A scheduling mode discrimination method for a river basin level water, wind and light integration base

Abstract

The application discloses a scheduling mode discrimination method for a river basin level water, wind and light integration base, obtains water power station group, new energy, energy storage and external sending channel operation data, and is divided into long-term, medium-term and short-term scheduling scales; a decoupling reference operation state of new energy priority external sending and water power execution certainty plan is constructed, decoupling discrimination indexes respectively representing channel congestion risk, reservoir regulation and storage demand and prediction error and peak regulation influence are calculated; according to historical samples and operation safety boundary, a tolerance threshold is calibrated, a comprehensive coupling degree is calculated and decoupling, critical coupling or coupling states are discriminated; according to the discrimination result, a water power certainty scheduling model or a water, wind and light, water, wind and light storage multi-energy complementary scheduling model of the corresponding scale is called, and the tolerance threshold is updated according to the actual operation result. The application can reduce unnecessary coupling calculation, improve scheduling efficiency, safety and adaptability.

Description

Technical Field

[0001] This invention relates to the field of new energy joint scheduling and reservoir optimized operation technology, and in particular to a scheduling mode discrimination method for watershed-level integrated water, wind and solar power bases. Background Technology

[0002] A basin-wide integrated hydropower, wind, and solar power base typically consists of a cascade of hydropower stations, wind farms, photovoltaic power stations, and transmission lines. Wind and solar power exhibit seasonal variations over the long term, and may experience consecutive days of high or low power generation over the medium term. In the short term, they are subject to prediction errors, counter-peak shaving, and random fluctuations. While hydropower has strong regulation capabilities, it is also constrained by flood control, water supply, ecological factors, navigation, water levels, flow rates, and the safe operation of generating units. Therefore, whether changes in renewable energy output need to be compensated by hydropower or energy storage depends on the installed capacity of renewable energy, the capacity of transmission lines, the adjustable capacity of reservoirs, and the regulation capacity of hydropower units. Not all operational scenarios require coupled hydropower, wind, and solar power scheduling.

[0003] Existing methods typically establish joint optimization models for hydropower, wind power, and solar power on long-term, day-ahead, intraday, or real-time timescales, directly utilizing hydropower or energy storage to compensate for fluctuations in renewable energy output and prediction errors. However, they lack a step to determine whether complementary dispatch needs to be initiated before model solving. When transmission capacity is sufficient, reservoir storage capacity is large, and the impact of renewable energy fluctuations is minimal, direct coupling optimization increases data preparation and model computation. Conversely, when transmission channels are congested, fluctuations persist for several days, or short-term prediction errors exceed the tolerance range of hydropower, timely activation of complementary dispatch at the appropriate scale is necessary. Therefore, it is crucial to assess the degree of coupling between hydropower and renewable energy based on long-term transmission channel utilization, medium-term reservoir storage needs, and short-term power generation reliability and peak-shaving requirements, and accordingly select a deterministic dispatch mode or a multi-energy complementary dispatch mode. Summary of the Invention

[0004] The technical problem to be solved by this invention is: how to accurately determine the coupling demand of hydropower and new energy at different time scales before optimizing the scheduling of hydropower, wind and solar power, and select the corresponding scheduling mode accordingly.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A method for determining the scheduling mode of a watershed-level integrated hydro-wind-solar base includes the following steps: S1. Obtain the operation data of hydropower station groups, new energy operation data, external transmission channel operation data, energy storage operation data and scheduling business parameters of the basin-level integrated water, wind and solar power base, perform time alignment and anomaly processing on the acquired data, and form a scheduling mode discrimination dataset; S2. Based on the scheduling cycle and scheduling step size, the scheduling mode discrimination is divided into long-term scale discrimination, medium-term scale discrimination and short-term scale discrimination. S3. Construct an uncoupled reference operating state when hydropower-wind-solar complementary scheduling is not carried out, in accordance with the principle of prioritizing the transmission or consumption of new energy and the execution of deterministic scheduling plans for hydropower station groups. S4. Based on the uncoupled reference operating state, calculate the long-term decoupling discrimination index, the medium-term decoupling discrimination index, and the short-term decoupling discrimination index respectively. The long-term decoupling discrimination index is used to characterize the utilization pressure of the transmission channel under the extreme output superposition of hydropower and new energy. The medium-term decoupling discrimination index is used to characterize the reservoir regulation demand required for hydropower to compensate for the continuous fluctuations of new energy. The short-term decoupling discrimination index is used to characterize the impact of new energy prediction error and output fluctuation on the reliability and peak-shaving capacity of hydropower generation. S5. Based on historical operation samples, typical scenario samples, planned grid-connected capacity and operation safety boundaries, determine the tolerance thresholds corresponding to each decoupling discrimination index. S6. Compare each decoupling discrimination index with the corresponding tolerance threshold to obtain the normalized limit violation coefficient corresponding to each scheduling time scale, and determine the corresponding comprehensive coupling degree based on the normalized limit violation coefficient. S7. When the overall coupling degree is not greater than the first mode switching threshold, the corresponding scheduling time scale is determined to be in a decoupled state; when the overall coupling degree is not less than the second mode switching threshold, the corresponding scheduling time scale is determined to be in a coupled state; when the overall coupling degree is between the first mode switching threshold and the second mode switching threshold, the corresponding scheduling time scale is determined to be in a critically coupled state. The first mode switching threshold and the second mode switching threshold are determined based on historical operation samples, scheduling risk analysis results, and mode switching requirements. The first mode switching threshold is less than the second mode switching threshold, and the second mode switching threshold is not less than 1. S8. For the decoupled state, call the hydropower deterministic scheduling model under the condition of priority transmission or priority consumption of new energy; for the coupled state, call the hydro-wind-solar multi-energy complementary scheduling model or hydro-wind-solar-storage multi-energy complementary scheduling model that matches the corresponding scheduling time scale; for the critical coupled state, select the scheduling model after performing the operation risk analysis.

