A cold source system monitoring decision method and device, a storage medium and an electronic device

Abstract

This invention discloses a method, apparatus, storage medium, and electronic device for monitoring and decision-making in a cold source system. The method includes collecting hydrological data, sensor data, blockage data, and cold source CFD simulation data. The sensor data includes acoustic-optical images, underwater acoustic signals, and cold source system operating data. The method constructs a cold source monitoring model, a cold source prediction model, and an auxiliary decision-making model. The hydrological data, sensor data, and cold source CFD simulation data are input into different branches of the cold source monitoring model and fused to obtain blockage monitoring data. The blockage monitoring data and a portion of the cold source system operating data are input into the cold source prediction model to obtain blockage prediction data. The blockage prediction data and a portion of the cold source system operating data are input into the auxiliary decision-making model to obtain cold source system operating decision data.

Description

Technical Field

[0001] This invention relates to the field of cold source system operation monitoring technology, specifically to a cold source system monitoring and decision-making method, device, storage medium, and electronic device. Background Technology

[0002] Nuclear power plant cooling systems consist of multiple subsystems, including the intake, sludge trap, circulating water system, and condenser. These subsystems are interconnected during operation and are influenced by multiple factors, such as the hydrological environment, marine biological activity, and unit operating conditions, exhibiting strong nonlinearity and multi-scale dynamic characteristics. Due to the complex and continuously changing system boundary conditions, current operating models still mainly rely on empirical procedures and local adjustments, lacking the ability to uniformly model and collaboratively analyze multi-source information. This makes it difficult to achieve accurate perception, dynamic prediction, and optimized control of the cooling system's operating status.

[0003] Existing monitoring technologies for nuclear power plant cold source systems suffer from at least the following problems: Cold source system operational data, hydrological environmental data, and biological invasion-related data are scattered, including data types such as images, acoustic signals, flow field parameters, and equipment operating parameters. These data exhibit significant multi-source heterogeneity and lack a unified feature representation and fusion mechanism, making it difficult to generate complete and continuous monitoring results for the system status. Furthermore, existing monitoring methods typically rely solely on single-modal data or local sensor information for blockage identification, failing to comprehensively consider the influence of hydrological conditions and CFD flow field characteristics on blockage distribution and migration. Consequently, they cannot accurately characterize the role of blockages in the cold source system. The spatial distribution and evolution of blockages in the system; for the dynamic relationship between the operating status of the cold source system and the development trend of blockages, existing methods lack effective time series modeling and prediction means, making it difficult to predict the future evolution of blockages and their impact on the operation of the circulating water system in advance; existing technologies usually rely on static operating procedures or human experience for control decisions, lacking a system-level optimization decision-making mechanism based on prediction results, and cannot achieve unified generation and coordinated optimization of risk warning levels, circulating water control parameters, filtration strategies and emergency response measures, resulting in response lag or inaccurate control under complex operating conditions and external disturbances.

[0004] Patent document CN116580537A discloses a method and system for monitoring and early warning of the status of heat sinks in nuclear power plants. The method includes acquiring multi-source monitoring data. First, it determines the alarm level for water intake environmental monitoring based on relevant data of marine water intake risk indicators, and then issues an alarm based on this alarm level. Next, it determines the alarm level for water intake facility status monitoring based on relevant data of water intake facility status indicators, and then issues an alarm based on this alarm level. Then, it generates a monitoring status curve based on the multi-source monitoring data. Finally, it inputs the monitoring status curve into a prediction and early warning model with deep learning capabilities, and uses the prediction and early warning model to issue an early warning for the heat sink status. However, this method does not address the problem that the difficulty in unifying and integrating multi-source system data in nuclear power plant cold source systems prevents the prediction results of blockages from being combined with other information for collaborative optimization of system decisions.

[0005] Patent document CN121279501A discloses a method for predicting the spread of marine nuclear pollution based on multimodal data fusion. The method includes constructing a marine nuclear pollution dataset, which comprises marine environmental data and marine nuclear pollution data; introducing graph-structured modeling based on the dataset, aggregating node features through graph convolution, and assigning cross-modal attention weights to the features; inputting the assigned features into ConvNeXt V2 for feature extraction to generate a feature map, and constructing a diffusion initialization network based on nuclear pollution features; constructing a diffusion evolution network, and based on the diffusion initialization network, combining continuously acquired marine environmental data and marine nuclear pollution data, obtaining nuclear pollution spread prediction results at different times through three diffusion evolutions, generating a prediction map, and calculating the centroid trajectory of the nuclear pollution concentration distribution after each diffusion prediction; however, it does not address the problem that the difficulty in unifying and fusing multi-source system data in nuclear power cold source systems prevents the prediction results of blockages from being combined with other information for collaborative optimization of system decisions.

[0006] In summary, neither of the two existing patents mentioned above solves the problem that the data from multiple systems in a nuclear power plant's cold source system are difficult to integrate and unify, making it impossible to combine blockage prediction results with other information for collaborative optimization of system decisions. Summary of the Invention

[0007] Based on the above-mentioned technical problems, this invention proposes a method, device, storage medium, and electronic device for monitoring and decision-making in a cold source system, which solves the problem that it is difficult to unify and integrate data from multiple sources in a nuclear power cold source system, making it impossible to combine blockage prediction results with other information for collaborative optimization of system decision-making.

[0008] To achieve the above objectives, this invention proposes a monitoring and decision-making method for cold source systems.

[0009] A method for monitoring and making decisions about a cold source system, comprising: Collect hydrological data, sensor data, blockage data, and cold source CFD simulation data. The sensor data includes acoustic-optical images, underwater acoustic signals, and cold source system operation data. Construct a cold source monitoring model, a cold source prediction model, and an auxiliary decision-making model, train them, and save them. The cold source monitoring model includes an identification branch and a feature extraction branch. The hydrological data, sensor data, and cold source CFD simulation data are input into different branches of the cold source monitoring model and then mapped and fused to obtain blockage monitoring data. The blockage monitoring data and a portion of the cold source system operation data are input into the cold source prediction model to obtain blockage prediction data. The blockage prediction data and a portion of the cold source system operation data are input into the auxiliary decision model to obtain cold source system operation decision data. The cold source system operation decision data includes at least one or more of the following: risk warning level, circulating water system control parameters, filtration system operation strategy, condenser cleaning strategy, and emergency response measures.

