A grid adaptive planning method considering the non-line substitution effect of virtual energy storage

Abstract

The present invention relates to a grid adaptive planning method that takes into account the non-line substitution effect of virtual energy storage, and belongs to the field of power planning. The method of the present invention regards the candidate cloud computing data center to be built as the virtual energy storage of the power grid, and constructs a virtual energy storage model based on the power consumption of the data center, the time domain and spatial migration of the computing load; constructs a construction cost model based on the candidate transmission lines, energy storage equipment and data center; constructs an operating cost model based on various energy outputs, loads and virtual energy storage of the power grid; and solves the construction plan of the transmission line, energy storage equipment and data center and the operation plan of the data center with the goal of minimizing the total cost of construction cost and operating cost. The method of the present invention obtains a grid planning that adapts to the allocation of multiple types of flexible resources by utilizing the time and space adjustable flexibility of the computing load energy consumption of the data center, which can reduce the line planning cost in the grid planning and improve the efficiency of new energy transmission.

Description

Technical Field

[0001] The present invention belongs to the field of electric power planning, and in particular relates to a power grid adaptive planning method taking into account the non-line substitution effect of virtual energy storage. Background Art

[0002] Solving the problem of imbalance between regional resource reserves and regional development has always been the focus and difficulty of the power industry. A widely supported and rapidly implemented strategy is to transmit excess electricity from resource-rich areas to resource-scarce load center areas through inter-regional transmission lines. This inter-regional transmission strategy effectively solves the urgent problem of long-term power shortage in some areas when power demand exceeds supply.

[0003] However, as economic development enters a new normal, electricity demand has gradually stabilized, and the electricity supply and demand relationship in various regions has gradually been balanced. At the same time, the proportion of renewable energy power generation in the power structure continues to increase, but the utilization efficiency of ultra-high voltage transmission lines is not satisfactory. Renewable energy power generation centers are usually located in areas far away from load centers, and the capacity of cross-regional power grid transmission is affected by many factors, including transmission distance, transmission efficiency, and load matching. These factors have led to a high level of wind and solar abandonment in some areas, and load centers often face supply and demand imbalances such as tight electricity consumption during peak hours. Long-distance load transfer is difficult to achieve, and high-voltage transmission lines are difficult to meet the instantaneous increase in electricity demand. Therefore, the traditional methods of solving the problems of transmission capacity limitations between regional power grids and insufficient power supply capacity within the power supply area have technical limitations and lack economic benefits.

[0004] In recent years, cloud computing technology has become a key technology for big data processing and storage. As the infrastructure of cloud computing technology, data centers are closely related to both digitalization and informatization, and the development of traditional industries. In particular, they are more closely related to energy-related fields. Data centers can migrate loads between data centers in different geographical locations through optical fibers, or migrate loads in different time periods through cloud computing task sequencing, thereby achieving real-time regulation of their power loads, allowing them to participate in the operation of the power system as a virtual energy storage with huge flexibility potential. How to rationally plan and build data centers and scientifically formulate data center operation plans are issues that need to be studied. Summary of the invention

[0005] In view of the above analysis, the present invention aims to provide a grid adaptive planning method taking into account the non-line substitution effect of virtual energy storage, and to establish a collaborative planning and operation model for multi-resource configuration, so as to reduce the construction redundancy of the grid for transmission pipelines and optimize the grid load distribution.

[0006] The method of the present invention specifically comprises the following steps:

[0007] The candidate cloud computing data center is regarded as the virtual energy storage of the power grid, and a virtual energy storage model is constructed based on the power consumption of the data center and the migration and exit of computing load in the time domain and space;

[0008] Constructing a construction cost model based on the candidate transmission lines, energy storage equipment and the data center;

[0009] An operation cost model is constructed based on the output of thermal power units in the power grid on each typical day, the operating status of thermal power units, the carbon emissions of thermal power units, the system load shedding, the new energy generation and actual new energy output in each typical day's uncertain scenario, and the charging power and discharging power of virtual energy storage;

[0010] The objective function is constructed with the goal of minimizing the total cost of construction cost and operation cost;

[0011] Based on the virtual energy storage model, equipment investment and construction constraints, system node balance constraints, line transmission capacity constraints, thermal power unit operation constraints, energy storage operation constraints, and new energy power generation constraints, the objective function is solved to obtain the investment and construction plan of the transmission line, energy storage equipment and the data center and the operation plan of the data center.

