Discharged water temperature abnormity identification and emergency processing method, device, equipment and medium

Abstract

The invention discloses a discharged water temperature abnormity identification and emergency processing method, device and equipment and a storage medium, and the method comprises the steps: collecting dam front vertical water temperature data, synchronously integrating data of a water quality sensor and a meteorological station, building a three-dimensional dynamic water temperature digital twinborn model based on a numerical model, real-time mapping and simulation of the fluid-heat-solid coupling effect of the reservoir area are achieved; setting a basic threshold according to a downstream ecological target, generating a dynamic threshold interval in combination with a seasonal probability distribution model, a meteorological disturbance factor and a water quality association rule, and dividing into multi-stage early warning according to water temperature deviation; when a preset early warning level is triggered, the digital twinborn model automatically adjusts the position of layered water taking equipment, a tail water mixing device or a solar heating system is started, and an ecological water supplementing strategy is linked; future water temperature changes are rehearsed through a digital twinborn model, and multiple sets of regulation and control plans are generated. According to the invention, a'monitoring-analysis-decision-execution 'full-closed-loop system is constructed, and accurate recognition and efficient processing of abnormal water temperature are realized.

Description

Technical Field

[0001] This invention relates to the fields of water conservancy engineering and ecological environment protection technology, and in particular to a method, device, equipment and medium for identifying and responding to abnormal discharge water temperature. Background Technology

[0002] In the field of water conservancy engineering, the construction and operation of reservoirs play a crucial role in the rational allocation and utilization of water resources. However, the temperature of water released from reservoirs has always been a significant factor affecting the downstream ecological environment, agricultural production, and the development of related industries.

[0003] Traditional methods for monitoring reservoir discharge water temperature primarily rely on single-point monitoring equipment. These devices typically only acquire water temperature information at specific locations, failing to comprehensively reflect the water temperature distribution across different depths and areas of the reservoir. Reservoir water temperature exhibits a clear vertical stratification, with the location and thickness of the thermocline dynamically changing with seasons, meteorological conditions, and the reservoir's operational status. Single-point monitoring cannot capture these complex variations, resulting in inaccurate and incomplete understanding of water temperature conditions, making it difficult to meet practical monitoring needs. Abnormal reservoir discharge water temperature has a significant impact on downstream ecosystems, agricultural irrigation, and fisheries production. Traditional methods suffer from problems such as limited monitoring (single-point sensors cannot capture vertical stratification), fixed thresholds (lack of integration of seasonal / meteorological / water quality factors), and delayed regulation (reliance on human experience). Summary of the Invention

[0004] The main purpose of this application is to provide a method, device, equipment and medium for identifying and handling abnormal discharge water temperature, which can solve the technical problems of incomplete monitoring, fixed thresholds and lack of intelligent emergency handling in the existing technology.

[0005] To achieve the above objectives, this application provides the following technical solution: a method for identifying and handling abnormal discharge water temperature, characterized in that the method includes: Collect vertical water temperature data in front of the dam, integrate water quality sensor and meteorological station data simultaneously, and construct a three-dimensional dynamic water temperature digital twin model based on numerical model to realize real-time mapping and simulation of the flow-heat-solid coupling effect in the reservoir area; Based on the downstream ecological objectives, a basic threshold is set, and a dynamic threshold range is generated by combining the seasonal probability distribution model, meteorological disturbance factors and water quality correlation rules. The range is divided into multiple warning levels according to the water temperature deviation. When the preset warning level is triggered, the digital twin model automatically adjusts the layer of the stratified water intake equipment, starts the tailwater mixing device or solar heating system, and links the ecological water replenishment strategy. By using digital twin models to predict future water temperature changes, multiple control plans are generated. Key data is stored on blockchain and combined with a remote operation and maintenance platform to achieve real-time monitoring and parameter adjustment of equipment.

[0006] As a further improvement to this application, the method further includes: High-frequency acquisition of vertical water temperature data, water quality parameters, and meteorological data; preliminary data processing through edge computing nodes to extract key feature parameters; acquisition of water quality parameters including at least conductivity, dissolved oxygen, and pH, and uploading to the data platform; meteorological data including air temperature, wind speed, sunshine duration, and precipitation, ensuring synchronous acquisition of meteorological data and water temperature data; The collected water temperature, water quality, and meteorological data from multiple sources were spatiotemporally aligned to unify timestamps and spatial coordinates, and a unified data format was constructed. The data was cleaned to remove noise and outliers. At the same time, derived features were extracted from the raw data to construct feature vectors, providing input for subsequent numerical models. The EFDC model was selected to construct the reservoir hydrodynamic module to simulate the water flow velocity and flow field distribution; the MIKEThermal module was coupled to construct the heat transfer model to simulate the heat conduction and convection process of water temperature; and the reservoir area was discretized using an unstructured grid. A 3D terrain model of the reservoir area is constructed based on a visualization platform, and real-time simulated water temperature field and flow field data are overlaid to realize a 3D visualization display of water temperature distribution and support interactive query. Historical data is used to calibrate model parameters so that the simulation results and measured data reach a preset degree of consistency. The model parameters are dynamically corrected by comparing the measured water temperature values ​​of the downstream monitoring section with the model output in real time.

