A method and system for optimizing scheduling of water-photovoltaic complementarity based on time scale

Abstract

The present invention provides a time-scale-based water-photovoltaic complementary optimization scheduling method and system, including real-time monitoring of the current photovoltaic capacity of the photovoltaic power generation system, the current hydropower capacity of the hydropower generation system, and the current load demand of the power grid; based on the current photovoltaic capacity, the current weather and the current time, predicting the photovoltaic predicted capacity of photovoltaic power generation after the first time; determining the maximum historical prediction error of the photovoltaic capacity under the same conditions as the current weather and the current time; predicting the predicted load demand of the power grid after the first time; according to the current photovoltaic capacity, the photovoltaic predicted capacity, the maximum historical prediction error, and the predicted load demand, under the constraint of meeting the power grid demand, with the goal of maximizing the use of photovoltaics, determining the hydropower capacity adjustment amount. The above scheme can improve the efficiency and accuracy of scheduling.

Description

Technical Field

[0001] The present invention relates to the field of power grid dispatching, and in particular to a time-scale-based water-photovoltaic complementary optimization dispatching method and system. Background Art

[0002] Hydropower and photovoltaic complementary optimization scheduling is a power system scheduling strategy that aims to maximize the advantages of hydropower and photovoltaic power generation to meet the load demand of the power system while reducing costs and increasing the utilization rate of renewable energy. Hydropower and photovoltaic complementary optimization scheduling takes into account two different renewable energy sources, hydropower and photovoltaic power generation, and reasonably schedules hydropower and photovoltaic power generation to make them complement each other and operate in coordination.

[0003] The existing technology usually predicts the future capacity of photovoltaic power generation systems based on historical data and weather forecasts, and adjusts it according to the existing user load. However, the capacity of photovoltaic power generation systems is affected by a variety of unexpected factors, and the existing technology does not consider the prediction error. At the same time, the prediction cycle of the existing technology is usually long. However, due to the influence of human activities such as concentrated commuting of users, user needs often have large variables in a short period of time. The existing technology is difficult to consider the frequently changing user demand load, resulting in inaccurate adjustment. Summary of the invention

[0004] In order to solve the problems in the prior art, the present invention provides a time-scale-based water-photovoltaic complementarity optimization scheduling method and system.

[0005] In one aspect of the present invention, a time-scale-based water-photovoltaic complementary optimization scheduling method is provided, comprising the following steps: real-time monitoring of the current photovoltaic capacity of the photovoltaic power generation system, the current hydropower capacity of the hydropower generation system, and the current load demand of the power grid; predicting the photovoltaic predicted capacity of photovoltaic power generation after a first time based on the current photovoltaic capacity, the current weather and the current time; determining the maximum historical prediction error of the photovoltaic capacity under the same conditions as the current weather and the current time; predicting the predicted load demand of the power grid after a first time; and determining the hydropower capacity adjustment amount based on the current photovoltaic capacity, the photovoltaic predicted capacity, the maximum historical prediction error, and the predicted load demand, with the goal of maximizing the utilization of photovoltaic power under the constraint of meeting the power grid demand.

[0006] Furthermore, the predicted load demand of the power grid after the first time is predicted specifically includes:

[0007] Obtain historical load data and historical weather data;

[0008] Dividing the historical load data according to the first time length, and calculating the load increase of each first time length;

[0009] Label each load increase based on whether it is a holiday, time period, or weather condition;

[0010] The load increase with the same label is determined as a same-label set, and the mean value of the same-label set + 2 times the standard deviation is used as the predicted load increase of the label corresponding to the same-label set;

[0011] The current label is determined according to the current time and weather data, and the corresponding predicted load increase is determined according to the current label; the predicted load demand of the power grid after the first time is calculated according to the current load and the predicted load increase.

[0012] Furthermore, the historical weather data at least includes temperature data.

[0013] Furthermore, the maximum historical prediction error is an error percentage or an absolute data.

