Method and system for coupling simulation of ice condition and water quality in freezing period of Yellow River

Abstract

The invention discloses a Yellow River ice condition and water quality coupling simulation method and system, and relates to the field of hydraulic engineering simulation, and the method comprises the following components: S1, physical entity layer construction, S2, data acquisition and transmission, S3, digital twinborn layer construction, and S4, application interaction. The physical entity layer comprising a hydraulic mechanical entity, an ice condition physical field and a water quality physical field is constructed, data are collected in real time in combination with multiple types of sensors, encryption transmission is performed by using a 5G / Beidou dual-mode communication technology, the real-time performance and accuracy of the data are ensured, and a digital twinborn body layer is provided with an ice condition physical field and a water quality physical field. A Kalman filtering algorithm is adopted to carry out fusion processing on data, data noise is effectively eliminated, a standardized model input data set is generated, particularly, dynamic coupling simulation is introduced, and dynamic correlation of hydraulic machinery operation parameters, ice condition parameters and water quality parameters is achieved by establishing a coupling correlation matrix; and the precision and reliability of ice condition and water quality simulation are obviously improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water conservancy simulation, in particular to a method and system for simulating ice conditions and water quality coupling during the ice sealing period of the Yellow River. BACKGROUND

[0002] The Yellow River, as the second longest river in China, has a vast watershed and complex geographical and climatic conditions. During the ice sealing period, the dynamic changes of ice conditions and water quality have a significant impact on water resource management, flood control and disaster reduction, and ecological environment protection in the coastal areas. During the ice sealing period, thick ice covers some sections of the Yellow River, which not only changes the water flow state and affects the operating efficiency of hydraulic machinery, but also may cause disasters such as ice jams and ice dams. At the same time, the changes in water quality under the ice layer are directly related to the sustainable use of water resources.

[0003] Traditional methods for simulating ice conditions and water quality during the ice sealing period of the Yellow River often have problems such as incomplete data acquisition, low simulation accuracy, and weak dynamic correlation. On the one hand, traditional methods rely heavily on offline monitoring data and empirical formulas, making it difficult to reflect the dynamic changes of ice conditions and water quality in real time, especially under extreme weather conditions. The real-time and accuracy of the data cannot be guaranteed. On the other hand, when dealing with the complex coupling relationship between hydraulic machinery operation, ice evolution, and water quality diffusion, traditional models often use simplified assumptions or single-factor analysis, ignoring the interaction between multiple factors, resulting in a large deviation between the simulation results and the actual situation. For example, in the simulation of ice evolution, traditional models may only consider the influence of water temperature on ice growth, while ignoring the comprehensive effect of hydraulic machinery disturbance and water quality thermodynamic characteristics on ice conditions, thus reducing the accuracy and reliability of the simulation.

[0004] In view of the shortcomings of traditional technology in simulating ice conditions and water quality during the ice sealing period of the Yellow River, the present application proposes a method and system for simulating ice conditions and water quality coupling during the ice sealing period of the Yellow River, which is particularly important. SUMMARY

[0005] The purpose of the present application is to overcome the shortcomings of the prior art and provide a method and system for simulating ice conditions and water quality coupling during the ice sealing period of the Yellow River. It can build a physical entity layer containing hydraulic machinery entities, ice physical fields, and water quality physical fields, combine real-time data collection from multiple types of sensors, and use advanced 5G / Beidou dual-mode communication technology for encrypted transmission to ensure the real-time and accuracy of the data. In the digital twin layer, Kalman filtering algorithm is used to fuse and process the data, effectively eliminating data noise, and introducing dynamic coupling simulation to realize the dynamic correlation of hydraulic machinery operating parameters, ice parameters, and water quality parameters, significantly improving the accuracy and reliability of the simulation.

[0006] The application provides the following technical solutions to solve the above technical problems: on the one hand, a method for simulating ice conditions and water quality coupling in the ice sealing period of the Yellow River, the specific steps of the method are as follows:

[0007] S1, physical entity layer construction: including water power mechanical entities, ice condition physical fields and water quality physical fields in the target monitoring area in the ice sealing period of the Yellow River; the water power mechanical entities are used for adjusting the water flow state in the target area, and the running state directly affects the ice condition evolution and water quality diffusion;

[0008] S2, data acquisition and transmission step: including multiple types of sensors and data transmission steps;

[0009] The sensors include: running parameter sensors installed on the water power mechanical entities, ice condition monitoring sensors arranged in the ice condition physical fields, and water quality monitoring sensors arranged in the water quality physical fields; the data transmission step adopts 5G / Beidou dual-mode communication technology, and encrypts the real-time data collected by the sensors and transmits the real-time data to the digital twin body layer, and simultaneously receives the simulation control instructions fed back by the digital twin body layer;

[0010] S3, digital twin body layer: including a three-dimensional fine model library, a data fusion engine, a dynamic coupling simulation and a twin body updating engine;

[0011] The three-dimensional fine model library includes: a three-dimensional fine model of water power machinery (constructed based on BIM+GIS technology, restoring the geometric structure, motion component correlation and running physical characteristics of the water power machinery), an ice condition evolution digital model (constructed based on hydrodynamics-thermodynamics coupling equations, including ice layer formation, growth, ablation and motion sub-models), and a water quality concentration field digital model (constructed based on convection-diffusion equations, considering the influence of ice layer coverage on the diffusion coefficient of water body, including pollutant migration and transformation sub-models);

[0012] The data fusion engine adopts Kalman filtering algorithm to fuse the real-time monitoring data transmitted by the data acquisition and transmission step and the initial parameters of the model, eliminate data noise, and generate a standardized data set for model input;

[0013] The dynamic coupling simulation realizes the dynamic correlation of the water power machinery running parameters, the ice condition parameters and the water quality parameters by establishing a coupling correlation matrix: when the water power machinery running parameters change, the water flow disturbance intensity is calculated through the hydrodynamic equation, which is used as the boundary condition of the ice condition evolution digital model to update the ice thickness and ice speed simulation results; during the ice condition evolution process, the ice layer thickness and ice speed change affect the convection-diffusion coefficient in the water quality concentration field digital model by changing the water body surface area and water flow resistance coefficient, and then the pollutant concentration distribution simulation results are updated; at the same time, the water quality parameters affect the ice layer growth / ablation rate calculation of the ice condition evolution digital model through the influence on the thermodynamic characteristics of the water body;

[0014] The twin updating engine adopts a multi-dimensional deviation weighted correction algorithm, dynamically corrects each model parameter in the three-dimensional fine model library according to the deviation between the real-time data of the data acquisition and transmission step and the dynamic coupling simulation result, and realizes the synchronous updating of the physical entity-digital twin.

[0015] S4, an application interaction step: including a visual interaction step and a decision support step;

[0016] The visual interaction step constructs a three-dimensional visualization interface based on the WebGL technology, intuitively displays the coupling simulation result of the digital twin layer, supports the user to perform zooming, rotating and cross-section cutting operations on the hydraulic machinery three-dimensional model, the ice evolution process and the water quality concentration field, and view the parameter values of any spatial point in real time; the decision support step internally embeds multi-scene decision algorithms, generates the hydraulic machinery scheduling suggestion, the ice disaster prevention and control early warning and the water quality improvement scheme according to the dynamic coupling simulation result, and feeds back to the physical entity layer construction through the data transmission step.

