Hydraulic engineering data intelligent monitoring method and system based on digital twinning

Abstract

The invention relates to the technical field of engineering data monitoring, in particular to a hydraulic engineering data intelligent monitoring method and system based on digital twinning, and the method comprises the following steps: collecting real-time operation data through physical sensors disposed at different nodes of a hydraulic engineering; a self-adaptive digital twinborn model capable of being automatically adjusted is generated through a meta-learning method, and the self-adaptive digital twinborn model dynamically adjusts the internal structure of the self-adaptive digital twinborn model according to the change of real-time data; and on the basis of the adaptive digital twinborn model, constructing a simulation module of a plurality of target scenes, and performing decision simulation in a virtual environment in real time. According to the method, the accuracy and the stability of the digital twinborn model in a complex environment are improved, the model can quickly adapt to real-time changes in the operation process of a water conservancy project through the self-adaptive mechanism, and the prediction precision and the response of the model in a dynamic complex scene are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering data monitoring, and in particular to an intelligent monitoring method and system for water conservancy engineering data based on digital twins. Background Art

[0002] With the continuous expansion of water conservancy project scale and increasingly complex management needs, how to efficiently and accurately monitor and manage the operation status of water conservancy projects has become an important issue that needs to be solved in the water conservancy industry. The existing water conservancy project monitoring system mainly relies on physical sensors and manual inspections. This traditional method has the following limitations:

[0003] Incomplete data collection: Traditional monitoring systems have a single data source and mainly rely on limited physical sensors to obtain basic data such as water level, flow velocity, and flow. They lack detailed monitoring of dynamically changing areas and are unable to fully reflect the actual operating status of water conservancy projects.

[0004] The model lacks adaptability: Most of the existing water conservancy project simulation models are based on static modeling with fixed parameters. They are unable to adjust their structures according to the dynamic changes of real-time data, resulting in reduced prediction accuracy of the model in emergencies (such as floods and climate change).

[0005] Lack of simulation verification and decision support: The current monitoring system lacks simulation capabilities for different operating strategies. Managers cannot predict the effects and risks of decisions in a virtual environment. Direct implementation of decisions can easily lead to waste of resources, increased overflow risks, or poor irrigation results.

[0006] Existing digital twin models are unable to dynamically adjust their internal structures. For example, they fail to improve calculation accuracy in areas with sudden changes in water levels or flow velocities, and are unable to accurately capture dynamic changes in high-risk areas.

[0007] Lack of simulation verification and feedback mechanism: The existing methods have not realized the simulation function based on multi-objective scenarios, and cannot predict the effects and risks of different operation strategies in scenarios such as flood control and irrigation, which limits the scientific nature of decision-making. Summary of the invention

[0008] The present invention provides a method and system for intelligent monitoring of water conservancy project data based on digital twins.

[0009] The intelligent monitoring method of water conservancy project data based on digital twins includes the following steps:

[0010] S1, data collection: real-time operation data including water level, flow velocity, flow rate, and meteorological parameters are collected through physical sensors deployed at different nodes of the water conservancy project;

[0011] S2, Adaptive digital twin model construction: Based on real-time operation data combined with historical data, an automatically adjustable adaptive digital twin model is generated through meta-learning methods. The adaptive digital twin model dynamically adjusts its internal structure according to the changes in real-time data, including:

[0012] a): Dynamically increase the grid resolution in areas where water level and flow velocity suddenly change;

[0013] b): Dynamically adjust data weights based on the timeliness and reliability of data sources, focus on high-value data, and reduce noise interference;

[0014] S3, Simulation: Based on the adaptive digital twin model, construct simulation modules for multiple target scenarios, including flood control and irrigation. By setting different operation strategies, real-time decision-making simulation is performed in a virtual environment to evaluate the effectiveness and risks of each strategy and predict the target results under different target scenarios.

