A pumped storage power station regulation method, device, equipment and storage medium

Abstract

The present application relates to the technical field of power station regulation, and discloses a pumped storage power station regulation method, device, equipment and storage medium, wherein, in the method, firstly, a long-period electricity price prediction model is used to obtain a basic electricity price trend of a prediction time period; secondly, a short-term correction model is used to introduce the currently collected electricity load, new energy output, inter-provincial tie-line plan and meteorological data to correct the long-period electricity price result, so as to obtain a spot electricity price prediction curve, so that the electricity price prediction result is more in line with the actual operation of the power market; then, based on the spot electricity price prediction curve, a peak-valley interval screening method is used to accurately identify the low-valley period and the peak period, so as to provide clear time boundaries for the pumped storage power station; finally, in combination with the physical operation constraints of the pumped storage power station, an optimal decision method is used to determine the pumping and power generation time window, so as to ensure that the pumped storage power station can quickly respond to the regulation requirements under the electricity price fluctuation scenario in the actual operation.

Description

Technical Field

[0001] This invention relates to the field of power plant control technology, specifically to a control method, device, equipment, and storage medium for a pumped storage power plant. Background Technology

[0002] Pumped storage power stations, as important peak-shaving and energy storage devices in the power system, derive their profits primarily from the peak-valley price difference in electricity prices. This means they consume electricity during off-peak hours to pump and store hydropower, and release the water to generate electricity during peak hours, profiting from the price difference. However, with the full-scale development of the electricity spot market, electricity prices are no longer fixed day-and-night time-of-use pricing, but rather exhibit characteristics of multiple peaks and valleys within a day, fluctuating in real time. Furthermore, they are influenced by various factors such as changes in grid load, disturbances in new energy output, weather conditions, and holiday characteristics, resulting in complex and ever-changing electricity price patterns.

[0003] The pumped storage power station control methods disclosed in related technologies rely on dispatchers to judge the peak and valley of electricity prices based on experience and formulate pumping and releasing plans. However, in actual operation, the efficiency of dispatchers making manual decisions is low and there is a significant lag, which makes it difficult to meet the rapid response requirements of current pumped storage power stations in the context of electricity price fluctuations. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, and storage medium for regulating pumped storage power stations, in order to solve the problem that the regulation methods for pumped storage power stations in related technologies are difficult to adapt to the rapid response requirements of current pumped storage power stations in the context of electricity price fluctuations.

[0005] In a first aspect, the present invention provides a method for regulating a pumped storage power station, the method comprising: Based on historical electricity supply and demand data and factors affecting electricity prices, a long-term electricity price forecast model is used to obtain long-term electricity price forecast results for the forecast period. The long-term electricity price forecast results include monthly electricity price forecast curves, electricity price trough forecast thresholds, and electricity price peak forecast thresholds. Based on the long-term electricity price forecast results for the forecast period, combined with the currently collected electricity load data, new energy output data, inter-provincial transmission line planning data and meteorological data, a short-term correction model is used for adjustment to obtain the spot electricity price forecast curve for the forecast period. Based on the spot electricity price forecast curve for the forecast period, and combined with the electricity price trough forecast threshold and electricity price peak forecast threshold for the forecast period, the peak-valley interval screening method is used to identify multiple trough periods and multiple peak periods for the forecast period. Based on multiple low-peak periods and multiple peak periods within the predicted time period, and combined with the physical operation constraints of the pumped storage power station, the optimal decision-making method is used to determine the pumping time window and power generation time window within the predicted time period.

[0006] The above implementation method first utilizes a long-term electricity price forecasting model to obtain the basic electricity price trend for the forecast period, providing a decision-making basis for subsequent regulation of pumped storage power stations. Second, a short-term correction model incorporates currently collected data on electricity load, renewable energy output, provincial interconnection plans, and meteorological data to correct the long-term electricity price results, yielding a spot price forecast curve that better reflects the actual operation of the electricity market. Then, based on the spot price forecast curve, a peak-valley interval screening method accurately identifies off-peak and peak periods, providing clear time boundaries for pumping and power generation at pumped storage power stations. Finally, considering the physical operational constraints of pumped storage power stations, an optimal decision-making method is used to determine the pumping and power generation time windows, ensuring that pumped storage power stations can quickly respond to regulation needs under fluctuating electricity prices during actual operation.

[0007] In one optional implementation, the spot electricity price forecast curve based on the forecast time period, combined with the electricity price trough forecast threshold and the electricity price peak forecast threshold for the forecast time period, is identified using a peak-valley interval filtering method to obtain multiple trough periods and multiple peak periods for the forecast time period, including: Based on the predicted low-price and high-price thresholds for the predicted time period, and combined with historical electricity price peak-valley characteristics, current load data, and renewable energy output data, a dynamic threshold evaluation method is used to obtain the low-price and high-price thresholds for the preset time period. Based on the spot electricity price forecast curve for the forecast period, and combined with the electricity price trough threshold and peak threshold for the preset period, the effective peak-valley interval screening method is used to remove instantaneous intervals with a duration lower than the preset threshold, thereby obtaining multiple trough periods and multiple peak periods for the forecast period.

[0008] Through the above implementation methods, based on the long-term electricity price forecast results, historical electricity price peak and valley characteristics, current load data, and renewable energy output data are further integrated. A dynamic threshold evaluation method is used to determine the electricity price low and high thresholds, ensuring that the criteria for determining peak and valley periods are adjusted according to market supply and demand and renewable energy output fluctuations. This improves the environmental adaptability and accuracy of peak and valley period identification. Subsequently, an effective peak and valley interval screening method is used to remove instantaneous electricity price peaks or valleys with a duration lower than the preset threshold, avoiding misjudgments caused by instantaneous electricity price fluctuations. This ensures that the identified low and high periods are practically operable, guaranteeing that the final low and high periods reflect both the macroeconomic laws of electricity price trends and the engineering feasibility of period continuity. This provides clear and reliable time boundaries for the subsequent pumping and power generation time windows of pumped storage power stations.

[0009] In one optional implementation, the determination of the pumping time window and power generation time window within the predicted time period, based on multiple low-peak periods and multiple peak periods within the predicted time period, combined with the physical operational constraints of the pumped storage power station, and using an optimal decision-making method, includes: Based on multiple low periods in the predicted time period, the optimal pumping time window is generated by using the pumping timing optimization method in the optimal decision-making method. Combined with the start-up response time of the pumping unit of the pumped storage power station, the optimal pumping time window is obtained. Based on multiple peak periods of the predicted time period, the optimal power generation time window is generated by using the power generation timing optimization method in the optimal decision-making method. Combined with the start-up response time of the generator units of the pumped storage power station, the optimal power generation time window is obtained. Based on the optimal pumping time window and the optimal power generation time window, and combined with the physical operation constraints of the pumped storage power station, the constraint verification method is used to verify the pumping time window and the power generation time window after verification, and output them as the pumping time window and the power generation time window for the predicted time period.

[0010] Through the above implementation methods, the pumping timing optimization method is used to screen candidate pumping time windows from multiple off-peak periods, and the actual start-up response time of the pumping unit of the pumped storage power station is used for correction. This ensures that the generated pumping window meets both economic requirements and the physical response capability of the equipment, avoiding execution deviations caused by start-up delays. Similarly, the power generation timing optimization method is used to determine the power generation window in combination with the start-up response time of the generator set, ensuring that the power generation command can be executed on time. Finally, the optimal pumping time window and the optimal power generation time window are physically constrained and verified through a constraint verification method to ensure that the final output pumping time window and power generation time window meet the constraints of equipment response physicality and system operation safety in actual operation.