[0017] Further, the data transmission step in S2 adopts an adaptive redundant encryption transmission protocol, which combines the advantages of 5G and Beidou communication technology, designs a dynamic redundancy strategy and a multi-layer encryption mechanism, the dynamic redundancy strategy adjusts the data redundancy according to the real-time communication quality, the communication quality is evaluated by the actual transmission rate and the actual transmission delay, the ratio of the actual transmission rate to the maximum transmission rate of the equipment accounts for 60% of the communication quality evaluation, and the ratio of the actual transmission delay to the maximum allowed delay (set to 500 milliseconds) accounts for 40%; when the communication quality index is not less than 0.8 (communication is good), the data redundancy is set to 1, that is, no data redundancy transmission is performed; when the communication quality index is between 0.5 and 0.8 (communication is general), the redundancy is set to 2, and two copies of the same data are transmitted; when the communication quality index is less than 0.5 (communication is poor), the redundancy is set to 3, and three copies of the same data are transmitted, the dynamic adjustment of the redundancy ensures that the data can be stably transmitted in different communication environments, avoids data loss, and the multi-layer encryption mechanism guarantees data security from the data acquisition end, the transmission process and the receiving end, the data acquisition end adopts the AES-256 symmetric encryption algorithm, the encryption key is dynamically generated by the system background and updated every 24 hours; the IPsec protocol is used to tunnel encrypt the data packet in the transmission process to prevent the data from being illegally intercepted in the middle of the transmission; the receiving end uses the RSA-4096 asymmetric encryption algorithm to verify the data integrity and ensure that the data has not been tampered with in the transmission process, which can effectively solve the problems of unstable communication and high data transmission security risk in remote sections of the Yellow River, and guarantee the safety and reliability of the monitoring data.

[0018] Further, the data fusion engine in S3 adopts a dynamic weight Kalman filter fusion algorithm. The algorithm introduces a data reliability weight coefficient to solve the poor robustness of traditional Kalman filter to abnormal data, according to the characteristics of strong data volatility during the ice sealing period of the Yellow River. The specific formula is as follows: wherein is the data after fusion at k time, Φ k is a state transition matrix, which is obtained by training based on historical data during the ice sealing period of the Yellow River, and is obtained by fitting the monitoring data of a certain section of the Yellow River during the ice sealing period in the past five years by using the least square method. k The mean value is 1.02, the standard deviation is 0.03, K k is the Kalman gain, Z k is the original data of the sensor at k time, H k is an observation matrix, ω k is a data reliability weight coefficient, and α is a weight distribution coefficient, which is 0.6. Through 100 comparative experiments, it is determined that the fusion error is the smallest when α = 0.6, σ ref is a rated standard deviation of the sensor, σ k is the standard deviation of the data at k time, which is calculated by using a sliding window method, and the window size is set to 5 groups of data. valid,k is the effective data amount in the sliding window at k time, N total,k is the total data amount in the window; P k is a covariance matrix, Q k is a process noise matrix, which is determined based on the intensity of environmental disturbance in historical data. Q k is 0.002 under the low-temperature environment during the ice sealing period. The algorithm dynamically adjusts ω k so that the error of the fused data is controlled within 5%, which is more than 15% higher than the precision of the traditional Kalman filter, and effectively eliminates the influence of sensor data volatility caused by low temperature and ice layer impact during the ice sealing period of the Yellow River.

[0019] Further, the ice condition evolution digital model in the dynamic coupling simulation in S3 adopts a multi-factor coupled ice condition evolution equation, which integrates the influence of water mechanical disturbance and water quality thermodynamic characteristics on ice condition, breaks through the limitation of traditional single evolution model considering only water temperature, and the specific formula is as follows: wherein is the change rate of ice layer thickness, h is the current ice layer thickness; β1, β2, β3 are coupling coefficients, which are determined by using multiple linear regression analysis based on the measured data during the ice sealing period from 2018 to 2023 in the section from Liujia Gorge to Qingtong Gorge on the upper reaches of the Yellow River, β1 = 0.8, β2 = 0.3, β3 = 0.5; Q m is the actual flow of water mechanical disturbance, Q0 is the design flow of the section, T0 is the freezing point temperature, T w is the water temperature, and Cw C represents the actual concentration of pollutants in the water body. ref v represents the baseline pollutant concentration for this river section. m The water flow velocity is calculated from the operating parameters of the hydraulic machinery. This equation is obtained through... This item reflects the disturbance of ice layer growth caused by changes in hydraulic mechanical flow rate, through... The item reflects the impact of pollutant concentration on the thermal conductivity of water (increased pollutant concentration reduces the thermal conductivity of water and slows down ice growth), through v m The equation reflects the scouring effect of water flow shear force on the ice layer. Tests show that the simulation error of the ice layer thickness is ≤3cm, which is 20% more accurate than the traditional model. It can accurately reflect the comprehensive impact of hydraulic machinery and water quality on the ice conditions during the Yellow River's ice-covered period.

[0020] Furthermore, the digital model of water quality concentration field in S3 adopts a pollutant diffusion correction equation under ice constraint, introducing the dual constraints of ice coverage and thickness to solve the problem that traditional models ignore the influence of ice on diffusion. The specific formula is as follows: Where C is the pollutant concentration, t is time, and v is the water flow velocity vector. The gradient represents the pollutant concentration; D0 is the water diffusion coefficient without ice; γ is the influence coefficient of ice thickness, with a value of 0.7, determined by comparing the diffusion coefficients under different ice thicknesses using measured data from the Kaifeng section of the lower Yellow River during the ice-covered period; h is the current ice thickness. max The maximum ice thickness in the history of this river section is represented by η, which is the ice coverage rate, ranging from 0 to 1. The data is obtained by drone aerial photography combined with image recognition technology and is updated every 2 hours. Here, S is the Laplace operator for pollutant concentration; S is the pollutant source and sink term. This equation is derived through... The term reflects the weakening effect of ice thickness on the diffusion coefficient (the thicker the ice, the smaller the diffusion coefficient), and the term (1-η) reflects the limitation of the diffusion space by the ice coverage area (the higher the coverage, the smaller the effective diffusion space). In the application of a section of the Yellow River, the simulation deviation of CODcr concentration by this equation is ≤8%, which is 12 percentage points lower than the deviation of the traditional model (ignoring the influence of ice), and can accurately simulate the migration and diffusion law of pollutants during the ice-covered period.

[0021] Furthermore, the twin update engine in S3 employs a multi-dimensional bias weighted correction algorithm to construct a parameter sensitivity weight matrix, enabling differentiated correction of different model parameters and avoiding model inaccuracies caused by traditional single correction methods. The specific correction process and formula are as follows: First, calculate the simulation bias of each model parameter: ΔP i =|P sim,i -P meas,i | / P meas,i ×100%, where ΔP iP represents the relative deviation of the i-th parameter. sim,i P is the simulated value of this parameter. meas,i These are measured values; the parameters include hydraulic machinery rotation speed (i=1), ice thickness (i=2), ice velocity (i=3), and pollutant concentration (i=4); then, a parameter sensitivity weight matrix is ​​constructed: Where S i Let i be the sensitivity coefficient of the i-th parameter. The sum of the sensitivity coefficients of all parameters (1.0 in this case); weight W i The sensitivity reflects the degree of influence of parameter deviation on the overall simulation results; the higher the sensitivity, the greater the weight. Next, the model parameter correction amount ΔP′ is calculated. i =ΔP i ×W i ×k、P new,i =P sim,i ±ΔP′ i , where ΔP′ i P represents the correction amount for the i-th parameter, and k is the correction coefficient, with a value of 1.2, determined through validation using 100 sets of historical data to ensure model convergence after correction; new,i The algorithm prioritizes correcting highly sensitive parameters (such as ice thickness) using a weight matrix to ensure the overall accuracy of the model. In its application at a hydropower station on the Yellow River, the data deviation after model correction was reduced from 12% to below 5%, significantly improving the synchronization between the digital twin and the physical entity.

