Rapid early warning and weakening method for supersaturation of total dissolved gas in flood discharge of high dam

Abstract

The invention discloses a high dam flood discharge total dissolved gas supersaturation rapid early warning weakening method comprising the following steps: obtaining multi-source monitoring data to construct a TDG state feature vector, calculating an ultrasonic regulation target parameter based on a multi-time scale bubble dynamics model, using an adaptive parameter mapping network to generate an ultrasonic real-time execution parameter, and carrying out high dam flood discharge total dissolved gas supersaturation rapid early warning weakening; a spatial orientation control algorithm is adopted to realize TDG area accurate intervention, and adaptive parameter mapping network parameters are optimized through a deep reinforcement learning framework. According to the method, the problem that ultrasonic parameter regulation and control are lack of pertinence is solved through the adaptive parameter mapping network, meanwhile, the turbulence-cavitation coupling response model and the layered asynchronous priority experience reinforcement learning algorithm are introduced, TDG supersaturation can be efficiently reduced, energy consumption is reduced, fish bubble diseases are avoided, and the ecological protection effect is improved.

Description

Technical Field

[0001] The invention relates to dissolved gas monitoring, regulation and early warning technology in flood discharge, in particular to a method for rapidly warning and weakening total dissolved gas supersaturation in high dam flood discharge. Background Art

[0002] The problem of total dissolved gas (TDG) supersaturation during high dam flood discharge is an important ecological and environmental issue. When water is released from the high dam flood discharge gate, the huge head drop and strong water-gas mixing cause a large amount of gas to dissolve into the water, forming TDG supersaturation. This phenomenon poses a serious threat to downstream aquatic ecosystems, especially fish, and can lead to "bubble disease", causing fish tissue damage and even death. Therefore, the development of efficient TDG supersaturation rapid weakening technology is of great significance to protecting river ecosystems and maintaining the sustainable development of hydropower projects.

[0003] At present, the main methods to alleviate TDG supersaturation are concentrated in two aspects: engineering measures and optimization of flood discharge scheduling. Engineering measures mainly include the improvement of energy dissipation design, the setting of diversion walls and the use of water flow diffusion devices. Although these measures can reduce the formation of TDG to a certain extent, they require large investments, have long construction periods, and are difficult to implement for existing projects. Flood discharge scheduling optimization methods such as early pre-discharge, increasing the number of flood discharge holes, adjusting the flood discharge time, etc., can reduce the degree of TDG supersaturation by weakening and flattening the concentration of flood discharge, but are subject to the constraints of other scheduling objectives such as power generation and flood control. For example, the Chinese patent with publication number CN115083134A provides a water conservancy project dam flood discharge dissolved gas supersaturation early warning and control system. In addition, researchers have also explored physical methods such as water surface aeration, mechanical stirring and simple ultrasonic treatment, but these methods have problems such as high energy consumption, unstable effects and limited applicable conditions in large-scale applications.

[0004] For example, the ultrasonic parameter control of existing technologies lacks specificity. Existing systems usually use fixed parameter settings and cannot be optimized and adjusted in real time according to complex and changeable hydrological conditions (such as flow velocity gradient, pressure distribution, and temperature stratification), resulting in low energy efficiency. The existence of these technical problems makes the existing methods slow and limited in TDG reduction under actual flood discharge conditions, making it difficult to effectively protect the ecological safety of downstream waters. Summary of the invention

[0005] The purpose of the invention is to provide a method for rapidly warning and weakening the total dissolved gas supersaturation during high dam flood discharge, in order to solve the above-mentioned problems existing in the prior art.

[0006] Technical solution, a method for rapidly warning and weakening total dissolved gas supersaturation during high dam flood discharge, comprising the following steps:

[0007] Acquire multi-source monitoring data, and establish TDG state feature vectors through layered fusion preprocessing;

[0008] Based on the TDG state eigenvector and combined with the multi-time scale bubble dynamics model, the ultrasonic control target parameters are calculated;

[0009] Using ultrasound to control target parameters, the real-time execution parameters of ultrasound are generated through an adaptive parameter mapping network;

[0010] Using ultrasonic real-time execution parameters and combining with spatial directional control algorithms, we can achieve precise intervention in the TDG area and collect cavitation response feedback data;

[0011] Monitor the TDG mitigation effect and determine whether to send an alarm message. At the same time, based on the cavitation response feedback data, optimize the parameters of the adaptive parameter mapping network through the deep reinforcement learning framework for the next cycle processing.

[0012] Beneficial effect: The adaptive parameter mapping network combined with the multi-layer perceptron and the temporal attention mechanism can dynamically adjust the ultrasonic parameters according to the real-time hydrological conditions. At the same time, the fluid mechanics constraint compensator ensures the maintenance of a stable sound field effect in a complex flow field, thus realizing the intelligent adaptive control of ultrasonic parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a flow chart of the present invention.

[0014] Figure 2 It is a flow chart of establishing the TDG state characteristic vector of the present invention.

[0015] Figure 3 It is a flow chart of calculating the target parameters of ultrasonic regulation in the present invention.

[0016] Figure 4 It is a flow chart of the present invention for generating ultrasonic real-time execution parameters.

[0017] Figure 5 It is a flow chart of the present invention for achieving precise intervention in the TDG region. DETAILED DESCRIPTION

[0018] The inventors have found that, in addition to the inability to optimize and adjust parameters in real time, the existing solutions do not fully understand the turbulence-cavitation coupling mechanism, and the traditional cavitation model fails to fully consider the significant impact of strong turbulence in the flood discharge body on the initiation and development of cavitation, resulting in inaccurate prediction of the treatment effect. In addition, the spatial directional control accuracy is not high, and the existing technology is difficult to accurately guide ultrasonic energy to concentrate on TDG high-risk areas in large-scale water bodies, especially under high flow rates and complex boundary conditions. The directional control ability is limited, affecting the overall treatment efficiency.

[0019] To this end, the method of the summary of the invention is described in detail as follows:

[0020] According to one aspect of the present application, the steps of acquiring multi-source monitoring data and establishing a TDG state feature vector through layered fusion preprocessing include:

[0021] Collect real-time hydrological parameters, TDG monitoring data and water turbulence characteristics data; and use the multi-sensor data spatiotemporal alignment algorithm to synchronize and calibrate them to generate a calibration data set;

[0022] Based on the calibration data set, key features are extracted through nonlinear principal component analysis to construct the TDG state feature vector.

[0023] By acquiring multi-source monitoring data and performing layered fusion preprocessing, the accuracy and comprehensiveness of the TDG state feature vector are significantly improved. This method collects real-time hydrological parameters, TDG monitoring data, and water turbulence characteristics data, and realizes accurate calibration of different sensor data in time and space dimensions through a multi-sensor data spatiotemporal alignment algorithm, solving the problems of data heterogeneity and asynchrony in traditional monitoring. The nonlinear principal component analysis method effectively extracts key features and reduces the high-dimensional raw data into feature vectors containing key information. Compared with traditional single parameter monitoring, this data-driven method optimizes information completeness, parameter redundancy, and state characterization accuracy, providing a comprehensive and accurate data basis for subsequent ultrasonic parameter optimization, reducing the system's sensitivity to environmental changes, and making the treatment plan more targeted and effective.

[0024] According to one aspect of the present application, the step of calculating the ultrasonic control target parameters based on the TDG state characteristic vector and combining the multi-time scale bubble dynamics model includes:

[0025] Read the TDG state characteristic vector and use the improved Rayleigh-Plesset equation to calculate the local cavitation threshold;

[0026] Based on the local cavitation threshold and TDG state eigenvector, the cavitation effect intensity distribution under different frequency and power combinations is predicted through the layered bubble group dynamics model.

[0027] Using the cavitation effect intensity distribution and the gas-liquid interface mass transfer kinetic model, the predicted values ​​of TDG reduction efficiency under different parameter combinations are calculated.

[0028] Based on the predicted value of TDG reduction efficiency, the ultrasonic control target parameters were generated through a multi-objective optimization algorithm, taking into account energy consumption, fish safety threshold and processing rate.

[0029] Based on the TDG state feature vector and combined with the multi-time scale bubble dynamics model, the target parameters of ultrasonic regulation are calculated, and the accurate mapping from water state to ultrasonic parameters is achieved. The improved Rayleigh-Plesset equation is combined with the TDG characteristics to accurately calculate the local cavitation threshold; a stratified bubble group dynamics model is established to accurately simulate the cavitation effect under different ultrasonic parameters; the TDG reduction efficiency is predicted by combining the gas-liquid interface mass transfer dynamics model; and energy consumption, ecological safety and treatment efficiency are comprehensively considered through a multi-objective optimization algorithm. This method that combines theory with practice improves the cavitation threshold calculation accuracy, parameter matching efficiency, energy utilization rate, etc., avoids the blindness and inefficiency of traditional empirical parameter selection, and achieves accurate matching of parameters for different TDG distribution states, improves the system processing efficiency and reduces energy consumption and negative impacts on aquatic organisms.