[0011] In one optional implementation, the physical operating constraints of the pumped storage power station include the maximum daily start-up and shutdown frequency of the pumped storage units, the minimum interval for switching operating conditions, the lower limit of the effective pumping duration, and the lower limit of the effective power generation duration. The step of verifying the optimal pumping time window and the optimal power generation time window using a constraint verification method, combined with the physical operating constraints of the pumped storage power station, to obtain the verified pumping time window and power generation time window, includes: Based on the optimal pumping time window and the optimal power generation time window, the corresponding duration is obtained respectively, and the duration is compared with the lower limit of the effective duration of pumping or the lower limit of the effective duration of power generation using the duration verification method, so as to obtain the pumping duration evaluation result and the power generation duration evaluation result. Based on the number of pumping and power generation times during the predicted time period, and combined with the maximum number of daily start-ups and shutdowns of the pumped storage power station, the maximum number of start-ups and shutdowns method is used for evaluation and verification to obtain the start-up and shutdown count evaluation result. Based on the difference between the end time of the adjacent optimal pumping time window and the start time of the optimal power generation time window, the switching interval is compared using the operating condition switching interval verification rule to obtain the switching interval evaluation result. Based on the combined evaluation results of pumping time, power generation time, start-stop frequency, and switching interval, the optimal pumping time window and optimal power generation time window that meet the preset conditions are output and used as the pumping time window and power generation time window after verification.

[0012] The above implementation method first obtains the duration of the optimal pumping time window and the optimal power generation time window, respectively, and compares them with the lower limits of the effective duration of pumping and power generation to ensure that each time window has sufficient physical duration to complete effective pumping or power generation operations. Then, based on the number of pumping and power generation operations within the predicted time period, and combined with the maximum number of daily unit start-ups and shutdowns, verification is performed to reduce the risk of unit lifespan loss due to excessively frequent start-ups and shutdowns within a day. Simultaneously, based on the difference between the end time of adjacent optimal pumping time windows and the start time of the optimal power generation time window, a comparison is made using operating condition switching interval verification rules to ensure sufficient interval time for physical recovery and safety between two adjacent operating condition switches. Finally, by combining the duration evaluation results, start-up / shutdown frequency evaluation results, and switching interval evaluation results, the optimal pumping time window and the optimal power generation time window that simultaneously satisfy all preset physical constraints are output. This ensures that the final control method, while pursuing economic optimization, fully respects the physical limits and operational safety requirements of the unit equipment, effectively avoiding the problem of theoretical optimal solutions being out of sync with engineering reality.

[0013] In an optional embodiment, when the duration of the pumping time window is less than the lower limit of the effective pumping duration or the duration of the power generation time window is less than the lower limit of the effective power generation duration, the method further includes: Based on the predicted time period, the pumping time window for the flat valley period is extracted using the pumping window supplementation method, and the supplemented pumping time window is obtained. Alternatively, based on the second-peak period in the predicted time period, the power generation time window of the second-peak period can be extracted using the power generation window supplementation method to obtain the supplemented power generation time window.

[0014] Through the above implementation methods, when the duration of the pumping time window is less than the lower limit of the effective pumping duration, a pumping time window supplementation method is used to extract available pumping time windows from the valley period of the predicted time period to supplement the original pumping time window; when the duration of the power generation time window is less than the lower limit of the effective power generation duration, a power generation time window supplementation method is used to extract available power generation time windows from the second peak period of the predicted time period to supplement the original power generation window, thereby avoiding the abandonment of the period due to the window being too short, and facilitating the extension of the working time to ensure the scheduling flexibility of the power plant when the pumping time window or the power generation time window is insufficient.

[0015] In one optional implementation, the long-term electricity price forecast result based on the forecast period is combined with currently collected electricity load data, renewable energy output data, inter-provincial transmission line planning data, and meteorological data, and adjusted using a short-term correction model to obtain a spot electricity price forecast curve for the forecast period, including: Based on the currently collected electricity load data, new energy output data, inter-provincial transmission line planning data, and meteorological data, the timestamp alignment method is used to synchronize the data at a unified time frequency to obtain synchronized multi-dimensional feature data. Based on the synchronized multidimensional feature data, the impact degree assessment method is used to analyze the impact degree of electricity load data, new energy output data, inter-provincial connection line planning data and meteorological data on electricity price data, and the deviation analysis results for each influencing factor are obtained. Based on the deviation analysis results of various influencing factors, the electricity price curve and peak-valley threshold in the long-term electricity price forecast results are adjusted using the deviation correction method, and the output is the spot electricity price forecast curve for the forecast period.

[0016] Through the above implementation method, based on long-term electricity price forecasting, multi-dimensional characteristic data such as currently collected load, renewable energy output, inter-provincial transmission line planning data, and meteorological data are introduced. A timestamp alignment method is used to unify the data sources to the same time frequency, eliminating the problem of inconsistent time bases for multi-source data and avoiding correction deviations caused by data misalignment. Then, an impact assessment method is used to analyze the degree of influence of electricity load data, renewable energy output data, inter-provincial transmission line planning data, and meteorological data on electricity price data, obtaining deviation analysis results for each influencing factor. This allows the short-term correction model to quantitatively identify the differentiated impact of different factors on electricity prices, rather than making equal corrections to all factors, thus improving the targeting and accuracy of the correction. Finally, based on the deviation analysis results of each influencing factor, a deviation correction method is used to adjust the electricity price curve and peak-valley threshold in the long-term electricity price forecast results, outputting a spot electricity price forecast curve. This allows the forecast curve to maintain the long-term trend framework while reflecting the short-term impact of real-time supply and demand changes in the spot market and renewable energy output fluctuations on electricity prices, making the identification of peak-valley periods more consistent with the actual operating state of the electricity market.

[0017] In one alternative implementation, it further includes: Based on the actual electricity price data and unit operation data of pumped storage power stations after the pumping time window and power generation time window are executed, combined with the long-term electricity price prediction results and spot electricity price prediction curve, the long-term electricity price prediction model, short-term correction model, peak-valley interval screening method and optimal decision method are optimized using iterative algorithms.

[0018] Through the above implementation methods, the sources of prediction deviations are identified using the actual operational data after execution, and the model parameters, threshold settings, and decision rules are iteratively adjusted accordingly. This allows the algorithm models and parameter settings of each link to evolve dynamically with the actual operational data, continuously adapting to the local power market characteristics and the actual response characteristics of the generating units, and avoiding the gradual decay of the control effect due to the model becoming out of touch with market changes due to fixed parameters.

[0019] In a second aspect, the present invention provides a control device for a pumped storage power station, the device comprising: The electricity price forecasting module is used to obtain long-term electricity price forecast results for the forecast period based on historical electricity supply and demand data and factors affecting electricity prices, using a long-term electricity price forecasting model. The long-term electricity price forecast results include monthly electricity price curves, electricity price trough forecast thresholds, and electricity price peak forecast thresholds. The electricity price correction module is used to adjust the long-term electricity price forecast results based on the forecast period, combined with the currently collected electricity load data, new energy output data, inter-provincial interconnection line plan data and meteorological data, using a short-term correction model to obtain the spot electricity price forecast curve for the forecast period. The peak-valley identification module is used to identify multiple low-valley periods and multiple peak periods based on the spot electricity price prediction curve for the prediction period, combined with the low-valley prediction threshold and the high-valley prediction threshold for the prediction period. The decision generation module is used to determine the pumping time window and power generation time window within the predicted time period based on multiple low-peak periods and multiple peak periods, combined with the physical operation constraints of the pumped storage power station, using the optimal decision method.

[0020] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the pumped storage power station control method of the first aspect or any corresponding embodiment described above.

[0021] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the pumped storage power station control method of the first aspect or any corresponding embodiment described above. Attached Figure Description

[0022] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the first type of pumped storage power station control method according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the second process of the pumped storage power station control method according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the third process of the pumped storage power station control method according to an embodiment of the present invention. Figure 4 This is a schematic diagram of the fourth process of the pumped storage power station control method according to an embodiment of the present invention. Figure 5 This is a schematic diagram of the fifth process of the pumped storage power station control method according to an embodiment of the present invention; Figure 6 This is a structural block diagram of a pumped storage power station control device according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0026] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0027] In the pumped storage power station control methods disclosed in related technologies, the identification of peak and valley periods and the decision-making on the timing of pumping and releasing water mainly rely on the experience of dispatchers based on historical load curves and electricity price change patterns, combined with the daily power grid operation status, weather forecasts, and holiday attributes. They subjectively judge the peak and valley periods that may occur during the day through manual analysis, and then manually compile pumping and power generation plans.

