Multi-time-scale layered decoupling time sequence operation simulation method considering extreme scene and related device

Abstract

The invention discloses a multi-time-scale layered decoupling time sequence operation simulation method considering an extreme scene and a related device, and relates to the field of power system operation simulation, and the method comprises the steps: obtaining the annual wind power, photovoltaic and hydropower output and load data of a power system in a horizontal year to form an original data set; preprocessing and dimensionality reduction are carried out on the original data set, a typical scene is selected by adopting a clustering method based on the original data set after dimensionality reduction compression, and an extreme scene is iteratively selected through a scene difference degree maximization principle; constructing a combined scene set based on the typical scene and the extreme scene, and establishing a power system time sequence operation simulation model considering the extreme scene and the maintenance plan; based on the model, performing multi-time scale layered decoupling operation simulation; based on the operation coupling constraint, after the consistency of long-time and short-time scale power balance simulation is determined, a capacity matching result is obtained, the solving efficiency can be improved, and meanwhile supply and demand balance of the system under the multi-time scale is guaranteed.

Description

Technical Field

[0001] This application relates to the field of power system operation simulation, and in particular to a multi-timescale hierarchical decoupled time-series operation simulation method and related apparatus that considers extreme scenarios. Background Technology

[0002] Currently, power balance analysis typically employs an 8760-hour time-series operation simulation model based on unit combination. However, with the growth of power system resources, the spatiotemporal coupling variables and constraints lead to slow convergence speeds and memory overflows, limiting its application in relevant scenarios. To improve the computational efficiency of operation simulation, a typical scenario method is often adopted, which selects a small number of representative scenarios to replace massive operation scenarios, thereby reducing the decision variables and constraints of the model. However, the power balance method based on typical scenarios has the following problems: First, when renewable energy reaches a certain proportion, seasonal power imbalances become increasingly apparent, and typical daily operation simulations are insufficient to measure long-term power shortages. Second, operation simulations based on typical scenarios cannot model the temporal coupling characteristics of multi-timescale energy storage.

[0003] Furthermore, with the increasing penetration rate of renewable energy, power balance analysis considering only typical annual scenarios is insufficient to guarantee safe and stable operation under extreme weather conditions. It is urgent to incorporate extreme weather scenarios into power balance simulations. Currently, methods primarily rely on experience or selecting the furthest point from the cluster center as the extreme scenario. However, experience-based methods cannot identify all extreme situations through rules, such as the timing mismatch between load demand and renewable energy output; selecting the point farthest from the cluster center as the extreme scenario fails to account for the existence of different types of extreme scenarios scattered at the edges of the same cluster. The selection of extreme scenarios serves only as a supplement to typical scenarios. As the penetration rate of renewable energy increases, the types of extreme scenarios will become more complex and diverse, requiring a systematic consideration of extreme scenario selection methods.

[0004] Therefore, constructing a high-proportion renewable energy power system operation simulation model that considers extreme scenarios and achieving accurate and rapid solution of the model is of great significance for power system planning and medium- and long-term power generation planning, and for ensuring the reliable operation of the power system. Summary of the Invention

[0005] The purpose of this application is to provide a multi-timescale hierarchical decoupled time-series operation simulation method and related devices that consider extreme scenarios. It can combine typical scenarios and extreme scenarios to construct combined scenarios that consider extreme scenario optimization; establish a power system time-series operation simulation model that considers extreme scenarios and maintenance plans, fully consider the coordinated operation of traditional power sources, renewable energy sources and multi-timescale energy storage resources, and realize the joint optimization of power generation plans and maintenance plans; and perform multi-timescale hierarchical decoupling on the annual 8760h operation simulation, which improves the solution efficiency while fully ensuring the supply and demand balance of the system under multiple time scales, and effectively simulates the energy transfer characteristics of energy storage under multiple time scales.

[0006] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a multi-timescale hierarchical decoupled temporal simulation method considering extreme scenarios, including: The raw dataset is constructed by acquiring annual wind power, photovoltaic, and hydropower output and load data for the power system. The original dataset is preprocessed and dimensionality reduced to obtain the original dataset after dimensionality reduction and compression. Based on the original dataset after dimensionality reduction and compression, a clustering method is used to select typical scenarios, and extreme scenarios are iteratively selected by maximizing scenario differences. A combined scenario set is constructed based on selected typical and extreme scenarios; Based on the aforementioned combined scenario set, a power system time-series operation simulation model considering extreme scenarios and maintenance plans is established; the power system time-series operation simulation model takes the minimization of power system operating costs as its objective function; the system operation constraints of the power system time-series operation simulation model include power balance constraints, thermal power unit operation constraints, nuclear power unit operation constraints, renewable energy unit operation constraints, multi-timescale energy storage operation constraints, and maintenance plan constraints; Based on the aforementioned power system time-series operation simulation model, multi-time-scale hierarchical decoupled operation simulation is performed. This multi-time-scale hierarchical decoupled operation simulation decouples the annual operation time of the power system into long-term power balance simulation and short-term power balance simulation. The long-term power balance simulation performs an annual simulation on a daily basis to optimize the daily utilization hours of generating units and the daily energy transfer of energy storage. The short-term power balance simulation performs an intraday simulation on the aforementioned combined scenario set on an hourly basis to optimize the output of generating units and the power of energy storage. Based on operational coupling constraints, after ensuring the consistency between long-term and short-term power balance simulations, the capacity allocation results of the power system are obtained.

