Scheduling method, device and equipment of hydropower station, storage medium and program product

Abstract

The invention relates to the technical field of power dispatching, and discloses a hydropower station dispatching method, device and equipment, a storage medium and a program product.The hydropower station dispatching method comprises the steps that power and load are predicted, and power and load predicted values are obtained; performing prediction deviation analysis on the power prediction value and the load prediction value to obtain a first probability distribution function and a second probability distribution function; performing scene analysis on the first probability distribution function and the second probability distribution function to generate a plurality of first power scenes and a plurality of first load scenes; clustering the plurality of first power scenes and the plurality of first load scenes to obtain a plurality of first powers, a plurality of loads, a plurality of first occurrence probabilities and a plurality of second occurrence probabilities; and determining a hydropower station scheduling scheme by taking the minimum expectation of the maximum value of the residual load, the minimum expectation of the fluctuation of the downstream tail water level and the minimum expectation of the fluctuation of the discharged flow as targets. According to the method, the peak regulation effect of the downstream is considered, the hydropower station is accurately scheduled, and the practicability is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power dispatching, and in particular to a dispatching method, device, equipment, storage medium and program product of a hydropower station. Background Art

[0002] At present, new energy systems based on clean energy are becoming more and more common. Multi-energy complementarity, as an important strategy in the new energy system, enhances the peak-shaving capacity of the power system and the stability of energy supply by integrating the advantages of different energy forms.

[0003] Renewable energy sources such as wind power and photovoltaic power are generally considered to be unadjustable power sources. When hydropower stations participate in peak regulation, hydropower plays a vital role in regulation. In a typical cascade hydropower structure, a reservoir with a large regulation capacity is usually built upstream, while a counter-regulation reservoir is configured downstream to reduce the impact of unstable water flow in the upstream reservoir on the downstream. The method of cascade hydropower stations participating in peak regulation in related technologies only considers the peak regulation of the upstream power station and the counter-regulation of the downstream power station, but does not consider the counter-regulation needs of the downstream power station, and does not fully utilize the peak regulation function of the downstream power station, resulting in the regulation process being inaccurate, not in line with the actual situation, and lacking practicality. Summary of the invention

[0004] In view of this, the present invention provides a dispatching method, device, equipment, storage medium and program product for a hydropower station to solve the problem that the regulation process is not accurate enough and does not conform to the actual situation.

[0005] In a first aspect, the present invention provides a scheduling method for a hydropower station, comprising: predicting the power of wind power or photovoltaic power according to historical power data to obtain a power prediction value, and predicting the load of the hydropower station according to historical load data to obtain a load prediction value; the historical power data is used to characterize the wind power or photovoltaic power data of a preset time period, and the historical load data is used to characterize the load data of the hydropower station in the preset time period; performing a prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; performing a prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation; performing a scenario analysis on the first probability distribution function to generate a plurality of first power scenarios; performing a scenario analysis on the second probability distribution function to generate a plurality of first load scenarios; clustering multiple first power scenarios to obtain multiple first powers and multiple first occurrence probabilities corresponding to multiple target power typical scenarios; clustering multiple first load scenarios to obtain multiple first loads and multiple second occurrence probabilities corresponding to multiple target load typical scenarios; multiple first occurrence probabilities are used to characterize the probability of occurrence of multiple target power typical scenarios, and multiple second occurrence probabilities are used to characterize the probability of occurrence of multiple target load typical scenarios; based on multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, with the minimum expected maximum value of residual load as the first optimization objective function, the minimum expected fluctuation of downstream tail water level as the second optimization objective function, and the minimum expected fluctuation of downstream discharge flow as the third optimization objective function, determine the dispatching plan of the hydropower station.

[0006] The present invention first predicts the power of the hydropower station according to the power data of wind power or photovoltaic power in a preset time period to obtain a power prediction value, and predicts the load of the hydropower station according to the historical load data used to characterize the load data of the hydropower station in the preset time period to obtain a load prediction value, and predicts the future power and load according to the historical data, which is convenient for subsequent analysis based on the future power and load, and provides a data basis for the generation of subsequent hydropower station scheduling plans. The present invention performs a prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; performs a prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation. The present invention performs a prediction deviation analysis on the prediction value to find out the deviation in the power prediction value and the load prediction value, which is convenient for analyzing the cause of the deviation, reducing the occurrence of the same problem, and making the subsequent hydropower station scheduling plan more accurate. The present invention performs scenario analysis on the first probability distribution function to generate multiple first power scenarios; performs scenario analysis on the second probability distribution function to generate multiple first load scenarios; performs clustering processing on the multiple first power scenarios to obtain multiple first powers and multiple first occurrence probabilities corresponding to multiple target power typical scenarios; performs clustering processing on the multiple first load scenarios to obtain multiple first loads and multiple second occurrence probabilities corresponding to multiple target load typical scenarios. The present invention analyzes typical scenarios of power and load to make the subsequently generated hydropower station scheduling plan more in line with actual conditions. The present invention is based on multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, with the expected minimum maximum value of the remaining load as the first optimization objective function, the expected minimum fluctuation of the downstream tail water level as the second optimization objective function, and the expected minimum fluctuation of the downstream discharge flow as the third optimization objective function to determine the hydropower station scheduling plan. Compared with the related technology that only considers the peak regulation of the upstream power station, the present invention integrates the operating restrictions of the cascade downstream counter-regulation reservoirs into the water-wind-solar multi-energy complementary scheduling, considers the counter-regulation needs of the downstream hydropower station, makes full use of the peak-shaving role of the downstream hydropower station, takes into account the uncertainty of wind power, photovoltaic output and day-ahead load forecasting, generates a hydropower station scheduling plan, effectively coordinates the relationship between the operating status of the cascade counter-regulation reservoirs and the scheduling capacity, and maximizes the scheduling role of water-wind-solar multi-energy complementarity under the comprehensive consideration of the operating restrictions of the cascade counter-regulation reservoirs, so that the regulation process is more accurate, more in line with the actual situation, and improves practicality.

[0007] In an optional embodiment, based on multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, with the expected minimum maximum value of the remaining load as the first optimization objective function, the expected minimum fluctuation of the downstream tail water level as the second optimization objective function, and the expected minimum fluctuation of the downstream discharge flow as the third optimization objective function, a hydropower station scheduling plan is determined, including: constructing a multi-objective scheduling model according to the multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities; the optimization objectives of the multi-objective scheduling model include taking the expected minimum maximum value of the remaining load as the first optimization objective function, the expected minimum fluctuation of the downstream tail water level as the second optimization objective function, and the expected minimum fluctuation of the downstream discharge flow as the third optimization objective function; the input of the multi-objective scheduling model is multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, and the output of the multi-objective scheduling model is an optimization solution set; the multi-objective scheduling model is solved to obtain an optimization solution set; the optimization solution set is used to characterize the solution set of optimization data for scheduling the hydropower station; select target scheduling data in the optimization solution set, and generate a hydropower station scheduling plan according to the target scheduling data; wherein the target scheduling data is used to characterize the optimal scheduling data.

[0008] The present invention integrates the operating constraints of the downstream counter-regulation reservoirs of the cascade into the water-wind-solar multi-energy complementary scheduling, takes into account the uncertainty of wind power, photovoltaic output and day-ahead load forecasting, combines multiple typical scenarios, and constructs a multi-objective optimization scheduling model under the constraints of cascade counter-regulation. The multi-objective scheduling model is solved to obtain an optimized solution set, which includes multiple scheduling data with good scheduling effects. The optimal scheduling data is selected from the optimized solution set to generate a hydropower station scheduling plan, thereby improving the accuracy of water-wind-solar multi-energy complementary scheduling.

