Digital twin basin reservoir group scheduling method and system

Abstract

The invention relates to the technical field of water conservancy projects. By providing the digital twinborn basin reservoir group scheduling method and system, the method comprises the following steps: constructing a real-time mirror image digital twinborn platform of a physical basin by fusing an Internet of Things sensing network, a hydrological and hydrodynamic model and a historical operation database; establishing a multi-dimensional value evaluation index system including flood control benefit, power generation benefit, water supply guarantee and ecological maintenance; according to the multi-dimensional value evaluation index system, collecting cost input and benefit output data of each scheduling subject through a preset digital platform to obtain a multi-subject value data set; calculating the benefit distribution proportion of each main body under different scheduling schemes through a value distribution algorithm; generating a candidate scheduling scheme set through a benefit distribution proportion, flood control safety constraints and ecological flow constraints; and performing simulation evaluation on the candidate scheduling scheme set, and outputting a final scheduling scheme so as to improve the execution fluency of the multi-agent cooperative scheduling scheme.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water conservancy engineering, in particular to a digital twin river basin reservoir group dispatching method and system. BACKGROUND

[0002] In the technical field of water conservancy engineering, joint dispatching of a river basin reservoir group is a core technical means for realizing multi-objective optimal utilization of water resources, and needs to take into account multiple objectives such as flood control, clean energy supply, water supply safety guarantee, and ecological health maintenance.

[0003] However, in traditional reservoir group dispatching technology, multiple subject objectives such as flood control, power generation, water supply, and ecological maintenance belong to different management subjects, are naturally in conflict due to interest demands, and lack a multi-subject quantitative coordination mechanism, often resulting in great resistance to implementation of a dispatching scheme and poor landing; existing technology often stays at a qualitative evaluation level for "benefit-cost" of a dispatching scheme, does not build a multi-dimensional quantitative model covering dimensions such as flood control and disaster reduction quantity and power generation benefit value, and is difficult to scientifically support multi-objective decision-making; in addition, interest distribution has long relied on "experience consultation" or "fixed proportion allocation", and has not quantified subject resource input and benefit output fairly through mathematical tools such as cooperative game theory, which easily intensifies multi-subject conflicts. SUMMARY

[0004] Therefore, it is necessary to provide a digital twin river basin reservoir group dispatching method and system to achieve the technical effects of improving the execution fluency of a multi-subject coordinated dispatching scheme, reducing cross-department collaboration resistance; improving the scientificity and accuracy of multi-objective dispatching decision-making, reducing the fuzziness of decision-making caused by qualitative evaluation; and improving the fairness and quantifiable level of an interest distribution mechanism, reducing the risk of intensifying conflicts due to unfair distribution.

[0005] In a first aspect, the present application provides a digital twin river basin reservoir group dispatching method, which comprises:

[0006] A real-time mirror digital twin platform of a physical river basin is constructed by fusing an Internet of Things sensing network, a hydrological and hydrodynamic model, and a historical operation database;

[0007] Based on the digital twin platform, a multi-dimensional value evaluation index system including flood control benefit, power generation benefit, water supply guarantee, and ecological maintenance is established;

[0008] According to the multi-dimensional value evaluation index system, cost input and benefit output data of each dispatching subject are collected through a preset digital platform to obtain a multi-subject value data set;

[0009] Based on the multi-subject value data set, an interest distribution proportion of each subject under different dispatching schemes is calculated through a value distribution algorithm;

[0010] A set of candidate scheduling schemes is generated by using the proportion of benefits distribution and the preset flood control safety constraints and preset ecological flow constraints.

[0011] The candidate scheduling scheme set is simulated and evaluated, and the final scheduling scheme is output.

[0012] In one embodiment, based on a digital twin platform, a multi-dimensional value assessment index system is established, including flood control benefits, power generation revenue, water supply security, and ecological maintenance, comprising:

[0013] Regarding flood control benefits, flood control value indicators are generated through quantitative analysis of flood risk maps and loss avoidance rates.

[0014] For power generation revenue, calculations are performed based on peak-valley electricity price curves and power generation efficiency models to generate power generation revenue indicators;

[0015] For water supply security, the water supply reliability rate and the expected value of water shortage are used for evaluation to generate water supply security indicators.

[0016] For ecological maintenance, ecological health indicators are generated by quantifying the ecological flow guarantee rate and water quality improvement degree.

[0017] By combining flood control value indicators, power generation revenue indicators, water supply security indicators, and ecological health indicators through a combined weighting method, a multi-dimensional evaluation indicator system is generated.

[0018] In one embodiment, to assess flood control benefits, a flood control value index is generated through quantitative analysis of flood risk maps and loss avoidance rates, including:

[0019] Acquire historical flood data and geographic information data of the target watershed, and construct a flood risk map based on the historical flood data and geographic information data to reflect the relationship between the inundation range and the degree of loss under different flood frequencies;

[0020] The following formula was used to collect data on the difference between actual flood losses and expected losses during multiple rounds of reservoir scheduling. Statistical analysis of the difference data was then performed to obtain the loss avoidance rate:

[0021]

[0022] Among them, R avoid E represents the loss avoidance rate, n represents the total number of reservoir scheduling rounds, and E represents the loss avoidance rate. i Let A represent the expected loss value in round i. i This represents the actual loss value in round i;

[0023] The potential loss value corresponding to each risk level in the flood risk map is correlated and mapped with the loss avoidance rate to generate the flood control risk impact value corresponding to each scheduling scheme.

[0024] The impact values ​​of flood control risks are normalized to generate flood control value indicators.

[0025] In one embodiment, for water supply security, an assessment is conducted using water supply reliability and expected water shortage values ​​to generate water supply safety indicators, including:

[0026] Obtain the daily water demand sequence of each water user within the target water supply cycle and the actual water supply sequence of the reservoir during the corresponding time period;

[0027] Based on the daily water demand sequence and the actual water supply sequence, the ratio of the number of days with sufficient water supply to the total number of monitoring days is calculated to generate the water supply reliability rate.

[0028] Collect data on reservoir inflow and water demand from water users during historical periods of continuous drought, and construct a water shortage probability distribution model.

[0029] By using a water shortage probability distribution model, the expected value of water shortage under different drought scenarios is calculated, and the expected value of water shortage is generated.