[0007] Optionally, the original dataset may be preprocessed and dimensionality reduced, specifically including: According to the formula The original dataset is standardized to obtain a standardized dataset; wherein, σ f Features F f variance X sc The standardized dataset; N sl The number of dimensions of the feature variables for each scenario; According to the formula Calculate the covariance matrix of the standardized dataset; Perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalued matrix. Among them, the eigenvalues ​​are λ 1, λ 1, ..., The feature vector is μ λ1 , μ λ1 ,..., ; Before choosing K Each feature variable represents the original data, according to the formula Determine the proportion of data variance to retain; Determine the feature matrix used to characterize the original data Finally, the dimensionality-reduced and compressed dataset is obtained. .

[0008] Optionally, the formula for combining the scene set is: ; In the formula, For extreme scenarios, N represents the number of typical scenarios. n N T These are the number of nodes and the scene duration, respectively. d , w , pv, hy These are respectively load demand, wind power output, photovoltaic power output, and hydropower output.

[0009] Optionally, the formula expression of the objective function is: ; in, T For collection of times within the year, t The index corresponds to the time. G , RE They are collections of thermal power units and renewable energy units, g , re For the corresponding unit subscript; , , , These represent the fuel cost, start-up cost, shutdown cost, and startup cost of a thermal power unit, respectively. , These represent the costs of curtailment and load shedding of renewable energy units, respectively. , These are the power curtailment from renewable energy units and the system load shedding power, respectively. , , , These are respectively the output of the thermal power unit, the start-up and shutdown status variables, the shutdown action variables, and the start-up action variables.

[0010] Optionally, the multi-timescale energy storage operation constraints consist of energy transfer constraints, upper and lower limits of charge / discharge power constraints, state of charge / discharge constraints, upper and lower limits of capacity constraints, and beginning and end energy balance constraints. The formulas for energy transfer constraints, upper and lower limits of charge / discharge power constraints, state of charge / discharge constraints, upper and lower limits of capacity constraints, and initial and final energy balance constraints are as follows: ; ; ; In the formula, , , These represent the energy storage capacity, charging power, and discharging power of the energy storage unit, respectively. , The state variables representing the charging and discharging power of the energy storage unit are 0. , / , / , These are the upper limit of energy storage capacity, the upper limit of charging / discharging power capacity, the lower limit of charging / discharging power capacity, and the minimum energy storage ratio for energy storage units. , These are the initial energy storage capacity and the final energy storage capacity, respectively. , , These are the self-release rate, charging power, and discharging efficiency of the energy storage unit, respectively. This refers to the time granularity of the energy storage unit's operation.

[0011] Optionally, the formulas for power balance constraints, thermal power unit operating constraints, nuclear power unit operating constraints, and renewable energy unit operating constraints are as follows: ; ; ; ; ; ; ; In the formula, N , ES It is a collection of nuclear power units and energy storage units. n , es For the corresponding unit subscript; , For nuclear power unit output and system load, , Output and upper limit of renewable energy units; , , / These are the minimum output ratio, ramp rate, and shortest continuous start-up / shutdown time for thermal power units, respectively. Minimum output ratio of nuclear power units / These represent the downhill and uphill ramp rates of thermal power units, / These represent the downhill and uphill ramp rates of the nuclear power unit, respectively.

[0012] Optionally, the maintenance plan constraints include maintenance start time constraints, maintenance status constraints, sequential maintenance constraints, mutually exclusive maintenance constraints, simultaneous maintenance constraints, and maintenance resource constraints.

[0013] Secondly, this application provides a multi-timescale hierarchical decoupled timing simulation device considering extreme scenarios, comprising: The data acquisition module is used to acquire annual wind power, photovoltaic, and hydropower output and load data of the power system to form the original dataset. The data processing module is used to preprocess and reduce the dimensionality of the original dataset to obtain the original dataset after dimensionality reduction and compression. The scene selection module is used to select typical scenes based on the original dataset after dimensionality reduction and compression, and to iteratively select extreme scenes by maximizing scene difference. The combination module is used to construct a set of combined scenarios based on selected typical and extreme scenarios; The model building module is used to establish a power system time-series operation simulation model that considers extreme scenarios and maintenance plans based on the combined scenario set. The power system time-series operation simulation model takes the minimization of power system operating costs as its objective function. The system operation constraints of the power system time-series operation simulation model include power balance constraints, thermal power unit operation constraints, nuclear power unit operation constraints, renewable energy unit operation constraints, multi-timescale energy storage operation constraints, and maintenance plan constraints. The simulation module is used to perform multi-timescale layered decoupled operation simulation based on the power system time-series operation simulation model. The multi-timescale layered decoupled operation simulation decouples the annual operation time of the power system into long-term power balance simulation and short-term power balance simulation. The long-term power balance simulation performs an annual simulation on a daily basis to optimize the daily utilization hours of generating units and the daily energy transfer of energy storage. The short-term power balance simulation performs intraday simulation on the combined scenario set on an hourly basis to optimize the unit output and energy storage power. The output module is used to determine the consistency between long-term and short-term power balance simulations based on operational coupling constraints, and then obtain the capacity allocation results of the power system.

[0014] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the multi-timescale hierarchical decoupled timing simulation method considering extreme scenarios described in any one of the above applications.

[0015] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the multi-timescale hierarchical decoupled timing simulation method considering extreme scenarios described above.