[0009] In an optional embodiment, the power of wind power or photovoltaic power is predicted based on historical power data to obtain a power prediction value, and the load of a hydropower station is predicted based on historical load data to obtain a load prediction value, including: inputting the historical power data into a trained power prediction model to obtain a power prediction value, the input of the power prediction model is the historical power data, and the output of the power prediction model is the power prediction value; inputting the historical load data into a trained load prediction model to obtain a load prediction value, the input of the load prediction model is the historical load data, and the output of the load prediction model is the load prediction value.

[0010] The present invention is based on a power prediction model and a load prediction model, and predicts power and load according to historical power data and historical load data, thereby ensuring the accuracy of power prediction values ​​and load prediction values, thereby improving the accuracy of subsequent generation of a hydropower station dispatching plan.

[0011] In an optional implementation, a prediction deviation analysis is performed on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; a prediction deviation analysis is performed on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation, including: taking the difference between the power measured value and the power prediction value as the first prediction deviation; performing a probability distribution analysis on the first prediction deviation to obtain a first probability distribution function; taking the difference between the load measured value and the load prediction value as the second prediction deviation; performing a probability distribution analysis on the second prediction deviation to obtain a second probability distribution function.

[0012] In an optional embodiment, a scenario analysis is performed on the first probability distribution function to generate a plurality of first typical power scenarios; a scenario analysis is performed on the second probability distribution function to generate a plurality of first typical load scenarios, including: performing random sampling a preset number of times on the first probability distribution function of the hydropower station scheduling scheme to obtain a plurality of first sampling results; generating a plurality of first typical power scenarios based on the plurality of first sampling results and the first prediction deviations corresponding to the plurality of first sampling results; performing random sampling a preset number of times on the second probability distribution function to obtain a plurality of second sampling results; generating a plurality of first typical load scenarios based on the plurality of second sampling results and the second prediction deviations corresponding to the plurality of first sampling results.

[0013] In an optional implementation, a plurality of first power scenarios are clustered to obtain a plurality of first powers and a plurality of first occurrence probabilities corresponding to a plurality of target power typical scenarios; a plurality of first load scenarios are clustered to obtain a plurality of first loads and a plurality of second occurrence probabilities corresponding to a plurality of target load typical scenarios, including: selecting a preset number of first power scenarios as initial power typical scenarios from a plurality of first power scenarios; calculating the distance from each first power scenario to the initial power typical scenario respectively to obtain a plurality of first target distances; assigning each first power scenario to the initial power typical scenario with the closest first target distance to obtain a plurality of first clustering groups; and grouping the first power scenarios according to the plurality of first clustering groups. The first power scenario in the group generates multiple target power typical scenarios; obtains multiple first powers and multiple first occurrence probabilities corresponding to the multiple target power typical scenarios; among the multiple first load scenarios, selects a preset number of first load scenarios as initial load typical scenarios; calculates the distance from each first load scenario to the initial load typical scenario respectively, and obtains multiple second target distances; assigns each first load scenario to the initial load typical scenario with the closest second target distance, and obtains multiple second clustering groups; generates multiple target load typical scenarios according to the first load scenarios in the multiple second clustering groups; obtains multiple first loads and multiple second occurrence probabilities corresponding to the multiple target load typical scenarios.

[0014] In an optional implementation, the first optimization objective function is:

[0015]

[0016] Among them, F1 represents the first optimization objective function, min represents the minimum value, max represents the maximum value, and p k represents the first occurrence probability corresponding to the target power typical scenario k or the second occurrence probability corresponding to the target load typical scenario k, K represents the total number of target power typical scenarios or the total number of target load typical scenarios, P k,t represents the first power of the target power typical scenario k in time period t or the first load of the target load typical scenario k in time period t; It represents the power generation of hydropower in time period t under the target power typical scenario k or target load typical scenario k. It represents the power generated by wind power in the target power typical scenario k or target load typical scenario k in time period t. It represents the power generation of photovoltaic in the target power typical scenario k or target load typical scenario k in the time period t.

[0017] In an optional implementation, the second optimization objective function is:

[0018]

[0019] Among them, F2 represents the second optimization objective function, min represents the minimum value, max represents the maximum value, and p k represents the first occurrence probability corresponding to the target power typical scenario k or the second occurrence probability corresponding to the target load typical scenario k, K represents the total number of target power typical scenarios or the total number of target load typical scenarios, Z k,t+1 It represents the water level in the time period t+1 under the target power typical scenario k or the target load typical scenario k, Z k,t It represents the water level in time period t under target power typical scenario k or target load typical scenario k.

[0020] In an optional implementation, the third optimization objective function is:

[0021]

[0022] Among them, F3 represents the third optimization objective function, min represents the minimum value, and p k represents the first occurrence probability corresponding to the target power typical scenario k or the second occurrence probability corresponding to the target load typical scenario k, Q k,t+1 represents the flow rate in time period t+1 under the target power typical scenario k or target load typical scenario k, Q k,t It represents the flow rate in time period t under the target power typical scenario k or the target load typical scenario k.

[0023] In a second aspect, the present invention provides a dispatching device for a hydropower station, comprising: a power and load prediction module, which is used to predict the power of wind power or photovoltaic power according to historical power data to obtain a power prediction value, and to predict the load of the hydropower station according to historical load data to obtain a load prediction value; the historical power data is used to characterize the wind power or photovoltaic power data of a preset time period, and the historical load data is used to characterize the load data of the hydropower station in a preset time period; a prediction deviation analysis module, which is used to perform prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; perform prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation; a scenario generation module, which is used to perform scenario analysis on the first probability distribution function to generate multiple first power scenarios; and perform scenario analysis on the second probability distribution function , generate multiple first load scenarios; a cluster analysis module, used to perform cluster analysis on multiple first power scenarios, and obtain multiple first powers and multiple first occurrence probabilities corresponding to multiple target power typical scenarios; perform cluster analysis on multiple first load scenarios, and obtain multiple first loads and multiple second occurrence probabilities corresponding to multiple target load typical scenarios; multiple first occurrence probabilities are used to characterize the probability of occurrence of multiple target power typical scenarios, and multiple second occurrence probabilities are used to characterize the probability of occurrence of multiple target load typical scenarios; an optimization scheduling module, used to determine the hydropower station scheduling plan based on multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, with the minimum expected maximum value of the remaining load as the first optimization objective function, the minimum expected fluctuation of the downstream tail water level as the second optimization objective function, and the minimum expected fluctuation of the downstream discharge flow as the third optimization objective function.

[0024] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the hydropower station scheduling method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0025] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for dispatching a hydropower station according to the first aspect or any corresponding embodiment thereof.

[0026] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions for causing a computer to execute the method for dispatching a hydropower station according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the related technologies, the drawings required for use in the specific embodiments or the related technical descriptions will be briefly introduced below. Obviously, the drawings described below are 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.

[0028] Figure 1 is a flow chart of a method for dispatching a hydropower station according to an embodiment of the present invention;

[0029] Figure 2 is a flow chart of another method for dispatching a hydropower station according to an embodiment of the present invention;

[0030] Figure 3 is a flow chart of another method for dispatching a hydropower station according to an embodiment of the present invention;

[0031] Figure 4 is a flow chart of a scheduling method for a hydropower station according to an embodiment of the present invention;

[0032] Figure 5 is a schematic diagram of power curves in different photovoltaic scenarios at different time periods according to an embodiment of the present invention;

[0033] Figure 6 is a schematic diagram of power curves in different wind power scenarios at different time periods according to an embodiment of the present invention;

[0034] Figure 7 is a schematic diagram of power curves under different load scenarios at different time periods according to an embodiment of the present invention;

[0035] Figure 8 is a structural block diagram of a dispatching device of a hydropower station according to an embodiment of the present invention;

[0036] Fig. 9 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0038] At present, the use of new energy, mainly clean energy, is becoming more and more widespread, and it is becoming more and more important to effectively improve the level of clean utilization of new energy. Multi-energy complementarity, as an important means of utilizing new energy, enhances the peak-shaving capacity of the power system and the stability of energy supply by integrating different forms of energy.