[0028] However, the pumped storage power station control methods disclosed in related technologies have at least the following technical defects: 1. The formation of peak and valley electricity prices is affected by a combination of factors, including real-time supply and demand in the electricity spot market, fluctuations in renewable energy output, adjustments to tie-line power, and changes in primary energy prices. Its changing pattern is complex and nonlinear. It is difficult to accurately predict the intraday electricity price inflection point by relying solely on historical experience. This can easily lead to situations where pumping is started only after the low-price period has passed, or grid connection and power generation are only started when the high-price period is about to end. 2. Differences in the experience and judgment logic of different dispatchers lead to inconsistent decision-making quality, resulting in unstable returns from the final pumped storage power station control method. 3. The fluctuation cycle of spot market electricity prices has been shortened to the minute level. It is impossible for humans to continuously track and respond quickly to such high-frequency changes. At the same time, manual scheduling plans require repeated verification of physical constraints such as the number of unit start-ups and shutdowns, the interval between operating conditions, and the duration of continuous operation. The decision-making process is time-consuming and inefficient, and it is difficult to quickly re-optimize the plan when encountering sudden grid situations or sharp fluctuations in electricity prices.

[0029] In summary, to address the shortcomings of the pumped storage power station control methods disclosed in the aforementioned related technologies, which are unable to meet the rapid response requirements under electricity price fluctuation scenarios in the electricity spot market environment, this invention provides a pumped storage power station control method. First, a long-cycle electricity price forecasting model is used to obtain the basic electricity price trend for the forecast period, providing a decision-making basis for subsequent pumped storage power station control. Second, a short-term correction model is used to incorporate currently collected electricity load, renewable energy output, inter-provincial transmission line plans, and meteorological data to correct the long-cycle electricity price results, obtaining a spot price forecast curve, making the electricity price forecast results more closely reflect the actual operation of the electricity market. Then, based on the spot price forecast curve, a peak-valley interval screening method is used to accurately identify low-valley and peak periods, providing clear time boundaries for the pumping and power generation timing of the pumped storage power station. Finally, combined with the physical operating constraints of the pumped storage power station, an optimal decision-making method is used to determine the pumping and power generation time windows, ensuring that the pumped storage power station can quickly respond to the control needs under electricity price fluctuation scenarios in actual operation.

[0030] According to an embodiment of the present invention, a method for regulating a pumped storage power station is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0031] This embodiment provides a control method for a pumped storage power station, which can be used in the control terminal of a pumped storage power station. Figure 1 This is a flowchart of a pumped storage power station control method according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps: S101, based on historical electricity supply and demand data and factors influencing electricity prices, uses a long-term electricity price forecasting model to obtain long-term electricity price forecast results for the forecast period. The long-term electricity price forecast results include monthly curves, electricity price trough forecast thresholds, and electricity price peak forecast thresholds.

[0032] Historical power supply and demand data are used to reflect the recorded operating status of the power grid system in the area where the target pumped storage power station is located. This includes historical load curves, output data of various power sources, central dispatch loads, tie line plans, and output of regulating resources.

[0033] Electricity price influencing factors are multidimensional factors that affect the formation of electricity spot market prices, including power source cost-output characteristics, electricity demand, primary energy prices, GDP growth rate of the region where the target pumped storage power station is located, and electricity consumption per unit of GDP.

[0034] The long-term electricity price forecasting model is a standardized forecasting model built based on the supply and demand logic of power source cost-output characteristics and electricity demand, with the constraint of maximizing social welfare, and combined with the charging and discharging price execution rules of pumped storage. It is used for standardized forecasting of electricity demand, power source and regulation resources, primary energy prices, market boundaries and output space, marginal unit supply curves, and electricity price based on the unified electricity market clearing model.

[0035] The forecast period refers to a long-term time range within the next month, used to grasp the macro trend of electricity price movement.

[0036] The long-term electricity price forecast results are the predicted values ​​of key indicators such as future electricity price trends and peak-valley price differences output by the long-term electricity price forecast model, which are used as the basis for subsequent short-term adjustments.

[0037] By establishing a quantitative correlation between electricity demand, power generation capacity, primary energy prices and electricity prices, electricity price forecasting is upgraded from experience-based judgment to reliable forecasting based on supply and demand. This provides a stable and reliable forecasting basis for subsequent short-term adjustments, ensures that peak-valley identification has a long-term perspective, and avoids the defects of systemic bias caused by ignoring long-term electricity price trends.

[0038] S102. Based on the long-term electricity price forecast results for the forecast period, combined with the currently collected electricity load data, new energy output data, inter-provincial transmission line planning data and meteorological data, a short-term correction model is used to adjust the forecast and obtain the spot electricity price forecast curve for the forecast period.

[0039] Load data is short-term load forecast data provided by the power trading platform to reflect real-time changes in power demand.

[0040] The renewable energy output data refers to the day-ahead output forecast data for wind power and photovoltaic power provided by the power trading platform.

[0041] The inter-provincial interconnection line plan data is the day-ahead forecast data of power transmission and reception for inter-provincial lines provided by the power trading platform.

[0042] Meteorological data includes temperature, wind speed, light intensity, and rainfall data for the predicted operating day.

[0043] By processing the collected data according to 96 sampling times evenly distributed over a 24-hour period, a multidimensional feature dataset was constructed.

[0044] The short-term correction model is a model that dynamically adjusts the long-term forecast base based on currently collected data. It is used to compare the electricity price trend of the long-term forecast with the current data, and to focus on analyzing the impact of factors such as load fluctuations, changes in renewable energy output, and sudden weather changes on the spot electricity price, so as to dynamically correct the electricity price curve and peak-valley threshold.

[0045] The spot electricity price forecast curve is a daily continuous electricity price forecast curve output after being corrected by a short-term correction model. It is based on a 15-minute time frame and is used for subsequent daily peak-valley identification.

[0046] By collecting multi-source characteristic data such as current electricity load, renewable energy output, inter-provincial transmission line plans, and meteorological data, and making dynamic corrections by comparing with long-term baselines, the electricity price can be adjusted in a timely manner when the output of renewable energy such as wind power and photovoltaics is significantly higher than the predicted value, and the electricity price can be adjusted in a timely manner when the load increases. This upgrades the electricity price forecast from "static trend" to "dynamic tracking", effectively responding to sudden changes in renewable energy output, sudden weather changes, and other emergencies, and providing an accurate data foundation for peak-valley identification.

[0047] S103, based on the spot electricity price forecast curve for the forecast period, combined with the electricity price trough forecast threshold and electricity price peak forecast threshold for the forecast period, the peak-valley interval screening method is used to identify multiple trough periods and multiple peak periods for the forecast period.

[0048] The peak-valley interval screening method is a method that dynamically identifies peak and valley periods of electricity prices based on the spot electricity price forecast curve. Specifically, it combines factors such as the long-term forecast peak-valley threshold range, the historical peak-valley characteristics of the past 7 days, and the daily load and renewable energy output to dynamically adjust the daily valley threshold and peak threshold of the forecast period. At the same time, by utilizing the duration of the effective peak-valley interval, it eliminates instantaneous invalid peaks and valleys, thereby determining the valley and peak periods that can meet the operational requirements of pumped storage power stations.

[0049] Off-peak periods are continuous periods when electricity prices are below the dynamic off-peak threshold, and are used to guide pumped storage power stations in pumping storage operations.

[0050] Peak periods are the continuous periods when electricity prices are higher than the dynamic peak threshold, and are used to guide pumped storage power stations in generating electricity.

[0051] By utilizing a triple mechanism of dynamic threshold setting, historical calibration, and invalid interval removal, the electricity price forecast curve for each day is precisely segmented into five types of time periods: low valley, average valley, average segment, secondary peak, and peak. This facilitates the provision of refined time period boundaries and electricity price characteristic information for subsequent water pumping and releasing decisions of pumped storage power stations.

[0052] S104, based on multiple low-peak periods and multiple peak periods of the predicted time period, combined with the physical operation constraints of the pumped storage power station, uses the optimal decision-making method to determine the pumping time window and power generation time window within the predicted time period.