[0007] As a further improvement to this application, the method further includes: Analyze the water temperature requirements of downstream ecologically sensitive objects and clarify the basic threshold ranges for different scenarios; Input historical water temperature data, seasonal attributes, typical meteorological parameters and water quality indicators, and extract seasonal features, meteorological features and water quality features to construct an ecological-water temperature correlation feature library; By using kernel density estimation or Gaussian mixture model, the historical water temperature data are fitted with a probability distribution according to the season to determine the mean and fluctuation range of water temperature in different seasons. Generate seasonal correction factors, form seasonally specific threshold offset rules, establish a meteorological-water temperature response model, and quantify the impact of sunshine duration and wind speed on water temperature. By using regression analysis or decision tree algorithms, the corrective weights of meteorological factors on the threshold are determined, and the threshold range is dynamically expanded or contracted. Define water quality-temperature coupling rules, use a rule engine to analyze water quality parameters in real time, and dynamically generate water quality correlation correction coefficients.

[0008] As a further improvement to this application, the multi-level early warning system includes: calculating the deviation between the measured water temperature and the dynamic threshold in real time, and dividing it into three levels of early warning based on the absolute value of the deviation; the method also includes setting a dynamic threshold update mechanism. The seasonal probability distribution model is updated regularly based on the latest data within the time period to adapt to interannual water temperature fluctuations; real-time reception of meteorological and water quality emergencies triggers immediate correction of threshold intervals, and the immediate correction completes the recalculation of deviation values ​​within a preset time period.

[0009] As a further improvement to this application, the method further includes: The digital twin model continuously receives data from various sensors. According to different warning levels, corresponding threshold values ​​for parameters such as water temperature, water quality, and water level are set. The digital twin model compares the collected data with the preset warning thresholds in real time. Once the data exceeds the threshold, the corresponding warning level is immediately triggered. The digital twin model automatically adjusts the stratification of the water intake equipment, including: The digital twin model uses its own simulation and analysis capabilities, combined with real-time collected water temperature data, to accurately identify the location of the thermocline in the reservoir and the water temperature distribution of different water layers. The target water temperature to be obtained is determined based on downstream ecological objectives, water demand, and early warning levels. Based on the target water temperature, the thermocline, and the water temperature distribution, the digital twin model calculates the optimal layer to which the stratified water intake device should be adjusted. By communicating with the device's control system, it automatically issues commands to control the device to move to the corresponding layer in order to obtain a water source close to the target water temperature. The stratified water intake device may include at least a valve valve and / or a stratified water intake port.

[0010] As a further improvement to this application, the method further includes: The digital twin model determines whether to activate the tailwater mixing device or the solar heating system based on the warning level and the current water temperature. If the water temperature deviates only slightly from the target value, the tailwater mixing device may be activated first. If the water temperature is severely low and the sunlight conditions permit, the solar heating system may be activated. When the tailwater mixing device needs to be started, the digital twin model sends a start command to the device's control system; the device usually mixes water of different temperatures thoroughly by disturbing the water flow to achieve a uniform water temperature. If the solar heating system is activated, the digital twin model calculates the required heating power based on real-time water temperature, light intensity, and other data, and sends corresponding control commands to the system; the solar heating system uses solar energy to heat the water and raise the water temperature to the target range; The ecological water replenishment strategy linkage includes: based on the warning level, downstream ecological water demand, and the current water level and flow of the reservoir, the digital twin model calculates the amount of water to be replenished and the appropriate time for water replenishment; The digital twin model sends control commands to the water replenishment equipment to adjust the equipment's opening degree or operating power, thereby achieving ecological water replenishment. During the water replenishment process, the digital twin model continuously monitors parameters such as water temperature, water level, and flow rate in the downstream water area to evaluate the water replenishment effect; if it is found that the water replenishment effect does not meet expectations, the water replenishment strategy is adjusted in a timely manner. The data and effects throughout the regulation process are fed back to the digital twin model, which then learns and optimizes itself based on this feedback.

[0011] As a further improvement to this application, the method of storing key data through blockchain and combining it with a remote operation and maintenance platform to achieve real-time monitoring and parameter adjustment of equipment includes: Based on the digital twin model, data such as vertical water temperature, meteorology, and equipment status are accessed in real time. The adjustable parameter combinations of the stratified water intake equipment opening degree and heating power are enumerated. The EFDC / MIKE model is driven to couple and simulate the water temperature evolution within a preset time period in the future. The LSTM model is superimposed to correct long-term prediction errors and generate water temperature change curves and thermocline migration trajectories at key downstream sections. Based on ecological indicators, at least three optimal plans are selected using the particle swarm optimization algorithm. The optimal plans include equipment control parameters such as the opening degree of the associated tongue valve and the height of the curtain wall. When a dynamic threshold triggers an alert, the system automatically extracts the alert level, control parameters, and verification data, encrypts and uploads them to the blockchain via smart contracts, completes the entire process of evidence storage, and supports blockchain traceability. The remote operation and maintenance platform integrates a 3D visualization interface to display the water temperature field, equipment status and pre-drill effect in real time. The manual mode supports parameter fine-tuning and one-click recall of preset plans. The automatic mode links equipment according to the warning level, completes parameter initialization, pushes warnings to mobile devices, and supports emergency operations and log storage. After regulation, the measured water temperature data downstream is fed back to the digital twin model, which automatically corrects parameters such as water flow mixing and heating efficiency, forming a closed loop of monitoring-prediction-execution-optimization.

