Comprehensive energy system uncertainty optimization scheduling method based on prediction driving

Abstract

The invention belongs to the field of energy system optimization scheduling, and particularly discloses a prediction driving-based integrated energy system uncertainty optimization scheduling method, which comprises the steps of establishing an integrated energy system, and constructing an optimization scheduling model of the integrated energy system; collecting meteorological energy consumption data in the research area based on a fixed time period, and preprocessing the meteorological energy consumption data to obtain preprocessed data; the meteorological energy consumption data comprises wind speed, illumination radiation intensity, temperature and power load data; the preprocessed data are input into a prediction model to obtain a prediction result of wind and light output and load, a prediction value interval is determined according to the prediction result to measure prediction uncertainty, and the prediction model is obtained by training historical data samples; and performing coupling optimization according to the objective function of the optimal scheduling model and the predicted value interval to obtain an optimal scheduling result of the system uncertainty. According to the invention, operation risks caused by prediction errors can be effectively reduced.

Description

Technical Field

[0001] The present application belongs to the field of energy system optimization and scheduling, and more specifically, to a prediction-driven integrated energy system uncertainty optimization and scheduling method. Background Art

[0002] Energy is the basic pillar of human survival and development. Achieving low-carbon energy economy is crucial to human and social development. Although the development and utilization of fossil fuels have promoted significant progress in human civilization, it has also brought severe challenges such as resource depletion and climate change. Accelerating energy transformation and ensuring sustainable energy utilization has become a global consensus. With the continuous expansion of renewable energy installed capacity, the construction of a modern energy system has made significant progress. The integrated energy system (IES) is considered to be an effective way to meet electricity demand by integrating various energy sources, technologies, and energy storage methods, and has gradually attracted widespread attention.

[0003] However, the operation of the integrated energy system faces many uncertainties. For example, affected by natural climate conditions, wind and solar energy show significant spatial imbalance and temporal instability, and have strong randomness, intermittency and uncertainty characteristics. Similarly, the uncertainty of electricity load stems from factors such as user behavior, market fluctuations and energy policies, and also presents considerable randomness and uncertainty. The combined effect of these uncertainties leads to complex and fluctuating demand patterns on both the supply and demand sides. This poses great challenges to the planning, scheduling and optimization of the integrated energy system.

[0004] Therefore, how to effectively deal with the uncertainty of the integrated energy system and reduce the operational risks caused by prediction errors is a technical problem that needs to be solved urgently. Summary of the invention

[0005] In view of the defects of the prior art, the purpose of this application is to provide a forecast-driven integrated energy system uncertainty optimization scheduling method, aiming to solve the problem that the prior art is difficult to cope with the uncertainty of the integrated energy system and difficult to deal with the operational risks caused by forecast errors.

[0006] To achieve the above objectives, in a first aspect, the present application provides a forecast-driven integrated energy system uncertainty optimization scheduling method, comprising: Establishing an integrated energy system and constructing an optimal dispatching model for the integrated energy system; Meteorological energy consumption data in the study area are collected based on a fixed time period, and the meteorological energy consumption data are preprocessed to obtain preprocessed data; the meteorological energy consumption data include: wind speed, light radiation intensity, temperature and power load data; Inputting the preprocessed data into a prediction model to obtain prediction results of wind and solar power output and load, and determining a prediction value interval according to the prediction results to measure prediction uncertainty, wherein the prediction model is obtained by training based on historical data samples; A coupled optimization is performed based on the objective function and the predicted value interval of the optimization scheduling model to obtain an optimization scheduling result of system uncertainty.

[0007] Optionally, comprehensive optimization is performed according to the coupling result of the objective function and the predicted value interval of the optimization scheduling model, including: Determine the system operation risk caused by prediction uncertainty according to the prediction value interval combined with a penalty factor, wherein the penalty factor includes a first penalty factor for power generation load surplus and a second penalty factor for power generation load deficit; The system operation risk is combined with the objective function of the optimization scheduling model for coupling optimization; The first penalty factor and the second penalty factor are both determined based on the time-of-use electricity price and the risk control factor, and the objective function includes maximizing the system economic benefits and minimizing the pollutant emissions.