[0030] The water supply reliability rate and the expected water shortage are weighted and integrated to generate a water supply security index.

[0031] In one embodiment, for ecological maintenance, ecological health indicators are generated by quantifying the ecological flow guarantee rate and water quality improvement degree, including:

[0032] Obtain the ecological flow demand sequence of key ecological zones in the target reservoir's control basin, and the actual outflow sequence within the corresponding scheduling cycle;

[0033] Based on the ecological flow demand sequence and the actual outflow sequence, the percentage of time the actual outflow reaches or exceeds the ecological flow demand is statistically analyzed to generate the ecological flow guarantee rate.

[0034] Collect water quality monitoring datasets of the target reservoir before the implementation of ecological scheduling and water quality monitoring datasets for the corresponding time period after the implementation;

[0035] Based on the water quality monitoring dataset before implementation and the water quality monitoring dataset for the corresponding period after implementation, the improvement rate of key water quality indicators is calculated, and the water quality improvement degree is generated.

[0036] The ecological flow guarantee rate and water quality improvement rate are weighted and integrated to generate ecological health indicators.

[0037] In one embodiment, a real-time mirror digital twin platform for the physical watershed is constructed by integrating an Internet of Things (IoT) sensing network, a hydrological and hydrodynamic model, and a historical operational database, including:

[0038] Real-time hydrological data and reservoir operation status data collected by the Internet of Things sensing network are fused and processed to obtain a dynamic monitoring dataset;

[0039] The dynamic monitoring dataset is input into the hydrological and hydrodynamic model to simulate the flood evolution path and water resource allocation process in the basin, and to generate a flood simulation result set.

[0040] Based on the flood simulation results set and engineering characteristic data in the historical operation database, a real-time mirror digital twin platform for the physical watershed is constructed.

[0041] In one embodiment, based on a multi-entity value dataset, a value allocation algorithm is used to calculate the benefit distribution ratio of each entity under different scheduling schemes, including:

[0042] A multi-agent negotiation model is constructed based on cooperative game theory, defining the contribution characteristic function of each agent in the scheduling alliance and generating a set of characteristic functions.

[0043] The set of characteristic functions is calculated iteratively using the Shapley value algorithm to generate the initial allocation ratio set;

[0044] A risk-contribution correction factor is introduced to correct the initial allocation ratio set, thereby generating the benefit allocation ratio.

[0045] Secondly, this application also provides a digital twin watershed reservoir group scheduling system, which includes:

[0046] The digital twin building module is used to construct a real-time mirror digital twin platform for a physical watershed by integrating IoT sensing networks, hydrological and hydrodynamic models, and historical operational databases.

[0047] The multi-dimensional assessment module is used to establish a multi-dimensional value assessment indicator system based on the digital twin platform, including flood control benefits, power generation revenue, water supply security and ecological maintenance.

[0048] The data collection module is used to collect cost input and benefit output data of each scheduling entity through a preset digital platform based on a multi-dimensional value assessment indicator system, so as to obtain a multi-entity value dataset.

[0049] The benefit allocation module is used to calculate the benefit allocation ratio of each entity under different scheduling schemes based on a multi-entity value dataset and a value allocation algorithm.

[0050] The scheme generation module is used to generate a set of candidate scheduling schemes based on the benefit distribution ratio and preset flood control safety constraints and preset ecological flow constraints.

[0051] The scheme evaluation module is used to simulate and evaluate the set of candidate scheduling schemes and output the final scheduling scheme.

[0052] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods in the first aspect of this application.

[0053] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods in the first aspect of this application.

[0054] The digital twin watershed reservoir group scheduling method and system provided in this application, by constructing a real-time mirror digital twin platform of the physical watershed, provides a unified and relatively accurate digital foundation for multi-entity collaborative scheduling, breaking down information barriers between different management entities. This enables all entities to conduct scheduling work based on a consistent understanding of the watershed status, thereby improving the smoothness of multi-entity collaborative scheduling schemes and reducing cross-departmental cooperation resistance. By establishing a multi-dimensional value assessment index system covering flood control benefits, power generation revenue, water supply security, and ecological maintenance, the scheduling assessment, which might otherwise remain at the qualitative level, is transformed into multi-dimensional quantitative analysis. This provides specific and comparable indicator basis for decision-making, thereby improving the scientificity and accuracy of multi-objective scheduling decisions and reducing the ambiguity of decisions caused by qualitative assessments. Based on the multi-dimensional value assessment index system, a multi-entity value dataset is collected, and then the benefit distribution ratio of each entity under different scheduling schemes is calculated through a value allocation algorithm. This transforms benefit distribution from relying on experience or fixed ratios to a fair and quantitative distribution based on data and algorithms, effectively improving the fairness and quantifiability of the benefit distribution mechanism and reducing the risk of conflicts arising from unfair distribution. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 A flowchart of a digital twin watershed reservoir group scheduling method in one embodiment of the present invention;

[0057] Figure 2 This is a flowchart of generating ecological health indicators based on ecological maintenance in one embodiment of the present invention, which quantifies ecological flow guarantee rate and water quality improvement degree.

[0058] Figure 3 This is a structural diagram of a digital twin watershed reservoir group scheduling system according to one embodiment of the present invention. Detailed Implementation

[0059] To make the above-mentioned objects, features, and advantages of this application more apparent and understandable, the specific embodiments of this application will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0060] First, the application scenarios of the embodiments of this application are described. In the embodiments of this application, a digital twin watershed reservoir group scheduling method and system are provided, applicable to but not limited to scenarios where there are multiple reservoirs within a watershed and these reservoirs need to be jointly scheduled to balance the needs of flood control, power generation, water supply, ecology, and other aspects.

[0061] As an illustration, the digital twin watershed reservoir group scheduling method and system provided in this application embodiment can also be applied to other application scenarios. This is only an example and does not limit the specific application scenarios.

[0062] like Figure 1 As shown, this application provides a digital twin watershed reservoir group scheduling method, which includes:

[0063] S101: By integrating IoT sensing networks, hydrological and hydrodynamic models, and historical operational databases, a real-time mirror digital twin platform for the physical watershed is constructed.