[0012] To achieve the above objectives, this application also provides the following technical solutions: A device for identifying and responding to abnormal discharge water temperature, applied to a method for identifying and responding to abnormal discharge water temperature, the device comprising: The data acquisition unit is used to collect vertical water temperature data in front of the dam, synchronously integrate water quality sensor and meteorological station data, and construct a three-dimensional dynamic water temperature digital twin model based on the numerical model to realize real-time mapping and simulation of the flow-heat-solid coupling effect in the reservoir area. The threshold setting unit is used to set a basic threshold based on downstream ecological targets, and generate a dynamic threshold range by combining a seasonal probability distribution model, meteorological disturbance factors and water quality correlation rules, and divide it into multiple warning levels according to water temperature deviation. The control unit is used to automatically adjust the layer of the stratified water intake equipment, start the tailwater mixing device or solar heating system, and link the ecological water replenishment strategy when a preset warning level is triggered. The prediction unit uses a digital twin model to predict future water temperature changes, generates multiple control plans, stores key data through blockchain, and combines a remote operation and maintenance platform to achieve real-time monitoring and parameter adjustment of the equipment.

[0013] To achieve the above objectives, this application also provides the following technical solutions: An electronic device includes a processor and a memory coupled to the processor, the memory storing program instructions executable by the processor; when the processor executes the program instructions stored in the memory, it implements the above-described method for identifying and handling abnormal discharge water temperature.

[0014] To achieve the above objectives, this application also provides the following technical solutions: A storage medium storing program instructions, which, when executed by a processor, enable the aforementioned method for identifying and handling abnormal discharge water temperatures.

[0015] Compared with existing technologies, this invention uses distributed fiber optic temperature sensors for vertical water temperature data acquisition, enabling comprehensive and accurate acquisition of water temperature information at different depths in the reservoir, overcoming the limitations of traditional single-point monitoring. The high precision and high resolution of the sensors allow even minute water temperature changes to be detected promptly, providing a reliable data foundation for subsequent anomaly identification and control. By combining data from water quality sensors and weather stations, multi-source data fusion is achieved, comprehensively considering the influence of various factors on water temperature and further improving the overall understanding of water temperature conditions.

[0016] Beneficial Effects: This invention also combines seasonal probability distribution models, meteorological disturbance factors, and water quality correlation rules, allowing the threshold to be dynamically adjusted according to actual conditions. This adaptive threshold setting method can more accurately reflect water temperature requirements under different periods and conditions, greatly improving the accuracy of anomaly identification. For example, during special periods such as fish spawning season, the threshold can be automatically adjusted according to ecological goals, promptly detecting abnormal water temperature conditions and providing strong protection for the survival and reproduction of aquatic organisms.

[0017] When an alert level is triggered, the digital twin model can automatically calculate and adjust the stratification of the water intake equipment, activate the tailwater mixing device or solar heating system, and simultaneously coordinate with the ecological water replenishment strategy. The entire process requires no human intervention, enabling a rapid response and significantly improving emergency response efficiency. Compared to traditional human experience-based decision-making, the method of this invention is more scientific and accurate, effectively avoiding decision-making errors caused by human factors, promptly resolving abnormal water temperature issues, and reducing the impact on the downstream ecological environment and related industries.

[0018] Furthermore, the closed-loop verification and decision support system of this invention is of great significance. The pre-simulation function of the digital twin model can predict water temperature change trends in advance, generate multiple control plans, and provide a scientific basis for decision-making. Blockchain notarization ensures the authenticity, integrity, and traceability of key data, which helps regulatory authorities supervise and manage reservoir operations. The establishment of a remote operation and maintenance platform allows staff to monitor equipment and adjust parameters anytime, anywhere, improving the convenience and efficiency of management. This invention, through innovative technical means and methods, effectively solves the problems existing in the monitoring and control of reservoir discharge water temperature, and has significant economic, ecological, and social benefits, providing strong support for the development of water conservancy projects and ecological environmental protection. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the steps of one embodiment of the method for identifying and handling abnormal discharge water temperature according to this application. Figure 2 This is a flowchart illustrating the data processing steps of the method for identifying and responding to abnormal discharge water temperature in this application. Figure 3 This is a schematic diagram of the functional modules of one embodiment of the method for identifying and handling abnormal discharge water temperature according to this application; Figure 4 This is a schematic diagram of the structure of an embodiment of the electronic device of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0021] As described in the background section, some industries that rely on suitable water temperatures, such as aquaculture and hot spring tourism, are also affected by abnormal downstream water temperatures. Existing technologies suffer from technical problems such as incomplete monitoring, fixed thresholds, and a lack of intelligent emergency response.

[0022] like Figure 1 As shown, to solve the above-mentioned technical problems, the present invention provides a method for identifying and handling abnormal discharge water temperature, including: S1. Collect vertical water temperature data in front of the dam, integrate water quality sensor and meteorological station data simultaneously, and construct a three-dimensional dynamic water temperature digital twin model based on the numerical model to realize real-time mapping and simulation of the flow-heat-solid coupling effect in the reservoir area. S2. Set basic thresholds based on downstream ecological targets, and generate dynamic threshold ranges by combining seasonal probability distribution models, meteorological disturbance factors and water quality correlation rules, and divide them into multi-level early warnings according to water temperature deviation. S3. When the preset warning level is triggered, the digital twin model automatically adjusts the layer of the stratified water intake equipment, starts the tailwater mixing device or solar heating system, and links the ecological water replenishment strategy. S4. Utilize digital twin models to predict future water temperature changes, generate multiple control plans, store key data through blockchain, and combine with a remote operation and maintenance platform to achieve real-time monitoring and parameter adjustment of equipment.