[0010] Furthermore, the hydrological data, sensor data, and cold source CFD simulation data are input into different branches of the cold source monitoring model and mapped and fused to obtain blockage monitoring data. This includes: inputting a portion of the hydrological data and sensor data into the identification branch to obtain blockage identification data; inputting the cold source CFD simulation data into the feature extraction branch to obtain cold source feature data; and mapping and fusing the blockage identification data and the cold source feature data to obtain the blockage monitoring data.

[0011] Furthermore, the cold source monitoring model also includes a fusion layer, which fuses the blockage identification data and the cold source feature data based on one of splicing, weighting, or attention mechanisms to obtain the blockage monitoring data.

[0012] Furthermore, the cold source monitoring model is constructed, including: constructing the network architecture of the identification branch and the feature extraction branch; training the identification branch by dividing the historical data in the collected hydrological data into training and validation sets; training the feature extraction branch by dividing the historical data in the cold source CFD simulation data into training and validation sets; and combining the trained identification branch, the feature extraction branch, and the fusion layer to form the cold source monitoring model.

[0013] Furthermore, the blockage identification data includes: biological species, confidence level, and approximate biomass.

[0014] Furthermore, the operating data of the cold source system includes: predictive data and decision constraint data; the predictive data includes circulating water flow rate, pressure, temperature, unit load, back pressure, circulating pump speed and condenser cleanliness status; the decision constraint data includes the strainer tension, intake flow field and intake flow velocity.

[0015] Further, inputting the blockage monitoring data and part of the cold source system operation data into the cold source prediction model to obtain blockage prediction data includes: inputting the prediction data and the blockage monitoring data into the cold source prediction model to obtain blockage prediction data, wherein the blockage prediction data is the blockage density, blockage spatial distribution, blockage outbreak probability and blockage migration trajectory corresponding to different timestamps in the next 24 hours.

[0016] Furthermore, inputting the blockage prediction data and part of the cold source system operation data into the auxiliary decision model to obtain cold source system operation decision data includes: inputting the decision constraint data and the blockage prediction data into the auxiliary decision model to obtain cold source system operation decision data.

[0017] Furthermore, the cold source prediction model includes constructing a multi-source data encoder, a trajectory extrapolation network, and a dynamic state prediction network, wherein the inputs of the trajectory extrapolation network and the dynamic state prediction network are the outputs of the multi-source data encoder.

[0018] Furthermore, the trajectory extrapolation network includes a temporal memory layer, which is used to train the trajectory extrapolation network by constructing a training set based on historical data in the blockage data and the corresponding hydrological data, and to save the parameters of the trajectory extrapolation network.

[0019] Furthermore, the blockage prediction data also includes debris barrier prediction data, which is obtained based on the dynamic state prediction network.

[0020] Furthermore, inputting the blockage prediction data and part of the cold source system operation data into the auxiliary decision model to obtain cold source system operation decision data also includes inputting the debris barrier prediction data and the blockage prediction data into the auxiliary decision model to obtain cold source system operation decision data.

[0021] To achieve the above objectives, the present invention also proposes a monitoring and decision-making device for a cold source system.

[0022] A cold source system monitoring and decision-making device, comprising: The acquisition module is used to acquire hydrological data, sensor data, blockage data, and cold source CFD simulation data. The sensor data includes acoustic-optical images, underwater acoustic signals, and cold source system operation data. The module is used to build a cold source monitoring model, a cold source prediction model, and an auxiliary decision-making model, and saves them after training. The cold source monitoring model includes an identification branch and a feature extraction branch. The monitoring module is used to input the hydrological data, the sensor data, and the cold source CFD simulation data into different branches of the cold source monitoring model and then map and fuse them to obtain blockage monitoring data. The decision module is used to input the blockage monitoring data and part of the cold source system operation data into the cold source prediction model to obtain blockage prediction data, and input the blockage prediction data and part of the cold source system operation data into the auxiliary decision model to obtain cold source system operation decision data. The cold source system operation decision data includes at least one or more of the following: risk warning level, circulating water system control parameters, filtration system operation strategy, condenser cleaning strategy, and emergency response measures.

[0023] Based on the above technical solution, the present invention has at least the following beneficial effects: 1. This invention proposes a method, device, storage medium, and electronic device for monitoring and decision-making in a cold source system. By introducing a multi-source information collaborative modeling mechanism based on hydrological data, sensor data, blockage data, and cold source CFD simulation data, the cold source system can achieve structured expression and correlation analysis of heterogeneous data within a unified framework. By setting up a cold source monitoring model that includes identification and feature extraction branches, it achieves joint representation based on physical simulation information and actual monitoring information, enabling the blockage state information to establish a correlation mapping relationship with the cold source flow field characteristics, thus improving the completeness of the representation of the spatial distribution of blockages and their impact on the system. By introducing a cold source prediction model, it achieves modeling of the evolution trend of blockage state over time and its coupling relationship with the operating state of the cold source system, giving the system a forward-looking analysis capability for future operational risks. Through an auxiliary decision-making model, the prediction results and system operating data are comprehensively analyzed, enabling the system to output operating decisions that match the risk level, thereby improving the risk response capability and the consistency of system decision-making under complex operating conditions and external disturbances.

[0024] 2. This invention proposes a method, device, storage medium, and electronic device for monitoring and decision-making in a cold source system. By setting an identification branch and a feature extraction branch based on CFD simulation in the cold source monitoring model, the semantic information of blockages based on acoustic-optical images and underwater acoustic signals, and the physical information of the flow field based on cold source CFD simulation, can be subjected to targeted feature learning and structured expression respectively. This avoids information interference caused by different data modalities in the same processing path and improves the accuracy of feature extraction from multi-source heterogeneous data. Combined with a fusion layer, the blockage identification data and the cold source feature data are mapped and fused using splicing, weighting, or attention mechanisms. This enables data with different physical properties and heterogeneous sources to be aligned and associated in a unified feature space. This improves the consistency and characterization integrity between monitoring results such as blockage type, confidence level, and approximate biomass and the cold source flow field structure, and enhances the ability to perceive the distribution state of blockages in complex flow environments.

