Flexible resource planning methods, devices, and computer equipment for new energy bases

Abstract

This application relates to a method, apparatus, and computer equipment for flexible resource planning in a new energy base. The method includes: constructing an objective function based on a full life-cycle technical resource model of the new energy base, with the goal of minimizing technical resources; constructing multi-dimensional scenario energy information for the new energy base under multiple time-series operating scenarios based on historical energy usage data; constructing constraints on the objective function based on the multi-dimensional scenario energy information; solving the objective function under these constraints to obtain initial resource planning strategies for various resources in the new energy base; and using the initial resource planning strategies to perform time-series operating simulations of the new energy base under each time-series operating scenario, iteratively and incrementally configuring the flexible resources in the new energy base to obtain a flexible resource planning strategy for the new energy base. This method can improve the accuracy of flexible resource planning and the overall resource utilization efficiency.

Description

Technical Field

[0001] This application relates to the field of power system planning technology, and in particular to a method, apparatus, computer equipment, storage medium and computer program product for flexible resource planning of a new energy base. Background Technology

[0002] In traditional power systems, flexibility resources are mainly provided by conventional power sources. However, with the significant increase in the proportion of renewable energy sources, the capacity of conventional power sources has further decreased. Relying solely on conventional power sources to provide flexibility is neither economical nor technically feasible. The contradiction between the gradually decreasing flexibility resources and the increasing demand for flexibility has become prominent. Therefore, it is necessary to fully explore the potential of flexibility resources in each link and plan them in a coordinated manner.

[0003] The core objective of flexible resource planning is to ensure that the system can maintain power balance economically and safely at any time and meet certain flexibility margin requirements by rationally allocating resources with rapid adjustment capabilities in the context of significantly increased uncertainty and volatility.

[0004] Currently, traditional flexibility planning methods typically employ deterministic reserve capacity methods. This approach is generally suitable for low-proportion renewable energy systems with relatively low system uncertainty, as deterministic reserve capacity methods can encompass system uncertainty. However, as the proportion of renewable energy increases, traditional flexibility planning methods cannot encompass system uncertainty, easily leading to insufficient power supply reliability under extreme operating conditions and resulting in investment redundancy. Summary of the Invention

[0005] Based on this, it is necessary to provide a flexible resource planning method, apparatus, computer equipment, computer-readable storage medium, and computer program product for new energy bases that can improve the accuracy of flexible resource planning and the overall resource utilization efficiency, in order to address the above-mentioned technical problems.

[0006] Firstly, this application provides a flexible resource planning method for new energy bases. The method includes:

[0007] With the goal of minimizing technical resources, an objective function is constructed based on the full life cycle technical resource model of the new energy base. The full life cycle technical resource model includes an initial resource input model, an operation and maintenance resource consumption model, an operation resource consumption model, and a penalty resource consumption model.

[0008] Based on the historical energy usage data of the new energy base, multi-dimensional scenario energy information of the new energy base under multiple time-series operation scenarios is constructed; the multi-dimensional scenario energy information includes the energy information of node units in multiple dimensions of the new energy base;

[0009] Based on the multi-dimensional scenario energy information, the constraints of the objective function are constructed, and the objective function is solved under the constraints to obtain the initial resource planning strategy for various resources in the new energy base.

[0010] Using the initial resource planning strategy, the new energy base is simulated for time-series operation under each of the time-series operation scenarios. The flexible resources in the new energy base are iteratively and incrementally configured to obtain the flexible resource planning strategy for the new energy base.

[0011] Secondly, this application also provides a flexible resource planning device for a new energy base. The device includes:

[0012] The objective function construction module is used to construct an objective function based on the full life cycle technical resource model of the new energy base with the goal of minimizing technical resources. The full life cycle technical resource model includes an initial resource input model, an operation and maintenance resource consumption model, an operation resource consumption model, and a penalty resource consumption model.

[0013] The scenario construction module is used to construct multi-dimensional scenario energy information of the new energy base under multiple time-series operation scenarios based on the historical energy usage data of the new energy base; the multi-dimensional scenario energy information includes the energy information of node units in multiple dimensions of the new energy base;

[0014] The solution module is used to construct the constraints of the objective function based on the multi-dimensional scenario energy information, solve the objective function under the constraints, and obtain the initial resource planning strategy for various resources in the new energy base.

[0015] The iterative optimization module is used to simulate the operation of the new energy base under each of the time-series operation scenarios using the initial resource planning strategy, and iteratively incrementally configure the flexibility resources in the new energy base to obtain the flexibility resource planning strategy of the new energy base.

[0016] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the above-described method.

[0017] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the above-described method.

[0018] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the above-described method.

[0019] The aforementioned flexible resource planning methods, devices, computer equipment, storage media, and computer program products for new energy bases aim to minimize technical resources. Based on a full life-cycle technical resource model of the new energy base, an objective function is constructed, enabling the final planning strategy to consider resource usage throughout the entire life-cycle, rather than simply focusing on inputs in a specific period, effectively improving the accuracy of flexible resource planning. Based on historical energy usage data of the new energy base, multi-dimensional scenario energy information is constructed for the new energy base under multiple time-series operating scenarios. Constraints on the objective function are then constructed based on this multi-dimensional scenario energy information. Solving the objective function under these constraints yields initial resource planning strategies for various resources in the new energy base, effectively improving the feasibility of flexible resource planning under various operating scenarios. Subsequently, the initial resource planning strategies are used to simulate the time-series operation of the new energy base under various time-series operating scenarios, iteratively and incrementally configuring flexible resources in the new energy base. This allows for a gradual approximation of a more accurate flexible resource planning strategy for the new energy base, effectively improving the accuracy of the flexible resource planning strategy and overall resource utilization efficiency. Attached Figure Description

[0020] Figure 1 This is an application environment diagram of a flexible resource planning method for a new energy base in one embodiment;

[0021] Figure 2 This is a flowchart illustrating a flexible resource planning method for a new energy base in one embodiment;

[0022] Figure 3 This is a flowchart illustrating the process of constructing multi-dimensional scenario energy information of a new energy base under multiple time-series operation scenarios based on historical energy usage data of the new energy base in one embodiment.

[0023] Figure 4 This is a flowchart illustrating the process of clustering initial time-series running scenarios according to a preset clustering type, based on the energy information of each initial scenario, to obtain multiple time-series running scenarios under the clustering type, and the multi-dimensional scenario energy information corresponding to each time-series running scenario.

[0024] Figure 5 This is a flowchart illustrating the process of using an initial resource planning strategy to simulate the operation of a new energy base under various time-series operation scenarios in one embodiment, iteratively and incrementally configuring the flexibility resources in the new energy base, and obtaining the flexibility resource planning strategy of the new energy base.

[0025] Figure 6 This is a flowchart illustrating the process of increasing the resource input of the target flexibility resource according to a preset increment step in one embodiment, thereby obtaining the round resource planning strategy of each flexibility resource in the current iteration round.

[0026] Figure 7 This is a flowchart illustrating a flexible resource planning method for a new energy base in another embodiment;

[0027] Figure 8 This is a structural block diagram of a flexible resource planning device for a new energy base in one embodiment;

[0028] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0029] 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.

[0030] The flexible resource planning method for new energy bases provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, the resource planning system 102 communicates with the server 104 via a network. A data storage system can store the data that the resource planning system 102 needs to process. The data storage system can be integrated into the resource planning system 102 or placed in the cloud or on another network server. The resource planning system 102 aims to minimize technical resources and constructs an objective function based on the full lifecycle technical resource model of the new energy base. This full lifecycle technical resource model includes an initial resource input model, an operation and maintenance resource consumption model, an operational resource consumption model, and a penalty resource consumption model. Subsequently, based on the historical energy usage data of the new energy base, multi-dimensional scenario energy information of the new energy base under multiple time-series operation scenarios is constructed. This multi-dimensional scenario energy information includes energy information of node units in multiple dimensions within the new energy base. Constraints are constructed for the objective function based on the multi-dimensional scenario energy information. The objective function is solved under these constraints to obtain the initial resource planning strategy for various resources in the new energy base. Using the initial resource planning strategy, time-series operation simulations of the new energy base are performed under various time-series operation scenarios. Iterative incremental configuration of the flexibility resources in the new energy base is then performed to obtain the flexibility resource planning strategy for the new energy base.

