indicated that the flow fluctuation has high complexity, and a strong hydrologic reaction and a potential torrential flood risk can be indicated;
The nonlinear complexity of the flow fluctuation can be revealed through calculation of the fractal dimension, and the accurate description of the flow change is further enhanced. The high fractal dimension means that the flow variation is more complex and irregular, and is generally closely related to the risk of sudden torrential flood outbreaks.
Degree of combined deflectionKurtosis degreeFractal dimensionCalculating a flow variation reference value, the calculated expression being: Wherein: and (3) representing a flow variation reference value, and comprehensively measuring the variation characteristics of the flow by combining three important characteristics of skewness, kurtosis and fractal dimension. Degree of deviation Kurtosis ofTogether reflect the asymmetry and extreme fluctuation of flow fluctuation, fractal dimensionThe complexity of the flow fluctuations is quantified. Finally, the flow rate variation reference valueThe degree and complexity of flow fluctuation can be quantified, so that an effective prediction basis is provided for the potential risk of torrential flood outbreak;
Flow rate variation reference value The method can effectively aggregate the multidimensional flow fluctuation characteristics into a digital risk assessment index, and provides accurate quantitative data support for mountain torrent risk prediction by quantifying the asymmetry, extreme fluctuation and complexity of the flow.
As can be seen from the flow fluctuation reference value, the larger the expression value of the flow fluctuation reference value generated by deep analysis and processing of the frequency and amplitude of the flow change in the flow domain by the characteristic engineering technology in the monitoring period, the more severe and complex the flow fluctuation in the flow domain is, and the larger the hydrologic response is usually accompanied, which is particularly important in the mountain torrent region. Frequent changes in flow and large fluctuations in magnitude generally indicate a high probability of extreme precipitation or other extreme climatic conditions occurring in the area in a short period of time, and an increased risk of torrential flood outbreaks. The flow variation reference value extracted by the feature engineering technology can quantify the intensity and complexity of these fluctuations. If the flow fluctuation reference value is higher, the hydrologic characteristics of the mountain torrent easy-to-develop area show irregular and severe fluctuation, which is usually caused by superposition effects of factors such as sudden precipitation, upstream water source injection, topography characteristics and the like, and the factors greatly promote the potential risk of mountain torrent. Conversely, if the flow variation reference value is lower, the flow variation reference value indicates that the hydrologic fluctuation in the flow field is smoother, the influence of the extreme weather event on the area is smaller, and the potential torrential flood risk is lower. Therefore, the flow variation reference value not only can be used as an effective quantification standard, but also can provide scientific basis for predicting the torrential flood risk.
The machine learning model is not particularly limited herein, and can realize the water level fluctuation reference valueAnd a flow rate variation reference valueComprehensive analysis is carried out to generate mountain torrent risk indexThe invention provides a specific implementation mode for realizing the technical scheme of the invention, namely the mountain torrent risk indexThe generated calculation formula is as follows: In which, in the process, 、Respectively the water level fluctuation reference valueAnd a flow rate variation reference valueIs a preset proportionality coefficient of (1), and、Are all greater than 0. Preset proportional coefficient [ ]And) Refers to that when the mountain torrent risk index is generated, the reference value of the water level fluctuation is used for the generation of the mountain torrent risk index) And flow variation reference value) A coefficient weighting the risk of torrential flood outbreak. The preset scaling factors have the function of setting the contribution ratio of different indexes to final risk assessment according to an actual hydrological model or historical data. In particular the number of the elements,Is the water level fluctuation reference value) Impact weight on the risk of torrential flood,Is the flow variation reference value) Impact weight on torrent risk. In practical application, the preset proportionality coefficients help the model dynamically adjust the influence degree on risk prediction according to the change of the water level and the flow, and ensure that the output of the model (namely, the mountain torrent risk index) can accurately reflect the relative contribution of different hydrologic characteristics to the mountain torrent risk. Thus, the first and second substrates are bonded together,AndThe model is a parameter obtained by fitting according to historical data or field experience, and the prediction accuracy of the model is usually ensured by tuning.
According to the mountain torrent risk indexes, for each mountain torrent plot, in the monitoring period, the larger the expression value of the water level fluctuation reference value generated by carrying out deep analysis and processing on the rising and falling amplitude of the water level in the flow area through the characteristic engineering technology, the larger the expression value of the flow fluctuation reference value generated by carrying out deep analysis and processing on the frequency and amplitude of the flow change in the flow area through the characteristic engineering technology, namely, the larger the expression value of the mountain torrent risk index generated when the potential mountain torrent is predicted through the machine learning model which is completed through training and put into use, the larger the risk of mountain torrent is indicated to be in the mountain torrent plot, otherwise, the smaller the risk of mountain torrent is indicated to be in the mountain torrent plot.