[0008] Optionally, it also includes: Plan the power generation or load demand in each future period according to the forecast interval to minimize the total operating risk of the system under the worst-case scenario; Define an auxiliary variable, wherein the auxiliary variable represents the risk value of the current period under the worst case scenario; Construct linear constraints. If the actual value reaches the upper limit of the forecast, calculate the risk when the planned value is lower than the upper limit. If the actual value reaches the lower limit of the forecast, calculate the risk when the planned value is higher than the lower limit. Determine that the optimal solution of the auxiliary variable corresponds to the upper limit or lower limit of the prediction interval, and use a linear programming algorithm to solve it to obtain the optimal plan value for each time period.

[0009] Optionally, the integrated energy system includes: a wind turbine subsystem, a photovoltaic panel subsystem, an electrolyzer subsystem, a hydrogen storage tank subsystem, and a fuel cell subsystem; The wind turbine subsystem model is shown below:

[0010] in, is the wind turbine output power, is the true wind speed, is the rated wind speed, and are the cut-in and cut-out wind speeds, is the rated output power, for the current moment; The photovoltaic panel subsystem model is as follows:

[0011]

[0012] in, Power for photovoltaic panels, is the rated output of the photovoltaic panel, is the penalty factor, is the temperature coefficient, is the real light radiation intensity, For standard test light radiation intensity, and Respectively represent the current temperature of the photovoltaic panel and the standard test temperature. and Represent the operating temperature and ambient temperature respectively; The electrolyzer subsystem model is shown below:

[0013]

[0014] in, The quality of hydrogen produced by the electrolyzer, Indicates the mass of hydrogen that can be generated per 1kWh of electricity. is the electrolyzer conversion efficiency, is the electrolyzer input, is the electrolyzer converter efficiency; The fuel cell subsystem model is as follows:

[0015]

[0016] in, is the mass of hydrogen consumed by the fuel cell, is the fuel cell output, is the amount of electricity that can be generated by 1kg of hydrogen, is the fuel cell efficiency, is the efficiency of the hydrogen storage tank; The hydrogen storage tank subsystem model is as follows: like , the electrolyzer works, produces hydrogen by electrolyzing water, and stores the hydrogen in the hydrogen storage tank. The balance formula of hydrogen in the hydrogen storage tank is:

[0017] like , the fuel cell works and consumes hydrogen to generate electricity. At this time, the balance formula of hydrogen in the hydrogen storage tank is:

[0018] in, for The amount of hydrogen remaining in the hydrogen storage tank at the moment, in kg; The maximum input of the electrolyzer and the maximum output of the fuel cell are affected not only by their own capacity, but also by the amount of hydrogen remaining in the current hydrogen storage tank. The mathematical expression is:

[0019]

[0020]

[0021]

[0022] in, and are the maximum input of the electrolyzer and the maximum output of the fuel cell, respectively. and are the electrolyzer and fuel cell capacities, and They are the upper and lower limits of the hydrogen storage tank capacity respectively.

[0023] Optionally, the method for determining the maximum economic benefit and the minimum pollutant emission of the system includes: Based on the output of wind turbines, photovoltaic panels, fuel cells and electrolyzers, combined with the time-of-use electricity prices of each time period, the power generation benefits are calculated; The power grid exchange income is obtained according to the power sales income and the power purchase cost, wherein the power sales income is determined according to the power sales time-of-use price and the power sales amount, and the power purchase cost is determined according to the power purchase time-of-use price and the power purchase amount; Based on the amount of wind and solar power abandoned, combined with the wind penalty coefficient and the solar penalty coefficient, the cost of wind and solar power abandoned is obtained; Constructing a first functional relationship of the economic benefits of the system according to the power generation benefits, the grid exchange benefits and the wind and solar power abandonment costs; Constructing a second functional relationship of the pollutant emission according to the power purchase amount and the emission coefficient; The first functional relationship and the second functional relationship are optimized under constraints to maximize the economic benefits of the system and minimize pollutant emissions.

[0024] Optionally, the constraint conditions include: power balance constraint, wind and solar output constraint, electrolyzer output constraint, hydrogen storage tank capacity constraint and fuel cell constraint; The power balance constraint is that the weighted sum of wind power generation, photovoltaic power generation, fuel cell power generation, electrolyzer input, power sales and power purchase is balanced with the power load demand; The wind and solar power output constraints are: the actual power generation of wind power and photovoltaic power is within the range of maximum power generation capacity and is not negative; The electrolytic cell output constraint is: the power consumption of the electrolytic cell is within the range of the maximum working capacity and is not negative; The hydrogen storage tank capacity constraint is: the hydrogen storage capacity of the hydrogen storage tank must be maintained between the minimum and maximum capacities, and maintain dynamic balance, with hydrogen production and hydrogen consumption mutually exclusive; The fuel cell constraint is that the power generation of the fuel cell is within the range of the maximum power generation capacity and is not negative.