[0015] Optionally, the physical sensor includes:

[0016] Water level sensors are deployed at reservoirs, rivers, and dam nodes to collect water level data in real time;

[0017] Flow velocity sensors are deployed in rivers and channels to collect water velocity data in real time to reflect the flow status of water;

[0018] Flow sensors are deployed at water flow control nodes in sluice gates and pipelines to collect water flow data in real time and monitor the delivery of water resources;

[0019] The meteorological data interface connects to the open data interface of the meteorological department to obtain meteorological data in the water conservancy project area, including rainfall parameters.

[0020] Optionally, the construction of the adaptive digital twin model in S2 specifically includes:

[0021] S21, real-time operation data D t and historical data D h The fusion is performed and the data is normalized, which is expressed as: D = {d s,t}∪{d s,h},s=1,2,…,n, where D represents the fused data set, d s,t represents the real-time data collected by the sth sensor at time t, d s,h represents the recorded data of the sth sensor at a historical moment, n represents the total number of sensors, and the data is normalized to ensure that the dimensions of different data are consistent;

[0022] S22, feature extraction: based on water level, flow velocity, flow and meteorological parameters, the key features of the data are extracted through time series analysis methods, including trend changes, periodic patterns and abnormal fluctuation characteristics;

[0023] S23, meta-learning model training: Using the extracted key features, a meta-learning method is used to train the initial digital twin model.

[0024] Optionally, the feature extraction in S22 specifically includes:

[0025] S221, Trend characteristics: Calculate the long-term trend of data by moving average method;

[0026] S222, periodic features: extracting periodic features through Fourier transform;

[0027] S223, Abnormal Fluctuation Characteristics: Detect outliers through the local anomaly factor method.

[0028] Optionally, the meta-learning method is used to train the initial adaptive digital twin model, with the goal of training a model M with fast generalization capability. θ , including:

[0029] Meta-learning optimization goal: optimize the initial parameters θ by minimizing the loss function L of the model in multiple tasks;

[0030] Model output: Through the trained meta-learning model, the real-time running data D t As input, it outputs the parameters and status of the adaptive digital twin model: in, represents the model after meta-learning optimization, and S(t) is the dynamic state of the adaptive digital twin model at time t, including water level, flow velocity, and flow prediction values.

[0031] Optionally, the dynamically increasing the grid resolution in the area where the water level or flow velocity changes suddenly includes:

[0032] By monitoring the changes in water level and flow rate data in real time, the change rate R is calculated. w (t) and R v (t), and identify the mutation area, and dynamically increase the grid resolution when the following conditions are met:

[0033] Among them, R w (t) is the water level change rate, W(t) is the water level data at time t, R v (t) is the velocity change rate, V(t) is the velocity data at time t, Δt is the time interval, δ w , δ v is the threshold for water level and flow rate changes.

[0034] Optionally, the dynamically adjusting the data weight based on the timeliness and reliability of the data source specifically includes:

[0035] The data weight is dynamically updated based on the timeliness and reliability of the data source, and the data source weight ω of the i-th group of data is calculated i (t): Among them, ω i (t) represents the weight of the i-th group of data at time t, α i represents the initial weight of the i-th group of data, which is obtained based on historical data evaluation, R i (t) represents the reliability of the i-th group of data, which is obtained based on the transmission delay and accuracy of the data, and n is the total number of data groups.

[0036] Optionally, the S3 specifically includes:

[0037] S31, target scenario definition: Based on the operation requirements of the water conservancy project, set the target scenario of the simulation module, including:

[0038] Flood prevention scenario: With water level monitoring data as the core, set the upper water level threshold W max , simulate the water level distribution and flow changes when floods come;

[0039] Irrigation scenario: Based on flow data and soil moisture requirements, define the irrigation water threshold Q target Simulate water resource allocation and soil moistening effects under different irrigation strategies;

[0040] S32, Operation strategy setting: Set a variety of operation strategies for different target scenarios, including:

[0041] In the flood prevention scenario, the effects of different flood discharge strategies are evaluated by controlling the opening degree of the sluice gate O(t);