[0006] Preferably, step S1 includes: S11. Obtain the runoff, reservoir water level, available storage capacity, water consumption rate, planned output, minimum output and maximum output of the hydropower station group; S12. Obtain historical power output, predicted power output, grid-connected installed capacity, extreme power output rate, prediction error, and simultaneous power output rate of wind farms and photovoltaic power plants. S13. Obtain the transmission capacity, planned transmission power, and safe transmission limits of the external transmission channel; S14. Obtain the rated power, rated capacity, state of charge, and charging / discharging power of the energy storage system; S15. Perform time-scale unification, abnormal data identification, and missing data repair on the acquired data.

[0007] Preferably, step S2 includes: Long-term timeframes correspond to monthly, ten-day, or high-water season scheduling. Medium-term timeframes correspond to weekly, ten-day, or consecutive days of scheduling. Short-term scales correspond to intraday, hourly, or 15-minute scheduling.

[0008] Preferably, the long-term scale decoupling discrimination index is calculated according to the following formula: ; In the formula, This is the long-term decoupling criterion for the transmission channel i. This represents the maximum power generation capacity of the hydropower station group corresponding to transmission channel i during the high-water season. This represents the maximum output of renewable energy during the high-water season for transmission channel i. The available transport capacity of the external transport channel i during the high-water season; The maximum value among the long-term scale decoupling discrimination indicators corresponding to each external transmission channel is determined as the long-term scale discrimination indicator for the watershed-level integrated water, wind, and solar power base.

[0009] Preferably, the calculation of the mid-scale decoupling discriminant index includes: S51. Calculate the water storage regulation demand corresponding to the hydropower station h under the condition of continuous large-scale renewable energy generation according to the following formula: ; S52. Calculate the drawdown regulation demand corresponding to the hydropower station h under the condition of continuous low renewable energy generation according to the following formula: ; In the formula, , For the set of intermediate scheduling periods, and These represent the extreme maximum and extreme minimum power output rates of new energy sources, respectively. For new energy installed capacity, Contribute to new energy forecasting For the water consumption rate of the hydropower station, For scheduling step size; S53. Calculate the mid-scale decoupling index corresponding to hydropower station h according to the following formula: ; In the formula, For hydroelectric power station The reservoir's storage capacity For hydroelectric power station The capacity to reduce storage capacity; The maximum value among the mid-scale decoupling discriminant indices corresponding to each hydropower station is determined as the mid-scale discriminant index.

[0010] Preferably, the short-term decoupling criteria include power generation reliability criteria and peak shaving impact criteria; The power generation reliability index is calculated according to the following formula: ; The peak-shaving impact index is calculated using the following formula: ; In the formula, The remaining prediction error of new energy sources after deducting the available compensation capacity of energy storage is included. For hydroelectric power station The minimum stable output, This is to account for the remaining peak-shaving demand of new energy sources after deducting the energy storage regulation capacity. and Hydropower stations The available upward adjustment capability and the available downward adjustment capability; The maximum value among the power generation reliability index and the peak-shaving impact index is determined as the short-term scale decoupling discrimination index corresponding to hydropower station h, and the maximum value among the short-term scale decoupling discrimination indices corresponding to each hydropower station is determined as the short-term scale discrimination index of the basin-level integrated hydro-wind-solar base.