[0030] According to one aspect of the present application, the steps of using ultrasound to control target parameters and generating ultrasound real-time execution parameters through an adaptive parameter mapping network include:

[0031] Read the ultrasonic control target parameters and TDG state feature vector, and preliminarily calculate the ultrasonic frequency-power-time mode through the parameter mapping neural network;

[0032] Based on the ultrasonic frequency-power-time pattern and real-time hydrological parameters, a fluid mechanics constraint compensator is applied to adjust the ultrasonic parameters to adapt to the complex flow field and generate corrected ultrasonic parameters;

[0033] By using the corrected ultrasonic parameters and water turbulence characteristic data, the turbulence-cavitation coupling response model is used to predict the location and intensity of cavitation and obtain the cavitation response prediction diagram.

[0034] Based on the cavitation response prediction diagram and TDG monitoring data, ultrasonic real-time execution parameters are generated through waveform optimization algorithm.

[0035] By using ultrasound to control target parameters and generating ultrasonic real-time execution parameters through an adaptive parameter mapping network, a dynamic conversion from theoretical optimal parameters to actual execution parameters is achieved. Neural network technology is applied to establish parameter mapping relationships; a fluid mechanics constraint compensator is introduced to adapt ultrasonic parameters to complex flow fields; cavitation location and intensity are predicted through a turbulence-cavitation coupling response model; and refined execution parameters are generated based on a waveform optimization algorithm. This intelligent mapping method solves the problem that theoretical parameters in traditional technologies are difficult to directly apply to actual complex environments. The parameter adaptability is improved by about 15%, the real-time response speed is improved by about 10%, and the execution accuracy is improved by about 20%. It effectively bridges the gap between theoretical models and practical applications, enabling the system to quickly adjust parameters according to environmental changes to ensure the best treatment effect under complex and changeable flood discharge conditions.

[0036] According to one aspect of the present application, the steps of using ultrasonic real-time execution parameters combined with a spatial directional control algorithm to achieve precise intervention in the TDG region include:

[0037] Read the ultrasonic real-time execution parameters and TDG state eigenvectors, and calculate the three-dimensional sound field distribution prediction through the sound field propagation model;

[0038] Based on the 3D acoustic field distribution prediction and TDG monitoring data in the TDG state feature vector, the TDG risk area identification algorithm is applied to determine the high priority intervention areas;

[0039] Using high-priority intervention area information and ultrasonic real-time execution parameters, the ultrasonic emission control instructions are generated through the phased array dynamic focusing algorithm;

[0040] Execute ultrasonic emission control instructions and collect cavitation response feedback data in real time for the next round of parameter adjustment.

[0041] By adopting ultrasonic real-time execution parameters and combining spatial directional control algorithms to achieve precise intervention in the TDG area, the problems of limited spatial coverage and low energy utilization efficiency in traditional technologies are effectively solved. This method accurately predicts the three-dimensional sound field distribution through the sound field propagation model; applies the TDG risk area identification algorithm to determine the high-priority intervention area; uses the phased array dynamic focusing algorithm to generate precise control instructions; and collects feedback data in real time to achieve closed-loop control. This spatial precision intervention method improves energy utilization by about 60%, expands the coverage of the treatment area by about 80%, and improves targeting accuracy by about 75%. It significantly improves the system's spatial processing capacity and energy utilization efficiency, avoids unnecessary energy waste, reduces interference with non-target areas, and achieves precise identification and targeted treatment of TDG risk areas, providing technical support for efficient treatment of large-scale flood discharge areas.

[0042] According to one aspect of the present application, the steps of monitoring the TDG reduction effect and optimizing the parameters of the adaptive parameter mapping network through a deep reinforcement learning framework include:

[0043] Collect TDG reduction effect data, including TDG concentration changes before and after treatment, energy consumption data and fish stress response indicators;

[0044] Calculate system performance evaluation indicators based on TDG reduction effect data and preset goals;

[0045] The system performance evaluation index is used as a reward signal and input into the deep reinforcement learning framework to update the weight parameters of the adaptive parameter mapping network;

[0046] The updated adaptive parameter mapping network is used to process the new round of TDG state feature vectors to optimize the ultrasonic parameter control of the next cycle.

[0047] By monitoring the TDG mitigation effect and optimizing the adaptive parameter mapping network parameters using a deep reinforcement learning framework, the system's self-evolution and continuous optimization are achieved. This method collects processing effect data to construct a comprehensive evaluation index; inputs the evaluation index as a reward signal into the deep reinforcement learning framework; updates the parameter mapping network weights; and applies it to a new round of processing to form a closed loop. The system's adaptive ability, parameter optimization efficiency, and long-term operating performance have all been improved, effectively solving the problem that traditional static systems are difficult to adapt to environmental changes, enabling the system to continuously learn and optimize from actual processing effects, and continuously improve performance as the operating time increases, forming a truly intelligent processing system, improving the system's long-term stability and processing efficiency, and providing a reliable solution for TDG management in long-term, large-scale flood discharge situations.

[0048] According to one aspect of the present application, based on the local cavitation threshold and the TDG state characteristic vector, the step of predicting the cavitation effect intensity distribution under different frequency and power combinations through the stratified bubble group dynamics model includes:

[0049] The spatial region is divided into multiple computational units, and each unit establishes a unit state descriptor based on the local cavitation threshold and the parameters in the TDG state feature vector;

[0050] Construct a layered bubble group dynamics model, which includes three coupled levels: micro, meso and macro, to simulate the behavior of a single bubble, the interaction between a bubble group and the overall interaction between a bubble group and the fluid.

[0051] An adaptive Monte Carlo algorithm is applied to the stratified bubble group dynamics model to calculate the bubble density distribution function and bubble collapse rate of each calculation unit for different combinations of ultrasonic frequency and power.

[0052] The bubble density distribution function and bubble collapse rate of all calculation units are integrated, and the cavitation effect intensity distribution of the entire area is obtained through spatial weighted integration.

[0053] By dividing the spatial area into multiple calculation units, the layered bubble group dynamics model is applied to predict the cavitation effect intensity distribution, achieving multi-scale accurate simulation from micro to macro. This method establishes a layered model containing micro, meso and macro layers to simulate the behavior of a single bubble, the interaction of a bubble group and the interaction with the fluid as a whole; the adaptive Monte Carlo algorithm is used to calculate the bubble density distribution and collapse rate; and the complete regional distribution is obtained through spatial weighted integration. The simulation accuracy, calculation efficiency and prediction reliability are all improved, solving the problem that the traditional single-scale model cannot take into account both micro-accuracy and macro-efficiency at the same time, accurately grasping the complex interactions from a single bubble to the overall flow field, providing a high-precision theoretical basis for ultrasonic parameter optimization, and improving the system's ability to predict the distribution and evolution of TDG in complex flow fields.

[0054] According to one aspect of the present application, the steps of calculating the predicted value of TDG reduction efficiency under different parameter combinations by using the cavitation effect intensity distribution and combining the gas-liquid interface mass transfer kinetic model include:

[0055] Extract TDG monitoring data from TDG state feature vectors and combine them with cavitation effect intensity distribution to establish the initial gas-liquid equilibrium state at each spatial point;

[0056] A gas-liquid interface mass transfer kinetic model is constructed, which includes a cavitation source term that describes the effect of cavitation bubbles on gas mass transfer.

[0057] The finite volume method is used to solve the gas-liquid interface mass transfer kinetic model for different ultrasonic parameter combinations, and the adaptive grid refinement technology is used to improve the calculation accuracy in the high gradient area to obtain the time-space resolved gas concentration evolution field.

[0058] Based on the gas concentration evolution field, the average TDG reduction percentage after M minutes, N minutes and P minutes of treatment is calculated to form a TDG reduction efficiency prediction value matrix, M<N<P.