[0016] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a multi-timescale hierarchical decoupled time-series operation simulation method and related apparatus considering extreme scenarios. It acquires and reduces the dimensionality of wind, solar, and hydropower output and load data; constructs a combined scenario set by selecting typical and extreme scenarios through clustering and maximizing dissimilarity; establishes a time-series operation simulation model with the goal of minimizing operating costs, including power balance, unit operation, energy storage, and maintenance constraints; performs hierarchical decoupled simulations at long timescales (daily power balance) and short timescales (hourly power balance); and outputs capacity allocation results after verifying the consistency between the two scales. This application significantly improves solution efficiency while ensuring computational accuracy through clear timescale division, model simplification based on data scenarios, and coupling constraint design. It transforms the originally coupled, year-round continuous problem into two parallel computational levels: daily energy optimization and intraday power optimization, while simultaneously achieving multi-scale characteristic simulation through dual representation of energy storage constraints. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is an overall flowchart of a multi-timescale hierarchical decoupling time-series simulation method considering extreme scenarios, according to an embodiment of this application. Figure 2 The logical framework for constructing combined scenarios that consider optimization for extreme scenarios in the embodiments of this application is as follows: Figure 3 This is a schematic diagram illustrating the selection of typical and extreme scenarios in the embodiments of this application; Figure 4 This is a schematic diagram of multi-timescale hierarchical decoupling in the 8760-hour time-series simulation of this application embodiment; Figure 5 This application provides a multi-timescale simulation model framework that considers extreme scenarios in its embodiments. Figure 6 This is a schematic diagram illustrating the long-term and short-term operational coupling relationship of long-term energy storage in the embodiments of this application; Figure 7 This is a comparison chart of the optimization results of different maintenance plans in the embodiments of this application; Figure 8 This is a comparison chart of the daily power balance operation results of different calculation examples in this application embodiment; Figure 9 This is a comparison chart of the typical running results of different examples in the embodiments of this application; Figure 10This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, this application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] In one exemplary embodiment, such as Figure 1 As shown, a multi-timescale hierarchical decoupled timing simulation method considering extreme scenarios is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server, and includes the following steps 101 to 107. Wherein: Step 101: Obtain the annual wind power, photovoltaic, and hydropower output and load data of the power system to form the original dataset; Step 102: Preprocess and reduce the dimensionality of the original dataset to obtain the original dataset after dimensionality reduction and compression; Step 103: Based on the original dataset after dimensionality reduction and compression, a clustering method is used to select typical scenarios, and extreme scenarios are iteratively selected based on the principle of maximizing scenario differences. Step 104: Construct a combined scenario set based on the selected typical and extreme scenarios; Step 105: Based on the combined scenario set, establish a power system time-series operation simulation model that considers extreme scenarios and maintenance plans; the power system time-series operation simulation model takes the minimization of power system operating costs as its objective function; the system operation constraints of the power system time-series operation simulation model include power balance constraints, thermal power unit operation constraints, nuclear power unit operation constraints, renewable energy unit operation constraints, multi-timescale energy storage operation constraints, and maintenance plan constraints; Step 106: Based on the power system time-series operation simulation model, perform multi-time-scale layered decoupled operation simulation; the multi-time-scale layered decoupled operation simulation decouples the annual operation time of the power system into long-term scale power balance simulation and short-term scale power balance simulation; the long-term scale power balance simulation performs annual simulation on a daily basis to optimize the daily utilization hours of generating units and the daily energy transfer of energy storage; the short-term scale power balance simulation performs intraday simulation on the combined scenario set on an hourly basis to optimize the unit output and energy storage power; Step 107: Based on operational coupling constraints, after determining the consistency between long-term power balance simulation and short-term power balance simulation, the capacity allocation result of the power system is obtained.

[0022] In an exemplary embodiment, when performing steps 101-107, the specific steps may be as follows: Step 101 specifically includes: The basic dataset is constructed by selecting wind power, solar power, and hydropower output and load data for each node throughout the entire 8760-hour period of the system's horizontal year. X Define the dataset. X Scenes in x As shown in the following formula: (1).

[0023] In the formula, d , w , pv, hy These refer to load demand, wind power, solar power output, and hydropower processing, respectively. N n , N T These represent the number of nodes and the duration of the scene, respectively. Scene Duration N T The selection of a timeframe is generally determined by wind and solar load characteristics. To highlight its intraday volatility, a 24-hour period is chosen as the duration of a single scene. N T = 24. Dataset X contains N s There are 1 scenario, and the number of dimensionality of the feature variables for each scenario is... N sl express, N sl = 3 N n N T .

[0024] Step 102 specifically includes: Principal Component Analysis (PCA) is used to preprocess the data to improve data processing efficiency. The process is as follows: Equation (2) standardizes the features using z-score. σ f Features F f variance X sc The dataset is standardized; Equation (3) calculates the dataset. X sc The covariance matrix is ​​then obtained; then, the covariance matrix is ​​decomposed into eigenvalues ​​in equation (4) from largest to smallest. λ 1, λ 1, ..., and the corresponding feature vectors μ λ1 , μ λ1 ,..., The matrix formed; when selected K When each feature variable represents the original data, the variance ratio of the retained data is shown in Equation (5); further, the feature matrix used to represent the original data is determined. The final dimensionality-reduced and compressed dataset is shown in Equation (6).

[0025] (2).

[0026] (3).

[0027] (4).

[0028] (5).

[0029] (6).

[0030] Step 103 specifically includes: The K-medoids clustering method is used to select typical scenarios so that the final cluster centers are the actual scenarios, rather than the average of each scenario. This avoids outliers interfering with the selection of cluster centers. The process is as follows.