[0011] Preferably, step S5 includes: S71. Classify historical operation samples according to hydropower station, external transmission channel, new energy transmission area, hydrological period and scheduling season; S72. Calculate the long-term scale decoupling discrimination index, medium-term scale decoupling discrimination index and short-term scale decoupling discrimination index corresponding to various historical operation samples; S73. Remove abnormal samples corresponding to missing data, equipment malfunctions, extreme maintenance, and abnormal power outages. S74. Use the quantile method to determine the initial tolerance threshold corresponding to each decoupling discrimination index; S75. Based on the safe transport limit of the external transmission channel, the safe water level of the reservoir, the adjustable capacity of the reservoir, the minimum output and peak-shaving capacity of the hydropower unit, the initial tolerance threshold is corrected to obtain the final tolerance threshold. S76. The tolerance threshold shall be updated on a rolling basis according to the changes in the installed capacity of new energy, the capacity of external transmission channels and the operating boundary of reservoirs.

[0012] Preferably, the overall coupling degree is calculated according to the following formula: ; ; In the formula, Let j be the normalized limit-crossing coefficient corresponding to the decoupling discrimination index j under the scheduling time scale q. To decouple the discrimination index, For the corresponding tolerance threshold, The overall coupling degree corresponding to the scheduling time scale q. As the indicator weight, The influence coefficient of the maximum limit exceeded indicator; The number of decoupling criteria corresponding to the scheduling time scale q; ,and ; ; .

[0013] Preferably, step S7 further includes: S91. When any decoupling judgment index reaches the corresponding safety limit threshold, the corresponding scheduling time scale is directly judged as a forced coupling state. S92. When the number of scheduling periods with a comprehensive coupling degree that is not less than the second mode switching threshold reaches the first preset number, the scheduling mode will be switched from decoupling to coupling. S93. When the number of scheduling periods with a comprehensive coupling degree not exceeding the first mode switching threshold reaches the second preset number, the scheduling mode will be switched from the coupled state to the decoupled state.

[0014] Preferably, it further includes: S101. Based on the scheduling time scale at which the coupling state is reached, select a long-term water-wind-solar complementary scheduling model, a medium-term water-wind-solar complementary scheduling model, or a short-term water-wind-solar-storage complementary scheduling model. S102. Output the hydropower output plan, wind power output plan, photovoltaic output plan, reservoir water level process, outflow, water abandonment flow, power of transmission channels and energy storage status corresponding to the selected scheduling model; S103. Obtain the actual new energy output, actual hydropower output, actual reservoir water level, and actual power of the external transmission channel after the scheduling is executed; S104. Update historical operating samples and corresponding tolerance thresholds based on actual operating data.

[0015] The beneficial effects of this invention are as follows: by constructing long-term, medium-term and short-term multi-scale decoupling discrimination indicators, the coupling demand of hydropower and new energy is judged before the calculation of the complementary scheduling model, and a deterministic scheduling mode or a multi-energy complementary scheduling mode is selected according to the discrimination results, reducing unnecessary complex coupling calculations; at the same time, it can identify risks such as channel congestion, insufficient reservoir storage and limited short-term peak-shaving capacity, and realize the adaptive switching of scheduling mode with the changes in the installed capacity of new energy and the operating boundary, thereby improving scheduling efficiency, safety and interpretability of results. Attached Figure Description