[0023] In practice, multi-source data collection is conducted in the waters in front of the dam to gain a comprehensive understanding of the water's condition.

[0024] In a further embodiment, such as Figure 2 As shown, it also includes: The S11 high-frequency collector acquires vertical water temperature data, water quality parameters, and meteorological data. The data is preliminarily processed through edge computing nodes to extract key feature parameters. The water quality parameters acquired include at least conductivity, dissolved oxygen, and pH, and are uploaded to the data platform. Meteorological data includes air temperature, wind speed, sunshine duration, and precipitation, ensuring that meteorological data is acquired synchronously with water temperature data. S12 performs spatiotemporal alignment on the collected water temperature, water quality, and meteorological data from multiple sources, unifying timestamps and spatial coordinates to construct a unified data format; it cleans the data to remove noise and outliers; and it extracts derived features from the raw data to construct feature vectors, providing input for subsequent numerical models. S13 uses the EFDC model to construct the reservoir hydrodynamic module to simulate water flow velocity and flow field distribution; it couples the MIKEThermal module to construct the heat transfer model to simulate the heat conduction and convection process of water temperature, and uses an unstructured grid to discretize the reservoir area. S14 constructs a three-dimensional terrain model of the reservoir area based on a visualization platform, overlays real-time simulated water temperature field and flow field data, realizes a three-dimensional visualization display of water temperature distribution, and supports interactive query; S15 uses historical data to calibrate model parameters, ensuring that the simulation results match the measured data to a preset degree of consistency. By comparing the measured water temperature values ​​at downstream monitoring sections with the model output in real time, the model parameters are dynamically corrected.

[0025] The high-frequency data acquisition and edge computing preprocessing includes sensor deployment parameters: Parameters are acquired using a distributed fiber optic sensor: vertical spacing Δz = 1m, Δz ranges from 0.5 to 2m, covering water depths of 0-50m, temperature measurement accuracy ±0.1℃, sampling frequency 1Hz, and real-time acquisition of water temperature Tz (℃) at depth z. In this embodiment, a water quality sensor with a YSIEXO2 buoy is selected, which monitors conductivity (EC, µS / cm, accuracy ±0.5%FS), dissolved oxygen (DO, mg / L, accuracy ±1%FS), and pH (accuracy ±0.01), acquiring data every 5 minutes.

[0026] Edge computing feature: Vertical water temperature gradient: G= Used to identify the thermocline (|G|>0.2℃ / m), Δz=1m is the sensor spacing.

[0027] Standard deviation of surface water temperature: σ= , Ti The water temperature is measured by five sensors located 0-5m below the water surface. The average surface water temperature (°C) is used to assess the uniformity of water mixing.

[0028] Multi-source data spatiotemporal alignment and feature engineering were performed. Data processing parameters included spatiotemporal alignment and data cleaning: timestamps were accurate to the second, and spatial coordinates were based on the dam's upstream origin (accuracy ±0.5m). An input vector containing 12 features was constructed (water temperature, EC, DO, pH, air temperature, wind speed, sunshine duration, precipitation, water temperature gradient, surface standard deviation, thermocline depth, and daily water temperature fluctuation). Data cleaning employed the 3σ criterion to remove outliers, and a sliding window (15 minutes) was used to extract derived features, achieving a data validity rate ≥98.7%.

[0029] EFDC model parameters: - Unstructured triangular mesh: 5m×5m resolution for the main channel, 10m×10m for the bay, simulating water flow velocity u (m / s) and flow field distribution. MIKE Thermal coupling: Reservoir bottom heat flux:

[0030] The thermal conductivity of the clay at the bottom of the reservoir, z b The elevation of the reservoir bottom (m) The temperature gradient at the bottom of the reservoir (°C / m). - Vertical stratification: σ-coordinate stratification (10-20 layers), with a grid thickness of ≤1m in the thermocline region.

[0031] Model calibration and dynamic correction calibration parameters: Root Mean Square Error (RMSE):

[0032] m=150 monitoring points, R after calibration ² ≥0.95, vertical water temperature simulation error ≤0.3°C. Real-time correction includes dynamic adjustment of model parameters within 10 minutes when the deviation between downstream measured and simulated values ​​is >0.5°C.