[0031] The resource planning system 102 can be integrated into the user terminal used by staff at the new energy base, or it can be integrated into the server 104. The user terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. The IoT device can be the energy management terminal of the new energy base. The portable wearable device can be a smartwatch, smart bracelet, head-mounted device, etc. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers.

[0032] In one embodiment, such as Figure 2 As shown, a flexible resource planning method for new energy bases is provided, which can be applied to... Figure 1 Taking resource planning system 102 as an example, the following steps are included:

[0033] S202 aims to minimize technical resources by constructing an objective function based on the full life-cycle technical resource model of the new energy base.

[0034] Among them, minimizing technical resources refers to the optimization goal of minimizing the total amount of all technical resources invested throughout the entire life cycle while ensuring the safe operation of the system. It means that we want to consume as little as possible the total amount of various technical resources, such as equipment, materials, energy, spare parts, and labor hours.

[0035] A new energy base is a specific geographical area that mainly consists of wind farms and photovoltaic power stations, and is equipped with energy storage and power transmission facilities. It is a large-scale power generation complex that utilizes renewable energy sources such as wind and solar power.

[0036] The life-cycle technical resource model includes an initial resource input model, an operation and maintenance resource consumption model, a running resource consumption model, and a penalty resource consumption model. The initial resource input model focuses on resource consumption before the new energy base is built, primarily used to calculate the total amount of technical resources required during planning, procurement, transportation, installation, and commissioning phases. The running resource consumption model focuses on resource consumption during the normal operation of the new energy base, primarily used to calculate the total amount of electricity, water, and oil continuously consumed in planned daily power generation, power supply, monitoring, and routine maintenance activities. The operation and maintenance resource consumption model focuses on resource consumption for regular maintenance and repair, primarily used to calculate the resources consumed in planned maintenance activities such as preventative inspections and regular maintenance to keep equipment running normally. The penalty resource consumption model focuses on the additional resources required when requirements are not met, primarily used to calculate the additional technical resources required when the operating status of the new energy base violates hard constraints such as safety, environmental protection, and reliability.

[0037] The objective function is a mathematical expression used to quantify the goal to be optimized. The objective function is mainly composed of decision variables, and the optimization process minimizes the objective function by adjusting the decision variables.

[0038] For example, the resource planning system can construct an objective function based on the full life cycle technical resource model of the new energy base with the goal of minimizing technical resources.

[0039] In one embodiment, taking the Shagohuang energy base as an example, the Shagohuang region is rich in solar and wind energy resources. The resource types involved in the planning may include: wind power (W), photovoltaic (PV) / solar thermal (CSP), thermal power (G), and electrochemical energy storage (ES). All of the above resources are included in the multi-stage capacity decision-making process; among them, thermal power, solar thermal, and energy storage are simultaneously allocated as core system flexibility resources. In addition, the system can utilize a certain amount of demand-side response resources for regulation, and its adjustable capacity is handled in the form of a known upper limit constraint, without separately planning the capacity for investment.

[0040] For the initial resource input model Operation and maintenance resource consumption model , runtime resource consumption model and penalty resource consumption model The following models are established respectively:

[0041] Among them, the initial resource input model The model expression is as follows:

[0042]

[0043] In the formula, For resources In the The unit capacity investment cost at each stage; The annual discount rate; This represents the number of years included in each stage.

[0044] Among them, the operation and maintenance resource consumption model The model expression is as follows:

[0045]

[0046] In the formula, For resources In the Annual unit fixed operation and maintenance cost for the phase; For resources In the The cumulative installed capacity at the end of the phase; for The present value factor for term equal annuities converts annualized fixed costs into the present value of the first period.

[0047] Among them, the runtime resource consumption model The model expression is as follows:

[0048]

[0049] In the formula, The unit cost of electricity generation for thermal power plants; It generates power for thermal power plants; The unit start-up cost of thermal power plants; This refers to the starting capacity of thermal power plants; Unit operating cost of solar thermal power; Powering photothermal polymerization; Unit operating cost of energy storage; , Aggregate charging and discharging power for energy storage; Compensation cost per unit of demand-side response; Power is invoked to respond to demand; Penalty cost per load shedding unit; This is the load shedding amount.

[0050] Among them, the penalty resource consumption model The model expression is as follows:

[0051]

[0052] In the formula, and These represent the penalty resource amount per unit of shortage, while and This indicates the corresponding deficit in renewable energy electricity and the deficit in carbon emission reduction.

[0053] With the goal of minimizing technical resources, the objective function based on the full life-cycle technical resource model of the new energy base is constructed as follows:

[0054] .

[0055] S204. Based on the historical energy usage data of the new energy base, construct multi-dimensional scenario energy information of the new energy base under multiple time-series operation scenarios.

[0056] Among them, historical energy usage data are information data that the new energy base observes and records during historical periods, used to reflect the source-load characteristics of the base. For example, measured data such as wind speed, irradiance, output, load, and temperature recorded in the history of the new energy base are usually in hourly or minute intervals.

[0057] Among them, a time-series operation scenario refers to a set of data sequences arranged in chronological order, covering a typical cycle, used to simulate the operating state of a new energy system on the time axis. It can be understood that multiple time-series operation scenarios can not only reflect the typical operating patterns of the system most of the time, but also capture the special states when the system approaches the safety boundary or a failure occurs.

[0058] Multidimensional energy information includes exogenous time-series parameters for each time-series operational scenario. These exogenous time-series parameters are energy information of multiple-dimensional node units in the new energy base at each moment under the corresponding time-series operational scenario. In the planning and operation optimization model, these parameters serve as known boundaries, are unaffected by resource capacity allocation decisions, and are generated solely from historical data statistics. Essentially, they are a set of values ​​arranged chronologically, given one by one at each moment. The multiple-dimensional node units can include various new energy power plants and load nodes within the new energy base.

[0059] In one embodiment, the time-series exogenous parameter may include the wind power aggregation output coefficient. Photovoltaic polymerization output coefficient Normalized collector heat power input calculated based on direct normal irradiance (DNI) This reflects the heat collection capacity of a unit of solar thermal power unit and the total system load power. wait.

[0060] For example, the resource planning system can construct multi-dimensional scenario energy information of the new energy base under multiple time-series operation scenarios based on the historical energy usage data of the new energy base.

[0061] S206. Based on the multi-dimensional scenario energy information, construct the constraints of the objective function, solve the objective function under the constraints, and obtain the initial resource planning strategy for various resources in the new energy base.

[0062] Among them, constraints are restrictive conditions that limit the range of values ​​of decision variables. They can be a set of inequalities or equations constructed based on physical laws, equipment limits, policy requirements, data boundaries, etc., and their function is to ensure the feasibility of the planning scheme.

[0063] Solving the objective function is the mathematical process of finding the optimal decision parameters. By solving the objective function, the initial resource planning strategies for various resources in the new energy base can be obtained.

[0064] In this context, the various resources within a new energy base refer to the complete set of technical equipment required for the generation, conversion, storage, and balance regulation of electricity. Each piece of equipment corresponds to a quantifiable capacity variable. For example, resource types may include wind power, photovoltaic (PV), solar thermal (CSP), thermal power, electrochemical energy storage, and demand-side response. Wind power corresponds to the total installed capacity of wind turbine generators, photovoltaic power corresponds to the total installed capacity of photovoltaic power generation arrays, CSP corresponds to the rated power of a solar thermal power plant, thermal power corresponds to the total installed capacity of thermal power generators, electrochemical energy storage corresponds to the rated power and continuous discharge duration of the energy storage system, and demand-side response corresponds to the responsive load regulation capacity.

[0065] The initial resource planning strategy is a resource capacity allocation scheme obtained with the goal of optimizing the economy throughout the entire life cycle, before flexibility iteration enhancement. That is, the installed capacity value of various types of technical resources at each stage is obtained by directly solving the objective function under constraints. The initial resource planning strategy can meet all operational constraints, but there may be a flexibility gap, which requires subsequent flexibility iteration enhancement.