[0016] Figure 1 This is a flowchart of the scheduling mode discrimination method for watershed-level integrated water, wind and solar power bases according to the present invention. Figure 2 This is a flowchart of the data acquisition and processing for scheduling mode discrimination in this invention; Figure 3 This is a flowchart illustrating the calculation process of the long-term, medium-term, and short-term decoupling discrimination indexes of this invention. Figure 4 This is a flowchart illustrating the calibration and updating process of the decoupling discrimination index tolerance threshold in this invention. Figure 5 This is a flowchart illustrating the scheduling mode discrimination and switching process based on comprehensive coupling degree of the present invention. Figure 6 This is a flowchart of the scheduling model invocation and operation result feedback update process of the present invention. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments. It should be understood that the following embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0018] Example 1: S1, such as Figure 2 As shown, operational data is obtained from cascade hydropower stations, wind farms, photovoltaic power stations, energy storage systems, power transmission channels, and dispatching systems. Operational data for hydropower station groups includes runoff, reservoir water level, available storage capacity, water consumption rate, planned output, minimum output, and maximum output; operational data for wind farms and photovoltaic power stations includes historical output, predicted output, grid-connected installed capacity, extreme output rate, prediction error, and output simultaneity rate; operational data for power transmission channels includes transmission capacity, planned power transmission, and safe transmission limits; operational data for energy storage systems includes rated power, rated capacity, state of charge, and charging / discharging power. The acquired data undergoes time-scale unification, outlier identification, and missing data repair. It is then categorized according to the scheduling time scale to form long-term, medium-term, and short-term datasets. S2. Based on the scheduling cycle and scheduling step size, the scheduling mode discrimination is divided into long-term scale discrimination, medium-term scale discrimination and short-term scale discrimination; Long-term timeframes correspond to monthly, ten-day, or high-water season scheduling, used to determine whether seasonal surges in renewable energy generation cause congestion in transmission channels and whether hydropower is needed for seasonal power transfer. Medium-term timeframes correspond to weekly, ten-day, or multi-day scheduling, used to determine whether reservoirs need to store water or regulate drawdowns for continuous large or small renewable energy generation over several days. Short-term timeframes correspond to intraday, hourly, or 15-minute scheduling, used to determine whether renewable energy forecasting errors, counter-peak shaving, and random fluctuations require hydropower or energy storage to provide flexibility support. S3, such as Figure 1 As shown, a non-coupled reference operating state is constructed when hydropower-wind-solar complementary scheduling is not carried out, based on the principle of prioritizing the transmission or consumption of new energy and the execution of deterministic scheduling plans for hydropower station groups. Under uncoupled reference operation, new energy sources are prioritized for external transmission or consumption based on predicted output, while hydropower stations operate according to their original deterministic dispatch plan. Based on planned hydropower output, predicted new energy output, reservoir water level, available storage capacity, and transmission capacity of external transmission channels, reference hydropower output, reference new energy output, reference reservoir water level, reference storage capacity, and reference transmission power of external transmission channels are obtained. S4, such as Figure 3 As shown, based on the uncoupled reference operating state, the long-term decoupling discrimination index, the medium-term decoupling discrimination index, and the short-term decoupling discrimination index are calculated respectively. Long-term decoupling criteria are used to characterize the pressure on transmission channels when the maximum hydropower generation capacity and the maximum output of new energy sources occur simultaneously during the high-water season. They are calculated using the following formula: ; In the formula, This is the long-term decoupling criterion for the transmission channel i. This represents the maximum power generation capacity of the hydropower station group corresponding to transmission channel i during the high-water season. This represents the maximum output of renewable energy during the high-water season for transmission channel i. The available transport capacity of the external transport channel i during the high-water season; The maximum value among the long-term scale decoupling discriminant indices corresponding to each transmission channel is determined as the long-term scale discriminant index, i.e.: ; In the formula, I represents the set of external transmission channels; The medium-scale decoupling discriminant index is used to characterize the amount of water storage required by a hydropower station to compensate for fluctuations in new energy sources under the conditions of continuous large or small new energy generation. Under the condition of continuous and large-scale generation of new energy sources, the water storage regulation demand corresponding to hydropower station h is calculated according to the following formula: ; Under the condition of continuous low-level renewable energy generation, the drawdown regulation demand corresponding to the hydropower station h is calculated according to the following formula: ; In the formula, , For the set of intermediate scheduling periods, and These represent the extreme maximum and extreme minimum power output rates of new energy sources, respectively. For new energy installed capacity, Contribute to new energy forecasting For the water consumption rate of the hydropower station, For scheduling step size; The mid-scale decoupling index corresponding to hydropower station h is calculated according to the following formula: ; In the formula, For hydroelectric power station The reservoir's storage capacity For hydroelectric power station The capacity of the reservoir can be reduced; The maximum value among the mid-scale decoupling discriminant indices corresponding to each hydropower station is determined as the mid-scale discriminant index, i.e.: ; In the formula, H represents the set of hydropower stations participating in the dispatch mode determination; Short-term decoupling criteria include power generation reliability indicators and peak-shaving impact indicators; The power generation reliability index is calculated according to the following formula: ; The peak-shaving impact index is calculated using the