[0022] Furthermore, the decision support step in S3 employs a multi-scenario weighted decision algorithm. This algorithm first determines the weights of the decision objectives using the analytic hierarchy process (AHP). The decision objectives include ice storm prevention, water quality assurance, and hydraulic machinery energy consumption. Five water conservancy experts are invited to score the importance of each objective. After a consistency check (consistency ratio CR = 0.08, less than 0.1, ensuring the weights are effective), the weights for ice storm prevention are determined to be 0.45, water quality assurance 0.35, and hydraulic machinery energy consumption 0.2. Next, at least three candidate decision schemes are generated for hydraulic machinery scheduling, corresponding to different gate openings. Then, each candidate scheme is further processed... The algorithm employs coupled simulations to calculate scores for each decision objective based on the simulation results. The ice disaster prevention score is calculated by combining the difference between the simulated ice thickness and the safe ice thickness; the water quality assurance score is calculated by combining the difference between the simulated pollutant concentration and the standard concentration; and the energy consumption score is calculated by combining the difference between the simulated energy consumption and the minimum energy consumption. Each objective score is out of 100. Finally, based on the weight of each objective and its corresponding score, the comprehensive score for each candidate scheme is calculated. The scheme with the highest comprehensive score is the recommended decision scheme. This algorithm comprehensively considers multiple dimensions of objectives, providing scientific and diverse decision-making suggestions for water conservancy scheduling during the Yellow River's ice-covered period, avoiding the one-sidedness caused by single-objective decision-making.

[0023] Further, the visual interaction step in S4 adopts a multi-scale dynamic rendering technology, which includes two parts of a model simplification strategy based on a level of detail and a real-time data-driven rendering mechanism. In terms of the model simplification strategy, the three-dimensional models of the hydraulic machinery, ice conditions, and water quality are divided into four levels of detail, and the switching of each level is determined according to the distance between the user's perspective and the model. When the distance between the user's perspective and the model is not more than 50 meters, the highest level of detail model is loaded, which includes detailed information such as mechanical component textures and ice layer textures. When the distance is between 50 meters and 100 meters, the second level model is loaded, which simplifies the model textures and retains key components. When the distance is between 100 meters and 200 meters, the third level model is loaded, which removes the model textures and only retains the geometric outline. When the distance exceeds 200 meters, the fourth level model is loaded, which only retains the bounding box of the model. Through this strategy, the model loading time is greatly shortened, and the real-time interaction fluency is improved. In terms of the real-time data-driven rendering mechanism, a mapping relationship between the monitoring data and the rendering parameters is established. The ice layer transparency changes with the ice thickness, and the greater the ice thickness, the higher the transparency, which intuitively shows the difference in ice thickness. The pollutant concentration is presented through color change, and the concentration from low to high corresponds to the color gradually transitioning from blue to red. The delay between data update and interface rendering is not more than 1 second, enabling users to observe the dynamic change process of ice conditions and water quality in real time and clearly.

[0024] In another aspect, a system for simulating ice conditions and water quality coupling during the ice sealing period of the Yellow River, the system comprising: a physical entity and digital model initialization module, a real-time data acquisition and transmission module, a data fusion and model input module, a dynamic coupling simulation operation module, and a twin body synchronous update and result output module;

[0025] The physical entity and digital model initialization module: determines the target monitoring area during the ice sealing period of the Yellow River, obtains the geographic information, hydraulic machinery entity parameters, initial ice condition data, and initial water quality data of the area; constructs a three-dimensional fine model of the hydraulic machinery based on BIM+GIS technology, constructs an ice condition evolution digital model based on hydrodynamic-thermodynamic coupling equations, and constructs a water quality concentration field digital model based on convection-diffusion equations; imports the above models into the three-dimensional fine model library of the digital twin body layer to complete the initialization of the digital twin body;

[0026] The real-time data acquisition and transmission module: starts the sensors of the data acquisition and transmission step, and the running parameter sensors acquire the speed, torque, and flow data of the hydraulic machinery in real time, the ice condition monitoring sensors acquire the ice thickness, ice speed, and ice layer temperature data in real time, and the water quality monitoring sensors acquire the pH value, dissolved oxygen concentration, and pollutant concentration data in real time; the data transmission step encrypts the collected real-time data and transmits it to the data fusion engine of the digital twin body layer through 5G / Beidou dual-mode communication technology;

[0027] The data fusion and model input module: the data fusion engine adopts a Kalman filtering algorithm to fuse the transmitted real-time monitoring data and the initial parameters of each model in the three-dimensional fine model library, eliminate abnormal data, and generate a standardized model input data set; the standardized data set is input into a hydraulic machinery sub-module, an ice condition evolution sub-module, and a water quality concentration field sub-module in dynamic coupling simulation;

[0028] The dynamic coupling simulation operation module: dynamic coupling simulation is based on a coupling correlation matrix to start multi-element dynamic coupling operation: the hydraulic machinery sub-module calculates the water flow velocity field and pressure field distribution through fluid mechanics equations according to the input operating parameters, and transmits the water flow disturbance intensity parameter to the ice condition evolution sub-module; the ice condition evolution sub-module takes the water flow disturbance intensity as a boundary condition, combines the input initial ice condition parameters, calculates the ice layer growth / ablation rate and ice velocity change through the hydrodynamic-thermodynamic coupling equation, generates real-time ice condition evolution results, and transmits the ice layer thickness and water flow resistance coefficient parameters to the water quality concentration field sub-module; the water quality concentration field sub-module corrects the water body diffusion coefficient according to the ice layer thickness, combines the water flow velocity field and the initial water quality parameters, calculates the pollutant migration path and concentration distribution through the convection-diffusion equation, and generates real-time water quality simulation results; at the same time, the water quality concentration field sub-module feeds back the water temperature and pollutant concentration parameters to the ice condition evolution sub-module to correct the ice layer thermodynamic property parameters, realizing closed-loop coupling operation of the three;

[0029] The twin body synchronous updating and result output module: the twin body updating engine compares the coupling simulation results and the real-time monitoring data, calculates the deviation value, and if the deviation value exceeds the preset threshold (such as ice thickness deviation > 5 cm, pollutant concentration deviation > 10%), the real-time interpolation algorithm is used to correct the parameters of the corresponding model in the three-dimensional fine model library (such as adjusting the thermodynamic coefficient of the ice condition evolution model and the diffusion coefficient of the water quality model), realizing the synchronization of the digital twin and the physical entity; the visual interactive step of the interactive step displays the coupling simulation results in a three-dimensional visual form, and the user can view the hydraulic machinery operating state, the ice condition evolution process, and the water quality concentration field distribution through the interface; the decision support step generates scheduling suggestions, warning information, and improvement schemes according to the simulation results, which are fed back to the physical entity layer construction through the data transmission step to guide the on-site operation.

[0030] Compared with the prior art, the method and system for coupling simulation of ice conditions and water quality in the Yellow River ice period have the following beneficial effects:

[0031] I, the application constructs a physical entity layer containing hydraulic machinery entities, ice conditions physical fields and water quality physical fields, combines multiple types of sensors to collect data in real time, and uses 5G / Beidou dual-mode communication technology for encrypted transmission, ensuring the real-time and accuracy of the data, in the digital twin body layer, the Kalman filter algorithm is used to fuse the data, effectively eliminating the data noise, generating a standardized model input data set, especially the introduction of dynamic coupling simulation, through the establishment of coupling correlation matrix, the dynamic correlation of hydraulic machinery operation parameters, ice condition parameters and water quality parameters is realized, which significantly improves the precision and reliability of ice condition and water quality simulation, for example, the ice condition evolution digital model uses a multi-factor coupled ice condition evolution equation, which integrates the influence of hydraulic machinery disturbance and water quality thermodynamic characteristics on ice conditions, so that the simulation error of ice thickness is controlled within 3cm, which improves the precision by 20% compared with the traditional model.