[0066] For example, multi-dimensional scenario energy information will affect the specific values ​​of each condition limit in the constraint adjustment conditions. Therefore, the resource planning system can construct the constraint conditions of the objective function based on the multi-dimensional scenario energy information, solve the objective function under the constraint conditions, and obtain the initial resource planning strategy for various resources in the new energy base.

[0067] In one embodiment, the resource planning system can use linear programming, integer programming, heuristic algorithms, etc., to search for the decision variable values ​​that minimize the objective function within the feasible region enclosed by the constraints, thereby obtaining the initial resource planning strategy for various resources in the new energy base.

[0068] S208 uses the initial resource planning strategy to simulate the operation of the new energy base under various time-series operation scenarios, iteratively and incrementally configures the flexibility resources in the new energy base, and obtains the flexibility resource planning strategy of the new energy base.

[0069] Among them, time-series operation simulation is a process of simulating system scheduling operation on a scenario-by-scenario and time-by-time basis after a given resource capacity plan is given, in order to verify the feasibility of the plan and calculate the operation indicators.

[0070] Iterative incremental configuration refers to an optimization process that gradually increases the capacity of flexible resources until the sufficiency requirements are met by repeatedly performing evaluation-selection-increase steps.

[0071] Flexible resources are power generation or energy storage equipment that can quickly adjust their output to track changes in net load, such as thermal power, solar thermal power, and electrochemical energy storage in new energy bases. These resources can adjust their output upward or downward in a short period of time to compensate for power imbalances caused by random fluctuations in renewable energy and load.

[0072] The flexible resource planning strategy is the final resource capacity scheme that simultaneously satisfies the constraints of optimal economy and sufficient flexibility after iterative incremental allocation.

[0073] For example, the resource planning system can use the initial resource planning strategy to perform time-series operation simulations of the new energy base under various time-series operation scenarios, iteratively and incrementally configure the flexibility resources in the new energy base, and obtain the flexibility resource planning strategy of the new energy base.

[0074] The aforementioned flexible resource planning method for new energy bases aims to minimize technical resources. Based on a full life-cycle technical resource model of the new energy base, an objective function is constructed. This ensures the final planning strategy considers resource usage throughout the entire life-cycle, rather than focusing solely on inputs at a specific time, effectively improving the accuracy of flexible resource planning. Based on historical energy usage data of the new energy base, multi-dimensional scenario energy information is constructed for various time-series operating scenarios. Constraints on the objective function are then established based on this multi-dimensional scenario energy information. Solving the objective function under these constraints yields initial resource planning strategies for various resources within the new energy base, effectively improving the feasibility of flexible resource planning across different operating scenarios. Subsequently, the initial resource planning strategies are used to simulate the operation of the new energy base under various time-series operating scenarios. Iterative incremental configuration of flexible resources within the new energy base allows for a gradual approximation of a more accurate flexible resource planning strategy, effectively improving the accuracy of the flexible resource planning strategy and overall resource utilization efficiency.

[0075] In one embodiment, such as Figure 3 As shown in S204, based on the historical energy usage data of the new energy base, multi-dimensional scenario energy information of the new energy base under multiple time-series operation scenarios is constructed, including:

[0076] S302, based on the historical energy usage data of the new energy base, fits the marginal probability distribution of the time-series power of each node unit in multiple dimensions of the new energy base.

[0077] Fitting a marginal probability distribution refers to estimating a parametric or non-parametric curve of a certain probability distribution form based on sample data. In other words, it is the process of using mathematical functions to approximate the distribution pattern of a set of data. Based on historical data, a probability distribution type can be selected and the parameters of the distribution can be estimated so that the statistical characteristics of the distribution are as consistent as possible with those of the actual data.

[0078] Among them, multi-dimensional node units refer to basic electrical points in the energy base that have independent power injection or outflow characteristics, such as a single new energy power station, a single load point, or a lumped equivalent source / load point.

[0079] Time-series power is an ordered numerical sequence of power changes over time, reflecting the dynamic process of power generation or consumption behavior on the time axis. It refers to the sequence of power values ​​recorded at fixed time intervals for each new energy power station or load node in a new energy base, arranged in chronological order. Among them, the power of the power station is called "output", and the power of the load is called "demand". Therefore, time-series power can also be understood as the time-series output / demand of each new energy power station and load.

[0080] Marginal probability distributions, in the context of multidimensional random variables, are probability distributions obtained by fixing one variable and calculating the probability of that variable taking each value. Marginal probability distributions describe the probability of that variable taking each value individually, without considering the values ​​of other variables.

[0081] For example, the resource planning system can fit the marginal probability distribution of the time-series power of each node unit in a new energy base across multiple dimensions based on the historical energy usage data of the new energy base.

[0082] In one embodiment, the resource planning system can fit the marginal probability distribution of time-series output / demand of each new energy power station and load using nonparametric kernel density estimation based on the historical energy usage data of the new energy base. Taking the time-series output / demand of each new energy power station and load, including the output coefficient of wind power stations, the output coefficient of photovoltaic power stations, the solar thermal collector irradiance, and the system load, as an example, the expression is as follows:

[0083]

[0084]

[0085]

[0086] Where n is the number of samples, h is the window width, K(·) is the kernel function, and u is the variable of the Gaussian kernel function.

[0087] S304, based on the probability distribution of each edge, uses an advanced probability generation model to construct the joint distribution of new energy bases.

[0088] Among them, advanced probabilistic generation models are mathematical models that can generate artificial samples that conform to joint probability distributions. They can learn the complex correlation structure between multiple random variables and randomly generate new multivariate sample sequences that are consistent with actual statistical characteristics.

[0089] In one embodiment, the advanced probabilistic generative model may include generative adversarial networks, variational autoencoders, etc.

[0090] Among them, the joint distribution is a complete probabilistic model used to describe the spatiotemporal dependence characteristics between various source and load variables in a new energy base. It can describe the values ​​of all random variables at the same time and is a probability distribution of multidimensional random variables. The joint distribution gives the probability that all variables fall within their respective ranges at the same time, and includes information on marginal distributions and all correlations between variables.

[0091] For example, the resource planning system can construct the joint distribution of new energy bases based on the probability distribution of each edge and using an advanced probability generation model.

[0092] In one embodiment, to accurately characterize the spatiotemporal interdependence between and within resources such as wind, solar, and load, the resource planning system can employ an advanced probabilistic generative model based on copula theory to construct a joint distribution describing the complex correlation structure among these variables, as shown in the following expression:

[0093]

[0094] in, for The marginal distribution function. C(·) is the Copula function, and m is the dimension of the random variable.

[0095] S306, repeatedly sample the joint distribution to obtain multi-dimensional scene energy information under multiple time-series operation scenarios.

[0096] Repeated sampling refers to the operation of independently drawing samples multiple times from an established joint distribution. Through repeated sampling, a large number of independent multivariate random sample sets can be generated according to the probability law of the joint distribution. Each sample contains the values ​​of all source load variables at a certain point in time.

[0097] For example, the resource planning system can repeatedly sample the joint distribution to obtain multi-dimensional scenario energy information under multiple time-series operating scenarios.

[0098] In one embodiment, the resource planning system can generate a distribution containing [data missing] by repeatedly sampling the aforementioned joint distribution. A collection of time-series execution scenarios Each time-series operational scenario s contains the following set of time-series exogenous parameters (t=1,…,T), namely: wind power aggregate output coefficient. Photovoltaic polymerization output coefficient Normalized collector heat power input calculated based on direct normal irradiance (DNI) This reflects the heat collection capacity of a unit of solar thermal power unit; the total load power of the system. .

[0099] In the above embodiments, by fitting marginal distributions based on historical data and constructing joint distributions using advanced probabilistic generation models, the nonlinear and asymmetric dependency structures between different variables and their respective time series can be accurately captured. The scenarios generated by repeated sampling retain the true spatiotemporal correlations. Compared with methods that independently sample or ignore correlations, the obtained multidimensional scenario energy information is more in line with reality, thereby improving the accuracy of subsequent flexible resource planning and operation optimization, and reducing decision-making biases and resource waste caused by source-load uncertainty.