following formula: ; In the formula, The remaining prediction error of new energy sources after deducting the available compensation capacity of energy storage is included. For hydroelectric power station The minimum stable output, This is to account for the remaining peak-shaving demand of new energy sources after deducting the energy storage regulation capacity. and Hydropower stations The available upward adjustment capability and the available downward adjustment capability; The maximum value among the power generation reliability index and the peak-shaving impact index is determined as the short-term decoupling discriminant index corresponding to the hydropower station h, that is: ; The maximum value among the short-term scale decoupling discriminant indices corresponding to each hydropower station is determined as the short-term scale discriminant index, i.e.: ; S5, such as Figure 4 As shown, based on historical operation samples, typical scenario samples, planned grid-connected capacity and operation safety boundaries, the tolerance thresholds corresponding to the long-term scale decoupling discrimination index, the medium-term scale decoupling discrimination index and the short-term scale decoupling discrimination index are determined respectively. Historical operation samples were classified according to hydropower stations, transmission channels, new energy transmission areas, hydrological periods, and scheduling seasons, and long-term scale decoupling discriminant indicators, medium-term scale decoupling discriminant indicators, and short-term scale decoupling discriminant indicators were calculated for each type of historical operation sample. Abnormal samples corresponding to missing data, equipment malfunctions, extreme maintenance, and abnormal power outages are removed. The quantile method is used to determine the initial tolerance threshold for each decoupling discrimination index. In this embodiment, the 80th percentile value is used as the initial tolerance threshold. Based on the safe transport limit of the external transmission channel, the safe water level of the reservoir, the adjustable capacity of the reservoir, the minimum output and peak-shaving capacity of the hydropower unit, the initial tolerance threshold is corrected to obtain the final tolerance threshold. When the installed capacity of new energy sources, the capacity of external transmission channels, or the operating boundary of reservoirs change, update the historical operating samples and recalibrate the tolerance threshold. S6. Compare each decoupling discrimination index with its corresponding tolerance threshold and calculate the normalized threshold violation coefficient: ; In the formula, Let j be the normalized limit-crossing coefficient corresponding to the decoupling discrimination index j under the scheduling time scale q. To decouple the discrimination index, This corresponds to the tolerance threshold; Calculate the overall coupling degree corresponding to the scheduling time scale q based on the normalized limit violation coefficient: ; in: ; In the formula, The overall coupling degree corresponding to the scheduling time scale q; The number of decoupling criteria corresponding to the scheduling time scale q; As the indicator weight; The influence coefficient of the maximum limit exceeded indicator; S7, such as Figure 5 As shown, the scheduling status of the corresponding scheduling time scale is determined based on the decoupling discrimination index, the safety limit threshold, and the comprehensive coupling degree. When any decoupling judgment index reaches the corresponding safety limit threshold, the corresponding scheduling time scale is directly judged as a forced coupling state; When none of the decoupling criteria reach the safety limit threshold and the overall coupling degree is not greater than the first mode switching threshold, the corresponding scheduling time scale is determined to be in a decoupling state. When none of the decoupling criteria reach the safety limit threshold and the overall coupling degree is not less than the second mode switching threshold, the corresponding scheduling time scale is determined to be in a coupled state. When the overall coupling degree is between the first mode switching threshold and the second mode switching threshold, the corresponding scheduling time scale is determined to be in a critical coupling state, and an operational risk analysis is performed. The first mode switching threshold and the second mode switching threshold are determined based on historical operation samples, scheduling risk analysis results, and mode switching requirements. The first mode switching threshold is less than the second mode switching threshold, and the second mode switching threshold is not less than 1. When the number of scheduling periods with a comprehensive coupling degree continuously not less than the second mode switching threshold reaches a first preset number, the scheduling mode is switched from decoupled to coupled. When the number of scheduling periods with a comprehensive coupling degree continuously not greater than the first mode switching threshold reaches a second preset number, the scheduling mode is switched from coupled to decoupled. In this embodiment, the first mode switching threshold is 0.9, and the second mode switching threshold is 1.1. S8, such as Figure 6 As shown, the corresponding scheduling model is invoked based on the scheduling mode determination result; When the corresponding scheduling time scale is in a decoupled state, the hydropower deterministic scheduling model under the condition of priority transmission or priority consumption of new energy is invoked. When the long-term scale is in a coupled state, the long-term hydro-wind-solar complementary scheduling model is invoked; when the medium-term scale is in a coupled state, the medium-term hydro-wind-solar complementary scheduling model is invoked; when the short-term scale is in a coupled state, the short-term hydro-wind-solar-storage complementary scheduling model is invoked. When the corresponding scheduling time scale is in a forced coupling state, the water-wind-solar multi-energy complementary scheduling model or the water-wind-solar-storage multi-energy complementary scheduling model that matches the scheduling time scale is directly invoked. When the corresponding scheduling time scale is in a critical coupling state, the hydropower deterministic scheduling model or the corresponding multi-energy complementary scheduling model is selected based on the results of the operational risk analysis. The scheduling model outputs hydropower output plan, wind power output plan, photovoltaic output plan, reservoir water level process, outflow, water abandonment flow, power of transmission channels, and energy storage status of charge. The system obtains the actual output of new energy sources, actual hydropower output, actual reservoir water level, and actual power of the transmission channels after the scheduling is executed, and updates the historical operation samples and corresponding tolerance thresholds based on the actual operation data.