[0033] In a further embodiment, dynamic threshold setting and anomaly classification are implemented, wherein ecologically goal-oriented basic thresholds are set based on the protected objects, and basic thresholds (°C) for different scenarios are clearly defined. The basic threshold table is shown below:

[0034] The seasonal division method uses the K-means algorithm to cluster the annual water temperature data, dividing it into four seasons. The characteristics of each season are as follows:

[0035] The threshold offset rule is estimated using kernel density: (h=0.5, K, ^{-1}x), calculate the 10th percentile (winter low temperature threshold) and 90th percentile (summer high temperature threshold) of seasonal water temperature. Standard deviation correction: Spring threshold upper limit = basic upper limit + 0.5σ_s (σ_s is the seasonal standard deviation, such as upper limit + 1°C if spring σ_s = 2°C); Winter threshold lower limit = basic lower limit - 1.0σ_s (lower limit - 1.5°C if winter σ_s = 1.5°C). Meteorological-water temperature response model - multiple linear regression modeling: Select annual meteorological data (n data sets) to establish the regression relationship between water temperature change ΔT and sunshine S and wind speed V: The t-test showed that the significance levels of both S and V were <0.01, thus the corrected formula (Formula 6) was determined: The dynamic threshold interval calculation, combined with decision tree weights (60% sunshine, 40% wind speed), results in the following final threshold: ; ; For example, when S=2h and V=8m / s, ΔT_{weather}=0.94°C, and the threshold range expands from 8-12°C to 7.4-12.4°C. S25. Multi-level early warning logic - water temperature deviation calculation (Formula 8): This formula allows for the accurate real-time calculation of the deviation between the measured water temperature and the dynamic threshold.

[0036] The warning levels are classified as follows: Level I warning: When the deviation is ≤1℃, the encrypted monitoring program is activated, the monitoring frequency is increased to 1 time / minute, and the relevant equipment is self-checked to ensure that the system is in normal operating condition. The whole process is automatically triggered and the response time does not exceed 10 seconds.

[0037] Level II warning: When the deviation is between 1-2℃, the system will automatically pre-adjust the water intake layer and start the mixing device for preheating to prepare for possible water temperature adjustment. The equipment linkage response time is controlled within 3 minutes.

[0038] Level III Warning: If the deviation exceeds 2°C, the entire system emergency response will be activated immediately, and key strategies such as heating or water replenishment will be initiated rapidly. The time from the triggering of the warning to the completion of the strategy execution shall not exceed 10 minutes.

[0039] Threshold update mechanism: Regular update: At 0:00 every day, the system automatically imports the latest 30 days of data, recalculates the seasonal quantiles, and completes the entire threshold update cycle in no more than 30 minutes to adapt to long-term changes in water temperature.

[0040] Real-time correction: When receiving weather warnings (such as orange alerts for cold waves) or sudden data such as water quality exceeding standards, the system can quickly recalculate the threshold within 2 minutes, with the correction delay controlled within 30 seconds, to ensure that the threshold can respond to changes in the external environment in a timely manner.

[0041] The digital twin model automatically adjusts the stratification of the water intake equipment, including: The digital twin model uses its own simulation and analysis capabilities, combined with real-time collected water temperature data, to accurately identify the location of the thermocline in the reservoir and the water temperature distribution of different water layers. The target water temperature to be obtained is determined based on downstream ecological objectives, water demand, and early warning levels. Based on the target water temperature, thermocline, and water temperature distribution, the digital twin model calculates the optimal layer to which the stratified water intake equipment should be adjusted. By communicating with the equipment's control system, it automatically issues commands to control the equipment to move to the corresponding layer in order to obtain a water source close to the target water temperature. The stratified water intake equipment may include at least a valve valve and / or a stratified water intake port.

[0042] The digital twin model determines whether to activate the tailwater mixing device or the solar heating system based on the warning level and the current water temperature. If the water temperature deviates only slightly from the target value, the tailwater mixing device may be activated first. If the water temperature is severely low and the sunlight conditions permit, the solar heating system may be activated. When the tailwater mixing device needs to be started, the digital twin model sends a start command to the device's control system; the device usually mixes water of different temperatures thoroughly by disturbing the water flow to achieve a uniform water temperature. If the solar heating system is activated, the digital twin model calculates the required heating power based on real-time water temperature, light intensity, and other data, and sends corresponding control commands to the system; the solar heating system uses solar energy to heat the water and raise the water temperature to the target range; Among them, the emergency control strategy automatically triggers stratified water intake intelligent control, including thermocline positioning: Canny algorithm for boundary detection: Positioning accuracy ±0.8m provides stratigraphic basis for the tongue-shaped valve / stratified intake. Target water temperature calculation: determined by combining ecological targets and early warning levels. For example, during a Level II warning .

[0043] Equipment control: BP neural network calculates the opening degree of the tongue valve. Input the depth of the thermocline Surface water temperature Output accuracy ±1℃, single adjustment time ≤90 seconds.

[0044] Multimodal water temperature regulation tailwater mixing: Right-angled baffles (1.8m high) and flow stabilizers (0.8m spacing), water temperature standard deviation ≤0.2℃ within 30 minutes. Solar heating: Power calculation: , The amount of water discharged. Energy consumption is reduced by more than 30%.

[0045] The coordinated ecological water replenishment strategy includes: based on the warning level, downstream ecological water demand, and the current water level and flow of the reservoir, the digital twin model calculates the amount of water that needs to be replenished and the appropriate time for water replenishment; The digital twin model sends control commands to the water replenishment equipment to adjust the equipment's opening degree or operating power, thereby achieving ecological water replenishment. During the water replenishment process, the digital twin model continuously monitors parameters such as water temperature, water level, and flow rate in the downstream water area to evaluate the water replenishment effect; if it is found that the water replenishment effect does not meet expectations, the water replenishment strategy is adjusted in a timely manner. The data and effects throughout the regulation process are fed back to the digital twin model, which then learns and optimizes itself based on this feedback.