[0025] Optionally, the preprocessed data is input into a prediction model to obtain uncertainty prediction results of wind and solar power output and load, including: The pre-processed data within N hours before the current time is input into the prediction model as input data to obtain the predicted data values ​​of wind and solar power output and load for the next M hours; Determine the lower bound and the upper bound of the prediction interval according to the predicted data value and the interval width, and determine the prediction interval; Wherein, N and M are both positive integers, and the interval width is determined according to the predicted data value and the target percentage coefficient.

[0026] Optionally, the prediction model training method includes: The preprocessed data within N hours before the current time is used as input data, and the predicted data value in the next M hours is used as output data; Normalizing the input data and output data, scaling the data range to a fixed interval, so as to improve the model training effect; The FEDformer algorithm is used for training to learn the mapping relationship between input and output through historical data; The test set is used to evaluate the model performance, and the root mean square error (RMSE) is used as the evaluation indicator to measure the deviation between the predicted value and the actual value.

[0027] In a second aspect, the present application provides an electronic device comprising: at least one memory for storing programs; and at least one processor for executing the programs stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method described in the first aspect or any possible implementation of the first aspect.

[0028] In a third aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method described in the first aspect or any possible implementation of the first aspect.

[0029] In a fourth aspect, the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method described in the first aspect or any possible implementation of the first aspect.

[0030] It can be understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0031] In general, the above technical solutions conceived by this application have the following beneficial effects compared with the prior art: (1) This application establishes an optimal dispatch model for an integrated energy system, effectively combines uncertainty prediction with the optimal dispatch model, and combines the prediction of meteorological energy consumption data to significantly improve the system's ability to respond to uncertain factors. By preprocessing and predicting uncertainty data such as wind speed, light radiation intensity, temperature, and power load, it can more accurately reflect the output of renewable energy, thereby effectively reducing the operational risks caused by prediction errors.

[0032] (2) The multi-objective optimization scheduling model established in this application takes into account the two key indicators of maximizing economic benefits and minimizing pollutant emissions. This application enables the scheduling plan to not only pursue economic benefits, but also pay attention to environmental protection, thus achieving an organic combination of economy and environmental protection, promoting the realization of sustainable development goals, and helping to promote the widespread application of green energy.

[0033] (3) The method proposed in this application allows the degree of uncertainty to be flexibly adjusted according to actual conditions to meet the conservative requirements of the scheduling scheme. This flexibility enables the system to better adapt to different operating environments and demand changes, improves the adaptability and robustness of scheduling, provides operators with greater decision-making space, and enhances the overall operating efficiency of the integrated energy system. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of a forecast-driven integrated energy system uncertainty optimization scheduling method provided in this application; Figure 2 It is a diagram of interval prediction and optimization decision results of an embodiment of the present application; Figure 2 (a) shows the relationship between wind power time and power. Figure 2 (b) shows the relationship between photoelectric time and power. Figure 2(c) shows the relationship between load time and power; Figure 3 is a diagram of optimization scheduling results under different uncertainty levels of the embodiment of the present application; Figure 3 (a), (b), (c), and (d) indicate uncertainty levels of 0.2, 0.4, 0.6, and 0.8, respectively; Figure 4 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0036] The term "and / or" in this article is a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The symbol " / " in this article indicates that the associated objects are in an or relationship, for example, A / B means A or B.

[0037] The terms "first" and "second" in the specification and claims herein are used to distinguish different objects rather than to describe a specific order of the objects. For example, a first response message and a second response message are used to distinguish different response messages rather than to describe a specific order of the response messages.

[0038] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0039] In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two. For example, multiple processing units refer to two or more processing units, etc.; multiple elements refer to two or more elements, etc.

[0040] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0041] Reference Figure 1 The present application provides a forecast-driven comprehensive energy system uncertainty optimization scheduling method, comprising: S101. Establishing an integrated energy system and constructing an optimal scheduling model for the integrated energy system; S102. Collect meteorological energy consumption data in the study area based on a fixed time period, and pre-process the meteorological energy consumption data to obtain pre-processed data; the meteorological energy consumption data includes: wind speed, light radiation intensity, temperature and power load data; S103. Input the preprocessed data into the prediction model to obtain the prediction results of wind and solar power output and load, and determine the prediction value interval according to the prediction results to measure the prediction uncertainty, wherein the prediction model is trained based on historical data samples; S104. Perform coupling optimization according to the objective function and prediction value interval of the optimization scheduling model to obtain an optimization scheduling result of system uncertainty.