[0053] The physical operating constraints of pumped storage power stations are the hard conditions that the units must meet in actual operation, including the maximum number of start-ups and shutdowns per day, the minimum interval for switching operating conditions, the lower limit of the effective pumping duration, and the lower limit of the effective power generation duration.

[0054] The optimal decision-making method is to prioritize pumping water during the off-peak hours when the electricity price is lowest, the duration is longest, and the volatility is lowest, while satisfying physical constraints; or to prioritize generating electricity during the peak hours when the electricity price is highest, the duration is longest, and the volatility is lowest. At the same time, the unit should be started in advance in conjunction with the unit's start-up response time to ensure that full-power generation or full-power pumped storage is carried out precisely at the beginning of the peak-valley period.

[0055] The pumping time window corresponds to the start and end times of the target pumped storage power station as output by the optimal decision method.

[0056] The power generation time window corresponds to the start and end times of the target pumped storage power station's power generation, as output by the optimal decision-making method.

[0057] By using off-peak hours as pumping time windows and peak hours as power generation time windows, the system maximizes the daily peak-valley price difference to generate revenue. At the same time, it protects the equipment of the target pumped storage power station and extends its lifespan by limiting the number of start-stop operations, the interval between operating conditions, and the lower limit of the duration. The final output is a pumping and releasing strategy that can be directly used to guide the target pumped storage power station.

[0058] The pumped storage power station regulation method provided in this invention first utilizes a long-cycle electricity price forecasting model to obtain the basic electricity price trend for the forecast period, providing a decision-making basis for subsequent regulation of the pumped storage power station. Second, a short-term correction model is used to incorporate currently collected electricity load, renewable energy output, inter-provincial transmission line plans, and meteorological data to correct the long-cycle electricity price results, resulting in a spot price forecast curve that better reflects the actual operation of the electricity market. Then, based on the spot price forecast curve, a peak-valley interval screening method is used to accurately identify low-valley and peak periods, providing clear time boundaries for the pumping and power generation timing of the pumped storage power station. Finally, considering the physical operating constraints of the pumped storage power station, an optimal decision-making method is used to determine the pumping and power generation time windows, ensuring that the pumped storage power station can quickly respond to regulation needs under electricity price fluctuation scenarios during actual operation.

[0059] This embodiment provides a control method for a pumped storage power station, which can be used in the control terminal of a pumped storage power station. Figure 2This is a flowchart of a pumped storage power station control method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: S201, based on historical electricity supply and demand data and factors influencing electricity prices, utilizes a long-term electricity price forecasting model to obtain long-term electricity price forecasts for the forecast period. The long-term electricity price forecast results include monthly curves, electricity price trough forecast thresholds, and electricity price peak forecast thresholds. For details, please refer to [link to relevant documentation]. Figure 1 S101 of the illustrated embodiment will not be described again here.

[0060] S202, based on the long-term electricity price forecast results for the forecast period, combined with the currently collected electricity load data, new energy output data, inter-provincial transmission line planning data and meteorological data, uses a short-term correction model for adjustment to obtain the spot electricity price forecast curve for the forecast period.

[0061] Specifically, S202 above includes: S2021, based on the currently collected electricity load data, new energy output data, inter-provincial connection line plan data and meteorological data, uses the timestamp alignment method to synchronize according to a unified time frequency to obtain synchronized multi-dimensional feature data.

[0062] Electricity load data is the load data provided by the trading platform on the previous day. It corresponds to the total electricity demand of the power system at each time period during operation and is used to reflect the real-time supply and demand tension of the power system in the area where the target pumped storage power station is located.

[0063] New energy output data includes wind power output and photovoltaic output forecast data, which correspond to the actual power transmitted to the grid by wind farms and photovoltaic power plants on the day-ahead short-term forecast operating days. Fluctuations in new energy output data directly affect the grid's net load and the output space of marginal pricing units, thereby affecting spot electricity prices.

[0064] The inter-provincial interconnection plan is the time-of-day power transmission or reception plan for the region between provinces, which was recently disclosed by the trading platform. For provinces that transmit power, the inter-provincial interconnection plan is treated as the load within the province, and for provinces that receive power, the inter-provincial interconnection plan is treated as the power output within the province.

[0065] Meteorological data includes temperature, wind speed, light intensity, and rainfall data for the predicted operating day. Meteorological conditions directly affect load levels; for example, high temperatures lead to an increase in cooling load and also affect the output of new energy sources.

[0066] Timestamp alignment is a technique that synchronizes data from different data sources according to a unified time coordinate. This eliminates discrepancies between the time base and sampling time of data such as electricity prices, load, renewable energy output, and meteorology, which may originate from different acquisition systems. Timestamp alignment ensures the comparability of data at the same point in time, avoiding analytical distortion caused by time delays. In this embodiment, daily data is aligned according to 96 time points throughout the day, i.e., a sampling time is constructed every 15 minutes. If the collected data lacks temporal characteristics and is a single value, it is copied and filled into the 96 time points.

[0067] Synchronization at a uniform time frequency refers to reconstructing and interpolating multi-source data at the same time interval, so that the multi-source data have a consistent sampling frequency on the time axis, thereby unifying the original data from different data sources to the same time granularity, which facilitates subsequent joint analysis and comparative calculations.

[0068] Multidimensional feature data is a comprehensive dataset that includes multiple dimensions such as load, new energy output, inter-provincial connection line plans, and meteorology, after being aligned with timestamps and placed in the same time coordinate system. Each set of timestamps contains data records of all the above-mentioned dimensions at the corresponding time point, which are used to provide structured input for subsequent deviation analysis.

[0069] By unifying multi-source characteristic data such as user load, renewable energy output, inter-provincial transmission line plans, and meteorology to the same time frequency, the time misalignment between various data acquisition systems is eliminated, ensuring that subsequent deviation analysis has a correct time correspondence. This provides an accurate, comparable, and structured data foundation for subsequent short-term correction models, thereby guaranteeing the accuracy and reliability of spot electricity price forecast curve adjustments.

[0070] S2022, based on synchronized multidimensional feature data, uses the impact assessment method to analyze the impact of electricity load data, new energy output data, inter-provincial transmission line planning data and meteorological data on electricity price data, and obtains the deviation analysis results for each influencing factor.

[0071] The impact assessment method is a quantitative analysis tool for analyzing the impact of factors such as load, renewable energy output, and weather on electricity prices. This includes comparing the trends of electricity demand and market boundary forecast data output by long-term electricity price forecast models with the differences between the currently collected characteristic data and the trends of various factors. Alternatively, it involves quantifying the correlation between fluctuations in various factors and fluctuations in electricity prices, thereby identifying the main influencing factors and their direction of action.

[0072] The deviation analysis results are a quantitative assessment of the amount of deviation caused by each influencing factor on electricity prices, as well as the direction and intensity of its influence.

[0073] By assessing the actual impact and magnitude of factors such as load, renewable energy, and weather on electricity prices, subsequent adjustments are made based on data-driven corrections. This avoids the crude adjustment of simply distributing weights equally among all influencing factors. Instead, the results of deviation analysis focus on a few key factors that actually have a significant impact, providing a precise quantitative basis for subsequent dynamic adjustments.

[0074] S2023, based on the deviation analysis results of various influencing factors, uses the deviation correction method to adjust the electricity price curve and peak-valley threshold in the long-term electricity price forecast results, and outputs the spot electricity price forecast curve for the forecast period.

[0075] The deviation correction method is an operational approach that quantitatively compensates and adjusts long-term electricity price forecasts based on deviation analysis results. It can be implemented as follows: when the actual load exceeds the forecast value, the electricity price for the corresponding period is increased by a preset correction coefficient; when the output of new energy sources exceeds the forecast value, the electricity price for the corresponding period is decreased according to the output deviation ratio; simultaneously, meteorological factors are weighted for correction, transforming qualitative deviation conclusions into quantitative electricity price adjustments, thus facilitating the key processing stages connecting deviation analysis and correction outputs.

[0076] The electricity price curve is a continuous curve showing the change of spot electricity prices over time. It is used to reflect the temporal distribution of electricity prices in the spot market. By adjusting the electricity price values ​​at each time point on the curve through correction methods, the electricity price curve can be made closer to the actual market conditions.