[0100] In one embodiment, S306, repeatedly sampling the joint distribution to obtain multi-dimensional scene energy information under multiple time-series operating scenarios may include: repeatedly sampling the joint distribution to obtain initial scene energy information under each initial time-series operating scenario; clustering each initial time-series operating scenario according to a preset clustering type based on the initial scene energy information to obtain multiple time-series operating scenarios under the clustering type, and the multi-dimensional scene energy information corresponding to each time-series operating scenario.

[0101] The initial time series running scenarios are unreduced candidate scenarios obtained by repeated sampling. After repeated sampling of the joint distribution, the resulting initial time series running scenarios are enormous, and directly using them for optimization would lead to the "curse of dimensionality." Therefore, it is necessary to extract representative scenarios for subsequent optimization solutions.

[0102] Among them, clustering type refers to the clustering target selected by the user to distinguish scenarios according to the analysis needs, which determines the input features and grouping basis of the clustering algorithm.

[0103] For example, the resource planning system can first perform repeated sampling on the joint distribution to obtain the initial scenario energy information under each initial time-series operating scenario. Then, according to a preset clustering type, the initial time-series operating scenarios are clustered based on the initial scenario energy information to obtain multiple time-series operating scenarios under the clustering type, as well as the multi-dimensional scenario energy information corresponding to each time-series operating scenario.

[0104] In the above embodiments, a large number of initial scenarios are generated by repeated sampling, which fully preserves the spatiotemporal correlation characteristics of the source load. Then, typical scenarios are selected by clustering. While significantly reducing the number of scenarios, the key power characteristics and their occurrence probabilities under different clustering types are preserved. This makes the subsequent planning model avoid the interference of sampling random noise and greatly reduce the computational burden, achieving a balance between accuracy and efficiency.

[0105] In one embodiment, the clustering type includes a typical scenario clustering type, and the multidimensional scenario energy information includes typical scenario energy information. According to a preset clustering type, each initial time-series running scenario is clustered based on its initial scenario energy information to obtain multiple time-series running scenarios under the clustering type, and the corresponding multidimensional scenario energy information for each time-series running scenario. This may include: extracting the time-series curve features of each initial time-series running scenario from its initial scenario energy information; and clustering each initial time-series running scenario based on the feature similarity between the time-series curve features to obtain each typical scenario, and the corresponding typical scenario energy information for each typical scenario.

[0106] Among them, the typical scenario clustering type is a clustering branch aimed at extracting representative scenarios under the normal operating conditions of the system. It is mainly used to generate typical scenarios, each of which represents a normal wind-solar-load combination pattern. The typical scenario energy information consists of time-series exogenous parameters recorded under typical scenarios. It can be understood as a time-series data matrix of each typical scenario generated by the typical scenario clustering type. Each value in the matrix is ​​the center value of the corresponding dimension of all initial time-series operating scenarios within the cluster, usually the mean.

[0107] Time series curve features are numerical indicators used to describe the shape or variation pattern of a time series curve. They are a set of values ​​calculated from the time series power sequence to characterize the curve's shape. Resource planning systems can transform the complete time series curve of each initial time series running scenario into a vector of fixed scenarios using feature engineering methods. This vector, or time series curve features, is used to calculate the similarity between scenarios.

[0108] Feature similarity is a quantitative indicator used to evaluate the degree of similarity between two feature vectors of curves. The larger the value, the more similar the two scenes are.

[0109] Typical scenarios refer to representative temporal scenarios that represent the commonalities of a certain initial scenario after clustering.

[0110] For example, the resource planning system can extract the time-series curve features of each initial time-series operating scenario from the energy information of each initial scenario. Then, for each time-series curve feature, the feature similarity between the time-series curve feature and other time-series curve features is determined. Based on the feature similarity, each initial time-series operating scenario is clustered to obtain each typical scenario and the corresponding typical scenario energy information.

[0111] In one embodiment, the resource planning system can cluster a small number of typical scenarios using K-means clustering sampling results, and use the full-time feature vector of each scenario (derived from...) , , , (assembled) to generate K t A typical daily scenario, with K clusters. t Determined by the elbow method:

[0112]

[0113]

[0114] In the formula, This represents the k-th typical scenario; This indicates the probability of this scenario occurring; To divide to the number The number of samples in each cluster.

[0115] In the above embodiments, by extracting time-series curve features and clustering based on similarity, massive initial scenarios are compressed into a small number of typical scenarios that retain shape features, effectively removing random noise and redundant information. Each typical scenario carries an accurate probability of occurrence, which can represent the overall shape of the original distribution with a concise set, thereby significantly reducing computational complexity in subsequent planning while maintaining the accuracy of key fluctuation information of the source load curve.

[0116] In one embodiment, the clustering type includes an extreme scenario clustering type, and the multidimensional scenario energy information includes extreme scenario energy information. For example... Figure 4 As shown, according to a preset clustering type, each initial time-series running scenario is clustered based on the energy information of each initial scenario, resulting in multiple time-series running scenarios under the clustering type, and multi-dimensional scenario energy information corresponding to each time-series running scenario, which may include:

[0117] S402 uses predefined vulnerability indicators to select initial risk scenarios from each initial time-series operation scenario based on information from each initial energy scenario.

[0118] Among them, the extreme scenario clustering type is a clustering branch aimed at extracting representative scenarios of the system under risky conditions. It can use extreme characteristic indicators reflecting the vulnerability of the system as the basis for clustering, and group risky scenarios with similar extreme behavior patterns into one category. Extreme scenario energy information consists of time-series exogenous parameters recorded under extreme scenarios. It can be understood as a time-series data matrix of each extreme scenario generated by the extreme scenario clustering type. Each value in the matrix is ​​the center value of the corresponding dimension of all initial time-series running scenarios within the cluster, usually the mean.

[0119] Among them, the predefined vulnerability index is a scoring function used to quantify the risk level of the initial scenario. The index value can measure the degree of inadequacy of the system's ability to withstand extreme events in the corresponding scenario. It can usually include power deficit index and power curtailment index. The higher the value of the vulnerability index, the higher the vulnerability of the system in that scenario and the greater the risk.

[0120] The initial risk scenarios are those that meet the initial time-series operation conditions and are retained after being screened by vulnerability indicators.

[0121] For example, the resource planning system can use predefined vulnerability indicators to determine the vulnerability indicator values ​​corresponding to each initial time-series operation scenario based on the information of each initial energy scenario. The vulnerability indicator values ​​are compared with preset indicator thresholds, and the initial time-series operation scenarios corresponding to vulnerability indicator values ​​higher than the preset indicator thresholds are identified as initial risk scenarios.

[0122] S404: Extract the extreme characteristic features of each initial risk scenario from the initial scenario energy information of each initial risk scenario.

[0123] Among them, extreme characteristic features are feature values ​​extracted from the initial risk scenario that can characterize the extreme nature of the initial risk scenario.

[0124] For example, the resource planning system can extract the extreme characteristics of each initial risk scenario from the initial scenario energy information of each initial risk scenario.

[0125] S406, based on the feature similarity between the extreme characteristics, cluster each initial risk scenario to obtain each extreme scenario and the corresponding extreme scenario energy information.

[0126] Among them, extreme scenarios refer to representative time-series scenarios obtained after clustering that represent the commonalities of a certain type of high-risk initial risk scenarios.

[0127] For example, the resource planning system can cluster each initial risk scenario based on the feature similarity between each extreme characteristic feature to obtain each extreme scenario and its corresponding extreme scenario energy information.

[0128] In one embodiment, to ensure the safety and robustness of the planning scheme under high-risk conditions, a comprehensive vulnerability index is defined. Each scene in the program is scored:

[0129]

[0130] Among them, the two sub-indicators characterize different types of system vulnerability:

[0131] (a) Power deficit index (characterizing the maximum power supply gap in the system within the scenario):

[0132]

[0133] In the formula, This is the maximum power deficit indicator; , For reference installed capacity; This is the reference value for the maximum output of the system's adjustable power supply; The higher this indicator, the more severe the power shortage in that scenario.