[0031] 1) Set the dataset X r The various scenes in the text are divided into different clusters, represented as follows: Ω 1, Ω 2, ..., Ω Np The corresponding number of various scenarios is represented as 101 , S 2, ...,S Np The classification objective of the K-medoids algorithm is to minimize the Euclidean distance from all scenes in each cluster to its cluster center, i.e. (7).

[0032] In the formula, o i It is a cluster Ω i The center point, x s For the dimensionality-reduced dataset X r One of the scenes.

[0033] 2) The silhouette coefficient method is used to determine the number of typical scenes. The core requirement is that scenes within the same cluster have sufficiently small differences, while scenes in different clusters have sufficiently large differences. For a given number of categories... N p Scene x s The contour coefficient is: (8).

[0034] In the formula, a s For the scene x s The average distance to other scenes in the same cluster; this value measures the cluster density. b s For the scene x s The silhouette coefficient is the average distance to the nearest cluster of scenes, measuring the inter-cluster separation. It's easy to see that the silhouette coefficient ranges from -1 to 1. For a single scene, a larger silhouette coefficient indicates a stronger scene... x s The stronger the compactness within its own cluster and the higher the separation from other clusters, the better the clustering effect.

[0035] 3) Given the number of clusters N p Calculate the average contour coefficient for all scenes, and select the cluster number that maximizes the contour coefficient as the final number of typical scenes. ,Right now: (9).

[0036] In the formula: (10).

[0037] Based on the above clustering algorithm and the principle for determining the number of clusters, the cluster center of each cluster is selected. o iAs a typical scenario, the typical scenario dataset is constructed as follows: (11).

[0038] This embodiment proposes an iterative selection method based on maximizing scene difference. Its key feature is that when selecting a representative extreme scene each time, it considers not only its difference from typical scenes but also its difference from existing extreme scenes, ensuring that the newly selected extreme scene provides the model with the maximum incremental information. Some definitions in the method are as follows: Definition 1.1 (Closest Points and Discrete Points): For a given cluster obtained from clustering... Ω i Select a threshold value for all points in the dataset. ε Define all and cluster centers o i The Euclidean distance is less than or equal to ε The points in the cluster are compact points, and the other points in the cluster are discrete points.

[0039] Definition 1.2 (Scene Differentiation): Defines a specific scene in the dataset. x s To scene set Ω The average Euclidean distance across all scenes is the scene's distance. x s Compared to scene set Ω Scene differences R s : (12).

[0040] Based on the selected typical scenarios, each cluster was chosen according to the principle of maximizing scenario differences. Ω i The most representative extreme scenario in the process is as follows: Step 1) Calculate the first i All points in the cluster and the cluster center o i The Euclidean distances are arranged in ascending order: ≤ ≤....≤ ; Step 2) Select a threshold ε Combining the distance sorting and definition 1.1 above, the first... i All points in a cluster are divided into compact points and discrete points, forming sets of compact points respectively. and discrete point set And establish a scene set. Its initial element is the cluster center. o i ; Step 3) For the set of discrete points For all points in the set, calculate the relationship between each point and the scene set according to Definition 1.2. The scene difference is defined as the point with the greatest scene difference as an extreme scene. x i,j ; Step 4) Extreme scenarios x i,j Add to scene collection Then, according to the method in step 1), calculate the set of discrete points. From the remaining points to extreme scenarios x i,j The Euclidean distance will be less than or equal to ε The point is moved to the close set of points in this extreme scenario. ; Step 5) Determine the set of discrete points Is it an empty set? If it is an empty set, end; otherwise, return to step 3).

[0041] Step 104 specifically includes: Based on the typical and extreme scenarios selected in the above steps, a combined scenario set is constructed for solving the power system operation simulation. The constructed combined scenario set is represented as follows: (13).

[0042] In the formula, Number of extreme scenarios, number of combined scenarios = + For each scenario, a weight is assigned based on its frequency of occurrence. The weight is defined by the number of close points contained in a representative scenario (typical scenario or extreme scenario), i.e., a typical scenario. o i The weight is the number of elements in its close set. |, extreme scenarios x i,j The weight is the number of elements in its close set. The combined scene weight vector is represented as: (14).

[0043] In the formula, , The first i A typical scenario and the first j The weights for each extreme scenario are given, and the sum of the elements in the weight vector represents the total number of scenarios within the simulation cycle. (15).

[0044] Step 105 specifically includes: Using system operating cost as the objective function, and considering the operating cost of conventional units, load shedding, and the penalty cost of renewable energy curtailment for a comprehensive analysis of the synergy of multiple energy systems, as shown in equation (16). (16).

[0045] In the formula, T For collection of times within the year, t This is the index for the corresponding time. G , RE They are collections of thermal power units and renewable energy units, g , re This is the subscript for the corresponding generator unit. , , , These represent the fuel cost, start-up cost, shutdown cost, and startup cost of a thermal power unit, respectively. , These represent the costs of curtailment and load shedding of renewable energy units, respectively. , These are the power curtailment from renewable energy units and the system load shedding power, respectively. , , , These are the output of the thermal power unit, the start-up and shutdown status (0-1 variables, where 1 represents the start-up status), the shutdown action (0-1 variables, where 1 represents the shutdown at time t), and the start-up action (0-1 variables, where 1 represents the start-up at time t).

[0046] To facilitate collaborative analysis of different types of energy storage across multiple time scales, a generalized model applicable to energy storage across multiple time scales was established. As shown in equations (17)-(19), these represent the energy transfer constraints, upper and lower limits of charge / discharge power constraints, state of charge / discharge constraints, upper and lower limits of capacity constraints, and initial and final energy balance constraints, respectively.