[0012] Furthermore, the construction of a virtual energy storage model based on the power consumption of the cloud computing data center and the migration of computing loads in and out in time and space includes: considering the power consumption of the data center as the capacity of the virtual energy storage, considering the migration of computing loads in the data center in time and space as charging of the virtual energy storage, and considering the migration of computing loads in time and space as discharging of the virtual energy storage, to construct the virtual energy storage model.

[0013] Furthermore, the virtual energy storage model is expressed as:

[0014]

[0015] in, and They represent the capacity of virtual energy storage at time t and time t-1 in a typical day r, i.e., the power consumption of the data center, in and Respectively represent the lower / upper limit of virtual energy storage capacity, represents the full load power of the data center server; M m,r,max Indicates the total number of servers in the data center;

[0016] and They represent the charging / discharging power of the virtual energy storage at time t in a typical day r respectively;

[0017] α VES represents the capacity factor of virtual energy storage;

[0018] β VES and κ VES They represent the virtual energy storage charging / discharging power coefficient, and Respectively represent the change amount of the delay-sensitive computing load and the delay-tolerant computing load of the data center after migration;

[0019] An energy conversion correction amount representing virtual energy storage;

[0020] and Respectively represent the charging / discharging state of the virtual energy storage,

[0021] and They represent the lower / upper limit of the virtual energy storage charging power, Indicates the number of servers required by the data center to process the load before computing the load migration;

[0022] and They represent the lower / upper limit of the virtual energy storage discharge power,

[0023] Indicates the idle power of the data center server;

[0024] Indicates the change in the number of servers required after the computing load of the data center is migrated;

[0025] and respectively represent the number of delay-sensitive computing loads and delay-tolerant computing loads of the data center before migration;

[0026] represents the floor function;

[0027] δ(*) represents Kronecker-δ function;

[0028] u m Indicates the efficiency of processing computing load per server in the data center;

[0029] H(*) represents the Heaviside function.

[0030] Furthermore, the construction cost model based on the candidate transmission lines, energy storage equipment and data center is expressed as:

[0031]

[0032] Among them, C Inv represents the construction cost; the subscripts l, m, and b represent the candidate transmission lines, energy storage equipment, and virtual energy storage, respectively; L cand , B, and VES are the collections of candidate transmission lines, energy storage equipment, and virtual energy storage, respectively; are 0 / 1 variables, representing the planning and construction status of transmission lines, energy storage equipment, and virtual energy storage; T k represents the life of the candidate transmission lines, energy storage equipment and virtual energy storage; s is the discount rate.

[0033] Furthermore, the operating cost model is expressed as:

[0034]

[0035] Among them, C Ope represents the operating cost; U is the uncertainty set in the operating cost model; Ξ is a 0 / 1 variable set; Ψ is a continuous variable set; v is a 0 / 1 variable, including the charging and discharging status of the energy storage device, whether there is an error in the new energy, and the operating status of the thermal power unit; p is a continuous variable, including the virtual energy storage operating variable, the energy storage device operating variable, and the thermal power unit operating variable;

[0036] Subscripts g, n, and d represent thermal power units, new energy equipment, and loads, respectively; c g P is the power generation cost coefficient of thermal power units; g,t,r Output for thermal power units; c e is the carbon emission cost coefficient; E g,t,r is the carbon emission of thermal power units; Pr VES is the cost coefficient of charging and discharging of virtual energy storage; and Respectively represent the charging / discharging power of virtual energy storage; c cur is the penalty coefficient for abandoning wind / solar power; and P n,t,r are the renewable energy generation and actual renewable energy output under uncertain scenarios; c LS is the load shedding penalty coefficient; ΔP d,t,r The system load shedding amount for the load node.

[0037] Furthermore, the equipment investment and construction constraints include:

[0038]

[0039] Among them, max Indicates the upper limit of construction cost; are the upper limit of the number of candidate transmission lines, virtual energy storage and energy storage equipment; P b ess The configured power of the energy storage device; is the configuration capacity of the energy storage device; τ ess is the continuous discharge time of the energy storage device.

[0040] Furthermore, the system node balance constraints include:

[0041]

[0042] Among them, G(s), N(s), B(s), L cand (s), D(s), and VES(s) represent the node set of thermal power units, the node set of new energy sources, the node set of energy storage devices, the set of transmission lines between nodes, the set of load nodes, and the set of virtual energy storage nodes, respectively; and are the discharge and charging power of the energy storage device respectively; P l,t,r is the transmission power of the transmission line; P d,t,r is the load size of the load node; ε PUE Energy utilization efficiency of virtual energy storage.