[0046] The ecological water replenishment strategy uses a one-dimensional water temperature diffusion equation: During the spawning period, the water level rise is maintained at ≤0.5m / hour by a PID controller, and the water temperature fluctuation is maintained at ≤0.5℃ / hour.

[0047] By using blockchain to store key data and combining it with a remote operation and maintenance platform, real-time monitoring and parameter adjustment of equipment can be achieved, including: Based on the digital twin model, data such as vertical water temperature, meteorology, and equipment status are accessed in real time. The adjustable parameter combinations of the stratified water intake equipment opening degree and heating power are enumerated to drive the EFDC / MIKE model to couple and simulate the water temperature evolution within a preset time period in the future. The LSTM model is superimposed to correct long-term prediction errors and generate water temperature change curves and thermocline migration trajectories at key downstream sections. Based on ecological indicators, at least three optimal plans are selected using the particle swarm optimization algorithm. The optimal plans include equipment control parameters such as the opening degree of the associated tongue valve and the height of the curtain wall. When a dynamic threshold triggers an alert, the system automatically extracts the alert level, control parameters, and verification data, encrypts and uploads them to the blockchain via smart contracts, completes the entire process of evidence storage, and supports blockchain traceability. The remote operation and maintenance platform integrates a 3D visualization interface to display the water temperature field, equipment status and pre-drill effect in real time. The manual mode supports parameter fine-tuning and one-click recall of preset plans. The automatic mode links equipment according to the warning level, completes parameter initialization, pushes warnings to mobile devices, and supports emergency operations and log storage. After regulation, the measured water temperature data downstream is fed back to the digital twin model, which automatically corrects parameters such as water flow mixing and heating efficiency, forming a closed loop of monitoring-prediction-execution-optimization.

[0048] Digital twin prediction and blockchain evidence storage are specifically achieved through a multi-objective prediction algorithm, including setting an objective function: ≤5% (power generation volatility) ≥0.5℃ / hour (water temperature recovery rate) ≥95% (ecological base flow). The particle swarm optimization algorithm enumerates 10+ parameter combinations to generate at least 3 optimal plans (such as tongue flap gate 60° + heating power 40kW).

[0049] Blockchain-based evidence storage system stores data including: early warning levels, Equipment parameters ( , ), verify data ( , The smart contract is encrypted and stored on the blockchain, with a storage delay of ≤2 minutes, and supports Merkle tree tracing.

[0050] Remote operation and maintenance platform enables 3D visualization: HSV color mapping of water temperature ( In manual mode, the parameter fine-tuning accuracy is ±1° (for tongue flap doors) and ±5cm (for curtain walls). In automatic mode, the pre-plan initialization is completed within 5 minutes, and the mobile terminal response time is ≤30 seconds.

[0051] Application in a large reservoir project: Data acquisition and modeling: Three vertical lines were deployed, each with 50 fiber optic sensors (1m apart). An edge computing node (Advantech UNO-3082G) was used to filter out abnormal data in real time and extract 12 feature parameters, with a data effectiveness rate of 98.7%.

[0052] The EFDC model has a mesh of 12,000 triangular elements, 15 vertical MIKE Thermal layers, a heat flux calculation frequency of 1 time / minute at the bottom of the reservoir, and a simulated water temperature RMSE of 0.25°C.

[0053] Dynamic threshold application: Winter threshold correction: KDE fitting of 2021-2023 data, 10th percentile 8°C, combined with... The lower limit of the basic threshold is changed from 8°C to 7°C.

[0054] Example of weather correction: , , The threshold range is 8-12°C → 7.1-12.9°C to avoid false alarms.

[0055] Emergency control measures: Level III warning: Gate opening 60° (8m below the thermocline), heating power 50kW, water temperature from 8°C to 12°C in 3 hours, ecological water replenishment flow increased by 15% (30m³ / s), downstream water temperature fluctuation ≤1°C in 24 hours.

[0056] Pre-drills and Evidence Storage: The multi-objective pre-drills generate 5 sets of contingency plans. The "Cold Wave Emergency" plan reduces energy consumption by 25% and power generation fluctuation by 4.8%. Control data is uploaded to the blockchain in real time, with an evidence storage delay of 1.2 seconds. The blockchain explorer can query the event hash value (such as 0x7a8b9c...).

[0057] like Figure 3 As shown, an apparatus for identifying and handling abnormal discharge water temperature according to a specific embodiment of the present invention is described. The apparatus includes: The data acquisition unit 201 is used to collect vertical water temperature data in front of the dam, and simultaneously integrates water quality sensor and meteorological station data. Based on the numerical model, a three-dimensional dynamic water temperature digital twin model is constructed to realize the real-time mapping and simulation of the flow-heat-solid coupling effect in the reservoir area. Threshold setting unit 202 is used to set a basic threshold based on downstream ecological targets, and generate a dynamic threshold range by combining a seasonal probability distribution model, meteorological disturbance factors and water quality correlation rules, and divide it into multiple warning levels according to water temperature deviation. The control unit 203 is used to automatically adjust the layer of the stratified water intake equipment, start the tailwater mixing device or solar heating system, and link the ecological water replenishment strategy when the preset warning level is triggered. Prediction unit 204 uses a digital twin model to predict future water temperature changes, generates multiple control plans, stores key data through blockchain, and combines a remote operation and maintenance platform to achieve real-time monitoring and parameter adjustment of equipment.