[0014] Furthermore, according to the current photovoltaic capacity, photovoltaic forecast capacity, maximum historical forecast error, forecast load demand, under the constraint of meeting grid demand, with the goal of maximizing photovoltaic utilization, the hydropower capacity adjustment amount is determined as follows:

[0015] To meet the grid demand, the sum of photovoltaic and hydropower must be greater than the predicted load demand, then:

[0016] P current_water +P adjust_water +P prdict_solar -P max_error ≥P prdict_need

[0017] and then

[0018] P adjust_water =min(P prdict_need -P current_water -P prdict_solar +P max_error )

[0019] Among them, P adjust_water is the hydropower capacity regulation; P current_water is the current hydropower capacity; P prdict_solar Forecasting production capacity for photovoltaics; P max_error is the maximum predicted production capacity error of photoelectricity; P prdict_need is to predict the load demand; min is to find the minimum value.

[0020] Another aspect of the present invention is to provide a water-photovoltaic complementary optimization scheduling system based on time scale, comprising the following modules:

[0021] Monitoring module, used to monitor the current photovoltaic power generation capacity of the photovoltaic power generation system, the current hydropower generation capacity of the hydropower generation system, and the current load demand of the power grid in real time;

[0022] A prediction module is used to predict the photovoltaic predicted capacity of photovoltaic power generation after the first time based on the current photovoltaic capacity, current weather and current time; determine the maximum historical prediction error of photovoltaic capacity under the same conditions as the current weather and current time; predict the predicted load demand of the power grid after the first time;

[0023] The determination module is used to determine the hydropower capacity adjustment amount based on the current photovoltaic capacity, photovoltaic forecast capacity, maximum historical forecast error, and forecast load demand, with the goal of maximizing the use of photovoltaic power under the constraint of meeting the grid demand.

[0024] Furthermore, the predicted load demand of the power grid after the first time is predicted specifically includes:

[0025] Obtain historical load data and historical weather data;

[0026] Dividing the historical load data according to the first time length, and calculating the load increase of each first time length;

[0027] Label each load increase based on whether it is a holiday, time period, or weather condition;

[0028] The load increase with the same label is determined as a same-label set, and the mean value of the same-label set + 2 times the standard deviation is used as the predicted load increase of the label corresponding to the same-label set;

[0029] The current label is determined according to the current time and weather data, and the corresponding predicted load increase is determined according to the current label; the predicted load demand of the power grid after the first time is calculated according to the current load and the predicted load increase.

[0030] Furthermore, the historical weather data at least includes temperature data.

[0031] Furthermore, the maximum historical prediction error is an error percentage or an absolute data.

[0032] Furthermore, according to the current photovoltaic capacity, photovoltaic forecast capacity, maximum historical forecast error, forecast load demand, under the constraint of meeting grid demand, with the goal of maximizing photovoltaic utilization, the hydropower capacity adjustment amount is determined as follows:

[0033] To meet the grid demand, the sum of photovoltaic and hydropower must be greater than the predicted load demand, then:

[0034] P current_water +P adjust_water +P prdict_solar -P max_error ≥P prdict_need

[0035] and then

[0036] P adjust_water =min(P prdict_need -P current_water -P prdict_solar +P max_error )

[0037] Among them, P adjust_water is the hydropower capacity regulation; P current_water is the current hydropower capacity; P prdict_solar Forecasting production capacity for photovoltaics; P max_error is the maximum predicted production capacity error of photoelectricity; P prdict_need is to predict the load demand; min is to find the minimum value.

[0038] The present invention can produce the following beneficial effects through the above technical solution:

[0039] By real-time monitoring of the production capacity of photovoltaic power generation systems and hydropower systems, and predicting and adjusting them according to current weather and electricity consumption conditions, the production capacity of photovoltaic power generation systems and hydropower systems can be maximized, thereby improving the utilization rate of renewable energy.