[0042] Specifically, through S101, a comprehensive energy system and an optimized dispatching model are established. The comprehensive energy system organically integrates various energy forms such as electricity, natural gas, and thermal energy. Through the coordinated conversion and complementary utilization of energy, energy utilization efficiency is improved, and energy consumption and environmental pollution are reduced. Through the optimized dispatching model, the optimal operating state of various energy equipment at different times can be determined, and the reasonable allocation and efficient utilization of energy can be achieved, ensuring the stable, economical, and environmentally friendly operation of the comprehensive energy system.

[0043] Secondly, the meteorological energy consumption data is collected and preprocessed through S102.

[0044] In order to accurately analyze and predict the operation of the integrated energy system, it is necessary to collect meteorological energy consumption data in the study area based on a fixed time period (the embodiment of the present application can preferably be 1 hour as a fixed time period). These data cover multiple key elements such as wind speed, light radiation intensity, temperature, and power load data. Wind speed and light radiation intensity are crucial for evaluating the potential of wind power generation and photovoltaic power generation, while temperature is closely related to energy demand and conversion efficiency. Power load data directly reflects the actual energy consumption of the system.

[0045] After collecting these raw data, it is usually necessary to preprocess them. Optionally, the preprocessing process usually includes data cleaning, removing outliers and erroneous data to ensure the accuracy and reliability of the data; data normalization, unifying data of different dimensions to the same scale range for subsequent analysis and calculation; data interpolation, reasonably supplementing missing data to make the data sequence complete and continuous. Preprocessed data can provide a better foundation for subsequent prediction and analysis.

[0046] Further, the uncertainty is predicted and the prediction value interval is determined through S103. In this embodiment, the preprocessed data is input into the prediction model, which is trained based on a large number of historical data samples. Through advanced algorithms such as machine learning and deep learning, the model can learn the complex relationship between meteorological factors and wind and solar output and power load.

[0047] The uncertainty prediction results of wind and solar power output and load output by the prediction model of this embodiment take into account various uncertain factors existing in actual operation, such as random changes in meteorological conditions and equipment failures.

[0048] Based on these uncertain prediction results, the prediction value range is determined. This prediction value range not only gives the central value of the prediction, but also reflects the degree of uncertainty of the prediction results, providing more comprehensive information for subsequent optimization scheduling.

[0049] Finally, the optimal scheduling result of the system uncertainty is obtained through S104 coupling optimization.

[0050] According to the objective function of the previously established integrated energy system optimization dispatch model, coupled optimization is performed in combination with the forecast value range of wind and solar power output and load. In the optimization process, the impact of uncertain factors on system operation is fully considered, and dispatch decisions are no longer made based solely on deterministic forecast values.

[0051] Furthermore, the comprehensive optimization is performed according to the coupling result of the objective function and the predicted value interval of the optimization scheduling model, including: Determine the system operation risk caused by prediction uncertainty according to the prediction value interval combined with a penalty factor, wherein the penalty factor includes a first penalty factor for power generation load surplus and a second penalty factor for power generation load deficit; The system operation risk is combined with the objective function of the optimization scheduling model for coupling optimization; The first penalty factor and the second penalty factor are both determined based on the time-of-use electricity price and the risk control factor, and the objective function includes maximizing the system economic benefits and minimizing the pollutant emissions.

[0052] Optionally, it also includes: Plan the power generation or load demand in each future period according to the forecast interval to minimize the total operating risk of the system under the worst-case scenario; Define an auxiliary variable, wherein the auxiliary variable represents the risk value of the current period under the worst case scenario; Construct linear constraints. If the actual value reaches the upper limit of the forecast, calculate the risk when the planned value is lower than the upper limit. If the actual value reaches the lower limit of the forecast, calculate the risk when the planned value is higher than the lower limit. Determine that the optimal solution of the auxiliary variable corresponds to the upper limit or lower limit of the prediction interval, and use a linear programming algorithm to solve it to obtain the optimal plan value for each time period.