[0025] 3. This invention proposes a cold source system monitoring and decision-making method, device, storage medium, and electronic device. By constructing a cold source prediction model that includes a multi-source data encoder and a trajectory extrapolation network, and introducing a time-series memory layer to perform long-term dependency modeling on historical blockage movement data and hydrological data, the system can characterize the nonlinear migration law of blockages under hydrodynamic conditions, and achieve continuous prediction of blockage density, spatial distribution, outbreak probability, and migration trajectory on a 24-hour scale, thereby improving the forward-looking analysis capability of the spatiotemporal evolution trend of blockages.

[0026] 4. This invention proposes a method, device, storage medium, and electronic equipment for monitoring and decision-making in a cold source system. By inputting blockage prediction data, cold source system operation data, decision constraint data, and debris barrier prediction data into an auxiliary decision-making model, the model simultaneously considers the coupling relationship between the external blockage evolution state and the internal debris barrier operation constraints. This enables the coordinated output of risk warning levels, circulating water system control parameters, filtration strategies, and emergency response measures, improving the consistency, constraint matching, and reliability of multi-objective decision-making under complex operating conditions. Attached Figure Description

[0027] The accompanying drawings, which form part of this specification, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 A flowchart of a cold source system monitoring and decision-making method according to one embodiment is shown; Figure 2 A schematic diagram of the structure of a cold source system monitoring and decision-making device according to one embodiment is shown; Figure 3 A schematic diagram of the structure of a cold source system monitoring and decision-making product according to one embodiment is shown; Figure 4 A schematic diagram of the structure of an electronic device according to one embodiment is shown; Figure 5 A schematic diagram of the tension-clogging rate curve of a debris-blocking net according to one embodiment is shown; Figure 6 A schematic diagram of the blockage migration trajectory of one embodiment is shown. Detailed Implementation

[0028] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0029] The present invention will be further described in detail below with reference to specific embodiments, which should not be construed as limiting the scope of protection claimed by the present invention.

[0030] Example

[0031] To address the problem that the difficulty in unifying and integrating multi-source system data in nuclear power cold source systems prevents the prediction results of blockages from being combined with other information for collaborative optimization of system decisions, this invention proposes a cold source system monitoring and decision-making method, device, storage medium, and electronic equipment.

[0032] The embodiments of this application will be described in detail below with reference to the accompanying drawings and examples.

[0033] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0034] To achieve the above objectives, the present invention also proposes a monitoring and decision-making method for a cold source system.

[0035] like Figure 1 The present invention illustrates a method for monitoring and making decisions for a cold source system according to an embodiment of the present invention. The process mainly includes the following steps: S101: Collect hydrological data, sensor data, blockage data, and cold source CFD simulation data. The sensor data includes acoustic-optical images, underwater acoustic signals, and cold source system operation data.

[0036] Furthermore, the operating data of the cold source system includes predictive data and decision constraint data; wherein the predictive data includes circulating water flow rate, pressure, temperature, unit load, back pressure, circulating pump speed, and condenser cleanliness status; the decision constraint data includes the strainer tension, intake flow field, and intake flow velocity.

[0037] Furthermore, the hydrological data includes seawater temperature, tidal level, and current field data; Furthermore, the cold source CFD simulation data is computational fluid dynamics simulation data established based on the cold source system structural model, used to characterize the flow characteristics such as velocity distribution, pressure distribution, streamline distribution, turbulence characteristics, stagnation regions, and migration paths of blockages within the cold source system.

[0038] Furthermore, the blockage data includes biological species, the distribution of blockage quantity, and the movement trajectory of blockages.

[0039] S102: Construct a cold source monitoring model, a cold source prediction model, and an auxiliary decision-making model, train them, and save them. The cold source monitoring model includes an identification branch and a feature extraction branch.

[0040] Furthermore, a framework for the cold source monitoring model is constructed, including an identification branch and a feature extraction branch. The identification branch is used to extract blockage features, the feature extraction branch is used to extract cold source flow field features, and the fusion layer is used to establish the correlation between blockage features and cold source flow field features.

[0041] Furthermore, the cold source monitoring model includes a fusion layer, which fuses the blockage identification data and the cold source feature data based on one of splicing, weighting, or attention mechanisms to obtain the blockage monitoring data.

[0042] Specifically, training datasets were constructed for the cold source monitoring model, the cold source prediction model, and the auxiliary decision-making model, respectively. The dataset for the cold source monitoring model identification branch included collected acoustic and optical images, underwater acoustic signals, seawater temperature, seawater tide level, and corresponding blockage labeling information. The blockage labeling information included biological species, biomass, and tag information. 80% of the above was divided into the training set, 20% into the validation set, and blockage data from some monitoring data was collected as the test set. Adding seawater temperature and seawater tide level data enabled the model to acquire the characteristics of different seawater areas corresponding to different biological species, and to more accurately identify biological species. Specifically, the dataset corresponding to the feature extraction branch includes a training dataset constructed from the collected cold source CFD simulation data and the corresponding cold source feature data. 70% of the data is divided into a training set, 15% into a validation set, and 15% into a test set. The cold source feature data is a set of cold source system flow field structure features extracted based on the cold source CFD simulation data. The set of cold source system flow field structure features includes system state representation vectors, spatial topological features, and external disturbance response mapping relationships.

[0043] Specifically, the dataset corresponding to the cold source prediction model includes prediction data such as circulating water flow rate, pressure, temperature, unit load, back pressure, circulating pump speed, and condenser cleanliness status, as well as blockage prediction training data including biological species, blockage density, blockage spatial distribution, blockage outbreak probability, and blockage migration trajectory. 70% of the above is divided into a training set, 15% into a validation set, and 15% into a test set. The cold source prediction model is used to obtain the implicit coupling relationship between the cold source system and the blockage outbreak process, which can provide more accurate blockage prediction for subsequent auxiliary decision-making models.