[0042] In the irrigation scenario, by adjusting the irrigation water flow Q in and irrigation duration T irrigate Evaluate water use efficiency and soil moisture distribution effects on crops;

[0043] S33, decision simulation and evaluation: Different operation strategies are input into the adaptive digital twin model, and the effects of each strategy are simulated in real time based on the simulation module, including:

[0044] Flood control scenario: Based on the simulation output of the water level distribution W(t,x,y) under different strategies combined with the water level upper limit threshold W max , calculate the overflow risk R flood , W(t,x,y) represents the water level at the coordinate position (x,y) at time t;

[0045] Irrigation scenario: Irrigation water volume Q based on simulation output used , soil moisture distribution M(t,x,y), calculate water resource utilization efficiency E irrigation , M(t,x,y) represents the soil moisture at the coordinate position (x,y) at time t;

[0046] S34, result prediction and risk assessment: Evaluate the effectiveness and risks of each operation strategy through the simulation output results, and form the target results.

[0047] Optionally, the target result includes:

[0048] In flood prevention scenarios, generate overflow risk reports and optimal flood discharge strategies;

[0049] In irrigation scenarios, water resource utilization efficiency reports and irrigation plans are generated.

[0050] The water conservancy project data intelligent monitoring system based on digital twins is used to implement the above-mentioned water conservancy project data intelligent monitoring method based on digital twins, and includes the following modules:

[0051] Data acquisition module: used to deploy physical sensors at different nodes of the water conservancy project to collect real-time operation data, including water level, flow velocity, flow and meteorological parameters, and transmit the data to the adaptive digital twin model construction module;

[0052] Adaptive digital twin model building module: used to generate an automatically adjustable adaptive digital twin model through meta-learning method based on real-time operation data and historical data. The adaptive digital twin model can dynamically adjust the internal structure according to real-time data, including dynamically increasing the grid resolution in areas with sudden changes in water level and flow velocity, and dynamically adjusting the data weight based on the timeliness and reliability of the data source;

[0053] Simulation module: It is used to build simulation modules for multiple target scenarios based on the adaptive digital twin model, set different operation strategies, conduct decision-making simulation in a virtual environment in real time, evaluate the effects and risks of each strategy, and predict the target results under different target scenarios;

[0054] Result output module: used to output the evaluation results of simulation, including overflow risk report, optimal flood discharge strategy, water resource utilization efficiency report and irrigation plan.

[0055] Beneficial effects of the present invention:

[0056] The present invention combines real-time operation data with historical data through a meta-learning method, and the adaptive digital twin model can automatically adjust its internal structure according to the dynamic changes of the data, dynamically increase the grid resolution in areas of sudden changes such as water level and flow rate, and effectively refine the simulation accuracy of high-risk areas, ensuring more accurate prediction and simulation of emergencies (such as floods). The data weight is dynamically adjusted based on the timeliness and reliability of the data source, the contribution of high-value data is enhanced, the impact of noise data is reduced, and the accuracy and stability of the digital twin model in complex environments are improved. This adaptive mechanism enables the model to quickly adapt to real-time changes in the operation of water conservancy projects, and improves the prediction accuracy and response speed of the model in dynamic and complex scenarios.

[0057] Based on the adaptive digital twin model, the present invention constructs simulation modules for target scenarios such as flood control and irrigation, and performs real-time decision-making simulation by setting different operation strategies to achieve:

[0058] Flood prevention scenario: Through real-time simulation of water level and flow velocity distribution, the overflow risk is dynamically calculated, and an overflow risk report and optimal flood discharge strategy are generated to minimize the overflow risk and effectively ensure the safety of water conservancy projects;

[0059] Irrigation scenario: Based on the simulation results of irrigation water volume and soil moisture distribution, the water resource utilization efficiency is quantified, and the optimal irrigation plan is output to achieve efficient allocation and utilization of water resources.