[0019] Example 2: S1. Select hydropower stations A, B, and C in a certain basin-level integrated hydro-wind-solar base as scheduling objects, and obtain the unit installed capacity of new energy, the maximum power generation capacity of hydropower, the grid-connected capacity of new energy, and the capacity of the transmission channel during the wet season from 2017 to 2023. S2. Classify the operating data from 2017 to 2023 according to hydropower stations and transmission channels, calculate the maximum installed capacity utilization hours of wind power and photovoltaic power during the wet season, and use the 80th percentile value to calibrate the long-term scale tolerance threshold. Table 1 Tolerance thresholds for long-term scale indicators

[0020]

[0021] S3. Obtain the maximum installed capacity utilization hours of wind power and photovoltaic power in the high-water month of 2023, the grid-connected capacity of new energy, the maximum power generation capacity of hydropower in the high-water month, and the transmission channel capacity in the high-water month for each hydropower station, and calculate the long-term scale decoupling discrimination index. Table 2 Long-term decoupling criteria Calculated value

[0022] The long-term calculated values ​​for hydropower stations A, B, and C are 0.688, 0.96, and 0.86, respectively, none of which exceed the corresponding calibration values ​​of 0.71, 1.13, and 0.88. Therefore, it is determined that the long-term scale is in a decoupled state in 2023, and the long-term deterministic scheduling model of hydropower under the condition of prioritizing the transmission of new energy is invoked. S4. Obtain daily new energy output data, average hydropower water consumption rate and reservoir regulation boundary in 2023, and use daily data in 2023 to calibrate the upper limit of medium-term water volume change; Table 3. Upper limit calibration data for medium-scale water quantity changes.

[0023] S5. Select August 16 to August 30, 2023, July 15 to July 29, 2023, June 16 to June 30, 2023, and January 1 to January 15, 2023 as typical scenarios, and obtain the maximum and minimum power output rates of wind power and photovoltaic power in each scenario. Table 4. Maximum and minimum power output rates of new energy sources in typical scenarios.

[0024] S6. Calculate the water demand changes of each hydropower station based on the new energy output rate, average daily output and hydropower station water consumption rate in typical scenarios. Table 5. Calculation data for the upper limit of water volume change in typical scenarios.

[0025] The calculated maximum water volume changes for hydropower stations A, B, and C are 47,215,400 m³, 47,010,600 m³, and 32,310,100 m³, respectively. None of these exceed the corresponding upper limits of water volume changes of 53,479,500 m³, 62,430,100 m³, and 51,000,200 m³. Therefore, it is determined that the mid-term scale in 2023 is in a decoupled state, and the mid-term deterministic scheduling model for hydropower under the condition of prioritizing the transmission of new energy is invoked.

[0026] Example 3: S1. Obtain the hourly hydropower output, wind power output simultaneity rate, and photovoltaic output simultaneity rate of hydropower stations A, B, and C in 2023, and select January 1, 2023 as a typical short-term scenario. Table 6 Input data for the short-term scale decoupling judgment formula

[0027] S2. Based on daily data from 2023, calibrate the tolerance thresholds corresponding to the power generation reliability indicators. Tolerance thresholds corresponding to peak shaving impact indicators And calculate the corresponding indicator values ​​based on the operating data as of January 1, 2023; Table 7. Calculation Output Data of Short-Term Scale Decoupling Judgment Formula

[0028] S3. Compare the short-term scale calculation values ​​of each hydropower station with the corresponding tolerance thresholds.

[0029] Hydropower Station A The calculated value of 0.67 is less than the calibrated value of 0.86. The calculated value of 0.11 is less than the calibrated value of 0.14; Hydropower Station B The calculated value of 0.31 is less than the calibrated value of 1.33. The calculated value of 0.07 is less than the calibrated value of 0.24; Hydropower Station C The calculated value of 0.39 is less than the calibrated value of 0.69. The calculated value of 0.02 is less than the calibrated value of 0.12; The power generation reliability index and peak-shaving impact index of the three hydropower stations did not exceed the corresponding tolerance threshold. Therefore, it was determined that the short-term scale was in a decoupled state on January 1, 2023, and the short-term deterministic scheduling model of hydropower under the condition of priority transmission of new energy was invoked. S4. When the number of scheduling periods with a continuous short-term comprehensive coupling degree not less than the second mode switching threshold reaches the first preset number, the short-term scheduling mode is switched from decoupled to coupled, and the short-term water, wind, solar and storage multi-energy complementary scheduling model is called. When the number of scheduling periods with a short-term comprehensive coupling degree not exceeding the first mode switching threshold reaches the second preset number, the short-term scheduling mode will be switched from the coupled state to the decoupled state, and the hydropower short-term deterministic scheduling model will be restored. S5. Execute the selected short-term scheduling model and output the hydropower output plan, wind power output plan, photovoltaic power output plan, reservoir water level process, outflow, water abandonment flow, power of transmission channels and energy storage status of charge. The system obtains the actual output of new energy sources, actual hydropower output, actual reservoir water level, and actual power of the transmission channels after the scheduling is executed, and updates the historical operation samples and corresponding tolerance thresholds based on the actual operation data.