[0064] For example, real-time hydrological data and reservoir operation parameters within the watershed are collected through an IoT sensing network. This data is then cleaned, verified, and standardized to generate continuous and reliable dynamic monitoring data. The processed dynamic monitoring data is input into a pre-constructed hydrological and hydrodynamic model. The model simulates the formation and evolution of floods within the watershed, as well as the allocation of water resources among different reservoirs, generating corresponding flood simulation results and water resource allocation simulation results.

[0065] Subsequently, the simulation results were integrated with static data such as characteristic parameters and historical scheduling records of various water conservancy projects in the basin stored in the historical operation database to construct a digital twin platform that can map the physical basin status in real time.

[0066] S102: Based on the digital twin platform, establish a multi-dimensional value assessment indicator system that includes flood control benefits, power generation revenue, water supply security, and ecological maintenance.

[0067] For example, based on a digital twin platform, indicators are constructed for four dimensions: flood control benefits, power generation revenue, water supply security, and ecological maintenance. For flood control benefits, quantitative indicators reflecting the degree of risk avoidance are determined by combining flood evolution data simulated by the platform with historical disaster records.

[0068] Regarding power generation revenue, an evaluation index linking power generation efficiency and electricity pricing mechanisms is established based on reservoir operation parameters and electricity market information provided by the platform. For water supply security, indicators reflecting water supply stability and shortage risk are designed based on real-time supply and demand data monitored by the platform.

[0069] For ecological maintenance, evaluation indicators related to ecological flow compliance and water quality are formulated based on hydrological and ecological data fed back by the platform. Then, these four dimensions of indicators are integrated through a scientific weighting method to generate a multi-dimensional value assessment indicator system.

[0070] S103: Based on the multi-dimensional value assessment index system, the cost input and benefit output data of each scheduling entity are collected through a pre-set digital platform to obtain a multi-entity value dataset.

[0071] For example, using a multi-dimensional value assessment index system as a framework, the cost input scope and benefit output type of each dispatching entity in dimensions such as flood control, power generation, water supply, and ecological maintenance are clarified, generating data collection standards and boundaries. Subsequently, based on a pre-set digital platform, according to the above standards, data on the cost input of human resources, materials, and funds generated by each entity during the dispatching process, as well as the benefit output data under the corresponding dimensions, are collected in a targeted manner.

[0072] The collected data is categorized, organized, and verified to ensure that the data matches the dimensions of the indicator system and is accurate and effective. The processed cost input and benefit output data of each entity are then summarized and integrated to obtain a multi-entity value dataset.

[0073] S104: Based on a multi-entity value dataset, calculate the profit distribution ratio of each entity under different scheduling schemes using a value allocation algorithm.

[0074] For example, based on a multi-entity value dataset, key features such as the cost input scale and benefit contribution of each scheduling entity in different potential scheduling schemes are extracted to clarify the role and function of each entity in the scheduling alliance. Then, a suitable value allocation algorithm is selected, incorporating the above feature parameters into the algorithm framework. The algorithm quantifies the correlation between the resource input and benefit output of each entity, analyzing the interest relationships and allocation logic among entities under different schemes.

[0075] Based on the above algorithm calculation process, and combined with the actual constraints of the scheduling scenario, the preliminary allocation results are verified and adjusted for rationality. Then, through algorithm calculation and verification adjustments, the benefit distribution ratio of each entity under different scheduling schemes is obtained.

[0076] S105: Generate a set of candidate scheduling schemes by using the benefit distribution ratio and preset flood control safety constraints and preset ecological flow constraints.

[0077] For example, based on the proportion of benefit distribution, the rights and interests boundaries and acceptance thresholds of each scheduling entity in the plan are clarified. At the same time, the limiting conditions of reservoir flood control water level and flood discharge capacity in flood control safety constraints, as well as the river base flow bottom line and discharge requirements during sensitive periods involved in ecological flow constraints, are transformed into specific parameters.

[0078] Subsequently, considering the total water resources volume, spatiotemporal distribution characteristics, and scheduling characteristics of each reservoir in the basin, and within the constraints mentioned above, multiple potential scheduling parameter combinations were constructed based on the benefit distribution ratio. These combinations included core elements such as reservoir water storage and release plans and time-period power output arrangements. Each parameter combination underwent compliance verification, eliminating schemes that did not meet the constraints or had obvious conflicts. The verified parameter combinations were then compiled to generate a candidate scheduling scheme set.

[0079] S106: Simulate and evaluate the set of candidate scheduling schemes, and output the final scheduling scheme.

[0080] For example, the parameters of each scheme in the candidate scheduling scheme set are imported into the digital twin platform. Based on the platform's real-time mapping capability of watershed hydrological processes and reservoir operation status, the execution process of each scheme under different hydrological scenarios is simulated to obtain multi-dimensional simulation results, including flood control effect, power generation efficiency, water supply stability and ecological response.

[0081] Subsequently, based on preset evaluation criteria, the simulation results of each scheme were comprehensively evaluated, and the schemes' performance in terms of satisfying the balance of benefit distribution, compliance with constraints, and multi-objective synergy was analyzed. The scheme with the best overall effectiveness in the simulation evaluation was selected, and its feasibility was verified to ensure that the scheme can adapt to the actual watershed scheduling needs.

[0082] This application provides a digital twin-based watershed reservoir group scheduling method in one embodiment. By constructing a real-time mirror digital twin platform of the physical watershed, it provides a unified and relatively accurate digital foundation for multi-entity collaborative scheduling, breaking down information barriers between different management entities. This enables each entity to conduct scheduling work based on a consistent understanding of the watershed status, thereby improving the smoothness of multi-entity collaborative scheduling schemes and reducing cross-departmental cooperation resistance. By establishing a multi-dimensional value assessment index system covering flood control benefits, power generation revenue, water supply security, and ecological maintenance, it transforms scheduling assessments that might otherwise remain at a qualitative level into multi-dimensional quantitative analysis, providing specific and comparable indicator basis for decision-making. This improves the scientificity and accuracy of multi-objective scheduling decisions and reduces the ambiguity of decisions caused by qualitative assessments. Based on the multi-dimensional value assessment index system, it collects multi-entity value datasets and then calculates the benefit distribution ratio of each entity under different scheduling schemes through a value allocation algorithm. This transforms benefit distribution from relying on experience or fixed ratios to a fair and quantitative distribution based on data and algorithms, effectively improving the fairness and quantifiability of the benefit distribution mechanism and reducing the risk of conflicts arising from unfair distribution.