[0077] Peak-valley thresholds are the critical values ​​used to determine the low and high electricity prices. They include a low-valley threshold to indicate periods of low electricity prices and a high-valley threshold to indicate periods of high electricity prices. The peak and low thresholds are dynamically adjusted upwards or downwards based on corrections to the spot electricity price curve to ensure the accuracy of peak-valley range division.

[0078] The spot electricity price forecast curve is the daily electricity price forecast curve output after short-term correction. It is the final result of superimposing the long-term forecast base and real-time deviation correction compensation, and is used as direct input data for subsequent peak-valley identification.

[0079] By incorporating multi-dimensional characteristic data such as current electricity load, renewable energy output, inter-provincial transmission line plans, and meteorology into long-term electricity price forecasts, and using timestamp alignment methods to unify the data sources to the same time frequency, the problem of inconsistent time bases for multi-source data is eliminated, avoiding correction biases caused by data misalignment. Then, an impact assessment method is used to analyze the degree of influence of load data, renewable energy output data, and meteorological data on electricity price data, obtaining deviation analysis results for each influencing factor. This allows the short-term correction model to quantitatively identify the differentiated impacts of different factors on electricity prices, rather than making equal corrections to all factors, thus improving the targeting and accuracy of the correction. Finally, based on the deviation analysis results of each influencing factor, the deviation correction method is used to adjust the electricity price curve and peak-valley threshold in the long-term electricity price forecast results, outputting a spot electricity price forecast curve. This ensures that the forecast curve maintains the long-term trend framework while reflecting the short-term impact of real-time supply and demand changes and renewable energy output fluctuations on electricity prices, making the identification of peak-valley periods more consistent with the actual operating state of the electricity market.

[0080] S203, based on the spot electricity price forecast curve for the forecast period, combines the low-price and high-price forecast thresholds for the forecast period, and uses a peak-valley interval filtering method to identify multiple low-price and high-price periods for the forecast period. For details, please refer to [link to details]. Figure 1 S103 of the illustrated embodiment will not be described again here.

[0081] S204, based on multiple low-peak and high-peak periods within the predicted timeframe, and considering the physical operational constraints of the pumped-storage power station, utilizes an optimal decision-making method to determine the pumping and power generation time windows within the predicted timeframe. For details, please refer to [link to relevant documentation]. Figure 1 S104 of the illustrated embodiment will not be described again here.

[0082] This embodiment provides a control method for a pumped storage power station, which can be used in the control terminal of a pumped storage power station. Figure 3 This is a flowchart of a pumped storage power station control method according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps: S301, based on historical electricity supply and demand data and factors influencing electricity prices, utilizes a long-term electricity price forecasting model to obtain long-term electricity price forecasts for the forecast period. The long-term electricity price forecast results include monthly electricity price curves, low-price forecast thresholds, and high-price forecast thresholds. For details, please refer to [link to relevant documentation]. Figure 1 S101 of the illustrated embodiment will not be described again here.

[0083] S302, based on the long-term electricity price forecast results for the forecast period, and combined with currently collected load data, renewable energy output data, inter-provincial transmission line planning data, and meteorological data, uses a short-term correction model for adjustments to obtain the spot electricity price forecast curve for the forecast period. For details, please refer to [link to details]. Figure 1 S101 of the illustrated embodiment will not be described again here.

[0084] S303, based on the spot electricity price forecast curve for the forecast period, combined with the electricity price trough forecast threshold and electricity price peak forecast threshold for the forecast period, uses the peak-valley interval screening method to identify multiple trough periods and multiple peak periods for the forecast period.

[0085] Specifically, S303 includes: S3031, based on the predicted low-price and high-price thresholds for the predicted time period, and combining historical electricity price peak-valley characteristics, current load data, and renewable energy output data, uses a dynamic threshold evaluation method to obtain the low-price and high-price thresholds for the preset time period.

[0086] Historical electricity price peak-valley characteristics are peak-valley feature information extracted from the historical time-of-use electricity prices of the past 7 days. This includes the average electricity price during peak hours, the average electricity price during off-peak hours, the peak-valley price difference, and the regularity of peak-valley occurrences. It is used to dynamically calibrate the threshold for the operating day, so that the threshold setting is consistent with the recent actual electricity price level, and avoids the threshold being too tight or too loose due to long-term prediction deviations.

[0087] The dynamic threshold evaluation method is an intelligent evaluation method that integrates multi-dimensional information such as long-term forecast base, historical and recent electricity price characteristics, current load and renewable energy output, and dynamically calculates the peak and valley thresholds for the day. It includes: using the long-term forecast threshold range as the basic framework, using the actual peak and valley electricity price levels of the past 7 days for upper and lower calibration, and then using the load level and renewable energy output of the day to finely adjust the thresholds, thereby achieving the goal of adaptively adjusting the thresholds according to changes in market environment and operating conditions.

[0088] The electricity price trough threshold is a critical value used to determine whether the electricity price is in a "trough". The continuous period of the electricity price curve that is lower than the electricity price trough threshold is identified as the trough period, which corresponds to the pumping time window of the pumped storage power station.

[0089] The peak electricity price threshold is a critical value used to determine whether the electricity price is at a "peak". The continuous period in the daily electricity price curve that is higher than the peak electricity price threshold is identified as the peak period, corresponding to the power generation time window of the pumped storage power station.

[0090] S3032, based on the spot electricity price prediction curve for the prediction period, combined with the electricity price trough threshold and peak threshold for the preset period, uses an effective peak-valley interval screening method to remove instantaneous intervals with a duration lower than the preset threshold, thus obtaining multiple trough periods and multiple peak periods for the prediction period.

[0091] The effective peak-valley interval screening method is a set of screening rules that determine the effectiveness of initially identified peak or valley periods after setting dynamic thresholds. By using the set duration standard of the effective peak-valley interval, instantaneous peak-valley intervals with insufficient duration are eliminated, thereby avoiding the probability of frequent switching of unit operating conditions due to short-term electricity price fluctuations and ensuring equipment safety.

[0092] Based on long-term electricity price forecasts, this method further integrates historical peak-valley characteristics, current load data, and renewable energy output data. A dynamic threshold evaluation method is used to determine peak and valley thresholds for electricity prices. This ensures that the criteria for determining peak-valley intervals are adjusted according to market supply and demand and fluctuations in renewable energy output, improving the environmental adaptability and accuracy of peak-valley period identification. Subsequently, an effective peak-valley interval screening method is used to eliminate instantaneous electricity price spikes or troughs with durations below the preset threshold, avoiding misjudgments caused by instantaneous price fluctuations. This ensures that the identified peak and valley periods are practically operable. The final peak and valley periods reflect both the macroeconomic trends of electricity prices and the engineering feasibility of period continuity, providing clear and reliable time boundaries for the subsequent pumping and power generation windows of pumped-storage power stations.

[0093] S304, based on multiple low-peak and high-peak periods within the predicted timeframe, and considering the physical operational constraints of the pumped-storage power station, utilizes an optimal decision-making method to determine the pumping and power generation time windows within the predicted timeframe. For details, please refer to [link to relevant documentation]. Figure 1 S104 of the illustrated embodiment will not be described again here.

[0094] This embodiment provides a control method for a pumped storage power station, which can be used in the control terminal of a pumped storage power station. Figure 4 This is a flowchart of a pumped storage power station control method according to an embodiment of the present invention, such as... Figure 4 As shown, the process includes the following steps: S401, based on historical electricity supply and demand data and factors influencing electricity prices, utilizes a long-term electricity price forecasting model to obtain long-term electricity price forecasts for the forecast period. These forecasts include monthly electricity price forecast curves, low-price forecast thresholds, and high-price forecast thresholds. For details, please refer to [link to relevant documentation]. Figure 1 S101 of the illustrated embodiment will not be described again here.

[0095] S402, based on long-term electricity price forecasts for the forecast period, and combined with currently collected electricity load data, renewable energy output data, inter-provincial transmission line planning data, and meteorological data, uses a short-term correction model for adjustments to obtain the spot electricity price forecast curve for the forecast period. For details, please refer to [link to details]. Figure 1 S101 of the illustrated embodiment will not be described again here.