[0044] Specifically, the dataset corresponding to the auxiliary decision-making model includes decision constraint data, preset blockage prediction data, and cold source output data. The decision constraint data includes the tension of the debris screen, the flow field at the water intake, and the flow velocity at the water intake. The preset blockage prediction data includes biological species, blockage density, spatial distribution of blockage, probability of blockage outbreak, and migration trajectory of blockage. The preset cold source output data includes risk warning level, circulating water system control parameters, filtration system operation strategy, condenser cleaning strategy, and emergency response measures. 70% of the above is divided into the training set, 15% into the validation set, and 15% into the test set.

[0045] Furthermore, in one embodiment, for the cold source monitoring model identification branch, 120,000 sets of samples, including historical acoustic and optical images, underwater acoustic signals, seawater temperature, seawater tide level, biological species, biomass, and tag information, are trained using the Adam optimizer with the following settings: initial learning rate: 0.001; batch size: 64; epoch: 100 rounds; Early trigger is activated when the error on the validation set no longer decreases for 10 consecutive rounds. The Stopping mechanism halts training and saves optimal parameters, outputting the species of organisms, the amount of debris retrieved per channel, and the total amount of debris. For the cold source monitoring model feature extraction branch, the cold source CFD simulation database and cold source feature data are labeled and constructed, including 50,000 sets of velocity field samples (nearshore flow velocity), 50,000 sets of pressure field samples (debris net tension), and 50,000 sets of streamline distribution samples (spatial distribution of blockages). Transfer learning is used for training, where the velocity field data includes nearshore flow velocity, the streamline distribution data includes spatial distribution of blockages, and the pressure field sample data includes debris net tension. The model is pre-trained for 50 rounds using the above flow field data, and then fine-tuned for 30 rounds using the cold source system CFD simulation data, finally obtaining the cold source feature extraction branch. The fusion layer includes a splicing layer and a weight calculation layer. The splicing layer splices different features entering the fusion layer into blockage monitoring data, and the weight calculation layer, i.e., the fully connected layer, calculates the weights of different features to obtain multiple weights.

[0046] Furthermore, the cold source prediction model includes a trajectory extrapolation network and a dynamic state prediction network, wherein the inputs of the trajectory extrapolation network and the dynamic state prediction network are the outputs of the multi-source data encoder.

[0047] Furthermore, the multi-source data encoder includes different input branches and feature alignment layers, including a system operation data branch and a congestion data branch. The system operation data branch is used to input prediction data, and the congestion data branch is used to input congestion monitoring data. The feature alignment layer is a fully connected layer used to fuse the data from the branches and output a multi-source encoded vector as the input to subsequent networks. The trajectory extrapolation network includes an input layer, a temporal memory layer, and a fully connected output layer. The input layer acquires congestion monitoring data corresponding to multiple timestamps, and the temporal memory layer adopts a recurrent neural network, including long short-term memory units. The network filters effective information through forget gates, input gates, and output gates. The fully connected layer maps the temporal features extracted by the long short-term memory unit to future spatiotemporal coordinates. The trajectory inference network can remember the long-term patterns of ocean currents and blockage drift (such as tidal influences) while filtering out short-term random fluctuation noise. The dynamic state prediction network includes a feature input layer, hidden layers, and an output layer. The feature input layer obtains the multi-source encoded vector output by the multi-source encoder. The hidden layer maps the multi-source encoded vector to a high-dimensional space using a Gaussian kernel function, fitting the tension-blockage rate curve. The tension-blockage rate curve of the debris barrier is shown below. Figure 5 As shown, the output layer directly outputs the predicted data of the debris barrier, including the debris barrier tension, blockage rate, water intake flow field, and water intake flow velocity.

[0048] Furthermore, a loss function is constructed for the cold source prediction model. The loss function can be the mean squared error loss function, and the Adam optimizer is used for backpropagation training. The initial learning rate is set to 1×10. - ³; During training, a cosine annealing strategy is used to gradually reduce the learning rate to 1×10. -5 The cold source prediction model is evaluated using a validation set. Training stops when the loss function of the validation set decreases below a threshold within a set number of consecutive rounds. The corresponding network parameters are then saved to obtain the trained cold source prediction model.

[0049] Furthermore, the auxiliary decision-making model includes a decision input layer, a temporal feature extraction layer, a two-branch regression prediction layer, and a decision evaluation layer. The decision input layer is used to acquire decision constraint data and blockage prediction data. In other embodiments, the debris barrier prediction data obtained by a dynamic state prediction network can be used instead of the decision constraint data. The temporal feature extraction layer includes a temporal convolutional network or a long short-term memory network to capture the dynamic coupling relationship between the blockage prediction data and the debris barrier tension and corresponding flow field velocity. The two-branch regression prediction layer includes a basic prediction branch and a measure intervention branch. The inputs to both branches are the outputs of the temporal feature extraction layer. The output of the basic prediction branch is a hidden representation of the current system state, expressed as follows: , in Based on the predicted output sequence, Based on the prediction mapping function, The output of the intervention branch is the system's natural evolution prediction, which represents the system's current state implicitly. , in To predict the output sequence, For the intervention mapping function, This is an implicit representation of the current state of the system. The measures control variables; the outputs of the above two branches are input into the decision evaluation layer, which includes the risk assessment head, control parameter head, and operation strategy decision head. The risk assessment head and control parameter head output the risk warning level and circulating water system regulation parameters. The operation strategy decision head outputs the filtration system operation strategy, condenser cleaning strategy, or emergency response measures. The decision evaluation layer outputs the following.

[0050] Furthermore, a multi-task joint training method is adopted for the auxiliary decision-making model. Loss functions are set for different branches, and the different loss functions are combined with the total loss function. Corresponding weights are set for each branch, and the parameters of the trained auxiliary decision-making model are saved.

[0051] S103: Input the hydrological data, the sensor data, and the cold source CFD simulation data into different branches of the cold source monitoring model and map and fuse them to obtain blockage monitoring data.

