CN113705874A

CN113705874A - New energy power grid evolution prediction method and device, computer equipment and storage medium

Info

Publication number: CN113705874A
Application number: CN202110952453.4A
Authority: CN
Inventors: 孙荣富; 徐海翔; 吴林林; 乔颖
Original assignee: Tsinghua University; State Grid Corp of China SGCC; Electric Power Research Institute of State Grid Jibei Electric Power Co Ltd
Current assignee: Tsinghua University; State Grid Corp of China SGCC; Electric Power Research Institute of State Grid Jibei Electric Power Co Ltd
Priority date: 2021-08-19
Filing date: 2021-08-19
Publication date: 2021-11-26
Anticipated expiration: 2041-08-19
Also published as: CN113705874B

Abstract

The application relates to a new energy power grid evolution prediction method and device, computer equipment and a storage medium. The method comprises the following steps: determining a key index set corresponding to a target area; the key index set comprises a power generation resource index, an energy storage index, a power demand side index, a flexibility resource index and a fuel price index which influence the development of a target regional power grid; carrying out numerical value prediction processing on each index in the key index set to obtain a key index sample set; processing the key index sample set by using a power grid evolution prediction algorithm to obtain multiple groups of evolution data corresponding to the target area; the evolution data comprises evolution data of a power generation resource structure corresponding to the target area and evolution data of a flexible resource structure corresponding to the target area. By adopting the method, the automatic intelligent prediction of the power grid evolution path can be realized, the influence of human factors is eliminated, the accuracy of the generated evolution path is ensured, and the predicted power grid evolution path has certain global coverage.

Description

New energy power grid evolution prediction method and device, computer equipment and storage medium

Technical Field

The application relates to the technical field of electric power, in particular to a new energy power grid evolution prediction method, a new energy power grid evolution prediction device, computer equipment and a storage medium.

Background

As flexibility resources (such as thermal power generating unit flexibility modification, pumped storage, battery storage, compressed air storage, load transfer type demand side response and load reduction type demand side response) continuously participate in evolution and development of the power grid, the evolution of the power grid faces a plurality of uncertain factors, so that the evolution of the power grid has depth uncertainty. The method has great challenges for identifying key driving factors of power grid evolution and judging paths of the power grid evolution.

In the traditional technology, the possible evolution path of the power grid is predicted mainly by human experience. A plurality of possible evolution paths are predicted in advance according to prior experience, and the possible evolution paths are used as rough evolution development paths of future systems.

However, the traditional power grid evolution path prediction method is greatly influenced by human factors, and only a few possible evolution paths of the power grid are predicted, so that the paths are used as rough evolution development paths of a future system, the prediction result lacks universality, and the accuracy of the evolution paths is difficult to guarantee in the prior art.

Disclosure of Invention

The application provides a new energy power grid evolution prediction method, a new energy power grid evolution prediction device, computer equipment and a storage medium, which can improve the global coverage of a power grid evolution path prediction and ensure the accuracy of the power grid evolution path prediction to a certain extent.

In a first aspect, a power grid evolution prediction method is provided, and the method includes: determining a key index set corresponding to a target area; the key index set comprises a power generation resource index influencing the development of a target regional power grid, an energy storage index influencing the development of the target regional power grid, a power demand side index influencing the development of the target regional power grid, a flexibility resource index and a fuel price index influencing the development of the target regional power grid; carrying out numerical value prediction processing on each index in the key index set to obtain a key index sample set; processing the key index sample set by using a power grid evolution prediction algorithm to obtain multiple groups of evolution data corresponding to the target area; the evolution data comprises evolution data of a power generation resource structure corresponding to the target area and evolution data of a flexible resource structure corresponding to the target area.

With reference to the first aspect, in a possible implementation manner of the first aspect, performing numerical prediction processing on each index in a set of key indexes to obtain a set of key index samples includes: aiming at each index in the key index set, determining an upper limit value and a lower limit value of each index in a prediction time period, and performing assignment processing on each index in the key index set according to the upper limit value and the lower limit value of each index; and randomly sampling the key index set subjected to assignment processing by adopting a Monte Carlo method to obtain the value of each index in each prediction time period, and determining a key index sample set according to the value of each index in each prediction time period.

With reference to the first aspect, in a possible implementation manner of the first aspect, processing a key index sample set by using a power grid evolution prediction algorithm to obtain multiple sets of evolution data corresponding to a target region includes: inputting the key index sample set into a target function in a power grid evolution prediction algorithm, and performing constraint processing on the key index sample set by using constraint conditions in the power grid evolution prediction algorithm; determining multiple sets of evolution data corresponding to the target area according to the output of the target function and the result of the constraint processing; wherein the objective function is: optimizing the sum of investment cost, fixed operation and maintenance cost and variable operation cost of the power grid in each prediction period by power grid modeling; the constraint conditions include: construction constraints of the power grid and operating cost constraints of the power grid.

With reference to the first aspect, in a possible implementation manner of the first aspect, the investment cost is a capital cost invested in building power generation resources and flexible resources in a power system; the fixed operation and maintenance cost of the power grid is the capital cost invested in daily maintenance of equipment built in the power system; the variable operating cost of the power grid is the cost of fuel consumed by the power generation resources in the power system when operating.

With reference to the third aspect, in a possible implementation manner of the third aspect, the construction constraint of the power grid includes at least one of: investment capacity constraint, total investment capacity constraint, renewable energy permeability constraint of the final stage of power grid evolution and carbon emission constraint of the final stage of power grid evolution of various resources on each power grid node in each prediction period; the operating cost constraint of the power grid is the capital cost required to maintain the normal operation of various power generation resources in the power system.

With reference to the first aspect, in a possible implementation manner of the first aspect, for any one key index in a set of key indexes, calculating a difference degree of the key index in different evolution data, and taking a maximum difference degree as an influence factor of the key index; determining one or more key indexes with influence factors larger than a preset threshold in the key index set as key driving factors of power grid evolution of a target region; the key driver is used to predict evolution data.

With reference to the first aspect, in a possible implementation manner of the first aspect, a flexible resource development index of a target region is calculated, where the flexible resource development index includes a flexible resource capacity requirement, a flexible resource marginal capacity requirement, a battery energy storage occurrence point, and a battery energy storage leading point; the flexibility resource capacity demand is the flexibility resource capacity of the system under different renewable energy electric quantity permeabilities, the flexibility resource marginal capacity demand is the flexibility resource capacity increment corresponding to a preset numerical value of permeability increase under different renewable energy electric quantity permeabilities, the battery energy storage appearance point is the renewable energy electric quantity permeability corresponding to the battery energy storage in the electric power system, and the battery energy storage leading point is the renewable energy electric quantity permeability corresponding to the battery energy storage capacity accounting for the preset proportion of the flexibility resource capacity.