[0096] S403, based on the spot electricity price forecast curve for the forecast period, combined with the electricity price trough forecast threshold and electricity price peak forecast threshold for the forecast period, uses the peak-valley interval screening method to identify multiple trough periods and multiple peak periods for the forecast period.

[0097] S404, based on multiple low-peak periods and multiple peak periods of the predicted time period, combined with the physical operation constraints of the pumped storage power station, uses the optimal decision-making method to determine the pumping time window and power generation time window within the predicted time period.

[0098] The physical operating constraints of pumped storage power stations include the maximum number of daily start-ups and shutdowns of pumped storage units, the minimum interval for switching operating conditions, the lower limit of the effective pumping duration, and the lower limit of the effective power generation duration.

[0099] Specifically, S404 includes: S4041, based on multiple low periods in the predicted time period, the optimal pumping time window is generated by using the pumping timing optimization method in the optimal decision-making method. Combined with the start-up response time of the pumping unit of the pumped storage power station, the optimal pumping time window is obtained.

[0100] The optimal pumping timing method is a candidate window selection method for pumping conditions. It prioritizes the period with the lowest average electricity price, the longest duration, and the smallest electricity price volatility among multiple off-peak periods as the optimal pumping time window.

[0101] The candidate pumping time window is one or more off-peak periods that meet the optimal electricity price conditions, which are initially screened based on the pumping timing optimization method. It still needs to be verified by subsequent start-up response time and physical constraints.

[0102] The start-up response time of a pumping unit refers to the time delay required for a pumped storage power station unit to pump water at full power from receiving the start-up command. This includes operational steps such as start-up preparation, power increase, and stable operation. By starting the pumping unit in advance before this time, it is ensured that pumping and storage can be carried out accurately at the beginning of the off-peak period.

[0103] The optimal pumping time window is the final start and end time of pumping operation determined after correction of the start-up response time. Its start time is the start time of the low-temperature period minus the start-up response time of the pumping unit, and its end time is aligned with the end time of the low-temperature period.

[0104] S4042: Based on multiple peak periods of the predicted time period, the optimal power generation time window is obtained by using the power generation timing optimization method in the optimal decision-making method. The optimal power generation time window is obtained by combining the start-up response time of the generator units of the pumped storage power station.

[0105] The optimal power generation timing method is a candidate window selection method for power generation conditions. It prioritizes the period with the highest average electricity price, the longest duration, and the lowest price volatility among multiple peak periods as the optimal power generation time window.

[0106] Candidate power generation time windows are one or more peak periods that meet the optimal electricity price conditions and are initially selected based on the power generation timing optimization method. They still need to undergo subsequent start-up response time verification and physical constraint verification.

[0107] The startup response time of a pumped storage unit is the time delay required from the issuance of a power generation command to actual grid connection and full power generation. This includes the generator unit's start-up preparation, speed-up, grid connection, and other operational steps. The unit must be started in advance before this time to ensure accurate grid connection at the start of peak hours.

[0108] The optimal power generation time window is the final start and end time of power generation operation determined after correction of the start-up response time. Its start time is the start time of the peak period minus the start-up response time of the generator unit, and its end time is aligned with the end time of the peak period.

[0109] By employing a pumping timing optimization method, the candidate window with the lowest electricity price, longest duration, and lowest volatility is identified from multiple off-peak periods. The pumping unit's start-up response time is advanced to ensure precise full-power pumping and storage at the beginning of the off-peak period, maximizing the absorption of low-price benefits and avoiding arbitrage losses due to start-up delays. Similarly, the power generation timing optimization method identifies the peak period with the highest electricity price, longest duration, and lowest volatility, and starts grid connection at full capacity in advance to ensure power generation begins at the start of the peak period, maximizing peak electricity sales revenue.

[0110] S4043, based on the optimal pumping time window and the optimal power generation time window, combined with the physical operation constraints of the pumped storage power station, uses the constraint verification method to verify and obtain the pumping time window and power generation time window after verification, and outputs them as the pumping time window and power generation time window for the predicted time period.

[0111] Specifically, S4043 includes: a1. Based on the optimal pumping time window and the optimal power generation time window, the corresponding duration is obtained respectively. The duration is then compared with the lower limit of the effective duration of pumping or the lower limit of the effective duration of power generation using the duration verification method to obtain the pumping duration evaluation result and the power generation duration evaluation result. a2, based on the number of pumping and power generation times during the predicted time period, combined with the maximum number of daily start-ups and shutdowns of the pumped storage power station, the maximum number of start-ups and shutdowns method is used for evaluation and verification to obtain the start-up and shutdown frequency evaluation result; a3. Based on the difference between the end time of adjacent optimal pumping time windows and the start time of optimal power generation time windows, the switching interval is compared using the operating condition switching interval verification rule to obtain the switching interval evaluation result. a4 integrates the evaluation results of pumping time, power generation time, start-stop frequency, and switching interval, and outputs the optimal pumping time window and optimal power generation time window that meet the preset conditions. These are then used as the pumping time window and power generation time window after successful verification.

[0112] First, the duration of the optimal pumping time window and the optimal power generation time window are obtained separately and compared with the lower limits of the effective duration of pumping and power generation to ensure that each time window has sufficient physical duration to complete effective pumping or power generation operations. Then, based on the number of pumping and power generation operations within the predicted time period, and combined with the maximum number of daily unit start-ups and shutdowns, verification is performed to reduce the risk of unit lifespan loss due to excessively frequent start-ups and shutdowns within a day. At the same time, based on the difference between the end time of adjacent optimal pumping time windows and the start time of optimal power generation time windows, a comparison is performed using the operating condition switching interval verification rule to ensure that there is sufficient interval time for physical recovery and safety separation between two adjacent operating condition switches. Finally, by combining the duration evaluation results, start-up and shutdown frequency evaluation results, and switching interval evaluation results, the optimal pumping time window and the optimal power generation time window that simultaneously meet all preset physical constraints are output. This ensures that the final control method fully respects the physical limits and operational safety requirements of the unit equipment while pursuing economic optimization, effectively avoiding the problem of the theoretical optimal solution being out of touch with engineering reality.

[0113] By employing a pumping timing optimization method to screen candidate pumping time windows from multiple off-peak periods and then modifying them based on the actual start-up response time of the pumping units in the pumped storage power station, the generated pumping windows are designed to satisfy both economic efficiency and the physical response capability of the equipment, avoiding execution deviations caused by start-up delays. Similarly, a power generation timing optimization method is used to determine the power generation window based on the start-up response time of the generator units, ensuring that power generation commands are executed on time. Finally, a constraint verification method is used to verify the physical operation constraints of the optimal pumping time window and the optimal power generation time window, ensuring that the final output pumping time window and power generation time window meet the constraints of equipment response physicality and system operation safety in actual operation.

[0114] In some optional embodiments, when the duration of the pumping time window is less than the lower limit of the effective pumping duration or the duration of the power generation time window is less than the lower limit of the effective power generation duration, the above-mentioned S404 further includes: S4044: Based on the valley period of the predicted time period, the pumping time window of the valley period is extracted using the pumping window supplementation method to obtain the supplemented pumping time window; or based on the secondary peak period in the predicted time period, the power generation time window of the secondary peak period is extracted using the power generation window supplementation method to obtain the supplemented power generation time window.

[0115] When the duration of the pumping time window is less than the lower limit of the effective pumping duration, the pumping time window supplementation method is used to extract available pumping time windows from the valley period of the predicted time period to supplement the original pumping window. When the duration of the power generation time window is less than the lower limit of the effective power generation duration, the power generation window supplementation method is used to extract available power generation time windows from the second peak period of the predicted time period to supplement the original power generation window. This avoids abandoning the period due to the window being too short, and ensures that the control scheme can still guarantee the scheduling flexibility of the power plant by extending the working time when the window is insufficient.