[0039] In the face of the problem of increasing peak-to-valley differences in power system loads in some regions, it is particularly important for hydropower stations to participate in peak load regulation of the power grid. Hydropower stations can use the regulation capacity of hydropower and cooperate with the intermittent characteristics of wind power and photovoltaic power generation to achieve more flexible and economical power production.

[0040] In a typical cascade hydropower structure, a reservoir with a large regulating capacity is usually built upstream, while a counter-regulating reservoir is configured downstream to reduce the impact of unstable water flow in the upstream reservoir on the downstream. For a water-wind-solar multi-energy complementary system with a typical cascade reservoir, how to give full play to the downstream peak-shaving capacity has become a difficult problem that needs to be solved urgently.

[0041] An embodiment of the present invention provides a scheduling method for a hydropower station, which integrates the operating restrictions of the downstream cascade counter-regulation reservoirs into the peak-shaving scheduling of a water-wind-solar multi-energy complementary system to achieve the effect of maximizing the peak-shaving effect.

[0042] According to an embodiment of the present invention, a method for dispatching a hydropower station is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0043] In this embodiment, a method for dispatching a hydropower station is provided, which can be used in a computer device. Figure 1 is a flow chart of a method for dispatching a hydropower station according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0044] Step S101, predicting the power of wind power or photovoltaic power according to historical power data to obtain a power prediction value, and predicting the load of the hydropower station according to historical load data to obtain a load prediction value; the historical power data is used to characterize the wind power or photovoltaic power data of a preset time period, and the historical load data is used to characterize the load data of the hydropower station in a preset time period.

[0045] In some optional embodiments, the power of wind power or photovoltaic power is predicted based on historical power data to obtain a power prediction value, and the load of a hydropower station is predicted based on historical load data to obtain a load prediction value, including: inputting the historical power data into a trained power prediction model to obtain a power prediction value, the input of the power prediction model is the historical power data, and the output of the power prediction model is the power prediction value; inputting the historical load data into a trained load prediction model to obtain a load prediction value, the input of the load prediction model is the historical load data, and the output of the load prediction model is the load prediction value.

[0046] In some optional implementations, the wind power and / or photovoltaic power is predicted based on historical power data to obtain a power prediction value, which may be a wind power and / or photovoltaic power prediction value.

[0047] Among them, the power forecast value is the wind power or photovoltaic power forecast value. Since the change of wind power or photovoltaic power is affected by meteorological factors, the unit's own characteristics and unit status, the historical power data includes historical wind power or photovoltaic power data, meteorological factors such as wind speed, radiation intensity and temperature at the time corresponding to the power data, and the unit status data of the hydropower station. The change of power grid load is affected by its own characteristics, meteorological factors and date type, so the historical load data includes historical load, meteorological and working day data.

[0048] In some optional embodiments, the power prediction model is a wind power / photovoltaic short-term power prediction model constructed based on a long short-term memory network (LSTM). The wind power / photovoltaic short-term power prediction model can realize short-term prediction of wind power or photovoltaic power. For example, the wind power or photovoltaic power of the next day is predicted based on the historical wind power or photovoltaic power data of the previous day.

[0049] In some optional implementations, the load forecasting model is a day-ahead load forecasting model constructed based on a long short-term memory network (LSTM). The day-ahead load forecasting model can realize the prediction of the day-ahead load. For example, the load of the next day is predicted using the historical load data of the previous day.

[0050] In some optional implementations, the dispatching method of the hydropower station further includes: predicting the power of wind power or photovoltaic power according to historical runoff data to obtain a runoff prediction value.

[0051] Among them, the power of wind power or photovoltaic power is predicted according to historical runoff data to obtain a runoff prediction value, including: inputting the historical runoff data into a trained runoff prediction model to obtain a runoff prediction value, the input of the runoff prediction model is the historical runoff data, and the output of the runoff prediction model is the runoff prediction value.

[0052] Among them, runoff changes are affected by hydrological and meteorological factors. Therefore, historical runoff data includes historical runoff, rainfall, evaporation and other data. The runoff prediction model is a day-ahead runoff prediction model based on the Long Short-Term Memory (LSTM) network. The day-ahead runoff prediction model is used to achieve day-ahead runoff prediction. For example, the runoff of the next day is predicted using the historical runoff data of the previous day.

[0053] Step S102, performing a prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; performing a prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation.

[0054] Among them, since the power prediction value is a wind power or photovoltaic power prediction value, a prediction deviation analysis is performed on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation, including: performing a prediction deviation analysis on the wind power prediction value and the wind power measured value to obtain the first probability distribution function corresponding to the first prediction deviation or performing a prediction deviation analysis on the photovoltaic power prediction value and the photovoltaic power measured value to obtain the first probability distribution function corresponding to the first prediction deviation.

[0055] In some optional implementations, the first prediction deviation is the difference between the power prediction value and the power measured value, and the formula for the first prediction deviation is:

[0056] E1=P 1 -P 2 ;

[0057] Among them, E1 is the first prediction deviation, P 1 is the predicted value of wind power or photovoltaic power, P 2 is the actual measured value of wind power or photovoltaic power.

[0058] Wherein, a kernel density estimation method is used to calculate a first probability distribution function corresponding to the first prediction deviation.

[0059] In some optional implementations, the second prediction deviation is the difference between the load prediction value and the load measured value, and the formula for the second prediction deviation is:

[0060] E2=P 3 -P 4 ;

[0061] Among them, E2 is the second prediction deviation, P 3 is the load forecast value, P 4 is the actual measured value of load.

[0062] Wherein, a kernel density estimation method is used to calculate the second probability distribution function corresponding to the second prediction deviation.

[0063] Step S103, performing scenario analysis on the first probability distribution function to generate a plurality of first power scenarios; performing scenario analysis on the second probability distribution function to generate a plurality of first load scenarios.

[0064] In some optional embodiments, multiple random samplings are performed in the first probability distribution function to obtain a first sampling result, and the first sampling result is added to the power prediction value to obtain a plurality of first power scenarios; multiple random samplings are performed in the second probability distribution function to obtain a second sampling result, and the second sampling result is added to the load prediction value to obtain a plurality of first load scenarios.

[0065] Exemplarily, the formula for the power corresponding to the first power scenario is:

[0066]

[0067] in, is the power in time period t under the i-th first power scenario; is the wind power or photovoltaic power forecast value in time period t, is the first prediction deviation of the vth sampling in time period t.

[0068] The load formula corresponding to the first load scenario is:

[0069]

[0070] in, is the load in time period t under the i-th first load scenario; is the load forecast value for time period t, is the second prediction deviation of the vth sample in time period t.

[0071] Step S104, clustering the multiple first power scenarios to obtain multiple first powers and multiple first occurrence probabilities corresponding to the multiple target power typical scenarios; clustering the multiple first load scenarios to obtain multiple first loads and multiple second occurrence probabilities corresponding to the multiple target load typical scenarios; the multiple first occurrence probabilities are used to characterize the probability of occurrence of the multiple target power typical scenarios, and the multiple second occurrence probabilities are used to characterize the probability of occurrence of the multiple target load typical scenarios.

[0072] In some optional embodiments, a preset number of scenes are randomly selected from a plurality of first power scenes as initial typical scenes, the distance from each first power scene to the initial typical scene is calculated respectively, and the first power scenes are assigned to the initial typical scenes with the closest distance, the average of the powers corresponding to all first power scenes in each initial typical scene is taken as the target power typical scene, and the target power typical scene to which each first power scene belongs is reallocated until the target power typical scene no longer changes, thereby obtaining multiple first powers corresponding to the final K target power typical scenes and multiple first occurrence probabilities of K1 target power typical scenes.