[0083] In one embodiment, based on a digital twin platform, a multi-dimensional value assessment index system is established, including flood control benefits, power generation revenue, water supply security, and ecological maintenance, comprising:

[0084] (1) Regarding flood control benefits, flood control value indicators are generated through quantitative analysis of flood risk maps and loss avoidance rates.

[0085] For example, based on a digital twin platform, focusing on the dimension of flood control benefits, historical flood data and geographic information are retrieved from the platform to construct a flood risk map that reflects the risk correlation under different flood scenarios. Simultaneously, information related to expected and actual losses during multiple rounds of dispatching is collected. Through statistical analysis of this information, the loss avoidance rate is determined. The potential risks presented in the flood risk map are correlated and mapped with the loss avoidance rate, and after quantitative processing, flood control value indicators are generated.

[0086] (2) For power generation revenue, the power generation revenue index is generated based on the peak-valley electricity price curve and the power generation efficiency model.

[0087] For example, regarding the power generation revenue dimension, reservoir operation status parameters and electricity market time-period price information are extracted from the platform. The electricity price benchmark for each time period is determined by combining the peak-valley electricity price curve. The power generation capacity for different time periods is calculated through the power generation efficiency model. The electricity price benchmark and the power generation capacity calculation results are combined to generate the power generation revenue index.

[0088] (3) For water supply security, the water supply reliability rate and the expected value of water shortage are used for evaluation to generate water supply security indicators.

[0089] For example, regarding water supply security, the platform uses real-time monitoring of user demand data and actual reservoir water supply records to calculate the percentage of time when water supply is met, thus determining the water supply reliability rate. Simultaneously, historical drought data on water inflow and outflow are retrieved to construct a water shortage probability analysis model. The model calculates the expected water shortage value, and a comprehensive evaluation of the water supply reliability rate and the expected water shortage value is conducted to generate a water supply security index.

[0090] (4) For ecological maintenance, ecological health indicators are generated by quantifying the ecological flow guarantee rate and water quality improvement.

[0091] For example, in the ecological maintenance dimension, the ecological flow demand standards for key ecological areas and the actual outflow data of reservoirs are obtained from the platform. The percentage of time the actual outflow meets the standards is statistically analyzed to determine the ecological flow guarantee rate. The water quality monitoring results before and after the implementation of ecological scheduling in the platform are compared, and the changes in key water quality parameters are analyzed to determine the degree of water quality improvement. The ecological flow guarantee rate and the degree of water quality improvement are weighted and integrated to generate ecological health indicators.

[0092] (5) By combining flood control value indicators, power generation revenue indicators, water supply security indicators and ecological health indicators through the combined weighting method, a multi-dimensional evaluation indicator system is generated.

[0093] For example, by adopting the combined weighting method, the importance of each indicator in the basin scheduling is comprehensively considered, and the flood control value indicator, power generation revenue indicator, water supply security indicator and ecological health indicator are weighted and systematically integrated to generate a multi-dimensional evaluation indicator system.

[0094] In one embodiment, to assess flood control benefits, a flood control value index is generated through quantitative analysis of flood risk maps and loss avoidance rates, including:

[0095] (1) Obtain historical flood data and geographic information data of the target watershed, and construct a flood risk map based on the historical flood data and geographic information data to reflect the relationship between the inundation range and the degree of loss under different flood frequencies.

[0096] For example, for a target watershed, detailed historical flood data, such as the inundation range, affected population, and economic losses, are collected from flood events that occurred at different times in history. At the same time, geographic information data, including elements such as elevation, land use type, and distribution of hydrological stations, are collected within the watershed. The two types of data are then standardized in format and their relevance is verified to ensure the integrity and matching of the data.

[0097] Based on verified historical flood data, this study analyzes the inundation boundaries and water depth distribution of floods under different flood return periods. Combining this with the vulnerability characteristics of various regions in geographic information data, a quantitative correlation model between inundation range and damage severity is established. This leads to the creation of a flood risk map that visually represents this correlation. Different flood return periods correspond to different flood frequencies.

[0098] (2) Using the following formula, collect the difference data between actual flood loss and expected loss during multiple rounds of reservoir scheduling, and perform statistical analysis on the difference data to obtain the loss avoidance rate:

[0099]

[0100] Among them, R avoid E represents the loss avoidance rate, n represents the total number of reservoir scheduling rounds, and E represents the loss avoidance rate. i Let A represent the expected loss value in round i. i This represents the actual loss value in round i.

[0101] For example, multiple representative reservoir scheduling instances are selected, and the expected loss range determined based on hydrological forecasts and watershed conditions before each scheduling is extracted, as well as the actual loss results obtained through on-site surveys and statistics after scheduling implementation. The difference between the expected loss and the actual loss for each round is calculated, and the differences from multiple rounds are summarized. The distribution pattern and average level of the above differences are analyzed through statistical methods to determine the loss avoidance rate, which comprehensively reflects the scheduling's effectiveness in avoiding losses.

[0102] (3) The potential loss value corresponding to each risk level in the flood risk map is correlated and mapped with the loss avoidance rate to generate the flood risk impact value corresponding to each scheduling scheme.

[0103] For example, the potential loss range corresponding to each risk level is extracted from the flood risk map to clarify the possible loss scale under different risk levels. The potential loss values ​​are then correlated with the determined loss avoidance rate. Through mapping calculation, the specific impact of different scheduling schemes on reducing flood risk is obtained, i.e., the flood risk impact value.

[0104] (4) Normalize the impact value of flood control risk to generate flood control value index.

[0105] For example, a unified numerical range standard is set to standardize the flood control risk impact values ​​of different scheduling schemes, so as to eliminate the interference caused by the difference in the magnitude of the original data, and make the processed values ​​able to intuitively reflect the relative advantages and disadvantages of the flood control benefits of each scheme, and generate flood control value indicators.

