Power distribution network low-carbon transformation planning method considering multi-source uncertainty

Abstract

The invention discloses a power distribution network low-carbon transformation planning method considering multi-source uncertainty, and the method comprises the following steps: 1, building a deterministic model of a low-carbon transformation planning model according to the parameters of a to-be-planned network frame and the parameters related to the planning operation; 2, constructing an uncertainty model, and obtaining a long-term uncertainty variable prediction value and a short-term uncertainty variable prediction value; step 3, constructing a low-carbon transformation planning model based on the step 1 and the step 2; 4, solving a planning problem by adopting a nested generation column constraint enhanced multi-target Harris eagle algorithm; and step 5, outputting all planning solutions, and selecting a preference solution according to a risk management and control target. According to the low-carbon transformation planning method for the power distribution network, for the current low-carbon transformation planning problem of the power distribution network, factors such as coordination of a renewable energy generator and a CCUS and multi-source uncertainty of renewable energy output, user load, external energy price, carbon price and equipment unit deployment cost are considered, and the power availability and the cost optimization planning result can be improved.

Description

Technical Field

[0001] The present invention relates to a distribution network robust planning method, and in particular to a distribution network low-carbon transformation planning method considering multi-source uncertainties. Background Art

[0002] The intensification of global warming has aroused widespread concern about controlling carbon emissions and implementing global low-carbon energy transformation programs. From the perspective of the global carbon emission structure, the power industry accounts for a relatively high proportion of global carbon emissions, up to about 40%. Therefore, decarbonization of the power industry is the key to achieving carbon neutrality. In this context, many countries have proposed their own transformation projects to achieve low-carbon goals in the power industry.

[0003] Mainstream low-carbon transformation solutions at the distribution network level can generally be divided into two categories. One is to deploy distributed renewable energy to replace fossil fuel installations or dependence on external power sources, such as wind turbines (WT) and photovoltaics (PV). This type of approach reduces the proportion of carbon emissions in the distribution network by increasing the share of clean original energy, because the electricity generated by these energy sources has almost no marginal emissions. However, smoothing the volatility of renewable energy output usually requires embedding other adjustment mechanisms or grid assets, including demand response (DR), energy storage (ES), multi-energy microgrids, market mechanisms, etc., to smooth the differences between time-varying loads and renewable energy output.

[0004] Emerging carbon capture technologies offer an alternative. By retrofitting high-carbon-emitting generators, carbon capture equipment (CCS) can extract CO2 from post-combustion exhaust and prevent it from leaking into the atmosphere by storing it. High operating and deployment costs encourage the use of captured CO2 (CCU) as a feedstock for other industrial applications to generate additional revenue, rather than just storage. The combination of CCS and CCU forms the basic concept of carbon capture, utilization and storage (CCUS).

[0005] There is a potential contradiction between the above two methods. Regional carbon emissions of energy systems generally consist of two parts: direct emissions from fossil fuel combustion and indirect emissions from imported external electricity. Only the former can be captured by CCS, but the increase in the proportion of local renewable energy generation will lead to a reduction in direct emissions. Sometimes CCS will even face extreme situations where there are no carbon emissions to capture, weakening its economic feasibility. Therefore, this contradiction should be considered in the low-carbon transformation of distribution networks.

[0006] Dealing with uncertainties is another major challenge in low-carbon transition planning. They usually originate from micro-operations, such as renewable energy and load fluctuations, equipment failures, external energy and carbon price changes, as well as macro-social and technological developments, such as energy policies and breakthroughs in emerging low-carbon technologies. These uncertainties involve different sources and different time scales. For example, retail electricity prices and available solar / wind resources based on real-time pricing mechanisms usually change every hour. As for carbon prices, their purchase prices fluctuate every day and their settlement is based on the fiscal year. Without loss of generality, it can be assumed that system operators can bring their annual unit carbon emission costs close to the average market price of the year through arbitrage or banking excess emission rights, according to the currently introduced carbon market mechanism. Equipment costs are mainly determined by the technology or industry policies in the target deployment year. Therefore, whether these multi-source uncertainties are fully considered will greatly affect the planning results of power availability and cost optimization. Summary of the invention

[0007] The purpose of the present invention is to provide a distribution network low-carbon transformation planning method taking into account multi-source uncertainties. Aiming at the current distribution network low-carbon transformation planning problem, the present invention takes into account factors such as the coordination of renewable energy generators and CCUS as well as multi-source uncertainties of renewable energy output, user load, external energy prices, carbon prices and equipment unit deployment costs, which can improve the planning results of power availability and cost optimization.