[0073] Exemplarily, the formula for respectively calculating the distance from each first power scenario to the target power typical scenario is:

[0074]

[0075] in, is the first distance from the first power scenario i to the target power typical scenario k, is the power in time period t under the i-th first power scenario, is the power of the target power typical scenario k in time period t, N1 is the total number of the first power scenarios, and K1 is the total number of the target power typical scenarios.

[0076] The power calculation formula of the target power typical scenario k in time period t is:

[0077]

[0078] in, is the power of the target power typical scenario k in time period t, is the number of the first power scenarios belonging to the target power typical scenario k, is the power of the target power typical scenario k in time period t.

[0079] In some optional implementations, a calculation formula for a first occurrence probability of a first power scenario target power typical scenario is:

[0080]

[0081] in, is the first occurrence probability, is the number of first power scenarios belonging to the target power typical scenario k, and N1 is the total number of first power scenarios.

[0082] In some optional embodiments, a preset number of scenes are randomly selected from a plurality of first load scenes as initial typical scenes, the distance from each first load scene to the initial typical scene is calculated respectively, and the first load scene is assigned to the initial typical scene with the nearest distance, the average of the loads corresponding to all first load scenes in each initial typical scene is taken as the target load typical scene, and the target load typical scene to which each first load scene belongs is reallocated until the target load typical scene no longer changes, thereby obtaining multiple first loads corresponding to the final K target load typical scenes and multiple second occurrence probabilities of K2 target load typical scenes.

[0083] Exemplarily, the formula for calculating the distance from each first load scenario to the target load typical scenario is:

[0084]

[0085] in, is the second distance from the first load scenario i to the target load typical scenario k, is the load in time period t under the i-th first load scenario, is the load of the target load typical scenario k in time period t, N2 is the total number of the first load scenario, and K2 is the total number of the target load typical scenario.

[0086] Among them, the calculation formula of the load of the target load typical scenario k in time period t is:

[0087]

[0088] in, is the load of the target load typical scenario k in time period t, is the number of the first load scenarios belonging to the target load typical scenario k, is the load of the typical scenario k of the target load in time period t.

[0089] In some optional implementations, the calculation formula for the second occurrence probability of the first load scenario target load typical scenario is:

[0090]

[0091] in, is the second occurrence probability, is the number of first load scenarios belonging to the target load typical scenario k, and N2 is the total number of first load scenarios.

[0092] Step S105, based on multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, determine the hydropower station scheduling plan with the expected minimum maximum value of the remaining load as the first optimization objective function, the expected minimum fluctuation of the downstream tail water level as the second optimization objective function, and the expected minimum fluctuation of the downstream discharge flow as the third optimization objective function.

[0093] The embodiment of the present invention dispatches a hydropower station according to a hydropower station dispatching plan, specifically refers to adjusting the parameters of the hydropower station according to the target dispatching data in the hydropower station dispatching plan. Exemplarily, the target dispatching data can be hydropower generation power, wind power generation power, photovoltaic power generation power, water level and flow rate, etc.

[0094] In some optional implementations, the first optimization objective function is:

[0095]

[0096] Among them, F1 represents the first optimization objective function, min represents the minimum value, max represents the maximum value, and p k represents the first occurrence probability corresponding to the target power typical scenario k or the second occurrence probability corresponding to the target load typical scenario k, K represents the total number of target power typical scenarios or the total number of target load typical scenarios, P k,t represents the first power of the target power typical scenario k in time period t or the first load of the target load typical scenario k in time period t; It represents the power generation of hydropower in time period t under the target power typical scenario k or target load typical scenario k. It represents the power generated by wind power in the target power typical scenario k or target load typical scenario k in time period t. It represents the power generation of photovoltaic in the target power typical scenario k or target load typical scenario k in the time period t.

[0097] In some optional implementations, the second optimization objective function is:

[0098]

[0099] Among them, F2 represents the second optimization objective function, min represents the minimum value, max represents the maximum value, and p k represents the first occurrence probability corresponding to the target power typical scenario k or the second occurrence probability corresponding to the target load typical scenario k, K represents the total number of target power typical scenarios or the total number of target load typical scenarios, Z k,t+1 It represents the water level in the time period t+1 under the target power typical scenario k or the target load typical scenario k, Z k,t It represents the water level in time period t under target power typical scenario k or target load typical scenario k.

[0100] In some optional implementations, the third optimization objective function is:

[0101]

[0102] Among them, F3 represents the third optimization objective function, min represents the minimum value, and p k represents the first occurrence probability corresponding to the target power typical scenario k or the second occurrence probability corresponding to the target load typical scenario k, Q k,t+1 represents the flow rate in time period t+1 under the target power typical scenario k or target load typical scenario k, Q k,t It represents the flow rate in time period t under the target power typical scenario k or the target load typical scenario k.

[0103] The scheduling method of the hydropower station provided in this embodiment first predicts the power of the hydropower station according to the power data of wind power or photovoltaic power in a preset time period to obtain a power prediction value, and predicts the load of the hydropower station according to the historical load data used to characterize the load data of the hydropower station in the preset time period to obtain a load prediction value, and predicts the future power and load according to the historical data, which is convenient for subsequent analysis based on the future power and load, and provides a data basis for the generation of subsequent hydropower station scheduling plans. The embodiment of the present invention performs a prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; performs a prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation. The embodiment of the present invention performs a prediction deviation analysis on the prediction value to find out the deviation in the power prediction value and the load prediction value, which is convenient for analyzing the cause of the deviation, reducing the occurrence of the same problem, and making the subsequent hydropower station scheduling plan more accurate. The embodiment of the present invention performs scenario analysis on the first probability distribution function to generate multiple first power scenarios; performs scenario analysis on the second probability distribution function to generate multiple first load scenarios; performs clustering processing on the multiple first power scenarios to obtain multiple first powers and multiple first occurrence probabilities corresponding to multiple target power typical scenarios; performs clustering processing on the multiple first load scenarios to obtain multiple first loads and multiple second occurrence probabilities corresponding to multiple target load typical scenarios. The present invention analyzes typical scenarios of power and load to make the subsequently generated hydropower station scheduling plan more in line with actual conditions. The embodiment of the present invention determines the hydropower station scheduling plan based on multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, with the expected minimum maximum value of the remaining load as the first optimization objective function, the expected minimum fluctuation of the downstream tail water level as the second optimization objective function, and the expected minimum fluctuation of the downstream discharge flow as the third optimization objective function. Compared with the related art that only considers the peak regulation of the upstream power station, the embodiment of the present invention integrates the operating restrictions of the cascade downstream counter-regulation reservoirs into the water-wind-solar multi-energy complementary scheduling, considers the counter-regulation needs of the downstream hydropower station, makes full use of the peak-shaving role of the downstream hydropower station, takes into account the uncertainty of wind power, photovoltaic output and day-ahead load forecasting, generates a hydropower station scheduling plan, effectively coordinates the relationship between the operating status of the cascade counter-regulation reservoirs and the scheduling capacity, and maximizes the scheduling role of water-wind-solar multi-energy complementarity while comprehensively considering the operating restrictions of the cascade counter-regulation reservoirs, so that the regulation process is more accurate, more in line with the actual situation, and improves practicality.

[0104] In this embodiment, a method for dispatching a hydropower station is provided, which can be used in a computer device. Figure 2 is a flow chart of another hydropower station scheduling method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0105] Step S201: predict the power of wind power or photovoltaic power according to historical power data to obtain a power prediction value, and predict the load of the hydropower station according to historical load data to obtain a load prediction value; the historical power data is used to represent the wind power or photovoltaic power data in a preset time period, and the historical load data is used to represent the load data of the hydropower station in a preset time period. For details, please refer to Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0106] Step S202: Perform a prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; perform a prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation. Figure 1 Step S102 of the illustrated embodiment will not be described in detail here.