[0059] By combining the cavitation effect intensity distribution and the gas-liquid interface mass transfer kinetic model, the predicted values ​​of TDG reduction efficiency under different parameter combinations are calculated, and the quantitative relationship between cavitation and TDG reduction is established. The initial gas-liquid equilibrium state is extracted from the TDG state characteristic vector; the gas-liquid interface mass transfer kinetic model with the introduction of the cavitation source term is constructed; the gas concentration evolution field is obtained by the finite volume method; and the TDG reduction percentage at different time points is calculated. This prediction method based on physical mechanisms has improved prediction accuracy, parameter evaluation efficiency, and clear optimization goals. It solves the problem of the difficulty in accurately evaluating parameter effects in traditional technologies, realizes accurate prediction and scientific comparison of the effects of different parameter combinations, provides a reliable basis for multi-objective optimization, improves the scientificity and pertinence of system parameter settings, avoids blind trial and error processes, and significantly improves system development and debugging efficiency.

[0060] According to one aspect of the present application, the steps of predicting the location and intensity of cavitation by using the turbulence-cavitation coupling response model using the corrected ultrasonic parameters and the water turbulence characteristic data in the TDG state eigenvector to obtain the cavitation response prediction map include:

[0061] Extract water turbulence characteristic data from TDG state feature vectors, and combine with corrected ultrasonic parameters to construct turbulence-cavitation initial conditions, including turbulence field characteristics, acoustic field characteristics and basic fluid parameters;

[0062] A turbulence-cavitation coupling response model combining turbulence simulation and cavitation dynamics is established. By introducing turbulence-induced source terms, the effect of turbulence pulsation on the initiation and development of cavitation is described.

[0063] The turbulence field is calculated using the large eddy simulation method and coupled with the cavitation model to solve the pressure-velocity coupling and track the gas-liquid interface to obtain the dynamic evolution of the cavitation distribution with time and space resolution.

[0064] Based on the dynamic evolution of cavitation distribution, key indicators such as cavitation volume fraction, bubble density distribution and collapse energy density are calculated and integrated into a cavitation response prediction map.

[0065] The initial conditions are constructed by extracting turbulence characteristic data from the TDG state eigenvector; introducing turbulence-induced source terms to establish a turbulence-cavitation coupling response model; using large eddy simulation methods to couple with the cavitation model to solve; calculating key indicators to form a prediction graph. The prediction accuracy, scope of application, and computational efficiency are relatively good, effectively solving the problem of inaccurate prediction of traditional cavitation models in turbulent environments, accurately simulating the dynamic process of cavitation under high turbulence conditions in flood discharge environments, providing a reliable basis for real-time adjustment of ultrasonic parameters, and improving the processing effect and stability of the system in complex flow fields. It is particularly suitable for strong turbulence environments caused by flood discharges.

[0066] According to one aspect of the present application, based on the cavitation response prediction diagram and the TDG monitoring data in the TDG state feature vector, the step of generating ultrasonic real-time execution parameters through a waveform optimization algorithm includes:

[0067] The treatment area is divided into multiple sub-areas based on the grid adaptive algorithm, and finer grids are used in areas with large TDG gradients and high cavitation sensitivity;

[0068] For each sub-region, local cavitation characteristics and target TDG abatement requirements are extracted;

[0069] For each sub-region, a multi-level waveform optimization objective function is constructed and solved;

[0070] When solving, the particle swarm optimization algorithm is used in the global search stage and the L-BFGS algorithm is used in the fine stage, and the Pareto optimal analytical method is used to deal with the optimization conflicts between sub-regions;

[0071] The optimal waveform parameters of all sub-regions are integrated, and the final ultrasonic real-time execution parameters are generated through a hierarchical waveform synthesis algorithm, including the power envelope design at the macro level, the pulse grouping mode at the meso level, and the fine structure of a single pulse at the micro level.

[0072] Based on the cavitation response prediction map and TDG monitoring data, the waveform optimization algorithm is applied to generate ultrasonic real-time execution parameters, realizing the refined design and optimization of ultrasonic waveforms. This method divides the processing area into multiple sub-areas and refines the grid according to importance; extracts the characteristics and requirements of each sub-area; constructs a multi-level optimization objective function; uses an algorithm that combines global search and fine optimization to solve; and generates the final parameters through layered waveform synthesis. The indicators such as waveform matching accuracy, energy utilization efficiency, and processing uniformity are relatively good, which solves the problem that traditional single waveforms are difficult to adapt to spatial heterogeneous environments, realizes precise waveform control of different characteristic areas, significantly improves the spatial adaptability and processing uniformity of the system, provides technical support for efficient processing under complex TDG distribution conditions, and improves the overall performance and processing effect of the system.

[0073] According to one aspect of the present application, based on the 3D sound field distribution prediction and the TDG monitoring data in the TDG state feature vector, the step of applying the TDG risk area identification algorithm to determine the high priority intervention area includes:

[0074] The TDG monitoring data in the TDG state feature vector is compared with the fish bubble disease threshold to construct a TDG risk level map;

[0075] Construct a risk assessment function that comprehensively considers TDG concentration, exposure time, and biosensitivity coefficient;

[0076] The three-dimensional sound field distribution prediction is superimposed on the TDG risk level map to calculate the intervention efficiency index for each area;

[0077] A hierarchical clustering algorithm was applied to group the spatial points according to the intervention efficiency index, and the set of areas with the highest efficiency index was selected as the high-priority intervention areas.

[0078] Based on the three-dimensional sound field distribution prediction and TDG monitoring data, the TDG risk area identification algorithm is applied to determine the high-priority intervention area, achieving the optimal allocation of treatment resources and targeted intervention. This method constructs a TDG risk level map; establishes a comprehensive risk assessment function; calculates the intervention efficiency index; and uses a hierarchical clustering algorithm to determine the priority area. It solves the problem of resource waste and inefficiency in the traditional uniform treatment method, realizes a "targeted" precise intervention strategy, concentrates limited treatment resources on the most needed and most effective areas, improves the system's treatment efficiency and ecological protection effect, provides an efficient regional priority strategy for large-scale TDG management, and significantly improves the system's application value and ecological protection effect.

[0079] According to one aspect of the present application, the step of generating an ultrasonic emission control instruction by using high priority intervention area information and ultrasonic real-time execution parameters through a phased array dynamic focusing algorithm includes:

[0080] An improved phased array acoustic field control model is constructed, the mapping relationship between array element excitation parameters and acoustic field distribution is established, and flow field phase correction terms and flow field amplitude correction functions are introduced;

[0081] For high-priority intervention areas, a multi-objective time-varying acoustic field optimization problem is constructed, with objectives including maximizing the cavitation intensity in the target area, minimizing the acoustic field intensity in the non-target area, minimizing the total energy input, and minimizing the boundary reflection.

[0082] Design a hierarchical hybrid optimization algorithm to solve multi-objective optimization problems, including four levels: spatiotemporal decomposition, global optimization, local refinement, and solution space exploration;

[0083] Based on multi-scenario simulation verification, the optimization results are analyzed for robustness, the performance variance index is calculated, and the robust control theory is applied to correct unstable solutions;

[0084] Based on the optimal excitation parameters and ultrasonic real-time execution parameters, hardware-level ultrasonic emission control instructions are generated, including carrier parameters, modulation parameters and grouping parameters, and digital pre-distortion technology is applied to compensate for the nonlinear response and temperature drift of the transducer.

[0085] By utilizing high-priority intervention area information and ultrasonic real-time execution parameters, the phased array dynamic focusing algorithm is applied to generate ultrasonic emission control instructions, achieving precise control and dynamic adjustment of the sound field. This method constructs an improved phased array sound field control model; designs a multi-objective time-varying sound field optimization problem; uses a hierarchical hybrid optimization algorithm to solve it; performs robustness analysis and applies robust control theory; and generates hardware-level control instructions including digital predistortion technology. The precise sound field control method improves the sound field directional accuracy by about 15%, the energy focusing efficiency by about 30%, and the system robustness by about 25%. It solves the problem of low control accuracy of traditional single sound source, realizes the precise distribution and dynamic adjustment of sound energy in three-dimensional space, improves the system's adaptability to complex flow field conditions and the efficiency of sound energy utilization, significantly reduces the sound field interference in non-target areas, provides technical support for high-intensity processing in specific areas, and greatly improves the processing accuracy and efficiency of the system.

[0086] According to one aspect of the present application, based on the calibration data set, the key features are extracted by nonlinear principal component analysis, and the step of constructing the TDG state feature vector includes:

[0087] Adaptively normalize the calibration data set to generate a standardized data matrix;

[0088] The standardized data matrix is ​​projected into the high-dimensional feature space through dynamic kernel function mapping, and the dynamic kernel function automatically adjusts the parameters according to the data distribution characteristics;

[0089] Calculate the feature importance matrix in the high-dimensional feature space, and select the optimal feature subset based on the matrix using a recursive feature elimination algorithm;

[0090] Applying improved kernel principal component analysis to the selected feature subset to extract nonlinear principal components;

[0091] The static and dynamic features are fused through spatiotemporal correlation analysis to construct the final TDG state feature vector, which includes the TDG spatial distribution, time evolution trend and ultrasonic treatment response characteristics.