[0116] This embodiment provides a control method for a pumped storage power station, which can be used in the control terminal of a pumped storage power station. Figure 5 This is a flowchart of a pumped storage power station control method according to an embodiment of the present invention, such as... Figure 5 As shown, the process includes the following steps: S501, based on historical electricity supply and demand data and factors influencing electricity prices, utilizes a long-term electricity price forecasting model to obtain long-term electricity price forecasts for the forecast period. These forecasts include monthly electricity price forecast curves, low-price forecast thresholds, and high-price forecast thresholds. For details, please refer to [link to relevant documentation]. Figure 1 S101 of the illustrated embodiment will not be described again here.

[0117] S502, based on long-term electricity price forecasts for the forecast period, combines currently collected electricity load data, renewable energy output data, inter-provincial transmission line planning data, and meteorological data, and uses a short-term correction model for adjustments to obtain the spot electricity price forecast curve for the forecast period. For details, please refer to [link to details]. Figure 1 S102 of the illustrated embodiment will not be described again here.

[0118] S503, based on the spot electricity price forecast curve for the forecast period, combines the low-price forecast threshold and the high-price forecast threshold for the forecast period, and uses a peak-valley interval filtering method to identify multiple low-price periods and multiple high-price periods for the forecast period. For details, please refer to [link to details]. Figure 1 S103 of the illustrated embodiment will not be described again here.

[0119] S504, based on multiple low-peak and high-peak periods within the predicted timeframe, and considering the physical operational constraints of the pumped-storage power station, utilizes an optimal decision-making method to determine the pumping and power generation time windows within the predicted timeframe. For details, please refer to [link to relevant documentation]. Figure 1 S104 of the illustrated embodiment will not be described again here.

[0120] S505, based on the actual electricity price data and unit operation data of pumped storage power stations after execution according to pumping time windows and power generation time windows, combines the long-term electricity price forecast results and spot electricity price forecast curves, and uses the deviation algorithm to optimize the long-term electricity price forecast model, short-term correction model, peak-valley interval screening method and optimal decision method.

[0121] By utilizing operational data after actual execution to identify the sources of prediction deviations, and accordingly iteratively adjusting model parameters, threshold settings, and decision rules, the algorithm models and parameter settings of each link can dynamically evolve with actual operational data, continuously adapting to local power market characteristics and actual unit response characteristics, and avoiding the gradual decline in control effectiveness due to the model becoming out of sync with market changes due to fixed parameters.

[0122] This embodiment also provides a pumped storage power station control device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0123] This embodiment provides a control device for a pumped storage power station, such as... Figure 6 As shown, it includes: The electricity price forecasting module 610 is used to obtain long-term electricity price forecasting results for the forecast period based on historical electricity supply and demand data and electricity price influencing factors, using a long-term electricity price forecasting model. The long-term electricity price forecasting results include monthly electricity price forecast curves, electricity price trough forecasting thresholds, and electricity price peak forecasting thresholds. The electricity price correction module 620 is used to adjust the long-term electricity price forecast results based on the forecast period, combined with the currently collected electricity load data, new energy output data, inter-provincial interconnection line plan data and meteorological data, using a short-term correction model to obtain the spot electricity price forecast curve for the forecast period. The peak-valley identification module 630 is used to identify multiple low-valley periods and multiple peak periods based on the spot electricity price prediction curve for the prediction period, combined with the low-valley prediction threshold and the high-valley prediction threshold for the prediction period, using the peak-valley interval filtering method. The decision generation module 640 is used to determine the pumping time window and power generation time window within the predicted time period based on multiple low-end periods and multiple peak periods, combined with the physical operation constraints of the pumped storage power station, using the optimal decision method.

[0124] In some alternative implementations, the electricity price adjustment module 620 includes: The time synchronization unit is used to synchronize the electricity load data, new energy output data, inter-provincial transmission line planning data and meteorological data collected at multiple sampling times using a timestamp alignment method at a uniform time frequency, so as to obtain synchronized multi-dimensional feature data. The deviation analysis unit is used to analyze the impact of electricity load data, new energy output data, inter-provincial transmission line planning data and meteorological data on electricity price data based on synchronized multi-dimensional feature data and using the impact degree assessment method, and obtain the deviation analysis results for each influencing factor. The curve adjustment unit is used to adjust the electricity price curve and peak-valley threshold in the long-term electricity price forecast results based on the deviation analysis results of various influencing factors and the deviation correction method, and outputs the spot electricity price forecast curve for the forecast period.

[0125] In some alternative implementations, the peak-valley identification module 630 includes: The threshold evaluation unit is used to obtain the low-price prediction threshold and high-price prediction threshold for the predicted time period based on the predicted low-price and high-price prediction thresholds for the predicted time period. It combines historical electricity price peak-valley characteristics, current load data and new energy output data, and uses a dynamic threshold evaluation method to obtain the low-price and high-price thresholds for the preset time period. The interval division unit is used to predict the spot electricity price curve based on the prediction time period. It combines the low-valley threshold and high-valley threshold of the preset time period and uses the effective peak-valley interval screening method to remove instantaneous intervals with a duration lower than the preset threshold, thereby obtaining multiple low-valley periods and multiple high-valley periods in the prediction time period.

[0126] In some alternative implementations, the decision generation module 640 includes: The pumping window construction unit is used to generate candidate pumping time windows based on multiple low periods of the predicted time period, using the pumping timing optimization method in the optimal decision method, and combined with the start-up response time of the pumping unit of the pumped storage power station to obtain the optimal pumping time window. The power generation window construction unit is used to generate candidate power generation time windows based on multiple peak periods of the predicted time period, using the power generation timing optimization method in the optimal decision method, and combined with the start-up response time of the generator units of the pumped storage power station to obtain the optimal power generation time window. The constraint evaluation unit is used to verify the optimal pumping time window and the optimal power generation time window based on the physical operation constraints of the pumped storage power station using the constraint verification method. The verification results in the pumping time window and the power generation time window after passing the verification, and the output is the pumping time window and the power generation time window for the predicted time period.

[0127] In some optional implementations, the physical operating constraints of the pumped storage power station include the maximum number of daily start-ups and shutdowns of the pumped storage units, the minimum interval for switching operating conditions, the lower limit of the effective pumping duration, and the lower limit of the effective power generation duration. The constraint evaluation unit is specifically used for: Based on the optimal pumping time window and the optimal power generation time window, the corresponding duration is obtained respectively. The duration is then compared with the lower limit of the effective pumping duration or the lower limit of the effective power generation duration using the duration verification method to obtain the pumping duration evaluation result and the power generation duration evaluation result. Based on the number of pumping and power generation times during the predicted time period, combined with the maximum number of daily start-ups and shutdowns of the pumped storage power station, the maximum number of start-ups and shutdowns method is used for evaluation and verification to obtain the start-up and shutdown frequency evaluation results. Based on the difference between the end time of adjacent optimal pumping time windows and the start time of optimal power generation time windows, the switching interval is compared using the operating condition switching interval verification rule to obtain the switching interval evaluation result. Based on the combined evaluation results of pumping time, power generation time, start-stop frequency, and switching interval, the optimal pumping time window and optimal power generation time window that meet the preset conditions are output, and these serve as the pumping time window and power generation time window after successful verification.

[0128] In some optional implementations, when the duration of the pumping time window is less than the lower limit of the effective pumping duration or the duration of the power generation time window is less than the lower limit of the effective power generation duration, the constraint evaluation unit is further specifically used for: Based on the valley period of the predicted time period, the pumping time window of the valley period is extracted using the pumping time period power generation window supplementation method to obtain the supplemented pumping time window; or based on the secondary peak period in the predicted time period, the power generation time window of the secondary peak period is extracted using the power generation window supplementation method to obtain the supplemented power generation time window.

[0129] In some alternative implementations, it also includes: The model optimization module is used to optimize the long-term electricity price prediction model, short-term correction model, peak-valley interval screening method, and optimal decision method based on the actual electricity price data and unit operation data of the pumped storage power station after execution according to the pumping time window and the power generation time window, combined with the long-term electricity price prediction results and the spot electricity price prediction curve, using iterative algorithms.

[0130] The pumped storage power station control device provided in this embodiment of the invention can execute the pumped storage power station control method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as in the corresponding embodiments described above, and will not be repeated here.