[0058] Figure 4 The hardware structure diagram of a computing device 30 for identifying and handling abnormal discharge water temperature is shown according to an embodiment of this specification. The computing device 30 may include at least one processor 301, a memory 302 (e.g., non-volatile memory), a main memory 303, and a communication interface 304, and the at least one processor 301, memory 302, main memory 303, and communication interface 304 are connected together via a bus 305. At least one processor 301 executes at least one computer-readable instruction stored or encoded in the memory 302.

[0059] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, 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 coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0060] 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 units described above can be implemented in hardware or as software functional units. The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0061] The specific embodiments of the invention have been described in detail above, but they are only examples, and this application is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications or substitutions to the invention are also within the scope of this application. Therefore, all equivalent changes, modifications, and improvements made without departing from the spirit and principles of this application should be covered within the scope of this application.

Claims

1. A method for identifying and handling abnormal discharge water temperature, characterized in that, The term includes: Collect vertical water temperature data in front of the dam, integrate water quality sensor and meteorological station data simultaneously, and construct a three-dimensional dynamic water temperature digital twin model based on numerical model to realize real-time mapping and simulation of the flow-heat-solid coupling effect in the reservoir area; Based on the downstream ecological objectives, a basic threshold is set, and a dynamic threshold range is generated by combining the seasonal probability distribution model, meteorological disturbance factors and water quality correlation rules. The range is divided into multiple warning levels according to the water temperature deviation. When the preset warning level is triggered, the digital twin model automatically adjusts the layer of the stratified water intake equipment, starts the tailwater mixing device or solar heating system, and links the ecological water replenishment strategy. By using digital twin models to predict future water temperature changes, multiple control plans are generated. Key data is stored on blockchain and combined with a remote operation and maintenance platform to achieve real-time monitoring and parameter adjustment of equipment.

2. The method for identifying and handling abnormal discharge water temperature according to claim 1, characterized in that, The method further includes: High-frequency acquisition of vertical water temperature data, water quality parameters, and meteorological data; preliminary data processing through edge computing nodes to extract key feature parameters; acquisition of water quality parameters including at least conductivity, dissolved oxygen, and pH, and uploading to the data platform; meteorological data including air temperature, wind speed, sunshine duration, and precipitation, ensuring synchronous acquisition of meteorological data and water temperature data; The collected water temperature, water quality, and meteorological data from multiple sources were spatiotemporally aligned to unify timestamps and spatial coordinates, and a unified data format was constructed. The data was cleaned to remove noise and outliers. At the same time, derived features were extracted from the raw data to construct feature vectors, providing input for subsequent numerical models. The EFDC model was selected to construct the reservoir hydrodynamic module to simulate the water flow velocity and flow field distribution; the MIKEThermal module was coupled to construct the heat transfer model to simulate the heat conduction and convection process of water temperature; and the reservoir area was discretized using an unstructured grid. A 3D terrain model of the reservoir area is constructed based on a visualization platform, and real-time simulated water temperature field and flow field data are overlaid to realize a 3D visualization display of water temperature distribution and support interactive query. Historical data is used to calibrate model parameters so that the simulation results and measured data reach a preset degree of consistency. The model parameters are dynamically corrected by comparing the measured water temperature values ​​of the downstream monitoring section with the model output in real time.

3. The method for identifying and handling abnormal discharge water temperature according to claim 1, characterized in that, The method further includes: Analyze the water temperature requirements of downstream ecologically sensitive objects and clarify the basic threshold ranges for different scenarios; Input historical water temperature data, seasonal attributes, typical meteorological parameters and water quality indicators, and extract seasonal features, meteorological features and water quality features to construct an ecological-water temperature correlation feature library; By using kernel density estimation or Gaussian mixture model, the historical water temperature data are fitted with a probability distribution according to the season to determine the mean and fluctuation range of water temperature in different seasons. Generate seasonal correction factors, form seasonally specific threshold offset rules, establish a meteorological-water temperature response model, and quantify the impact of sunshine duration and wind speed on water temperature. By using regression analysis or decision tree algorithms, the corrective weights of meteorological factors on the threshold are determined, and the threshold range is dynamically expanded or contracted. Define water quality-temperature coupling rules, use a rule engine to analyze water quality parameters in real time, and dynamically generate water quality correlation correction coefficients.

4. The method for identifying and handling abnormal discharge water temperature according to claim 3, characterized in that, The multi-level early warning system includes: calculating the deviation between the measured water temperature and the dynamic threshold in real time, and classifying it into three levels of early warning based on the absolute value of the deviation; the method also includes setting a dynamic threshold update mechanism. The seasonal probability distribution model is updated regularly based on the latest data within the time period to adapt to interannual water temperature fluctuations; real-time reception of meteorological and water quality emergencies triggers immediate correction of threshold intervals, and the immediate correction completes the recalculation of deviation values ​​within a preset time period.