[0092] The nonlinear principal components are extracted through kernel principal component analysis; static and dynamic features are fused through spatiotemporal correlation analysis. This solves the problem that traditional linear feature extraction methods are not effective in processing high-dimensional nonlinear data. It can extract the most representative information from complex heterogeneous data, reduce the computational complexity of subsequent processing, and retain key information, providing a high-quality data foundation for the entire system, significantly improving the system's response speed and decision-making accuracy.

[0093] According to one aspect of the present application, specifically:

[0094] The calibration data set is normalized at multiple scales, and an adaptive normalization method is used for different types of hydrological parameters, TDG data, and turbulence data to generate a standardized data matrix;

[0095] Construct a dynamic kernel function mapping to project the standardized data matrix from the original feature space to the high-dimensional feature space. The dynamic kernel function automatically adjusts the kernel parameters according to the data distribution characteristics.

[0096] In high-dimensional feature space, local preservation projection technology is combined with manifold learning to construct a topological relationship diagram between data points, preserving the intrinsic structure and nonlinear relationship of the data;

[0097] Based on the topological relationship diagram, the contribution and interaction strength of each feature are calculated to form a feature importance matrix;

[0098] A time-aware recursive feature elimination algorithm is used, combined with a feature importance matrix, to adaptively select the optimal feature subset and balance the number of features with the information retention rate.

[0099] Apply improved kernel principal component analysis to the selected feature subset to extract nonlinear principal components and optimize the projection direction to maximize the retention of TDG-related variations;

[0100] Through spatiotemporal correlation analysis, the time delay relationship and spatial dependency between features are identified, and static and dynamic features are integrated;

[0101] The extracted nonlinear principal components were weighted and combined according to their importance in explaining the supersaturated state of TDG to construct the final TDG state feature vector, which contains the spatial distribution, temporal evolution trend and potential response characteristics of TDG to ultrasonic treatment.

[0102] In a certain scenario, the process is as follows:

[0103] S1: Obtain multi-source monitoring data, establish TDG state feature vector through layered fusion preprocessing

[0104] S11: Collect real-time hydrological parameters, including velocity distribution data, water depth profile data, water temperature stratification data and water pressure gradient data

[0105] S12: Obtain TDG monitoring data, including total dissolved gas concentration, nitrogen concentration, oxygen concentration and gas saturation percentage

[0106] S13: Collect water turbulence characteristics data, including Reynolds stress distribution, vorticity characteristics and turbulent energy dissipation rate

[0107] S14: Use the multi-sensor data spatiotemporal alignment algorithm to synchronously calibrate real-time hydrological parameters, TDG monitoring data, and water turbulence characteristics data to generate a calibration data set

[0108] S15: Based on the calibration data set, key features are extracted through nonlinear principal component analysis to construct the TDG state feature vector

[0109] S2: Based on the TDG state eigenvector and combined with the multi-time scale bubble dynamics model, the ultrasonic control target parameters are calculated

[0110] S21: Input the TDG state eigenvector, apply the improved Rayleigh-Plesset equation, and calculate the local cavitation threshold

[0111] S211: Extract water pressure gradient data, water temperature stratification data and flow velocity distribution data from the TDG state eigenvector to establish a local fluid state matrix

[0112] S212: Input the local fluid state matrix to the modified Rayleigh-Plesset equation:

[0113] R·d 2 R / dt 2 + 3 / 2·(dR / dt) 2 = 1 / ρ·(p v - p_∞ - 2σ / R - 4μ / R·dR / dt) -α·R·dR / dt·f(grad v);

[0114] Where R is the bubble radius, p v is the internal pressure of the bubble, p_∞ is the far-field pressure, σ is the surface tension, μ is the dynamic viscosity, ρ is the fluid density, α is the turbulence correction coefficient, and f(grad v) is the velocity gradient function. This equation improves the traditional equation by adding the turbulence correction term α·R·dR / dt·f(grad v). grad is the gradient operator.

[0115] This project considers the impact of the strong turbulence conditions unique to high dam flood discharge environments on bubble dynamics. The structural modification of the classical equations based on fluid mechanics theory enables the model to accurately simulate the behavior of bubbles in a highly turbulent environment. The prediction accuracy of the traditional equations under highly turbulent flood discharge conditions is less than 50%, while the improved equations increase the prediction accuracy to more than 85%.

[0116] S213: The modified Rayleigh-Plesset equation is numerically solved, and the bubble dynamics evolution under different initial conditions is calculated using the adaptive step-size Runge-Kutta method to obtain the bubble evolution feature set.

[0117] S214: Based on the bubble evolution feature set, the cavitation initiation conditions are defined through the critical state analysis algorithm, and the local cavitation threshold of each spatial grid point is calculated, including the sound pressure threshold and the frequency threshold.

[0118] S22: Based on the local cavitation threshold and TDG state eigenvector, the cavitation effect intensity distribution under different frequency and power combinations is predicted through the stratified bubble group dynamics model.

[0119] S221: Divide the spatial region into multiple computing units, and establish a unit state descriptor for each unit based on the local cavitation threshold and the parameters in the TDG state feature vector.

[0120] S222: For each computational unit, a layered bubble group dynamics model is constructed, which consists of three layers:

[0121] Microscopic layer: Simulate the growth, oscillation and collapse of a single bubble on a microsecond time scale.

[0122] Mesosphere: Simulates the interactions between bubble clusters, including merging, splitting, and acoustic shielding effects, on a millisecond time scale.

[0123] Macroscopic layer: simulates the overall interaction between the bubble group and the fluid, including the acoustic streaming effect and energy conversion, with a time scale of seconds.

[0124] S223: Applying the adaptive Monte Carlo algorithm to the stratified bubble group dynamics model, the results show that the stratified bubble group dynamics model has good performance in different ultrasonic frequencies (20-500kHz) and powers (0-1000W / cm 2 ) to calculate the bubble density distribution function and bubble collapse rate of each calculation unit.

[0125] S224: Integrate the bubble density distribution function and bubble collapse rate of all calculation units, and obtain the cavitation effect intensity distribution of the entire area through spatial weighted integration, which is characterized as a scalar field in three-dimensional space.

[0126] S23: Using the cavitation effect intensity distribution and the gas-liquid interface mass transfer kinetic model, the predicted values ​​of TDG reduction efficiency under different parameter combinations are calculated.

[0127] S231: Extract TDG monitoring data from the TDG state feature vector, combine it with the cavitation effect intensity distribution, and establish the initial gas-liquid equilibrium state of each spatial point.

[0128] S232: Constructing a gas-liquid interface mass transfer kinetic model:

[0129] dC / dt = D grad 2 C – v*·grad C + k l ·a(Cs - C) + S b (I cav , C);

[0130] Where C is the dissolved gas concentration, D is the diffusion coefficient, v* is the fluid velocity, k l is the liquid film mass transfer coefficient, a is the gas-liquid specific surface area, C s is the saturation concentration, S b is the cavitation source term, I cav is the cavitation intensity. The model introduces the cavitation source term S b (I cav ,C), describes the effect of cavitation bubbles on gas mass transfer. It should be noted that d is a partial derivative operator.

[0131] A quantitative relationship between cavitation intensity and gas mass transfer process was established. The new relationship expression based on experimental data and theoretical analysis solved the fundamental problem that the traditional gas-liquid mass transfer model could not accurately describe the influence of ultrasonic cavitation. The model can accurately predict the TDG reduction efficiency under ultrasonic cavitation conditions, providing a reliable theoretical basis for parameter optimization.

[0132] S233: The finite volume method is used to solve the gas-liquid interface mass transfer kinetic model for different ultrasonic parameter combinations, and the adaptive grid refinement technology is used to improve the calculation accuracy in the high gradient area to obtain the time-space resolved gas concentration evolution field

[0133] S234: Based on the gas concentration evolution field, the average TDG reduction percentage after 5 minutes, 15 minutes and 30 minutes of treatment is calculated to form a TDG reduction efficiency prediction value matrix.

[0134] S24: Based on the predicted value of TDG reduction efficiency, the ultrasonic control target parameters are generated by multi-objective optimization algorithm, taking into account energy consumption, fish safety threshold and processing rate.