[0060] This simulation verification mechanism avoids the waste of resources and potential risks caused by direct operations, allowing managers to evaluate the effectiveness and risks of various strategies in advance in a virtual environment, and ultimately make scientific and reasonable decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 A schematic diagram of a monitoring method flow chart of an embodiment of the present invention;

[0063] Figure 2 Schematic diagram of the monitoring system functional modules according to an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0065] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0066] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0067] like Figure 1 As shown in the figure, the intelligent monitoring method of water conservancy project data based on digital twins includes the following steps:

[0068] S1, data collection: real-time operation data including water level, flow velocity, flow, and meteorological parameters are collected through physical sensors deployed at different nodes of the water conservancy project;

[0069] S2, Adaptive digital twin model construction: Based on real-time operation data combined with historical data, an automatically adjustable adaptive digital twin model is generated through meta-learning methods. The adaptive digital twin model dynamically adjusts its internal structure according to the changes in real-time data, including:

[0070] a): Dynamically increase the grid resolution in areas where water level and flow velocity suddenly change;

[0071] b): Dynamically adjust data weights based on the timeliness and reliability of data sources, focus on high-value data, and reduce noise interference;

[0072] S3, Simulation: Based on the adaptive digital twin model, construct simulation modules for multiple target scenarios, including flood control and irrigation. By setting different operation strategies, real-time decision-making simulation is performed in a virtual environment to evaluate the effectiveness and risks of each strategy and predict the target results under different target scenarios.

[0073] Physical sensors include:

[0074] Water level sensors are deployed at reservoirs, rivers, and dam nodes to collect water level data in real time;

[0075] Flow velocity sensors are deployed in rivers and channels to collect water velocity data in real time to reflect the flow status of water;

[0076] Flow sensors are deployed at water flow control nodes in sluice gates and pipelines to collect water flow data in real time and monitor the delivery of water resources;

[0077] The meteorological data interface connects to the open data interface of the meteorological department to obtain meteorological data in the water conservancy project area, including rainfall parameters.

[0078] The construction of the adaptive digital twin model in S2 specifically includes:

[0079] S21, real-time operation data D t and historical data D h The fusion is performed and the data is normalized, which is expressed as: D = {d s,t}∪{d s,h},s=1,2,…,n, where D represents the fused data set, d s,t represents the real-time data collected by the sth sensor at time t, d s,h represents the recorded data of the sth sensor at a historical moment, n represents the total number of sensors, and the data is normalized to ensure that the dimensions of different data are consistent;

[0080] S22, feature extraction: based on water level, flow velocity, flow and meteorological parameters, the key features of the data are extracted through time series analysis methods, including trend changes, periodic patterns and abnormal fluctuation characteristics;

[0081] S23, meta-learning model training: Using the extracted key features, the meta-learning method is used to train the initial digital twin model so that it has the ability to quickly adapt to different operating scenarios. Through the rapid generalization characteristics of meta-learning and combined with real-time input data, the parameters and structure of the digital twin model are dynamically updated to improve the accuracy and adaptability of the model in complex environments.

[0082] The feature extraction in S22 specifically includes:

[0083] S221, Trend characteristics: Calculate the long-term trend of data by sliding average method:

[0084] Among them, T i (t) represents the trend characteristics of the i-th group of data at time t, k represents the size of the sliding window, and d s,t represents the real-time data collected by the sth sensor at time t, d s,t The whole represents the data point;

[0085] S222, Periodic Features: Extracting Periodic Features by Fourier Transform:

[0086] Among them, P i (f) represents the power spectral density of the i-th group of data at frequency f, N represents the total number of time points, j is an imaginary unit, d s,t represents the real-time data collected by the sth sensor at time t, e -j2πft / N is the complex exponential term in Fourier transform;

[0087] S223, abnormal fluctuation characteristics: Detecting outliers by local anomaly factor method:

[0088] Among them, LOF(d s,t ) represents the data point d s,t The local anomaly factor indicates the degree of abnormality of the data point relative to its neighborhood data. The larger the value, the higher the probability that the data point is an outlier. k (d s,t ) represents the data point d s,t The k nearest neighbor set, i.e., the k data points closest to |N k (d s,t )| represents the data point d i,t The number of k nearest neighbors is equal to k, lrd(d s,t ) represents the data point d s,t The local reachable density of .