[0040] By real-time monitoring of grid load demand and forecasting of photovoltaic production capacity, as well as determining the maximum historical forecast error, it is possible to more accurately predict future electricity demand, make targeted adjustments to power generation plans and grid operation strategies, and improve the stability and reliability of the grid.

[0041] Load demand forecasting can be done simply and quickly, and hourly forecasts can usually be made using one or two years of data. Compared with forecasting methods such as machine learning, it has low hardware requirements and requires less data. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0044] Below, the invention is preferably described in conjunction with the accompanying drawings and specific implementation methods.

[0045] This embodiment solves the above problem through the following steps:

[0046] In one embodiment, reference Figure 1 The present invention provides a time-scale-based optimization scheduling method for water-photovoltaic complementarity.

[0047] Hydropower (hydroelectricity) is a method of generating electricity by converting the kinetic energy of water flow into electrical energy. It is usually achieved by building a dam to form a reservoir to store water, and using the drop in water flow to drive a turbine generator to generate electricity.

[0048] Hydropower generation is stable, controllable, and predictable in the long term. Hydropower stations can adjust water flow according to demand, making the generation of electricity highly controllable and providing sustainable power supply in areas with abundant water resources.

[0049] Photovoltaic power generation is a power generation method that uses sunlight to directly convert into electrical energy. Photovoltaic panels convert solar photons into electrical energy.

[0050] Photovoltaic power generation is clean, pollution-free, and noise-free, and is an environmentally friendly energy source. However, photovoltaic power generation is greatly affected by weather and sunshine conditions, has unstable production capacity, and cannot generate electricity at night.

[0051] Photovoltaic power generation is greatly affected by the weather and cannot generate electricity when there is insufficient solar energy, while hydropower generation is stable and controllable. Through the complementary scheduling of hydropower and photovoltaic power generation, hydropower generation can be used to maintain a stable supply of the power grid when photovoltaic power generation is insufficient.

[0052] Hydropower and photovoltaics are two major renewable energy sources. Through complementary scheduling, we can maximize the use of these two resources, reduce dependence on traditional fossil energy, and achieve a high proportion of clean energy. Complementary scheduling of hydropower and photovoltaics can solve the problem of large fluctuations and instability in photovoltaic power generation to a certain extent, while promoting the rational use of renewable energy and the stable operation of the power grid.

[0053] If the capacity of photovoltaic power generation or hydropower generation is insufficient, it may lead to insufficient power supply to the grid, affect the power demand of users, and even cause power system failure. Therefore, timely monitoring of capacity and load demand can help dispatchers make timely adjustments to ensure the balance between supply and demand. At the same time, real-time monitoring of capacity and load demand can provide real-time data support for optimizing dispatching strategies to maximize the use of renewable energy, reduce power generation costs, and improve the power supply quality of the grid.

[0054] Real-time monitoring of capacity and load demand can also help respond to emergencies. For example, if a photovoltaic power generation system fails or a sudden change in weather causes a decrease in photovoltaic power generation capacity, the dispatcher can adjust the operating status of the hydropower generator set in time to make up for the lack of photovoltaic power generation and ensure the stable power supply of the power grid. Therefore, this embodiment needs to first monitor the current photovoltaic capacity of the photovoltaic power generation system, the current hydropower capacity of the hydropower generation system, and the current load demand of the power grid in real time.

[0055] Specifically, sensors can be installed on photovoltaic power generation systems, hydropower generation systems and key nodes of power grids to monitor production capacity and load demand in real time. For example, light intensity sensors can be installed on photovoltaic power generation systems, water level sensors can be installed on hydropower generation systems, and current and voltage sensors can be installed on various nodes of the power grid. The data collected by the sensors can be collected in real time. The data acquisition system can use various traditional or modern communication technologies to transmit and save the data to the central control system.