[0043] Furthermore, the line transmission capacity constraint is expressed as:

[0044]

[0045] Among them, i and j are nodes of the power grid system; θ i,t.r and θ j,t,r Respectively represent the voltage phase angles of nodes i and j; and Respectively represent the upper and lower limits of the voltage phase angle of node j; l is the reactance of the line; P l max is the maximum transmission power of the line; M big is a very large positive number.

[0046] Furthermore, the renewable energy generation constraint is expressed as:

[0047]

[0048] Among them, p n,t,r is the per unit output value of renewable energy at time t in a typical day r; is the per unit value of power generation of renewable energy at time t in a typical day r; and Whether there is a new energy output prediction error, which is a 0 / 1 variable; and New energy output prediction error; Γ T budget uncertainty for time; It is the electric field installed capacity of the new energy node.

[0049] Furthermore, a two-stage robust optimization method is used to find the optimal solution to the objective function, thereby obtaining candidate transmission lines, energy storage equipment, and the construction plan of the virtual energy storage, as well as the operation plan of the virtual energy storage.

[0050] The present invention can achieve at least one of the following beneficial effects:

[0051] By considering the migration of computing loads in cloud computing data centers in time and space as the charging behavior of virtual energy storage, and the migration of computing loads in time and space as the discharging behavior of virtual energy storage, a virtual energy storage model is constructed. While ensuring data integrity and sufficient computing resources, online loads can be processed remotely and data center power loads can be transferred between regions. This breaks through the physical space limitations and technical limitations of traditional methods for solving power supply congestion problems and solves the problem of temporary power shortages in some areas.

[0052] Aiming at the scenario where the pressure of high proportion of new energy transmission surges and the transmission line is seriously blocked, and considering the non-line substitution effect shown by the spatiotemporal adjustable flexibility of the computing load energy consumption of the data center, a power grid planning and operation model that adapts to the allocation of multiple types of flexible resources is proposed. The investment and construction costs of transmission lines, energy storage equipment and data centers, as well as the grid operation costs including thermal power output, new energy output and virtual energy storage are considered, and the objective function is constructed. The investment and construction plans of the transmission lines, energy storage equipment and the data center that achieve the minimum total cost, as well as the operation plans of each energy source and the data center are solved, so as to reduce the redundancy of the grid's investment and construction of transmission pipelines, optimize the grid load distribution, and further improve the operation and adjustment flexibility on the grid demand side, thereby changing the real-time supply and demand balance of electricity in a larger range.

[0053] By taking into account the uncertainty of wind power / photovoltaic output, the robust optimization method is used to deal with the uncertainty in the system. The collaborative planning model is divided into a two-stage robust optimization model, in which the first stage is the construction cost model and the second stage is the operation problem. The construction decision results of the first stage are used to better formulate the decision strategy of the second stage to deal with the impact of uncertainty in the system. This method can reduce the conservatism of the model solution and improve the robustness.

[0054] Other features and advantages of the present invention will be described in the following description, and some advantages may become apparent from the description, or may be understood by practicing the present invention. The purpose and other advantages of the present invention may be realized and obtained through the contents particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings are only used for the purpose of illustrating specific embodiments and are not to be considered as limiting the present invention. In the entire drawings, the same reference symbols represent the same components;

[0056] Figure 1 It is a schematic diagram of the steps of the method of the present invention;

[0057] Figure 2 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0058] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0059] A specific embodiment of the present invention discloses a grid adaptability planning method taking into account the non-line substitution effect of virtual energy storage, which specifically includes the following steps:

[0060] Step S01: Consider the candidate cloud computing data center to be built as the virtual energy storage of the power grid, and build a virtual energy storage model based on the power consumption of the data center and the migration in and out of the computing load in the time domain and space, including: considering the power consumption of the data center as the capacity of the virtual energy storage, considering the migration in of the computing load in the time domain and space of the data center as the charging of the virtual energy storage, and considering the migration out of the computing load in the time domain and space as the discharge of the virtual energy storage, so as to build the virtual energy storage model;

[0061] Step S02: constructing a construction cost model based on the candidate transmission lines, energy storage equipment and the data center;

[0062] Grid adaptability planning includes planning and decision-making for candidate transmission lines, energy storage equipment and data centers based on the existing thermal power units, new energy units (including wind turbines and photovoltaic units), existing energy storage equipment and transmission lines in the grid;