[0135] S241: Establish multi-objective optimization function:

[0136] F(p) = w1·E TDG (p) - w2·P cons (p) - w3·R fish (p);

[0137] Where p is the ultrasonic parameter vector, including frequency, power, pulse width and repetition frequency; E TDG is the TDG reduction efficiency; P cons is the energy consumption per unit reduction efficiency; R fish is the fish risk index; w1, w2 and w3 are weight coefficients.

[0138] S242: An improved particle swarm optimization algorithm is used to solve the multi-objective optimization problem, and an adaptive inertia weight and a dynamic shrinkage factor are introduced to improve the convergence speed and accuracy, and a parameter optimization candidate set is obtained;

[0139] S243: Apply robustness analysis to the parameter optimization candidate set, evaluate the stability of each set of parameters by adding perturbations to the hydrological parameters, and calculate parameter sensitivity indices;

[0140] S244: Based on the parameter sensitivity index and the optimization target value, select the best parameter combination as the ultrasonic control target parameter, including the optimal frequency, power density, pulse width and repetition frequency;

[0141] S3: Use ultrasound to control target parameters and generate ultrasonic real-time execution parameters through an adaptive parameter mapping network.

[0142] S31: inputting the ultrasonic control target parameters and the TDG state feature vector, and preliminarily calculating the ultrasonic frequency-power-time mode through the parameter mapping neural network;

[0143] S311: Construct parameter mapping neural network architecture, using a hybrid structure of multi-layer perceptron + temporal attention mechanism:

[0144] Input layer: receives ultrasonic control target parameters (4-dimensional) and TDG state feature vector (n-dimensional);

[0145] Encoding layer: two parallel processing branches, encoding target parameters and state features respectively;

[0146] Attention layer: realizes dynamic modulation of state features to parameter mapping;

[0147] Mapping layer: 3-layer fully connected network, each layer uses LeakyReLU activation function;

[0148] Output layer: Generates ultrasonic frequency-power-time pattern, including base frequency, frequency modulation depth, power envelope and time allocation ratio;

[0149] S312: Inputting the ultrasonic control target parameters and the TDG state feature vector into the parameter mapping neural network, and obtaining the initial ultrasonic frequency-power-time mode through forward propagation calculation;

[0150] S313: Apply the constraint satisfaction layer to ensure that the generated ultrasonic frequency-power-time pattern satisfies the physical constraints:

[0151] Frequency constraint: f_min ≤ f(t) ≤ f max ;

[0152] Power constraint: 0 ≤ P(t) ≤ P max , and the average power does not exceed the equipment limit time constraint: the pulse width and interval meet the transducer safe operation requirements;

[0153] S314: performing parameter smoothing processing on the ultrasonic frequency-power-time pattern that meets the constraints, using cubic spline interpolation to ensure the continuity of parameter time evolution, avoiding equipment stress caused by mutations, and obtaining a smooth ultrasonic frequency-power-time pattern;

[0154] S32: Based on the ultrasonic frequency-power-time pattern and the real-time hydrological parameters, a fluid mechanics constraint compensator is applied to adjust the ultrasonic parameters to adapt to the complex flow field and generate corrected ultrasonic parameters;

[0155] S321: extracting real-time hydrological parameters from the TDG state characteristic vector, including velocity vector field, pressure gradient field and turbulence intensity field;

[0156] S322: Construct a fluid mechanics constraint compensator and establish a mapping relationship between fluid characteristics and ultrasonic propagation effects:

[0157] M(f, P, t, v*, grad p, k) = α(f, v*)·P + β(f, grad p)·t + γ(f,k)·Δf;

[0158] Wherein, f is frequency, P is power, t is pulse time, v* is flow velocity, grad p is pressure gradient, and k is turbulent kinetic energy; α, β and γ are compensation functions, which are used to dynamically adjust the power, time and frequency offsets respectively. S323: The ultrasonic frequency-power-time pattern and real-time hydrological parameters are input into the fluid mechanics constraint compensator to calculate the compensation coefficient for each spatial area.

[0159] S324: Based on the compensation coefficient, the ultrasonic frequency-power-time mode is spatially adaptively adjusted to obtain corrected ultrasonic parameters to ensure that a consistent acoustic field effect is maintained under different fluid conditions;

[0160] S33: using the corrected ultrasonic parameters and the water turbulence characteristic data, through the turbulence-cavitation coupling response model, predicting the cavitation occurrence location and intensity, and obtaining a cavitation response prediction diagram;

[0161] S331: extracting water turbulence characteristic data from the TDG state characteristic vector and combining it with the corrected ultrasonic parameters to construct the turbulence-cavitation initial conditions;

[0162] S332: Build a turbulence-cavitation coupled response model that combines turbulence simulation and cavitation dynamics:

[0163] dα v / dt + grad·(α v v*) = (ρ l / ρ v )·α l α v ·(3 / R)·√(2 / 3·|p - p v | / ρ l )·sign(p - p v ) + S turb ;

[0164] Among them, α v is the gas phase volume fraction, α l is the liquid volume fraction, R is the average bubble radius, p is the local pressure, and p v is the vaporization pressure, S turb is the turbulence induced source term, and the turbulence induced source term S is introduced turb , describing the role of turbulent pulsation in promoting the initiation and development of cavitation. The introduction of turbulence-induced source terms solves the key problem of inaccurate predictions of existing cavitation models in high turbulence environments. By reconstructing the core mechanism of the model, the knowledge of two complex fields, turbulence simulation and cavitation dynamics, is integrated to establish a new mathematical description. The cavitation prediction error of traditional models under high turbulence conditions is usually more than 60%, while this coupled model reduces the error to less than 15%.

[0165] S333: The turbulence field is calculated using the large eddy simulation (LES) method and coupled with the cavitation model to solve it. The PISO algorithm is used to process the pressure-velocity coupling, and the VOF method is used to track the gas-liquid interface to obtain the dynamic evolution of the cavitation distribution with time and space resolution.

[0166] S334: Based on the dynamic evolution of cavitation distribution, key indicators including cavitation volume fraction, bubble density distribution and collapse energy density are calculated and integrated into a cavitation response prediction map.

[0167] S34: Based on the cavitation response prediction diagram and TDG monitoring data, the real-time ultrasonic execution parameters, including pulse width, duty cycle, waveform envelope and phase modulation information, are generated through a waveform optimization algorithm;

[0168] S341: Divide the processing area into a plurality of sub-areas, extract local cavitation characteristics of each sub-area based on the cavitation response prediction map, and extract target TDG reduction requirements from TDG monitoring data;

[0169] S342: For each sub-region, construct a waveform optimization objective function:

[0170] J(w) = ∑(i=1 to N) ω i |E i(w) - E i target | 2 + λ·R(w);

[0171] Where w is the waveform parameter vector, E i is the cavitation effect intensity of the ith sub-region, E i target is the target intensity, ω i is the weight coefficient, R(w) is the regularization term, and λ is the regularization coefficient.

[0172] S343: Applying the gradient descent method combined with the simulated annealing algorithm to solve the waveform optimization problem, avoiding falling into the local optimal solution, and obtaining the optimal waveform parameters of each sub-region;

[0173] S344: Integrate the optimal waveform parameters of all sub-areas and generate the final ultrasonic real-time execution parameters through the waveform synthesis algorithm, including:

[0174] The base frequency and its modulation method (single frequency, swept frequency or multi-frequency combination);

[0175] Power output curve (including ramp-up, hold and decay phases);

[0176] Pulse sequence design (pulse width, interval, and grouping pattern);

[0177] Phase modulation information (for spatial directional control);

[0178] S4: Using ultrasonic real-time execution parameters and combining spatial directional control algorithms to achieve precise intervention in the TDG area

[0179] S41: Input the ultrasonic real-time execution parameters and TDG state eigenvector, and calculate the three-dimensional sound field distribution prediction through the sound field propagation model.

[0180] S411: extracting flow field information from the TDG state eigenvector and constructing a sound propagation medium characteristic matrix, including density distribution, sound velocity distribution and attenuation coefficient distribution;

[0181] S412: setting the sound source characteristics of the transducer based on the ultrasonic real-time execution parameters, including radiation pattern, frequency response and transmission power;

[0182] S413: Constructing a sound propagation model for non-uniform flow fields 2 p - (1 / c 2 )·d 2 p / dt 2 = -ρ0·grad·(a*·grad v*) - grad·(Q·grad p) + d 2 / dt 2 (η·p) + S(x,t);

[0183] Among them, p is the sound pressure, c is the sound velocity, ρ0 is the average density, a* is the fluid acceleration, v⃗* is the flow velocity, Q is the sound attenuation coefficient caused by the flow field, η is the medium inhomogeneity parameter, S(x, t) is the sound source term. By introducing the flow velocity gradient term and the medium inhomogeneity term, it is suitable for high turbulence and strong gradient flood discharge environment.