[0106] In one embodiment, for water supply security, an assessment is conducted using water supply reliability and expected water shortage values ​​to generate water supply safety indicators, including:

[0107] (1) Obtain the daily water demand sequence of each water user within the target water supply cycle and the actual water supply sequence of the reservoir during the corresponding time period.

[0108] For example, a target water supply cycle is determined, and the daily water demand of each water user within that cycle is collected to generate a continuous daily water demand sequence. Simultaneously, data on the actual water volume supplied by the reservoir to each water user during the corresponding time period is collected and compiled into an actual water supply sequence. The time nodes and data integrity of the two sequences are then verified and matched. The water users include those engaged in urban life, industrial production, and agricultural irrigation.

[0109] (2) Based on the daily water demand sequence and the actual water supply sequence, the ratio of the number of days the water supply is met to the total number of monitoring days is calculated to generate the water supply reliability rate.

[0110] For example, based on the verified daily water demand sequence and the actual water supply sequence, the actual water supply is compared day by day to see if it meets the daily water demand. The number of days that the water supply meets the demand is counted, and then the proportion of this number of days to the total number of monitoring days in the target water supply cycle is calculated to generate the water supply reliability rate, so as to reflect the stability of the water supply under normal operating conditions.

[0111] (3) Collect data on reservoir inflow and water demand from water users during historical periods of continuous drought, and construct a water shortage probability distribution model.

[0112] For example, by screening for consecutive drought periods in history, data such as the natural inflow of water into the reservoir and changes in the reservoir's water storage during the aforementioned periods, as well as the water demand data of each water user during the same period, and after outlier processing and standardization transformation of the data, a water shortage probability distribution model that can reflect the possibility and degree of water shortage under different drought intensities is constructed using probabilistic statistical methods.

[0113] (4) Calculate the expected value of water shortage under different drought scenarios using the water shortage probability distribution model, and generate the expected value of water shortage.

[0114] For example, using the constructed water shortage probability distribution model, the scale of the gap between reservoir water supply and demand under different drought scenarios is simulated. The average expected value of water shortage under each scenario is obtained through model calculation, that is, the expected value of water shortage is generated to reflect the water supply risk under extreme drought conditions. Among them, different drought scenarios include mild drought, moderate drought, and severe drought.

[0115] (5) The water supply reliability rate and the expected value of water shortage are weighted and integrated to generate water supply security indicators.

[0116] For example, based on the core objectives and actual needs of water supply security, the weighting coefficients of water supply reliability and expected water shortage are determined, and the two are integrated by weighted summation to combine the evaluation results of different dimensions into a unified quantitative indicator, thereby generating a water supply security indicator.

[0117] like Figure 2 As shown, for ecological maintenance, ecological health indicators are generated by quantifying the ecological flow guarantee rate and water quality improvement degree, including:

[0118] S201: Obtain the ecological flow demand sequence of the key ecological zone in the target reservoir's control basin, and the actual discharge flow sequence within the corresponding scheduling cycle.

[0119] For example, key ecological zones within the target reservoir's control basin are identified, and based on ecological protection standards and the basin's ecological characteristics, the minimum flow thresholds required for each key ecological zone at different times are determined, generating a continuous ecological flow demand sequence.

[0120] Simultaneously, actual discharge data from the reservoir to key ecological areas within the corresponding scheduling cycle were collected and compiled into actual discharge flow sequences. The consistency of the time granularity and coverage of the two types of sequences was then verified. Key ecological areas include fish spawning grounds and wetland reserves.

[0121] S202: Based on the ecological flow demand sequence and the actual outflow sequence, calculate the percentage of time during which the actual outflow reaches or exceeds the ecological flow demand, and generate the ecological flow guarantee rate.

[0122] For example, based on the verified ecological flow demand sequence and the actual outflow sequence, the actual outflow is compared time-by-time to see if it reaches or exceeds the ecological flow demand of the same period. The total duration of the time period that meets the demand is calculated, and the proportion of this duration to the total duration of the scheduling cycle is calculated to generate the ecological flow guarantee rate, so as to reflect the stability of the ecological flow supply.

[0123] S203: Collect water quality monitoring datasets of the target reservoir before the implementation of ecological scheduling and the corresponding water quality monitoring datasets after the implementation.

[0124] For example, a typical time period before the implementation of ecological scheduling is selected, and water quality monitoring data of key ecological areas are collected during this period to generate a preliminary water quality monitoring dataset. After the implementation of ecological scheduling, water quality data from the same monitoring points are collected during the same period corresponding to the preliminary time period to generate a subsequent water quality monitoring dataset. The monitoring indicators and sampling frequencies of the two datasets are then checked for compatibility. The water quality monitoring data includes indicators such as dissolved oxygen, ammonia nitrogen, and total phosphorus.

[0125] S204: Based on the water quality monitoring dataset before execution and the water quality monitoring dataset for the corresponding period after execution, calculate the improvement rate of key water quality indicators and generate the water quality improvement degree.

[0126] For example, based on the verified water quality monitoring datasets before and after the adjustment, the change in the value of each key water quality indicator after the adjustment is calculated relative to the value before the adjustment. The improvement of each indicator is then considered to generate a water quality improvement rate reflecting the effectiveness of the water quality improvement. This includes metrics such as the improvement percentage and the reduction rate.

[0127] S205: The ecological flow guarantee rate and water quality improvement are weighted and integrated to generate ecological health indicators.

[0128] For example, based on the core objectives of watershed ecological protection, the weight allocation of ecological flow guarantee rate and water quality improvement is determined, and the two indicators are integrated into a unified quantitative result using a weighted comprehensive method to generate an ecological health indicator. Core objectives include prioritizing the guarantee of ecological base flow or focusing on improving water quality.

[0129] In one embodiment, a real-time mirror digital twin platform for the physical watershed is constructed by integrating an Internet of Things (IoT) sensing network, a hydrological and hydrodynamic model, and a historical operational database, including:

[0130] (1) The real-time hydrological data and reservoir operation status data collected by the Internet of Things sensing network are fused and processed to obtain a dynamic monitoring dataset.

[0131] For example, an IoT sensing network deployed within the watershed continuously collects real-time hydrological data and reservoir operational status data. This raw data undergoes cleaning, format standardization, and spatiotemporal alignment to ensure consistency and continuity, and is then integrated into a dynamic monitoring dataset. The IoT sensing network includes devices such as hydrological stations, water level gauges, and dam sensors. Real-time hydrological data includes rainfall, runoff, and water level changes. Reservoir operational status data includes gate opening, water storage capacity, and outflow.

[0132] (2) Input the dynamic monitoring dataset into the hydrological and hydrodynamic model to simulate the evolution path of the basin flood and the water resource scheduling process, and generate a flood simulation result set.

[0133] For example, the processed dynamic monitoring dataset is used as input parameters and imported into a pre-built hydrological and hydrodynamic model. The model simulates the complete path of a flood from formation to evolution within the basin based on real-time data, while simultaneously simulating the allocation of water resources among reservoirs under different scheduling strategies, generating a flood simulation result set containing various simulation parameters and results. The pre-built hydrological and hydrodynamic model includes a basin runoff generation and confluence calculation module, a flood wave propagation module, etc. The complete path of a flood from formation to evolution includes changes in flow velocity and inundation extent. The allocation process includes water distribution and water level regulation.

[0134] (3) Based on the flood simulation results set and the engineering characteristic data in the historical operation database, a real-time mirror digital twin platform of the physical watershed is constructed.

[0135] For example, information such as the characteristic data of water conservancy projects within the basin, historical hydrological process records, and past scheduling schemes stored in the historical operation database are extracted and spatiotemporally correlated and fused with the flood simulation result set. Through 3D modeling and real-time rendering technology, a real-time mirror digital twin platform capable of dynamically mapping the physical basin's topography, hydrological status, and engineering operation is constructed. This includes water conservancy project characteristic data such as reservoir capacity curves, dam parameters, and canal water conveyance capacity.

[0136] In one embodiment, based on a multi-entity value dataset, a value allocation algorithm is used to calculate the benefit distribution ratio of each entity under different scheduling schemes, including:

[0137] (1) Based on the theory of cooperative game theory, a multi-agent negotiation model is constructed, the contribution characteristic function of each agent in the scheduling alliance is defined, and a set of characteristic functions is generated.

[0138] For example, based on a multi-agent value dataset, a cooperative game theory framework is introduced to clarify the roles and responsibilities of each scheduling agent in the scheduling alliance, thus constructing a multi-agent negotiation model. In this model, based on the resource input scale, benefit contribution level, and alliance synergy of each agent, characteristic functions that quantify the value contribution of each agent in different alliance combinations are defined, and a set of characteristic functions is generated. These characteristic functions include: scheduling agents such as flood control management departments, power generation companies, and water supply units; resource input scale such as funds, facilities, and technology; and benefit contribution level such as flood control and disaster reduction, power generation, and water supply.

[0139] (2) The set of characteristic functions is calculated iteratively by the Shapley value algorithm to generate the initial allocation ratio set.

[0140] For example, the Shapley value algorithm is selected as the core calculation tool, and the feature function set is incorporated into the algorithm operation framework. By sorting, weighting and accumulating the marginal contributions of each subject in all possible alliance subsets, the basic interest share that each subject should receive under different scheduling schemes is iteratively calculated to generate an initial allocation ratio set.

[0141] (3) Introduce a risk-contribution correction factor to modify the initial allocation ratio set and generate the benefit allocation ratio.

[0142] For example, this paper analyzes the risk factors that various entities may face during the implementation of different scheduling schemes, and designs a risk-contribution correction factor based on the matching degree between the entities' actual contributions and risk-bearing capabilities. The proportions in the initial allocation ratio set are dynamically adjusted according to the correction factor, so that the allocation results reflect both the differences in contributions and the fairness of risk-bearing. The corrected distribution ratios of benefits for each entity under different scheduling schemes are then generated. Risk factors include scheduling failure risk and revenue fluctuation risk.

[0143] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0144] In one embodiment, such as Figure 3 As shown, this application also provides a digital twin watershed reservoir group scheduling system 300, which includes:

[0145] Digital twin building module 301 is used to build a real-time mirror digital twin platform for a physical watershed by integrating IoT sensing networks, hydrological and hydrodynamic models and historical operation databases.

[0146] The multi-dimensional assessment module 302 is used to establish a multi-dimensional value assessment indicator system based on the digital twin platform, including flood control benefits, power generation revenue, water supply security and ecological maintenance.

[0147] The data collection module 303 is used to collect cost input and benefit output data of each scheduling entity through a preset digital platform according to a multi-dimensional value assessment index system, so as to obtain a multi-entity value dataset.

[0148] The benefit allocation module 304 is used to calculate the benefit allocation ratio of each subject under different scheduling schemes based on a multi-subject value dataset and a value allocation algorithm.

[0149] The scheme generation module 305 is used to generate a set of candidate scheduling schemes based on the benefit distribution ratio and preset flood control safety constraints and preset ecological flow constraints.

[0150] The scheme evaluation module 306 is used to simulate and evaluate the set of candidate scheduling schemes and output the final scheduling scheme.

[0151] Specifically, the digital twin construction module 301 collects real-time hydrological data and reservoir operation status data within the watershed through an IoT sensing network. This data is then cleaned, deduplicated, standardized, and spatiotemporally matched to generate a dynamic monitoring dataset. This dynamic monitoring dataset is input into a hydrological and hydrodynamic model to simulate the path of floods from their generation to their evolution, and the allocation of water resources among reservoirs, generating a flood simulation result set. Furthermore, engineering characteristic data such as structural parameters, functional characteristics, and historical operation records of water conservancy projects within the watershed, stored in a historical operation database, are retrieved and integrated with the flood simulation result set to construct a digital twin platform that can reflect the entire physical watershed in real time.

[0152] Based on the aforementioned digital twin platform, the multi-dimensional assessment module 302 constructs flood control value indicators by combining flood risk simulation data and historical disaster loss records from the platform for flood control benefits; power generation revenue indicators by considering reservoir power generation efficiency, water inflow, and electricity pricing mechanisms from the platform; water supply security indicators by considering water demand and actual water supply performance monitored by the platform; and ecological health indicators by considering ecological flow data and water quality monitoring results from the platform. Finally, the module integrates these four types of indicators using a combined weighting method to generate a multi-dimensional value assessment indicator system.

[0153] The data collection module 303 is based on a multi-dimensional value assessment index system, which clarifies the specific scope of cost input and benefit output of each dispatching entity in dimensions such as flood control, power generation, water supply, and ecological maintenance. It collects the above data in a targeted manner through a preset digital platform, and after classification, verification, and integration, generates a multi-entity value dataset.

[0154] The benefit allocation module 304, based on a multi-agent value dataset, introduces cooperative game theory to construct a multi-agent negotiation model. It defines contribution characteristic functions for each agent in different scheduling alliances, including resource input and benefit contribution, generating a set of characteristic functions. The Shapley value algorithm is used to iteratively calculate the characteristic function set to obtain an initial allocation ratio set. Combining the matching degree between the risk level borne by each agent during the scheduling process and their actual contribution, a risk-contribution correction factor is introduced to correct the initial allocation ratio set, resulting in the benefit allocation ratio for each agent under different scheduling schemes.

[0155] The scheme generation module 305, based on the benefit distribution ratio and combined with flood control safety constraints and ecological flow constraints, comprehensively considers the total amount and spatiotemporal distribution characteristics of watershed water resources, the scheduling capacity of each reservoir, etc., to construct multiple sets of scheduling parameter combinations, including reservoir water storage and release periods, flow allocation, and output arrangements. The above parameter combinations are then checked for compliance, and combinations that do not meet the constraints are eliminated, generating a set of candidate scheduling schemes.

[0156] The scheme evaluation module 306 inputs each scheme from the candidate scheduling scheme set into the digital twin platform, simulates the execution process of each scheme under different hydrological scenarios, and obtains simulation results of each scheme in terms of flood control effect, power generation benefit, water supply security, and ecological impact. Based on the preset evaluation criteria, the simulation results of each scheme are comprehensively evaluated from multiple dimensions, and the scheme with the best overall performance is selected. After feasibility verification, the final scheduling scheme is output.

[0157] The multi-dimensional evaluation module 302 is also used for:

[0158] Based on flood control benefits, flood control value indicators are generated through quantitative analysis of flood risk maps and loss avoidance rates.

[0159] Based on power generation revenue, a power generation revenue index is generated by calculating based on peak-valley electricity price curves and power generation efficiency models.

[0160] Based on water supply security, water supply reliability rate and expected water shortage are used for evaluation to generate water supply security indicators.

[0161] Based on ecological maintenance, ecological health indicators are generated by quantifying the ecological flow guarantee rate and water quality improvement degree.

[0162] By combining flood control value indicators, power generation revenue indicators, water supply security indicators, and ecological health indicators through a combined weighting method, a multi-dimensional evaluation indicator system is generated.

[0163] The multi-dimensional evaluation module 302 is also used for:

[0164] Acquire historical flood data and geographic information data of the target watershed, and construct a flood risk map based on the historical flood data and geographic information data to reflect the relationship between the inundation range and the degree of loss under different flood frequencies;

[0165] The following formula was used to collect data on the difference between actual flood losses and expected losses during multiple rounds of reservoir scheduling. Statistical analysis of the difference data was then performed to obtain the loss avoidance rate:

[0166]

[0167] Among them, R avoid E represents the loss avoidance rate, n represents the total number of reservoir scheduling rounds, and E represents the loss avoidance rate. i Let A represent the expected loss value in round i. i This represents the actual loss value in round i;

[0168] The potential loss value corresponding to each risk level in the flood risk map is correlated and mapped with the loss avoidance rate to generate the flood control risk impact value corresponding to each scheduling scheme.

[0169] The impact values ​​of flood control risks are normalized to generate flood control value indicators.

[0170] The multi-dimensional evaluation module 302 is also used for:

[0171] Obtain the daily water demand sequence of each water user within the target water supply cycle and the actual water supply sequence of the reservoir during the corresponding time period;

[0172] Based on the daily water demand sequence and the actual water supply sequence, the ratio of the number of days with sufficient water supply to the total number of monitoring days is calculated to generate the water supply reliability rate.

[0173] Collect data on reservoir inflow and water demand from water users during historical periods of continuous drought, and construct a water shortage probability distribution model.

[0174] By using a water shortage probability distribution model, the expected value of water shortage under different drought scenarios is calculated, and the expected value of water shortage is generated.

[0175] The water supply reliability rate and the expected water shortage are weighted and integrated to generate a water supply security index.

[0176] The multi-dimensional evaluation module 302 is also used for:

[0177] Obtain the ecological flow demand sequence of key ecological zones in the target reservoir's control basin, and the actual outflow sequence within the corresponding scheduling cycle;

[0178] Based on the ecological flow demand sequence and the actual outflow sequence, the percentage of time the actual outflow reaches or exceeds the ecological flow demand is statistically analyzed to generate the ecological flow guarantee rate.

[0179] Collect water quality monitoring datasets of the target reservoir before the implementation of ecological scheduling and water quality monitoring datasets for the corresponding time period after the implementation;

[0180] Based on the water quality monitoring dataset before implementation and the water quality monitoring dataset for the corresponding period after implementation, the improvement rate of key water quality indicators is calculated, and the water quality improvement degree is generated.

[0181] The ecological flow guarantee rate and water quality improvement rate are weighted and integrated to generate ecological health indicators.

[0182] In one embodiment, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0183] In one embodiment, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described method embodiments.

[0184] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and 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 modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0185] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A digital twin-based method for scheduling watershed reservoir groups, characterized in that: The method includes: By integrating IoT sensing networks, hydrological and hydrodynamic models, and historical operational databases, a real-time mirror digital twin platform for the physical watershed is constructed. Based on the aforementioned digital twin platform, a multi-dimensional value assessment index system is established, including flood control benefits, power generation revenue, water supply security, and ecological maintenance. Based on the aforementioned multi-dimensional value assessment index system, cost input and benefit output data of each scheduling entity are collected through a pre-set digital platform to obtain a multi-entity value dataset. Based on the multi-entity value dataset, the benefit distribution ratio of each entity under different scheduling schemes is calculated using a value allocation algorithm. A set of candidate scheduling schemes is generated based on the aforementioned benefit distribution ratio, preset flood control safety constraints, and preset ecological flow constraints. The candidate scheduling scheme set is simulated and evaluated, and the final scheduling scheme is output.

2. The digital twin watershed reservoir group scheduling method according to claim 1, characterized in that, Based on the aforementioned digital twin platform, a multi-dimensional value assessment index system is established, encompassing flood control benefits, power generation revenue, water supply security, and ecological maintenance. This system includes: Regarding the aforementioned flood control benefits, flood control value indicators are generated through quantitative analysis of flood risk maps and loss avoidance rates. The power generation revenue is calculated based on the peak-valley electricity price curve and the power generation efficiency model to generate a power generation revenue index. For the aforementioned water supply security, the water supply reliability rate and the expected value of water shortage are used for evaluation to generate water supply security indicators; For the aforementioned ecological maintenance, ecological health indicators are generated based on the ecological flow guarantee rate and water quality improvement degree. The flood control value index, the power generation revenue index, the water supply security index, and the ecological health index are integrated by combining the weighting method to generate the multi-dimensional evaluation index system.

3. The digital twin watershed reservoir group scheduling method according to claim 2, characterized in that, Regarding the aforementioned flood control benefits, flood control value indicators are generated through quantitative analysis of flood risk maps and loss avoidance rates, including: Acquire historical flood data and geographic information data of the target watershed, and construct a flood risk map based on the historical flood data and geographic information data to reflect the correlation between the inundation range and the degree of loss under different flood frequencies; The following formula is used to collect data on the difference between actual flood losses and expected losses during multiple rounds of reservoir scheduling. Statistical analysis of this difference data yields the loss avoidance rate: Among them, R avoid E represents the loss avoidance rate, n represents the total number of reservoir scheduling rounds, and E represents the loss avoidance rate. i Let A represent the expected loss value in round i. i This represents the actual loss value in round i; The potential loss value corresponding to each risk level in the flood risk map is correlated and mapped with the loss avoidance rate to generate the flood control risk impact value corresponding to each scheduling scheme. The flood control risk impact value is normalized to generate the flood control value index.

4. The digital twin watershed reservoir group scheduling method according to claim 2, characterized in that, The aforementioned water supply security is assessed using water supply reliability and expected water shortage values ​​to generate water supply safety indicators, including: Obtain the daily water demand sequence of each water user within the target water supply cycle and the actual water supply sequence of the reservoir during the corresponding time period; Based on the daily water demand sequence and the actual water supply sequence, the ratio of the number of days with water supply satisfaction to the total number of monitoring days is calculated to generate the water supply reliability rate. Collect data on reservoir inflow and water demand from water users during historical periods of continuous drought, and construct a water shortage probability distribution model. Using the water shortage probability distribution model, the expected value of water shortage under different drought scenarios is calculated, and the expected value of water shortage is generated. The water supply reliability rate and the expected water shortage value are weighted and fused to generate the water supply safety index.

5. The digital twin watershed reservoir group scheduling method according to claim 2, characterized in that, The aforementioned ecological maintenance is quantified based on the ecological flow guarantee rate and water quality improvement degree to generate ecological health indicators, including: Obtain the ecological flow demand sequence of key ecological zones in the target reservoir's control basin, and the actual outflow sequence within the corresponding scheduling cycle; Based on the ecological flow demand sequence and the actual outflow sequence, the percentage of time during which the actual outflow reaches or exceeds the ecological flow demand is calculated to generate the ecological flow guarantee rate. Collect water quality monitoring datasets of the target reservoir before the implementation of ecological scheduling and water quality monitoring datasets for the corresponding time period after the implementation. Based on the water quality monitoring dataset before execution and the water quality monitoring dataset for the corresponding time period after execution, the improvement rate of key water quality indicators is calculated, and the water quality improvement degree is generated. The ecological flow guarantee rate and the water quality improvement rate are weighted and fused to generate the ecological health index.

6. The digital twin watershed reservoir group scheduling method according to claim 1, characterized in that, The aforementioned method, which integrates IoT sensing networks, hydrological and hydrodynamic models, and historical operational databases to construct a real-time mirror digital twin platform for the physical watershed, includes: The real-time hydrological data and reservoir operation status data collected by the Internet of Things sensing network are fused and processed to obtain a dynamic monitoring dataset; The dynamic monitoring dataset is input into the hydrological and hydrodynamic model to simulate the flood evolution path and water resource allocation process in the basin, and to generate a flood simulation result set. Based on the flood simulation result set and the engineering characteristic data in the historical operation database, a real-time mirror digital twin platform for the physical watershed is constructed.

7. The digital twin watershed reservoir group scheduling method according to claim 1, characterized in that, The step of calculating the benefit distribution ratio of each entity under different scheduling schemes based on the multi-entity value dataset and using a value allocation algorithm includes: A multi-agent negotiation model is constructed based on cooperative game theory, defining the contribution characteristic function of each agent in the scheduling alliance and generating a set of characteristic functions. The set of characteristic functions is iteratively calculated using the Shapley value algorithm to generate an initial allocation ratio set; A risk-contribution correction factor is introduced to correct the initial allocation ratio set, thereby generating the benefit allocation ratio.

8. A digital twin watershed reservoir group scheduling system, characterized in that, The system includes: The digital twin building module is used to construct a real-time mirror digital twin platform for a physical watershed by integrating IoT sensing networks, hydrological and hydrodynamic models, and historical operational databases. The multi-dimensional evaluation module is used to establish a multi-dimensional value evaluation index system based on the digital twin platform, including flood control benefits, power generation revenue, water supply security and ecological maintenance. The data collection module is used to collect cost input and benefit output data of each scheduling entity through a preset digital platform according to the multi-dimensional value assessment index system, so as to obtain a multi-entity value dataset. The benefit allocation module is used to calculate the benefit allocation ratio of each entity under different scheduling schemes based on the multi-entity value dataset and through a value allocation algorithm. The scheme generation module is used to generate a set of candidate scheduling schemes based on the aforementioned benefit distribution ratio and preset flood control safety constraints and preset ecological flow constraints. The scheme evaluation module is used to simulate and evaluate the set of candidate scheduling schemes and output the final scheduling scheme.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the digital twin watershed reservoir group scheduling method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the digital twin watershed reservoir group scheduling method as described in any one of claims 1 to 7.

Images