[0063] Step S03, constructing an operation cost model based on the output of thermal power units in the power grid on each typical day, the operating status of thermal power units, the carbon emissions of thermal power units, the load shedding of the system, the new energy power generation and actual new energy output of each typical day's uncertain scenario, the charging power and the discharging power of the virtual energy storage; the typical day can be obtained through historical data, using the K-means clustering method to simulate the operation of the power grid system;

[0064] Step S04, constructing an objective function with the goal of minimizing the total cost of construction cost and operation cost;

[0065] Step S05, solving the objective function based on the virtual energy storage model, equipment construction constraints, system node balance constraints, line transmission capacity constraints, thermal power unit operation constraints, energy storage operation constraints, and new energy power generation constraints to obtain the construction plan of the transmission line, energy storage equipment and the data center and the operation plan of the data center.

[0066] In this embodiment, by constructing a virtual energy storage model, while ensuring data completeness and sufficient computing resources, the remote processing of online loads is realized, and the power load of the data center is transferred between regions. It breaks through the physical space limitations and technical limitations of the traditional method of solving the power supply congestion problem, and solves the problem of temporary power shortage in some areas. By establishing a collaborative planning and operation model for multi-resource configuration, considering the investment and construction costs of transmission lines, energy storage equipment and data centers, as well as the grid operation costs including thermal power output, new energy output, and virtual energy storage, the total cost of the transmission lines, energy storage equipment and the data center investment and construction plan, as well as the operation output of each energy source and the operation plan of the data center are obtained, reducing the redundancy of the grid's investment and construction of transmission pipelines, optimizing the grid load distribution, and further improving the operation and adjustment flexibility on the grid demand side, thereby changing the real-time supply and demand balance of electricity in a larger range.

[0067] Specifically, in step S01, the virtual energy storage model is expressed as:

[0068]

[0069] in, and They represent the capacity of virtual energy storage at time t and time t-1 in a typical day r, i.e., the power consumption of the data center at the corresponding time, which are variables. It should be noted that the data center as virtual energy storage is not real energy storage, but a form of energy storage is used to represent the flexibility of the data center. The actual physical meaning of the energy storage capacity is the power consumption of the data center.

[0070] in and Respectively represent the lower / upper limit of virtual energy storage capacity, which are constants. represents the full load power of the data center server, a constant; M m,r,max Represents the total number of servers in the data center, which is a constant;

[0071] and They respectively represent the charging / discharging power of the virtual energy storage at time t in a typical day r, and actually represent the increase / decrease in the computing load that the data center needs to process after the computing load migration, both of which are variables; it should be noted that after the computing load migration refers to the time after the data center computing load energy consumption is used to respond to the power grid demand, and correspondingly, before the computing load migration refers to the time before the data center computing load energy consumption is used to respond to the power grid demand;

[0072] α VES represents the capacity factor of virtual energy storage, which is a constant;

[0073] β VES and κ VES They represent the virtual energy storage charging / discharging power coefficients, which are related to the data center calculation load migration state and are 0 / 1 variables. and Respectively represent the change amount of the delay-sensitive computing load and the delay-tolerant computing load of the data center after migration;

[0074] The energy conversion correction amount representing the virtual energy storage actually represents the capacity correction obtained by correcting the number of servers in the data center processing the computing load during the computing load migration process, which is a variable;

[0075] and They represent the charging / discharging status of the virtual energy storage, and actually represent whether the computing load that the data center needs to handle after the computing load migration is increased or decreased compared to before the migration. They are variables.

[0076] and They represent the lower / upper limit of the virtual energy storage charging power, which actually represent the lower / upper limit of the computing load increment that the data center needs to handle after the computing load migration, and are constants. Indicates the number of servers required by the data center to process the load before computing the load migration;

[0077] and They represent the lower / upper limit of the virtual energy storage discharge power, which actually represent the lower / upper limit of the computing load reduction that the data center needs to handle after the computing load migration. They are constants.

[0078] It represents the idle power of the data center server, which is a constant;

[0079] represents the change in the number of servers required after the computing load of the data center is migrated, which is a variable;

[0080] and respectively represent the number of the delay-sensitive computing load and the delay-tolerant computing load of the data center before migration, which are constants;

[0081] represents the floor function;

[0082] δ(*) represents the Kronecker-δ function, i.e., δ(x=0)=1, otherwise it is 0;

[0083] u m represents the computing load processing efficiency of the unit server of the data center, which is a constant;

[0084] H(*) represents the Heaviside function, which is a unit step function and takes the value of 0 or 1 depending on the conditions.