[0107] Step S203: Perform scenario analysis on the first probability distribution function to generate multiple first power scenarios; perform scenario analysis on the second probability distribution function to generate multiple first load scenarios. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0108] Step S204: Cluster multiple first power scenarios to obtain multiple first powers and multiple first occurrence probabilities corresponding to multiple target power typical scenarios; cluster multiple first load scenarios to obtain multiple first loads and multiple second occurrence probabilities corresponding to multiple target load typical scenarios; multiple first occurrence probabilities are used to characterize the probability of occurrence of multiple target power typical scenarios, and multiple second occurrence probabilities are used to characterize the probability of occurrence of multiple target load typical scenarios. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0109] Step S205, based on multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, with the expected minimum maximum value of the remaining load as the first optimization objective function, the expected minimum fluctuation of the downstream water level as the second optimization objective function, and the expected minimum fluctuation of the downstream discharge flow as the third optimization objective function, determine the hydropower station scheduling plan, and schedule the hydropower station according to the hydropower station scheduling plan.

[0110] Specifically, the above step S205 includes:

[0111] Step S2051, constructing a multi-objective scheduling model according to multiple first powers, multiple first occurrence probabilities, multiple first loads, and multiple second occurrence probabilities.

[0112] Among them, the optimization objectives of the multi-objective scheduling model include minimizing the expected maximum value of the remaining load as the first optimization objective function, minimizing the expected fluctuation of the downstream water level as the second optimization objective function, and minimizing the expected fluctuation of the downstream discharge flow as the third optimization objective function; the input of the multi-objective scheduling model is multiple first powers, multiple first occurrence probabilities, multiple first loads, and multiple second occurrence probabilities, and the output of the multi-objective scheduling model is the optimization solution set.

[0113] In some optional embodiments, based on multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, the expected minimum maximum value of the remaining load is taken as the first optimization objective function to construct a short-term peak-shaving model of the water-wind-solar multi-energy complementary system, the expected minimum downstream tail water level fluctuation is taken as the second optimization objective function to construct a downstream tail water level fluctuation model, and the expected minimum downstream discharge flow volatility is taken as the third optimization objective function to construct a flow limitation model; the short-term peak-shaving model of the water-wind-solar multi-energy complementary system, the downstream tail water level fluctuation model and the flow limitation model are integrated to obtain a multi-objective scheduling model.

[0114] Among them, the constraints of the short-term peak-shaving model of the water-wind-solar multi-energy complementary system include: power station output limitation, reservoir water level limitation, downstream flow limitation and water balance equation, etc.; there are counter-regulation limitations in the downstream, and it is necessary to build a downstream tail water level fluctuation model based on the short-term peak-shaving model of the water-wind-solar multi-energy complementary system with the goal of minimizing the expected fluctuation of the downstream tail water level; downstream counter-regulation has restrictions on the downstream flow, therefore, a flow limitation model is constructed with the goal of minimizing the expected fluctuation of the downstream flow.

[0115] Step S2052, solving the multi-objective scheduling model to obtain an optimization solution set; the optimization solution set is used to characterize the solution set of optimization data for scheduling the hydropower station.

[0116] Among them, the embodiment of the present invention adopts a decomposition-based multi-objective evolutionary algorithm (MOEA / D algorithm) to solve the multi-objective scheduling model and obtain a Pareto optimization solution set. The MOEA / D algorithm decomposes the multi-objective optimization problem into a set of single-objective optimization sub-problems, and approximates the solution set of the original multi-objective optimization problem by solving these sub-problems.

[0117] Step S2053, selecting target scheduling data from the optimization solution set, and generating a hydropower station scheduling plan based on the target scheduling data; wherein the target scheduling data is used to characterize the optimal scheduling data.

[0118] The embodiment of the present invention integrates the operating constraints of the downstream counter-regulation reservoirs of the cascade into the water-wind-solar multi-energy complementary scheduling, takes into account the uncertainty of wind power, photovoltaic output and day-ahead load forecasting, combines multiple typical scenarios, and constructs a multi-objective optimization scheduling model under the constraints of cascade counter-regulation. The multi-objective scheduling model is solved to obtain an optimized solution set, which includes multiple scheduling data with good scheduling effects. The optimal scheduling data is selected from the optimized solution set to generate a hydropower station scheduling plan, thereby improving the accuracy of water-wind-solar multi-energy complementary scheduling.

[0119] In this embodiment, a method for dispatching a hydropower station is provided, which can be used in a computer device. Figure 3 is a flow chart of another method for dispatching a hydropower station according to an embodiment of the present invention, such as Figure 3 As shown, the process includes the following steps:

[0120] Step S301: predict the power of wind power or photovoltaic power according to historical power data to obtain a power prediction value, and predict the load of the hydropower station according to historical load data to obtain a load prediction value; the historical power data is used to represent the wind power or photovoltaic power data in a preset time period, and the historical load data is used to represent the load data of the hydropower station in a preset time period. For details, please refer to Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0121] Step S302, performing a prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; performing a prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation.

[0122] Specifically, the above step S302 includes:

[0123] Step S3021: taking the difference between the actual power value and the predicted power value as a first prediction deviation; performing probability distribution analysis on the first prediction deviation to obtain a first probability distribution function.

[0124] The probability distribution analysis may be to use a kernel density estimation method to calculate a first probability distribution function corresponding to the first prediction deviation.

[0125] Step S3022: taking the difference between the actual load value and the predicted load value as the second prediction deviation; performing probability distribution analysis on the second prediction deviation to obtain a second probability distribution function.

[0126] The probability distribution analysis may be to use a kernel density estimation method to calculate a second probability distribution function corresponding to the second prediction deviation.

[0127] Step S303: Perform scenario analysis on the first probability distribution function to generate multiple first power scenarios; perform scenario analysis on the second probability distribution function to generate multiple first load scenarios.

[0128] Specifically, the above step S303 includes:

[0129] Step S3031, performing random sampling of a preset number of times on the first probability distribution function to obtain a plurality of first sampling results; and generating a plurality of first power typical scenarios according to the plurality of first sampling results and the first prediction deviations corresponding to the plurality of first sampling results.

[0130] Among them, multiple first sampling results and the first prediction deviation corresponding to each sampling result are added together to obtain multiple first power typical scenarios.

[0131] Among them, the preset number of times can be set according to actual conditions.

[0132] Step S3032, performing random sampling of a preset number of times on the second probability distribution function to obtain a plurality of second sampling results; and generating a plurality of first load typical scenarios according to second prediction deviations corresponding to the plurality of second sampling results and the plurality of first sampling results.

[0133] Among them, multiple second sampling results and the second prediction deviation corresponding to each sampling result are added together to obtain multiple first load typical scenarios.

[0134] Among them, the preset number of times can be set according to actual conditions.

[0135] Step S304, clustering the multiple first power scenarios to obtain multiple first powers and multiple first occurrence probabilities corresponding to the multiple target power typical scenarios; clustering the multiple first load scenarios to obtain multiple first loads and multiple second occurrence probabilities corresponding to the multiple target load typical scenarios; the multiple first occurrence probabilities are used to characterize the probability of occurrence of the multiple target power typical scenarios, and the multiple second occurrence probabilities are used to characterize the probability of occurrence of the multiple target load typical scenarios.

[0136] Specifically, the above step S304 includes:

[0137] Step S3041, among multiple first power scenarios, select a preset number of first power scenarios as initial power typical scenarios; calculate the distance from each first power scenario to the initial power typical scenario respectively, and obtain multiple first target distances; assign each first power scenario to the initial power typical scenario with the closest first target distance, and obtain multiple first cluster groups; generate multiple target power typical scenarios based on the first power scenarios in the multiple first cluster groups; obtain multiple first powers and multiple first occurrence probabilities corresponding to the multiple target power typical scenarios.