[0032] II, the system of the application not only has high-precision simulation capability, but also has built-in multi-scenario decision algorithm, which can generate hydraulic machinery scheduling suggestions, ice disaster prevention and control early warning and water quality improvement schemes according to the simulation results, through visual interactive steps, users can intuitively view the hydraulic machinery operation state, ice evolution process and water quality concentration field distribution, so as to make scientific decisions quickly, for example, in the aspect of ice disaster prevention and control, the system can calculate the prevention and control score according to the difference between the simulated ice thickness and the safe ice thickness, and generate the corresponding early warning information, which effectively enhances the emergency response capability, in addition, the application of multi-scenario weighted decision algorithm makes the decision process consider the ice disaster prevention and control, water quality guarantee and hydraulic machinery energy consumption and other multi-dimensional targets, avoiding the one-sidedness problem caused by single target decision, and providing comprehensive and scientific decision support for the water conservancy scheduling of the Yellow River ice period.

[0033] Other advantages, objects, and features of the application will be in part apparent and in part pointed out hereinafter in the specification, and in part will be observed by persons skilled in the art upon examination of the following specification, or can be learned by practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and those skilled in the art can obtain other drawings according to these drawings without creating any inventive labor.

[0035] Figure 1 It is a method flow operation diagram for ice condition and water quality coupling simulation in the ice period of the Yellow River;

[0036] Figure 2It is a system flow operation chart for ice and water quality coupling simulation in ice sealing period of the Yellow River. DETAILED DESCRIPTION

[0037] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined object of the application, the specific embodiments, structures, features and effects according to the present application are described in detail below in combination with the drawings and preferred embodiments.

[0038] Embodiment one

[0039] A high-latitude section of the Yellow River is determined as a target monitoring area. The air temperature in this area is below freezing point for a long time in winter, and the ice sealing period lasts for 4-5 months. The continuously thickening ice layer is easy to cause disasters such as ice jam and ice dam, which leads to river overflow and threatens the safety of residents along the coast. At the same time, there is an industrial park around, and the discharge of industrial wastewater may cause water pollution. Therefore, the changes of ice and water quality need to be paid attention to simultaneously. A physical entity layer is constructed in the area. The water power mechanical entity includes devices such as gates and water pumps. These devices can change the water flow velocity and flow by adjusting the opening degree and rotating speed. The running state of these devices directly determines the impact force of water flow on the ice layer and the diffusion efficiency of pollutants, and plays a key role in regulating and controlling the evolution direction of ice and the pollution range of water quality. The ice physical field covers the whole river section, which can comprehensively capture the physical characteristics such as the thickness, speed and temperature of the ice layer, and provide a real physical scene support for subsequent ice simulation, so as to avoid that the simulation results deviate too much from the actual ice conditions due to incomplete physical field coverage. The water quality physical field mainly monitors indexes such as pH value, dissolved oxygen and pollutant concentration of water body, which can reflect the pollution degree of water body in real time and provide basic data for water quality simulation, so as to ensure that the simulation results can accurately reflect the actual water quality. By constructing the three types of entities, a complete physical scene is formed, which lays a realistic foundation for subsequent data collection and simulation operation.

[0040] The running parameter sensors are installed on the water power mechanical entity to collect data such as gate opening degree, water pump rotating speed, torque and flow in real time. These data can accurately reflect the running state of the water power mechanical entity, and provide a basis for judging whether the device is working normally and whether the running parameters need to be adjusted, so as to avoid that the subsequent simulation results are distorted due to device failure or abnormal parameters. In the ice physical field, the ice monitoring sensors are evenly arranged according to the river section, nearshore and river center area to collect data such as ice layer thickness, ice speed and ice layer temperature in real time. The differences and changes of ice conditions at different positions can be comprehensively mastered, so as to prevent the local ice disaster (such as local ice layer being too thick) from being missed due to single monitoring point. In the water quality physical field, the monitoring points are set at key positions such as the vicinity of the sewage outlet, upstream and downstream of the river section to collect data such as water pH value, dissolved oxygen concentration and various pollutant concentrations in real time. The emission source and diffusion track of pollutants can be captured in time, which provides data support for tracing the pollution source and evaluating the pollution influence range.

[0041] Data transmission employs 5G / BeiDou dual-mode communication technology and utilizes an adaptive redundancy encryption transmission protocol. This protocol determines communication quality by real-time evaluation of actual transmission rate and latency, dynamically adjusting data redundancy. In severe weather conditions such as blizzards and strong winds that weaken 5G signals, increased redundancy ensures data is not lost or interrupted, guaranteeing data transmission stability and preventing disruptions to the simulation process. Regarding data security, the data acquisition end uses the AES-256 symmetric encryption algorithm, with the encryption key dynamically updated by the system backend every 24 hours, preventing data theft or tampering during acquisition. During transmission, the IPsec protocol is used for tunnel encryption of data packets, isolating external network attacks and ensuring data security during transmission, preventing the illegal use of monitoring information due to data leakage. The receiving end uses the RSA-4096 asymmetric encryption algorithm to verify data integrity, effectively identifying whether data has been tampered with during transmission, ensuring the authenticity and reliability of data entering the digital twin layer, and providing high-quality data input for subsequent data fusion and simulation operations.

[0042] The 3D high-resolution model library contains a 3D high-resolution model of hydraulic machinery, a digital model of ice condition evolution, and a digital model of water quality concentration field for this river section. The hydraulic machinery 3D high-resolution model, built using BIM+GIS technology, clearly presents the structural details and installation locations of gates and pumps, allowing simulation personnel to intuitively understand the spatial distribution and working principles of the equipment. This facilitates accurate parameter setting during simulations and avoids errors due to a lack of understanding of the equipment structure. The ice condition evolution digital model is built based on the hydrodynamic-thermodynamic coupling equation, which fully considers the combined effects of water flow dynamics and temperature on the ice layer. It accurately simulates the growth and melting processes of the ice layer under different water flow and temperature conditions, avoiding discrepancies between simulation results and actual ice conditions due to the neglect of any factor. The water quality concentration field digital model, built based on the convection-diffusion equation, accurately describes the migration and diffusion patterns of pollutants in the water body. It can predict the diffusion range and concentration changes of pollutants under different water flow conditions, providing a scientific basis for assessing pollution impacts and formulating pollution control measures. Importing these three types of models into the digital twin layer completes model initialization, providing a standardized model foundation for subsequent data fusion and coupled simulation.

[0043] The data fusion engine employs a dynamic weighted Kalman filter fusion algorithm to fuse real-time monitoring data transmitted during the data acquisition and transmission process with the initial parameters of the model, introducing a data credibility weight coefficient. The specific formula is as follows: in For the data fused at time k, Φ k Let K be the state transition matrix. k For Kalman gain, Z k H represents the raw sensor data at time k. kis the observation matrix, ω k is the data reliability weight coefficient, α is the weight distribution coefficient, σ ref is the sensor rated standard deviation, σ k is the standard deviation of data at time k, N valid,k is the amount of valid data in the sliding window at time k, N total,k is the total amount of data in the window; P k is the covariance matrix, Q k is the process noise matrix, this algorithm is aimed at the characteristics of large temperature fluctuations and unstable flow of the Yellow River during the ice period, the data reliability weight coefficient is introduced, different weights are given to the data collected by different sensors, for example, higher weight is given to the imported sensor data with high precision and good stability, and lower weight is given to the domestic sensor data which is greatly affected by the environment, through this differentiated weight distribution, the data noise can be effectively eliminated, the abnormal data (such as ice thickness abnormal data caused by the ice layer extrusion of the sensor) can be removed, and the standardized data set for model input is generated, this processing process can avoid the deviation of simulation results caused by errors in the original data, ensure the authenticity, accuracy and uniformity of the data input into the model, provide reliable data support for subsequent dynamic coupling simulation, and improve the credibility of the simulation results.