[0085] Specifically, in step S02, the construction cost model based on the candidate transmission lines, energy storage equipment and data center is expressed as:

[0086]

[0087] Among them, C Inv represents the construction cost; the subscripts l, m, and b represent the candidate transmission lines, energy storage equipment, and virtual energy storage, respectively; L cand , B, and VES are the candidate transmission line set, energy storage equipment set, and virtual energy storage set, respectively; is a 0 / 1 variable, and is a variable, which respectively represents the planned construction status of the transmission line, energy storage equipment, and virtual energy storage. 1 means planned construction, and 0 means no planned construction; T k It represents the life of the candidate transmission lines, energy storage equipment and virtual energy storage, which is a quantitative value; s is the discount rate, which is a quantitative value.

[0088] Specifically, in step S03, the operating cost model is expressed as:

[0089]

[0090] Among them, C Operepresents the operating cost; U is the uncertainty set in the operating cost model; Ξ is a 0 / 1 variable set; Ψ is a continuous variable set; v is a 0 / 1 variable, including the charging and discharging state of the energy storage device (exemplarily, 1 represents charging and 0 represents discharging), whether there is an error in the new energy (exemplarily, 1 represents an error and 0 represents no error), and the operating state of the thermal power unit (exemplarily, 1 represents operation and 0 represents shutdown); p is a continuous variable, including virtual energy storage operation variables, energy storage device operation variables and thermal power unit operation variables;

[0091] π r is the probability of a typical day r, quantitative; the calculation method is 1 / the total number of dates that cluster to obtain the typical day;

[0092] Subscripts g, n, and d represent thermal power units, new energy equipment, and loads, respectively; c g P is the power generation cost coefficient of thermal power units, quantitative; g,t,r is the output of thermal power unit, which is a variable; c e is the carbon emission cost coefficient, which is quantitative; E g,t,r is the carbon emission of thermal power units, which is a variable; Pr VES is the virtual energy storage charging and discharging cost coefficient, which is quantitative; and They represent the charging / discharging power of virtual energy storage, which are variables; c cur is the penalty coefficient for wind / solar abandonment, which is quantitative; and P n,t,r are the renewable energy power generation and actual renewable energy output under uncertain scenarios, respectively, which are variables; c LS is the load shedding penalty coefficient, which is quantitative; ΔP d,t,r is the system load shedding amount of the load node, which is a variable.

[0093] Specifically, in step S04, an objective function is constructed with the minimum total cost of construction cost and operation cost as the goal, which is expressed as: min C total =C Inv +C Ope .

[0094] Specifically, in step S05, the equipment investment constraint is expressed as:

[0095]

[0096] Among them, max It represents the upper limit of construction cost, which is quantitative; They are the upper limits of the number of transmission lines, virtual energy storage and energy storage equipment that are candidates for investment and construction, which are quantitative; is the configured power of the energy storage device, which is a variable; is the configuration capacity of the energy storage device, is a variable; τess It is the continuous discharge time of the energy storage device, which is quantitative.

[0097] Specifically, in step S05, the system node balance constraint is one of the key factors to ensure the stable operation of the power grid, which means that at each node of the power grid, the power entering the node must be equal to the power leaving the node plus the power consumed at the node, expressed as:

[0098]

[0099] Among them, G(s), N(s), B(s), L cand (s), D(s), and VES(s) represent the node set of thermal power units, the node set of new energy sources, the node set of energy storage devices, the set of transmission lines between nodes, the set of load nodes, and the set of virtual energy storage nodes, respectively; and are the discharge and charging power of the energy storage device, respectively, and are variables; P l,t,r is the transmission power of the transmission line, which is a variable; P d,t,r is the load size of the load node, which is quantitative; ε PUE It is the energy utilization efficiency of virtual energy storage, which represents the ratio of the total power consumption of the data center to the power consumption required for computing load, and is a variable.

[0100] Specifically, in step S05, the line transmission capacity constraint is expressed as:

[0101]

[0102] Among them, i and j are nodes of the power grid system; θ i,t.r and θ j,t,r Respectively represent the voltage phase angle of nodes i and j, variables; and They represent the upper and lower limits of the voltage phase angle of node j, which are quantitative; x l is the reactance of the line, which is quantitative; P l max is the maximum transmission power of the line, which is quantitative; M big is a very large positive number, which is a quantitative value.