[0138] The average of the first power scenarios in the multiple first cluster groups is taken as the target power typical scenario to obtain multiple target power typical scenarios.

[0139] In some optional embodiments, a preset number of scenes are randomly selected from multiple first power scenes as initial typical scenes, the distance from each first power scene to the initial typical scene is calculated respectively, and the first power scene is assigned to the initial typical scene with the closest distance, the average of the powers corresponding to all first power scenes in each initial typical scene is taken as the target power typical scene, and the target power typical scene to which each first power scene belongs is reallocated until the target power typical scene no longer changes, thereby obtaining multiple first powers corresponding to the final K target power typical scenes and multiple first occurrence probabilities of the target power typical scenes.

[0140] Step S3042, among multiple first load scenarios, select a preset number of first load scenarios as initial load typical scenarios; calculate the distance from each first load scenario to the initial load typical scenario respectively, and obtain multiple second target distances; assign each first load scenario to the initial load typical scenario with the closest second target distance, and obtain multiple second clustering groups; generate multiple target load typical scenarios based on the first load scenarios in the multiple second clustering groups; obtain multiple first loads and multiple second occurrence probabilities corresponding to the multiple target load typical scenarios.

[0141] The average of the first load scenarios in the multiple second cluster groups is used as the target load typical scenario to obtain multiple target load typical scenarios.

[0142] In some optional embodiments, a preset number of scenes are randomly selected from a plurality of first load scenes as initial typical scenes, the distance from each first load scene to the initial typical scene is calculated respectively, and the first load scenes are assigned to the initial typical scenes with the closest distance, the average of the loads corresponding to all first load scenes in each initial typical scene is taken as the target load typical scene, and the target load typical scenes to which each first load scene belongs are reallocated until the target load typical scene no longer changes, thereby obtaining multiple first loads corresponding to the final K target load typical scenes and multiple second occurrence probabilities of the target load typical scenes.

[0143] Step S305: Based on the multiple first powers, multiple first occurrence probabilities, multiple first loads, and multiple second occurrence probabilities, the hydropower station dispatching plan is determined with the expected minimum of the maximum value of the remaining load as the first optimization objective function, the expected minimum of the downstream tail water level fluctuation as the second optimization objective function, and the expected minimum of the downstream discharge flow fluctuation as the third optimization objective function. Figure 1Step S105 of the illustrated embodiment will not be described in detail here.

[0144] In this embodiment, a method for dispatching a hydropower station is provided, which can be used in a computer device. Figure 4 FIG. 1 is a flow chart of a scheduling method for a hydropower station according to another embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0145] Step S401: forecasting wind power or photovoltaic short-term power, day-ahead load curve, and day-ahead runoff.

[0146] In some optional implementations, the change of wind power / photovoltaic power is affected by meteorological factors, the characteristics and status of the unit itself, etc. Meteorological factors such as wind speed, radiation intensity, temperature and unit status data corresponding to the wind power / photovoltaic power data are collected as model input factors, and a wind power / photovoltaic short-term power prediction model is constructed based on the LSTM network to achieve short-term prediction of wind and solar power.

[0147] In some optional implementations, the change of wind power / photovoltaic power is affected by meteorological factors, the characteristics and status of the unit itself, etc. Meteorological factors such as wind speed, radiation intensity, temperature and unit status data corresponding to the wind power / photovoltaic power data are collected as model input factors, and a wind power / photovoltaic short-term power prediction model is constructed based on the LSTM network to achieve short-term prediction of wind and solar power.

[0148] In some optional implementations, runoff changes are affected by hydrological and meteorological factors. Historical runoff, rainfall, evaporation and other data are collected as key input factors of the model, and a day-ahead runoff prediction model is constructed based on the LSTM network to achieve day-ahead runoff prediction.

[0149] Step S402: uncertainty analysis of wind power or photovoltaic prediction error and load prediction error.

[0150] In some optional implementations, the prediction deviation is calculated based on the wind power / photovoltaic power prediction results and the actual observation results, and the probability distribution function of the prediction deviation is calculated using a kernel density estimation method.

[0151] In some optional implementations, the prediction deviation is calculated based on the load prediction results and the actual observation results, and the probability distribution function of the prediction deviation is calculated using a kernel density estimation method.

[0152] Step S403: Generate typical scenarios of wind power or photovoltaic power and load.

[0153] In some optional implementations, multiple random samplings are performed in the probability distribution of the prediction deviation, and the sampling results are added to the prediction results to obtain multiple power / load scenarios that may occur at that moment.

[0154] In some optional implementations, the generated scenes are clustered and reduced, and an initial scene is randomly selected from multiple scenes as the initial typical scene. The distance from each scene to each typical scene is calculated respectively, and it is assigned to the typical scene with the closest distance. After all scenes are assigned, the mean of all scenes in the typical scene at each time is used as the new typical scene, and the typical scene to which each scene belongs is reallocated until the typical scene does not change. Finally, multiple new typical scenes and their occurrence probabilities are obtained.

[0155] Step S404: constructing a short-term multi-objective scheduling model.

[0156] In some optional embodiments, on the basis of generating typical power / load scenarios, a short-term peak-shaving model of the water-wind-solar multi-energy complementary system is constructed with the goal of minimizing the expected maximum value of the residual load. There are counter-regulation restrictions in the downstream, and it is necessary to construct a downstream tail water level fluctuation model based on the short-term peak-shaving model with the goal of minimizing the expected fluctuation of the downstream tail water level. The downstream counter-regulation has restrictions on the downstream flow, and a flow limitation model is constructed with the goal of minimizing the expected volatility of the downstream flow. A short-term multi-objective scheduling model is constructed by the short-term peak-shaving model of the water-wind-solar multi-energy complementary system, the downstream tail water level fluctuation model and the flow limitation model.

[0157] Step S405: a multi-objective intelligent optimization algorithm is used to perform optimization and solution.

[0158] In some optional implementations, the MOEA / D algorithm is used to solve the multi-objective scheduling model, and finally a set of Pareto optimal solution sets of the multi-objective scheduling model is obtained, and a hydropower station scheduling plan is generated according to the Pareto optimal solution set.

[0159] For example, Figure 5 As shown, it is a schematic diagram of power curves under different photovoltaic scenes at different time periods. The power corresponding to different photovoltaic scenes at different time periods is different. The power corresponding to photovoltaic scene 3 is greater than the power corresponding to photovoltaic scene 1, which is greater than the power corresponding to photovoltaic scene 2. Figure 6 As shown, it is a schematic diagram of power curves under different wind power scenarios at different time periods. The power corresponding to different time periods under different wind power scenarios is different. The power corresponding to wind power scenario 6 is greater than the power corresponding to wind power scenario 5, which is greater than the power corresponding to wind power scenario 4. Figure 7 As shown, it is a schematic diagram of power curves under different load scenarios at different time periods. The loads corresponding to different time periods under different load scenarios are different. The load corresponding to load scenario 9 is greater than the load corresponding to load scenario 7, which is greater than the load corresponding to load scenario 8.

[0160] In this embodiment, a dispatching device for a hydropower station is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0161] This embodiment provides a dispatching device for a hydropower station, such as Figure 8 As shown, including:

[0162] The power and load forecasting module 801 is used to forecast the power of wind power or photovoltaic power according to historical power data to obtain a power forecast value, and to forecast the load of a hydropower station according to historical load data to obtain a load forecast value; the historical power data is used to characterize the wind power or photovoltaic power data in a preset time period, and the historical load data is used to characterize the load data of a hydropower station in a preset time period.

[0163] The prediction deviation analysis module 802 is used to perform prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; and to perform prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation.