[0131] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0132] The following is a detailed reference. Figure 7 This diagram illustrates a suitable structural schematic for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 701, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 702 or a program loaded from memory 708 into random access memory (RAM) 703. The RAM 703 also stores various programs and data required for the operation of the electronic device. The processor 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0133] Typically, the following devices can be connected to I / O interface 705: input devices 706 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 707 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 708 including, for example, magnetic tapes, hard disks, etc.; and communication devices 709. Communication device 709 allows electronic devices to exchange data via wireless or wired communication with other devices. Although Figure 7 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0134] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 709, or installed from a memory 708, or installed from a ROM 702. When the computer program is executed by the processor 701, it performs the functions defined in the pumped storage power station control method of the embodiments of the present invention.

[0135] Figure 7The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0136] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the pumped-storage power station control method shown in the above embodiments is implemented.

[0137] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0138] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for regulating a pumped storage power station, characterized in that, The method includes: Based on historical electricity supply and demand data and factors affecting electricity prices, a long-term electricity price forecast model is used to obtain long-term electricity price forecast results for the forecast period. The long-term electricity price forecast results include monthly electricity price forecast curves, electricity price trough forecast thresholds, and electricity price peak forecast thresholds. Based on the long-term electricity price forecast results for the forecast period, combined with the currently collected electricity load data, new energy output data, inter-provincial transmission line planning data and meteorological data, a short-term correction model is used for adjustment to obtain the spot electricity price forecast curve for the forecast period. Based on the spot electricity price forecast curve for the forecast period, and combined with the peak and trough forecast thresholds for the forecast period, the peak-trough interval screening method is used to identify multiple trough periods and multiple peak periods for the forecast period. Based on multiple low-peak periods and multiple peak periods within the predicted time period, and combined with the physical operation constraints of the pumped storage power station, the optimal decision-making method is used to determine the pumping time window and power generation time window within the predicted time period.

2. The method according to claim 1, characterized in that, The spot electricity price forecast curve based on the forecast time period, combined with the peak and trough price forecast thresholds for the forecast time period, is identified using a peak-trough interval filtering method to obtain multiple trough periods and multiple peak periods for the forecast time period, including: Based on the predicted low-price and high-price thresholds for the predicted time period, and combined with historical electricity price peak-valley characteristics, current load data, and renewable energy output data, a dynamic threshold evaluation method is used to obtain the low-price and high-price thresholds for the preset time period. Based on the spot electricity price forecast curve for the forecast period, and combined with the electricity price trough threshold and peak threshold for the preset period, the effective peak-valley interval screening method is used to remove instantaneous intervals with a duration lower than the preset threshold, thereby obtaining multiple trough periods and multiple peak periods for the forecast period.

3. The method according to claim 1, characterized in that, Based on multiple low-peak and high-peak periods within the predicted time period, and combined with the physical operational constraints of the pumped storage power station, the optimal decision-making method is used to determine the pumping time window and power generation time window within the predicted time period, including: Based on multiple low periods in the predicted time period, the optimal pumping time window is generated by using the pumping timing optimization method in the optimal decision-making method. Combined with the start-up response time of the pumping unit of the pumped storage power station, the optimal pumping time window is obtained. Based on multiple peak periods of the predicted time period, the optimal power generation time window is generated by using the power generation timing optimization method in the optimal decision-making method. Combined with the start-up response time of the generator units of the pumped storage power station, the optimal power generation time window is obtained. Based on the optimal pumping time window and the optimal power generation time window, and combined with the physical operation constraints of the pumped storage power station, the constraint verification method is used to verify the pumping time window and the power generation time window after verification, and output them as the pumping time window and the power generation time window for the predicted time period.

4. The method according to claim 3, characterized in that, The physical operating constraints of the pumped storage power station include the maximum number of daily start-ups and shutdowns of the pumped storage units, the minimum interval for switching operating conditions, the lower limit of the effective pumping duration, and the lower limit of the effective power generation duration. Based on the optimal pumping time window and the optimal power generation time window, and in conjunction with the physical operating constraints of the pumped storage power station, a constraint verification method is used to verify the optimal pumping time window and the optimal power generation time window, resulting in the verified pumping time window and power generation time window, including: Based on the optimal pumping time window and the optimal power generation time window, the corresponding duration is obtained respectively, and the duration is compared with the lower limit of the effective duration of pumping or the lower limit of the effective duration of power generation using the duration verification method, so as to obtain the pumping duration evaluation result and the power generation duration evaluation result. Based on the number of pumping and power generation times during the predicted time period, and combined with the maximum number of daily start-ups and shutdowns of the pumped storage power station, the maximum number of start-ups and shutdowns method is used for evaluation and verification to obtain the start-up and shutdown count evaluation result. Based on the difference between the end time of the adjacent optimal pumping time window and the start time of the optimal power generation time window, the switching interval is compared using the operating condition switching interval verification rule to obtain the switching interval evaluation result. Based on the combined evaluation results of pumping time, power generation time, start-stop frequency, and switching interval, the optimal pumping time window and optimal power generation time window that meet the preset conditions are output and used as the pumping time window and power generation time window after verification.

5. The method according to claim 3, characterized in that, When the duration of the pumping time window is less than the lower limit of the effective pumping duration or the duration of the power generation time window is less than the lower limit of the effective power generation duration, the method further includes: Based on the predicted time period, the pumping time window for the flat valley period is extracted using the pumping window supplementation method, and the supplemented pumping time window is obtained. Alternatively, based on the second-peak period in the predicted time period, the power generation time window of the second-peak period can be extracted using the power generation window supplementation method to obtain the supplemented power generation time window.

6. The method according to claim 1, characterized in that, The long-term electricity price forecast results based on the forecast period are combined with currently collected electricity load data, renewable energy output data, inter-provincial transmission line planning data, and meteorological data, and adjusted using a short-term correction model to obtain the spot electricity price forecast curve for the forecast period, including: Based on the currently collected electricity load data, new energy output data, inter-provincial transmission line planning data and meteorological data, the timestamp alignment method is used to synchronize the data at a unified time frequency to obtain synchronized multi-dimensional feature data. Based on the synchronized multidimensional feature data, the impact degree assessment method is used to analyze the impact degree of electricity load data, new energy output data, inter-provincial connection line planning data and meteorological data on electricity price data, and the deviation analysis results for each influencing factor are obtained. Based on the deviation analysis results of various influencing factors, the electricity price curve and peak-valley threshold in the long-term electricity price forecast results are adjusted using the deviation correction method, and the output is the spot electricity price forecast curve for the forecast period.

7. The method according to any one of claims 1 to 6, characterized in that, Also includes: Based on the actual electricity price data and unit operation data of pumped storage power stations after the pumping time window and power generation time window are executed, combined with the long-term electricity price prediction results and spot electricity price prediction curve, the long-term electricity price prediction model, short-term correction model, peak-valley interval screening method and optimal decision method are optimized using iterative algorithms.

8. A control device for a pumped storage power station, characterized in that, The device includes: The electricity price forecasting module is used to obtain long-term electricity price forecasting results for the forecast period based on historical electricity supply and demand data and factors affecting electricity prices, using a long-term electricity price forecasting model. The long-term electricity price forecasting results include monthly electricity price forecasting curves, peak electricity price forecasting thresholds, and trough electricity price forecasting thresholds. The electricity price correction module is used to adjust the long-term electricity price forecast results based on the forecast period, combined with the currently collected electricity load data, new energy output data, inter-provincial interconnection line plan data and meteorological data, using a short-term correction model to obtain the spot electricity price forecast curve for the forecast period. The peak-valley identification module is used to identify multiple low-valley periods and multiple peak periods based on the spot electricity price prediction curve for the prediction period, combined with the peak and low price prediction thresholds for the prediction period. The decision generation module is used to determine the pumping time window and power generation time window within the predicted time period based on multiple low-peak periods and multiple peak periods, combined with the physical operation constraints of the pumped storage power station, using the optimal decision method.

9. An electronic device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform the pumped storage power station control method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the pumped storage power station control method according to any one of claims 1 to 7.

Images