The method comprises the following specific steps of comparing and analyzing a mountain torrent risk index generated when a potential mountain torrent is predicted by a machine learning model which is completed through training and is put into use with a preset mountain torrent risk index reference threshold value, and predicting the potential mountain torrent:
If the mountain torrent risk index is smaller than or equal to the mountain torrent risk index reference threshold, a normal state signal is generated, and the state of the mountain torrent is relatively stable, so that the risk of torrent is avoided.
When a potential flood burst risk exists in the flood burst area, the time-space sensitivity threshold of the flood burst area is intelligently adjusted by combining the real-time burst prediction result of the flood burst area and the characteristics of the flood burst prediction result along with time and space changes, and the time-space sensitivity is dynamically improved in the potential flood burst area so as to enhance the response capability to the risk change, and ensure that a river basin manager can obtain more accurate and timely decision support to cope with the potential threat caused by the flood burst;
When a potential torrent outbreak risk exists in the torrent easy-to-develop area, the time-space sensitivity threshold of the torrent easy-to-develop area is intelligently adjusted by combining the real-time outbreak prediction result of the torrent easy-to-develop area and the characteristics of the torrent easy-to-develop area along with time and space changes, and the specific steps are as follows:
Firstly, calculating a space-time sensitivity threshold of a mountain torrent prone region, wherein the space-time sensitivity threshold is dynamically adjusted according to a mountain torrent risk index and a mountain torrent risk index reference threshold, and the space-time sensitivity threshold is calculated according to the following formula: Wherein, the method comprises the steps of, wherein, As a threshold value for the temporal-spatial sensitivity,And is a mountain torrent risk index which represents the mountain torrent risk of the current river basin,The threshold is referenced for the mountain torrent risk index, typically by historical data, expert judgment or analog analysis settings,The time change rate of the mountain torrent risk index, namely the change speed of the mountain torrent risk index along with time reflects the change trend of the mountain torrent risk, especially in sudden weather or hydrologic events,AndRespectively representing the contribution weights of the current torrent risk index and the torrent risk index reference threshold value to the time space sensitivity threshold value for the preset weighting coefficient,An influence factor representing time variation, and weighing the influence degree of the time variation in the river basin on risk response;
The function of this step is to dynamically adjust the spatiotemporal sensitivity threshold based on the gap between the real-time torrent risk index and the torrent risk index reference threshold, and the rate of change of the torrent risk index over time. Dynamically boosting spatiotemporal sensitivity means that when the torrent risk index approaches or exceeds the torrent risk index reference threshold, the spatiotemporal sensitivity threshold is raised to enhance the responsiveness to changes. This adjustment ensures that the system reacts quickly when the risk is drastically changed.
After the time-space sensitivity threshold is adjusted, re-evaluating the flood burst risk of the flood burst area, determining a risk level according to the new time-space sensitivity threshold, and intelligently adjusting the time-space sensitivity threshold of the flood burst area when the potential flood burst risk exists in the flood burst area (namely, when the flood risk index is larger than the reference threshold of the flood risk index), wherein the dynamic adjustment expression of the new time-space sensitivity threshold is as follows: Wherein: is a time-space response adjustment factor used for controlling the dynamic magnification of a time-space sensitivity threshold value when the flood burst risk is changed, And in order to adjust the space-time sensitivity threshold, dynamically improving the influence based on the deviation between the torrent risk index and the torrent risk index reference threshold, and enhancing the response capability to torrent outbreak risk.
By monitoring and predicting the risk change of the mountain torrent prone area in real time, the space-time sensitivity threshold of the area is intelligently adjusted, so that the response capability of the system to the potential threat of mountain torrent is enhanced. When the risk changes severely, the system can dynamically improve the space-time sensitivity, so that the early warning of the torrential flood is more accurate and timely. Through the dynamic adjustment, a river basin manager can be ensured to quickly obtain reliable risk assessment when facing sudden weather changes or hydrologic anomalies, and timely take disaster prevention and reduction measures, so that the threat of torrential flood outbreak to life and property is reduced to the greatest extent, and the efficiency and effect of emergency response are improved. The process improves the flexibility of the early warning system, so that the early warning system can effectively cope with torrential flood outbreaks with different risk levels, and the prevention and control measures are ensured to be effectively executed at the most critical moment.