[0056] The capacity of the photovoltaic power generation system is affected by weather conditions, such as solar radiation intensity, cloud cover, etc. Therefore, based on real-time monitoring of light intensity, predicting the photovoltaic power generation capacity in the future can improve the accuracy of scheduling. In this way, the system can prepare in advance and reasonably adjust the operation strategy of the photovoltaic power generation system to meet the grid demand in the future period. Therefore, this embodiment also predicts the photovoltaic predicted capacity of photovoltaic power generation after the first time based on the current photovoltaic capacity, current weather and current time.

[0057] In order to realize photovoltaic production capacity prediction, it is necessary to collect current weather data, including solar radiation intensity, cloud cover, wind speed, etc. These data can be obtained through meteorological stations, meteorological satellites, weather forecasts and other channels.

[0058] Analyze historical weather data and photovoltaic power generation system capacity data, and establish a relationship model between weather and photovoltaic power generation capacity. Statistical methods, machine learning algorithms, etc. can be used for data analysis and modeling. Based on the results of historical data analysis, a prediction model for photovoltaic predicted capacity is constructed. The model can be a time series model based on statistical methods, or a regression model, neural network model, etc. based on machine learning. Or any other method in the prior art, such as the "Short-term prediction model for output power of photovoltaic power generation system based on improved BP neural network" disclosed by Ding Ming et al.

[0059] After real-time monitoring of the current weather conditions, the constructed prediction model is used to predict the photovoltaic power generation capacity in the future. This prediction can be calculated forward for a period of time based on the current time point, such as predicting the photovoltaic power generation capacity in the next 1 hour, 3 hours or longer.

[0060] Since the prediction results of photovoltaic power generation are often accompanied by certain uncertainties, even under the same weather conditions, the prediction results will fluctuate due to the complexity of the photovoltaic power generation system itself and the uncertainty of weather forecasts. Due to different times and different weather conditions, the probability of sudden weather changes is different. For example, in summer, when the air temperature is high, the probability of sudden rain is high, while in winter, the probability of sudden weather changes is small. Therefore, the prediction errors at different time points will also be different. In order to reduce the fluctuations of the power grid and ensure the electricity needs of users, this implementation further needs to consider the prediction error. Therefore, based on the current weather and the current time, the maximum historical prediction error is determined.

[0061] Specifically, historical forecast data of photovoltaic power generation systems are collected, including predicted photovoltaic power generation capacity and actually observed photovoltaic power generation capacity. These data should include a sufficiently long time period, covering various weather conditions and time points. The historical forecast data are analyzed to calculate the forecast error for each forecast time point. The forecast error can be calculated by the difference between the predicted value and the actual observed value. Further, the error can be expressed as a percentage to apply different basic absolute values. For each forecast time point, find the maximum value of the historical forecast error (which can be expressed as a percentage). This maximum historical forecast error can be used as the maximum forecast error (percentage) under the current weather and current time.

[0062] In order to maintain the supply and demand balance of the power system, the power system needs to adjust the power generation and power supply capacity according to the expected load demand to ensure sufficient power supply and avoid insufficient or excessive power supply. Based on the expected load demand, the power generation system can reasonably plan the operation of the generator set and adjust the power generation strategy to meet the power demand in the future period while minimizing unnecessary power generation costs. Therefore, the power grid predicts the load demand based on the current power consumption, time and weather.

[0063] The specific steps are as follows:

[0064] First, obtain historical load data and historical weather data. The historical load data can be obtained based on the records of the power company. The historical load data is recorded according to a preset period, such as one minute, five minutes, etc. Since the use of civil electricity is highly correlated with the weather, the corresponding weather data also needs to be obtained. The weather data should at least include the temperature.

[0065] The historical load data is divided according to the first time length, and the load increase of each first time length is calculated. For example, the first time length is measured in half an hour. For example, the time can be divided into 0:00 to 0:30, 0:30 to 1:00, 1:00 to 1:30..., then a day can be divided into 48 time periods. The finer the division, the more accurate the prediction, but the more data needs to be processed and the larger the amount of data required. When implementing this embodiment, those skilled in the art can select a suitable division length based on the existing data.