[0134] (b) Energy curtailment index (depicting the difficulty of integrating new energy sources):

[0135]

[0136] In the formula, This is a reference value for the minimum technical output of the system. The value in parentheses is the difference between the output of new energy sources and (load minus minimum output of thermal power). A positive value indicates that there are difficulties in absorption and that wind and solar power may be curtailed at that moment.

[0137] from Screening Above the threshold In scenarios where extreme candidate sets are formed, clustering is performed again to generate... An extreme typical scenario:

[0138]

[0139] In the formula, To represent the j-th typical scenario; This indicates the probability of this scenario occurring; The number of samples assigned to the j-th cluster.

[0140] In the above embodiments, high-risk initial risk scenarios are accurately identified from a massive number of initial scenarios through vulnerability index pre-screening, significantly reducing the scope of analysis. Subsequently, extreme characteristic features are extracted and clustered to obtain a small number of extreme scenarios and their probability of occurrence that can represent different extreme modes, avoiding the dilution of extreme information by regular scenarios. This method enables subsequent planning models to be targeted to enhance their ability to cope with extreme scenarios, improving the robustness and power supply reliability of new energy bases under limited computing resources.

[0141] In one embodiment, after obtaining the typical scenarios and extreme scenarios, the resource planning system can merge the typical scenarios and extreme scenarios to form a complete multi-scenario set for the input objective function:

[0142]

[0143] Since both subsets originate from the same After merging, the weights of each scenario are normalized to ensure that the sum of the days represented by each scenario is exactly equal to 365 days.

[0144]

[0145] Normalized This refers to the weight of the number of representative days in the year used in the resource consumption model and all constraints. Temporal exogenous parameters contained in each scene It is passed to the objective function as an exogenous input.

[0146] In one embodiment, such as Figure 5As shown in S208, the initial resource planning strategy is used to simulate the operation of the new energy base under various time-series operation scenarios. The flexibility resources in the new energy base are iteratively and incrementally configured to obtain the flexibility resource planning strategy for the new energy base, which may include:

[0147] S502 uses the initial resource planning strategy and multi-dimensional scenario energy information to perform time-series operation simulations on the new energy base under various time-series operation scenarios, and obtains the flexibility availability and flexibility demand of multiple flexibility resources in the new energy base.

[0148] Among them, the available flexibility is the actual amount of flexible adjustment capability that can be provided in the simulation, that is, the capacity of the flexible resources to provide upward or downward adjustment power during the simulation period.

[0149] Flexibility demand is the amount of upward or downward adjustment power required from flexibility resources, caused by net load fluctuations, forecasting errors, or accidents.

[0150] For example, the resource planning system can use the initial resource planning strategy and multi-dimensional scenario energy information to perform time-series operation simulations on the new energy base under various time-series operation scenarios, and obtain the flexibility availability and flexibility demand of each of the multiple flexibility resources in the new energy base.

[0151] In one embodiment, the calculation process for available flexibility is as follows:

[0152] The current iteration configuration scheme (First iteration) Initial resource planning strategy The timing simulation was substituted in to obtain various resources at each stage. Scene ,time The system's operating status is used to calculate the upward adjustment (of the system). ) or adjust downwards ( The available flexibility offered by the direction :

[0153]

[0154] The flexibility contribution of various resources is determined by the margin in the current operating state:

[0155] Of this, the flexibility contribution of thermal power is:

[0156]

[0157] Increased flexibility is the difference between online capacity and current output; decreased flexibility is the difference between current output and minimum technical output.

[0158] The contribution of photothermal flexibility is:

[0159]

[0160] The flexibility of increasing solar thermal power output is limited by both the upper limit of rated power generation and the thermal storage margin (output cannot be increased when thermal storage is insufficient), while the flexibility of decreasing output output is equal to the space that can be reduced from the current output.

[0161] The contribution of energy storage to flexibility is:

[0162]

[0163] The energy storage up-adjustment flexibility is the difference between the rated power and the current discharge power (which can increase discharge), while the down-adjustment flexibility is the difference between the rated power and the current charging power (which can increase charging).

[0164] The contribution of demand-side response flexibility is:

[0165] .

[0166] In one embodiment, the flexibility requirement is calculated as follows:

[0167] The system in the scene ,time The flexibility requirement is determined by the fluctuations in net load between adjacent time points. Net load is defined as the remaining demand after deducting wind and solar power output from the total system load.

[0168]

[0169] In the formula, , , Obtained through timing simulation. The flexibility requirements for upward and downward adjustments are as follows:

[0170] .

[0171] S504, based on the availability and demand of each flexibility resource, constructs a flexibility adequacy index for each flexibility resource.

[0172] Among them, the flexibility adequacy index is a quantitative value that measures whether the supply of flexibility can meet the demand, and is used to determine whether the system meets the flexibility constraints.

[0173] For example, a resource planning system can construct a flexibility adequacy index for each flexibility resource based on the availability and demand of each flexibility.

[0174] In one embodiment, the resource planning system can be based on In all scenarios, the following two flexibility adequacy metrics are defined using normalized probability weights:

[0175] The probability of inflexibility (LOFP) measures the frequency of inflexibility events.

[0176]

[0177] Expectations of Lack of Flexibility (LOFE) are used to measure the expected severity of the flexibility gap:

[0178]

[0179] In the formula, This is an indicator function. These correspond to upward and downward adjustments, respectively. The two metrics, when combined, quantify the supply-demand gap in system flexibility: LOFP reflects the probability characteristics of insufficiency events, while LOFE reflects the magnitude of the gap.

[0180] The flexibility adequacy constraint requires that for all times and direction satisfy:

[0181]

[0182] In the formula, The system flexibility adequacy constraint threshold is pre-set by planning requirements.

[0183] S506, based on the preset allocation step size, determine the target flexible resource with the largest net resource gain incremental rate among all flexible resources.

[0184] The preset increment step size is a fixed incremental value for increasing the amount of resources invested in each iteration, used to increase the capacity of the target flexibility resources in each iteration.

[0185] The incremental rate of net resource gain is the rate of change of net resource gain resulting from the increase of target flexibility resources in each iteration. It can be defined as the ratio of overall net resource gain to capacity increment.

[0186] Target flexibility resources are the flexibility resources that are selected to be increased in each iteration round.

[0187] For example, the resource planning system can determine the net resource gain increment rate of each flexible resource based on a preset increment step size, and identify the flexible resource with the largest net resource gain increment rate as the target flexible resource.

[0188] S508: Increase the resource input of the target flexibility resource according to the preset increment step size to obtain the round resource planning strategy of each flexibility resource in the current iteration round.

[0189] Among them, the round resource planning strategy is the resource investment plan obtained after the current iteration round ends. That is, based on the previous round strategy, the new strategy is formed by increasing the target flexibility resource investment according to the preset allocation step size. It is the output of this round of iteration and the input of the next round of iteration.

[0190] For example, the resource planning system can increase the resource input of the target flexibility resource according to a preset incremental step size based on the previous strategy, so as to obtain the round resource planning strategy of each flexibility resource in the current iteration round.

[0191] S510: If the round-based resource planning strategy does not meet the flexibility adequacy index, the round-based resource planning strategy will be used as the initial resource planning strategy. The process will then return to the execution step of using the initial resource planning strategy and multi-dimensional scenario energy information to perform time-series operation simulation of the new energy base under various time-series operation scenarios.

[0192] Among them, the round resource planning strategy does not meet the flexibility adequacy index, indicating that the round resource planning strategy of the current iteration round has failed to meet the preset flexibility requirements and needs to be further iterated and optimized.

[0193] For example, the resource planning system can determine the flexibility adequacy constraint requirements of the system based on the round resource planning strategy and the flexibility adequacy index. If it is determined that the round resource planning strategy does not meet the flexibility adequacy index, the round resource planning strategy can be used as the initial resource planning strategy, and the system can return to the execution of step S502, that is, the step of using the initial resource planning strategy and multi-dimensional scenario energy information to perform time-series operation simulation of the new energy base under each time-series operation scenario.

[0194] S512, under the condition that the round-based resource planning strategy meets the flexibility adequacy index, the round-based resource planning strategy is determined as the flexible resource planning strategy for the new energy base.

[0195] For example, if the resource planning system determines that the round-based resource planning strategy meets the flexibility adequacy index, the round-based resource planning strategy can be determined as the flexible resource planning strategy for the new energy base.