The dynamic estimation method of the parameters of the basin hydrologic model based on the digital twin technology can effectively solve the problem that the model in the prior art can not adjust weight in time and miss the early warning of torrent. The method establishes an accurate drainage basin model by integrating multi-mode data (such as rainfall, water level, flow and the like) and geographic features, and identifies a mountain torrent prone area by adopting a spatial clustering technology. The key risk features are extracted through a feature engineering technology, prediction is carried out by utilizing machine learning, and the system can dynamically respond to the regional risk change according to real-time data and the torrent outbreak prediction result by combining with intelligent adjustment of space-time sensitivity. The method and the system greatly improve the accuracy and response speed of the torrential flood risk prediction, ensure that a river basin manager can timely obtain more accurate and dynamic decision support, and effectively prevent and reduce the influence of torrential flood disasters.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas with a large amount of data collected for software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
While certain exemplary embodiments of the present invention have been described above by way of illustration only, it will be apparent to those of ordinary skill in the art that modifications may be made to the described embodiments in various different ways without departing from the spirit and scope of the invention. Accordingly, the drawings and description are to be regarded as illustrative in nature and not as restrictive of the scope of the invention, which is defined by the appended claims.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510258551.6A CN119761265B (en) | 2025-03-06 | 2025-03-06 | Dynamic estimation method for parameters of drainage basin hydrologic model based on digital twin technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510258551.6A CN119761265B (en) | 2025-03-06 | 2025-03-06 | Dynamic estimation method for parameters of drainage basin hydrologic model based on digital twin technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119761265A true CN119761265A (en) | 2025-04-04 |
| CN119761265B CN119761265B (en) | 2025-06-03 |
Family
ID=95196541
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510258551.6A Active CN119761265B (en) | 2025-03-06 | 2025-03-06 | Dynamic estimation method for parameters of drainage basin hydrologic model based on digital twin technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119761265B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120086718A (en) * | 2025-04-30 | 2025-06-03 | 水发规划设计有限公司 | A watershed digital twin construction method and system based on multi-source data fusion |
| CN120126300A (en) * | 2025-05-14 | 2025-06-10 | 长沙理工大学 | Highway engineering disaster monitoring and early warning system based on machine learning |
| CN120387684A (en) * | 2025-06-30 | 2025-07-29 | 广东省科学院广州地理研究所 | Watershed ecological risk monitoring method, device, electronic device and storage medium |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150019262A1 (en) * | 2013-07-11 | 2015-01-15 | Corelogic Solutions, Llc | Method and system for generating a flash flood risk score |
| CN106803223A (en) * | 2017-01-17 | 2017-06-06 | 中国水利水电科学研究院 | A method of mountain torrent disaster risk assessment based on polymorphic system theory |
| CN109785586A (en) * | 2019-03-04 | 2019-05-21 | 四川大学 | Early warning method of mountain torrent disaster zone in small watershed in mountainous area |
| US20190354873A1 (en) * | 2018-02-16 | 2019-11-21 | Lucas Pescarmona | Analysis system and hydrology management for basin rivers |
| CN113222283A (en) * | 2021-05-31 | 2021-08-06 | 中国水利水电科学研究院 | Mountain torrent forecasting and early warning method and system based on digital twin |
| WO2023030513A1 (en) * | 2021-09-05 | 2023-03-09 | 汉熵通信有限公司 | Internet of things system |
| CN117057508A (en) * | 2023-08-24 | 2023-11-14 | 广州市城市规划勘测设计研究院 | A flash flood and debris flow disaster risk identification method, device, terminal and medium |
| US20230400603A1 (en) * | 2022-06-09 | 2023-12-14 | Wuhan University | Method of projecting future flood risks under changing hydrological cycles |
| CN117727158A (en) * | 2023-12-07 | 2024-03-19 | 华海智汇技术有限公司 | Mountain torrent monitoring and early warning method and device based on sound characteristics |
| CN118587030A (en) * | 2024-08-06 | 2024-09-03 | 安徽金海迪尔信息技术有限责任公司 | A method of irrigation area allocation based on big data |
| US20240403514A1 (en) * | 2023-06-02 | 2024-12-05 | Dalian University Of Technology | Method, system and device for allocating total emission reduction of water environmental pollution load |
| US20250060512A1 (en) * | 2023-08-17 | 2025-02-20 | Mcmaster University | Rapid Deep Learning-Based Flood Losses/Risk Prediction Tool, Methods of Making and Uses Thereof |
-
2025
- 2025-03-06 CN CN202510258551.6A patent/CN119761265B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150019262A1 (en) * | 2013-07-11 | 2015-01-15 | Corelogic Solutions, Llc | Method and system for generating a flash flood risk score |
| CN106803223A (en) * | 2017-01-17 | 2017-06-06 | 中国水利水电科学研究院 | A method of mountain torrent disaster risk assessment based on polymorphic system theory |