[0089] The meta-learning method is used to train the initial adaptive digital twin model, with the goal of training a model M with fast generalization ability. θ , including:

[0090] Meta-learning optimization goal: Optimize the initial parameter θ by minimizing the loss function L of the model in multiple tasks, expressed as: Among them, θ * represents the optimized model parameters, D u represents the data set of the u-th task, p(D) represents the data distribution of the task, L is the loss function of the task, and the mean square error loss function is used: y i represents the true value, represents the model prediction value;

[0091] Model output: Through the trained meta-learning model, the real-time running data D t As input, the parameters and status of the adaptive digital twin model are output: in, represents the model after meta-learning optimization, S(t) is the dynamic state of the adaptive digital twin model at time t, including the water level, flow velocity, and flow prediction values;

[0092] Through the above data fusion, feature extraction and meta-learning model training, real-time operation data is mapped with historical data to form an automatically adjustable adaptive digital twin model, whose core structure is:

[0093] M θ (D) = f(T, P, LOF), where M θ For the generated digital twin model, T is the trend characteristic of the data, P is the periodic characteristic, and LOF is the abnormal fluctuation characteristic.

[0094] Dynamically increase the grid resolution in areas with sudden changes in water level and flow velocity, including:

[0095] By monitoring the changes in water level and flow rate data in real time, calculate the change rate R w (t) and R v (t), and identify the mutation area, and dynamically increase the grid resolution when the following conditions are met:

[0096] Among them, R w (t) is the water level change rate, W(t) is the water level data at time t, R v (t) is the velocity change rate, V(t) is the velocity data at time t, Δt is the time interval, δ w , δ v It is the threshold of water level and flow rate change, which is preset as the standard for triggering dynamic adjustment. When the above conditions are met, the grid resolution of the mutation area is automatically increased, and high-precision simulation and prediction are achieved by refining the grid units.

[0097] Dynamically adjust data weights based on the timeliness and reliability of data sources, including:

[0098] The data weight is dynamically updated based on the timeliness and reliability of the data source, and the data source weight ω of the i-th group of data is calculated i (t): Among them, ω i (t) represents the weight of the i-th group of data at time t, α irepresents the initial weight of the i-th group of data, which is obtained based on historical data evaluation, R i (t) represents the reliability of the i-th group of data, which is obtained based on the transmission delay and accuracy of the data. n is the total number of data groups. By dynamically adjusting the weights, the contribution of high-value data (such as high-timeliness, low-noise data) is increased, and the interference of low-reliability data on the model is reduced.

[0099] S3 specifically includes:

[0100] S31, target scenario definition: Based on the operation requirements of the water conservancy project, set the target scenario of the simulation module, including:

[0101] Flood prevention scenario: With water level monitoring data as the core, set the upper water level threshold W max , simulate the water level distribution and flow changes when floods come;

[0102] Irrigation scenario: Based on flow data and soil moisture requirements, define the irrigation water threshold Q target Simulate water resource allocation and soil moistening effects under different irrigation strategies;

[0103] S32, operation strategy setting: Set a variety of operation strategies for different target scenarios, including:

[0104] In the flood prevention scenario, the effects of different flood discharge strategies are evaluated by controlling the opening degree of the sluice gate O(t);

[0105] In the irrigation scenario, by adjusting the irrigation water flow Q in and irrigation duration T irrigate Evaluate water use efficiency and soil moisture distribution effects on crops;

[0106] S33, decision simulation and evaluation: Different operation strategies are input into the adaptive digital twin model, and the effects of each strategy are simulated in real time based on the simulation module, including:

[0107] Flood control scenario: Based on the simulation output of the water level distribution W(t,x,y) under different strategies combined with the water level upper limit threshold W max , calculate the overflow risk R flood , W(t,x,y) represents the water level at the coordinate position (x,y) at time t;

[0108] Irrigation scenario: Irrigation water volume Q based on simulation output used , soil moisture distribution M(t,x,y), calculate water resource utilization efficiency E irrigation , M(t,x,y) represents the soil moisture at the coordinate position (x,y) at time t;

[0109] S34, result prediction and risk assessment: Evaluate the effectiveness and risks of each operation strategy through the simulation output results, and form the target results.

[0110] Target outcomes include:

[0111] In the flood control scenario, an overflow risk report and optimal flood discharge strategy are generated. The specific calculation is:

[0112] Overflow risk R flood calculate: Among them, [z] + It means that when z>0, z is taken, otherwise it is taken as 0, z=W(t,x,y)-W max , A region is the total area of ​​the calculation region, R flood Indicates the overall overflow risk value of the region. The larger the value, the higher the risk.

[0113] Optimal flood discharge strategy generation: Through simulation of different flood discharge strategies, the strategy with the least overflow risk is selected: opt =argmin O(t) R flood (O(t)), where O(t) is the degree of opening of the sluice gate at time t, R flood (O(t)) represents the overflow risk value corresponding to the flood discharge strategy O(t), O opt This is the optimal flood discharge strategy, indicating the degree of sluice gate opening with the lowest risk;

[0114] Overflow risk report generation: Generate an overflow risk report based on simulation results, including visualization of water level distribution and high-risk areas, and overflow risk value R flood .

[0115] In the irrigation scenario, a water resource utilization efficiency report and irrigation plan are generated. The specific calculation is:

[0116] Water resource utilization efficiency irrigation calculate:

[0117] Among them, M target (x, y) is the target soil moisture distribution, M initial (x,y) represents the soil moisture distribution before irrigation, Q used is the actual amount of water consumed for irrigation, A region is the total area of ​​the irrigated area, E irrigation It indicates the contribution rate of unit water volume to soil moisture. The higher the value, the higher the efficiency.

[0118] Optimal irrigation plan generation: By simulating different irrigation strategies, the plan with the highest water resource utilization efficiency is selected: Among them, Q in is the input irrigation water flow, T irrigate is the irrigation duration, E irrigation (Q in |T irrigate ) represents the corresponding water resource utilization efficiency, Q opt It is the optimal irrigation plan, indicating the most efficient water volume and duration configuration plan;

[0119] Water resource utilization efficiency report generation: Generate a water resource utilization efficiency report based on simulation results, including visualization results of irrigation water volume and humidity distribution, water resource utilization efficiency E irrigation , optimal irrigation solution Q opt .

[0120] like Figure 2 As shown, the water conservancy project data intelligent monitoring system based on digital twins is used to implement the above-mentioned water conservancy project data intelligent monitoring method based on digital twins, and includes the following modules:

[0121] Data acquisition module: used to deploy physical sensors at different nodes of the water conservancy project to collect real-time operation data, including water level, flow velocity, flow and meteorological parameters, and transmit the data to the adaptive digital twin model construction module;

[0122] Adaptive digital twin model building module: used to generate an automatically adjustable adaptive digital twin model through meta-learning methods based on real-time operation data and historical data. The adaptive digital twin model can dynamically adjust the internal structure according to real-time data, including dynamically increasing the grid resolution in areas with sudden changes in water level and flow velocity, and dynamically adjusting the data weight based on the timeliness and reliability of the data source;

[0123] Simulation module: It is used to build simulation modules for multiple target scenarios based on the adaptive digital twin model, set different operation strategies, conduct decision-making simulation in a virtual environment in real time, evaluate the effects and risks of each strategy, and predict the target results under different target scenarios;

[0124] Result output module: used to output the evaluation results of simulation, including overflow risk report, optimal flood discharge strategy, water resource utilization efficiency report and irrigation plan, to provide support for the actual decision-making of water conservancy projects.