[0053] Specifically, the system operation risk caused by prediction uncertainty in this application can be expressed as:

[0054]

[0055] in, , is the actual power generation / load demand, To plan power generation / forecast demand, For 1 hour. is the penalty factor for generation / load surplus, is the penalty factor for generation / load deficit, and The definition is as follows:

[0056]

[0057] in, It is a time-of-use electricity price. is the risk control factor, and its value is in [0,1].

[0058] However, in practice, the actual power generation / load demand is often unknown when making decisions, so the introduction of auxiliary variables , reshape the optimization problem into:

[0059]

[0060] in, is the optimal operating risk corresponding to the worst case.

[0061] Next, and Bringing in the constraints, we get:

[0062]

[0063] This function is a step function, and the optimal value can only be obtained at the boundary. Therefore, the optimization problem has a solution when and only when the boundary is taken, which can be further expressed as:

[0064] In summary, the uncertainty optimization problem based on the prediction interval is transformed into a linear programming problem.

[0065] Optionally, the integrated energy system includes: a wind turbine subsystem, a photovoltaic panel subsystem, an electrolyzer subsystem, a hydrogen storage tank subsystem, and a fuel cell subsystem; The wind turbine subsystem model is shown below:

[0066] in, is the wind turbine output power, is the true wind speed, is the rated wind speed, and are the cut-in and cut-out wind speeds, is the rated output power, for the current moment; The photovoltaic panel subsystem model is as follows:

[0067]

[0068] in, Power for photovoltaic panels, is the rated output of the photovoltaic panel, is the penalty factor, is the temperature coefficient, is the real light radiation intensity, For standard test light radiation intensity, and Respectively represent the current temperature of the photovoltaic panel and the standard test temperature. and Represent the operating temperature and ambient temperature respectively; The electrolyzer subsystem model is shown below:

[0069]

[0070] in, The quality of hydrogen produced by the electrolyzer, Indicates the mass of hydrogen that can be generated per 1kWh of electricity. is the electrolyzer conversion efficiency, is the electrolyzer input, is the electrolyzer converter efficiency; The fuel cell subsystem model is as follows:

[0071]

[0072] in, is the mass of hydrogen consumed by the fuel cell, is the fuel cell output, is the amount of electricity that can be generated by 1kg of hydrogen, is the fuel cell efficiency, is the efficiency of the hydrogen storage tank; The hydrogen storage tank subsystem model is as follows: like , the electrolyzer works, produces hydrogen by electrolyzing water, and stores the hydrogen in the hydrogen storage tank. The balance formula of hydrogen in the hydrogen storage tank is:

[0073] like , the fuel cell works and consumes hydrogen to generate electricity. At this time, the balance formula of hydrogen in the hydrogen storage tank is:

[0074] in, for The amount of hydrogen remaining in the hydrogen storage tank at the moment, in kg; The maximum input of the electrolyzer and the maximum output of the fuel cell are affected not only by their own capacity, but also by the amount of hydrogen remaining in the current hydrogen storage tank. The mathematical expression is:

[0075]

[0076]

[0077]

[0078] in, and are the maximum input of the electrolyzer and the maximum output of the fuel cell, respectively. and are the electrolyzer and fuel cell capacities, and They are the upper and lower limits of the hydrogen storage tank capacity respectively.

[0079] Optionally, the method for determining the maximum economic benefit and the minimum pollutant emission of the system includes: Based on the output of wind turbines, photovoltaic panels, fuel cells and electrolyzers, combined with the time-of-use electricity prices of each time period, the power generation benefits are calculated; The power grid exchange income is obtained according to the power sales income and the power purchase cost, wherein the power sales income is determined according to the power sales time-of-use price and the power sales amount, and the power purchase cost is determined according to the power purchase time-of-use price and the power purchase amount; Based on the amount of wind and solar power abandoned, combined with the wind penalty coefficient and the solar penalty coefficient, the cost of wind and solar power abandoned is obtained; Constructing a first functional relationship of the economic benefits of the system according to the power generation benefits, the grid exchange benefits and the wind and solar power abandonment costs; Constructing a second functional relationship of the pollutant emission according to the power purchase amount and the emission coefficient; The first functional relationship and the second functional relationship are optimized under constraints to maximize the economic benefits of the system and minimize pollutant emissions.