[0052] Furthermore, the identification branch is used to extract blockage-related features from hydrological and sensor data. Specifically, the input layer is used to obtain multi-source heterogeneous inputs, including acoustic-optical images, underwater acoustic signals, and hydrological data. The acoustic-optical images are input into a convolutional neural network, and the target contour, texture features, and morphological features in the images are extracted through convolutional and pooling layers. The underwater acoustic signals are input into a one-dimensional convolutional network or a recurrent neural network to extract acoustic reflection features and time-series features. The hydrological data, namely seawater temperature, seawater tide level, and current field data, are input into a fully connected network for feature mapping. Subsequently, the features extracted by each branch are fused, and blockage identification data is output through a classification head and a regression head. The blockage identification data includes biological species, identification confidence, and approximate biomass. The classification head outputs biological types, such as jellyfish or krill, which are possible aquatic organisms, while the regression head outputs confidence and approximate biomass.

[0053] Furthermore, the feature extraction branch is used to extract the flow characteristics of the cold source system from the cold source CFD simulation data; specifically, the cold source CFD simulation data is input into the input layer of the convolutional neural network, and the flow field distribution characteristics, velocity field characteristics, pressure field characteristics, and eddy current distribution characteristics are extracted through the convolutional layer, i.e., the feature extraction layer; then the extracted features are mapped to a unified feature space through a fully connected layer, and finally the output layer outputs the cold source feature data.

[0054] Specifically, in this embodiment, the blockage monitoring data output by the cold source monitoring model is shown in Table 1. The blockage monitoring data predicted at the corresponding timestamp includes biological species, net tension (kN), single-channel net retrieval volume (catties / hour), total net volume (catties / hour), spatial distribution of blockages, and nearshore flow velocity (m / s).

[0055] Table 1. Blockage Monitoring Data

[0056] S104: Input the blockage monitoring data and part of the cold source system operation data into the cold source prediction model to obtain blockage prediction data, and input the blockage prediction data and part of the cold source system operation data into the auxiliary decision model to obtain cold source system operation decision data. The cold source system operation decision data includes at least one or more of the following: risk warning level, circulating water system control parameters, filtration system operation strategy, condenser cleaning strategy, and emergency response measures.

[0057] Specifically, the data in Table 1 above are input into the blockage prediction model to obtain blockage prediction data for 24 hours, including biological species: jellyfish as the dominant species; spatial distribution of the culprits: diffusion and accumulation towards the intake area A and the area in front of the debris barrier; and the migration trajectory of the blockages as follows. Figure 6 As shown in the image.

[0058] Specifically, the above blockage prediction data is input into the decision support model, and the output includes the following: Risk level: Level III (high risk), corresponding risk score: 0.92; Output circulating water system control parameters: circulating water pump operating power increased by about 10%–15%, water intake flow increased by about 8%; Output filtration system operation strategy: the screen is switched to high-frequency cleaning operation mode, the front area enhanced filtration unit of the screen is activated, and the cleaning frequency is increased to more than 1.5 times the baseline operating condition; Output condenser cleaning strategy: the preventive online cleaning mechanism is triggered, the cleaning cycle is shortened from the conventional 72h to 48h, and the local flushing intensity is increased to enhanced mode; when the risk level reaches Level III and the outbreak probability exceeds 0.9, the output emergency response measures include: activating the backup water intake channel B area switching operation, limiting the unit load increase rate, linking and implementing the biological outbreak early warning mechanism, and activating the screen emergency enhanced interception mode.

[0059] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0060] Based on another aspect of the embodiments of this application, the present invention also provides a monitoring and decision-making device for a cold source system. For example... Figure 2 As shown, the device includes: The acquisition module 201 is used to acquire hydrological data, sensor data, blockage data and cold source CFD simulation data. The sensor data includes acoustic-optical images, underwater acoustic signals and cold source system operation data. Module 202 is used to build a cold source monitoring model, a cold source prediction model, and an auxiliary decision-making model, and save them after training. The cold source monitoring model includes an identification branch and a feature extraction branch. Monitoring module 203 is used to input the hydrological data, the sensor data and the cold source CFD simulation data into different branches of the cold source monitoring model and map and fuse them to obtain blockage monitoring data; The decision module 204 is used to input the blockage monitoring data and part of the cold source system operation data into the cold source prediction model to obtain blockage prediction data, and input the blockage prediction data and part of the cold source system operation data into the auxiliary decision model to obtain cold source system operation decision data. The cold source system operation decision data includes at least one or more of the following: risk warning level, circulating water system control parameters, filtration system operation strategy, condenser cleaning strategy, and emergency response measures.

[0061] As an optional solution, the above-mentioned device is also used to: input the hydrological data, the sensor data, and the cold source CFD simulation data into different branches of the cold source monitoring model and map and fuse them to obtain blockage monitoring data, including: inputting part of the hydrological data and the sensor data into the identification branch to obtain blockage identification data, inputting the cold source CFD simulation data into the feature extraction branch to obtain cold source feature data, and mapping and fusing the blockage identification data and the cold source feature data to obtain the blockage monitoring data.

[0062] As an optional solution, the above-mentioned device is also used to: the cold source monitoring model further includes a fusion layer, which fuses the blockage identification data mapping and the cold source feature data based on one of splicing, weighting or attention mechanisms to obtain the blockage monitoring data.

[0063] As an optional solution, the above-mentioned device is also used to: construct a cold source monitoring model, including: constructing the network architecture of the identification branch and the feature extraction branch; training the identification branch by dividing the historical data in the collected hydrological data into training and validation sets; training the feature extraction branch by dividing the historical data in the cold source CFD simulation data into training and validation sets; and combining the trained identification branch, the feature extraction branch, and the fusion layer to form the cold source monitoring model.

[0064] As an alternative, the above-mentioned device is also used for: the blockage identification data, including: biological species, confidence level, and approximate biomass.