[0125] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0126] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. The intelligent monitoring method of water conservancy project data based on digital twin is characterized by: The following steps are involved: S1, data collection: real-time operation data including water level, flow velocity, flow rate, and meteorological parameters are collected through physical sensors deployed at different nodes of the water conservancy project; S2, Adaptive digital twin model construction: Based on real-time operation data combined with historical data, an automatically adjustable adaptive digital twin model is generated through meta-learning methods. The adaptive digital twin model dynamically adjusts its internal structure according to the changes in real-time data, including: a): Dynamically increase the grid resolution in areas where water level and flow velocity suddenly change; b): Dynamically adjust data weights based on the timeliness and reliability of data sources; S3, Simulation: Based on the adaptive digital twin model, construct simulation modules for multiple target scenarios, including flood control and irrigation. By setting different operation strategies, real-time decision-making simulation is performed in a virtual environment to evaluate the effectiveness and risks of each strategy and predict the target results under different target scenarios.

2. The method for intelligent monitoring of water conservancy project data based on digital twins according to claim 1 is characterized in that: The physical sensor comprises: Water level sensors are deployed at reservoirs, rivers, and dam nodes to collect water level data in real time; Flow velocity sensors are deployed in rivers and channels to collect water velocity data in real time to reflect the flow status of water; Flow sensors are deployed at water flow control nodes in sluice gates and pipelines to collect water flow data in real time and monitor the delivery of water resources; The meteorological data interface connects to the open data interface of the meteorological department to obtain meteorological data in the water conservancy project area, including rainfall parameters.

3. The method for intelligent monitoring of water conservancy project data based on digital twins according to claim 1 is characterized in that: The construction of the adaptive digital twin model in S2 specifically includes: S21, real-time operation data D t and historical data D h The fusion is performed and the data is normalized, which is expressed as: D = {d s,t }∪{d s,h },s=1,2,…,n, where D represents the fused data set, d s,t represents the real-time data collected by the sth sensor at time t, d s,h represents the recorded data of the sth sensor at a historical moment, n represents the total number of sensors, and the data is normalized to ensure that the dimensions of different data are consistent; S22, feature extraction: based on water level, flow velocity, flow and meteorological parameters, the key features of the data are extracted through time series analysis methods, including trend changes, periodic patterns and abnormal fluctuation characteristics; S23, meta-learning model training: Using the extracted key features, a meta-learning method is used to train the initial digital twin model.

4. The method for intelligent monitoring of water conservancy project data based on digital twins according to claim 3 is characterized in that: The feature extraction in S22 specifically includes: S221, Trend characteristics: Calculate the long-term trend of data by moving average method; S222, periodic features: extracting periodic features through Fourier transform; S223, Abnormal Fluctuation Characteristics: Detect outliers through the local anomaly factor method.

5. The method for intelligent monitoring of water conservancy project data based on digital twins according to claim 4 is characterized in that: The meta-learning method is used to train the initial adaptive digital twin model, with the goal of training a model M with fast generalization ability. θ , specifically including: Meta-learning optimization goal: optimize the initial parameters θ by minimizing the loss function L of the model in multiple tasks; Model output: Through the trained meta-learning model, the real-time running data D t As input, the parameters and status of the adaptive digital twin model are output: in, represents the model after meta-learning optimization, and S(t) is the dynamic state of the adaptive digital twin model at time t, including water level, flow velocity, and flow prediction values.

6. The method for intelligent monitoring of water conservancy project data based on digital twins according to claim 1 is characterized in that: The dynamic increase of the grid resolution in the area of ​​water level and flow velocity mutation specifically includes: By monitoring the changes in water level and flow rate data in real time, the change rate R is calculated. w (t) and R v (t), and identify the mutation area, and dynamically increase the grid resolution when the following conditions are met: Among them, R w (t) is the water level change rate, W(t) is the water level data at time t, R v (t) is the velocity change rate, V(t) is the velocity data at time t, Δt is the time interval, δ w , δ v is the threshold for water level and flow rate changes.