[0196] In the above embodiments, the availability and demand of resources in each scenario are accurately quantified through time-series simulation, an sufficiency index is constructed, and the incremental step size is guided by the net gain incremental rate to form a closed-loop iterative optimization. Each round only adds the resource with the highest gain to avoid blind expansion. The process terminates when the flexibility constraint is met, so as to achieve the system sufficiency goal with the minimum resource input and take into account the economy and reliability of flexible resource planning.

[0197] Furthermore, in one embodiment, such as Figure 6As shown in S508, the resource allocation of the target flexibility resource is increased according to the preset increment step size to obtain the round resource planning strategy for each flexibility resource in the current iteration round, which may include:

[0198] S602, for each flexible resource, increase the resource input of the flexible resource according to the preset increase step size to obtain the increase input of the flexible resource.

[0199] For example, for each flexible resource, the resource planning system can increase the resource input of the flexible resource according to a preset increment step to obtain the additional input of the flexible resource.

[0200] S604, based on the increased allocation, determines the first total resource gain value of the new energy base in the current iteration round under the condition of increased flexible resource allocation.

[0201] The first total resource gain value is the total system benefit obtained through simulation calculation in the current iteration round, assuming that a certain flexibility resource is increased.

[0202] For example, the resource planning system can perform operational simulation calculations based on the increased input amount, under the condition of flexible resource allocation, to determine the first total resource gain value of the new energy base in the current iteration round.

[0203] S606. Based on the first total resource gain value and the second total resource gain value of the new energy base in the previous iteration with the increase in flexible resources, determine the net resource gain increment rate of flexible resources.

[0204] The second total resource gain value is the total system benefit obtained through simulation calculation in the previous iteration, assuming that the same flexibility resources have been increased.

[0205] For example, the resource planning system can obtain the second total resource gain value of the new energy base in the previous iteration when flexible resources are increased, and determine the net resource gain increment rate of flexible resources based on the first total resource gain value and the second total resource gain value of the new energy base in the previous iteration when flexible resources are increased.

[0206] In one embodiment, in the current solution Based on this, resource planning systems can provide flexible resources Additional unit capacity The ratio of the overall net resource gain to the capacity increase is defined as the net resource gain incremental rate:

[0207]

[0208]

[0209] In the formula, To increase resource allocation Afterwards, the increment of the total resource gain value of the system operation (including the benefits of reducing wind and solar curtailment and the benefits of reducing load shedding); Based on The results were obtained from time-series simulation calculations. This is the amount by which the probability of insufficient flexibility is reduced after the addition of additional features; The expected reduction in flexibility; , The probability of insufficient flexibility and the expected unit penalty resource consumption coefficient; This indicates that the overall net resource gain from increasing the allocation of this resource is positive; The larger the value, the higher the overall net resource gain in terms of economy and flexibility per unit of investment, and the more it should be prioritized for allocation. The power generation resource gain of all types of power sources is calculated based on the exchanged resource value for grid-connected power. Exchange resource value for carbon emissions; Carbon emissions per unit of electricity generated by thermal power plants. Carbon emissions and resource consumption from thermal power generation; This represents the comprehensive marginal resource consumption of thermal power plants. This refers to the resource consumption for starting up thermal power plants; This is a load shedding penalty.

[0210] S608, the flexibility resource corresponding to the largest incremental increase among the incremental rates of net resource gains is determined as the target flexibility resource.

[0211] For example, the resource planning system can sort the net resource gain increment rates corresponding to each flexibility resource in descending order, and determine the flexibility resource corresponding to the largest increment rate among the net resource gain increment rates as the target flexibility resource.

[0212] In the above embodiments, the marginal gain resulting from the incremental allocation step size is calculated on a resource-by-resource basis, i.e., the incremental rate of net resource gain, and resources with the largest gain are prioritized for allocation. This ensures that the resource increment in each iteration is invested in the most efficient flexibility measures. This method avoids inefficient average allocation, quickly approaches the optimal configuration within a finite number of steps, meets the system sufficiency requirements with the minimum total resource input, and achieves an efficient balance between cost and flexibility.

[0213] In one embodiment, such as Figure 7 As shown, a flexible resource planning method for new energy bases is proposed. The method is illustrated using an application scenario of a new energy base planning project in the desert region. The method specifically includes the following steps:

[0214] S701, based on the historical energy usage data of the new energy base, uses nonparametric kernel density estimation to fit the marginal probability distributions of wind farm output coefficient, photovoltaic output coefficient, solar thermal collector irradiance, and system load.

[0215] S702 employs an advanced probabilistic generation model based on Copula theory to construct a joint distribution that describes the complex correlation structure among resource variables.

[0216] S703 involves repeated sampling of the joint distribution to generate a large-scale initial scene set containing multiple scenes.

[0217] Each scenario s contains a corresponding set of temporal exogenous parameters.

[0218] S704 extracts typical and extreme scene sets from a large initial scene set.

[0219] S705 merges the typical scenario set and the extreme scenario set to construct a complete multi-scenario set.

[0220] S706, based on the initial resource input model, operation and maintenance resource consumption model, running resource consumption model and penalty resource consumption model, constructs an objective function with the goal of minimizing technical resources.

[0221] S707: Based on the temporal exogenous parameters corresponding to each scenario in the complete multi-scenario set, construct the constraints of the objective function.

[0222] In one embodiment, constraints may include upper-level investment constraints and lower-level operational constraints, with the upper and lower levels coupled through an input capacity variable. Constraints may include:

[0223] Constraint (1), Input capacity constraint, limits the feasible domain of new capacity for each type of resource in a single stage:

[0224]

[0225] Multi-stage capacity recursion constraints (Stage 0 is the existing stock at the initial planning stage) (Take 0 when there is no stock):

[0226]

[0227] In the formula, For resources In the stage The single-phase increase in capacity limit.

[0228] The following constraints (2)-(9) apply to all All scenarios All moments All of these hold true; the temporal exogenous parameters of the scene. It is taken directly from the scene data.

[0229] Constraint (2), renewable energy power penetration rate constraint, requiring that by the end of the planning period (stage) The weighted net wind and solar power generation across all scenarios meets the set penetration rate target. :

[0230]

[0231] The weighted total electricity consumption at the end of the period is:

[0232]

[0233] In the formula, This refers to the shortfall in renewable energy power. This represents the load shedding amount at the end of the period.

[0234] Constraint (3) Total carbon emission constraint, limiting the annual carbon emissions of thermal power generation at the end of the planning period to not exceed the carbon emission ceiling target, and introducing excess allowances. Soften:

[0235]

[0236] In the formula, Carbon emissions per unit of electricity generated by thermal power plants; The target value for the carbon emission ceiling at the end of the planning period; Corresponding penalty costs .

[0237] Constraint (4), Real-time power supply and demand balance constraint: At any stage, in any scenario, and at any time, the sum of the aggregated output of various power sources, the cross-regional exchange power, the energy storage charging and discharging, and the demand response within the system must equal the load (including load shedding correction):

[0238]

[0239] In the formula, , These represent the inter-regional switching power received and transmitted within the region; load These are exogenous parameters in the scene data; This is the load shedding amount.

[0240] Constraint (5), thermal power operation constraints: In this method, thermal power is treated as an aggregated capacity block at the planning level to accumulate the total installed capacity. As the upper limit of aggregate capacity, it does not distinguish between individual units. A continuous variable is introduced. Representation stage Scene ,time The online capacity ratio, and the startup capacity variable. :

[0241] Output upper and lower limits:

[0242]

[0243] Climbing speed constraint:

[0244]

[0245] Start / stop logic constraints:

[0246]

[0247] In the formula, The minimum technical output ratio for thermal power; , The uphill and downhill ramp rates; online capacity ratio Continuous relaxation is performed at the planning layer to ensure the solvability of the model in large-scale, multi-scenario applications.