[0065] As an optional solution, the above-mentioned device is also used for: cold source system operating data, including: predictive data and decision constraint data; the predictive data includes circulating water flow rate, pressure, temperature, unit load, back pressure, circulating pump speed and condenser cleanliness status; the decision constraint data includes the strainer tension, water intake flow field and water intake velocity.

[0066] As an optional solution, the above-mentioned device is also used to: input the blockage monitoring data and part of the cold source system operation data into the cold source prediction model to obtain blockage prediction data, including: inputting the prediction data and the blockage monitoring data into the cold source prediction model to obtain blockage prediction data, wherein the blockage prediction data is the blockage density, blockage spatial distribution, blockage outbreak probability and blockage migration trajectory corresponding to different timestamps in the next 24 hours.

[0067] As an optional solution, the above-mentioned device is further used to: input the blockage prediction data and part of the cold source system operation data into the auxiliary decision model to obtain cold source system operation decision data, including: inputting the decision constraint data and the blockage prediction data into the auxiliary decision model to obtain the cold source system operation decision data.

[0068] As an optional solution, the above-mentioned device is also used for: the cold source prediction model, including constructing a multi-source data encoder, a trajectory extrapolation network and a dynamic state prediction network, wherein the inputs of the trajectory extrapolation network and the dynamic state prediction network are the outputs of the multi-source data encoder.

[0069] As an optional solution, the above-mentioned device is also used for: the trajectory extrapolation network including a temporal memory layer, constructing a training set based on historical data in the blockage data and the corresponding hydrological data to train the trajectory extrapolation network, and saving the parameters of the trajectory extrapolation network.

[0070] As an optional solution, the above-mentioned device is also used to: the blockage prediction data further includes debris barrier prediction data, the debris barrier prediction data being obtained based on the dynamic state prediction network.

[0071] As an optional solution, the above-mentioned device is also used to: input the blockage prediction data and part of the cold source system operation data into the auxiliary decision model to obtain cold source system operation decision data, and further includes inputting the debris net prediction data and the blockage prediction data into the auxiliary decision model to obtain cold source system operation decision data.

[0072] In this application embodiment, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0073] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0074] According to one aspect of this application, a computer program product is provided, the computer program product comprising a computer program.

[0075] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0076] Figure 3 A schematic block diagram of a computer system architecture for implementing an electronic device according to embodiments of the present application is shown.

[0077] It should be noted that, Figure 3 The computer system 300 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0078] like Figure 3 As shown, the computer system 300 includes a central processing unit (CPU) 301, which is coupled to a read-only memory (ROM) 302. The CPU 301 can perform various appropriate actions and processes based on programs stored in the ROM or loaded from storage section 308 into the random access memory (RAM). The RAM also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output interface 305 (I / O interface) is also connected to the bus 304.

[0079] The following components are connected to the input / output interface 305: Input section 306, including but not limited to one or more of the following: keyboard, mouse, touch screen, touchpad, buttons, microphone, camera, biometric device, and various sensors; Output section 307, including but not limited to one or more of the following: display (e.g., liquid crystal display (LCD), organic light-emitting diode display (OLED), mini-LED display, micro-LED display, e-ink screen, head-up display (HUD), etc.), speaker, buzzer, indicator light, vibration device, etc.); Storage section 308, including but not limited to hard disk, solid-state drive, flash memory, etc. (the specific types can be found in the relevant description of memory in this application); and Communication section 309, including one or more of the following: wired communication modules (e.g., local area network (LAN) card, modem, Ethernet interface, CAN bus interface, USB interface, serial interface, etc.) and / or wireless communication modules (e.g., Wi-Fi module, Bluetooth module, Zigbee module, LoRa module, 4G / 5G communication module, NB-IoT module, etc.). The communication unit 309 performs communication processing via networks such as the Internet, local area network, wide area network, and vehicle networks (such as CAN bus, vehicle Ethernet). The driver 310 can also be connected to the input / output interface 305 as needed. Removable media 311, such as disks, optical disks, magneto-optical disks, semiconductor memory, USB flash drives, memory cards (such as SD cards, TF cards), etc., can be installed on the driver 310 as needed so that computer programs read from them can be installed into the storage unit 308 as required.

[0080] Specifically, according to embodiments of this application, the processes described in the various method flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit 301, it performs various functions defined in the system of this application.

[0081] According to another aspect of the embodiments of this application, an electronic device for a monitoring and decision-making method for a cold source system is also provided. This embodiment uses this electronic device as an example of a terminal device. Figure 4As shown, the electronic device includes a memory 402 and a processor 404. The memory 402 stores a computer program, and the processor 404 is configured to execute the steps in any of the above method embodiments via the computer program.

[0082] Optionally, in this embodiment, the aforementioned electronic device may be located in at least one of a plurality of network devices in a computer network.

[0083] Optionally, in this embodiment, the processor may be configured to execute the methods in the embodiments of this application via a computer program.

[0084] Alternatively, as those skilled in the art will understand, Figure 4 The structure shown is for illustrative purposes only. Figure 4 This does not limit the structure of the aforementioned electronic devices. For example, the electronic device may also include components that are more... Figure 4 The more or fewer components shown (such as network interfaces, etc.), or having the same Figure 4 The different configurations shown.

[0085] The memory 402 can be used to store software programs and modules, such as the program instructions / modules corresponding to the cold source system monitoring and decision-making method and device in this embodiment. The processor 404 is coupled to the memory 402. By running the software programs and modules stored in the memory 402, various functional applications and data processing are executed, thereby realizing the aforementioned cold source system monitoring and decision-making method. In some instances, the memory 402 may further include a memory remotely located relative to the processor 404. These remote memories can be connected to a terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Specifically, the memory 402 can be used, but is not limited to, to store collected hydrological data, sensor data, blockage data, and cold source CFD simulation data. As an example, such as... Figure 4 As shown, the memory 402 may include, but is not limited to, the acquisition module 201, construction module 202, monitoring module 203, and decision module 204 of the aforementioned cold source system monitoring and decision-making device. Furthermore, it may include, but is not limited to, other module units in the aforementioned device, which will not be elaborated upon in this example.