7. The method for intelligent monitoring of water conservancy project data based on digital twins according to claim 1 is characterized in that: The dynamic adjustment of data weight based on the timeliness and reliability of the data source specifically includes: The data weight is dynamically updated based on the timeliness and reliability of the data source, and the data source weight ω of the i-th group of data is calculated i (t): Among them, ω i (t) represents the weight of the i-th group of data at time t, α i represents the initial weight of the i-th group of data, which is obtained based on historical data evaluation, R i (t) represents the reliability of the i-th group of data, which is obtained based on the transmission delay and accuracy of the data, and n is the total number of data groups.

8. The method for intelligent monitoring of water conservancy project data based on digital twins according to claim 1 is characterized in that: The S3 specifically includes: S31, target scenario definition: Based on the operation requirements of the water conservancy project, set the target scenario of the simulation module, including: Flood prevention scenario: With water level monitoring data as the core, set the upper water level threshold W max , simulate the water level distribution and flow changes when floods come; Irrigation scenario: Based on flow data and soil moisture requirements, define the irrigation water threshold Q target Simulate water resource allocation and soil moistening effects under different irrigation strategies; S32, Operation strategy setting: Set a variety of operation strategies for different target scenarios, including: In the flood prevention scenario, the effects of different flood discharge strategies are evaluated by controlling the opening degree of the sluice gate O(t); In the irrigation scenario, by adjusting the irrigation water flow Q in and irrigation duration T irrigate Evaluate water use efficiency and soil moisture distribution effects on crops; S33, decision simulation and evaluation: Different operation strategies are input into the adaptive digital twin model, and the effects of each strategy are simulated in real time based on the simulation module, including: Flood control scenario: Based on the simulation output of the water level distribution W(t,x,y) under different strategies combined with the water level upper limit threshold W max , calculate the overflow risk R flood , W(t,x,y) represents the water level at the coordinate position (x,y) at time t; Irrigation scenario: Irrigation water volume Q based on simulation output used , soil moisture distribution M(t,x,y), calculate water resource utilization efficiency E irrigation , M(t,x,y) represents the soil moisture at the coordinate position (x,y) at time t; S34, result prediction and risk assessment: Evaluate the effectiveness and risks of each operation strategy through the simulation output results, and form the target results.

9. The method for intelligent monitoring of water conservancy project data based on digital twins according to claim 8 is characterized in that: The target outcomes include: In flood prevention scenarios, generate overflow risk reports and optimal flood discharge strategies; In irrigation scenarios, water resource utilization efficiency reports and irrigation plans are generated.

10. A water conservancy project data intelligent monitoring system based on digital twins, used to implement the water conservancy project data intelligent monitoring method based on digital twins as described in any one of claims 1 to 9, characterized in that: Includes the following modules: Data acquisition module: used to deploy physical sensors at different nodes of the water conservancy project to collect real-time operation data, including water level, flow velocity, flow and meteorological parameters, and transmit the data to the adaptive digital twin model construction module; Adaptive digital twin model building module: used to generate an automatically adjustable adaptive digital twin model through meta-learning method based on real-time operation data and historical data. The adaptive digital twin model can dynamically adjust the internal structure according to real-time data, including dynamically increasing the grid resolution in areas with sudden changes in water level and flow velocity, and dynamically adjusting the data weight based on the timeliness and reliability of the data source; Simulation module: It is used to build simulation modules for multiple target scenarios based on the adaptive digital twin model, set different operation strategies, conduct decision-making simulation in a virtual environment in real time, evaluate the effects and risks of each strategy, and predict the target results under different target scenarios; Result output module: used to output the evaluation results of simulation, including overflow risk report, optimal flood discharge strategy, water resource utilization efficiency report and irrigation plan.

Images