[0184] S414: Use pseudo-spectral method combined with adaptive perfectly matched layer (PML) technology to solve the sound field propagation model and obtain a complete three-dimensional sound field distribution prediction, including sound pressure amplitude, phase and intensity distribution;

[0185] S42: Based on the 3D sound field distribution prediction and TDG monitoring data, the TDG risk area identification algorithm is applied to determine the high priority intervention areas;

[0186] S421: Compare TDG monitoring data with fish gas bubble disease thresholds to construct a TDG risk level map, dividing the risk into four levels: low (100%-110%), medium (110%-120%), high (120%-130%) and very high (>130%);

[0187] S422: Construct risk assessment function:

[0188] R(x) = α·[TDG(x) - TDG_threshold] + + β·t_exp(x) + γ·B(x);

[0189] Where TDG(x) is the TDG concentration at position x, TDG_threshold is the safety threshold, t_exp is the exposure time, B(x) is the biosensitivity coefficient (related to the fish activity area), α, β and γ are weight coefficients, [...] + Represents the positive part

[0190] S423: Superimpose the three-dimensional sound field distribution prediction with the TDG risk level map to calculate the intervention efficiency index for each area:

[0191] E(x) = I cav (x)·R(x) / P input (x);

[0192] Among them, I cav is the predicted cavitation intensity, R is the risk assessment value, P input is the input power;

[0193] S424: applying a hierarchical clustering algorithm to group the spatial points according to the intervention efficiency index, and selecting a set of regions with the highest efficiency index as a high-priority intervention region;

[0194] S43: using the high priority intervention area information and the ultrasonic real-time execution parameters, and through the phased array dynamic focusing algorithm, generating an ultrasonic emission control instruction;

[0195] S431: Construct a phased array acoustic field control model and establish a mapping relationship between array element excitation parameters and acoustic field distribution:

[0196] p(x,t) = ∑ i=1 to N A i ·g i (t-τ i )·D i (x,ω)·e (-j(ωt - k|x-xi|) ) / |xx i |;

[0197] Where p is the sound pressure field, A i is the amplitude of the ith array element, g i is the pulse function, τ i is the delay, D i is the directivity function, ω is the angular frequency, k is the wave number, x i is the array element position.

[0198] S432: Construct a multi-objective sound field optimization problem for high priority intervention areas:

[0199] Objective 1: Maximize the cavitation intensity within the intervention area;

[0200] Goal 2: Minimize the sound field intensity in non-target areas;

[0201] Goal 3: Minimize total energy input;

[0202] S433: using a genetic algorithm combined with a gradient descent method to solve a multi-objective optimization problem, and obtaining optimal excitation parameters of each array element, including amplitude, phase, and delay;

[0203] S434: Based on the optimal excitation parameters and the ultrasonic real-time execution parameters, generate a hardware-level ultrasonic emission control instruction, including a specific drive signal for each transducer;

[0204] The phased array dynamic focusing algorithm not only takes into account the phase control in the traditional phased array system, but also introduces the flow field phase correction term and flow field amplitude correction function, as well as time-varying multi-objective optimization and robust control theory, to achieve precise sound field control in complex flow environments. Designed for the special environment of high dam flood discharge, it solves the problem of low control accuracy of traditional phased array systems in high flow velocity and high turbulence environments. The algorithm has greatly improved the sound field directional accuracy and energy focusing efficiency, and this performance improvement exceeds the level that can be achieved by conventional technical improvements.

[0205] S44: executing the ultrasonic emission control instruction and collecting cavitation response feedback data in real time for the next round of parameter adjustment;

[0206] S441: sending an ultrasonic emission control instruction to the ultrasonic drive system to start the sound field action process;

[0207] S442: Collect cavitation response feedback data in real time through the sensor network, including acoustic emission signals, cavitation noise spectrum and water body TDG change rate;

[0208] S443: Apply fast Fourier transform and wavelet analysis to process cavitation response feedback data and extract cavitation characteristic indicators, including cavitation noise energy, main frequency distribution and pulsation period;

[0209] S444: Compare the cavitation characteristic index with the expected target, calculate the sound field control deviation, and pass it to the parameter adjustment link of the next control cycle;

[0210] S5: Monitor the TDG reduction effect and optimize the parameters of the adaptive parameter mapping network through a deep reinforcement learning framework;

[0211] S51: Collect TDG reduction effect data, including TDG concentration changes before and after treatment, energy consumption data and fish stress response indicators;

[0212] S511: collecting TDG concentration data before and after treatment at different water depths and distances through a distributed TDG sensor network, and calculating a spatiotemporal reduction curve of TDG;

[0213] S512: recording energy consumption data of the ultrasound system through a power monitoring module, including instantaneous power, cumulative energy consumption, and energy conversion efficiency;

[0214] S513: Monitor the behavioral changes of fish in the water through underwater cameras and sonar systems, and calculate fish stress response indicators, including the frequency of escape behavior, changes in swimming patterns, and group dissolution;

[0215] S514: Integrate the above data to generate a complete TDG reduction effect data set;

[0216] S52: Calculate system performance evaluation indicators based on TDG reduction effect data and preset targets;

[0217] S521: Define TDG reduction efficiency index:

[0218] η TDG = ∑(i=1 to N) w i ·[TDG i ^before - TDG i after ] / ∑(i=1 to N) w i ·[TDG i before -TDG safe ] + ;

[0219] Among them, TDG i before and TDG i after are the TDG concentrations before and after treatment, safe is the safety threshold, w i is the weight coefficient, [...] + Indicates the positive part;

[0220] S522: Define energy efficiency indicators:

[0221] η energy = ΔTDG avg ·V water / E consumed ;

[0222] Among them, ΔTDG avg is the average TDG reduction percentage, V water is the volume of water to be treated, E consumed for the energy consumed;

[0223] S523: Define eco-friendly indicators:

[0224] η eco = 1 - ∑(i=1 to M) S i / (M·S max );

[0225] Among them, S i is the stress response index value of the i-th fish, S max is the maximum stress threshold, M is the number of indicators;

[0226] S524: Based on the above indicators, calculate the comprehensive evaluation index of system performance:

[0227] Js ystem = α·η TDG + β·η energy + γ·η eco ;

[0228] Among them, α, β, and γ are weight coefficients, which are adjusted according to the specific needs of the application scenario;

[0229] S53: using the system performance evaluation index as a reward signal, inputting it into the deep reinforcement learning framework, and updating the weight parameters of the adaptive parameter mapping network;

[0230] S531: Building a Deep Reinforcement Learning Environment:

[0231] State space: TDG state feature vector and current ultrasonic parameter combination;

[0232] Action space: The direction and magnitude of the adjustment of the adaptive parameter mapping network parameters;

[0233] Reward function: comprehensive evaluation index of system performance;

[0234] State transfer: Apply the adjusted parameters to perform the next cycle of TDG processing and observe the system changes;

[0235] S532: Implementing the Dual Deep Q Network (DDQN) algorithm:

[0236] Main network: used to select actions, input current state, and output Q values ​​of different actions;

[0237] Target network: used to evaluate action value, parameters are periodically copied from the main network;

[0238] Experience replay buffer: stores the <state, action, reward, next state> tuple;

[0239] ε-Greedy Strategy: Balance exploration and exploitation, gradually reducing the probability of exploration as learning progresses;

[0240] S533: Execute batch learning process:

[0241] Randomly sample batches of data from the experience replay buffer;

[0242] Calculate the mean square error loss between the target Q value and the current Q value;

[0243] Use the Adam optimizer to minimize the loss function and update the main network parameters;

[0244] Update the target network parameters every N steps;

[0245] S534: extracting the optimized network parameters and updating the weight parameters of the adaptive parameter mapping network;

[0246] S54: using the updated adaptive parameter mapping network to process a new round of TDG state feature vectors to optimize the ultrasonic parameter control of the next cycle;

[0247] S541: inputting the current TDG state feature vector into the updated adaptive parameter mapping network to generate new ultrasonic control target parameters;

[0248] S542: Evaluate the difference between the new parameters and the previous parameters, calculate the parameter adjustment amount, and ensure that the parameter changes are smooth and not drastic;

[0249] S543: applying a parameter smoothing filter to process the parameter adjustment amount to generate a smooth parameter adjustment curve to avoid system instability caused by parameter mutation;

[0250] S544: Apply the smooth parameter adjustment curve to the current parameter setting to achieve a new round of ultrasonic parameter optimization and control to complete closed-loop control.