[0164] The scenario generation module 803 is used to perform scenario analysis on the first probability distribution function to generate multiple first power scenarios; and to perform scenario analysis on the second probability distribution function to generate multiple first load scenarios.

[0165] The cluster analysis module 804 is used to perform cluster analysis on multiple first power scenarios to obtain multiple first powers and multiple first occurrence probabilities corresponding to multiple target power typical scenarios; perform cluster analysis on multiple first load scenarios to obtain multiple first loads and multiple second occurrence probabilities corresponding to multiple target load typical scenarios; the multiple first occurrence probabilities are used to characterize the probability of occurrence of multiple target power typical scenarios, and the multiple second occurrence probabilities are used to characterize the probability of occurrence of multiple target load typical scenarios.

[0166] The optimization scheduling module 805 is used to determine the hydropower station scheduling plan based on multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, with the expected minimum maximum value of the remaining load as the first optimization objective function, the expected minimum fluctuation of the downstream water level as the second optimization objective function, and the expected minimum fluctuation of the downstream discharge flow as the third optimization objective function.

[0167] In some optional implementations, the optimization scheduling module 805 includes:

[0168] A model building unit is used to build a multi-objective scheduling model according to multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities; the optimization objectives of the multi-objective scheduling model include taking the expected minimum maximum value of the remaining load as the first optimization objective function, taking the expected minimum fluctuation of the downstream tail water level as the second optimization objective function, and taking the expected minimum fluctuation of the downstream discharge flow as the third optimization objective function; the input of the multi-objective scheduling model is multiple first powers, multiple first occurrence probabilities, multiple first loads and multiple second occurrence probabilities, and the output of the multi-objective scheduling model is an optimization solution set.

[0169] The model solving unit is used to solve the multi-objective scheduling model to obtain an optimization solution set; the optimization solution set is used to represent the solution set of optimization data for scheduling the hydropower station.

[0170] The hydropower station dispatching scheme generating unit is used to select target dispatching data in the optimization solution set, and generate a hydropower station dispatching scheme according to the target dispatching data; wherein the target dispatching data is used to characterize the optimal dispatching data.

[0171] In some optional implementations, the power and load prediction module 801 includes:

[0172] The power prediction unit is used to input the historical power data into the trained power prediction model to obtain the power prediction value. The input of the power prediction model is the historical power data, and the output of the power prediction model is the power prediction value.

[0173] The load prediction unit is used to input historical load data into the trained load prediction model to obtain a load prediction value. The input of the load prediction model is the historical load data, and the output of the load prediction model is the load prediction value.

[0174] In some optional implementations, the prediction deviation analysis module 802 includes:

[0175] The first prediction deviation analysis unit is used to take the difference between the actual power value and the predicted power value as the first prediction deviation; perform probability distribution analysis on the first prediction deviation to obtain a first probability distribution function.

[0176] The second prediction deviation analysis unit is used to take the difference between the load measured value and the load predicted value as the second prediction deviation; perform probability distribution analysis on the second prediction deviation to obtain a second probability distribution function.

[0177] In some optional implementations, the scene generation module 803 includes:

[0178] The power scenario generation unit is used to randomly sample the first probability distribution function of the hydropower station scheduling plan a preset number of times to obtain multiple first sampling results; and generate multiple first power typical scenarios based on the multiple first sampling results and the first prediction deviations corresponding to the multiple first sampling results.

[0179] The load scenario generating unit is used to perform random sampling of a preset number of times on the second probability distribution function to obtain a plurality of second sampling results; and to generate a plurality of first load typical scenarios according to the second prediction deviations corresponding to the plurality of second sampling results and the plurality of first sampling results.

[0180] In some optional implementations, the cluster analysis module 804 includes:

[0181] The first cluster analysis unit is used to select a preset number of first power scenarios as initial power typical scenarios from multiple first power scenarios; calculate the distance from each first power scenario to the initial power typical scenario respectively to obtain multiple first target distances; assign each first power scenario to the initial power typical scenario with the closest first target distance to obtain multiple first cluster groups; generate multiple target power typical scenarios based on the first power scenarios in the multiple first cluster groups; obtain multiple first powers and multiple first occurrence probabilities corresponding to the multiple target power typical scenarios.

[0182] The second clustering analysis unit is used to select a preset number of first load scenarios as initial load typical scenarios from multiple first load scenarios; calculate the distance from each first load scenario to the initial load typical scenario to obtain multiple second target distances; assign each first load scenario to the initial load typical scenario with the closest second target distance to obtain multiple second clustering groups; generate multiple target load typical scenarios based on the first load scenarios in the multiple second clustering groups; obtain multiple first loads and multiple second occurrence probabilities corresponding to the multiple target load typical scenarios.

[0183] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0184] The dispatching device of the hydropower station in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, or other devices that can provide the above functions.

[0185] The embodiment of the present invention also provides a computer device having the above Figure 8 The dispatching device of the hydropower station is shown.

[0186] See also Fig. 9 , Fig. 9 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Fig. 9 As shown, the computer device includes: one or more processors 910, memory 920, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Fig. 9 A processor 910 is taken as an example.

[0187] The processor 910 may be a central processing unit, a network processor or a combination thereof. The processor 910 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable logic gate array, a general purpose array logic or any combination thereof.

[0188] The memory 920 stores instructions executable by at least one processor 910, so that the at least one processor 910 executes the method shown in the above embodiment.

[0189] The memory 920 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 920 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 920 may optionally include a memory remotely arranged relative to the processor 910, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0190] The memory 920 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 920 may also include a combination of the above types of memory.

[0191] The computer device also includes a communication interface 930, which is used for the computer device to communicate with other devices or a communication network.

[0192] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0193] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of computer program instructions in computer-readable media includes, but is not limited to, source files, executable files, installation package files, etc., and accordingly, the way in which computer program instructions are executed by a computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0194] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for dispatching a hydropower station, characterized in that: The method comprises: Predicting the power of wind power or photovoltaic power according to historical power data to obtain a power prediction value, and predicting the load of the hydropower station according to historical load data to obtain a load prediction value; the historical power data is used to characterize the wind power or photovoltaic power data in a preset time period, and the historical load data is used to characterize the load data of the hydropower station in a preset time period; Performing a prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; performing a prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation; Performing scenario analysis on the first probability distribution function to generate multiple first power scenarios; performing scenario analysis on the second probability distribution function to generate multiple first load scenarios; Clustering the multiple first power scenarios to obtain multiple first powers and multiple first occurrence probabilities corresponding to multiple target power typical scenarios; clustering the multiple first load scenarios to obtain multiple first loads and multiple second occurrence probabilities corresponding to multiple target load typical scenarios; the multiple first occurrence probabilities are used to characterize the probability of occurrence of multiple target power typical scenarios, and the multiple second occurrence probabilities are used to characterize the probability of occurrence of multiple target load typical scenarios; Based on the multiple first powers, the multiple first occurrence probabilities, the multiple first loads and the multiple second occurrence probabilities, a hydropower station scheduling plan is determined with the expected minimum maximum value of the remaining load as the first optimization objective function, the expected minimum fluctuation of the downstream water level as the second optimization objective function, and the expected minimum fluctuation of the downstream discharge flow as the third optimization objective function.