[0008] In order to achieve the above object, the solution of the present invention is:

[0009] A distribution network low-carbon transformation planning method considering multi-source uncertainty includes the following steps:

[0010] Step 1: construct a deterministic model of the low-carbon transformation planning model based on the grid parameters to be planned and the parameters involved in the planning operation;

[0011] Step 2: construct an uncertainty model to obtain the predicted values ​​of long-term uncertainty variables and short-term uncertainty variables;

[0012] Step 3: Based on Step 1 and Step 2, a low-carbon transformation planning model is constructed;

[0013] Step 4, using the multi-objective Harris Eagle algorithm strengthened by nested generated column constraints to solve the planning problem;

[0014] Step 5: Output all planning solutions and select the preferred solution based on the risk management objectives.

[0015] In step 1 above, the goal of the low-carbon transformation planning model is to minimize the investment cost of new equipment and the total operating cost of the energy supply system during the target period, and its constraints include capacity planning constraints and operation constraints.

[0016] The equipment subject to the above-mentioned operating constraints include generators, new energy units, carbon capture equipment, electric gas production, energy storage equipment, carbon storage equipment and loads.

[0017] In step 2 above, the information gap decision model based on envelope constraints is used to model long-term uncertainty, which is expressed as:

[0018]

[0019] in, The uncertainty set of equipment planning cost for the information gap decision model based on envelope constraints, is the predicted value of equipment planning cost, α eq is the uncertainty margin, is the uncertainty set of carbon quota prices in the information gap decision model based on envelope constraints, π ER is the carbon quota price, Predicting prices for carbon allowances.

[0020] In step 2 above, the uncertainty budget model is used to model short-term uncertainty, which is expressed as:

[0021]

[0022] in, rs s,t , The predicted value, Δ is the preset deviation size, is a positive deviation 0-1 variable, is a negative deviation 0-1 variable, Γ RS , Γ L , Γ E , Γ G is the uncertainty budget parameter.

[0023] In step 3 above, the expression of the low-carbon transformation planning model is:

[0024]

[0025] (1)-(24)

[0026] in, is the user's expected planning cost, RB is the parameter for setting system uncertainty, C is the total cost, C = C INV +σ year C OP , σ year The number of planned operating years.

[0027] The specific process of step 4 above is:

[0028] Step 41, obtaining a long-term uncertainty variable prediction value;

[0029] Step 42, based on the upper limit value of the long-term uncertainty variable, solving the lower problem RO to obtain a first worst scenario; based on the lower limit value of the long-term uncertainty variable, solving the lower problem RO to obtain a second worst scenario;

[0030] Step 43: Substitute the first worst scenario and the second worst scenario into the lower problem RO, and solve the planning model to obtain the Pareto solution.

[0031] Step 44, let re take the values ​​of 1, ..., nRep respectively, and perform the following operations:

[0032] Based on the re-th solution of the long-term uncertainty variable, solve the lower-level problem RO and determine whether the constraint is established. If the constraint is established, output the corresponding planning result, otherwise cancel the solution;

[0033] Step 45, obtain all planning solutions.