5. The method for identifying and handling abnormal discharge water temperature according to claim 1, characterized in that, The method further includes: The digital twin model continuously receives data from various sensors. According to different warning levels, corresponding threshold values ​​for parameters such as water temperature, water quality, and water level are set. The digital twin model compares the collected data with the preset warning thresholds in real time. Once the data exceeds the threshold, the corresponding warning level is immediately triggered. The digital twin model automatically adjusts the stratification of the water intake equipment, including: The digital twin model uses its own simulation and analysis capabilities, combined with real-time collected water temperature data, to accurately identify the location of the thermocline in the reservoir and the water temperature distribution of different water layers. The target water temperature to be obtained is determined based on downstream ecological objectives, water demand, and early warning levels. Based on the target water temperature, the thermocline, and the water temperature distribution, the digital twin model calculates the optimal layer to which the stratified water intake device should be adjusted. By communicating with the device's control system, it automatically issues commands to control the device to move to the corresponding layer in order to obtain a water source close to the target water temperature. The stratified water intake device may include at least a valve valve and / or a stratified water intake port.

6. The method for identifying and handling abnormal discharge water temperature according to claim 5, characterized in that, The method further includes: The digital twin model determines whether to activate the tailwater mixing device or the solar heating system based on the warning level and the current water temperature. If the water temperature deviates only slightly from the target value, the tailwater mixing device may be activated first. If the water temperature is severely low and the sunlight conditions permit, the solar heating system may be activated. When the tailwater mixing device needs to be started, the digital twin model sends a start command to the device's control system; the device usually mixes water of different temperatures thoroughly by disturbing the water flow to achieve a uniform water temperature. If the solar heating system is activated, the digital twin model calculates the required heating power based on real-time water temperature, light intensity, and other data, and sends corresponding control commands to the system; the solar heating system uses solar energy to heat the water and raise the water temperature to the target range; The ecological water replenishment strategy linkage includes: based on the warning level, downstream ecological water demand, and the current water level and flow of the reservoir, the digital twin model calculates the amount of water to be replenished and the appropriate time for water replenishment; The digital twin model sends control commands to the water replenishment equipment to adjust the equipment's opening degree or operating power, thereby achieving ecological water replenishment. During the water replenishment process, the digital twin model continuously monitors parameters such as water temperature, water level, and flow rate in the downstream water area to evaluate the water replenishment effect; if it is found that the water replenishment effect does not meet expectations, the water replenishment strategy is adjusted in a timely manner. The data and effects throughout the regulation process are fed back to the digital twin model, which then learns and optimizes itself based on this feedback.

7. The method for identifying and handling abnormal discharge water temperature according to claim 1, characterized in that, The method of using blockchain to store key data and combining it with a remote operation and maintenance platform to achieve real-time monitoring and parameter adjustment of equipment includes: Based on the digital twin model, data such as vertical water temperature, meteorology, and equipment status are accessed in real time. The adjustable parameter combinations of the stratified water intake equipment opening degree and heating power are enumerated. The EFDC / MIKE model is driven to couple and simulate the water temperature evolution within a preset time period in the future. The LSTM model is superimposed to correct long-term prediction errors and generate water temperature change curves and thermocline migration trajectories at key downstream sections. Based on ecological indicators, at least three optimal plans are selected using the particle swarm optimization algorithm. The optimal plans include equipment control parameters such as the opening degree of the associated tongue valve and the height of the curtain wall. When a dynamic threshold triggers an alert, the system automatically extracts the alert level, control parameters, and verification data, encrypts and uploads them to the blockchain via smart contracts, completes the entire process of evidence storage, and supports blockchain traceability. The remote operation and maintenance platform integrates a 3D visualization interface to display the water temperature field, equipment status and pre-drill effect in real time. The manual mode supports parameter fine-tuning and one-click recall of preset plans. The automatic mode links equipment according to the warning level, completes parameter initialization, pushes warnings to mobile devices, and supports emergency operations and log storage. After regulation, the measured water temperature data downstream is fed back to the digital twin model, which automatically corrects parameters such as water flow mixing and heating efficiency, forming a closed loop of monitoring-prediction-execution-optimization.

8. A device for identifying and handling abnormal discharge water temperature, applied to the method for identifying and handling abnormal discharge water temperature as described in any one of claims 1 to 7, characterized in that, The device for identifying and handling abnormal discharge water temperature includes: The data acquisition unit is used to collect vertical water temperature data in front of the dam, synchronously integrate water quality sensor and meteorological station data, and construct a three-dimensional dynamic water temperature digital twin model based on the numerical model to realize real-time mapping and simulation of the flow-heat-solid coupling effect in the reservoir area. The threshold setting unit is used to set a basic threshold based on downstream ecological targets, and generate a dynamic threshold range by combining a seasonal probability distribution model, meteorological disturbance factors and water quality correlation rules, and divide it into multiple warning levels according to water temperature deviation. The control unit is used to automatically adjust the layer of the stratified water intake equipment, start the tailwater mixing device or solar heating system, and link the ecological water replenishment strategy when a preset warning level is triggered. The prediction unit uses a digital twin model to predict future water temperature changes, generates multiple control plans, stores key data through blockchain, and combines a remote operation and maintenance platform to achieve real-time monitoring and parameter adjustment of the equipment.

9. An electronic device, characterized in that, The method includes a processor and a memory coupled to the processor, the memory storing program instructions executable by the processor; when the processor executes the program instructions stored in the memory, it implements the method for identifying and handling abnormal discharge water temperature as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores program instructions, which, when executed by a processor, enable the method for identifying and handling abnormal discharge water temperature as described in any one of claims 1 to 7.

Images