2. The method according to claim 1, characterized in that The method of determining the hydropower station scheduling plan based on the multiple first powers, the multiple first occurrence probabilities, the multiple first loads, and the multiple second occurrence probabilities, taking the expected minimum of the maximum value of the remaining load as the first optimization objective function, the expected minimum of the downstream tail water level fluctuation as the second optimization objective function, and the expected minimum of the downstream discharge flow volatility as the third optimization objective function, comprises: A multi-objective scheduling model is constructed according to the multiple first powers, the multiple first occurrence probabilities, the multiple first loads, and the multiple second occurrence probabilities; the optimization objectives of the multi-objective scheduling model include the first optimization objective function with the minimum expected maximum value of the remaining load as the target, the second optimization objective function with the minimum expected fluctuation of the downstream tail water level as the target, and the third optimization objective function with the minimum expected fluctuation of the downstream discharge flow as the target; the input of the multi-objective scheduling model is the multiple first powers, the multiple first occurrence probabilities, the multiple first loads, and the multiple second occurrence probabilities, and the output of the multi-objective scheduling model is an optimization solution set; Solving the multi-objective scheduling model to obtain the optimization solution set; the optimization solution set is used to represent the solution set of optimization data for scheduling the hydropower station; Select target scheduling data from the optimization solution set, and generate the hydropower station scheduling plan based on the target scheduling data; wherein the target scheduling data is used to characterize the optimal scheduling data.

3. The method according to claim 1 or 2, characterized in that: The method of predicting the power of wind power or photovoltaic power according to historical power data to obtain a power prediction value, and predicting the load of a hydropower station according to historical load data to obtain a load prediction value, includes: Inputting the historical power data into a trained power prediction model to obtain the power prediction value, wherein the input of the power prediction model is the historical power data, and the output of the power prediction model is the power prediction value; The historical load data is input into a trained load prediction model to obtain the load prediction value. The input of the load prediction model is the historical load data, and the output of the load prediction model is the load prediction value.

4. The method according to claim 1 or 2, characterized in that: The performing of prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; and performing of prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation include: Taking the difference between the actual power value and the predicted power value as the first prediction deviation; Performing probability distribution analysis on the first prediction deviation to obtain the first probability distribution function; taking the difference between the load measured value and the load predicted value as the second prediction deviation; Perform a probability distribution analysis on the second prediction deviation to obtain the second probability distribution function.

5. The method according to claim 1 or 2, characterized in that: The performing scenario analysis on the first probability distribution function to generate a plurality of first power typical scenarios; and the performing scenario analysis on the second probability distribution function to generate a plurality of first load typical scenarios include: Performing random sampling on the first probability distribution function for a preset number of times to obtain a plurality of first sampling results; generating the plurality of first power typical scenarios according to the plurality of first sampling results and the first prediction deviations corresponding to the plurality of first sampling results; Performing random sampling on the second probability distribution function for a preset number of times to obtain a plurality of second sampling results; The plurality of first load typical scenarios are generated according to the second prediction deviations corresponding to the plurality of second sampling results and the plurality of first sampling results.

6. The method according to claim 1 or 2, characterized in that: The clustering process is performed on the multiple first power scenarios to obtain multiple first powers and multiple first occurrence probabilities corresponding to multiple target power typical scenarios; Clustering the multiple first load scenarios to obtain multiple first loads and multiple second occurrence probabilities corresponding to multiple target load typical scenarios includes: Among the multiple first power scenarios, a preset number of first power scenarios are selected as initial power typical scenarios; the distance from each first power scenario to the initial power typical scenario is calculated respectively to obtain multiple first target distances; each first power scenario is assigned to the initial power typical scenario with the closest first target distance to obtain multiple first cluster groups; based on the first power scenarios in the multiple first cluster groups, multiple target power typical scenarios are generated; multiple first powers and multiple first occurrence probabilities corresponding to the multiple target power typical scenarios are obtained; Among the multiple first load scenarios, a preset number of first load scenarios are selected as initial load typical scenarios; the distance from each first load scenario to the initial load typical scenario is calculated respectively to obtain multiple second target distances; each first load scenario is assigned to the initial load typical scenario with the closest second target distance to obtain multiple second cluster groups; based on the first load scenarios in the multiple second cluster groups, multiple target load typical scenarios are generated; and multiple first loads and multiple second occurrence probabilities corresponding to the multiple target load typical scenarios are obtained.

7. The method according to claim 1 or 2, characterized in that: The first optimization objective function is: Among them, F1 represents the first optimization objective function, min represents the minimum value, max represents the maximum value, and p k represents the first occurrence probability corresponding to the target power typical scenario k or the second occurrence probability corresponding to the target load typical scenario k, K represents the total number of target power typical scenarios or the total number of target load typical scenarios, P k,t represents the first power of the target power typical scenario k in time period t or the first load of the target load typical scenario k in time period t; represents the power generation of hydropower in the target power typical scenario k or target load typical scenario k in time period t, represents the power generated by wind power in time period t under the target power typical scenario k or target load typical scenario k, It represents the power generation of photovoltaic power in the target power typical scenario k or the target load typical scenario k in the time period t.

8. The method according to claim 1 or 2, characterized in that: The second optimization objective function is: Among them, F2 represents the second optimization objective function, min represents the minimum value, max represents the maximum value, and p k represents the first occurrence probability corresponding to the target power typical scenario k or the second occurrence probability corresponding to the target load typical scenario k, K represents the total number of target power typical scenarios or the total number of target load typical scenarios, Z k,t+1 represents the water level in the time period t+1 under the target power typical scenario k or the target load typical scenario k, Z k,t It represents the water level in time period t under the target power typical scenario k or the target load typical scenario k.

9. The method according to claim 1 or 2, characterized in that: The third optimization objective function is: Among them, F3 represents the third optimization objective function, min represents the minimum value, and p k represents the first occurrence probability corresponding to the target power typical scenario k or the second occurrence probability corresponding to the target load typical scenario k, Q k,t+1 represents the flow rate in the time period t+1 under the target power typical scenario k or the target load typical scenario k, Q k,t It represents the flow rate in time period t under the target power typical scenario k or the target load typical scenario k.

10. A dispatching device for a hydropower station, characterized in that: The device comprises: A power and load forecasting module, used to forecast the power of wind power or photovoltaic power according to historical power data to obtain a power forecast value, and to forecast the load of a hydropower station according to historical load data to obtain a load forecast value; the historical power data is used to characterize the wind power or photovoltaic power data in a preset time period, and the historical load data is used to characterize the load data of a hydropower station in a preset time period; A prediction deviation analysis module is used to perform a prediction deviation analysis on the power prediction value and the power measured value to obtain a first probability distribution function corresponding to the first prediction deviation; and perform a prediction deviation analysis on the load prediction value and the load measured value to obtain a second probability distribution function corresponding to the second prediction deviation; A scenario generation module, configured to perform scenario analysis on the first probability distribution function to generate a plurality of first power scenarios; and perform scenario analysis on the second probability distribution function to generate a plurality of first load scenarios; A cluster analysis module, used to perform cluster analysis on the multiple first power scenarios to obtain multiple first powers and multiple first occurrence probabilities corresponding to multiple target power typical scenarios; perform cluster analysis on the multiple first load scenarios to obtain multiple first loads and multiple second occurrence probabilities corresponding to multiple target load typical scenarios; the multiple first occurrence probabilities are used to characterize the probability of occurrence of multiple target power typical scenarios, and the multiple second occurrence probabilities are used to characterize the probability of occurrence of multiple target load typical scenarios; The optimization scheduling module is used to determine the hydropower station scheduling plan based on the multiple first powers, the multiple first occurrence probabilities, the multiple first loads and the multiple second occurrence probabilities, with the expected minimum maximum value of the remaining load as the first optimization objective function, the expected minimum fluctuation of the downstream tail water level as the second optimization objective function, and the expected minimum fluctuation of the downstream discharge flow as the third optimization objective function.

11. A computer device, characterized in that: include: 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 scheduling method for a hydropower station according to any one of claims 1 to 9 by executing the computer instructions.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the hydropower station scheduling method according to any one of claims 1 to 9.

13. A computer program product, characterized in that It comprises computer instructions, and the computer instructions are used to make a computer execute the dispatching method of the hydropower station according to any one of claims 1 to 9.

Images