[0044] The dynamic coupling simulation realizes the dynamic correlation of the hydraulic machinery operating parameters, ice condition parameters and water quality parameters by establishing the coupling correlation matrix, breaks the limitation of independent operation of each parameter in traditional simulation, makes the simulation results more in line with the actual situation of mutual influence of each factor, when the hydraulic machinery operating parameters change, such as the increase of gate opening degree leading to the increase of flow, the water flow disturbance intensity is calculated through the hydrodynamic equation, and it is used as the boundary condition of the ice evolution digital model to update the simulation results of ice thickness and ice speed, at this time, the increase of water flow disturbance may slow down the ice layer thickening speed or even cause part of the ice layer to melt, the specific formula is as follows: wherein is the ice layer thickness change rate, h is the current ice layer thickness; β1, β2, β3 are coupling coefficients; Q m is the actual flow of hydraulic machinery, Q0 is the design flow of the river section, T0 is the freezing point temperature, T w is the water temperature, C w is the actual pollutant concentration of water body, C ref is the reference pollutant concentration of the river section, v mThe water flow velocity is the speed of the water flow, and the process can accurately reflect the direct influence of the hydraulic machinery on the ice conditions, avoiding the distortion of the ice condition simulation caused by ignoring the effect of the equipment; during the evolution of the ice conditions, the changes in the ice layer thickness and ice speed will change the surface area of the water body and the water flow resistance coefficient, thereby affecting the convection and diffusion coefficient in the water quality concentration field digital model, so that the diffusion speed and range of the pollutants change, for example, the decrease in the surface area of the water body and the increase in the water flow resistance caused by the thickening of the ice layer may reduce the diffusion speed of the pollutants, and through such associated operations, the indirect influence of the ice conditions on the water quality can be truly simulated, and the inaccuracy of the water quality simulation results caused by the separation of the ice conditions and the water quality can be prevented; at the same time, the water quality parameters such as the water temperature and the pollutant concentration will affect the thermodynamic characteristics of the water body, and react on the ice layer growth / melting rate calculation of the ice condition evolution digital model, for example, the excessively high pollutant concentration may change the specific heat capacity of the water body, affect the thermodynamic equilibrium of the ice layer, and thus change the growth or melting rate of the ice layer, and the feedback process can reflect the reverse action of the water quality on the ice conditions, so that the coupled simulation is more comprehensive and more logical.

[0045] In the simulation process, the ice condition evolution digital model adopts a multi-factor coupled ice condition evolution equation that integrates the influence of the water quality thermodynamic characteristics and the hydraulic machinery disturbance on the ice conditions, can consider the influence of multiple key factors on the ice conditions at the same time, and avoids the one-sidedness of the ice condition evolution results caused by single-factor simulation; the water quality concentration field digital model adopts a pollutant diffusion correction equation under the constraint of the ice layer, introduces the dual constraints of the ice layer coverage and thickness, can fully consider the limiting effect of the ice layer on the pollutant diffusion, and avoids the large difference between the pollutant diffusion simulation results and the actual situation caused by ignoring the influence of the ice layer, and the specific formula is as follows: where C is the pollutant concentration, t is the time, v is the water flow velocity vector, is the pollutant concentration gradient; D0 is the water body diffusion coefficient without the ice layer, γ is the ice layer thickness influence coefficient; h is the current ice layer thickness, h max is the historical maximum ice thickness of the river section, and η is the ice layer coverage; is the Laplace operator of the pollutant concentration; S is the pollutant source and sink term, and through the application of these two special equations, the precision of the ice condition and water quality coupled simulation is further improved.

[0046] The twin body updating engine adopts a multi-dimensional deviation weighted correction algorithm, compares the real-time data of the data collection and transmission steps with the dynamic coupled simulation results, calculates the deviation of each model parameter, first calculates the simulation deviation of each model parameter: ΔP i = |P sim,i -P meas,i | / P meas,i × 100%, where ΔP i is the relative deviation of the i-th parameter, P sim,i is the simulation value of the parameter, and P meas,i is the measured value; then a parameter sensitivity weight matrix is constructed: wherein S i is the sensitivity coefficient of the i-th parameter, is the sum of all parameter sensitivity coefficients; the weight W i reflects the degree of influence of the parameter deviation on the overall simulation results, the higher the sensitivity, the greater the weight; then the parameter correction amount ΔP is calculated ′ i = ΔP i × W i × k, P new,i = P sim,i ± ΔP ′ i , wherein ΔP ′ i is the correction amount of the i-th parameter, k is the correction coefficient; P new,i is the corrected parameter value, the algorithm first constructs a parameter sensitivity weight matrix to determine the degree of influence of different parameter deviations on the overall simulation results, higher weights are given to parameters with high sensitivity (such as ice thickness, pollutant concentration), and lower weights are given to parameters with low sensitivity (such as water secondary mineral content), this weight distribution can make the correction process more targeted, avoid over-correction of parameters with less influence, leading to resource waste, when the deviation value exceeds the preset threshold, calculate the parameter correction amount according to the deviation value and the weight, dynamically correct each model parameter in the three-dimensional fine model library, for example, if the ice thickness simulation value deviates too much from the measured value, calculate the correction amount according to the weight, adjust the related parameters in the ice evolution digital model, ensure that the model simulation results and the actual situation always maintain high consistency, this correction process can real-time compensate for the simulation error caused by the deviation of the initial parameter setting, environmental changes and other factors, let the digital twin layer can continuously and accurately reflect the actual situation of the physical entity layer, provide reliable simulation results for subsequent application interaction.

[0047] Based on WebGL technology to build a three-dimensional visualization interface, and using multi-scale dynamic rendering technology, this technology divides the three-dimensional models of hydraulic machinery, ice conditions, and water quality into four levels of detail, when the user views the entire river section from a global perspective, the system automatically switches to a low-detail level model, which can reduce data computation and ensure smooth interface operation, avoiding interface lag caused by too many model details; when the user focuses on a local area (such as a specific water conservancy machinery or water quality near a sewage outlet), the system switches to a high-detail level model, which can clearly show the device structure (such as the gear transmission structure of the gate) and the details of the water quality concentration distribution (such as the gradient change of the high-concentration area of pollutants), allowing users to intuitively understand the specific situation of the local area.

[0048] Meanwhile, the mapping relationship between the monitoring data and the rendering parameters is established, the ice layer transparency changes with the ice thickness, the greater the ice thickness, the higher the transparency, and the user can intuitively observe the difference in ice layer thickness at different positions through the interface, quickly identify the areas with excessively thick or thin ice layer, and provide intuitive reference for ice condition prevention and control. In addition, the user can also clearly view the running state of the hydraulic machinery (such as the gate opening degree, the water pump working state) and the distribution of the water quality concentration field (such as the position and diffusion trend of the high concentration area of pollutants), without the need to analyze through complex data reports, the overall condition of the ice condition and water quality in the monitoring area can be quickly mastered, and the information acquisition efficiency is improved.

[0049] The decision support step is built-in multi-scenario weighted decision algorithm, first, the decision target weight is determined through the analytic hierarchy process, the winter ice disaster prevention and control of this river section is crucial (the weight is set to 0.5), followed by water quality guarantee (the weight is set to 0.3), and finally the energy consumption of the hydraulic machinery (the weight is set to 0.2), this weight distribution can clearly determine the priority direction of the decision, ensure that the final scheme can meet the key requirements first, and at least three candidate decision schemes are generated for the hydraulic machinery scheduling, scheme one is to appropriately reduce the water pump flow, slow down the water flow, promote the stability of the ice layer, and reduce the risk of ice disaster, but the slow water flow may hinder the diffusion of pollutants, and aggravate local pollution; scheme two is to maintain the existing water pump flow, and adjust the gate opening degree at the same time, balance the ice condition stability and the diffusion of pollutants, reduce the risk of ice disaster while guaranteeing the normal diffusion of pollutants; scheme three is to appropriately increase the water pump flow, speed up the water flow, and promote the diffusion of pollutants, but the enhanced water flow impact may cause the ice layer to be unstable, and increase the risk of ice disaster.

