Comprehensive system multi-time scale scheduling method considering prediction error

Abstract

The invention belongs to the technical field of electric power system scheduling, and particularly discloses an integrated system multi-time scale scheduling method considering prediction errors, which comprises the following steps: establishing an operation constraint model of an integrated system; processing the uncertainty parameters of the integrated system by using an adjustable robust optimization method, and constructing an uncertainty interval of each uncertainty parameter; in a day-ahead scheduling stage, establishing a master-slave game optimization scheduling model based on the uncertainty interval and the operation constraint model to generate a day-ahead scheduling plan; and in an intra-day scheduling stage, optimizing and adjusting the day-ahead scheduling plan in real time based on a prediction error by adopting a self-adaptive step length double-closed-loop model prediction control method so as to obtain an optimized scheduling scheme of the integrated system. According to the method and the device, the prediction error of the renewable energy source and the load can be effectively processed in multi-time-scale optimization scheduling, so that the economy and the robustness of system operation are balanced, and the scheduling precision is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power system dispatching, and more particularly to a comprehensive system multi-time scale scheduling method considering prediction errors. BACKGROUND

[0002] Energy systems have become the main form of energy utilization, and water, wind, light and hydrogen combined power generation can effectively promote green development.

[0003] However, renewable energy has indirectness and randomness. With the continuous expansion of renewable energy, it brings difficulties to the user load side. Although the existing technology considers the uncertainty of the source and load in the dispatching process, the optimization method is often too conservative in the decision-making process, which may cause economic losses.

[0004] In addition, the uncertainty of renewable energy and load leads to an increase in prediction error with the increase of time scale. The traditional model predictive control method relies on a single closed-loop mechanism and fails to fully utilize other information that helps to enhance system reliability. Therefore, how to effectively handle the prediction error of renewable energy and load in multi-time scale optimization scheduling to balance the economy and robustness of system operation and improve the scheduling accuracy is a technical problem that needs to be solved at present. SUMMARY

[0005] In view of the defects of the prior art, the purpose of the present application is to balance the economy and robustness of system operation and improve the scheduling accuracy by considering the prediction error of renewable energy and load in multi-time scale optimization scheduling.

[0006] To achieve the above-mentioned purpose, in a first aspect, the present application provides a comprehensive system multi-time scale scheduling method considering prediction errors, comprising: establishing an operation constraint model of a comprehensive system; the comprehensive system comprises a wind power subsystem, a photovoltaic subsystem, a water power subsystem and a hydrogen energy storage subsystem; the operation constraint model comprises a wind power output model, a photovoltaic output model, a water power output model and a hydrogen energy storage system model; processing the uncertainty parameters of the comprehensive system by using an adjustable robust optimization method to construct the uncertainty interval of each uncertainty parameter; the uncertainty parameters include runoff, wind and light output and load; In the day-ahead scheduling stage, a master-slave game optimization scheduling model is established based on the uncertainty interval and the operation constraint model to generate a day-ahead scheduling plan; In the intra-day scheduling stage, an adaptive step double-closed-loop model predictive control method is used to optimize and adjust the day-ahead scheduling plan based on the prediction error to obtain an optimized scheduling scheme of the comprehensive system.

[0007] Optionally, the method for constructing the wind power output model includes: Based on the real-time wind speed at the hub height of the wind turbine at the target time, and the cut-in wind speed, rated wind speed, and cut-out wind speed of the wind turbine, different output ranges are divided, and the output power of the wind turbine in each output range is determined to construct a set of equations relating power and wind speed as a wind power output model.

[0008] Optionally, the method for constructing the photovoltaic output model includes: Based on the standard operating temperature of the photovoltaic panel, the ambient temperature and actual light intensity of the photovoltaic panel at the target time, the actual operating temperature of the photovoltaic panel is determined. Based on the actual operating temperature and temperature coefficient of the photovoltaic panel, the power loss caused by the temperature rise at the target time is determined; Based on the power loss, rated output of the photovoltaic panel, standard illuminance, actual illuminance, and converter efficiency, the output power of the photovoltaic panel at the target time is determined, and a set of equations relating power and photovoltaic parameters is constructed as a photovoltaic output model.

[0009] Optionally, the hydropower output model includes at least one of the following constraints: Water balance constraints; Reservoir capacity constraints; Downflow constraint; Hydropower station output constraints; Head constraint; Constraints related to the relationship between reservoir water level and reservoir capacity; Constraints on the relationship between tailwater level and discharge flow.

[0010] Optionally, the method for constructing the hydrogen energy storage system model includes: A dynamic equilibrium relationship is established based on the current hydrogen storage capacity, the previous hydrogen storage capacity, the current hydrogen production mass, and the current hydrogen sales mass. Based on the unit hydrogen production mass, the output power of the electrolyzer, the working efficiency of the electrolyzer, and the converter efficiency, the hydrogen production mass of the electrolyzer at the current moment is determined to establish a hydrogen production model for the electrolyzer. The upper and lower limits of the hydrogen storage capacity of the hydrogen storage tank are set, and the hydrogen storage capacity boundary conditions at the beginning and end of the scheduling cycle are set for the hydrogen energy storage system.

[0011] Optionally, an adjustable robust optimization method is used to handle the uncertainty parameters of the synthesized system, including: Based on the predicted maximum and minimum values ​​of each uncertainty parameter, determine the summation mean and the difference mean; An adjustable robustness coefficient is introduced and multiplied by the mean difference to obtain the uncertainty deviation; Based on the uncertainty deviation and the summation mean, the upper limit and lower limit of the uncertainty interval are determined to obtain the uncertainty interval; A robust auxiliary coefficient is introduced to add uncertainty constraints to the uncertainty interval.

[0012] Optionally, the master-slave game optimization scheduling model includes an upper-level sub-model and a lower-level sub-model; the upper-level objective function of the upper-level sub-model is to maximize the operational efficiency index, and the lower-level objective function of the lower-level sub-model is to minimize the user expenditure index. The upper-level objective function is constructed based on data on electricity transmission, hydrogen transmission, green energy benefits, wind and solar curtailment penalties, carbon emission data, equipment operation and maintenance data, and purchased electricity data. The constraints of the upper-level objective function include electricity sales price constraints, power balance constraints, electricity purchase constraints, and wind and solar power output constraints. The lower-level objective function is constructed based on the current electricity price, transferable load, reduceable load, actual load, and subsidy unit price, which is obtained by adjusting flexible load. The lower-level objective function is constrained by the load reduction range, the load transfer range, and energy conservation. The load reduction range is determined based on the maximum allowable load reduction rate, the reduceable load, and the actual load, while the load transfer range is determined based on the maximum transferable load rate, the transferable load, and the actual load.

[0013] Optionally, the method of employing an adaptive step-size dual-closed-loop model predictive control to optimize and adjust the day-ahead scheduling plan in real time based on the prediction error includes: By monitoring the prediction error between the predicted and actual values ​​of wind and solar power output, runoff and load in real time, the random fluctuation of each uncertainty parameter is determined based on the prediction error value of each uncertainty parameter, and the first decision index of the random variable at the current moment is determined based on the random fluctuation. By monitoring the prediction deviation between the predicted and actual values ​​of the system's operating status in real time, a second decision indicator for the operating status at the current moment is determined based on the deviation value. Based on the weighted combination of the first decision indicator and the second decision indicator, a step size adjustment decision indicator is dynamically generated. The step size adjustment decision index is compared with a preset threshold, and the day-ahead scheduling plan is adaptively optimized and adjusted in real time based on the comparison result.

[0014] Optionally, during the intraday scheduling phase, the day-ahead scheduling plan is optimized and adjusted in real time, and solved using the optimized upper-level objective function. The optimized upper-level objective function is constructed based on the predicted time domain data of electricity transmission, hydrogen transmission, green energy benefits, wind and solar curtailment penalties, carbon emission data, equipment operation and maintenance data, and purchased electricity data during the intraday rolling phase. The predicted time domain is determined based on the current time, the number of control time domains, and the interval of the scheduling time period.

[0015] In a second aspect, this application provides an electronic device, comprising: at least one memory for storing a program; and at least one processor for executing the program stored in the memory, wherein when the program stored in the memory is executed, the processor is configured to execute the method described in the first aspect or any possible implementation thereof.

[0016] Thirdly, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to perform the method described in the first aspect or any possible implementation thereof.

[0017] Fourthly, this application provides a computer program product that, when run on a processor, causes the processor to perform the method described in the first aspect or any possible implementation thereof.

[0018] It is understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0019] Overall, the technical solutions conceived in this application have the following beneficial effects compared with the prior art: (1) In the day-ahead stage, this application adopts an adjustable robust optimization method to quantify the uncertainty of the prediction error into an adjustable range, so that the generated scheduling plan can flexibly balance economic costs and operational risks through robustness coefficients; in the intraday stage, it uses a double closed-loop feedback based on prediction error to dynamically adjust the optimization step size, so as to achieve accurate real-time correction of the day-ahead plan, thereby achieving a balance between the economy and robustness of system operation and improving scheduling accuracy.

[0020] (2) This application is based on an adjustable robust optimization master-slave game optimization scheduling model, which can alleviate the uncertainty of source load through adjustable coefficients. By selecting appropriate adjustable coefficients, a balance between economy and security can be achieved.

[0021] (3) This application can closely track the prediction error of random variables and operating income, and adaptively adjust the rolling time step based on real-time feedback data, which can effectively improve the economic benefits and operating efficiency of the system. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the integrated system multi-timescale scheduling method considering prediction errors provided in the embodiments of this application; Figure 2 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0024] In this article, the term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The symbol " / " in this article indicates that the related objects are in an "or" relationship; for example, A / B means A or B.

[0025] The terms "first" and "second," etc., used in the specification and claims herein are used to distinguish different objects, not to describe a specific order of objects. For example, "first response message" and "second response message," etc., are used to distinguish different response messages, not to describe a specific order of response messages.

[0026] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0027] In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more, for example, multiple processing units means two or more processing units, multiple elements means two or more elements, etc.

[0028] The embodiments of this application are described below with reference to the accompanying drawings.

[0029] Reference Figure 1 This application provides a multi-timescale scheduling method for a comprehensive system that considers prediction errors, including: S101. Establish an operational constraint model for the integrated system; the integrated system includes a wind power system, a photovoltaic subsystem, a hydropower system, and a hydrogen energy storage subsystem; the operational constraint model includes a wind power output model, a photovoltaic power output model, a hydropower output model, and a hydrogen energy storage system model; S102. The uncertain parameters of the integrated system are processed using an adjustable robust optimization method to construct the uncertainty intervals of each uncertain parameter; the uncertain parameters include runoff, wind and solar power output, and load; S103. In the day-ahead scheduling phase, a master-slave game optimization scheduling model is established based on the uncertainty interval and the operational constraint model to generate the day-ahead scheduling plan; S104. During the intraday scheduling phase, an adaptive step-size dual-closed-loop model predictive control method is adopted to optimize and adjust the intraday scheduling plan based on the prediction error, thereby obtaining the optimized scheduling scheme of the integrated system.

[0030] Specifically, step S101 aims to construct a mathematical model that accurately describes the physical operation of the integrated water-wind-solar-hydrogen energy system. Specifically, the operational constraint model includes: a wind power output model, whose output is a piecewise function of real-time wind speed; a photovoltaic power output model, whose output is related to light intensity and ambient temperature; a hydropower output model, which includes complex hydraulic-electric coupling constraints such as water balance, reservoir capacity, outflow, head, and output characteristics; and a hydrogen energy storage system model, describing the energy conversion and mass balance relationships during hydrogen production in electrolyzers and hydrogen storage in hydrogen tanks.

[0031] In step S102, the core is to use adjustable robust optimization coefficients to transform the inherent randomness of runoff, wind and solar power output, and load into a deterministic form that the optimization model can handle. By constructing uncertainty intervals for each uncertain parameter, the fuzzy question of how the parameters might fluctuate is transformed into a clear mathematical set.

[0032] Furthermore, in the day-ahead scheduling phase, a two-layer master-slave game-theoretic optimization scheduling model is established based on the set of operational constraints and uncertainties obtained in the first two steps. The upper layer of this model represents the integrated system operator, whose goal is to optimize its operational strategy while satisfying all robust constraints; the lower layer represents electricity users, whose goal is to adjust their electricity consumption behavior according to the electricity price signals released by the upper layer to optimize electricity costs. The two are coupled and mutually influential through electricity prices and load demand.

[0033] The output of this game theory model is a robust day-ahead scheduling plan. This plan not only specifies in detail the output of each generator unit, the operating mode of the hydrogen energy system, and the power exchange with the grid in the future time period, but also ensures that the system can still operate stably even if the output and load of renewable energy in reality fluctuate adversely within its forecast range, thus significantly improving the day-ahead plan's ability to cope with uncertainty.

[0034] Furthermore, the intraday scheduling phase optimizes the framework of the day-ahead plan. This step employs an adaptive step-size dual-closed-loop model predictive control method. The first closed loop senses the intensity of randomness in the external environment by monitoring the prediction errors of wind and solar power output, runoff, and load. The second closed loop senses the degree of deviation between the actual system operation and the expected operation by monitoring the comprehensive index error of the system's operating status.

[0035] Based on the aforementioned dual closed-loop feedback signals, this method dynamically adjusts the time step size for rolling optimization: when the prediction error is large and the system fluctuates drastically, the step size is automatically shortened to improve control accuracy; when the system is running smoothly, the step size is automatically extended to improve computational efficiency. For optimization problems in the finite time domain, the latest adjustment instructions for the current day's plan are generated, thereby achieving closed-loop feedback and real-time precise control of the entire system.

[0036] It should be noted that the optimization models involved in the embodiments of this application, including the complex nonlinear master-slave game model in step S103 and the model predictive control optimization problem in each rolling window in step S104, are all solved using the high-performance mathematical programming solver Gurobi.

[0037] Gurobi can efficiently handle large-scale, multi-constraint optimization problems. Through its powerful algorithm kernel, it can quickly and accurately calculate the optimal solution or high-quality feasible solution of the model, thereby ensuring the feasibility and timeliness of the entire method in practice.

[0038] Optionally, the method for constructing the wind power output model includes: Based on the real-time wind speed at the hub height of the wind turbine at the target time, and the cut-in wind speed, rated wind speed, and cut-out wind speed of the wind turbine, different output ranges are divided, and the output power of the wind turbine in each output range is determined to construct a set of equations relating power and wind speed as a wind power output model.

[0039] Specifically, the method for constructing the wind power output model in this application embodiment is shown in the following formula:

[0040] in, For wind turbines in Output power at any moment For wind turbines in Real-time wind speed at wheel hub height This refers to the cut-in wind speed of the wind turbine. To cut off the wind speed, Rated wind speed, This refers to the rated power of the wind turbine.

[0041] Optionally, the method for constructing the photovoltaic output model includes: Based on the standard operating temperature of the photovoltaic panel, the ambient temperature and actual light intensity of the photovoltaic panel at the target time, the actual operating temperature of the photovoltaic panel is determined. Based on the actual operating temperature and temperature coefficient of the photovoltaic panel, the power loss caused by the temperature rise at the target time is determined; Based on the power loss, rated output of the photovoltaic panel, standard illuminance, actual illuminance, and converter efficiency, the output power of the photovoltaic panel at the target time is determined, and a set of equations relating power and photovoltaic parameters is constructed as a photovoltaic output model.

[0042] Specifically, the method for constructing the photovoltaic power output model is shown in the following formula:

[0043]

[0044]

[0045] in, For photovoltaic panels in Output power at any moment This refers to the rated output of the photovoltaic panel. and They are respectively in The light intensity at a given time and the standard light intensity, For the efficiency of the DC-DC converter, for Power loss due to temperature rise at all times. For temperature coefficient, The operating temperature of the photovoltaic panel. For solar panels in The ambient temperature at any given time This refers to the standard operating temperature of photovoltaic panels.

[0046] Optionally, the hydropower output model includes at least one of the following constraints: Water balance constraints; Reservoir capacity constraints; Downflow constraint; Hydropower station output constraints; Head constraint; Constraints related to the relationship between reservoir water level and reservoir capacity; Constraints on the relationship between tailwater level and discharge flow.

[0047] Specifically, the method for constructing the hydropower output model in this application embodiment is shown in the following formula: 1) Water balance constraint

[0048]

[0049] in, and They are respectively Time and The reservoir capacity at any given time for The actual inbound flow at any given time for The flow rate at any given moment, for Power generation flow rate at any given moment for The amount of water discarded at any given moment.

[0050] 2) Storage capacity constraints

[0051] in, and These represent the minimum and maximum reservoir capacity, respectively.

[0052] 3) Downflow constraint

[0053] in, and These are the minimum and maximum values ​​of the discharge flow, respectively.

[0054] 4) Power output constraints of hydropower stations

[0055]

[0056] in, for The hydroelectric power station outputs power at all times. A function describing the relationship between power output, power generation flow rate, and water head of a hydropower station. For the water head, and These represent the minimum and maximum output values ​​of the hydropower station, respectively.

[0057] 5) Head constraint

[0058] in, and They are respectively Time and The reservoir water level at any time, for The tailwater level at that moment.

[0059] 6) Reservoir water level-capacity relationship

[0060] in, This represents a function that describes the relationship between water level and reservoir capacity.

[0061] 7) Tailwater level - discharge flow rate relationship

[0062] in, This represents a function that describes the relationship between tailwater level and discharge flow.

[0063] Optionally, the method for constructing the hydrogen energy storage system model includes: A dynamic equilibrium relationship is established based on the current hydrogen storage capacity, the previous hydrogen storage capacity, the current hydrogen production mass, and the current hydrogen sales mass. Based on the unit hydrogen production mass, the output power of the electrolyzer, the working efficiency of the electrolyzer, and the converter efficiency, the hydrogen production mass of the electrolyzer at the current moment is determined to establish a hydrogen production model for the electrolyzer. The upper and lower limits of the hydrogen storage capacity of the hydrogen storage tank are set, and the hydrogen storage capacity boundary conditions at the beginning and end of the scheduling cycle are set for the hydrogen energy storage system.

[0064] Specifically, the hydrogen energy storage system proposed in this application includes a hydrogen storage tank and an electrolyzer, and its relevant constraints are as follows:

[0065]

[0066]

[0067]

[0068] in, and Hydrogen storage tanks Time and The amount of hydrogen stored at all times, Electrolytic cell The mass of hydrogen produced at any given time Hydrogen storage tank The quality of hydrogen sold at any given time The mass of hydrogen that can be produced from 1 kWh of electricity. for The output power of the electrolytic cell at all times To improve the working efficiency of the electrolytic cell. For the efficiency of electrolytic DC-DC converters, and These are the lower and upper limits of the hydrogen storage tank capacity, respectively. and These represent the initial capacity of the hydrogen storage tank and the capacity at the end of the scheduling period, respectively.

[0069] Optionally, an adjustable robust optimization method is used to handle the uncertainty parameters of the synthesized system, including: Based on the predicted maximum and minimum values ​​of each uncertainty parameter, determine the summation mean and the difference mean; An adjustable robustness coefficient is introduced and multiplied by the mean difference to obtain the uncertainty deviation; Based on the uncertainty deviation and the summation mean, the upper limit and lower limit of the uncertainty interval are determined to obtain the uncertainty interval; A robust auxiliary coefficient is introduced to add uncertainty constraints to the uncertainty interval.

[0070] Specifically, this application embodiment utilizes the adjustable robustness coefficient method to handle uncertainties in runoff, wind and solar power output, and load. Taking wind and solar power output as an example, the adjustable robustness coefficient method is detailed below:

[0071]

[0072]

[0073] in, and They are respectively The maximum and minimum values ​​of power output at any given moment. for The mean of the sum of the maximum and minimum output values ​​of the wind power at any given moment. for The mean of the difference between the maximum and minimum output values ​​of the wind power at any given time. for The robustness coefficient of always being in the spotlight.

[0074] Accordingly, the following uncertainty constraints are added to the wind and solar power output:

[0075]

[0076]

[0077]

[0078] in, , and This is an introduced robust auxiliary coefficient. The representation of the uncertainty intervals for runoff and load is consistent with that for wind and solar power output. To ensure that the load considered by operators during the intraday rolling phase can meet user demand, this application sets the initial load as the upper limit of the load uncertainty interval.

[0079] Optionally, the master-slave game optimization scheduling model includes an upper-level sub-model and a lower-level sub-model; the upper-level objective function of the upper-level sub-model is to maximize the operational efficiency index, and the lower-level objective function of the lower-level sub-model is to minimize the user expenditure index. The upper-level objective function is constructed based on data on electricity transmission, hydrogen transmission, green energy benefits, wind and solar curtailment penalties, carbon emission data, equipment operation and maintenance data, and purchased electricity data. The constraints of the upper-level objective function include electricity sales price constraints, power balance constraints, electricity purchase constraints, and wind and solar power output constraints. The lower-level objective function is constructed based on the current electricity price, transferable load, reduceable load, actual load, and subsidy unit price, which is obtained by adjusting flexible load. The lower-level objective function is constrained by the load reduction range, the load transfer range, and energy conservation. The load reduction range is determined based on the maximum allowable load reduction rate, the reduceable load, and the actual load, while the load transfer range is determined based on the maximum transferable load rate, the transferable load, and the actual load.

[0080] During the current scheduling phase, a master-slave game optimization scheduling model based on adjustable robust optimization is established. The main players in the model are the integrated water, wind, solar and hydrogen system operators, and the slaves are the users.

[0081] During the day-ahead scheduling phase, a master-slave game-theoretic optimization scheduling model is established based on the uncertainty handling results of runoff, wind and solar power output, and load. The main players in the model are the integrated hydro-wind-solar-hydropower system operators and the slaves are the users. The upper-level model's optimization objective is to maximize the operator's daily operating revenue, while the lower-level model's optimization objective is to minimize the user's electricity purchase cost. The model expression is as follows: The operator's daily operating revenue maximization model, i.e., the upper-level model, is as follows: 1) Objective function

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092] in, For the operator's daily operating revenue, This refers to electricity sales revenue, i.e., data on electricity transmission to other regions. For data on hydrogen sales revenue and hydrogen energy transmission, Green certificate revenue is equivalent to green energy benefits. The penalty cost for wind and solar power curtailment is the amount of wind and solar power curtailment penalty. This refers to the tiered carbon trading cost, i.e., carbon emission data. The equipment operation and maintenance cost is the equipment operation and maintenance data. This refers to data on the cost of purchasing electricity from the grid, i.e., purchased electricity. for The electricity price at any given time, for Electricity sold to users by operators at all times For the price of hydrogen, This is the price of one green certificate, where one green certificate corresponds to 1 MWh of green electricity. for The actual total output of wind and solar power at any given time For green electricity quota coefficient, For the optimized load, Carbon emission rights obtained through green certificate conversion. This is the green hydrogen quota conversion factor. The penalty coefficient for wind and solar power curtailment. To account for the amount of wind and solar power curtailed, The base price for carbon trading, For the trading volume of carbon emission rights, The length of the carbon emission range, For price growth rate, For the system's carbon quota, Carbon allowance factor for electricity purchased from the grid, Electricity purchased from the power grid, This represents the actual carbon emissions generated by the system. The carbon emission factor for purchasing electricity from the active distribution network. This is a conversion factor, determined by the base trading prices in the carbon market and the green certificate market. This represents the operation and maintenance cost coefficient for wind turbines and solar panels. This represents the operation and maintenance cost coefficient for hydropower units. This represents the operation and maintenance cost coefficient of the electrolytic cell. This refers to the electricity purchase price.

[0093] 2) The constraints are as follows: Electricity price constraints:

[0094]

[0095] in, For grid connection electricity price, The electricity price sold by the power grid. This represents the maximum permissible average electricity price.

[0096] Power balance constraints:

[0097] Electricity purchase constraints:

[0098] in, This represents the maximum amount of electricity that an operator can purchase from the grid.

[0099] Wind and solar power output constraints:

[0100]

[0101] in, and They are respectively Always consider the wind and solar power output and runoff after uncertainties.

[0102] The model for minimizing user electricity purchase costs, i.e., the lower-level model, is as follows: Objective function:

[0103]

[0104] in, For users' electricity purchase costs, To obtain the subsidy unit price by adjusting flexible load, for Transferable load at any time, for The load can be reduced at any time. for The actual load at any given time.

[0105] 2) Constraints

[0106]

[0107]

[0108] in, The maximum allowable load reduction rate, This represents the maximum transferable load factor.

[0109] Optionally, during the intraday scheduling phase, the day-ahead scheduling plan is optimized and adjusted in real time, and solved using the optimized upper-level objective function. The optimized upper-level objective function is constructed based on the predicted time domain data of electricity transmission, hydrogen transmission, green energy benefits, wind and solar curtailment penalties, carbon emission data, equipment operation and maintenance data, and purchased electricity data during the intraday rolling phase. The predicted time domain is determined based on the current time, the number of control time domains, and the interval of the scheduling time period.

[0110] Specifically, in this embodiment, during the intraday scheduling phase, an intraday rolling scheduling method based on adaptive step-size dual closed-loop model predictive control is proposed to adjust the day-ahead scheduling plan of the integrated system operator.

[0111] During the intraday scheduling phase, an intraday rolling scheduling method based on adaptive step-size dual-closed-loop model predictive control is proposed to adjust the day-ahead scheduling plan of the integrated system operator. During the intraday scheduling phase, the objective function remains maximizing the operator's daily operating revenue, and its model expression is as follows: 1) Objective function

[0112]

[0113]

[0114]

[0115]

[0116]

[0117]

[0118]

[0119]

[0120]

[0121]

[0122] in, For the operator's operating revenue during the intraday rolling phase, For various revenues and costs during the intraday rolling phase, For the current moment, To predict the number of control time domains contained in the time domain, The interval for scheduling is equivalent to the control time domain.

[0123] Optionally, the method of employing an adaptive step-size dual-closed-loop model predictive control to optimize and adjust the day-ahead scheduling plan in real time based on the prediction error includes: By monitoring the prediction error between the predicted and actual values ​​of wind and solar power output, runoff and load in real time, the random fluctuation of each uncertainty parameter is determined based on the prediction error value of each uncertainty parameter, and the first decision index of the random variable at the current moment is determined based on the random fluctuation. By monitoring the prediction deviation between the predicted and actual values ​​of the system's operating status in real time, a second decision indicator for the operating status at the current moment is determined based on the deviation value. Based on the weighted combination of the first decision indicator and the second decision indicator, a step size adjustment decision indicator is dynamically generated. The step size adjustment decision index is compared with a preset threshold, and the day-ahead scheduling plan is adaptively optimized and adjusted in real time based on the comparison result.

[0124] Specifically, in addition to meeting the day-ahead constraints, the equipment must ensure that its intraday operating status is consistent with the status determined in the day-ahead phase.

[0125] The adaptive step-size dual-feedback model predictive control method refers to dynamically adjusting domain parameters, including the time step, simultaneously using the prediction errors of random variables and operating revenue, and adding feedback from the prediction errors of random variables and operating revenue, thus forming a dual closed-loop feedback mechanism. The random variables include runoff, wind power output, photovoltaic power output, and load. The adaptive step-size strategy is as follows:

[0126]

[0127] in, To predict the time domain, To control the time domain, for Time step of the moment This is the minimum sampling interval. The decision index is adjusted adaptively based on a variable step size. The expression for the decision index is as follows:

[0128]

[0129]

[0130] in, In order to be in Time prediction Every moment of glory is a testament to hard work. In order to be in The true value of consistently glamorous contributions. In order to be in Time prediction The flow of time, In order to be in The true value of runoff at any given time. exist Time prediction The load of time, In order to be in The actual value of the load at any given time. In order to be in Time prediction Revenue per moment In order to be in The true value of constantly running income. for Variable step size decision indicators at any given time. For random variables in Decision indicators at any given moment For operating revenue in Decision indicators at any given moment and , where is the decision coefficient.

[0131] Based on the above decision-making indicators, The following adjustments have been made:

[0132] when When the time step decreases to ,when When the time step increases to .

[0133] Furthermore, to illustrate the superiority of the adaptive step-size dual-feedback model predictive control method proposed in the embodiments of this application, the following three comparison scenarios were constructed, and the comparison results are shown in Table 1.

[0134] Scenario 1: The adaptive step-size dual-feedback model predictive control method proposed in this application.

[0135] Scenario 2: Adaptive step-size single feedback model predictive control method.

[0136] Scenario 3: Traditional model predictive control method.

[0137] Table 1. Results of running scenarios 1-3

[0138] As shown in Table 1, Scenario 1 fully utilizes renewable energy, load information, and operating revenue to adjust the time step and scheduling plan, resulting in significantly better scheduling performance compared to Scenario 2 and Scenario 3. Based on the day-ahead scheduling plan, Scenario 2 performs intraday rolling and real-time adjustments to gradually reduce operational deviations, increasing operating revenue by 0.62% compared to Scenario 2. Unlike the traditional model predictive control method proposed in Scenario 3, the adaptive step-size dual-feedback model predictive control method utilizes dual closed-loop feedback of renewable energy, load, and operating revenue prediction errors to achieve adaptive adjustment of the time step, thereby improving scheduling accuracy and operating revenue, with an increase of 0.73% in operating revenue.

[0139] To illustrate the impact of uncertainty on scheduling results, different robustness coefficients were set, and the scheduling results are shown in Table 2.

[0140] Table 2 Scheduling results under different robustness coefficients

[0141] As shown in Table 2, the robustness of the system increases with the increase of the robustness coefficient. The larger the robustness coefficient, the more conservative the decisions made by the system. Therefore, the operator's operating revenue continues to decrease, while the user's operating costs continue to increase. This is because after the operator assumes the uncertainty risk of renewable energy, the change in its revenue will be passed on to users through pricing strategies. In addition, users' own electricity demand will also respond elastically to changes in electricity prices. Ultimately, users' operating costs are affected by both electricity prices and electricity consumption. In particular, when the operator's operating costs increase, it will pass on this cost to users by raising electricity prices, thus leading to an increase in user costs. In actual dispatching, the robustness coefficient can be selected according to the risk level acceptable to both operators and users.

[0142] Reference Figure 2 Based on the methods described in the above embodiments, this application provides an electronic device that may include a processor 210, a communications interface 220, a memory 230, and a communication bus 240. The processor 210, communications interface 220, and memory 230 communicate with each other via the communication bus 240. The processor 210 may call logical instructions stored in the memory 230 to execute the methods described in the above embodiments.

[0143] Furthermore, the logical instructions in the aforementioned memory 230 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0144] Based on the methods in the above embodiments, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0145] Based on the methods in the above embodiments, this application provides a computer program product that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0146] It is understood that the processor in the embodiments of this application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0147] The method steps in this application embodiment can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which 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, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from 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 reside in an ASIC.

[0148] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially as 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, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0149] It is understood that the various numerical designations used in the embodiments of this application are merely for the convenience of description and are not intended to limit the scope of the embodiments of this application.

[0150] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A multi-time-scale scheduling method for a comprehensive system considering prediction errors, characterized in that, include: An operational constraint model for the integrated system is established; the integrated system includes a wind power system, a photovoltaic subsystem, a hydropower system, and a hydrogen energy storage subsystem; the operational constraint model includes a wind power output model, a photovoltaic power output model, a hydropower output model, and a hydrogen energy storage system model; An adjustable robust optimization method is used to process the uncertain parameters of the integrated system and to construct the uncertainty intervals for each uncertain parameter; the uncertain parameters include runoff, wind and solar power output, and load. During the day-ahead scheduling phase, a master-slave game optimization scheduling model is established based on the aforementioned uncertainty interval and operational constraint model to generate the day-ahead scheduling plan; During the intraday scheduling phase, an adaptive step-size dual-closed-loop model predictive control method is adopted to optimize and adjust the intraday scheduling plan based on the prediction error, thereby obtaining the optimized scheduling scheme of the integrated system.

2. The integrated system multi-timescale scheduling method considering prediction errors according to claim 1, characterized in that, The method for constructing the wind power output model includes: Based on the real-time wind speed at the hub height of the wind turbine at the target time, and the cut-in wind speed, rated wind speed, and cut-out wind speed of the wind turbine, different output ranges are divided, and the output power of the wind turbine in each output range is determined to construct a set of equations relating power and wind speed as a wind power output model.

3. The integrated system multi-timescale scheduling method considering prediction errors according to claim 1, characterized in that, The method for constructing the photovoltaic power output model includes: Based on the standard operating temperature of the photovoltaic panel, the ambient temperature and actual light intensity of the photovoltaic panel at the target time, the actual operating temperature of the photovoltaic panel is determined. Based on the actual operating temperature and temperature coefficient of the photovoltaic panel, the power loss caused by the temperature rise at the target time is determined; Based on the power loss, rated output of the photovoltaic panel, standard illuminance, actual illuminance, and converter efficiency, the output power of the photovoltaic panel at the target time is determined, and a set of equations relating power and photovoltaic parameters is constructed as a photovoltaic output model.

4. The integrated system multi-timescale scheduling method considering prediction errors according to claim 1, characterized in that, The hydropower output model includes at least one of the following constraints: Water balance constraints; Reservoir capacity constraints; Downflow constraint; Hydropower station output constraints; Head constraint; Constraints related to the relationship between reservoir water level and reservoir capacity; Constraints on the relationship between tailwater level and discharge flow.

5. The integrated system multi-time-scale scheduling method considering prediction errors according to claim 1, characterized in that, The method for constructing the hydrogen energy storage system model includes: A dynamic equilibrium relationship is established based on the current hydrogen storage capacity, the previous hydrogen storage capacity, the current hydrogen production mass, and the current hydrogen sales mass. Based on the unit hydrogen production mass, the output power of the electrolyzer, the working efficiency of the electrolyzer, and the converter efficiency, the hydrogen production mass of the electrolyzer at the current moment is determined to establish a hydrogen production model for the electrolyzer. The upper and lower limits of the hydrogen storage capacity of the hydrogen storage tank are set, and the hydrogen storage capacity boundary conditions at the beginning and end of the scheduling cycle are set for the hydrogen energy storage system.

6. The integrated system multi-timescale scheduling method considering prediction errors according to claim 1, characterized in that, The uncertainties of the synthesized system are addressed using an adjustable robust optimization method, including: Based on the predicted maximum and minimum values ​​of each uncertainty parameter, determine the summation mean and the difference mean; An adjustable robustness coefficient is introduced and multiplied by the mean difference to obtain the uncertainty deviation; Based on the uncertainty deviation and the summation mean, the upper limit and lower limit of the uncertainty interval are determined to obtain the uncertainty interval; A robust auxiliary coefficient is introduced to add uncertainty constraints to the uncertainty interval.

7. The integrated system multi-time-scale scheduling method considering prediction errors according to claim 1, characterized in that, The master-slave game optimization scheduling model includes an upper-level sub-model and a lower-level sub-model; the upper-level objective function of the upper-level sub-model is to maximize the operational efficiency index, and the lower-level objective function of the lower-level sub-model is to minimize the user expenditure index. The upper-level objective function is constructed based on data on electricity transmission, hydrogen transmission, green energy benefits, wind and solar curtailment penalties, carbon emission data, equipment operation and maintenance data, and purchased electricity data. The constraints of the upper-level objective function include electricity sales price constraints, power balance constraints, electricity purchase constraints, and wind and solar power output constraints. The lower-level objective function is constructed based on the current electricity price, transferable load, reduceable load, actual load, and subsidy unit price, which is obtained by adjusting flexible load. The lower-level objective function is constrained by the load reduction range, the load transfer range, and energy conservation. The load reduction range is determined based on the maximum allowable load reduction rate, the reduceable load, and the actual load, while the load transfer range is determined based on the maximum transferable load rate, the transferable load, and the actual load.

8. The integrated system multi-time-scale scheduling method considering prediction errors according to claim 1, characterized in that, The adaptive step-size dual-closed-loop model predictive control method optimizes and adjusts the day-ahead scheduling plan in real time based on the prediction error, including: By monitoring the prediction error between the predicted and actual values ​​of wind and solar power output, runoff and load in real time, the random fluctuation of each uncertainty parameter is determined based on the prediction error value of each uncertainty parameter, and the first decision index of the random variable at the current moment is determined based on the random fluctuation. By monitoring the prediction deviation between the predicted and actual values ​​of the system's operating status in real time, a second decision indicator for the operating status at the current moment is determined based on the deviation value. Based on the weighted combination of the first decision indicator and the second decision indicator, a step size adjustment decision indicator is dynamically generated. The step size adjustment decision index is compared with a preset threshold, and the day-ahead scheduling plan is adaptively optimized and adjusted in real time based on the comparison result.

9. The integrated system multi-timescale scheduling method considering prediction errors according to claim 7, characterized in that, During the intraday scheduling phase, the day-ahead scheduling plan is optimized and adjusted in real time, and the optimized upper-level objective function is used to solve for it. The optimized upper-level objective function is constructed based on the predicted time domain data of electricity transmission, hydrogen transmission, green energy benefits, wind and solar curtailment penalties, carbon emission data, equipment operation and maintenance data, and purchased electricity data during the intraday rolling phase. The predicted time domain is determined according to the current time, the number of control time domains, and the interval of the scheduling time period.

10. An electronic device, characterized in that, include: At least one memory for storing computer programs; At least one processor is configured to execute a program stored in the memory, wherein when the program stored in the memory is executed, the processor is configured to perform the method as described in any one of claims 1-9.

Images