[0086] Optionally, the transmission device 406 described above is used to receive or send data via a network. Specific examples of the network described above may include wired networks and wireless networks. In one example, the transmission device 406 includes a Network Interface Controller (NIC), which can be connected to other network devices and routers via a network cable to communicate with the Internet or a local area network. In another example, the transmission device 406 is a Radio Frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0087] In addition, the aforementioned electronic device also includes: a display 408 for displaying the aforementioned hydrological data, sensor data, blockage data and cold source CFD simulation data; and a connection bus 410 for connecting the various module components in the aforementioned electronic device.

[0088] In the embodiments of this application, the processor can be any hardware unit with data processing and instruction execution capabilities, including but not limited to a central processing unit (CPU), microprocessor, microcontroller unit (MCU), digital signal processor (DSP), graphics processing unit (GPU), neural processing unit (NPU), tensor processing unit (TPU), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), complex programmable logic device (CPLD), system-on-chip (SoC), microcontroller, or any combination of the above devices. The processor can be a single-core processor or a multi-core processor; it can be integrated into a single chip or distributed across multiple chips or multiple physical devices.

[0089] In embodiments of this application, the memory may include random access memory (RAM), such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), etc. The memory may also include non-volatile memory, such as read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory (including but not limited to NAND Flash, NOR Flash), solid-state drive (SSD), embedded multimedia card (eMMC), universal flash storage (UFS), magnetic storage devices (such as hard disks), ferroelectric memory (FRAM), magnetoresistive memory (MRAM), phase-change memory (PCM), resistive random access memory (ReRAM), or one or more other non-volatile solid-state storage media.

[0090] In this application, "memory," "computer-readable storage medium," and "computer-readable medium" refer to the same or related concepts and can be used interchangeably. Specifically, when the aforementioned types of memory are used to store computer program instructions and / or data, they constitute the computer-readable storage medium as referred to in this application; conversely, the physical implementation of the computer-readable storage medium as referred to in this application can be any one or more of the aforementioned types of memory. Those skilled in the art should understand that the computer-readable storage medium can be a single medium or a distributed medium (e.g., a combination of storage media distributed across multiple physical devices or multiple chips).

[0091] In this application, the computer-readable storage medium is a non-transitory computer-readable storage medium, which refers to a physical storage device capable of storing information in a non-transitory manner. This includes, but is not limited to, various types of random access memory, read-only memory, flash memory, solid-state drives, magnetic storage devices, and other storage media with physical form, but excludes transiently propagating signals themselves (such as electromagnetic carrier waves, modulation signals, etc.). Those skilled in the art should understand that "non-transitory" here means that the storage medium has a persistent physical form relative to transiently propagating signals, and does not limit the storage medium to be non-volatile memory; although the aforementioned types of random access memory are volatile memories, they still fall under the category of non-transitory computer-readable storage media as defined in this application.

[0092] According to one aspect of this application, a computer-readable storage medium is provided, wherein a processor of an electronic device reads computer instructions from the computer-readable storage medium, and executes the computer instructions, causing the electronic device to perform a cold source system monitoring and decision-making method provided in one of the various alternative implementations of the above-described cold source system monitoring and decision-making aspects.

[0093] Optionally, in this embodiment, the computer-readable storage medium described above may be configured to store methods for performing the embodiments of this application.

[0094] Optionally, in this embodiment, those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device, and the program can be stored in a computer-readable storage medium.

[0095] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0096] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause one or more electronic devices to execute all or part of the steps of the methods described in the various embodiments of this application.

[0097] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0098] In the several embodiments provided in this application, it should be understood that the disclosed application can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0099] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0100] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0101] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0102] In summary, as can be seen from the above description, the embodiments of the present invention achieve the following technical effects: 1. This invention proposes a method, device, storage medium, and electronic device for monitoring and decision-making in a cold source system. By introducing a multi-source information collaborative modeling mechanism based on hydrological data, sensor data, blockage data, and cold source CFD simulation data, the cold source system can achieve structured expression and correlation analysis of heterogeneous data within a unified framework. By setting up a cold source monitoring model that includes identification and feature extraction branches, it achieves joint representation based on physical simulation information and actual monitoring information, enabling the blockage state information to establish a correlation mapping relationship with the cold source flow field characteristics, thus improving the completeness of the representation of the spatial distribution of blockages and their impact on the system. By introducing a cold source prediction model, it achieves modeling of the evolution trend of blockage state over time and its coupling relationship with the operating state of the cold source system, giving the system a forward-looking analysis capability for future operational risks. Through an auxiliary decision-making model, the prediction results and system operating data are comprehensively analyzed, enabling the system to output operating decisions that match the risk level, thereby improving the risk response capability and the consistency of system decision-making under complex operating conditions and external disturbances.

[0103] 2. This invention proposes a method, device, storage medium, and electronic device for monitoring and decision-making in a cold source system. By setting an identification branch and a feature extraction branch based on CFD simulation in the cold source monitoring model, the semantic information of blockages based on acoustic-optical images and underwater acoustic signals, and the physical information of the flow field based on cold source CFD simulation, can be subjected to targeted feature learning and structured expression respectively. This avoids information interference caused by different data modalities in the same processing path and improves the accuracy of feature extraction from multi-source heterogeneous data. Combined with a fusion layer, the blockage identification data and the cold source feature data are mapped and fused using splicing, weighting, or attention mechanisms. This enables data with different physical properties and heterogeneous sources to be aligned and associated in a unified feature space. This improves the consistency and characterization integrity between monitoring results such as blockage type, confidence level, and approximate biomass and the cold source flow field structure, and enhances the ability to perceive the distribution state of blockages in complex flow environments.

[0104] 3. This invention proposes a cold source system monitoring and decision-making method, device, storage medium, and electronic device. By constructing a cold source prediction model that includes a multi-source data encoder and a trajectory extrapolation network, and introducing a time-series memory layer to perform long-term dependency modeling on historical blockage movement data and hydrological data, the system can characterize the nonlinear migration law of blockages under hydrodynamic conditions, and achieve continuous prediction of blockage density, spatial distribution, outbreak probability, and migration trajectory on a 24-hour scale, thereby improving the forward-looking analysis capability of the spatiotemporal evolution trend of blockages.