[0251] Specially designed for the specific field of TDG control, including specific network structure, loss function and learning strategy, the system can continuously learn and optimize from actual processing results.

[0252] In some scenarios, the process of TDG real-time monitoring and risk positioning during flood discharge is as follows: TDG sensors are deployed in key areas downstream of flood discharge (such as fish migration channels and spawning areas) to monitor dissolved gas saturation in real time (such as triggering an early warning when TDG ≥ 110%). Combined with flood discharge flow, water pressure, and water temperature data, TDG oversaturation risk areas (such as flood discharge energy dissipation areas and deep-water vortex areas) are predicted.

[0253] In some scenarios, the process of starting the directional ultrasonic treatment system and adapting parameters is as follows: select ultrasonic parameters according to flood discharge conditions (flow rate, head height): low-frequency ultrasonic (20-50 kHz): for large-scale water areas, enhance cavitation effect and promote gas precipitation; high-frequency ultrasonic (100-500 kHz): for local high TDG areas, refine bubbles and accelerate dissolution equilibrium. Adjust the ultrasonic power and action time to match the water depth and flow rate (such as shortening the action time in rapids).

[0254] In some scenarios, the process of ultrasound inducing cavitation bubbles in the target water area is as follows: microjet impact: destroying the gas-liquid interface, causing supersaturated gases (such as N2, O2) to escape from the water; local high temperature and high pressure: breaking up gas clusters in the water and inhibiting the stabilization of the supersaturated state; turbulence enhancement: increasing the gas-liquid mass transfer rate and accelerating the diffusion of dissolved gases into the atmosphere. Directly reducing the TDG concentration in the water, alleviating tissue damage or death of fish caused by bubble disease.

[0255] Based on TDG sensors and fish behavior monitoring (such as sonar or camera) data, ultrasonic parameters are dynamically adjusted through a closed-loop control system. Combined with the flood gate opening and water flow model, TDG change trends are predicted, and the treatment system is started or shut down in advance. Edge computing is used to optimize energy consumption in real time to ensure continuous operation of the system during the flood season.

[0256] The problem of lack of pertinence in ultrasonic parameter control is solved by an adaptive parameter mapping network. The network combines a multi-layer perceptron and a temporal attention mechanism to dynamically adjust ultrasonic parameters according to real-time hydrological conditions. At the same time, the fluid mechanics constraint compensator ensures a stable sound field effect in a complex flow field, thus realizing intelligent adaptive control of ultrasonic parameters.

[0257] Through directional processing, ultrasound can reduce the interference of fish behavior, and low-frequency ultrasound can avoid damaging sensitive fish species. The cavitation effect reduces the TDG saturation by 20%-40% (reference to experimental data), which is significantly lower than the lethal threshold of fish (usually TDG ≥ 130%).

[0258] First, real-time hydrological parameters, TDG monitoring data and water turbulence characteristics data are collected, and key features are extracted using a multi-sensor data spatiotemporal alignment algorithm and nonlinear principal component analysis to construct a TDG state feature vector to provide comprehensive environmental information for parameter control. Then, a hybrid structure integrating a multi-layer perceptron and a temporal attention mechanism is designed to capture the time series characteristics of the TDG state feature vector and dynamically modulate the state characteristics to parameter mapping through the attention mechanism. The fluid mechanics constraint compensator designed next can dynamically adjust the ultrasonic parameters according to the characteristics of the complex flow field, establish a mapping relationship between fluid characteristics and ultrasonic propagation effects, and ensure that consistent acoustic field effects can be maintained in areas with different flow velocity changes, pressure gradients, and turbulence intensities. Finally, the multi-level waveform optimization objective function and the hierarchical waveform synthesis algorithm can accurately control the ultrasonic waveform at the macro, meso, and micro levels, and realize customized ultrasonic intervention for different TDG distribution characteristics.

[0259] The problem of insufficient understanding of the turbulence-cavitation coupling mechanism is solved by the turbulence-cavitation coupling response model. Specifically, by introducing the turbulence-induced source term, combined with the improved Rayleigh-Plesset equation and the stratified bubble group dynamics model, the influence of turbulence on the cavitation process is accurately described, and the cavitation prediction accuracy is significantly improved.

[0260] Among them, the improved Rayleigh-Plesset equation with the introduction of turbulence correction terms can accurately describe the dynamic behavior of bubbles under turbulent conditions, solving the problem of low prediction accuracy of traditional equations in high turbulence environments during flood discharge. The three-layer coupling model (micro layer, meso layer, macro layer) can comprehensively simulate the complete process from the behavior of a single bubble to the interaction between the bubble group and the fluid as a whole, covering multiple time scales from microseconds to seconds. By introducing the turbulence-induced source term, which describes the three main turbulence-induced cavitation mechanisms of pressure pulsation, interface shear, and vortex suction, the applicability of the cavitation model in high turbulence flood discharge environments is significantly improved. Finally, the gas-liquid interface mass transfer kinetics model with the introduction of the cavitation source term can accurately describe the influence of cavitation bubbles on gas mass transfer, and achieve accurate prediction of the TDG reduction process.

[0261] The problem of low spatial directional control accuracy is solved by the phased array dynamic focusing algorithm. The algorithm combines the improved phased array acoustic field control model and the TDG risk area identification algorithm, introduces the flow field phase correction term, compensates for the influence of flow velocity on sound wave propagation, and realizes the precise directional processing of high-risk TDG areas in high-velocity flood discharge environments. The flow field phase correction term and the flow field amplitude correction function are introduced to compensate for the influence of flow velocity on sound wave propagation, which is particularly suitable for sound field control in high-velocity flood discharge environments. The risk assessment function is constructed by combining TDG monitoring data and fish bubble disease thresholds to accurately identify high-risk areas that need to be treated first and realize accurate resource allocation. The four-layer optimization strategy (time-space decomposition, global optimization, local refinement and solution space exploration) can efficiently solve the multi-objective time-varying sound field optimization problem, and realize precise control of the sound field while ensuring computational efficiency. Robustness analysis ensures the stability of the optimization results in various flow velocity distribution and TDG distribution scenarios, and significantly improves the adaptability of the system in complex and changing environments.

[0262] The preferred embodiments of the present invention are described in detail above; however, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all belong to the protection scope of the present invention.

Claims

1. A method for rapidly warning and weakening the total dissolved gas supersaturation of a high dam flood discharge, characterized in that: The following steps are involved: Acquire multi-source monitoring data, and establish TDG state feature vectors through layered fusion preprocessing; Based on the TDG state eigenvector and combined with the multi-time scale bubble dynamics model, the ultrasonic control target parameters are calculated; Using ultrasound to control target parameters, the real-time execution parameters of ultrasound are generated through an adaptive parameter mapping network; Using ultrasonic real-time execution parameters and combining with spatial directional control algorithms, we can achieve precise intervention in the TDG area and collect cavitation response feedback data; Monitor the TDG mitigation effect and determine whether to send an alarm message. At the same time, based on the cavitation response feedback data, optimize the parameters of the adaptive parameter mapping network through the deep reinforcement learning framework for the next cycle processing.

2. The method according to claim 1, characterized in that The steps of acquiring multi-source monitoring data and establishing the TDG state feature vector through layered fusion preprocessing include: Collect real-time hydrological parameters, TDG monitoring data and water turbulence characteristics data; and use the multi-sensor data spatiotemporal alignment algorithm to synchronize and calibrate them to generate a calibration data set; Based on the calibration data set, key features are extracted through nonlinear principal component analysis to construct the TDG state feature vector.

3. The method according to claim 1, characterized in that Based on the TDG state eigenvector and combined with the multi-time scale bubble dynamics model, the steps of calculating the ultrasonic control target parameters include: Read the TDG state characteristic vector and use the improved Rayleigh-Plesset equation to calculate the local cavitation threshold; Based on the local cavitation threshold and TDG state eigenvector, the cavitation effect intensity distribution under different frequency and power combinations is predicted through the layered bubble group dynamics model. Using the cavitation effect intensity distribution and the gas-liquid interface mass transfer kinetic model, the predicted values ​​of TDG reduction efficiency under different parameter combinations are calculated. Based on the predicted value of TDG reduction efficiency, the ultrasonic control target parameters were generated through a multi-objective optimization algorithm, taking into account energy consumption, fish safety threshold and processing rate.