[0047] (17).

[0048] (18).

[0049] (19).

[0050] In the formula, , , These represent the energy storage capacity and charging / discharging power of the energy storage unit, respectively. , The state of charge / discharge power of the energy storage unit is represented by a 0-1 variable. , / , / , These are the upper limit of energy storage capacity, the upper limit of charging / discharging power capacity, the lower limit of charging / discharging power capacity, and the minimum energy storage ratio for energy storage units. , For initial energy storage capacity and final energy storage capacity, , , For the self-release rate and charge / discharge efficiency of energy storage units, This refers to the time granularity of the energy storage unit's operation.

[0051] Then, a collaborative operation model for traditional power sources, renewable energy sources, and multi-timescale energy storage is established. Equations (20)-(26) represent the system power balance constraints, the upper and lower limits of thermal power unit output, the unit combination constraints and ramping constraints, the upper and lower limits of nuclear power unit output and ramping constraints, and the operation constraints of renewable energy units, respectively. The meanings of energy storage operation constraints and related variables are referenced in equations (17)-(19).

[0052] (20).

[0053] (twenty one).

[0054] (twenty two).

[0055] (twenty three).

[0056] (twenty four).

[0057] (25).

[0058] (26).

[0059] In the formula, N , ES It is a collection of nuclear power units and energy storage units. n , es This is the subscript for the corresponding generator unit. , For nuclear power unit output and system load, , The power output and upper limit of renewable energy units. , , / These are the minimum output ratio, ramp rate, and shortest continuous start / stop time for thermal power units, respectively. This represents the minimum output ratio of the nuclear power unit. / These represent the downhill and uphill ramp rates of thermal power units, / These represent the downhill and uphill ramp rates of the nuclear power unit, respectively.

[0060] The constraints related to the maintenance plan are shown in equations (27)-(32), namely, maintenance start time constraint, maintenance status constraint, sequential maintenance constraint, mutually exclusive maintenance constraint, simultaneous maintenance constraint and maintenance resource constraint.

[0061] (27).

[0062] (28).

[0063] (29).

[0064] (30).

[0065] (31).

[0066] (32)

[0067] In the formula, For the assembly of units awaiting maintenance. j The corresponding unit subscripts include thermal power units and nuclear power units; This variable represents the start-up state of the unit under maintenance, ranging from 0 to 1 (1 indicates that maintenance has begun). , , These represent the earliest and latest start dates of the unit's intended maintenance period, as well as the duration of the maintenance. For overhaul of the unit j 1. Needs to compare with the unit j The time interval in advance; S t This is the maximum number of units that can be inspected and maintained simultaneously.

[0068] Step 106 specifically includes: The annual 8760-hour operation simulation was decoupled, and power balance simulations were performed on a long-term scale, on a daily basis, covering 365 calendar days of the year. On a short-term scale, combined scenarios were simulated for intraday power balance, on an hourly basis. Figure 3 As shown.

[0069] Restructure system operating costs; system operating costs Decoupling is considered in terms of operating costs for short-term power balance. Operating costs for long-term power balance The mean values ​​are shown in equations (33)-(35).

[0070] (33)

[0071] (34).

[0072] (35)

[0073] In the formula, w This is a weighting factor for long-term operating costs, which can be adjusted to suit simulation needs under different conditions. D , S These are the sets of natural days within the year and the sets of combined scenarios, respectively, with subscripts... d , s For the corresponding set element. T s This represents the set of times within a day for the combined scenario. t This is the index for the corresponding time. , , These are the daily utilization hours of thermal power units, the amount of electricity curtailed by renewable energy units, and the system load shedding, respectively. , , , , , These are the output of thermal power units, the 0-1 variables of start-up and shutdown status (1 represents start-up status), the 0-1 variables of shutdown action (1 represents shutdown at time t), the 0-1 variables of start-up action (1 represents start-up at time t), the power curtailment of renewable energy units, and the power shedding of loads under the combined scenario.

[0074] Among them, the system operation simulation constraints for the entire year of 8760h are decoupled into long-term scale operation constraints and short-term scale operation constraints. Long-term scale operation constraints include: system power balance constraints, as shown in Equation (36), annual and daily utilization hours constraints of thermal power units, as shown in Equation (37), utilization hours constraints of nuclear power units, as shown in Equation (38), operation constraints of renewable energy units, as shown in Equation (39), and energy storage operation constraints, as shown in Equations (40)-(42). Based on the energy storage generalization model, power transfer is carried out on a daily basis.

[0075] (36)

[0076] (37)

[0077] (38).

[0078] (39)

[0079] (40)

[0080] (41).

[0081] (42)

[0082] In the formula, , These represent the installed capacity and daily utilization hours of the nuclear power unit, respectively. This refers to the number of hours within a day. , , , These represent the upper limit of daily utilization hours for thermal power units and nuclear power units, and the 0-1 variable representing maintenance status (0 represents maintenance status). , These refer to the power generation capacity of renewable energy units and their upper limits. , These represent the daily load and daily load shedding of the system, respectively. The meanings of the variables and parameters related to the operation constraints of the energy storage unit are referenced in equations (17)-(19).

[0083] The operational constraints within the short-timescale combined scenario are similar to the 8760h time-series operational constraints, where the system power balance constraint is shown in equation (43): (43)

[0084] In the formula, , , / , These represent the combined scenarios of nuclear power unit output, renewable energy unit output, and energy storage unit charging / discharging power and load demand.