[0050] According to the target weight and the score of the corresponding scheme (such as the scores of scheme two in ice disaster prevention and control, water quality guarantee, and energy consumption are 80, 85, and 90 respectively), the comprehensive score of each candidate scheme is calculated, and the scheme with the highest comprehensive score is the recommended decision scheme for execution, if the comprehensive score of scheme two is the highest, the system generates the hydraulic machinery scheduling suggestion (such as the specific gate opening degree adjustment range, the water pump flow retention value) of this scheme, to guide the staff to operate accurately; at the same time, according to the ice condition simulation result, if there is a risk of local ice layer being too thick, an ice disaster prevention and control warning is generated, prompting the relevant departments to make ice breaking preparations in advance; according to the water quality simulation result, if the concentration of pollutants in a certain area is close to the threshold, a water quality improvement scheme (such as adjusting the discharge time of the sewage outlet or increasing the sewage treatment measures) is generated, and these information is fed back to the physical entity layer through the data transmission step, to realize accurate control of the ice condition and water quality in the monitoring area, and guarantee the safe operation of the river section in winter.

[0051] Embodiment two

[0052] A certain section of the middle and lower reaches of the Yellow River is selected as the target monitoring area. In this area, the temperature rises in spring, and the ice layer gradually melts. During the melting process of the ice layer, ice flowers may be discharged, and ice floods may occur. At the same time, agricultural activities are frequent around the area, and spring irrigation and agricultural non-point source pollution may affect water quality. The physical entity layer includes hydraulic machinery entities used to regulate water flow, such as dams and water diversion gates. The operation of these devices can adjust the water level, flow direction, and flow rate of the river section, which has an important influence on the melting speed of the ice layer and the migration path of pollutants. In addition, the ice condition physical field of the river section is constructed, focusing on monitoring the thickness change, ice speed, and ice flower distribution during the melting process of the ice layer, as well as the water quality physical field, monitoring the pH value, dissolved oxygen, nitrogen and phosphorus concentration, and pesticide residue of the water body.

[0053] On the hydraulic machinery entity, install the operating parameter sensor to collect real-time data such as dam body displacement, gate opening height, water flow, and pressure of the water diversion gate, accurately grasp the operation state of the hydraulic machinery, in the ice condition physical field, arrange ice condition monitoring sensors, set monitoring points in shallow areas where the ice layer melts quickly, river bends, and river sections where ice flowers are easy to accumulate, collect real-time data such as ice layer thickness, ice speed, ice flower concentration, and ice layer temperature, track the melting dynamics of the ice layer, in the water quality physical field, arrange water quality monitoring sensors, set monitoring points at the inlet of the tributary where agricultural non-point source pollution may enter, the irrigation water discharge outlet, and the key section of the river section, collect real-time data such as water pH value, dissolved oxygen concentration, nitrogen and phosphorus concentration, and pesticide residue, monitor water quality changes.

[0054] Data transmission uses 5G / Beidou dual-mode communication technology, and also uses adaptive redundant encryption transmission protocol. In spring, the river section may experience strong winds, sandstorms, and other weather conditions, affecting communication quality. The protocol will evaluate the communication quality in real time based on the actual transmission rate and transmission delay, dynamically adjust the data redundancy, and ensure stable data transmission under complex weather conditions. In terms of data security, a multi-layer encryption mechanism is strictly implemented. The data collection end uses the AES-256 symmetric encryption algorithm, and the encryption key is updated dynamically by the system background every 24 hours. IPsec protocol is used to tunnel encrypt data packets during transmission. The receiving end uses the RSA-4096 asymmetric encryption algorithm to verify data integrity, safely transmits real-time data collected to the digital twin layer, and receives simulation control instructions fed back from the digital twin layer.

[0055] In the three-dimensional fine model library, based on BIM+GIS technology, three-dimensional fine models of river dams and water diversion gates and other hydraulic machinery are constructed to clearly show the structural composition and spatial position relationship of the equipment; based on the water dynamics-thermodynamics coupling equation, an ice condition evolution digital model is constructed, and the simulation algorithm of the ice layer ablation process is optimized to adapt to the rapid changes in the ice layer caused by the rise in spring temperature; based on the convection-diffusion equation, a water quality concentration field digital model is constructed, considering the non-point source characteristics of agricultural non-point source pollution, and the simulation of the migration and diffusion of pollutants in the water body is optimized, and these models are imported into the three-dimensional fine model library of the digital twin body layer to complete the initialization.

[0056] The data fusion engine adopts a dynamic weight Kalman filter fusion algorithm to fuse the transmitted real-time monitoring data and the initial parameters of each model in the three-dimensional fine model library. During the spring ice layer ablation period, the data fluctuates greatly, such as the ice layer thickness, which may change rapidly due to sudden temperature changes, and the pollutant concentration, which may fluctuate due to the carrying of non-point source pollution by rainfall runoff. This algorithm introduces a data reliability weight coefficient to give reasonable weights to data from different time periods and different monitoring points, and eliminates abnormal data such as ice thickness abnormal data caused by ice flowers hitting the sensor, generating standardized model input data sets to provide reliable data support for subsequent simulation.

[0057] Dynamic coupling simulation starts multi-element dynamic coupling operation according to the coupling correlation matrix. When the operating parameters of the hydraulic machinery change, such as the increase in the opening height of the water diversion gate leading to an increase in the diversion quantity, the water flow disturbance intensity is calculated through the water dynamics equation, which is used as the boundary condition of the ice condition evolution digital model to update the simulation results of ice thickness and ice speed. The increase in diversion quantity may accelerate the flow velocity and promote the ablation of the ice layer, while changing the movement trajectory of the ice flowers.

[0058] During the ice condition evolution process, the decrease in ice layer thickness and the increase in ice speed will change the water surface area and the water flow resistance coefficient, thereby affecting the convection-diffusion coefficient in the water quality concentration field digital model, which may accelerate the diffusion speed of pollutants and update the simulation results of pollutant concentration distribution. At the same time, water quality parameters such as water temperature and nitrogen and phosphorus concentrations will affect the thermodynamic properties of the water body, which will react on the ice layer ablation rate calculation of the ice condition evolution digital model. For example, high nitrogen and phosphorus concentrations in the water body may affect the freezing point of the water body, thereby slightly changing the ablation rate of the ice layer.

[0059] During the simulation process, the ice condition evolution digital model uses a multi-factor coupled ice condition evolution equation to fully consider the influence of hydraulic machinery disturbance and water quality thermodynamic properties on ice layer ablation; the water quality concentration field digital model uses a pollutant diffusion correction equation under the constraint of the ice layer, introduces double constraints of ice layer coverage and thickness, and combines the characteristics of spring agricultural non-point source pollution to more accurately simulate the migration and diffusion law of pollutants.