[0248] Constraint (6), Solar thermal operation constraints: The solar thermal power plant adopts the thermal storage-power generation decoupled model (TES), which is modeled as a aggregated capacity block, and the upper limit of rated power generation is... The thermal power input is determined by the normalized irradiance factor from the scene data. Determined by the product of the current period's cumulative installed capacity:

[0249] Power generation range:

[0250]

[0251] Power generation regulation rate:

[0252]

[0253] Recurrence relation for thermal storage state:

[0254]

[0255] Thermal storage capacity constraints:

[0256]

[0257] In the formula, The input coefficient for the normalized heat collection power calculated based on DNI in the scenario data of Section 1 is an exogenous parameter that is given independently for each scenario and time. For the heat collection efficiency of the heat collection field; For the thermoelectric conversion efficiency of the steam turbine; The rated duration of the thermal storage system; For a moment The amount of thermal energy stored in a thermal storage system.

[0258] Constraint (7): Wind and solar power output constraints, actual output is non-negative and does not exceed the maximum generateable power driven by scenario data. Wind and solar power output coefficients. Exogenous parameters from the scene generated in the previous scenario:

[0259]

[0260]

[0261] Constraint (8), power flow constraints of transmission sections:

[0262]

[0263] In the formula, This represents the upper limit of the external transmission capacity of the cross-section.

[0264] Constraint (9), Energy storage operation constraints: The energy storage system is also treated as an aggregated capacity block, and the upper limit of the rated charge and discharge power is... :

[0265] Charge and discharge power constraints:

[0266]

[0267] Recurrence relation for state of charge:

[0268]

[0269] Charge range constraints:

[0270]

[0271] In the formula, For charge and discharge efficiency; The rated continuous charge and discharge duration for energy storage; For a moment The charge capacity.

[0272] Constraint (10), demand-side response constraint, the adjustable capacity of the demand-side response is treated as a known upper limit constraint and is not included in the planning decision:

[0273]

[0274]

[0275] In the formula, This represents the maximum response power limit for a single time period in demand response; This is the maximum cumulative response power limit for a single day.

[0276] S708 combines the objective function with the constraints to form a MILP planning model oriented towards minimizing comprehensive resource consumption throughout the entire life cycle.

[0277] S709, solve the planning model to obtain the initial optimal installed capacity scheme for each type of resource at each stage, and determine the initial optimal installed capacity scheme as the round resource planning strategy for the first iteration round.

[0278] Among them, the initial optimal installed capacity scheme .

[0279] S710 reads the complete multi-scene set and various resource flexibility-related parameters, and sets the iteration count k=0.

[0280] S711 substitutes the current iteration's resource planning strategy into the complete multi-scenario centralized time-series simulation to calculate the available flexibility and flexibility requirements at each time point and in each direction.

[0281] S712 calculates the flexibility adequacy index based on flexibility availability and flexibility requirements.

[0282] S713 determines whether the resource planning strategy meets the flexibility adequacy index. If yes, it is executed; otherwise, S714 is executed.

[0283] S714, according to the preset allocation step size, assuming that the three types of flexibility resources are allocated in sequence, the allocation amount of each flexibility resource is obtained.

[0284] Among them, the step size for increasing the allocation is Three types of flexible resources are .

[0285] S715, for each flexible resource, performs time-series simulations according to the amount of additional allocation, and determines the first total resource gain value of the system in the current iteration round under the condition that each flexible resource is increased.

[0286] S716. Based on the first total resource gain value and the second total resource gain value of the new energy base in the previous iteration with the increase in flexible resources, determine the net resource gain increment rate of flexible resources.

[0287] S717, the flexibility resource corresponding to the largest incremental increase among the incremental rates of net resource gains is determined as the target flexibility resource.

[0288] S718, increase the resource input of the target flexibility resource according to the preset increment step size, obtain the round resource planning strategy of each flexibility resource in the current iteration round, let k+1, and return to execute S711.

[0289] S719 defines the current iteration's resource planning strategy as the flexible resource planning strategy for the new energy base.

[0290] The aforementioned method employs a hierarchical scenario generation strategy to construct a set of coupled 'typical + extreme' scenarios. Considering the technical and economic efficiency throughout the entire lifecycle, it establishes a method for the coordinated optimization of flexible resources in new energy bases and proposes a solution algorithm, providing a reference method for the coordinated optimization of flexible resources in new energy bases. The objective function of the integrated optimization model for planning and operation considers the optimal resource consumption across all stages and adopts a two-stage algorithm that maximizes the incremental rate of net resource gain during the flexible resource allocation process, ensuring optimal overall resource consumption for new energy bases.

[0291] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0292] Based on the same inventive concept, this application also provides a flexible resource planning device for a new energy base to implement the flexible resource planning method for a new energy base as described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the flexible resource planning device for a new energy base provided below can be found in the limitations of the flexible resource planning method for a new energy base described above, and will not be repeated here.

[0293] In one embodiment, such as Figure 8 As shown, a flexible resource planning device 800 for a new energy base is provided, including: an objective function construction module 801, an operation scenario construction module 802, a solution module 803, and an iterative optimization module 804, wherein:

[0294] The objective function construction module 801 is used to construct an objective function based on the full life cycle technical resource model of the new energy base with the goal of minimizing technical resources. The full life cycle technical resource model includes an initial resource input model, an operation and maintenance resource consumption model, an operation resource consumption model, and a penalty resource consumption model.

[0295] The scenario construction module 802 is used to construct multi-dimensional scenario energy information of the new energy base under multiple time-series operation scenarios based on the historical energy usage data of the new energy base; the multi-dimensional scenario energy information includes the energy information of node units in multiple dimensions in the new energy base.

[0296] The solver module 803 is used to construct the constraints of the objective function based on the multi-dimensional scenario energy information, solve the objective function under the constraints, and obtain the initial resource planning strategy for various resources in the new energy base.

[0297] The iterative optimization module 804 is used to simulate the operation of the new energy base under various time-series operation scenarios using the initial resource planning strategy, iteratively and incrementally configure the flexibility resources in the new energy base, and obtain the flexibility resource planning strategy of the new energy base.

[0298] In one embodiment, the scenario construction module 802 is used to: fit the edge probability distribution of the time-series power of each node unit in the new energy base according to the historical energy usage data of the new energy base; construct the joint distribution of the new energy base using an advanced probability generation model based on each edge probability distribution; the joint distribution is used to describe the spatiotemporal dependence characteristics between each source-load variable in the new energy base; and repeatedly sample the joint distribution to obtain multi-dimensional scenario energy information under multiple time-series operation scenarios.

[0299] In one embodiment, the scenario construction module 802 is used to: repeatedly sample the joint distribution to obtain the initial scenario energy information under each initial time-series scenario; and cluster each initial time-series scenario according to a preset clustering type based on the initial scenario energy information to obtain multiple time-series scenarios under the clustering type, as well as the multi-dimensional scenario energy information corresponding to each time-series scenario.

[0300] In one embodiment, the clustering type includes a typical scenario clustering type; the multidimensional scenario energy information includes typical scenario energy information. The running scenario construction module 802 is used to: extract the time-series curve features of each initial time-series running scenario from the energy information of each initial scenario; and cluster each initial time-series running scenario based on the feature similarity between the features of each time-series curve to obtain each typical scenario and the typical scenario energy information corresponding to each typical scenario.

[0301] In one embodiment, the clustering type includes an extreme scenario clustering type; the multidimensional scenario energy information includes extreme scenario energy information. The scenario construction module 802 is used to: use predefined vulnerability indicators to select initial risk scenarios from each initial time-series running scenario based on each initial energy scenario information; extract the extreme characteristic features of each initial risk scenario from its respective initial scenario energy information; and cluster each initial risk scenario based on the feature similarity between the extreme characteristic features to obtain each extreme scenario and its corresponding extreme scenario energy information.

[0302] In one embodiment, the iterative optimization module 804 is used to: use an initial resource planning strategy and multi-dimensional scenario energy information to perform time-series operation simulations of the new energy base under various time-series operation scenarios, and obtain the flexibility availability and flexibility demand of each of the multiple flexibility resources in the new energy base; construct a flexibility sufficiency index for each flexibility resource based on the flexibility availability and flexibility demand; determine the target flexibility resource with the maximum net resource gain incremental rate among the flexibility resources based on a preset allocation step size; increase the resource input of the target flexibility resource according to the preset allocation step size, and obtain the round resource planning strategy for each flexibility resource in the current iteration round; if the round resource planning strategy does not meet the flexibility sufficiency index, use the round resource planning strategy as the initial resource planning strategy, and return to execute the step of using the initial resource planning strategy and multi-dimensional scenario energy information to perform time-series operation simulations of the new energy base under various time-series operation scenarios; if the round resource planning strategy meets the flexibility sufficiency index, determine the round resource planning strategy as the flexibility resource planning strategy for the new energy base.