Claims

1. A method for determining the scheduling mode of a watershed-level integrated water, wind, and solar power base, characterized in that, Includes the following steps: S1. Obtain the operation data of hydropower station groups, new energy operation data, external transmission channel operation data, energy storage operation data and scheduling business parameters of the basin-level integrated water, wind and solar power base, perform time alignment and anomaly processing on the acquired data, and form a scheduling mode discrimination dataset; S2. Based on the scheduling cycle and scheduling step size, the scheduling mode discrimination is divided into long-term scale discrimination, medium-term scale discrimination and short-term scale discrimination; S3. Construct an uncoupled reference operating state when hydropower-wind-solar complementary scheduling is not carried out, in accordance with the principle of prioritizing the transmission or consumption of new energy and the execution of deterministic scheduling plans for hydropower station groups. S4. Based on the uncoupled reference operating state, calculate the long-term decoupling discrimination index, the medium-term decoupling discrimination index, and the short-term decoupling discrimination index respectively. Long-term decoupling criteria are used to characterize the pressure on transmission channels under extreme power output superposition of hydropower and new energy. Medium-term decoupling criteria are used to characterize the reservoir regulation demand required for hydropower to compensate for continuous fluctuations in new energy. Short-term decoupling criteria are used to characterize the impact of new energy prediction errors and power output fluctuations on the reliability and peak-shaving capacity of hydropower generation. S5. Based on historical operation samples, typical scenario samples, planned grid-connected capacity and operation safety boundaries, determine the tolerance thresholds corresponding to each decoupling discrimination index. S6. Compare each decoupling discrimination index with the corresponding tolerance threshold to obtain the normalized limit violation coefficient corresponding to each scheduling time scale, and determine the corresponding comprehensive coupling degree based on the normalized limit violation coefficient. S7. When the overall coupling degree is not greater than the first mode switching threshold, the corresponding scheduling time scale is determined to be in a decoupled state; when the overall coupling degree is not less than the second mode switching threshold, the corresponding scheduling time scale is determined to be in a coupled state; when the overall coupling degree is between the first mode switching threshold and the second mode switching threshold, the corresponding scheduling time scale is determined to be in a critically coupled state. The first mode switching threshold and the second mode switching threshold are determined based on historical operation samples, scheduling risk analysis results, and mode switching requirements. The first mode switching threshold is less than the second mode switching threshold, and the second mode switching threshold is not less than 1. S8. For the decoupled state, call the hydropower deterministic scheduling model under the condition of priority transmission or priority consumption of new energy; for the coupled state, call the hydro-wind-solar multi-energy complementary scheduling model or hydro-wind-solar-storage multi-energy complementary scheduling model that matches the corresponding scheduling time scale; for the critical coupled state, select the scheduling model after performing the operation risk analysis.

2. The scheduling mode discrimination method for watershed-level integrated water, wind, and solar power bases according to claim 1, characterized in that, Step S1 includes: S11. Obtain the runoff, reservoir water level, available storage capacity, water consumption rate, planned output, minimum output and maximum output of the hydropower station group; S12. Obtain historical power output, predicted power output, grid-connected installed capacity, extreme power output rate, prediction error, and simultaneous power output rate of wind farms and photovoltaic power plants. S13. Obtain the transmission capacity, planned transmission power, and safe transmission limits of the external transmission channel; S14. Obtain the rated power, rated capacity, state of charge, and charging / discharging power of the energy storage system; S15. Perform time-scale unification, abnormal data identification, and missing data repair on the acquired data.

3. The scheduling mode discrimination method for watershed-level integrated water, wind, and solar power bases according to claim 1, characterized in that, Step S2 includes: Long-term timeframes correspond to monthly, ten-day, or high-water season scheduling. Medium-term timeframes correspond to weekly, ten-day, or consecutive days of scheduling. Short-term scales correspond to intraday, hourly, or 15-minute scheduling.

4. The scheduling mode discrimination method for watershed-level integrated water, wind, and solar power bases according to claim 1, characterized in that, The long-term decoupling index is calculated according to the following formula: ； In the formula, This is the long-term decoupling criterion for the transmission channel i. This represents the maximum power generation capacity of the hydropower station group corresponding to transmission channel i during the high-water season. This represents the calculated maximum output of new energy sources corresponding to transmission channel i during the high-water season. The available transport capacity of the external transport channel i during the high-water season; The maximum value among the long-term scale decoupling discrimination indicators corresponding to each external transmission channel is determined as the long-term scale discrimination indicator for the watershed-level integrated water, wind, and solar power base.