[0085] Other constraints include upper and lower limits of thermal power output, ramping constraints, start-up and shutdown time constraints, nuclear power operation constraints, renewable energy operation constraints, and energy storage operation constraints. The specific construction references (17)-(26).

[0086] Step 107 specifically includes: The combined scenarios maintain coupling in both short- and long-term operational simulations, fully ensuring supply and demand balance across multiple time scales and effectively simulating the energy transfer characteristics of the energy storage system at different time scales. Taking long-term energy storage as an example, the sum of its daily net charge and discharge equals the medium- to long-term natural daily charge and discharge, and the coupling relationship is as follows: Figure 5As shown. The operational coupling constraints are shown in equations (44)-(47). Equations (44)-(45) respectively restrict the output power of thermal power and renewable energy in the combined scenario to be consistent with the corresponding natural day. Equation (46) restricts the SOC of LDES at the initial moment of the combined scenario to be equal to the SOC of the previous natural day of the combined scenario. Equation (47) restricts the SOC of LDES at the end of each day of the combined scenario to be equal to the SOC of the corresponding natural day.

[0087] (44)

[0088] (45)

[0089] (46)

[0090] (47)

[0091] In the formula, For combined scenes s The corresponding natural day. , Each is a combined scene s Initial and final time of energy storage units es The energy storage capacity. The coupling constraints of nuclear power units are similar to those of thermal power units, and will not be elaborated further.

[0092] In addition, in some embodiments, it also includes: The improved Garver-6 node system was tested, and the solution accuracy and efficiency of the multi-timescale hierarchical decoupled time-series operation simulation method considering extreme scenarios were analyzed. The results show that the solution time of complex power system operation simulation is significantly reduced while ensuring the accuracy of the model solution.

[0093] That is, by considering typical scenarios and extreme weather conditions, the improved Garver-6 node system is used for simulation, which effectively realizes the simulation of high-proportion renewable energy power system operation considering extreme scenarios.

[0094] This application also provides specific embodiments for further explanation.

[0095] 1) Example setup: The method proposed in this embodiment is tested using an improved Garver-6 node system to simulate the proposed multi-timescale hierarchical decoupled time-series operation considering extreme scenarios. The improved Garver-6 node system includes three thermal power plants, two nuclear power plants, one 200MW hydropower station, one 300MW wind farm, and one 250MW photovoltaic power station, with a 200MW energy storage system and a continuous discharge time of 10 hours. The peak load of the system in a horizontal year is 700MW. Table 1 shows the remaining technical parameters.

[0096] Table 1. Relevant technical parameters for system operation

[0097] To comprehensively evaluate the effectiveness of the proposed multi-timescale hierarchical decoupled time-series simulation method that considers extreme scenarios, five different types of simulation cases were configured, as shown in Table 2. Power system operation simulation calculations were then performed according to the different simulation case configuration requirements.

[0098] Table 2 Example Settings

[0099] System operation simulations were conducted using five methods, from Examples 1 to Examples 5. For the method proposed in this application in Example 5, firstly, based on the K-medoids clustering method, seven typical scenarios were selected, and a scenario difference threshold ε=c was set. We set c = 0.72. Based on the principle of maximizing scene difference, the 75 outliers were further divided into 8 extreme scenarios and assigned different weights. Compared with the traditional method that only considers the farthest outliers, this method can cover a wider range of scene types and avoid scene redundancy. Finally, a combined scene set was formed for simulation and solution. The selected extreme scenario weights and their proportion throughout the year are shown in Table 3.

[0100] Table 3. Weights and proportions of selected extreme scenarios

[0101] To demonstrate the efficiency difference between the proposed method and that considering only typical scenarios, the number of typical scenarios was increased in Example 2, with 30 typical scenarios (k=30) selected for comparison. In Example 3, seven cluster centers were selected as typical scenarios, and eight points furthest from the cluster were selected as extreme scenarios. The optimized results of each example are shown in Table 4. Based on the simulation results of Example 1 considering 8760 hours of time-series simulation, the differences in solution efficiency and accuracy of the remaining examples are compared, as shown in Table 5.

[0102] Table 4 Results of different examples

[0103] Table 5. Computational efficiency and error for different examples

[0104] Table 5 shows the computational efficiency and solution accuracy of different examples. Compared with example 1, example 5 reduces the computation time by 98%, greatly improving the solution efficiency. Compared with examples 2, 3, and 4, the total cost deviation of the proposed method is only 1.06%, and the solution accuracy is the highest among all the comparative examples.

[0105] Figure 7 Comparing the unit maintenance situations of Example 1 and Example 5, it can be seen that the unit maintenance situations of the two are quite similar, concentrated in the autumn and winter seasons when the net load is small. This corresponds to the fluctuation characteristics of the system net load and shows obvious seasonality. However, the other examples cannot simulate the power balance characteristics on a long time scale and cannot take into account the impact of system unit maintenance.

[0106] To visually demonstrate the system's long-term power balance operation simulation... Figure 8 The simulation results of the annual daily power balance operation of Examples 1 and 5 were compared. The renewable energy output in each example exhibited significant seasonal fluctuations. Power generation was highest in autumn and winter (October to February of the following year), followed by spring (March to May), and lowest in summer (June to September). Load shedding and power curtailment occurred during the summer peak net load and the winter trough net load periods, respectively. This demonstrates that the daily power balance model in Example 5 can accurately and quickly simulate the long-term operation of the system, while the simulations in Examples 2, 3, and 4, based on typical days, cannot analyze the long-term power imbalance problem of the system.