In a second aspect, there is provided a power grid evolution prediction apparatus, the apparatus comprising: the determining module is used for determining a key index set corresponding to the target area; the key index set comprises a power generation resource index influencing the development of a target area power grid, an energy storage index influencing the development of the target area power grid, a power demand side index influencing the development of the target area power grid, a flexibility resource index and a fuel price index influencing the development of the target area power grid;

the numerical value processing module is used for carrying out numerical value prediction processing on each index in the key index set to obtain a key index sample set;

the evolution data generation module is used for processing the key index sample set by utilizing a power grid evolution prediction algorithm to obtain multiple groups of evolution data corresponding to the target area; the evolution data comprises evolution data of a power generation resource structure corresponding to the target area and evolution data of a flexible resource structure corresponding to the target area.

In a third aspect, a computer device is provided, comprising a memory and a processor, the memory storing a computer program. The steps of the method described in the first aspect or any one of the possible implementations of the first aspect above are implemented when a processor executes a computer program.

In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to the first aspect or any one of the possible implementation forms of the first aspect.

The application provides a new energy power grid evolution prediction method, a new energy power grid evolution prediction device, computer equipment and a storage medium, wherein a key index set corresponding to a target area can be determined, and numerical prediction processing is performed on each index in the key index set to obtain a key index sample set; finally, processing the key index sample set by using a power grid evolution prediction algorithm to obtain multiple sets of evolution data corresponding to the target area; the evolution data are used for representing the evolution trend, the evolution path and the like of the power grid. The key index set comprises a power generation resource index influencing the development of a target regional power grid, an energy storage index influencing the development of the target regional power grid, an electricity demand side index influencing the development of the target regional power grid, a flexibility resource index and a fuel price index influencing the development of the target regional power grid. Therefore, the computer in the application can realize automatic intelligent prediction of the power grid evolution path based on the power grid evolution prediction algorithm, and solves the problem that in the prior art, the accuracy of the evolution path is difficult to guarantee due to the fact that the power grid evolution path prediction is carried out manually. Influence of human factors on power grid evolution prediction can be eliminated based on a power grid evolution prediction algorithm. In addition, the key index set basically covers various indexes influencing the power grid development, specifically comprises some indexes related to the flexibility resource allocation, uncertainty factors such as the flexibility resource allocation are fully considered when the power grid path is predicted, therefore, a relatively comprehensive evolution path can be obtained based on the key index set and a power grid evolution prediction algorithm, and the predicted power grid evolution path has certain global coverage.

Drawings

FIG. 1 is a schematic flow chart of a power grid evolution prediction method in one embodiment;

FIG. 2 is a flow diagram illustrating a numerical prediction processing method according to one embodiment;

FIG. 3 is a schematic flow chart diagram of a method of evolving data analysis in one embodiment;

FIG. 4 is a diagram illustrating a flexible modification path of a thermal power generating unit in one embodiment;

FIG. 5 is a schematic diagram of key driver identification based on time varying patterns in one embodiment;

FIG. 6 is a schematic diagram of the current installation situation and the load situation in the northwest region of an embodiment;

FIG. 7 is a diagram of the system evolution path and its distribution in one embodiment;

FIG. 8 is a schematic diagram of flexible resource demand with different power rates of renewable energy in one embodiment;

FIG. 9 is a schematic diagram of a statistical case of battery energy storage in a large number of evolution paths in one embodiment;

FIG. 10 is a diagram of clustering results of renewable energy permeability evolution data in one embodiment;

FIG. 11 is a graph illustrating the mean of key indicators corresponding to different classes of paths in one embodiment;

FIG. 12 is a graph of the average distance between classes for key indicators in one embodiment;

FIG. 13 is a block diagram of the generation and analysis of the mass evolutionary paths in one embodiment;

FIG. 14 is a schematic structural diagram of a power grid evolution prediction device in one embodiment;

FIG. 15 is a schematic diagram of a structure of analyzing power grid evolution data in one embodiment;

FIG. 16 is a diagram illustrating a structure of adjusting power grid evolution data according to an embodiment;

FIG. 17 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The embodiment of the application provides a power grid evolution prediction method, which is applied to computer equipment, and as shown in fig. 1, the method comprises the following steps:

step 101, determining a key index set corresponding to a target area; the key index set comprises a power generation resource index influencing the development of a target regional power grid, an energy storage index influencing the development of the target regional power grid, a power demand side index influencing the development of the target regional power grid, a flexibility resource index and a fuel price index influencing the development of the target regional power grid;

in order to realize the automatic prediction of the power grid evolution path, multiple uncertain factors influencing the power grid development need to be determined, so that the power grid evolution path is predicted based on the uncertain factors, and the predicted evolution path can take the influence of the uncertain factors into account. In specific implementation, when power grid evolution prediction is performed on a certain region, some key indexes influencing power grid evolution of the region can be determined. For example, when power grid evolution prediction is performed on a target area, a key index set of the target area is determined first, where the key index set includes some key indexes that affect the power grid evolution direction of the target area.

In one possible implementation, the set of key indicators includes a power generation resource indicator, an energy storage indicator, a power demand side indicator, a flexibility resource indicator, and a fuel price indicator that affect the development of the target regional power grid.

The power generation resource index may be an investment parameter of each power generation resource in the target region, and the investment parameter may represent investment details of the corresponding power generation resource, for example, may be an investment price of each power generation resource. For example, the power generation resource may be coal electricity, gas electricity, hydroelectric power, wind electricity, photovoltaic, photo-thermal.

The energy storage index may be energy storage efficiency of a target area flexible resource, and the energy storage efficiency may represent emergency capacity of the power system, for example, the energy storage efficiency may be lithium battery energy storage efficiency and compressed air energy storage efficiency. The lithium battery energy storage mainly utilizes chemical reaction of chemical substances in the lithium ion battery to store or release electric power, and the compressed air energy storage mainly utilizes residual electric power generated in the load valley of a power grid to compress air and store the air in a high-voltage sealing facility, and the air is released at the peak of power utilization to drive a gas turbine to generate power.

The electricity demand side index may be an electricity demand response of the target area power system, and the electricity demand response may represent the reliability of the operation of the power system, and may be, for example, an investment price of the electricity demand side response, a compensation price of the electricity demand side response, and a capacity potential of the electricity demand side response. The investment price responded by the electricity demand side is the capital invested for ensuring the balance of supply and demand and the reliability of system operation when the power system is in need of users or in power shortage, and illustratively, the investment price responded by the electricity demand side can be the unit investment price responded by the transfer demand side and the unit investment price responded by the reduction demand side; the compensation price responded by the electricity demand side is that the user reduces the electricity demand when the system needs or the electricity is in tension so as to obtain direct compensation or preferential electricity price in other periods, and the compensation price responded by the electricity demand side can be a transfer type demand side response unit compensation price and a reduction type demand side response unit compensation price; the capacity potential responded by the electricity demand side is the maximum electric power which can be provided for a user or electric equipment by the electric power system when the user needs or the electric power is in tension, and the capacity potential responded by the electricity demand side can be a transfer type demand side response capacity potential and a reduction type demand side response capacity potential.

The flexible resource indicator may be an investment parameter of various flexible resources of the target area, and the investment parameter may characterize investment details of the corresponding flexible resource, for example, may be an investment price of the various flexible resources or a maximum investment capacity in each prediction period (a maximum amount of money that the power system can invest on the flexible resources in each prediction period). For example, the investment price of the flexible resource can be the unit investment price of pumped storage, the unit investment price of lithium battery energy storage, and the unit investment price of compressed air energy storage; the pumped storage is a hydropower station which pumps water to an upper reservoir by using electric energy in the low ebb of the electric load and discharges water to a lower reservoir to generate electricity in the peak period of the electric load. When the electric power consumed by electric equipment in the electric power system is small, the residual electric power is used for pumping water to an upper reservoir, and the electric energy is converted into the potential energy of the water; when the electric power consumed by the electric equipment in the electric power system is large, the water is discharged to the lower reservoir, and the potential energy of the water is converted into electric energy. The maximum investment capacity of the flexible resource in each prediction period can be the maximum investment capacity of the lithium battery energy storage in each stage and the maximum investment capacity of the compressed air energy storage in each stage. The maximum investment capacity of the lithium battery energy storage at each stage is the maximum fund which can be invested on the lithium battery energy storage of the power system in each prediction time interval; the maximum investment capacity of the compressed air energy storage at each stage is the maximum fund which can be invested on the compressed air energy storage equipment by the power system in each prediction period.

The fuel price index may be a price of fuel consumed when various power generation resources of the target region are operated. For example, the fuel price may be a coal-electricity fuel price, a gas-electricity fuel price.

The following table 1 provides a set of key indicators for the embodiment of the present application, which includes 21 key indicators (uncertainty factors) that affect the power grid development evolution.

In specific implementation, the key index set of the target area may be determined according to the actual development condition of the power grid in the target area. The computer device may receive a set of key indicators input by a user, or the set of key indicators may be pre-stored in the computer device.

TABLE 1

102, performing numerical prediction processing on each key index in a key index set to obtain a key index sample set;

in the embodiment of the application, in order to determine the evolution data corresponding to the target region based on the critical index set, each index in the critical index set needs to be assigned. For example, a numerical prediction process may be performed on each key index in the set of key indices to obtain a sample set of key indices.

In specific implementation, each key index in the key index set can be quantized, and a random sampling method is adopted to sample a quantized result to obtain a key index sample set. The quantization may be to determine a possible value of each index, and the sampling may be to sample possible values of each index. The data obtained by one-time sampling is a group of key index samples, and the data obtained by multiple times of sampling is a key index sample set.

In a possible implementation manner, the determined uncertainty factors affecting the power grid development are quantized, and the quantized data can be obtained by analyzing and predicting historical data or determined according to other related documents; the quantized uncertainty factors influencing the power grid development are sampled, and a random sampling method can be adopted for sampling.

103, processing the key index sample set by using a power grid evolution prediction algorithm to obtain multiple sets of evolution data corresponding to the target area; the evolution data comprises evolution data of a power generation resource structure corresponding to the target area and evolution data of a flexible resource structure corresponding to the target area.

In the prior art, the power grid evolution path is manually predicted based on human experience, and in the embodiment of the application, the computer equipment can obtain the power grid evolution path through an automatic processing algorithm. For example, when power grid evolution prediction is performed on a target region, a key index sample set corresponding to the region may be processed according to a power grid evolution prediction algorithm to obtain multiple sets of evolution data corresponding to the region.

In specific implementation, the key index sample set is input into a power grid evolution prediction algorithm, and reasonable evolution of each index sample under certain constraint conditions can be determined, so that power grid evolution data of the region can be constructed according to the evolved index samples.

In one possible implementation, the evolution data is used to characterize an evolution trend of the power grid or an evolution path of the power grid. The multiple sets of evolution data can determine multiple power grid evolution paths of the target region. For example, the evolution data may include evolution data of power generation resource structures corresponding to the target region and evolution data of flexible resource structures corresponding to the target region.

The evolution data is used for representing an evolution trend, and can be a change trend of a power generation resource structure and a flexible resource structure corresponding to the target area in the coming years. The power generation resource structure refers to the configuration of various power generation resources deployed in a target area, and specifically may be the percentage of the construction quantity of the various power generation resources in the target area to the total construction quantity. The flexible resource structure refers to the configuration of various flexible resources deployed in the target area, and specifically may be the percentage of the number of the various flexible resources in the target area to the total number of the flexible resources.

In the power grid evolution prediction method provided by the embodiment of the application, the computer can realize automatic intelligent prediction of the power grid evolution path based on the power grid evolution prediction algorithm, and the problem that in the prior art, the accuracy of the evolution path is difficult to guarantee due to the fact that the power grid evolution path prediction is carried out manually is solved. Influence of human factors on power grid evolution prediction can be eliminated based on a power grid evolution prediction algorithm. In addition, the key index set basically covers various indexes influencing the power grid development, specifically comprises some indexes related to the flexibility resource allocation, and uncertainty factors such as the flexibility resource allocation are fully considered when the power grid path is predicted, so that a relatively comprehensive evolution path can be obtained based on the key index set and a power grid evolution prediction algorithm, and the predicted power grid evolution path has certain global coverage.

In step 102, the computer device may obtain a sample set of key indicators by quantizing and sampling the set of key indicators. For example, the specific implementation of "performing numerical prediction processing on each index in the set of key indexes to obtain a sample set of key indexes" referred to above includes the steps shown in fig. 2:

step 201, determining an upper limit value and a lower limit value of each index in a prediction time period for each index in the key index set, and performing assignment processing on each index in the key index set according to the upper limit value and the lower limit value of each index.

In the embodiment of the application, in order to generate specific power grid evolution data, a key index set needs to be quantized, and the specific data represents the key index set. For example, each index in the set of key indexes may be assigned to obtain an assigned set of key indexes.

In specific implementation, each key index in a key index set is divided into a plurality of prediction time periods, an upper limit value and a lower limit value (an upper limit and a lower limit) of each key index in each prediction time period are determined, the key index set is represented in a multi-stage (a plurality of prediction time periods) and interval (an upper limit value interval and a lower limit value interval) mode, and assignment processing of the key index set is completed;

in one possible implementation, each key index in the set of key indices is divided into a plurality of prediction periods. For example, with reference to table 2, 5 prediction periods each of 5 years may be specifically divided between 2025 and 2050 years; and determining an upper limit value and a lower limit value of each key index in a prediction period, wherein the upper limit value and the lower limit value of each year can be unchanged in one prediction period, and the upper limit value and the lower limit value of each key index in the prediction period can be obtained by analyzing and predicting historical data or can be determined according to other related documents.

TABLE 2

Step 202, randomly sampling the key index set subjected to assignment processing by using a monte carlo method, obtaining a numerical value of each index in each prediction time period, and determining a key index sample set according to the numerical value of each index in each prediction time period.

In order to perform algorithm-based processing on the key index set in the embodiment of the application, the valued key index set needs to be sampled to obtain a key index sample set.

In the concrete implementation, random sampling is carried out in the interval of the upper limit value and the lower limit value of each prediction time interval of each key index, a group of data obtained by random sampling is used as a group of key index samples, and the data obtained by multiple sampling is a key index sample set.

In a possible implementation manner, the assigned key index set is sampled randomly by adopting a monte carlo method, the sampling result is a specific value of each key index in the upper and lower limit intervals of each prediction time period, the specific value represents data of the key index in the prediction time period, the sampling result of one time is a group of key index samples, and the assigned key index set is sampled randomly for multiple times, so that a key index sample set is obtained.

As shown in table 1, for 21 uncertainty factors (key index sets) that influence power grid evolution development and are determined by the present application, dividing the 21 uncertainty factors into 5 prediction periods from 2025 to 2050, where each prediction period is 5 years, and then determining an upper limit value and a lower limit value of each uncertainty factor in each prediction period to obtain a quantization result of the 21 uncertainty factors; and then randomly sampling each uncertainty factor in the upper limit interval and the lower limit interval of each prediction time interval, wherein the sampled numerical value represents the data of the uncertainty factor in the prediction time interval, the result of one-time sampling is a group of key index samples, and the quantized key index set is randomly sampled for multiple times to obtain a key index sample set.

The quantitative results of the 21 uncertainty factors (key index set) in table 1 are shown in table 2. Taking the uncertainty factor of the coal-electricity unit investment price in table 2 as an example: in the table, 2025 to 2030 years are a prediction period, 2030 to 2035 years are a prediction period, and 5 prediction periods are shared between 2025 and 2050 years; 559$/kW (1 dollar per kW) is the lowest unit investment price for the coal-electricity unit investment price between 2025 and 2030, which is the lower limit value for the prediction period of 2025 to 2030, and 621$/kW is the highest unit investment price for the coal-electricity unit investment price between 2025 and 2030, which is the upper limit value for the prediction period of 2025 to 2030. When the coal and electricity unit investment price is sampled in the prediction period of 2025 to 2030, a numerical value between 559$/kW and 621$/kW is randomly extracted to serve as data of the prediction period of the coal and electricity unit investment price in the period of 2025 to 2030, 21 uncertainty factors are randomly sampled once in each prediction period, and the obtained sampling result is a group of key index samples.

The embodiment of the application provides a method for performing numerical prediction processing on each key index in a key index set to obtain a key index sample set. Specifically, aiming at a key index set influencing the evolution and development of a power grid of a target region, dividing each key index in the key index set into a plurality of prediction time periods, determining an upper limit value and a lower limit value (an upper limit and a lower limit) in each prediction time period, and finishing assignment processing of the key index set; and then randomly sampling each prediction time interval of each key index within the interval of the upper limit value and the lower limit value of the prediction time interval, taking a group of data obtained by random sampling as a group of key index samples, and obtaining data obtained by multiple times of sampling as a key index sample set. Therefore, the method and the device have the advantages that numerical prediction processing is carried out on the key index set influencing the power grid evolution development of the target area, the key index set is quantized and sampled to obtain the key index sample set, and the uncertainty factors (key index set) influencing the power grid evolution development are represented in a specific numerical value form so as to be conveniently used for a power grid evolution prediction algorithm subsequently to generate a large number of power grid evolution paths. Compared with the prior art, the power grid evolution path is predicted by means of human experience, the power grid evolution path is predicted by means of specific samples, influence of human factors is reduced, and accuracy of power grid evolution path prediction is improved.

In step 103, the computer device may process the key index sample set by using a power grid evolution prediction algorithm to obtain multiple sets of evolution data corresponding to the target region.

In the prior art, the power grid evolution path is predicted mainly by means of human experience, and the power grid evolution path is automatically generated through a power grid evolution prediction algorithm. For example, when power grid evolution prediction is performed on a target area, a key index sample set is input into a power grid evolution prediction algorithm, and multiple sets of evolution data corresponding to the area are obtained.

In specific implementation, each group of samples in the key index sample set is input to a target function in a power grid evolution prediction algorithm, then constraint processing is performed on the output of the target function by using constraint conditions, and the processed result is corresponding evolution data. Namely, the output evolution data not only meets the objective function of the power grid evolution prediction algorithm, but also meets the constraint condition.

In a possible implementation manner, the power grid evolution prediction algorithm is used for outputting a group of evolution data according to each group of samples in the key index sample set if the samples meet the objective function and the constraint condition. Wherein the objective function is: the sum of the investment cost, the fixed operation and maintenance cost and the variable operation cost of the power grid at each stage of power grid evolution is optimal; the constraint conditions include: the construction constraints of the power grid evolution and the operating cost constraints of the power grid.

The objective function includes investment cost of each stage of power grid evolution, fixed operation and maintenance cost of the power grid, and variable operation cost of the power grid. Wherein, the investment cost of each stage of power grid evolution refers to: capital cost invested by power generation resources and flexible resources used for building the power system in each prediction period; the fixed operation and maintenance cost of the power grid refers to that: capital costs invested in routine maintenance of equipment built in the power system; the variable operation cost of the power grid refers to that: the cost of fuel consumed in the operation of the power generation resources in the power system. In addition, the objective function not only considers that the sum of the investment cost of each stage of power grid evolution, the fixed operation and maintenance cost of the power grid and the variable operation cost of the power grid is minimum, but also considers the time value of capital. When the power grid evolution path is explored, the time interval is defined from 2025 to 2050, and 5 years are taken as a prediction period, the time span is longer, so that the same fund has different values in different prediction periods, namely the value of 1 piece of money in 2025 is different from that of 1 piece of money in 2050 (for example, 1 piece of money in 2025 can buy 2 pieces of money in 2050). Therefore, when the objective function considers various capital costs invested in each stage of power grid evolution, the capital values of each prediction period need to be unified into the same prediction period, that is, the time value of the capital is considered.

The constraint conditions include a construction constraint of power grid evolution of the target region and an operation cost constraint of power grid evolution. Wherein, the construction constraint of power grid evolution refers to: in the power system, constraints on various unit investment capacities and constraints on evolution targets; the method specifically comprises the following four aspects: investment capacity constraint, total investment capacity constraint, renewable energy permeability constraint of the final stage of power grid evolution and carbon emission constraint of the final stage of power grid evolution of various resources on each power grid node in each prediction stage. Wherein, the investment capacity constraint of each stage of power grid evolution refers to: in each prediction time period, the capital cost invested by constructing various resources in each provincial power system; the total investment capacity constraint refers to: in the whole prediction period, the capital cost invested by constructing various resources in the whole power system; the electric quantity permeability constraint of the renewable energy source refers to: the final stage of evolution (the prediction period from 2045 years to 2050 years) needs to reach the preset target of the electric quantity permeability of renewable energy sources (the electric quantity generated by the renewable energy sources); the carbon emission constraint refers to: maximum carbon emissions limited by the final stage of evolution (this prediction period of 2045 to 2050).

The power grid operating cost constraint refers to: in the power grid system, the capital cost required by the normal operation of various resources (power generation equipment) is supplied. The method comprises the following steps that operation constraints of the thermal power generating unit are considered in a clustering mode; the hydroelectric generating set considers the operation constraint by three-section output; the output of the wind power, photovoltaic and photo-thermal units is less than the maximum power generation capacity; the energy storage considers the power, energy and cycle constraint of the energy storage; assuming that the power transfer between nodes is a non-blocking case, the power flow between nodes takes into account the operational constraints in the traffic flow model.

The objective function and constraints described above are introduced below in conjunction with specific equations:

the objective function in the power grid evolution prediction algorithm is that the sum of the investment cost, the fixed operation and maintenance cost and the variable operation cost of each evolution stage is minimum, and the time value of capital is considered, as shown in formulas (1) to (5):

min C^INV+C^OM+C^OPR+C^Target (1)

C^Target＝c^VREQ^VRE+c^CarbQ^Carb (5)

wherein, formula (1) represents an objective function; the formula (2) represents the total investment cost, including the sum of the investment costs of each stage and each type of resource; the formula (3) represents the total operation and maintenance cost, including the sum of the fixed operation and maintenance costs of each stage and each type of resource; equation (4) represents the total operating cost, including the sum of the operating costs for each stage, each type of resource, and the load shedding penalty cost, estimated on a typical day; equation (5) represents a forward penalty cost comprising the sum of a penalty cost for not meeting the renewable energy permeability and a penalty cost for not meeting the carbon emission.

Wherein, C^INVRepresenting the total investment cost for a plurality of forecast periods; c^OMRepresenting a plurality of prediction time interval fixed operation and maintenance costs; c^OPRRepresenting a plurality of prediction period variable operating costs; c^TargetRepresenting penalty cost of gap with a preset target, wherein the preset target can be the electric quantity permeability of the renewable energy source and can also be the carbon emission; d is the annual percentage, i.e., the annual interest rate at which future assets are converted to realized values; x represents various power generation resources and flexibility resources;

representing the investment cost per unit volume of a certain resource in the nth forecasting period;

represents the construction capacity of the resource X in the nth prediction period of the kth node (the power system in province as a unit); z represents the number of years included in each prediction period; f is the annual unit fixed operation and maintenance cost;

represents the total capacity of resource X in the nth prediction period; assuming that there are S typical operation scenes in each phase (simulating the cooperative optimization operation of various types of power supplies with different proportions in the power system), the number of days corresponding to each scene in a year is ρ_s；

Representing the unit power generation cost of the thermal power generating unit;

expressing the starting cost (cost of fuel consumed by starting the thermal power unit) of the thermal power unit per unit capacity, wherein G expresses the type of the thermal power unit, including a coal power unit and a gas power unit;

representing the power generation power of the live-fire generator set on the node k at the moment t under the typical scene s of the prediction time interval n;

representing the starting capacity of the live working motor group of the node k at the moment t under a typical scene s in a prediction period n;

representing the unit operation cost of the energy storage equipment, wherein ES represents the type of the energy storage equipment, including pumped storage, battery storage and compressed air storage;

a charging power representing stored energy;

a discharge power representing stored energy;

a unit offset cost representing a demand side response, wherein DR represents a type of demand side response, including a load shedding type and a load shifting type;

a response power representing a demand-side response;

represents the unit penalty cost of load shedding (in case of accident, disconnecting part of the load from the grid in order to maintain the power balance and stability of the power system);

representing the load shedding power; c. C^VREThe penalty cost of unit deficit amount compared with the target when the target of the renewable energy permeability is not finished is represented; c. C^CarbA penalty cost per deficit as compared to the target when the carbon emissions target is incomplete; q^VRERepresenting a corresponding renewable energy power shortage amount; q^CarbRepresenting the corresponding carbon emission abatement deficit amount.

The constraint conditions in the power grid evolution prediction algorithm are the construction constraint of the power grid and the operation cost constraint of the power grid, and the formulas (6) to (9):

the formula (6) restricts the investment capacity of each resource on each node in each stage; the equation (7) represents the relationship between the investment capacity and the total capacity; formula (8) constrains the renewable energy permeability target that needs to be reached at the final stage of evolution; equation (9) constrains the upper carbon emission target at the final stage of evolution.

Wherein the content of the first and second substances,

representing the upper limit of the investment capacity of the resource X on the k nodes in the n stages;

representing the generated power of the wind power in the last stage N;

representing the generated power of the photovoltaic in the last phase N;

representing the photo-thermal generated power in the last phase N; ζ represents a renewable energy source electrical capacity permeability target; d_k,n＝N,t,sRepresenting the load power; e.g. of the type^GRepresenting the carbon emission corresponding to the power generation of a thermal power unit; v^CarbRepresents a target value for the upper carbon emission limit.

The embodiment of the application provides a method for processing a key index sample set based on a power grid evolution prediction algorithm so as to generate a large amount of evolution data. Specifically, for each group of samples in the key index sample set, if the sample meets an objective function (the sum of the investment cost of each stage of power grid evolution, the fixed operation and maintenance cost of the power grid and the variable operation cost of the power grid is optimal) and a constraint condition (the construction constraint of the power grid evolution and the operation cost constraint of the power grid), a group of evolution data is output according to the sample. As can be seen, the embodiment of the present application performs processing based on a power grid evolution prediction algorithm on the key index sample set, and generates a large amount of evolution data. Compared with the prior art that several power grid evolution development paths are predicted manually, the method and the device for predicting the power grid evolution development paths automatically generate massive evolution paths based on reliable specific data and by means of a power grid evolution prediction algorithm, can eliminate the influence of human factors on power grid evolution prediction, and improve the accuracy of power grid evolution path prediction.

In the method provided by the embodiment of the application, evolution data obtained by automatic prediction can be analyzed, some key driving factors influencing power grid evolution are determined, the key driving factors are used for predicting the power grid evolution data, and a guiding effect is provided for the subsequent power grid actual evolution. Specifically, the method provided in the embodiment of the present application further includes the steps shown in fig. 3:

step 301, calculating the difference degree of the index in different evolution data aiming at any index in the key index set, and taking the maximum difference degree as the influence factor of the index.

In order to determine some key indexes (key driving factors) which have a large influence on the power grid evolution development in the key index set, a specific numerical value needs to be used for representing the influence degree of the key indexes on the power grid evolution.

In the concrete implementation, the difference degree of the key index in different evolution data is calculated, the maximum difference degree is used as the influence factor of the key index, and the influence degree of the key index on the power grid evolution is represented.

Calculating the difference degree of the key indexes in different evolution data, wherein the specific process is as follows (10) - (13):

y_i＝[y_i,1,y_i,2,...,y_i,N] (10)

the formula (10) represents a time series composed of multiple-stage values; equation (11) represents the mean of each key index in the corresponding subspace; equation (12) represents normalizing the time series of key indicators; equation (13) represents the average distance between different clusters of the normalized time series.

Wherein, y_iRepresenting the ith key index; y is_i,1,y_i,2,...,y_i,NThe value of the ith key index in the prediction time interval 1 to the prediction time interval N is represented; r represents the number of the massive path clusters;

representing key indexes corresponding to the generalization paths belonging to the r-th class; Ψ_iRepresenting the whole value space of the key index;

representing the value space of the key indexes in a cluster;

representing a normalized time series; mu.s_i,nRepresenting the mean value of the key index in the whole value space corresponding to the nth prediction time interval; sigma_i,nAnd expressing the standard deviation of the key index in the whole value space corresponding to the nth prediction time interval. Dist_iAnd the average distance of the key indexes among different clustering evolution paths is represented.

Finally, the calculated average distance is the difference degree of the key index in different evolution data, the maximum difference degree is taken as the influence factor of the key index,

step 302, determining one or more indexes with influence factors larger than a preset threshold in the key index set as key driving factors of power grid evolution of the target area.

In order to determine one or more key driving factors which have a large influence on the evolution development of the power grid from the key index set, a threshold needs to be preset for the influence factor corresponding to each key index, and the corresponding key index of which the influence factor is larger than the preset threshold is determined as a key driving factor.

And step 303, predicting subsequently generated evolution data according to the key driving factors.

And determining key driving factors which have a large influence on the power grid evolution development, and when power grid evolution data of the target area are generated subsequently, predicting the power grid evolution data by adjusting the key driving factors.

The embodiment of the application provides a method for determining some key indexes (key driving factors) which have large influence on power grid evolution development in a key index set. Specifically, a time sequence formed by multi-stage values of any one key index is determined, then the generated massive paths are clustered, a value space of the key index in the same cluster is determined, the mean value of the key index in the value space is calculated, the time sequence of the key index is standardized based on the mean value, the distance between different clusters of the standardized parameters is calculated and used as the difference degree of the key index in different evolution data, the difference degrees are sorted, and the maximum difference degree is used as an influence factor of the index; obtaining an influence factor of each key index in a key index set, comparing the influence factor with a preset threshold, and determining a key index corresponding to the influence factor larger than the preset threshold as a key driving factor of target regional power grid evolution; in the subsequent power grid evolution development process, the subsequent power grid evolution path (evolution data) can be adjusted by influencing the key driving factors.

In the method provided by the embodiment of the present application, the method may further calculate a flexible resource development index of the target area, including: the method comprises the following steps of flexible resource capacity requirement, flexible resource marginal capacity requirement, battery energy storage appearance point and battery energy storage leading point. Adjusting multiple sets of evolution data according to the flexible resource development index, and planning subsequently generated evolution data according to the development index;

in the embodiment of the application, in order to adjust the evolution data generated by the power grid evolution prediction algorithm, the distribution condition of the flexible resources and the battery energy storage in the power system needs to be obtained, and the flexible resource development index is determined. Specifically, the flexible resource capacity requirement, the flexible resource marginal capacity requirement, the battery energy storage occurrence point and the battery energy storage leading point of the target area are calculated according to a formula.

The flexible resource capacity requirement is the flexible resource capacity of the system under different renewable energy electric quantity permeability, namely the installed capacity (rated power of a generator set) which the flexible resource needs to reach under different renewable energy generating capacity targets; the flexibility resource marginal capacity requirement is that under different renewable energy electric quantity permeabilities, the permeability is increased by a flexibility resource capacity increment corresponding to a preset numerical value, namely, under different renewable energy electric quantity targets, the renewable energy electric quantity is increased by 1 percentage point, and the increment corresponding to the installed capacity of the flexibility resource is required; the battery energy storage appearance point is the corresponding permeability of the electric quantity of the renewable energy when the battery energy storage appears in the power system, and the battery energy storage main guide point is the corresponding permeability of the electric quantity of the renewable energy when the battery energy storage capacity accounts for the preset proportion of the capacity of the active resource. The predetermined ratio may be 50%.

Calculation of flexible resource capacity, as in equation (14):

wherein the content of the first and second substances,

representing flexible resource capacity; zeta₀Representing a certain level of renewable energy charge penetrationAnd (4) rate.

Calculating the flexibility resource marginal capacity as formula (15):

wherein the content of the first and second substances,

representing capacity increments of the flexible resource; Δ ζ represents the renewable energy permeability increase.

Calculating the battery energy storage appearance point, as formula (16):

calculating the main energy storage guide point of the battery according to the formula (17):

wherein

Representing the total capacity of the battery to store energy.

The embodiment of the application provides a method for calculating a flexible resource development index of a target area. Specifically, the flexibility resource capacity requirement, the flexibility resource marginal capacity requirement, the battery energy storage occurrence point and the battery energy storage leading point of the target area are calculated according to a formula, the flexibility resource is adjusted according to a calculation result so as to achieve a preset renewable energy electric quantity permeability target, the distribution condition (occurrence frequency) of the battery energy storage resource in the power system at present can also be judged according to the renewable energy electric quantity permeability, and the flexibility of the power system is improved; when the permeability of the electric quantity of the renewable energy source in the electric power system is abnormally changed, the permeability of the electric quantity of the renewable energy source can be interfered by adjusting the flexible resource, and the reliability of the electric power system is improved to a certain extent.

The embodiment of the invention takes northwest regions (including Shaanxi, Gansu, Qinghai, Ningxia and Xinjiang, for example) as examples, and the power grid evolution prediction method provided by the invention is explained in detail. Specifically, consider the 2020 to 2050 power grid evolution data. The current installation and load conditions of each province of the northwest power grid are shown in fig. 4, and the proportion of wind power and photovoltaic installation is 38.8%. Assuming that the annual growth rate of the local load is 3%, the outgoing load capacity remains unchanged.

Applied to the specific scene, the power grid evolution prediction method provided by the embodiment of the application comprises the following steps:

and S1, determining the value ranges of a plurality of uncertain factors in each prediction stage according to historical documents, and finishing assignment of the uncertain factors.

The uncertainty factor may also be referred to as a key index, and a plurality of uncertainty factors form the aforementioned key index set.

And S2, dividing by taking 5 years as a prediction time period, and setting the renewable energy permeability target of the evolution final state (2050) as 80%.

And S3, sampling the assigned uncertainty factors for 2500 times, generating evolution data under the three thermal power flexibility transformation paths shown in the figure 4, and generating 7500 evolution data. It should be noted that the setting of the renewable energy permeability target defines the power generation space of the conventional power supply, and the carbon emission amount of the conventional power supply is relatively determined, that is, the renewable energy permeability target and the carbon emission target have a certain degree of correlation, so the carbon emission target is not set separately in this chapter.

The evolution path of carbon emission of the system is shown in fig. 7(d), and annual carbon emission shows a trend of first reaching a peak and then descending. From the average path, the annual carbon emission will still increase by about 12%, and the final annual carbon emission will decrease by 57% compared with 2020 under the target of 80% renewable energy permeability. The carbon emission rises in the early stage of evolution because the reduction of carbon emission caused by the increase of renewable energy in the early stage cannot offset the increase of carbon emission caused by the increase of load and the increase of thermal power generation amount.

Fig. 6 shows the demand situation of the flexible resources (flexible resource capacity, flexible resource boundary capacity) under the massive evolution path under different renewable energy electric quantity permeabilities, and the evolution data participating in statistics are all generated under the thermal power deep reconstruction path. With the increase of the electric quantity permeability of the renewable energy, not only the demand of the flexible resource capacity is increased, but also the demand of the marginal capacity of the flexible resource is on the rising trend. The flexibility capacity required for a permeability of between 70-80% is about 23 times the permeability of between 10-20%, with a corresponding increase of about 11 times for marginal flexibility requirements. Therefore, when the permeability of renewable energy is higher, the resource demand for further improving the permeability is more, and the position of flexible resources in the system will become more important.

Fig. 7 shows the statistical situation of the battery energy storage in the massive evolution path. It can be seen that battery energy storage is probably intervened at the earliest stage of low-proportion development of renewable energy sources, and a system is most probably intervened when the electric quantity permeability of the renewable energy sources is between 40 and 45 percent; meanwhile, the renewable energy source is most likely to become a dominant flexible resource when the permeability of the renewable energy source is between 60 and 65 percent.

As mentioned earlier, the evolution of the power grid is influenced by different factors, such as technology, market, public, etc. The identification process shown in fig. 5 is shown by taking the identification of the key driving factors of the renewable energy permeability evolution in the technical level as an example, and then the key driving factors influencing various aspects of the power grid evolution are analyzed.

For renewable energy permeability evolution paths, the generated large number of paths are first clustered, here into 4 classes, as shown in fig. 8. These 4 classes correspond to 4 modes of permeability evolution: a fast growth mode, a slow growth mode, a front-fast-back slow growth mode, a front-slow-back fast growth mode. The key indicators corresponding to each type of evolution data are averaged and normalized, as shown in fig. 9. It can be seen that the unit investment cost, the battery energy storage cost and the coal price curve of renewable energy are greatly different among different categories, which indicates that the permeability of renewable energy is more sensitive to the indexes. The average distance of each index curve is calculated and ranked, and the result is shown in fig. 10. It can be seen that the significant factors are coal price, unit investment cost of wind power, maximum investable capacity of battery energy storage at each stage, unit investment cost of photovoltaic, and unit investment cost of battery energy storage in turn, which are consistent with the results in fig. 9.

Based on the above process, the key indexes affecting the evolution paths in different aspects are further analyzed, and the calculated values of the average distance indexes are listed in table 3. In each aspect of the evolution, the main influencing factors are basically the same, including the unit investment cost of renewable energy sources and energy storage, the coal price and the maximum investable capacity of battery energy storage in each prediction period. The relative importance of these factors is somewhat different in every respect. In the technical aspect, the maximum investable capacity of the battery energy storage in each stage is the most main influence factor, and the configuration of the battery energy storage is the economic choice, and the configuration capacity of the battery energy storage is closely related to the maximum configurable upper limit of the battery energy storage. The most important factor in the market and public areas is coal price.

TABLE 3

From the above calculation process, the annual carbon emission will still increase by about 12%, and the annual carbon emission in 2050 will decrease by 57% compared with 2020. The flexibility requirement and the marginal flexibility requirement of the system increase along with the increase of the permeability of the renewable energy source, wherein the battery energy storage is most likely to intervene in the system when the permeability of the electricity quantity of the renewable energy source is between 40 and 45 percent, and becomes a dominant flexibility resource when the permeability is between 60 and 65 percent. The main driving factors influencing the power grid evolution include the unit investment cost of renewable energy and stored energy, the coal price and the maximum investable capacity of battery stored energy at each stage. The method has clear calculation thought and good universality, and is suitable for popularization and application.

Fig. 13 is a frame diagram of generation and analysis of a mass evolution path according to the present application. Specifically, firstly, determining uncertainty factors (key index set) influencing the evolution development of a target regional power grid, then dividing the uncertainty factors into a plurality of prediction periods, determining an upper limit value and a lower limit value of each prediction period, and completing the quantification of the uncertainty factors; randomly sampling each uncertainty factor in an upper limit value interval and a lower limit value interval in each prediction period to obtain a key index sample, inputting the key index sample to a evolvement path generation module (power grid evolution prediction algorithm), and correspondingly generating a group of evolution data if the key index sample meets a target function and a constraint condition; finally, analyzing a large amount of generated evolution data to obtain key driving factors which have great influence on the evolution development of the power grid; the method can be realized, and can also calculate the flexible resource development index for adjusting the generated power grid evolution data.

It should be understood that although the various steps in the flow charts of fig. 1-3 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 1-3 may include multiple steps or phases, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps or phases.

In one embodiment, as shown in fig. 14, there is provided a power grid evolution prediction apparatus, including:

a determining module 1401, configured to determine a set of key indicators corresponding to a target region; the key index set comprises a power generation resource index influencing the development of a target area power grid, an energy storage index influencing the development of the target area power grid, an electricity demand side index influencing the development of the target area power grid, a flexibility resource index and a fuel price index influencing the development of the target area power grid;

a numerical processing module 1402, configured to perform numerical prediction processing on each index in the key index set to obtain a key index sample set;

an evolution data generation module 1403, configured to process the key index sample set by using a power grid evolution prediction algorithm to obtain multiple sets of evolution data corresponding to the target region; the evolution data comprises evolution data of the power generation resource structure corresponding to the target area and evolution data of the flexible resource structure corresponding to the target area.

In one embodiment, the numerical processing module 1402 is specifically configured to perform numerical prediction processing on each index in the set of key indexes to obtain a set of key index samples, and includes:

aiming at each index in the key index set, determining an upper limit value and a lower limit value of each index in a prediction time period, and performing assignment processing on each index in the key index set according to the upper limit value and the lower limit value of each index;

and randomly sampling the key index set subjected to assignment processing by adopting a Monte Carlo method to obtain the value of each index in each prediction time period, and determining a key index sample set according to the value of each index in each prediction time period.

In one embodiment, the evolution data generating module 1403 is specifically configured to input the key index sample set into an objective function in the power grid evolution prediction algorithm, and perform constraint processing on the key index sample set by using a constraint condition in the power grid evolution prediction algorithm;

determining multiple sets of evolution data corresponding to the target area according to the output of the target function and the result of the constraint processing;

wherein the objective function is: the sum of the investment cost of power grid evolution in each prediction period, the fixed operation and maintenance cost of the power grid and the variable operation cost of the power grid is optimal; the constraint conditions include: construction constraints of the grid and operating cost constraints of the grid.

In one embodiment, the investment cost is the capital cost invested in building power generation resources and flexibility resources in the power system; the fixed operation and maintenance cost of the power grid is the capital cost invested in daily maintenance of equipment built in the power system; the variable operating cost of the grid is the cost of fuel consumed by the power generation resources in the power system when operating.