[0303] In one embodiment, the iterative optimization module 804 is configured to: for each flexible resource, increase the resource input of the flexible resource according to a preset incremental step size to obtain the incremental input of the flexible resource; based on the incremental input, determine the first total resource gain value of the new energy base in the current iteration round when the flexible resource is increased; based on the first total resource gain value and the second total resource gain value of the new energy base in the previous iteration round when the flexible resource is increased, determine the net resource gain incremental rate of the flexible resource; and determine the flexible resource corresponding to the largest incremental rate among the net resource gain incremental rates as the target flexible resource.

[0304] The various modules in the flexible resource planning device of the aforementioned new energy base can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0305] In one embodiment, a computer device is provided, which may be a server integrating a resource planning system, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing 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 an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores data related to a flexible resource planning method for new energy bases. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a flexible resource planning method for new energy bases.

[0306] Those skilled in the art will understand that Figure 9 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.

[0307] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the specific steps of the above-described embodiment of the flexible resource planning method for new energy bases.

[0308] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, it implements the specific steps of the above-described embodiment of the flexible resource planning method for new energy bases.

[0309] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the specific steps of the above-described embodiment of the flexible resource planning method for new energy bases.

[0310] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the acquisition, storage, processing, and transmission of the data all comply with relevant laws and regulations.

[0311] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0312] 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.

[0313] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for flexible resource planning of a new energy base, characterized in that, The method includes: With the goal of minimizing technical resources, an objective function is constructed based on the full life cycle technical resource model of the new energy base. The full life cycle technical resource model includes an initial resource input model, an operation and maintenance resource consumption model, an operation resource consumption model, and a penalty resource consumption model. Based on the historical energy usage data of the new energy base, multi-dimensional scenario energy information of the new energy base under multiple time-series operation scenarios is constructed; the multi-dimensional scenario energy information includes the energy information of node units in multiple dimensions of the new energy base; Based on the multi-dimensional scenario energy information, the constraints of the objective function are constructed, and the objective function is solved under the constraints to obtain the initial resource planning strategy for various resources in the new energy base. Using the initial resource planning strategy, the new energy base is simulated for time-series operation under each of the time-series operation scenarios. The flexible resources in the new energy base are iteratively and incrementally configured to obtain the flexible resource planning strategy for the new energy base.

2. The method of claim 1, wherein, The process of constructing multi-dimensional scenario energy information for the new energy base under multiple time-series operation scenarios based on historical energy usage data includes: Based on the historical energy usage data of the new energy base, fit the marginal probability distribution of the time-series power of each node unit in the new energy base in multiple dimensions; Based on the aforementioned marginal probability distributions, an advanced probability generation model is used to construct the joint distribution of the new energy base; the joint distribution is used to describe the spatiotemporal dependence characteristics between the source and load variables in the new energy base. Repeated sampling of the joint distribution yields multi-dimensional scene energy information under multiple time-series operating scenarios.

3. The method of claim 2, wherein, The method involves repeatedly sampling the joint distribution to obtain multi-dimensional scene energy information under multiple time-series operating scenarios, including: Repeated sampling is performed on the joint distribution to obtain the initial scenario energy information under each initial time series running scenario; According to the preset clustering type, each initial time-series running scenario is clustered based on the initial scenario energy information to obtain multiple time-series running scenarios under the clustering type, as well as multi-dimensional scenario energy information corresponding to each time-series running scenario.

4. The method of claim 3, wherein, The clustering type includes typical scenario clustering type; the multidimensional scenario energy information includes typical scenario energy information; The step involves clustering each initial time-series running scenario according to a preset clustering type, based on the initial scenario energy information, to obtain multiple time-series running scenarios under the clustering type, and multi-dimensional scenario energy information corresponding to each time-series running scenario, including: Extract the time-series curve features of each initial time-series running scenario from the energy information of each initial scenario; Based on the feature similarity between the features of each time series curve, each initial time series operation scenario is clustered to obtain each typical scenario and the typical scenario energy information corresponding to each typical scenario.

5. The method of claim 3, wherein, The clustering type includes extreme scenario clustering type; the multidimensional scenario energy information includes extreme scenario energy information; The step involves clustering each initial time-series running scenario according to a preset clustering type, based on the initial scenario energy information, to obtain multiple time-series running scenarios under the clustering type, and multi-dimensional scenario energy information corresponding to each time-series running scenario, including: Using predefined vulnerability indicators, initial risk scenarios are selected from each of the initial time-series operation scenarios based on the initial energy scenario information. Extract the extreme characteristic features of each initial risk scenario from the initial scenario energy information of each initial risk scenario; Based on the feature similarity between the extreme characteristics, the initial risk scenarios are clustered to obtain the extreme scenarios and their corresponding extreme scenario energy information.

6. The method according to any one of claims 1 to 5, characterized in that, The step of using the initial resource planning strategy and the multi-dimensional scenario energy information to incrementally iteratively configure the flexible resources in the new energy base to obtain the flexible resource planning strategy for the new energy base includes: Using the initial resource planning strategy and the multi-dimensional scenario energy information, the new energy base is simulated under each of the time-series operation scenarios to obtain the flexibility availability and flexibility demand of each of the multiple flexibility resources in the new energy base. Based on the available flexibility and the required flexibility, construct a flexibility sufficiency index for each flexibility resource. Based on a preset incremental step size, determine the target flexible resource among the flexible resources that has the largest net resource gain incremental rate; Increase the resource input of the target flexibility resources according to the preset incremental step size to obtain the round resource planning strategy of each flexibility resource in the current iteration round; If the round-based resource planning strategy does not meet the flexibility adequacy index, the round-based resource planning strategy is used as the initial resource planning strategy, and the process returns to the step of using the initial resource planning strategy and the multi-dimensional scenario energy information to perform time-series operation simulation on the new energy base in each of the time-series operation scenarios. If the resource planning strategy of the specified round meets the flexibility adequacy index, the resource planning strategy of the specified round will be determined as the flexible resource planning strategy of the new energy base.

7. The method of claim 6, wherein, The step of determining the target flexible resources with a slight increase in net resource gain among the various flexible resources based on a preset incremental step size includes: For each of the aforementioned flexible resources, the resource input of the flexible resource is increased according to a preset incremental step size to obtain the incremental input of the flexible resource. Based on the increased allocation, determine the first total resource gain value of the new energy base in the current iteration round under the condition of increased flexible resource allocation; Based on the first total resource gain value and the second total resource gain value of the new energy base in the previous iteration when the flexible resources were increased, the net resource gain increment rate of the flexible resources is determined. The flexibility resource corresponding to the largest incremental increase in the net resource gain rate is determined as the target flexibility resource.

8. A flexible resource planning device for a new energy base, characterized by, The device includes: The objective function construction module is used to construct an objective function based on the full life cycle technical resource model of the new energy base with the goal of minimizing technical resources. The full life cycle technical resource model includes an initial resource input model, an operation and maintenance resource consumption model, an operation resource consumption model, and a penalty resource consumption model. The scenario construction module is used to construct multi-dimensional scenario energy information of the new energy base under multiple time-series operation scenarios based on the historical energy usage data of the new energy base; the multi-dimensional scenario energy information includes the energy information of node units in multiple dimensions of the new energy base; The solution module is used to construct the constraints of the objective function based on the multi-dimensional scenario energy information, solve the objective function under the constraints, and obtain the initial resource planning strategy for various resources in the new energy base. The iterative optimization module is used to simulate the operation of the new energy base under each of the time-series operation scenarios using the initial resource planning strategy, and iteratively incrementally configure the flexibility resources in the new energy base to obtain the flexibility resource planning strategy of the new energy base.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images