[0107] Figure 9 The simulation results of typical intraday operation under various examples were compared. Figure (a) shows the power output of thermal power units. It can be seen that the power output curves of Example 5 and Example 1 are the closest, especially in the 3rd and 4th typical days when their power output is completely consistent. In contrast, Example 2 and Example 4 have larger errors. Figure (b) compares the SOC variation curves of energy storage units during typical days. It can be found that the variation patterns of Example 5 and Example 1 are relatively consistent, while Example 2 and 4 differ significantly from Example 1 and have smaller overall fluctuations. Therefore, the proposed method can more effectively simulate the hourly operation of the system during typical intraday days.

[0108] The results analysis shows that the proposed multi-timescale hierarchical decoupled time-series simulation method that considers extreme scenarios can effectively simulate the system's annual operating characteristics while improving solution efficiency, and can effectively ensure power supply reliability in the face of extreme weather.

[0109] Based on the same inventive concept, this application also provides a multi-timescale hierarchical decoupling time-series simulation apparatus for implementing the above-described multi-timescale hierarchical decoupling time-series simulation method considering extreme scenarios. The solution provided by this apparatus is similar to the implementation scheme described in the above-described method. Therefore, the specific limitations of one or more embodiments of the multi-timescale hierarchical decoupling time-series simulation apparatus considering extreme scenarios provided below can be found in the above-described limitations of the multi-timescale hierarchical decoupling time-series simulation method considering extreme scenarios, and will not be repeated here.

[0110] In one exemplary embodiment, a multi-timescale hierarchical decoupled timing simulation apparatus considering extreme scenarios is provided, comprising: The data acquisition module is used to acquire annual wind power, photovoltaic, and hydropower output and load data of the power system to form the original dataset. The data processing module is used to preprocess and reduce the dimensionality of the original dataset to obtain the original dataset after dimensionality reduction and compression. The scene selection module is used to select typical scenes based on the original dataset after dimensionality reduction and compression, and to iteratively select extreme scenes by maximizing scene difference. The combination module is used to construct a set of combined scenarios based on selected typical and extreme scenarios; The model building module is used to establish a power system time-series operation simulation model that considers extreme scenarios and maintenance plans based on the combined scenario set. The power system time-series operation simulation model takes the minimization of power system operating costs as its objective function. The system operation constraints of the power system time-series operation simulation model include power balance constraints, thermal power unit operation constraints, nuclear power unit operation constraints, renewable energy unit operation constraints, multi-timescale energy storage operation constraints, and maintenance plan constraints. The simulation module is used to perform multi-timescale layered decoupled operation simulation based on the power system time-series operation simulation model. The multi-timescale layered decoupled operation simulation decouples the annual operation time of the power system into long-term power balance simulation and short-term power balance simulation. The long-term power balance simulation performs an annual simulation on a daily basis to optimize the daily utilization hours of generating units and the daily energy transfer of energy storage. The short-term power balance simulation performs intraday simulation on the combined scenario set on an hourly basis to optimize the unit output and energy storage power. The output module is used to determine the consistency between long-term and short-term power balance simulations based on operational coupling constraints, and then obtain the capacity allocation results of the power system.

[0111] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 10 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores multi-timescale hierarchical decoupling timing simulation data considering extreme scenarios. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a multi-timescale hierarchical decoupling timing simulation method considering extreme scenarios.

[0112] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0113] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0114] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0115] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0116] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A multi-time scale hierarchical decoupled time series operational simulation method considering extreme scenarios, characterized in that, The method comprises the following steps: obtaining annual wind power, photovoltaic power, hydropower output and load data of a power system in a year to form an original data set; preprocessing and dimensionality reduction are performed on the original data set to obtain a dimensionally reduced and compressed original data set; based on the dimensionally reduced and compressed original data set, a clustering method is used to select typical scenarios, and extreme scenarios are iteratively selected by maximizing the difference between scenarios; a combined scenario set is constructed based on the selected typical scenarios and extreme scenarios; a power system time sequence operation simulation model considering extreme scenarios and maintenance plans is established based on the combined scenario set; the power system time sequence operation simulation model takes minimization of power system operation cost as an objective function; system operation constraints of the power system time sequence operation simulation model include power balance constraints, thermal power unit operation constraints, nuclear power unit operation constraints, renewable energy unit operation constraints, multi-time scale energy storage operation constraints and maintenance plan constraints; based on the power system time sequence operation simulation model, multi-time scale hierarchical decoupling operation simulation is performed; the multi-time scale hierarchical decoupling operation simulation decouples the annual operation time of the power system into long-time scale electric quantity balance simulation and short-time scale power balance simulation; the long-time scale electric quantity balance simulation is performed in days to optimize daily utilization hours of units and daily energy transfer of energy storage; the short-time scale power balance simulation is performed in hours on the combined scenario set to optimize unit output and energy storage power; after the consistency of the long-time scale electric quantity balance simulation and the short-time scale power balance simulation is determined based on operation coupling constraints, capacity ratio results of the power system are obtained.

2. The multi-time scale hierarchical decoupled chronological operational simulation method considering extreme scenarios according to claim 1, characterized in that, The preprocessing and dimensionality reduction of the original data set specifically include: According to the formula The original data set is standardized to obtain a standardized data set; wherein σ f The feature F f The variance of the feature X sc The standardized data set; N sl The feature variable dimension number of each scene; According to the formula , the covariance matrix of the normalized data set is calculated; performing eigen-decomposition on the covariance matrix to obtain an eigen-decomposed matrix ; wherein the eigenvalues are λ 1, λ 1,..., , and the eigenvectors are μ λ1 , μ λ1 ,..., ; Pre-selection K The original data is represented by a feature variable, and the reserved data variance ratio is determined according to the formula . determining a feature matrix for characterizing the raw data eventually obtaining a reduced dimension compressed dataset .