In one embodiment, the construction constraints of the power grid include at least one of:

investment capacity constraint, total investment capacity constraint, renewable energy permeability constraint of the final stage of power grid evolution and carbon emission constraint of the final stage of power grid evolution of various resources on each power grid node in each prediction period;

the operating cost constraint of the power grid is the capital cost required to maintain the normal operation of various power generation resources in the power system.

In one embodiment, as shown in fig. 15, the grid evolution prediction device further comprises a grid evolution data analysis module 1404.

The power grid evolution data analysis module is specifically used for calculating the difference degree of the key indexes in different evolution data aiming at any one key index in the key index set, and taking the maximum difference degree as the influence factor of the key indexes;

determining one or more key indexes with influence factors larger than a preset threshold in the key index set as key driving factors of power grid evolution of a target region; the key driving factors are used for predicting power grid evolution data.

In one embodiment, as shown in fig. 16, the grid evolution prediction apparatus further comprises a grid evolution data adjustment module 1405.

The power grid evolution data adjusting module is specifically used for calculating a flexible resource development index of a target area, wherein the flexible resource development index comprises a flexible resource capacity requirement, a flexible resource marginal capacity requirement, a battery energy storage occurrence point and a battery energy storage main guide point;

adjusting multiple sets of evolution data according to the flexible resource development index, and planning subsequently generated evolution data according to the development index;

the flexibility resource capacity demand is the flexibility active resource capacity of the system under different renewable energy electric quantity permeabilities, the flexibility resource marginal capacity demand is the flexibility resource capacity increment corresponding to a preset numerical value of permeability increase under different renewable energy electric quantity permeabilities, the battery energy storage appearance point is the renewable energy electric quantity permeability corresponding to the battery energy storage in the electric power system, and the battery energy storage leading point is the renewable energy electric quantity permeability corresponding to the battery energy storage capacity accounting for the preset proportion of the flexibility resource capacity.

For specific limitations of the power grid evolution prediction device, reference may be made to the above limitations of the power grid evolution prediction method, which are not described herein again. All or part of the modules in the power grid evolution prediction device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 17. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing a set of key indicators, a set of key indicator samples and evolution data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement the power grid evolution prediction method according to the embodiment of the present application.

Those skilled in the art will appreciate that the architecture shown in fig. 17 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

determining a key index set corresponding to a target area; the key index set comprises a power generation resource index influencing the development of a target regional power grid, an energy storage index influencing the development of the target regional power grid, a power demand side index influencing the development of the target regional power grid, a flexibility resource index and a fuel price index influencing the development of the target regional power grid;

carrying out numerical value prediction processing on each index in the key index set to obtain a key index sample set;

processing the key index sample set by using a power grid evolution prediction algorithm to obtain multiple groups of evolution data corresponding to the target area; the evolution data comprises evolution data of a power generation resource structure corresponding to the target area and evolution data of a flexible resource structure corresponding to the target area.

In one embodiment, the processor, when executing the computer program, implements:

inputting the key index sample set into a target function in a power grid evolution prediction algorithm, and performing constraint processing on the key index sample set by using constraint conditions in the power grid evolution prediction algorithm;

calculating the difference degree of the key indexes in different evolution data aiming at any one key index in the key index set, and taking the maximum difference degree as an influence factor of the key indexes;

calculating a flexible resource development index of the target area, wherein the flexible resource development index comprises a flexible resource capacity requirement, a flexible resource marginal capacity requirement, a battery energy storage occurrence point and a battery energy storage leading point;

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. 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), for example.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent application shall be subject to the appended claims.

Claims

1. A power grid evolution prediction method, characterized in that the method comprises:

determining a key index set corresponding to a target area; the key index set comprises a power generation resource index influencing the development of the target regional power grid, an energy storage index influencing the development of the target regional power grid, a power demand side index influencing the development of the target regional power grid, a flexibility resource index and a fuel price index influencing the development of the target regional power grid;

performing numerical prediction processing on each index in the key index set to obtain a key index sample set;

processing the key index sample set by using a power grid evolution prediction algorithm to obtain multiple sets of evolution data corresponding to the target area; the evolution data comprises evolution data of a power generation resource structure corresponding to the target area and evolution data of a flexible resource structure corresponding to the target area.

2. The method according to claim 1, wherein the performing a numerical prediction process on each index in the set of key indexes to obtain a sample set of key indexes comprises:

for each index in the key index set, determining an upper limit value and a lower limit value of each index in a prediction time period, and performing assignment processing on each index in the key index set according to the upper limit value and the lower limit value of each index;

and randomly sampling the key index set subjected to the assignment processing by adopting a Monte Carlo method to obtain the value of each index in each prediction time period, and determining the key index sample set according to the value of each index in each prediction time period.

3. The method according to claim 1, wherein the processing the set of key indicator samples by using a power grid evolution prediction algorithm to obtain multiple sets of evolution data corresponding to the target region comprises:

inputting the key index sample set into an objective function in the power grid evolution prediction algorithm, and performing constraint processing on the key index sample set by using constraint conditions in the power grid evolution prediction algorithm;

determining multiple sets of evolution data corresponding to the target region according to the output of the target function and the result of constraint processing;

wherein the objective function is: the sum of the investment cost of power grid evolution in each prediction period, the fixed operation and maintenance cost of the power grid and the variable operation cost of the power grid is optimal; the constraint conditions include: construction constraints of the power grid and operating cost constraints of the power grid.

4. The method of claim 3, wherein the investment cost is a capital cost invested in building power generation resources and flexibility resources in the power system; the fixed operation and maintenance cost of the power grid is the capital cost invested in daily maintenance of equipment built in the power system; the grid variable operating cost is the cost of fuel consumed by the operation of the power generation resources in the power system.

5. The method of claim 3, wherein the construction constraints of the power grid include at least one of:

6. The method according to claim 1, characterized in that it comprises:

determining one or more key indexes with influence factors larger than a preset threshold in the key index set as key driving factors of the power grid evolution of the target region; the key driving factors are used for predicting power grid evolution data.

7. The method of claim 1, further comprising:

the flexibility resource capacity demand is the flexibility resource capacity of the system under different renewable energy electric quantity permeabilities, the flexibility resource marginal capacity demand is the flexibility resource capacity increment corresponding to a preset numerical value of permeability increase under different renewable energy electric quantity permeabilities, the battery energy storage appearance point is the renewable energy electric quantity permeability corresponding to the battery energy storage in the electric power system, and the battery energy storage main guide point is the renewable energy electric quantity permeability corresponding to the battery energy storage capacity accounting for the preset proportion of the flexibility resource capacity.

8. A power grid evolution prediction apparatus, characterized in that the apparatus comprises:

the determining module is used for determining a key index set corresponding to the target area; the key index set comprises a power generation resource index influencing the development of the target regional power grid, an energy storage index influencing the development of the target regional power grid, a power demand side index influencing the development of the target regional power grid, a flexibility resource index and a fuel price index influencing the development of the target regional power grid;

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.