[0060] The twin body updating engine adopts a multi-dimensional deviation weighted correction algorithm. By comparing the real-time monitoring data with the dynamic coupling simulation results, the deviation value is calculated, and the parameter sensitivity weight matrix is constructed. During the spring ice layer melting period, the ice layer thickness, ice velocity, and pollutant concentration have a greater impact on the simulation results, and thus have a higher sensitivity and weight. If the measured ice layer melting rate of a certain monitoring point deviates from the simulation result by more than a preset threshold, the correction amount is calculated according to the parameter sensitivity weight and the deviation value, and the related parameters in the ice evolution digital model are corrected; if the measured pollutant concentration of a certain area deviates greatly from the simulation result, the correction amount is also calculated, and the parameters of the water quality concentration field digital model are corrected to ensure that the model can accurately reflect the ice conditions and water quality conditions of the river section during the spring ice melting period.

[0061] A three-dimensional visualization interface is constructed based on WebGL technology, and a multi-scale dynamic rendering technology is adopted to divide the three-dimensional models of hydraulic machinery, ice conditions, and water quality into four levels of detail. When a user views the overall ice melting conditions and the macroscopic distribution of water quality of the entire river section, the system calls the low-detail level model to ensure fast loading and smooth operation of the interface; when the user needs to view the running state of a certain water intake gate, the surrounding ice melting details, or the water quality concentration distribution near a certain agricultural water discharge outlet in detail, the system automatically switches to the high-detail level model to clearly display the relevant details.

[0062] A mapping relationship between the monitoring data and the rendering parameters is established. The ice layer transparency changes with the ice thickness, and the smaller the ice thickness, the lower the transparency. Users can intuitively observe the progress of ice melting and the differences in ice thickness in different areas through the interface, and clearly view the running parameters of the hydraulic machinery, such as the opening height of the dam gate, the flow of the water intake gate, and the distribution of the water quality concentration field, such as the location and diffusion trend of high-concentration areas of nitrogen and phosphorus.

[0063] The decision support step adopts a multi-scenario weighted decision algorithm. The decision target weight is determined by the analytic hierarchy process. The spring ice flood prevention and control is the most critical, with the highest weight, followed by water quality guarantee, and finally hydraulic machinery energy consumption. At least three candidate decision schemes are generated for the hydraulic machinery scheduling. Scheme one is to appropriately reduce the dam water level, reduce the discharge flow, slow down the ice floe discharge speed, and reduce the ice flood risk, but may affect the downstream irrigation water and pollutant diffusion; scheme two is to maintain the existing dam water level, adjust the opening degree of the water intake gate, and balance the ice flood prevention and control, irrigation water, and pollutant diffusion; and scheme three is to appropriately increase the dam water level, increase the discharge flow, speed up the ice floe discharge, and avoid ice floe accumulation, but may increase the ice flood risk.

[0064] According to the target weight and the score of the corresponding scheme, the comprehensive score of each candidate scheme is calculated. If the comprehensive score of the second scheme is the highest, the system generates the hydraulic mechanical scheduling suggestion of the scheme to guide the staff to adjust the opening degree of the diversion gate. At the same time, according to the ice condition simulation result, if there is an ice flower accumulation risk in a certain river section, a flood control warning is generated to prompt the relevant departments to make good flood control preparation. According to the water quality simulation result, if the nitrogen and phosphorus concentration in a certain area is too high, a water quality improvement scheme is generated, such as reasonably controlling agricultural fertilization amount, optimizing irrigation mode to reduce the content of pollutants in return water, etc. And through the data transmission step, these information is fed back to the physical entity layer, to ensure the safe operation of the river section in the spring ice melting period and the water quality meets the standard.

[0065] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with a preferred embodiment, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content without departing from the scope of the present application, and any equivalent embodiments with equivalent changes are equivalent to the above embodiments. Any modification, change, and modification of the above embodiments according to the technical essence of the present application are still within the scope of the present application.

Claims

1. A method for coupled simulation of ice conditions and water quality during the Yellow River's ice-covered period, characterized in that, The specific steps of this method are as follows: S1. Construction of physical entity layer: including hydraulic mechanical entities, ice condition physical field and water quality physical field within the target monitoring area during the Yellow River ice-covered period; the hydraulic mechanical entities are used to regulate the water flow state within the target area, and their operating state directly affects the ice condition evolution and water quality diffusion; S2. Data acquisition and transmission steps: including data transmission steps for multiple types of sensors; The sensors include: an operating parameter sensor installed on the hydraulic machinery entity, an ice condition monitoring sensor arranged in the ice condition physical field, and a water quality monitoring sensor arranged in the water quality physical field; the data transmission step adopts 5G / BeiDou dual-mode communication technology to encrypt and transmit the real-time data collected by the sensors to the digital twin layer, while receiving the analog control commands fed back from the digital twin layer; S3, Digital Twin Layer: Includes a 3D fine model library, data fusion engine, dynamic coupling simulation, and twin update engine; The three-dimensional fine model library includes: three-dimensional fine model of hydraulic machinery, digital model of ice condition evolution, and digital model of water quality concentration field; The data fusion engine uses the Kalman filter algorithm to fuse the real-time monitoring data transmitted during the data acquisition and transmission steps with the initial parameters of the model, thereby eliminating data noise and generating a standardized dataset for model input. The dynamic coupling simulation establishes a coupling correlation matrix to dynamically correlate hydraulic machinery operating parameters, ice condition parameters, and water quality parameters. When the hydraulic machinery operating parameters change, the intensity of water flow disturbance is calculated using hydrodynamic equations and used as the boundary condition of the ice condition evolution digital model to update the simulation results of ice thickness and ice velocity. During the ice condition evolution process, changes in ice thickness and ice velocity affect the convection diffusion coefficient in the water quality concentration field digital model by altering the water surface area and water flow resistance coefficient, thereby updating the simulation results of pollutant concentration distribution. Simultaneously, water quality parameters affect the calculation of ice growth / melting rate in the ice condition evolution digital model by influencing the thermodynamic properties of the water body. The twin update engine uses a multi-dimensional bias weighted correction algorithm to dynamically correct the parameters of each model in the three-dimensional fine model library based on the deviation between the real-time data of the data acquisition and transmission steps and the dynamic coupling simulation results. S4. Application Interaction Steps: including visual interaction steps and decision support steps; The visualization interaction step is based on WebGL technology to build a three-dimensional visualization interface; the decision support step has a built-in multi-scenario decision algorithm, which generates hydraulic machinery scheduling suggestions, ice disaster prevention and early warning and water quality improvement schemes based on the dynamic coupling simulation results, and feeds them back to the physical entity layer through the data transmission step.

2. The method for coupled simulation of ice conditions and water quality during the Yellow River ice-covered period according to claim 1, characterized in that, The data transmission step in S2 employs an adaptive redundancy encryption transmission protocol. This protocol combines the advantages of 5G and BeiDou communication technologies, designing a dynamic redundancy strategy and a multi-layer encryption mechanism. The dynamic redundancy strategy adjusts the data redundancy based on real-time communication quality. Communication quality is comprehensively evaluated through actual transmission rate and actual transmission delay. By dynamically adjusting redundancy, it ensures stable data transmission under different communication environments. The multi-layer encryption mechanism safeguards data security at three stages: the data acquisition end, the transmission process, and the receiving end. The data acquisition end uses the AES-256 symmetric encryption algorithm, with the encryption key dynamically generated by the system backend and updated every 24 hours. During transmission, the IPsec protocol is used to tunnel-encrypt data packets. The receiving end uses the RSA-4096 asymmetric encryption algorithm to verify data integrity.

3. The method for coupled simulation of ice conditions and water quality during the Yellow River ice-covered period according to claim 1, characterized in that, The data fusion engine in S3 employs a dynamic weighted Kalman filter fusion algorithm. This algorithm addresses the high volatility of data during the Yellow River's ice-covered period by introducing a data reliability weight coefficient, the specific formula of which is as follows: in For the data fused at time k, Φ k Let K be the state transition matrix. k For Kalman gain, Z k H represents the raw sensor data at time k. k Let ω be the observation matrix. k Here, α is the weighting coefficient for data credibility, and σ is the weighting allocation coefficient. ref σ is the rated standard deviation of the sensor. k Calculate the standard deviation N for the data at time k. valid,k Let N be the number of valid data points within the sliding window at time k. total,k P represents the total amount of data in the window. k Let Q be the covariance matrix. k This is the process noise matrix.