[0105] 4. This invention proposes a method, device, storage medium, and electronic equipment for monitoring and decision-making in a cold source system. By inputting blockage prediction data, cold source system operation data, decision constraint data, and debris barrier prediction data into an auxiliary decision-making model, the model simultaneously considers the coupling relationship between the external blockage evolution state and the internal debris barrier operation constraints. This enables the coordinated output of risk warning levels, circulating water system control parameters, filtration strategies, and emergency response measures, improving the consistency, constraint matching, and reliability of multi-objective decision-making under complex operating conditions.

[0106] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0107] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0108] It should be noted that, in the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

Claims

1. A cold source system monitoring and decision method, characterized by, include: Collect hydrological data, sensor data, blockage data, and cold source CFD simulation data. The sensor data includes acoustic-optical images, underwater acoustic signals, and cold source system operation data. Construct a cold source monitoring model, a cold source prediction model, and an auxiliary decision-making model, train them, and save them. The cold source monitoring model includes an identification branch and a feature extraction branch. The hydrological data, sensor data, and cold source CFD simulation data are input into different branches of the cold source monitoring model and then mapped and fused to obtain blockage monitoring data. The blockage monitoring data and a portion of the cold source system operation data are input into the cold source prediction model to obtain blockage prediction data. The blockage prediction data and a portion of the cold source system operation data are input into the auxiliary decision model to obtain cold source system operation decision data. The cold source system operation decision data includes at least one or more of the following: risk warning level, circulating water system control parameters, filtration system operation strategy, condenser cleaning strategy, and emergency response measures.

2. The method of claim 1, wherein, The hydrological data, sensor data, and cold source CFD simulation data are input into different branches of the cold source monitoring model and then mapped and fused to obtain blockage monitoring data, including: The hydrological data and sensor data are input into the identification branch to obtain blockage identification data. The cold source CFD simulation data is input into the feature extraction branch to obtain cold source feature data. The blockage identification data and cold source feature data are mapped and fused to obtain blockage monitoring data.

3. The method of claim 2, wherein, The cold source monitoring model also includes a fusion layer, which fuses the blockage identification data and the cold source feature data based on one of splicing, weighting or attention mechanisms to obtain the blockage monitoring data.

4. The method according to claim 3, characterized in that, Construct a cold source monitoring model, including: The network architecture of the identification branch and the feature extraction branch is constructed. The identification branch is trained by dividing the historical data in the collected hydrological data into training set and validation set. The feature extraction branch is trained by dividing the historical data in the cold source CFD simulation data into training set and validation set. The trained identification branch, the feature extraction branch and the fusion layer are combined to form the cold source monitoring model.

5. The method according to claim 2, characterized in that, The obstruction identification data includes: Species, confidence level, and approximate biomass.

6. The method according to claim 1, characterized in that, Cold source system operating data, including: Forecast data and decision constraint data; The predicted data includes circulating water flow rate, pressure, temperature, unit load, back pressure, circulating pump speed, and condenser cleanliness status. The decision constraint data includes the tension of the debris barrier, the flow field at the water intake, and the flow velocity at the water intake.

7. The method according to claim 6, characterized in that, The blockage monitoring data and some of the cold source system operation data are input into the cold source prediction model to obtain blockage prediction data, including: The predicted data and the blockage monitoring data are input into the cold source prediction model to obtain the blockage prediction data. The blockage prediction data includes the blockage density, spatial distribution of blockage, probability of blockage outbreak, and migration trajectory of blockage at different timestamps in the next 24 hours.

8. The method according to claim 6, characterized in that, The blockage prediction data and a portion of the cold source system operation data are input into the auxiliary decision model to obtain cold source system operation decision data, including: The decision constraint data and the blockage prediction data are input into the auxiliary decision model to obtain the operating decision data of the cold source system.

9. The method according to claim 1, characterized in that, The cold source prediction model includes constructing a multi-source data encoder, a trajectory extrapolation network, and a dynamic state prediction network, wherein the inputs of the trajectory extrapolation network and the dynamic state prediction network are the outputs of the multi-source data encoder.

10. The method according to claim 9, characterized in that, The trajectory extrapolation network includes a temporal memory layer. The trajectory extrapolation network is trained by constructing a training set based on historical data in the blockage data and the corresponding hydrological data, and the parameters of the trajectory extrapolation network are saved.

11. The method according to claim 9, characterized in that, The blockage prediction data also includes debris barrier prediction data, which is obtained based on the dynamic state prediction network.

12. The method according to claim 11, characterized in that, The blockage prediction data and a portion of the cold source system operation data are input into the auxiliary decision model to obtain cold source system operation decision data, which also includes: The predicted data of the debris barrier and the predicted data of the blockage are input into the auxiliary decision model to obtain the operation decision data of the cold source system.

13. A device for monitoring and decision-making in a cold source system, characterized in that, include: The acquisition module is used to acquire hydrological data, sensor data, blockage data, and cold source CFD simulation data. The sensor data includes acoustic-optical images, underwater acoustic signals, and cold source system operation data. The module is used to build a cold source monitoring model, a cold source prediction model, and an auxiliary decision-making model, and saves them after training. The cold source monitoring model includes an identification branch and a feature extraction branch. The monitoring module is used to input the hydrological data, the sensor data, and the cold source CFD simulation data into different branches of the cold source monitoring model and then map and fuse them to obtain blockage monitoring data. The decision module is used to input the blockage monitoring data and part of the cold source system operation data into the cold source prediction model to obtain blockage prediction data, and input the blockage prediction data and part of the cold source system operation data into the auxiliary decision model to obtain cold source system operation decision data. The cold source system operation decision data includes at least one or more of the following: risk warning level, circulating water system control parameters, filtration system operation strategy, condenser cleaning strategy, and emergency response measures.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein the computer program, when executed by an electronic device, performs the method according to any one of claims 1 to 12.

15. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1 to 12.

16. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to execute the method according to any one of claims 1 to 12 through the computer program.

Images