| US20190354873A1 (en) * | 2018-02-16 | 2019-11-21 | Lucas Pescarmona | Analysis system and hydrology management for basin rivers |
| CN109785586A (en) * | 2019-03-04 | 2019-05-21 | 四川大学 | Early warning method of mountain torrent disaster zone in small watershed in mountainous area |
| CN113222283A (en) * | 2021-05-31 | 2021-08-06 | 中国水利水电科学研究院 | Mountain torrent forecasting and early warning method and system based on digital twin |
| WO2023030513A1 (en) * | 2021-09-05 | 2023-03-09 | 汉熵通信有限公司 | Internet of things system |
| US20230400603A1 (en) * | 2022-06-09 | 2023-12-14 | Wuhan University | Method of projecting future flood risks under changing hydrological cycles |
| US20240403514A1 (en) * | 2023-06-02 | 2024-12-05 | Dalian University Of Technology | Method, system and device for allocating total emission reduction of water environmental pollution load |
| US20250060512A1 (en) * | 2023-08-17 | 2025-02-20 | Mcmaster University | Rapid Deep Learning-Based Flood Losses/Risk Prediction Tool, Methods of Making and Uses Thereof |
| CN117057508A (en) * | 2023-08-24 | 2023-11-14 | 广州市城市规划勘测设计研究院 | A flash flood and debris flow disaster risk identification method, device, terminal and medium |
| CN117727158A (en) * | 2023-12-07 | 2024-03-19 | 华海智汇技术有限公司 | Mountain torrent monitoring and early warning method and device based on sound characteristics |
| CN118587030A (en) * | 2024-08-06 | 2024-09-03 | 安徽金海迪尔信息技术有限责任公司 | A method of irrigation area allocation based on big data |
Non-Patent Citations (2)
| Title |
|---|
| ZHEWEI LIU ET AL.: "Interpretable machine learning for predicting urban flash flood hotspots using intertwined land and built-environment features", 《COMPUTERS, ENVIRONMENT AND URBAN SYSTEMS》, vol. 110, 30 June 2024 (2024-06-30) * |
| 孙京东;万金红;张葆蔚;马建明;李云鹏;: "四川绵竹"8・13"山洪泥石流灾害调查", 人民珠江, no. 01, 25 February 2015 (2015-02-25) * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120086718A (en) * | 2025-04-30 | 2025-06-03 | 水发规划设计有限公司 | A watershed digital twin construction method and system based on multi-source data fusion |
| CN120126300A (en) * | 2025-05-14 | 2025-06-10 | 长沙理工大学 | Highway engineering disaster monitoring and early warning system based on machine learning |
| CN120387684A (en) * | 2025-06-30 | 2025-07-29 | 广东省科学院广州地理研究所 | Watershed ecological risk monitoring method, device, electronic device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119761265B (en) | 2025-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN119761265B (en) | Dynamic estimation method for parameters of drainage basin hydrologic model based on digital twin technology | |
| US11238356B2 (en) | Method of predicting streamflow data | |
| CN108961688B (en) | Geological disaster monitoring and early warning method under support of big data | |
| Zhao et al. | Optimally integrating multi-source products for improving long series precipitation precision by using machine learning methods | |
| CN119418508B (en) | Real-time river water level monitoring and early warning method and system based on deep learning | |
| CN108375808A (en) | Dense fog forecasting procedures of the NRIET based on machine learning | |
| CN114611778B (en) | Reservoir water level early warning method and system based on warehousing flow | |
| CN118822042B (en) | A data-driven real-time flood forecasting method and system | |
| CN119047852B (en) | A method and system for predicting rainstorm disaster risk | |
| CN120450432B (en) | A flood control risk identification and early warning system and method for reservoir groups | |
| CN120526540B (en) | Intelligent risk perception and multi-stage progressive early warning system for flood storage and detention areas | |
| CN119559770A (en) | A water level monitoring and management method based on multi-source information fusion | |
| CN119091069B (en) | Three-dimensional geographic target construction analysis system based on neural network model | |
| CN117787726A (en) | A road natural disaster risk assessment method based on supervised learning | |
| CN120354787B (en) | A river water regime forecasting method and system based on digital twin | |
| CN119647935B (en) | A regional geological disaster risk intelligent assessment method and system | |
| CN119559764A (en) | A dynamic early warning method for debris flow in mountainous areas based on deep learning | |
| CN120913366A (en) | Storm lightning disaster risk early warning method and system based on analytic hierarchy process | |
| CN117608005B (en) | A method and system for studying and judging the frequency of regional heavy rains | |
| CN118245930A (en) | Hydrologic event liquid level prediction method, hydrologic event liquid level prediction device, storage medium and computer equipment | |
| CN120450452B (en) | Hydrological weather coupling forecast early warning system | |
| CN120085395B (en) | Method and system for forecasting sandstorm areas | |
| CN119066563B (en) | Intelligent monitoring system of secondary flood and sediment based on multi-source data | |
| CN120598341B (en) | Sea level rising influence risk assessment method, medium and system based on multi-source data fusion | |
| CN120726765B (en) | A method and device for coordinated flood warning issuance between upstream and downstream areas |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |