A fast and high-fidelity integrated operation risk assessment method for watershed water and scenery storage

Abstract

The present application belongs to the technical field of water, wind, light and storage multi-capability complementary risk evaluation, and relates to a kind of fast high-fidelity integrated operation risk evaluation method of watershed water, wind, light and storage. Firstly, the triggering and transmission mechanism of multi-dimensional risk under multiple uncertainty conditions is analyzed, and the risk generation principle is revealed. Secondly, a risk index quantification method based on boundary damage depth is proposed, which can quantify multi-dimensional risk according to the current operating state and boundary conditions, and help the simulation algorithm directly output risk assessment results. A panoramic time series simulation model integrating expert knowledge is constructed. The explicit water-electricity-storage scheduling capability is introduced into the operation simulation framework through expert experience rules, which not only helps to determine the priority of multi-dimensional risk avoidance, but also ensures the interpretability and reliability of the simulation results. The present application also solves the low efficiency problem caused by the introduction of a large number of 0-1 variables to realize the nonlinear curve of water and electricity and the charging and discharging logic of energy storage, as well as the difficulty of identifying the water-electricity-storage scheduling sequence.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of water, wind, light and storage multi-energy complementary risk evaluation, and relates to a rapid and high-fidelity watershed water, wind, light and storage integrated operation risk evaluation method. BACKGROUND

[0002] Due to the regulation ability of the hydropower station and the water-wind-light multi-energy complementary characteristics, the complementary operation of variable renewable energy and hydropower has become an important way to improve the effective consumption capacity of variable renewable energy and guarantee the reliable supply of electricity. This is particularly important in areas rich in water resources. Large-scale storage is integrated to enhance the regulation ability, and ultimately forms a water-wind-light-storage integrated energy base.

[0003] In order to ensure the continuous complementary operation of the water-wind-light-storage integrated energy base, meet the comprehensive utilization requirements of the reservoir in navigation, flood control, ecology, etc., and reduce the wear of the water turbine, the operator needs to conduct timely and forward-looking risk assessment on the agreed dispatching plan, so as to maintain sufficient spinning reserve or hydropower storage to deal with unexpected accidents and fluctuations of runoff and variable renewable energy. The traditional risk assessment method usually starts from risk optimization simulation under uncertain scenarios to identify risk indicators (such as power shortage, curtailment and other constraint violations) that the current dispatching decision may involve, and then statistically analyzes the simulation results to quantify the risk level, including the probability distribution and expected value of the risk indicators.

[0004] However, the traditional method for evaluating the operation risk of the water-wind-light-storage integrated energy base faces two major challenges:

[0005] Multi-dimensional uncertainty: There are combinations of uncertain scenarios of wind power, photovoltaic and runoff in the operation environment of the water-wind-light-storage integrated energy base, which leads to huge calculation pressure.

[0006] Temporal and spatial coupling and nonlinear characteristics of cascade hydropower stations: There is a hydraulic connection between cascade hydropower reservoirs (the drainage of the upstream hydropower station constitutes an important part of the inflow of the downstream station), and the water power output function has nonlinear characteristics (describing the nonlinear relationship between the flow rate of the water turbine, the water head and the hydraulic power generation efficiency); These characteristics not only increase the complexity of solving mathematical models, but also increase the linearization approximation error of commonly used solving methods such as mixed integer linear programming (MILP) and linear programming (LP).

[0007] Therefore, how to fully consider the complex operation characteristics of hydropower and realize rapid and accurate risk evaluation of the water-wind-light-storage integrated energy base has become a challenge to be solved. SUMMARY

[0008] This invention combines a risk indicator system based on operational boundary failure with expert knowledge rules to construct a rapid and high-fidelity risk assessment framework for integrated hydro-wind-solar-storage energy bases. First, it analyzes the triggering and transmission mechanisms of multidimensional risks under multiple uncertainties, revealing the principles of risk generation. Second, it proposes a risk indicator quantification method based on boundary damage depth, which can quantify multidimensional risks according to the current operating state and boundary conditions, assisting simulation algorithms in directly outputting risk assessment results. A panoramic time-series simulation model integrating expert knowledge is constructed. By explicitly incorporating hydropower-storage dispatch capabilities into the operational simulation framework through expert experience rules, it not only helps determine the priority of multidimensional risk avoidance but also ensures the interpretability and reliability of the simulation results.

[0009] Technical solution of the present invention:

[0010] A rapid and high-fidelity method for risk assessment of integrated watershed hydropower, wind power, solar power, and energy storage operations, comprising the following steps:

[0011] Step (1): Initial calculation conditions;

[0012] This includes a set of output scenarios characterizing the uncertainty of new energy sources, historical runoff observation data, basic data of hydropower stations and comprehensive utilization operation boundaries, operation boundaries of DC transmission lines, and day-ahead dispatch plans to be evaluated.

[0013] Step (2): Construct a risk quantification index system based on the depth of boundary breach;

[0014] Indicator 1: Renewable Energy Abandonment Risk (VPAR). When hydropower is operating at its minimum technical output and energy storage (ES) charging power reaches its maximum value (considering energy storage capacity), the output of the hydro-wind-solar-storage complementary power generation still exceeds the power capacity boundary of the DC transmission line; the excess portion is defined as the renewable energy abandonment risk.

[0015]

[0016] In the formula: This is the quantitative value of the renewable energy curtailment risk for power plant i branch k, in MW; It is the actual output value of the renewable energy in time period t of power plant i branch k under the scenario s of uncertain renewable energy output, in MW; It is the charging power (MW) of the energy storage power station at branch plant k of power station i in time period t under the uncertain output scenario of new energy. The power capacity (MW) of the DC transmission line corresponding to power plant i branch k is given by the power plant i branch k. It is a set of uncertainty scenarios related to new energy. It is a set of scheduling periods; It is a collection of power stations; It is a collection of branch plants under power station i; It is the number of hours in a scheduling period, h.

[0017] Indicator 2: Spilled water risk (SWR). Spilled water risk arises when both the reservoir capacity boundary and the turbine flow boundary are breached simultaneously.

[0018]

[0019] In the formula: This refers to the risk of water wastage at power station i. 3 / s; It is the reservoir capacity of power station i in time period t under the uncertain output scenario of new energy source s, in 10,000 m³. 3 ; This refers to the upper limit of the reservoir capacity of the power station, in 10,000 m³. 3 ; It is the number of seconds in a scheduling period, s; The outflow m from reservoir i at power station s during time period t is the outflow rate of power station i in scenario s. 3 / s; This is the upper limit of the power generation flow of power station i, m 3 / s.

[0020] Indicator 3: Generation Agreement Deviation Risk (GADR). When a hydro-wind-solar-storage complementary power generation system fails to generate electricity as planned, the GADR is defined as the deviation between the actual power generation of the system and the planned output boundary in the dispatch scheme, expressed as the absolute value of the deviation.

[0021]

[0022] In the formula, It is the performance deviation risk of power plant i branch k, MW; It is the actual complementary power output (MW) of power plant i branch k in time period t under scenario s; It is the planned output of power plant i in MW during time period t.

[0023] Step (3): Construct expert experience knowledge rules to form a risk assessment simulation framework;

[0024] The current operating status (such as renewable energy output deviation, turbine overflow, reservoir water level, and energy storage state of charge (SOC)) is input into the rule base, and rule matching determines the coordinated scheduling of multiple power sources in the complementary system. The specific rules of this invention are as follows: First, energy storage and hydropower output are controlled to 0, and renewable energy curtailment risk is controlled by ensuring the physical boundary of the power capacity of the transmission lines; then, during periods when complementary output (at which point only renewable energy output is greater than 0) exceeds the planned output, the constraints of energy storage capacity and charging power boundaries are considered, and positive performance deviation risk is controlled through energy storage charging; during periods when complementary output is lower than the planned output, the constraints of hydropower output boundaries are considered, and hydropower output is increased to make complementary output equal to the planned output; during periods when complementary output is higher than the planned output, hydropower output is kept at 0 to control performance deviation risk; next, reservoir capacity boundaries and power generation flow boundaries are considered to minimize reservoir water curtailment, and hydropower output is updated according to the hydropower water consumption rate. At this point, the performance deviation risk degrades from the optimal to suboptimal of the aforementioned steps. The above rules enable the priority control of risk management to be optimized sequentially for risks of new energy curtailment, water curtailment, and performance deviation.

[0025] Step (4): Construct a rapid, high-fidelity, multi-dimensional risk assessment model framework;

[0026] A multidimensional risk assessment framework based on expert knowledge and with clear interpretability is constructed. This framework aims to replace traditional global search methods with ordered boundary constraint-guided risk assessment, achieving efficient and effective risk assessment, thereby significantly reducing the search space and computation time. The framework process is as follows.

[0027] (4.1) Initialize the algorithm framework: Input the scenario set that characterizes the uncertainty of wind power and photovoltaic power, the complementary power output plan in the day-ahead scheduling scheme, and the initial state of energy storage (ES) and reservoir;

[0028] (4.2) Iterate through the scenarios, objects, and time periods in sequence to initialize. .

[0029] (4.3) Conduct risk assessment for the s-th scenario and the t-th time period: determine the output of renewable energy. Contributing to the plan The deviation between them. Based on The relative deviation is combined with the risk avoidance priority of expert knowledge, and the risk assessment algorithm a–d is called in sequence to calculate the risk index after compensating for the deviation in renewable energy output.

[0030] (4.4) Settings Repeat step (4.3) until the time set is complete. Processing of all time periods.

[0031] (4.5) Settings Repeat steps (4.3)-(4.4) until the scene set is complete. The system processes all scenarios to obtain risk quantification results for all time periods in all scenarios.

[0032] (4.6) Output the simulation and evaluation results of risk avoidance.

[0033] The risk assessment algorithms a–d described are the hydropower output calculation strategies for the VPAR assessment algorithm, power state simulation algorithm, SWR assessment algorithm, and GADR assessment algorithm, respectively. Details are as follows:

[0034] Algorithm a: VPAR assessment algorithm. Maintaining hydropower at its minimum technical output level while increasing energy storage charging power to mitigate VPAR, and calculating... First, input the line capacity boundary. Energy storage charge boundary Energy storage power boundary Hours of the time period Energy storage charging and discharging efficiency and new energy power generation scenarios Then determine if... If so, then perform the following assignments respectively: If not, then order Then, update the energy storage state of charge based on the current energy storage capacity: .in, This represents the energy storage charging and discharging power; negative values ​​indicate charging, and positive values ​​indicate discharging. Contribute to the actual consumption of new energy; For storing the current load of energy; This involves calculating the curtailment power of renewable energy. After traversing all scenarios and time periods, the final calculation of the risk of renewable energy curtailment is performed.

[0035] .

[0036] Algorithm b: Power state simulation algorithm. It initializes hydropower output by minimizing the GADR (Gross Energy Demand Ratio), while considering the physical limitations of the hydropower station's output ceiling. First, it inputs the day-ahead planned complementary output. The recent hydropower retention plan has been implemented. Hydropower station output upper limit Hydropower station output lower boundary .when season ;when season Next, a new judgment will be made: when season ;when Finally, let , .in, This refers to the actual power output transmitted from the hydropower plant. It is the actual total output of the hydropower plant.

[0037] Algorithm c: SWR assessment algorithm. Adjusts reservoir outflow and storage capacity to mitigate SWR, and calculates... First, input the lower limit of the reservoir outflow. Upper limit of reservoir capacity Hydropower generation flow limit Then, algorithm e is used, with... Calculate and order ,if ,make By traversing each reservoir from upstream to downstream, the calculations were obtained. Determine if If so: then let Update the corresponding If at this time Then let , .make Determine if If so: then let Update the corresponding If at this time Then let .in, The power generation flow rate of the hydropower branch plant; This refers to the outflow from the reservoir. For reservoir power generation flow; These are the water discharge rate and the water shortage rate, respectively. These are intermediate variables: the adjustment of reservoir power generation flow due to boundary exceedance; the adjustment of water discharge flow; and the adjustment of outflow flow. After traversing all scenarios and time periods, the risk of water discharge is calculated:

[0038] .

[0039] Algorithm d: GADR evaluation algorithm. Based on the hydraulic state adjustment in Algorithm c, the hydropower output is updated, then the energy storage operation state is adjusted to avoid GADR, and calculations are performed. First, input the water consumption rate. and power generation flow adjustment Then, iterate through the branch plants of the hydroelectric power station: for each branch plant, according to Calculate the adjustment amount of power transmission output from the hydropower plant. And calculate the energy storage output adjustment amount. To make up for Next, order After traversing all scenarios and time periods, calculate the performance deviation risk:

[0040] .

[0041] This section details the call to algorithm e mentioned in algorithm c. This algorithm is based on a bivariate nonlinear function trial calculation algorithm, which calculates the hydropower generation flow and the water discharge flow by inputting the hydropower output, as follows:

[0042] Algorithm e: A hydropower output calculation strategy based on a bivariate nonlinear function trial calculation algorithm. Deriving the power generation flow rate from hydropower output requires iterative calculations. Therefore, an electricity-determined water flow simulation algorithm is proposed: the convergence accuracy is set to a very small value. The simulation algorithm for determining water level using electricity is as follows: First, input the tailrace water level and discharge curve of the hydropower station. Water level and reservoir capacity curve Water head consumption rate curve Hydropower station branch plant output Reservoir power generation flow , upstream reservoir outflow Reservoir capacity , number of seconds within the time period Head loss Natural runoff flowing into the reservoir Let the intermediate variable and initialize .

[0043] make If the current power station is the upstream power station in the basin ( ),make ,otherwise Finally, let , ,if only This process will continue in a loop.

[0044] Finally, ,return .

[0045] in This refers to the real-time tailwater level of the reservoir. Real-time reservoir water level; This represents the real-time water head of the reservoir.

[0046] The beneficial effects of this invention are as follows: This invention solves the inefficiency caused by the introduction of a large number of 0-1 variables in traditional optimization simulation methods to realize the nonlinear curve of hydropower and the charging and discharging logic of energy storage, as well as the difficulty in identifying the scheduling sequence of hydropower-energy storage, and provides technical support for risk assessment of integrated operation of watershed hydropower, wind power, solar power and energy storage. Attached Figure Description

[0047] Figure 1 It is the triggering and transmission mechanism of multidimensional risks under the influence of multiple uncertainties;

[0048] Figure 2 It is a flowchart for embedding expert knowledge into a risk assessment framework;

[0049] Figure 3 This is a risk assessment framework diagram. Detailed Implementation

[0050] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0051] This embodiment uses an integrated hydropower, wind power, solar power, and energy storage base located in the lower reaches of the Jinsha River as a case study. The total installed capacity of hydropower, solar power, and wind power in this base is 44,800 MW, 7,109 MW, and 3,377 MW, respectively. Due to the long-term joint dispatch of cascade hydropower in the lower reaches of the Jinsha River, there is considerable experience regarding the risks of hydropower operation. When large-scale new energy sources are planned and integrated, the giant hydropower stations and energy storage will play a regulatory role, inevitably altering the original risk characteristics. Therefore, this embodiment uses the integrated hydropower, wind power, solar power, and energy storage base in the planning stage of the lower reaches of the Jinsha River as a simulation case to develop a rapid, high-fidelity risk assessment algorithm, providing technical support for risk assessment of the complementary operation of hydropower, wind power, solar power, and energy storage.

[0052] Step (1): Initial calculation conditions;

[0053] This includes a set of output scenarios characterizing the uncertainty of new energy sources, historical runoff observation data, basic data of hydropower stations and comprehensive utilization operation boundaries, operation boundaries of DC transmission lines, and day-ahead dispatch plans to be evaluated.

[0054] Step (2): Construct a risk quantification index system based on the depth of boundary breach;

[0055] For integrated hydro-wind-solar-storage energy bases, the uncertainties in runoff, wind power, and solar power output may make it difficult for the system to strictly adhere to the power generation plans and corresponding boundaries agreed upon by the grid. Therefore, hydropower and energy storage (ES) are needed to mitigate and compensate for the fluctuations in renewable energy. However, when hydropower operation exceeds its operational boundaries, it may trigger multi-dimensional risks involving complex coupling relationships between hydraulics and electricity. Figure 1 This demonstrates the triggering and transmission mechanisms of such multidimensional risks.

[0056] Indicator 1: Renewable Energy Abandonment Risk (VPAR). When hydropower is operating at its minimum technical output and energy storage (ES) charging power reaches its maximum value (considering energy storage capacity), the output of the hydro-wind-solar-storage complementary power generation still exceeds the power capacity boundary of the DC transmission line; the excess portion is defined as the renewable energy abandonment risk.

[0057]

[0058] In the formula: This is the quantitative value of the renewable energy curtailment risk for power plant i branch k, in MW; It is the actual output value of the renewable energy in time period t of power plant i branch k under the scenario s of uncertain renewable energy output, in MW; It refers to the energy storage power station at branch plant k of power station i in the time period t under the uncertain output scenario of new energy; The power capacity (MW) of the DC transmission line corresponding to power plant i branch k is given by the power plant i branch k. It is a set of uncertainty scenarios related to new energy. It is a set of scheduling periods; It is a collection of power stations; It is a collection of branch plants under power station i; It is the number of hours in a scheduling period, h.

[0059] Indicator 2: Spilled water risk (SWR). Spilled water risk arises when both the reservoir capacity boundary and the turbine flow boundary are breached simultaneously.

[0060]

[0061] In the formula: This refers to the risk of water wastage at power station i. 3 / s; It is the reservoir capacity of power station i in time period t under the uncertain output scenario of new energy source s, in 10,000 m³. 3 ; This refers to the upper limit of the reservoir capacity of the power station, in 10,000 m³. 3 ; It is the number of seconds in a scheduling period, s; The outflow m from reservoir i at power station s during time period t is the outflow rate of power station i in scenario s. 3 / s; This is the upper limit of the power generation flow of power station i, m 3 / s.

[0062] Indicator 3: Generation Agreement Deviation Risk (GADR). When a hydro-wind-solar-storage complementary power generation system fails to generate electricity as planned, the GADR is defined as the deviation between the actual power generation of the system and the planned output boundary in the dispatch scheme, expressed as the absolute value of the deviation.

[0063]

[0064] In the formula, It is the performance deviation risk of power plant i branch k, MW; It is the actual complementary power output (MW) of power plant i branch k in time period t under scenario s; It is the planned output of power plant i in MW during time period t.

[0065] Step (3): Construct expert experience knowledge rules to form a risk assessment simulation framework;

[0066] The risk aversion rules adopted by dispatchers or the scheduling priorities of hydropower and energy storage have a decisive impact on the results of multidimensional risk quantification. For example, when power generation needs to be reduced, if energy storage responds first, the adjustment amount of hydropower will be smaller, or even unnecessary. In this case, the SWR is significantly lower than when hydropower responds first. Therefore, an expert knowledge method based on dispatchers' risk aversion preferences and scheduling sequence experience is proposed to fully utilize the flexibility of hydropower and energy storage to achieve accurate and efficient risk assessment. Expert knowledge expresses condition-action mapping rules through logical judgments (IF–THEN). A complete set of rules covering all possible scenarios forms the expert knowledge rule base, which is integrated into the risk assessment framework. The current operating status (such as new energy output deviation, turbine overflow, reservoir water level, and energy storage state of charge (SOC)) is input into the rule base, and rule matching determines the coordinated scheduling of multiple power sources in the complementary system. The specific rules of this invention are as follows: First, the output of energy storage and hydropower is controlled to zero, and the risk of new energy curtailment is controlled by ensuring the physical boundary of the power capacity of the transmission lines. Then, during periods when the complementary output (at which point only the output of new energy is greater than zero) exceeds the planned output, the constraints of the energy storage charge capacity and charging power boundary are considered, and the risk of positive performance deviation is controlled by energy storage charging. During periods when the complementary output is lower than the planned output, the constraint of the hydropower output boundary is considered, and the hydropower output is increased to make the complementary output equal to the planned output. During periods when the complementary output is higher than the planned output, the hydropower output is kept at zero to control the performance deviation risk. Next, the reservoir capacity boundary and the power generation flow boundary are considered to minimize reservoir water curtailment, and the hydropower output is updated according to the hydropower water consumption rate. At this point, the performance deviation risk will degrade from the optimal to the suboptimal of the aforementioned steps. Through the above rules, the risk management priority control of new energy curtailment risk, water curtailment risk, and performance deviation risk can be optimized sequentially. Figure 2 This paper demonstrates the process of embedding expert knowledge into a risk assessment optimization framework. Without introducing binary variables, the knowledge rules clarify multi-dimensional risk aversion priorities to guide the flexible adjustment of complementary systems, satisfying the different risk preferences of schedulers and thus generating risk assessment results that better reflect actual operating conditions. Simultaneously, this method preserves the interpretability of the decision-making process.

[0067] Step (4): Construct a rapid, high-fidelity, multi-dimensional risk assessment model framework;

[0068] A multidimensional risk assessment framework based on expert knowledge and with clear interpretability is constructed. This framework aims to replace traditional global search methods with ordered boundary constraint-guided risk assessment, achieving efficient and effective risk assessment, thereby significantly reducing the search space and computation time. The framework flowchart is shown below. Figure 3 As shown, the framework process is as follows.

[0069] (4.1) Initialize the algorithm framework: Input the scenario set that characterizes the uncertainty of wind power and photovoltaic power, the complementary power output plan in the day-ahead scheduling scheme, and the initial state of energy storage (ES) and reservoir;

[0070] (4.2) Iterate through the scenarios, objects, and time periods in sequence to initialize. .

[0071] (4.3) Conduct risk assessment for the s-th scenario and the t-th time period: determine the output of renewable energy. Contributing to the plan The deviation between them. Based on The relative deviation is combined with the risk aversion priority of expert knowledge, and the algorithm ad is called in sequence to calculate the risk index after compensating for the deviation in renewable energy output.

[0072] (4.4) Settings Repeat step (4.3) until the time set is complete. Processing of all time periods.

[0073] (4.5) Settings Repeat steps (4.3)–(4.4) until the scene set is complete. The system processes all scenarios to obtain risk quantification results for all time periods in all scenarios.

[0074] (4.6) Output the simulation and evaluation results of risk avoidance.

[0075] For the risk assessment process within the proposed methodological framework, four algorithms were developed: VPAR assessment algorithm, power state simulation algorithm, SWR assessment algorithm, and GADR assessment algorithm. Details are as follows:

[0076] Algorithm a: VPAR assessment algorithm. Maintaining hydropower at its minimum technical output level while increasing energy storage charging power to mitigate VPAR, and calculating... First, input the line capacity boundary. Energy storage charge boundary Energy storage power boundary Hours of the time period Energy storage charging and discharging efficiency and new energy power generation scenarios Then determine if... If so, then perform the following assignments respectively: If not, then order Then, update the energy storage state of charge based on the current energy storage capacity: .in, This represents the energy storage charging and discharging power; negative values ​​indicate charging, and positive values ​​indicate discharging. Contribute to the actual consumption of new energy; For storing the current load of energy; This involves calculating the curtailment power of renewable energy. After traversing all scenarios and time periods, the final calculation of the risk of renewable energy curtailment is performed.

[0077] .

[0078] Algorithm b: Power state simulation algorithm. It initializes hydropower output by minimizing the GADR (Gross Energy Demand Ratio), while considering the physical limitations of the hydropower station's output ceiling. First, it inputs the day-ahead planned complementary output. The recent hydropower retention plan has been implemented. Hydropower station output upper limit Hydropower station output lower boundary .when season ;when season Next, a new judgment will be made: when season ;when hour, Finally, let .in, This refers to the actual power output transmitted from the hydropower plant. It is the actual total output of the hydropower plant.

[0079] Algorithm c: SWR assessment algorithm. Adjusts reservoir outflow and storage capacity to mitigate SWR, and calculates... First, input the lower limit of the reservoir outflow. Upper limit of reservoir capacity Hydropower generation flow limit Then, algorithm e is used, with... Calculate and order ,if ,make By traversing each reservoir from upstream to downstream, the calculations were obtained. Determine if If so: then let Update the corresponding If at this time Then let , .make Determine if If so: then let Update the corresponding If at this time Then let .in, The power generation flow rate of the hydropower branch plant; This refers to the outflow from the reservoir. For reservoir power generation flow; These are the water discharge rate and the water shortage rate, respectively. These are intermediate variables: the adjustment of reservoir power generation flow due to boundary exceedance; the adjustment of water discharge flow; and the adjustment of outflow flow. After traversing all scenarios and time periods, the risk of water discharge is calculated:

[0080] .

[0081] Algorithm d: GADR evaluation algorithm. Based on the hydraulic state adjustment in Algorithm c, the hydropower output is updated, then the energy storage operation state is adjusted to avoid GADR, and calculations are performed. First, input the water consumption rate. and power generation flow adjustment Then, iterate through the branch plants of the hydroelectric power station: for each branch plant, according to Calculate the adjustment amount of power transmission output from the hydropower plant. And calculate the energy storage output adjustment amount. To make up for Next, order After traversing all scenarios and time periods, calculate the performance deviation risk:

[0082] .

[0083] This section details the call to algorithm e mentioned in algorithm c. This algorithm is based on a bivariate nonlinear function trial calculation algorithm, which calculates the hydropower generation flow and the water discharge flow by inputting the hydropower output, as follows:

[0084] Algorithm e: A hydropower output calculation strategy based on a bivariate nonlinear function trial calculation algorithm. Deriving the power generation flow rate from hydropower output requires iterative calculations. Therefore, an electricity-determined water flow simulation algorithm is proposed: the convergence accuracy is set to a very small value. The simulation algorithm for determining water level using electricity is as follows: First, input the tailrace water level and discharge curve of the hydropower station. Water level and reservoir capacity curve Water head consumption rate curve Hydropower station branch plant output Reservoir power generation flow , upstream reservoir outflow Reservoir capacity , number of seconds within the time period Head loss Natural runoff flowing into the reservoir Let the intermediate variable and initialize .

[0085] make If the current power station is the upstream power station in the basin ( ),make ,otherwise Finally, let , ,if only This process will continue in a loop.

[0086] Finally, ,return .

[0087] in This refers to the real-time tailwater level of the reservoir. Real-time reservoir water level; This represents the real-time water head of the reservoir.

[0088] Accurate risk assessment depends on the accuracy of simulation algorithms in solving nonlinear constraints, primarily reflected in hydraulic calculations related to reservoir water level and flow rate. Water consumption rate is a crucial quantitative parameter for measuring hydraulic conversion efficiency and can be used as a quantitative indicator of simulation accuracy. Table 1 compares the root mean square error (MAE), mean square error (MSE), and coefficient of determination (R²) of the head-water consumption rate functions fitted by the two algorithms with the actual water consumption rate curves. Specifically, the MAE values ​​calculated by the comparative algorithm (CA) are 0.10 m³ / kWh, 0.09 m³ / kWh, and 0.01 m³ / kWh, respectively. In contrast, the proposed algorithm (PA) achieves all of these values ​​below 0.01 m³ / kWh, representing reductions of 98.5%, 96.3%, and 82.0% compared to CA, respectively. Furthermore, PA's MSE is reduced by 99.9%, 99.9%, and 96.4% compared to CA, respectively. Therefore, the proposed algorithm significantly improves computational efficiency compared to traditional algorithms that optimize the solver. Table 2 shows the solution times of CA and PA under five typical inflow conditions. From dry to wet, PA's solution efficiency is 13423 times, 11162 times, 5176 times, 12397 times, and 19034 times that of CA in a scenario of 10, respectively. In a scenario of 100, PA's solution efficiency is 213231 times, 51941 times, 5612 times, 26226 times, and 28886 times that of CA, respectively. In a scenario of 200, PA's solution efficiency is 239326 times, 54493 times, 6718 times, 32936 times, and 34805 times that of CA, respectively. It is evident that as the number of scenarios increases, PA significantly improves the reduction rate of solution time, demonstrating its significant computational efficiency advantage. In summary, the risk assessment algorithm for the integrated watershed hydro-wind-solar-storage base proposed in this invention can significantly improve simulation accuracy while greatly optimizing computational efficiency, providing an effective basis for forward-looking risk assessment and rapid response.

[0089] Table 1. Comparison of simulation accuracy between the proposed algorithm (this invention) and traditional comparison algorithms.

[0090]

[0091] Table 2 Comparison of computational efficiency between the proposed algorithm (this invention) and traditional comparison algorithms.

[0092]

Claims

1. A fast and high-fidelity integrated operation risk assessment method for watershed water and scenery, characterized in that, The steps are as follows: Step (1): initial calculation condition; The output scene set characterizing new energy uncertainty, runoff historical observation data, hydropower station basic data, and comprehensive utilization operation boundary, DC transmission line operation boundary, and day-ahead scheduling plan to be evaluated are included. Step (2): constructing a risk quantification index system based on boundary damage depth; Index 1: new energy curtailment risk VPAR; when the hydropower is operated at the minimum technical output and the storage charging power reaches the maximum value, the water-wind-light-storage complementary output still exceeds the power capacity boundary of the DC transmission line; the excess part is defined as the new energy curtailment risk; Index 2: water abandonment risk SWR; when the reservoir capacity boundary and the water turbine flow boundary are broken at the same time, the water abandonment risk is generated; Index 3: performance deviation risk GADR; when the water-wind-light-storage complementary power generation system cannot generate power according to the plan, the performance deviation risk is defined as the deviation between the actual power generation of the system and the output plan boundary in the scheduling scheme, and the absolute value of the deviation represents the size of GADR; Step (3): constructing expert experience knowledge rules to form a risk evaluation simulation framework; The current operating state is input into the rule base, and the coordinated scheduling of various power sources in the complementary system is determined by rule matching; Step (4): constructing a fast and high-fidelity multi-dimensional risk assessment model framework; (4.1) initialization algorithm framework: input the scene set characterizing wind power and photovoltaic uncertainty, the complementary output plan in the day-ahead scheduling scheme, and the initial state of the storage and the reservoir; (4.2) Iterating over scenarios, objects, and time periods in order, initialize ; (4.3) Risk assessment for the s-th scenario, the t-th time period: determine the deviation between the renewable energy output and the planned output ; based on the relative deviation between and the risk aversion priority of expert knowledge, sequentially call the risk assessment algorithms a-d to calculate the risk index after compensating for the deviation of the renewable energy output; (4.4) Set , repeat step (4.3) until the processing of all time periods in the time set is completed; (4.5) setting , repeating steps (4.3)-(4.4) until the processing of all scenes in the scene set is completed, thereby obtaining the risk quantification results for all scene periods; (4.6) output risk avoidance simulation and evaluation results; The risk assessment algorithms a-d are respectively VPAR evaluation algorithm, power state simulation algorithm, SWR evaluation algorithm, and hydropower output calculation strategy of GADR evaluation algorithm; the specific steps are as follows: Algorithm a: VPAR evaluation algorithm; keep the water and electricity at the minimum technical output level, while increasing the energy storage charging power to relieve VPAR, and calculate ; first input the line capacity boundary , the energy storage state of charge boundary , the energy storage power boundary , the time period hours , the energy storage charging and discharging efficiency and the new energy output scene ; then determine whether : if yes, respectively, the following assignment is made: , ; if not, let ; then update the energy storage state of charge according to the current period energy storage power: ; wherein, is the energy storage charging and discharging power, negative value represents charging, positive value represents discharging; is the actual consumption of new energy output; is the current state of charge of energy storage; is the new energy curtailment power; is the charging power of the energy storage power station of the power plant i branch k under the new energy uncertainty output scene s in the time period t; after traversing all the scenes and time periods, finally calculate the new energy curtailment risk: ； Algorithm b: power state simulation algorithm; initialize the water and electricity output by minimizing the GADR, while considering the physical limit of the upper limit of the water and electricity station output; first input the day-ahead plan complementary output , day-ahead water and electricity retention plan output , upper limit of water and electricity station output , lower limit of water and electricity station output ; when , let ; when , let ; next, make a new judgment: when , let ; when , ; finally, let ; wherein, is the actual water and electricity substation export power; is the actual total output of the water and electricity substation; Algorithm c: SWR evaluation algorithm; adjust the reservoir outflow and reservoir storage to mitigate SWR, and calculate ; first input the lower limit of reservoir outflow , the upper limit of reservoir capacity , the upper limit of hydropower generation flow ; then use algorithm e to calculate , and let , if , let ; traverse each reservoir from upstream to downstream to calculate ; judge whether , if yes: let , update the corresponding , if at this time, let ; let ; judge whether , if yes: let , update the corresponding , if at this time, let ; wherein is the hydropower plant generation flow; is the reservoir outflow; is the reservoir generation flow; is the abandoned water flow and the water shortage flow, respectively; is an intermediate variable, which is the reservoir generation flow adjustment amount, the abandoned water flow adjustment amount and the outflow adjustment amount due to boundary overrun; after traversing all scenarios and time periods, calculate the abandoned water risk:​ ； Algorithm d: GADR evaluation algorithm; update the hydropower output according to the hydraulic state adjusted in algorithm c, then adjust the operating state of the energy storage to avoid GADR, and calculate ; first, input the water consumption rate and the generation flow adjustment amount ; then, traverse the sub-stations of the hydropower station: for each sub-station, calculate the hydropower sub-station export output adjustment amount according to and , and calculate the energy storage output adjustment amount to make up for ; then, let ; after traversing all scenarios and time periods, calculate the compliance deviation risk: ； Algorithm e: water power output calculation strategy based on binary nonlinear function trial algorithm Iterative calculation is needed to derive power output from water power output. The power-to-water simulation algorithm is proposed, which sets the convergence accuracy to a very small value . The power-to-water simulation algorithm flow is as follows: first, input the tailwater level-discharge curve of the hydropower station , the water level-storage capacity curve , the water head-water consumption rate curve , the power output of the hydropower station , the power generation flow of the reservoir , the outflow of the upstream reservoir , the reservoir storage capacity , the number of seconds in the period , the water head loss , the natural runoff into the reservoir ; let the intermediate variable , and initialize ; Let , if the current power station is the most upstream power station of the river basin , let , otherwise ; finally, let , , as long as , this step is looped. Finally let , return ; wherein is the real-time reservoir tailwater level; is the real-time reservoir water level; is the real-time reservoir head.