[0066] Since the historical load data has been obtained before, the load increase in each time period can be calculated based on the historical load data. It can be predicted that when the load increases, the load increase is positive, and when the load decreases, the load increase is negative.

[0067] Since holidays, time periods (mainly human concentrated activity time periods, such as commuting time periods, etc.) and weather have the greatest impact on load increase, in order to extract the same influence, each load increase is further labeled according to whether it is a holiday, time period, or weather condition. For example, from 10:00 to 10:30 on a Sunday with a temperature of 18 degrees, the load increase is 3500kwh, then the following data can be generated:

[0068] 18 degrees - Holidays - 10:00 to 10:30 3500kw

[0069] On a Monday with a temperature of 21 degrees Celsius, from 23:00 to 23:30, the load increase is -4500kwh, then the following data can be generated:

[0070] 21 degrees - weekdays - 23:00 to 23:30 -4500kw

[0071] Similarly, a similar labeling operation can be performed on each load increase.

[0072] The electricity usage conditions with the same temperature, holiday conditions, and time periods are similar. In order to determine the specific load change amount for each similar electricity usage condition, this embodiment further determines the load increase amount with the same label as a same-label set, and uses the mean + 2 times the standard deviation of the same-label set as the predicted load increase amount of the label corresponding to the same-label set.

[0073] Exemplarily, for the label "18 degrees - holidays - 10 to 10:30", 200 load increase data can be determined, and these 200 load increase data are regarded as a set with the same label. It should be noted that when determining the temperature label, the temperature within a certain range can be mapped to a value, such as 17.5 to 18.5 degrees are all regarded as 18 degrees. Calculate the mean and standard deviation of these 200 load increase data, then the mean + 2 times the standard deviation is the 97.5% percentile data, that is, the mean + 2 times the standard deviation of the set with the same label is greater than 97.5% of the 200 data. Under non-extreme conditions, this value can cover the situation where the temperature is 18 degrees, it is a holiday, and the load increase is from 10 to 10:30. Using this value as the predicted value can ensure normal electricity consumption for users under non-extreme conditions.

[0074] After obtaining the current time and weather data, the current label can be determined according to the current time and weather data. For example, the current date is January 1, 2024, New Year's Day, 19:00, and the temperature is 16 degrees, then the current label is 16 degrees-holiday-19:00 to 19:30. In the above steps, the predicted load increase of each label is calculated, and the predicted load increase of 16 degrees-holiday-19:00 to 19:30 can be found, so as to determine the corresponding predicted load increase according to the current label; the predicted load demand of the power grid after the first time is calculated according to the current load and the predicted load increase, that is, the current load is added to the predicted load increase to obtain the predicted load demand of the power grid after the first time.

[0075] The above method can be used to quickly and easily forecast load demand, usually using one or two years of data. Compared with forecasting methods such as machine learning, it has low hardware requirements and less data demand.

[0076] According to the current hydropower capacity, photovoltaic forecast capacity, the maximum historical photovoltaic forecast capacity error, and the forecast load demand, under the constraint of meeting the grid demand, the hydropower capacity adjustment is determined with the goal of maximizing the use of photovoltaic power. Among them, when the forecast value is larger than the actual value, the maximum historical photovoltaic forecast capacity error is positive, otherwise it is negative; when the hydropower capacity is adjusted upward, the hydropower capacity adjustment is positive, otherwise it is negative.

[0077] To meet the grid demand, the sum of photovoltaic and hydropower must be greater than the predicted load demand, then:

[0078] P current_water +P adjust_water +P prdict_solar -P max_error ≥P prdict_need

[0079] and then

[0080] P adjust_water=min(P prdict_need -P current_water -P prdict_solar +P max_error )

[0081] Among them, P adjust_water is the hydropower capacity regulation; P current_water is the current hydropower capacity; P prdict_solar Forecasting production capacity for photovoltaics; P max_error is the maximum predicted production capacity error of photoelectricity; P prdict_need is to predict the load demand; min is to find the minimum value.