[0080] Optionally, the constraint conditions include: power balance constraint, wind and solar output constraint, electrolyzer output constraint, hydrogen storage tank capacity constraint and fuel cell constraint; The power balance constraint is that the weighted sum of wind power generation, photovoltaic power generation, fuel cell power generation, electrolyzer input, power sales and power purchase is balanced with the power load demand; The wind and solar power output constraints are: the actual power generation of wind power and photovoltaic power is within the range of maximum power generation capacity and is not negative; The electrolytic cell output constraint is: the power consumption of the electrolytic cell is within the range of the maximum working capacity and is not negative; The hydrogen storage tank capacity constraint is: the hydrogen storage capacity of the hydrogen storage tank must be maintained between the minimum and maximum capacities, and maintain dynamic balance, with hydrogen production and hydrogen consumption mutually exclusive; The fuel cell constraint is that the power generation of the fuel cell is within the range of the maximum power generation capacity and is not negative.

[0081] Specifically, the calculation formula for economic benefits is as follows: (1) Maximizing economic revenue (ER):

[0082]

[0083]

[0084]

[0085] in, For power generation efficiency, For the grid exchange income, Cost of curtailing wind and solar power. For electricity sales revenue, The cost of purchasing electricity, Time-of-use electricity price, To generate power for wind turbines, Power for photovoltaic panels, Powering fuel cells, is the electrolyzer input, To purchase electricity, For electricity sales, and They represent the time-of-use electricity price for purchasing electricity and the time-of-use electricity price for selling electricity, and are the amount of wind abandonment and solar abandonment, and are the wind penalty coefficient and the light penalty coefficient.

[0086] (2) Minimize pollutant emissions (PE):

[0087] in, is the emission factor.

[0088] The constraints are as follows: (1) Power balance constraints:

[0089] in, Indicates the power load demand.

[0090] (2) Wind and solar power output constraints:

[0091]

[0092] in, and Represent the maximum output of wind and light respectively.

[0093] (3) Electrolyzer output constraints:

[0094] (4) Hydrogen storage tank capacity constraints:

[0095]

[0096]

[0097] (5) Fuel cell constraints:

[0098] Optionally, the preprocessed data is input into a prediction model to obtain uncertainty prediction results of wind and solar power output and load, including: The pre-processed data within N hours before the current time is input into the prediction model as input data to obtain the predicted data values ​​of wind and solar power output and load for the next M hours; Determine the lower bound and the upper bound of the prediction interval according to the predicted data value and the interval width, and determine the prediction interval; Wherein, N and M are both positive integers, and the interval width is determined according to the predicted data value and the target percentage coefficient.

[0099] Optionally, the prediction model training method includes: The preprocessed data within N hours before the current time is used as input data, and the predicted data value in the next M hours is used as output data; Normalizing the input data and output data, scaling the data range to a fixed interval, so as to improve the model training effect; The FEDformer algorithm is used for training to learn the mapping relationship between input and output through historical data; The test set is used to evaluate the model performance, and the root mean square error (RMSE) is used as the evaluation indicator to measure the deviation between the predicted value and the actual value.

[0100] Specifically, this embodiment is a process of performing prediction using a deep learning algorithm, comprising the following steps: Step S41: Use deep learning algorithm to predict wind and solar power output and power load in the next 24 hours. The specific mathematical formula is:

[0101] in, represents the predicted value, Represents input data. is a mapping function, which corresponds to the FEDformer algorithm in the embodiment of the present application.

[0102] Specifically, step S411: extract input data and output data. In one embodiment of the present application, the input data is the wind, light, and load data for the previous 96 hours, and the output data is the t +1, t +2, …… , t +24-hour wind, light and load forecast values.

[0103] That is to say, the above N can be the data within 96 hours before the current time, and M can be the data within 24 hours after the current time.

[0104] Step S412: using the min-max method to normalize the data.

[0105] Step S413: using the FEDformer algorithm to perform model training.

[0106] Step S414: Use the trained model for the test set and use RMSE as the standard to evaluate the model performance.

[0107] Step S42: construct a prediction interval. The specific mathematical expression is:

[0108] in, is the prediction interval, and represent the lower and upper bounds of the prediction interval, respectively. Indicates the interval width. In one embodiment of the present application, Take 5% of the predicted value.

[0109] In order to prove the effectiveness of the method proposed in this application, the wind power output, photovoltaic output and load demand of a certain region for one year were selected as the research objects. The specific data scale is hourly, and a total of 8760 hours of data are included.

[0110] First, the uncertainty of wind, solar and load was predicted, and on this basis, the decision of wind and solar output and load demand was made. The results are as follows: Figure 2 As shown. Figure 2 In the figure, the gray area represents the forecast interval, the blue dashed line represents the true value, and the orange solid line represents the forecast-driven optimization day-ahead planning decision result.

[0111] Secondly, the conservatism of the scheduling scheme under different degrees of uncertainty is discussed. The results are as follows: Figure 3 In one embodiment of the present application, four situations are considered, and the corresponding uncertainty levels are 0.2, 0.4, 0.6 and 0.8 respectively. The higher the uncertainty level, the more conservative the solution. Figure 3 (a)-(d) show that as the degree of uncertainty increases, the electricity sales at 11:00 decrease significantly, the fuel cell discharge decreases, and the electricity purchase increases. This is mainly because in terms of day-ahead planning, more electricity needs to be purchased to meet the worst-case scenario. Generally speaking, the greater the uncertainty faced by IES in day-ahead scheduling, the more conservative the scheduling plan will be.

[0112] In addition, Table 1 shows the economic benefits, pollutant emissions and system operation risks of the scheduling scheme under different degrees of uncertainty. A positive number indicates electricity sales, and a negative number indicates electricity purchases.

[0113] Table 1 Economic benefits, pollutant emissions and system operation risks under different degrees of uncertainty

[0114] According to Table 1, when the degree of uncertainty increases from 0.2 to 0.8, the economic benefit decreases from 23,161.12 yuan to 20,364.59 yuan, 17,621.46 yuan, and 14,912.43 yuan, and the pollutant emissions increase from 454,263.76 tons to 507,262.32 tons, 564,234.50 tons, and 624,854.86 tons. However, it is worth noting that although the economic benefits and pollutant emissions are relatively optimal when the degree of uncertainty is small, this does not mean that the scheduling scheme obtained in this scenario is optimal. This is because the scheme corresponds to the day-ahead plan. Due to the uncertainty of prediction and decision-making, there may be a certain degree of imbalance between the day-ahead plan and the true value, and this imbalance will bring additional uncertainty risks. As can be seen from Table 1, with the increase of uncertainty, the system operation risk decreases from 8554.69 to 7485.35, 5703.13, and 3208.01, respectively, which shows that although the economic benefits and pollutant emissions are relatively optimal, the system operation risk is relatively high. Therefore, when the degree of uncertainty is high, the scheduling plan is relatively conservative.

[0115] Reference Figure 4 Based on the method in the above embodiment, the embodiment of the present application provides an electronic device, which may include: a processor (Processor) 410, a communication interface (Communications Interface) 420, a memory (Memory) 430 and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 can call the logic instructions in the memory 430 to execute the method in the above embodiment.

[0116] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0117] Based on the method in the above embodiment, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method in the above embodiment.

[0118] Based on the method in the above embodiment, an embodiment of the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method in the above embodiment.

[0119] It is understandable that the processor in the embodiment of the present application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0120] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0121] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0122] It should be understood that the various numerical numbers involved in the embodiments of the present application are only used for the convenience of description and are not used to limit the scope of the embodiments of the present application.

[0123] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A forecast-driven integrated energy system uncertainty optimization scheduling method, characterized in that: include: Establishing an integrated energy system and constructing an optimal dispatching model for the integrated energy system; Collecting meteorological energy consumption data in a study area based on a fixed time period, and preprocessing the meteorological energy consumption data to obtain preprocessed data; The meteorological energy consumption data include: wind speed, light radiation intensity, temperature and power load data; Inputting the preprocessed data into a prediction model to obtain prediction results of wind and solar power output and load, and determining a prediction value interval according to the prediction results to measure prediction uncertainty, wherein the prediction model is obtained by training based on historical data samples; A coupled optimization is performed based on the objective function and the predicted value interval of the optimization scheduling model to obtain an optimization scheduling result of system uncertainty.

2. The forecast-driven integrated energy system uncertainty optimization scheduling method according to claim 1 is characterized in that: Comprehensive optimization is performed based on the coupling results of the objective function and the predicted value interval of the optimization scheduling model, including: Determine the system operation risk caused by prediction uncertainty according to the prediction value interval combined with a penalty factor, wherein the penalty factor includes a first penalty factor for power generation load surplus and a second penalty factor for power generation load deficit; The system operation risk is combined with the objective function of the optimization scheduling model for coupling optimization; The first penalty factor and the second penalty factor are both determined based on the time-of-use electricity price and the risk control factor, and the objective function includes maximizing the system economic benefits and minimizing the pollutant emissions.

3. The uncertainty optimization scheduling method for integrated energy system based on prediction drive according to claim 2 is characterized in that: Also includes: Plan the power generation or load demand in each future period according to the forecast interval to minimize the total operating risk of the system under the worst-case scenario; Define an auxiliary variable, wherein the auxiliary variable represents the risk value of the current period under the worst case scenario; Construct linear constraints. If the actual value reaches the upper limit of the forecast, calculate the risk when the planned value is lower than the upper limit. If the actual value reaches the lower limit of the forecast, calculate the risk when the planned value is higher than the lower limit. Determine that the optimal solution of the auxiliary variable corresponds to the upper limit or lower limit of the prediction interval, and use a linear programming algorithm to solve it to obtain the optimal plan value for each time period.

4. The uncertainty optimization scheduling method for integrated energy system based on prediction drive according to claim 2 is characterized in that: The method for determining the maximum economic benefit and the minimum pollutant emission of the system includes: Based on the output of wind turbines, photovoltaic panels, fuel cells and electrolyzers, combined with the time-of-use electricity prices of each time period, the power generation benefits are calculated; The power grid exchange income is obtained according to the power sales income and the power purchase cost, wherein the power sales income is determined according to the power sales time-of-use price and the power sales amount, and the power purchase cost is determined according to the power purchase time-of-use price and the power purchase amount; Based on the amount of wind and solar power abandoned, combined with the wind penalty coefficient and the solar penalty coefficient, the cost of wind and solar power abandoned is obtained; Constructing a first functional relationship of the economic benefits of the system according to the power generation benefits, the grid exchange benefits and the wind and solar power abandonment costs; Constructing a second functional relationship of the pollutant emission according to the power purchase amount and the emission coefficient; The first functional relationship and the second functional relationship are optimized under constraints to maximize the economic benefits of the system and minimize pollutant emissions.

5. The uncertainty optimization scheduling method for integrated energy system based on prediction drive according to claim 4 is characterized in that: The constraints include: power balance constraints, wind and solar output constraints, electrolyzer output constraints, hydrogen storage tank capacity constraints and fuel cell constraints; The power balance constraint is that the weighted sum of wind power generation, photovoltaic power generation, fuel cell power generation, electrolyzer input, power sales and power purchase is balanced with the power load demand; The wind and solar power output constraints are: the actual power generation of wind power and photovoltaic power is within the range of maximum power generation capacity and is not negative; The electrolytic cell output constraint is: the power consumption of the electrolytic cell is within the range of the maximum working capacity and is not negative; The hydrogen storage tank capacity constraint is: the hydrogen storage capacity of the hydrogen storage tank must be maintained between the minimum and maximum capacities, and maintain dynamic balance, with hydrogen production and hydrogen consumption mutually exclusive; The fuel cell constraint is that the power generation of the fuel cell is within the range of the maximum power generation capacity and is not negative.

6. The forecast-driven integrated energy system uncertainty optimization scheduling method according to claim 1 is characterized in that: The pre-processed data is input into the prediction model to obtain the uncertainty prediction results of wind and solar power output and load, including: The pre-processed data within N hours before the current time is input into the prediction model as input data to obtain the predicted data values ​​of wind and solar power output and load for the next M hours; Determine the lower bound and the upper bound of the prediction interval according to the predicted data value and the interval width, and determine the prediction interval; Wherein, N and M are both positive integers, and the interval width is determined according to the predicted data value and the target percentage coefficient.

7. The forecast-driven integrated energy system uncertainty optimization scheduling method according to claim 6 is characterized in that: The training method of the prediction model includes: The preprocessed data within N hours before the current time is used as input data, and the predicted data value in the next M hours is used as output data; Normalizing the input data and output data, scaling the data range to a fixed interval, so as to improve the model training effect; The FEDformer algorithm is used for training to learn the mapping relationship between input and output through historical data; The test set is used to evaluate the model performance, and the root mean square error (RMSE) is used as the evaluation indicator to measure the deviation between the predicted value and the actual value.

8. An electronic device, characterized in that: include: at least one memory for storing a computer program; At least one processor is used to execute the program stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method according to any one of claims 1 to 7.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program runs on a processor, the processor is caused to execute the method according to any one of claims 1 to 7.

Images