3. The multi-time scale hierarchical decoupled chronological operational simulation method considering extreme scenarios according to claim 1, characterized in that, The formula expression of the combined scenario set is: ； wherein, is the number of extreme scenarios, is the number of typical scenarios; N n , N T are the number of nodes and the length of the scenario, respectively; d , w , pv, hy are the load demand, wind power output, photovoltaic power output and hydropower output, respectively.

4. The multi-time scale hierarchical decoupled chronological operational simulation method considering extreme scenarios according to claim 1, wherein, The formula expression of the objective function is: ； wherein, T is the set of time instants within a year, t is the corresponding time instant index; G , RE are the sets of thermal power units and renewable energy units, respectively, g , re is the corresponding unit index; , , , represent the fuel cost, the on-operation cost, the shutdown cost and the startup cost of thermal power units, respectively, , represent the curtailment cost and the load shedding cost of renewable energy units, respectively, , are the curtailment power of renewable energy units and the load shedding power of the system, respectively, , , , are the output, the on-off state variable, the shutdown action variable and the startup action variable of thermal power units, respectively.

5. The multi-time scale hierarchical decoupled chronological operational simulation method considering extreme scenarios according to claim 1, wherein, The multi-time scale energy storage operation constraints are composed of energy storage energy transfer constraints, charge / discharge power upper and lower limit constraints, charge / discharge state constraints, capacity upper and lower limit constraints and start / end energy balance constraints; The formula expressions of the energy storage energy transfer constraints, the charge / discharge power upper and lower limit constraints, the charge / discharge state constraints, the capacity upper and lower limit constraints and the start / end energy balance constraints are as follows: ； ； ； wherein, , , respectively represent the energy storage capacity, the charging power and the discharging power of the energy storage unit, , represent the charging power and the discharging power state 0 variable of the energy storage unit, , , , respectively represent the upper limit of the energy storage capacity, the upper limit of the charge / discharge power capacity, the lower limit of the charge / discharge power capacity and the minimum energy storage ratio of the energy storage unit, , respectively represent the initial energy storage capacity and the final energy storage capacity, , , respectively represent the self-discharge rate, the charging power and the discharging efficiency of the energy storage unit, is the time granularity of the operation of the energy storage unit.​​ 6. The multi-time scale hierarchical decoupled chronological operational simulation method considering extreme scenarios according to claim 1, wherein, The formula expressions of the power balance constraints, the thermal power unit operation constraints, the nuclear power unit operation constraints and the renewable energy unit operation constraints are as follows: ； ； ； ； ； ； ； In the formula, N , ES are nuclear and energy storage unit sets, respectively, n , es is the corresponding unit subscript; , are nuclear unit output and system load, , are renewable energy unit output and output upper limit; , , / are the minimum output ratio, ramp rate and minimum continuous start / stop time of thermal units, respectively; the minimum output ratio of nuclear units, / are the down and up ramp rates of thermal units, respectively, / are the down and up ramp rates of nuclear units, respectively.

7. The multi-time scale hierarchical decoupled chronological operational simulation method considering extreme scenarios according to claim 1, wherein, The maintenance plan constraints include maintenance start time constraints, maintenance state constraints, sequential maintenance constraints, mutually exclusive maintenance constraints, simultaneous maintenance constraints and maintenance resource constraints.

8. A multi-time scale hierarchical decoupled chronological operation simulation apparatus considering extreme scenarios, characterized in that, The method comprises the following steps: a data acquisition module is configured to obtain annual wind power, photovoltaic power, hydropower output and load data of a power system in a year to form an original data set; a data processing module is configured to preprocess and perform dimensionality reduction on the original data set to obtain a dimensionally reduced and compressed original data set; a scenario selection module is configured to select typical scenarios based on the dimensionally reduced and compressed original data set by using a clustering method, and iteratively select extreme scenarios by maximizing the difference between scenarios; a combination module is configured to construct a combined scenario set based on the selected typical scenarios and extreme scenarios; and a simulation module is configured to perform multi-time scale hierarchical decoupling operation simulation based on the combined scenario set. The model construction module is configured to establish a power system time sequence operation simulation model considering extreme scenarios and maintenance plans based on the combined scenario set; the power system time sequence operation simulation model takes power system operation cost minimization as an objective function; system operation constraints of the power system time sequence operation simulation model include power balance constraints, thermal power unit operation constraints, nuclear power unit operation constraints, renewable energy unit operation constraints, multi-time scale energy storage operation constraints, and maintenance plan constraints; The simulation module is configured to perform multi-time scale hierarchical decoupling operation simulation based on the power system time sequence operation simulation model; the multi-time scale hierarchical decoupling operation simulation decouples the annual operation time of the power system into long-time scale electric quantity balance simulation and short-time scale power balance simulation; the long-time scale electric quantity balance simulation performs annual simulation in days, and optimizes unit daily utilization hours and energy storage daily energy transfer; The short-time scale power balance simulation performs intra-day simulation on the combined scenario set in hours, and optimizes unit output and energy storage power; The output module is configured to obtain capacity allocation results of the power system after determining consistency of the long-time scale electric quantity balance simulation and the short-time scale power balance simulation based on operation coupling constraints.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method of claim 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method of claim 1-7.

Images