[0082] Through the above method, by real-time monitoring of the production capacity of the photovoltaic power generation system and the hydropower generation system, and predicting and adjusting according to the current weather and electricity consumption, the production capacity of the photovoltaic power generation system and the hydropower generation system can be maximized, thereby improving the utilization rate of renewable energy.

[0083] By real-time monitoring of grid load demand and forecasting of photovoltaic production capacity, as well as determining the maximum historical forecast error, it is possible to more accurately predict future electricity demand, make targeted adjustments to power generation plans and grid operation strategies, and improve the stability and reliability of the grid.

[0084] Load demand forecasting can be done simply and quickly, and hourly forecasts can usually be made using one or two years of data. Compared with forecasting methods such as machine learning, it has low hardware requirements and requires less data.

[0085] On the other hand, the present invention also provides a water-photovoltaic complementary optimization scheduling system based on time scale, comprising:

[0086] Monitoring module, used to monitor the current photovoltaic power generation capacity of the photovoltaic power generation system, the current hydropower generation capacity of the hydropower generation system, and the current load demand of the power grid in real time;

[0087] A prediction module is used to predict the photovoltaic predicted capacity of photovoltaic power generation after the first time based on the current photovoltaic capacity, current weather and current time; determine the maximum historical prediction error of photovoltaic capacity under the same conditions as the current weather and current time; predict the predicted load demand of the power grid after the first time;

[0088] The determination module is used to determine the hydropower capacity adjustment amount based on the current photovoltaic capacity, photovoltaic forecast capacity, maximum historical forecast error, and forecast load demand, with the goal of maximizing the use of photovoltaic power under the constraint of meeting the grid demand.

[0089] Furthermore, the specific implementation methods of the above-mentioned time-scale-based water-photovoltaic complementary optimization scheduling system are the same as a time-scale-based water-photovoltaic complementary optimization scheduling method, and all further technical solutions in a time-scale-based water-photovoltaic complementary optimization scheduling method are completely introduced into a time-scale-based water-photovoltaic complementary optimization scheduling system.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

[0091] The present invention does not particularly specify the structure of some modules, which shall be subject to the contents recorded in the prior art. The prior art mentioned in the above background technology section and the specific embodiment section of the present invention can be used as part of the present invention to understand the meaning of some technical features or parameters. The scope of protection of the present invention shall be subject to the contents actually recorded in the claims.

Claims

1. A time-scale-based optimization scheduling method for water-photovoltaic complementarity, characterized in that The method comprises the following steps: Real-time monitoring of the current photovoltaic power generation capacity of the photovoltaic power generation system, the current hydropower generation capacity of the hydropower generation system, and the current load demand of the power grid; Based on the current photovoltaic capacity, current weather and current time, predict the photovoltaic capacity of photovoltaic power generation after the first time; Determine the maximum historical forecast error of photovoltaic power generation under the same conditions as the current weather and the current time; Predict the load demand of the power grid after the first forecast; Based on the current photovoltaic capacity, photovoltaic forecast capacity, the maximum historical forecast error, and forecast load demand, the hydropower capacity adjustment is determined with the goal of maximizing the use of photovoltaic power under the constraint of meeting the grid demand; The predicted load demand of the power grid after the first prediction includes: Obtain historical load data and historical weather data; Dividing the historical load data according to the first time length, and calculating the load increase of each first time length; Each load increase is labeled according to whether it is a holiday, time period, or weather condition; the load increases with the same label are determined as a same-label set, and the mean + 2 times the standard deviation of the same-label set is used as the predicted load increase of the label corresponding to the same-label set; The current label is determined according to the current time and weather data, and the corresponding predicted load increase is determined according to the current label; the predicted load demand of the power grid after the first time is calculated according to the current load and the predicted load increase.