5. The scheduling mode discrimination method for watershed-level integrated water, wind, and solar power bases according to claim 1, characterized in that, The calculation of the mid-scale decoupling discriminant index includes: S51. Calculate the water storage regulation demand corresponding to the hydropower station h under the condition of continuous large-scale renewable energy generation according to the following formula: ； S52. Calculate the drawdown regulation demand corresponding to the hydropower station h under the condition of continuous low renewable energy generation according to the following formula: ； In the formula, , For the set of intermediate scheduling periods, and These represent the extreme maximum and extreme minimum power output rates of new energy sources, respectively. For new energy installed capacity, Contribute to new energy forecasting For the water consumption rate of the hydropower station, For scheduling step size; S53. Calculate the mid-scale decoupling index corresponding to hydropower station h according to the following formula: ； In the formula, For hydroelectric power station The reservoir's storage capacity For hydroelectric power station The capacity of the reservoir can be reduced; The maximum value among the mid-scale decoupling discriminant indices corresponding to each hydropower station is determined as the mid-scale discriminant index.

6. The scheduling mode discrimination method for watershed-level integrated water, wind, and solar power bases according to claim 1, characterized in that, Short-term decoupling criteria include power generation reliability indicators and peak-shaving impact indicators; The power generation reliability index is calculated according to the following formula: ； The peak-shaving impact index is calculated using the following formula: ； In the formula, The remaining prediction error of new energy sources after deducting the available compensation capacity of energy storage is included. For hydroelectric power station The minimum stable output, This is to account for the remaining peak-shaving demand of new energy sources after deducting the energy storage regulation capacity. and Hydropower stations The available upward adjustment capability and the available downward adjustment capability; The maximum value among the power generation reliability index and the peak-shaving impact index is determined as the short-term scale decoupling discrimination index corresponding to hydropower station h, and the maximum value among the short-term scale decoupling discrimination indices corresponding to each hydropower station is determined as the short-term scale discrimination index of the basin-level integrated hydro-wind-solar base.

7. The scheduling mode discrimination method for watershed-level integrated water, wind, and solar power bases according to claim 1, characterized in that, Step S5 includes: S71. Classify historical operation samples according to hydropower station, external transmission channel, new energy transmission area, hydrological period and scheduling season; S72. Calculate the long-term scale decoupling discrimination index, medium-term scale decoupling discrimination index and short-term scale decoupling discrimination index corresponding to various historical operation samples; S73. Remove abnormal samples corresponding to missing data, equipment malfunctions, extreme maintenance, and abnormal power outages. S74. Use the quantile method to determine the initial tolerance threshold corresponding to each decoupling discrimination index; S75. Based on the safe transport limit of the external transmission channel, the safe water level of the reservoir, the adjustable capacity of the reservoir, the minimum output and peak-shaving capacity of the hydropower unit, the initial tolerance threshold is corrected to obtain the final tolerance threshold. S76. The tolerance threshold shall be updated on a rolling basis according to the changes in the installed capacity of new energy, the capacity of external transmission channels and the operating boundary of reservoirs.

8. The scheduling mode discrimination method for watershed-level integrated water, wind, and solar power bases according to claim 1, characterized in that, The overall coupling degree is calculated according to the following formula: ； ； In the formula, Let j be the normalized limit-crossing coefficient corresponding to the decoupling discrimination index j under the scheduling time scale q. To decouple the discrimination index, For the corresponding tolerance threshold, Let q be the overall coupling degree corresponding to the scheduling time scale. As the indicator weight, The maximum impact coefficient of the out-of-limit indicator. The number of decoupling criteria corresponding to the scheduling time scale q; ,and ; ; .

9. The scheduling mode discrimination method for watershed-level integrated water, wind, and solar power bases according to claim 1, characterized in that, Step S7 also includes: S91. When any decoupling judgment index reaches the corresponding safety limit threshold, the corresponding scheduling time scale is directly judged as a forced coupling state. S92. When the number of scheduling periods with a comprehensive coupling degree that is not less than the second mode switching threshold reaches the first preset number, the scheduling mode will be switched from decoupling to coupling. S93. When the number of scheduling periods with a comprehensive coupling degree not exceeding the first mode switching threshold reaches the second preset number, the scheduling mode will be switched from the coupled state to the decoupled state.

10. The scheduling mode discrimination method for watershed-level integrated water, wind, and solar power bases according to claim 1, characterized in that, Also includes: S101. Based on the scheduling time scale at which the coupling state is reached, select a long-term water-wind-solar complementary scheduling model, a medium-term water-wind-solar complementary scheduling model, or a short-term water-wind-solar-storage complementary scheduling model. S102. Output the hydropower output plan, wind power output plan, photovoltaic output plan, reservoir water level process, outflow, water abandonment flow, power of transmission channels and energy storage status corresponding to the selected scheduling model; S103. Obtain the actual new energy output, actual hydropower output, actual reservoir water level, and actual power of the external transmission channel after the scheduling is executed; S104. Update historical operating samples and corresponding tolerance thresholds based on actual operating data.

Images