[0103] Specifically, in step S05, the operation constraints of the thermal power unit are expressed as:

[0104]

[0105] Among them, P g max and P g min They represent the lower and upper limits of the unit output respectively, which are quantitative; u g,t,rIndicates the operating status of the unit, which is a 0 / 1 variable (1 means running, 0 means shut down); and Respectively represent the time that the unit has been continuously started and stopped, which are variables; and They represent the startup and shutdown time constraints of the unit, which are quantitative; and They respectively represent the up and down climbing limits of the unit and are quantitative.

[0106] Specifically, in step S05, the energy storage operation constraint is expressed as:

[0107]

[0108] in, and are 0 / 1 variables (1 for charging and 0 for discharging) of the energy storage device’s charging / discharging working state, respectively; and are the charging and discharging power of the energy storage device, respectively, and are variables; P b ess is the configuration power of the energy storage device, which is a variable; e b,t,r is the storage capacity of the energy storage device, is a variable; η ch and η dis are the charging / discharging efficiency of the energy storage device, which is quantitative; is the configured capacity of the energy storage device and is a variable.

[0109] Specifically, in step S05, the renewable energy generation constraint is expressed as:

[0110]

[0111] Among them, p n,t,r is the per unit output value of the renewable energy at time t in a typical day r, which is determined by the output curve of the typical day; is the power generation per unit value of renewable energy at time t in a typical day r, which is determined by the output curve of the typical day; and P n,t,r They are the renewable energy power generation and actual renewable energy output under uncertain scenarios, which are variables; and is whether there is a new energy output prediction error, which is a 0 / 1 variable, and is a variable; and The prediction error of renewable energy output is quantitative; Γ T For uncertain budget of time, for quantitative; is the electric field installed capacity of the new energy node, which is quantitative.

[0112] It should be noted that the output curve of a typical day is obtained by clustering historical data based on the K-means method, and p is determined based on the output curve. n,t,r and value.

[0113] Furthermore, in step S05, a two-stage robust optimization method is used to find the optimal solution for the objective function.

[0114] Specifically, a two-stage robust optimization model is constructed based on the objective function. The first stage is the construction cost model, and the second stage is the operation problem of a typical day. The specific matrix form expression is:

[0115]

[0116] where x and c represent the binary and continuous variable vectors in the first stage; v and p represent the binary and continuous variable vectors in the second stage problem; p u represents the uncertain variable vector; a, b, f, g, d, A, B, D, Q, F, M are the corresponding coefficient matrices.

[0117] Furthermore, for the two-stage robust optimization model, the column and constraint generation (C&CG) algorithm is used to solve it. The C&CG algorithm decomposes the original problem into the planning main problem and the operation sub-problem under uncertain scenarios and solves them iteratively.

[0118] Main question:

[0119] The goal of the main problem is to minimize the total cost. The constraints include equipment investment constraints, thermal power unit operation constraints, energy storage operation constraints, and C&CG constraints returned in the sub-problems.

[0120]

[0121] Among them, p z is the continuous variable vector of the second stage after the zth iteration; is the worst scenario in the second stage after the zth iteration, because the subproblem has been solved in the zth iteration. , so it is considered as a known parameter. is the cutting plane constraint and η is an auxiliary variable.

[0122] Sub-questions:

[0123] The second stage sub-problem is the system operation problem under the given planning and construction decision of the main problem. The sub-problem is a max-min problem and cannot be solved directly. Therefore, it is necessary to perform a dual transformation on the inner min problem in the sub-problem and convert it into a max problem for solution.

[0124]

[0125] Since the subproblem contains 0 / 1 variables, the subproblem is a mixed integer linear programming (MILP) problem and cannot be directly transformed into a dual. In order to solve this problem, a method of introducing auxiliary variables is proposed. First, 0-1 auxiliary variables are introduced into the subproblem. Define the original 0 / 1 variable v as a continuous variable and constrain v to be equal to The following formula is obtained:

[0126]

[0127] After introducing the auxiliary variables, the inner layer in formula (1) is transformed into a duality, and the following formula is obtained:

[0128]

[0129] According to the minimax inequality, we can get the following relationship:

[0130]

[0131] In the objective function of formula (2), The optimal value of can be determined by enumerating the possible combinations of 0 and 1.

[0132]

[0133] In formula (3), Ω = {δ≤λ,,δ≤0}. Through model transformation, the inner min problem can be eliminated, and then the outer max problem is merged with the inner max problem. The objective function of formula (2) can be converted to:

[0134]

[0135] Finally, the sub-problem objective function is converted to the formula (4), which contains bilinear terms. This embodiment uses the large M method for linearization.