4. The method for coupled simulation of ice conditions and water quality during the Yellow River ice-covered period according to claim 1, characterized in that, The ice condition evolution digital model in the dynamic coupling simulation of S3 adopts a multi-factor coupled ice condition evolution equation, integrating the influence of hydraulic mechanical disturbance and water quality thermodynamic properties on ice conditions. The specific formula is as follows: in The ice thickness change rate is given by h, where h is the current ice thickness; β1, β2, and β3 are coupling coefficients; Q m The actual flow rate of the hydraulic machinery is given by T0, the design flow rate of the river section is given by Q0, and the freezing point temperature is given by T. w For water temperature, C w C represents the actual concentration of pollutants in the water body. ref v represents the baseline pollutant concentration for this river section. m The velocity is the water flow velocity.

5. The method for coupled simulation of ice conditions and water quality during the Yellow River ice-covered period according to claim 1, characterized in that, The digital model for water concentration field in S3 adopts a pollutant diffusion correction equation under ice constraints, introducing dual constraints of ice coverage and thickness. The specific formula is as follows: Where C is the pollutant concentration, t is time, and v is the water flow velocity vector. The pollutant concentration gradient is represented by D0, where D0 is the water diffusion coefficient without ice, γ is the influence coefficient of ice thickness, and h is the current ice thickness. max η represents the historical maximum ice thickness for this river section, and η is the ice coverage rate. is the Laplace operator for pollutant concentration; S is the pollutant source and sink term.

6. The method for coupled simulation of ice conditions and water quality during the Yellow River ice-covered period according to claim 1, characterized in that, The twin update engine in S3 employs a multi-dimensional bias weighted correction algorithm to construct a parameter sensitivity weight matrix. The specific correction process and formula are as follows: First, calculate the simulation bias of each model parameter: ΔP i =|P sim,i -P meas,i | / P meas,i ×100%, where ΔP i P represents the relative deviation of the i-th parameter. sim,i P is the simulated value of this parameter. meas,i These are the measured values; then, a parameter sensitivity weight matrix is ​​constructed: Where S i Let i be the sensitivity coefficient of the i-th parameter. The sum of the sensitivity coefficients of all parameters; weight W i The sensitivity reflects the degree of influence of parameter deviation on the overall simulation results; the higher the sensitivity, the greater the weight. Next, the model parameter correction amount ΔP′ is calculated. i =ΔP i ×W i ×k、P new,i =P sim,i ±ΔP′ i , where ΔP′ i P is the correction amount for the i-th parameter, and k is the correction coefficient; new,i These are the corrected parameter values.

7. The method for coupled simulation of ice conditions and water quality during the Yellow River ice-covered period according to claim 1, characterized in that, The decision support step in S3 adopts a multi-scenario weighted decision algorithm. The algorithm first determines the weight of the decision objectives through the analytic hierarchy process. The decision objectives include ice disaster prevention and control, water quality protection, and hydraulic machinery energy consumption. Then, at least three candidate decision schemes are generated for hydraulic machinery scheduling. Finally, based on the weight of each objective and the corresponding score, the comprehensive score of each candidate scheme is calculated, and the scheme with the highest comprehensive score is the recommended decision scheme.

8. The method for coupled simulation of ice conditions and water quality during the Yellow River ice-covered period according to claim 1, characterized in that, The visualization interaction step in S4 adopts multi-scale dynamic rendering technology, which includes two parts: a model simplification strategy based on the level of detail and a real-time data-driven rendering mechanism. In terms of the model simplification strategy, the three-dimensional models of hydraulic machinery, ice conditions, and water quality are divided into four levels of detail. The switching between each level is determined based on the distance between the user's perspective and the model. In terms of the real-time data-driven rendering mechanism, a mapping relationship between monitoring data and rendering parameters is established. The transparency of the ice layer changes with the ice thickness. The greater the ice thickness, the higher the transparency, which intuitively shows the difference in ice layer thickness.

9. A system for coupled simulation of ice conditions and water quality during the Yellow River's ice-covered period, applicable to the method for coupled simulation of ice conditions and water quality during the Yellow River's ice-covered period as described in any one of claims 1-8, characterized in that, The system includes: a physical entity and digital model initialization module, a real-time data acquisition and transmission module, a data fusion and model input module, a dynamic coupling simulation calculation module, and a twin synchronization update and result output module. The physical entity and digital model initialization module: determines the target monitoring area during the Yellow River ice-covered period, and obtains the geographic information, hydraulic machinery entity parameters, initial ice condition data, and initial water quality data of the area; constructs a three-dimensional fine model of hydraulic machinery based on BIM+GIS technology, constructs a digital model of ice condition evolution based on hydrodynamic-thermodynamic coupling equations, constructs a digital model of water quality concentration field based on convection-diffusion equations, and imports the above models into the three-dimensional fine model library of the digital twin layer to complete the digital twin initialization; The real-time data acquisition and transmission module includes: sensors that initiate the data acquisition and transmission process; a parameter sensor that collects real-time data on the rotational speed, torque, and flow rate of the hydraulic machinery; an ice condition monitoring sensor that collects real-time data on ice thickness, ice velocity, and ice temperature; and a water quality monitoring sensor that collects real-time data on pH value, dissolved oxygen concentration, and pollutant concentration. The data transmission process encrypts and transmits the collected real-time data to the data fusion engine of the digital twin layer using 5G / BeiDou dual-mode communication technology. The data fusion and model input module: The data fusion engine uses the Kalman filter algorithm to fuse the transmitted real-time monitoring data with the initial parameters of each model in the 3D fine model library, remove abnormal data, and generate a standardized model input dataset; the standardized dataset is then input into the hydraulic machinery submodule, ice condition evolution submodule, and water quality concentration field submodule in the dynamic coupling simulation respectively; The dynamic coupling simulation module is based on a coupling correlation matrix and initiates multi-factor dynamic coupling calculations: the hydraulic machinery submodule calculates the water flow velocity field and pressure field distribution using fluid dynamics equations based on the input operating parameters, and transmits the water flow disturbance intensity parameters to the ice condition evolution submodule; the ice condition evolution submodule uses the water flow disturbance intensity as a boundary condition, combined with the input initial ice condition parameters, and calculates the ice layer growth / melting rate and ice velocity change using hydrodynamic-thermodynamic coupling equations to generate real-time ice condition evolution results, and transmits the ice layer thickness and water flow resistance coefficient parameters to the water quality concentration field submodule; the water quality concentration field submodule corrects the water diffusion coefficient based on the ice layer thickness, and calculates the pollutant migration path and concentration distribution using convection diffusion equations based on the water flow velocity field and initial water quality parameters, generating real-time water quality simulation results; simultaneously, the water quality concentration field submodule feeds back the water temperature and pollutant concentration parameters to the ice condition evolution submodule to correct the ice layer thermodynamic characteristic parameters; The twin synchronization update and result output module: The twin update engine compares the coupled simulation results with real-time monitoring data, calculates the deviation value, and if the deviation value exceeds the preset threshold, it uses a real-time interpolation algorithm to correct the parameters of the corresponding model in the three-dimensional fine model library; The visualization interaction step of the application interaction step displays the coupled simulation results in a three-dimensional visualization form, and users can view the hydraulic machinery operation status, ice condition evolution process and water quality concentration field distribution through the interface; The decision support step generates scheduling suggestions, early warning information and improvement plans based on the simulation results.

Images