2. The method according to claim 1, wherein, The indexes in step (2) are as follows: Index 1: new energy curtailment risk VPAR; ; In the formula: is the new energy curtailment risk quantitative value of power station i substation k, MW; is the actual output value of new energy of power station i substation k in time period t under new energy uncertainty output scenario s, MW; is the charging power of energy storage power station in time period t under new energy uncertainty output scenario s of power station i substation k, MW; is the power capacity of the direct current transmission line corresponding to power station i substation k, MW; is the new energy uncertainty scenario number set; is the dispatching time period set; is the power station set; is the substation set under power station i; is the number of hours in a dispatching time period; Index 2: water abandonment risk SWR; ; wherein: is the risk of abandoned water of power station i; is the reservoir capacity of power station i at time period t under the new energy uncertainty output scenario s; is the upper limit of the reservoir capacity of power station i; is the number of seconds in a dispatch period; is the outflow of power station i reservoir at time period t under scenario s; is the upper limit of the power generation flow of power station i; Index 3: performance deviation risk GADR; ; wherein is the performance deviation risk of power plant i substation k, MW; is the actual complementary output of power plant i substation k in time period t under scenario s, MW; is the planned output of power plant i substation k in time period t.

3. The method according to claim 1, wherein, In step (3), the specific rules are as follows: first, control the storage and hydropower output to 0, control the new energy curtailment risk by guaranteeing the power capacity physical boundary of the transmission line; then, in the period when the complementary output exceeds the planned output, consider the restriction of the storage charging capacity and charging power boundary, control the positive performance deviation risk by storage charging; in the period when the complementary output is lower than the planned output, consider the restriction of the hydropower output boundary, increase the hydropower output to make the complementary output equal to the planned output, and control the performance deviation risk by keeping the hydropower output to 0 when the complementary output is higher than the planned output; next, consider the reservoir capacity boundary and power generation flow boundary to minimize the reservoir water abandonment, and update the hydropower output according to the water consumption rate of the hydropower, at this time, the performance deviation risk will be degraded from the optimal to the suboptimal in the foregoing steps.

Images