2. The method for optimizing the scheduling of water-photovoltaic complementarity based on time scale according to claim 1 is characterized in that The historical weather data at least includes temperature data.

3. The method for optimizing the scheduling of water-photovoltaic complementarity based on time scale according to claim 1 is characterized in that The maximum historical forecast error is an error percentage or an absolute data.

4. The method for optimizing the scheduling of water-photovoltaic complementarity based on time scale according to claim 1 is characterized in that According to the current photovoltaic capacity, photovoltaic forecast capacity, maximum historical forecast error, forecast load demand, and under the constraint of meeting the grid demand, with the goal of maximizing the use of photovoltaic power, the hydropower capacity adjustment is determined as follows: To meet the grid demand, the sum of photovoltaic and hydropower must be greater than the predicted load demand, then: P current_water +P adjust_water +P prdict_solar -P max_error ≥P prdict_need and then P adjust_water =min(P prdict_need -P current_water -P prdict_solar +P max_error ) Among them, P adjust_water is the hydropower capacity regulation; P current_water is the current hydropower capacity; P prdict_solar Forecasting production capacity for photovoltaics; P max_error is the maximum predicted production capacity error of photoelectricity; P prdict_need is to predict the load demand; min is to find the minimum value.

5. A water-photovoltaic complementary optimization scheduling system based on time scale, characterized in that The system includes the following modules: Monitoring module, used to monitor the current photovoltaic power generation capacity of the photovoltaic power generation system, the current hydropower generation capacity of the hydropower generation system, and the current load demand of the power grid in real time; A prediction module, used to predict the photovoltaic predicted capacity of photovoltaic power generation after the first time based on the current photovoltaic capacity, current weather and current time; Determine the maximum historical forecast error of photovoltaic power generation under the same conditions as the current weather and the current time; predict the load demand of the power grid after the first time; A determination module is used to determine the hydropower capacity adjustment amount based on the current photovoltaic capacity, photovoltaic forecast capacity, maximum historical forecast error, and forecast load demand, with the goal of maximizing the use of photovoltaic power under the constraint of meeting the grid demand; The predicted load demand of the power grid after the first prediction includes: Obtain historical load data and historical weather data; Dividing the historical load data according to the first time length, and calculating the load increase of each first time length; Each load increase is labeled according to whether it is a holiday, time period, or weather condition; the load increases with the same label are determined as a same-label set, and the mean + 2 times the standard deviation of the same-label set is used as the predicted load increase of the label corresponding to the same-label set; The current label is determined according to the current time and weather data, and the corresponding predicted load increase is determined according to the current label; the predicted load demand of the power grid after the first time is calculated according to the current load and the predicted load increase.

6. The time-scale-based water-photovoltaic complementary optimization scheduling system according to claim 5 is characterized in that The historical weather data at least includes temperature data.

7. The time-scale-based water-photovoltaic complementary optimization scheduling system according to claim 5 is characterized in that The maximum historical forecast error is an error percentage or an absolute data.

8. The time-scale-based water-photovoltaic complementary optimization scheduling system according to claim 5 is characterized in that According to the current photovoltaic capacity, photovoltaic forecast capacity, maximum historical forecast error, forecast load demand, and under the constraint of meeting the grid demand, with the goal of maximizing the use of photovoltaic power, the hydropower capacity adjustment is determined as follows: To meet the grid demand, the sum of photovoltaic and hydropower must be greater than the predicted load demand, then: P current_water +P adjust_water +P prdict_solar -P max_error ≥P prdict_need and then P adjust_water =min(P prdict_need -P current_water -P prdict_solar +P max_error ) Among them, P adjust_water is the hydropower capacity regulation; P current_water is the current hydropower capacity; P prdict_solar Forecasting production capacity for photovoltaics; P max_error is the maximum predicted capacity error of photoelectricity; P prdict_need is to predict the load demand; min is to find the minimum value.

Images