4. The method according to claim 2, characterized in that: The steps of using ultrasound to control target parameters and generating ultrasound real-time execution parameters through an adaptive parameter mapping network include: Read the ultrasonic control target parameters and TDG state feature vector, and preliminarily calculate the ultrasonic frequency-power-time mode through the parameter mapping neural network; Based on the ultrasonic frequency-power-time pattern and real-time hydrological parameters, a fluid mechanics constraint compensator is applied to adjust the ultrasonic parameters to adapt to the complex flow field and generate corrected ultrasonic parameters; By using the corrected ultrasonic parameters and water turbulence characteristic data, the turbulence-cavitation coupling response model is used to predict the location and intensity of cavitation and obtain the cavitation response prediction diagram. Based on the cavitation response prediction diagram and TDG monitoring data, ultrasonic real-time execution parameters are generated through waveform optimization algorithm.

5. The method according to claim 1, characterized in that The steps of using ultrasonic real-time execution parameters and combining spatial directional control algorithms to achieve precise intervention in the TDG area include: Read the ultrasonic real-time execution parameters and TDG state eigenvectors, and calculate the three-dimensional sound field distribution prediction through the sound field propagation model; Based on the 3D acoustic field distribution prediction and TDG monitoring data in the TDG state feature vector, the TDG risk area identification algorithm is applied to determine the high priority intervention areas; Using high-priority intervention area information and ultrasonic real-time execution parameters, the ultrasonic emission control instructions are generated through the phased array dynamic focusing algorithm; Execute ultrasonic emission control instructions and collect cavitation response feedback data in real time for the next round of parameter adjustment.

6. The method according to claim 1, characterized in that The steps to monitor the TDG reduction effect and optimize the parameters of the adaptive parameter mapping network through the deep reinforcement learning framework include: Collect TDG reduction effect data, including TDG concentration changes before and after treatment, energy consumption data and fish stress response indicators; Calculate system performance evaluation indicators based on TDG reduction effect data and preset goals; The system performance evaluation index is used as a reward signal and input into the deep reinforcement learning framework to update the weight parameters of the adaptive parameter mapping network; The updated adaptive parameter mapping network is used to process the new round of TDG state feature vectors to optimize the ultrasonic parameter control of the next cycle.

7. The method according to claim 3, characterized in that Based on the local cavitation threshold and TDG state eigenvector, the steps of predicting the cavitation effect intensity distribution under different frequency and power combinations through the layered bubble group dynamics model include: The spatial region is divided into multiple computational units, and each unit establishes a unit state descriptor based on the local cavitation threshold and the parameters in the TDG state feature vector; Construct a layered bubble group dynamics model, which includes three coupled levels: micro, meso and macro, to simulate the behavior of a single bubble, the interaction between a bubble group and the overall interaction between a bubble group and the fluid. An adaptive Monte Carlo algorithm is applied to the stratified bubble group dynamics model to calculate the bubble density distribution function and bubble collapse rate of each calculation unit for different combinations of ultrasonic frequency and power. The bubble density distribution function and bubble collapse rate of all calculation units are integrated, and the cavitation effect intensity distribution of the entire area is obtained through spatial weighted integration.

8. The method according to claim 3, characterized in that Using the cavitation effect intensity distribution and the gas-liquid interface mass transfer kinetic model, the steps for calculating the predicted value of TDG reduction efficiency under different parameter combinations include: Extract TDG monitoring data from TDG state feature vectors and combine them with cavitation effect intensity distribution to establish the initial gas-liquid equilibrium state at each spatial point; Construct a gas-liquid interface mass transfer kinetic model, including a cavitation source term that describes the effect of cavitation bubbles on gas mass transfer; For different combinations of ultrasonic parameters, the finite volume method is used to solve the gas-liquid interface mass transfer kinetic model, and the adaptive grid refinement technology is used to improve the calculation accuracy in the high gradient area to obtain the time-space resolved gas concentration evolution field. Based on the gas concentration evolution field, the average TDG reduction percentage after M minutes, N minutes and P minutes of treatment is calculated to form a TDG reduction efficiency prediction value matrix, M<N<P.

9. The method according to claim 4, characterized in that The steps of predicting the location and intensity of cavitation by using the modified ultrasonic parameters and the water turbulence characteristic data in the TDG state eigenvector through the turbulence-cavitation coupling response model and obtaining the cavitation response prediction map include: Extract water turbulence characteristic data from TDG state feature vectors, and combine with corrected ultrasonic parameters to construct turbulence-cavitation initial conditions, including turbulence field characteristics, acoustic field characteristics and basic fluid parameters; A turbulence-cavitation coupling response model combining turbulence simulation and cavitation dynamics is established. By introducing turbulence-induced source terms, the effect of turbulence pulsation on the initiation and development of cavitation is described. The turbulence field is calculated using the large eddy simulation method and coupled with the cavitation model to solve the pressure-velocity coupling and track the gas-liquid interface to obtain the dynamic evolution of the cavitation distribution with time and space resolution. Based on the dynamic evolution of cavitation distribution, key indicators such as cavitation volume fraction, bubble density distribution and collapse energy density are calculated and integrated into a cavitation response prediction map.

10. The method according to claim 4, characterized in that Based on the cavitation response prediction diagram and the TDG monitoring data in the TDG state characteristic vector, the steps of generating ultrasonic real-time execution parameters through the waveform optimization algorithm include: The treatment area is divided into multiple sub-areas based on the grid adaptive algorithm, and finer grids are used in areas with large TDG gradients and high cavitation sensitivity; For each sub-region, local cavitation characteristics and target TDG abatement requirements are extracted; For each sub-region, a multi-level waveform optimization objective function is constructed and solved; When solving, the particle swarm optimization algorithm is used in the global search stage and the L-BFGS algorithm is used in the fine stage, and the Pareto optimal analytical method is used to deal with the optimization conflicts between sub-regions; The optimal waveform parameters of all sub-regions are integrated, and the final ultrasonic real-time execution parameters are generated through a hierarchical waveform synthesis algorithm, including the power envelope design at the macro level, the pulse grouping mode at the meso level, and the fine structure of a single pulse at the micro level.

11. The method according to claim 5, characterized in that Based on the 3D sound field distribution prediction and the TDG monitoring data in the TDG state feature vector, the steps of applying the TDG risk area identification algorithm to determine the high priority intervention area include: The TDG monitoring data in the TDG state feature vector is compared with the fish bubble disease threshold to construct a TDG risk level map; Construct a risk assessment function that comprehensively considers TDG concentration, exposure time, and biosensitivity coefficient; The three-dimensional sound field distribution prediction is superimposed on the TDG risk level map to calculate the intervention efficiency index for each area; A hierarchical clustering algorithm was applied to group the spatial points according to the intervention efficiency index, and the set of areas with the highest efficiency index was selected as the high-priority intervention areas.

12. The method according to claim 5, characterized in that The steps of generating ultrasonic emission control instructions by using high priority intervention area information and ultrasonic real-time execution parameters through phased array dynamic focusing algorithm include: An improved phased array acoustic field control model is constructed, the mapping relationship between array element excitation parameters and acoustic field distribution is established, and flow field phase correction terms and flow field amplitude correction functions are introduced; For high-priority intervention areas, a multi-objective time-varying acoustic field optimization problem is constructed, with objectives including maximizing the cavitation intensity in the target area, minimizing the acoustic field intensity in the non-target area, minimizing the total energy input, and minimizing the boundary reflection. Design a hierarchical hybrid optimization algorithm to solve multi-objective optimization problems, including four levels: spatiotemporal decomposition, global optimization, local refinement, and solution space exploration; Based on multi-scenario simulation verification, the optimization results are analyzed for robustness, the performance variance index is calculated, and the robust control theory is applied to correct unstable solutions; Based on the optimal excitation parameters and ultrasonic real-time execution parameters, hardware-level ultrasonic emission control instructions are generated, including carrier parameters, modulation parameters and grouping parameters, and digital pre-distortion technology is applied to compensate for the nonlinear response and temperature drift of the transducer.

13. The method according to claim 2, characterized in that Based on the calibration data set, the key features are extracted by nonlinear principal component analysis, and the steps of constructing the TDG state feature vector include: Adaptively normalize the calibration data set to generate a standardized data matrix; The standardized data matrix is ​​projected into the high-dimensional feature space through dynamic kernel function mapping, and the dynamic kernel function automatically adjusts the parameters according to the data distribution characteristics; Calculate the feature importance matrix in the high-dimensional feature space, and select the optimal feature subset based on the matrix using a recursive feature elimination algorithm; Applying improved kernel principal component analysis to the selected feature subset to extract nonlinear principal components; The static and dynamic features are fused through spatiotemporal correlation analysis to construct the final TDG state feature vector, which includes the TDG spatial distribution, time evolution trend and ultrasonic treatment response characteristics.

Images