[0136] Through the above processing, the sub-problem is converted into a solvable form, and the solution of the sub-problem considering the energy storage operation 0 / 1 variable is realized. Finally, by solving the sub-problem, the worst scenario p of the new energy output in the uncertain scenario is obtained. worst , by adding new variables, C&CG cut set constraints are generated and added to the main problem for iterative solution.

[0137] The cut set constraint relation is as follows:

[0138]

[0139] Among them, p k is the continuous variable vector of the second stage after the kth iteration; is the worst scenario in the second phase after the kth iteration.

[0140] Finally, the solution is obtained for the construction and investment plans of transmission lines, energy storage equipment and data centers, as well as the operation plans of various energy sources in the power grid, including the output of thermal power units at each time period t on each typical day, carbon emissions, the charging and discharging status and power of each virtual energy storage (i.e., the operation plan of the data center), the power generation of various renewable energy sources and the actual output of renewable energy sources, as well as the load of each load node.

[0141] The present embodiment discloses a grid adaptive planning method that takes into account the non-line substitution effect of virtual energy storage. By targeting scenarios where the pressure of high-proportion new energy transmission surges and transmission lines are severely blocked, and taking into account the non-line substitution effect exhibited by the spatiotemporal adjustable flexibility of data center computing load energy consumption, a grid planning and operation model that adapts to the allocation of multiple types of flexible resources is provided. The objective function is constructed by considering the investment and construction costs of transmission lines, energy storage equipment, and data centers, as well as the grid operation costs including thermal power output, new energy output, and virtual energy storage. The investment and construction plans of transmission lines, energy storage equipment, and the data centers that minimize the total cost, as well as the operation plans of the energy operation outputs and data centers, are solved to obtain the investment and construction plans of the transmission lines, energy storage equipment, and the data centers that minimize the total cost, as well as the operation plans of the energy operation outputs and the data centers, thereby reducing the redundancy of the grid's investment and construction of transmission pipelines, optimizing the grid load distribution, and further improving the operation and adjustment flexibility on the grid demand side, thereby changing the real-time supply and demand balance of electricity over a larger range. By taking into account the uncertainty of wind power / photovoltaic output, the robust optimization method is used to deal with the uncertainty in the system. The collaborative planning model is divided into a two-stage robust optimization model, in which the first stage is the construction cost model and the second stage is the operation problem. The construction decision results of the first stage are used to better formulate the decision strategy of the second stage to deal with the impact of uncertainty in the system. This method can reduce the conservatism of the model solution and improve the robustness.

[0142] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A grid adaptive planning method taking into account the non-line substitution effect of virtual energy storage, characterized in that: The steps include: The candidate cloud computing data center is regarded as the virtual energy storage of the power grid, and a virtual energy storage model is constructed based on the power consumption of the data center and the migration and exit of computing load in the time domain and space; the virtual energy storage model is expressed as: in, and They represent the capacity of virtual energy storage at time t and time t-1 in a typical day r, i.e., the power consumption of the data center, in and Respectively represent the lower / upper limit of virtual energy storage capacity, represents the full load power of the data center server; M m,r,max Indicates the total number of servers in the data center; and They represent the charging / discharging power of the virtual energy storage at time t in a typical day r; α VES Represents the capacity factor of virtual energy storage; β VES and κ VES They represent the virtual energy storage charging / discharging power coefficient, and Respectively represent the change amount of the delay-sensitive computing load and the delay-tolerant computing load of the data center after migration; An energy conversion correction amount representing virtual energy storage; and Respectively represent the charging / discharging state of the virtual energy storage, and They represent the lower / upper limit of the virtual energy storage charging power, Indicates the number of servers required by the data center to process the load before computing the load migration; and They represent the lower / upper limit of the virtual energy storage discharge power, Indicates the idle power of the data center server; Indicates the change in the number of servers required after the computing load of the data center is migrated; and respectively represent the number of delay-sensitive computing loads and delay-tolerant computing loads of the data center before migration; represents the floor function; δ(*) represents the Kronecker-δ function; u m represents the efficiency of processing computing load per server in the data center; H(*) represents the Heaviside function; Constructing a construction cost model based on the candidate transmission lines, energy storage equipment and the data center; An operation cost model is constructed based on the output of thermal power units in the power grid on each typical day, the operating status of thermal power units, the carbon emissions of thermal power units, the system load shedding, the new energy generation and actual new energy output in each typical day's uncertain scenario, and the charging power and discharging power of virtual energy storage; The objective function is constructed with the goal of minimizing the total cost of construction cost and operation cost; Based on the virtual energy storage model, equipment investment and construction constraints, system node balance constraints, line transmission capacity constraints, thermal power unit operation constraints, energy storage operation constraints, and new energy power generation constraints, the objective function is solved to obtain the investment and construction plan of the transmission line, energy storage equipment and the data center and the operation plan of the data center.

2. The planning method according to claim 1, characterized in that: The construction of a virtual energy storage model based on the power consumption of the data center and the migration of computing loads in and out in time and space includes: considering the power consumption of the data center as the capacity of the virtual energy storage, considering the migration of computing loads in the data center in time and space as virtual energy storage charging, and considering the migration of computing loads in time and space as virtual energy storage discharge, to construct a virtual energy storage model.

3. The planning method according to claim 2, characterized in that: The construction cost model based on the candidate transmission lines, energy storage equipment and data center is expressed as: Among them, C Inv represents the construction cost; the subscripts l, m, and b represent the candidate transmission lines, energy storage equipment, and virtual energy storage, respectively; L cand , B, and VES are the collections of candidate transmission lines, energy storage equipment, and virtual energy storage, respectively; are 0 / 1 variables, representing the planning and construction status of transmission lines, energy storage equipment, and virtual energy storage; T k represents the life of the candidate transmission lines, energy storage equipment and virtual energy storage; s is the discount rate.

4. The planning method according to claim 3, characterized in that: The operating cost model is expressed as: Among them, C Ope represents the operating cost; U is the uncertainty set in the operating cost model; Ξ is a 0 / 1 variable set; Ψ is a continuous variable set; v is a 0 / 1 variable, including the charging and discharging status of the energy storage device, whether there is an error in the new energy, and the operating status of the thermal power unit; p is a continuous variable, including the virtual energy storage operating variable, the energy storage device operating variable, and the thermal power unit operating variable; Subscripts g, n, and d represent thermal power units, new energy equipment, and loads, respectively; c g P is the power generation cost coefficient of thermal power units; g,t,r Output for thermal power units; c e is the carbon emission cost coefficient; E g,t,r is the carbon emission of thermal power units; Pr VES is the cost coefficient of virtual energy storage charging and discharging; and Respectively represent the charging / discharging power of virtual energy storage; c cur is the penalty coefficient for abandoning wind / solar power; and P n,t,r are the renewable energy generation and actual renewable energy output under uncertain scenarios; c LS is the load shedding penalty coefficient; ΔP d,t,r The system load shedding amount for the load node.

5. The planning method according to claim 4, characterized in that: The equipment investment constraints include: Among them, max Indicates the upper limit of construction cost; They are the upper limit of the number of transmission lines, virtual energy storage and energy storage equipment that are candidates for investment and construction; The configured power of the energy storage device; is the configuration capacity of the energy storage device; τ ess is the continuous discharge time of the energy storage device.

6. The planning method according to claim 5, characterized in that: The system node balance constraints include: Among them, G(s), N(s), B(s), L cand (s), D(s), and VES(s) represent the node set of thermal power units, the node set of new energy sources, the node set of energy storage devices, the set of transmission lines between nodes, the set of load nodes, and the set of virtual energy storage nodes, respectively; and are the discharge and charging power of the energy storage device respectively; P l,t,r is the transmission power of the transmission line; P d,t,r is the load size of the load node; ε PUE Energy utilization efficiency of virtual energy storage.

7. The planning method according to claim 6, characterized in that: The line transmission capacity constraint is expressed as: Among them, i and j are nodes of the power grid system; θ i,t.r and θ j,t,r Represent the voltage phase angles of nodes i and j respectively; and Respectively represent the upper and lower limits of the voltage phase angle of node j; l is the reactance of the line; P l max is the maximum transmission power of the line; M big is a very large positive number.

8. The planning method according to claim 7, characterized in that: The renewable energy generation constraint is expressed as: Among them, p n,t,r is the per unit output value of renewable energy at time t in a typical day r; is the per unit value of power generation of renewable energy at time t in a typical day r; and Whether there is a new energy output prediction error, which is a 0 / 1 variable; and New energy output prediction error; Γ T budget uncertainty for time; It is the electric field installed capacity of the new energy node.

9. The planning method according to claim 8, characterized in that: A two-stage robust optimization method is used to find the optimal solution to the objective function, and the candidate transmission lines, energy storage equipment and construction plans of the virtual energy storage and the operation plan of the virtual energy storage are obtained.

Images