[0034] After adopting the above scheme, the beneficial effects of the present invention are:

[0035] The present invention comprehensively considers the multiple uncertain factors faced in the transformation process of the distribution network, such as equipment deployment cost, carbon quota price, wind and solar resource fluctuation, etc., classifies and models the uncertainty factors on different time scales, adopts the uncertainty set model based on information gap decision technology (IGDT) to characterize the medium- and long-term uncertainty factors such as equipment deployment cost and carbon quota price, and adopts the uncertainty set model based on uncertain budget to characterize the short-term uncertainty factors such as wind speed, light and load forecast error, and constructs a two-layer robust optimization model for the low-carbon transformation planning of the distribution network. In view of the non-convex nonlinear characteristics of the model, a nested generated column constraint-enhanced multi-objective Harris Eagle algorithm is designed to achieve effective solution to the problem, laying a foundation for the collaborative planning of low-carbon equipment in low-carbon distribution networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of the multi-objective Harris Eagle algorithm with nested generated column constraint reinforcement in the present invention;

[0037] Figure 2 It is a system network diagram;

[0038] Figure 3 It is the running scene graph;

[0039] Among them, (a) is the total load, (b) is the sunlight, (c) is the wind speed, and (d) is the electricity price;

[0040] Figure 4 is the solution result;

[0041] Figure 5It is a flow chart of the present invention. DETAILED DESCRIPTION

[0042] like Figure 5 As shown, the present invention provides a distribution network low-carbon transformation planning method considering multi-source uncertainty, comprising the following steps:

[0043] S101: Obtaining grid parameters to be planned, including key parameters such as network topology, line impedance, line capacity, and voltage upper and lower limits;

[0044] S102: Obtaining parameters involved in planning and operation, including upper and lower limits of planned capacity, locations of planned equipment nodes, operation and maintenance costs, etc.;

[0045] S103: Obtain the predicted values ​​of long-term uncertain variables, including the average carbon price in the operating year, equipment deployment cost and other predicted values;

[0046] S104: Obtaining short-term uncertain variable prediction values, including wind speed, photovoltaic, load, upper power source power purchase price, natural gas sales price and other prediction values ​​in specific operation scenarios;

[0047] S105: Construct a corresponding low-carbon transformation planning model according to equations (1)-(24);

[0048] S106: The proposed nested generated column constraint-enhanced multi-objective Harris Hawk algorithm is used to solve the corresponding planning problem;

[0049] S107: Output all planning solutions and select preferred solutions based on risk management objectives.

[0050] The technical solutions and beneficial effects of the present invention will be described in detail below with reference to the accompanying drawings.

[0051] 1 Model construction

[0052] The main equipment considered in the distribution network low-carbon transformation planning model includes fossil fuel generators, renewable energy generators, energy storage, carbon capture and storage (CCS), and power-to-gas (P2G). Generators provide energy to meet the needs of park users and power power equipment such as CCS and P2G. CCS captures carbon dioxide emitted by fossil fuel generators. The captured carbon dioxide can be stored or input into P2G as raw material to produce methane through CO2+4H2→CH4+2H2O to obtain additional income. First, the deterministic model of the distribution network low-carbon transformation planning is introduced.

[0053] 1.1 Objective Function

[0054] The goal of the planning model is to minimize the investment cost of new equipment and the total operating cost of the energy supply system during the target period. The total investment cost expression is

[0055]

[0056] The operating cost expression is

[0057] C OP =C MT +C TR +C ENV +C CUR +C UN (2)

[0058] Among them, ρ eq , Cap eq is the 0-1 variable and planned capacity of the equipment planning, and eq is the equipment name variable, including wind power (WT), photovoltaic (PV), energy storage (ES), carbon capture equipment (CCS) and power to gas (P2G). MT , C TR , C ENV , C CUR , C UN They are maintenance cost, transaction cost, environmental cost, power abandonment penalty and non-supply penalty, and their calculation formulas are

[0059]

[0060] Among them, c GEN 、c GEN,on 、c GEN,off 、c WT 、c PV 、c ES 、c CCS 、c P2G They are the maintenance cost per unit power of generator, the cost of a single startup, the cost of a single shutdown, the maintenance cost per unit power of wind power, the maintenance cost per unit power of photovoltaic power, the maintenance cost per unit power of CCS, and the maintenance cost per unit power of P2G. They are respectively the generator power generation, startup and shutdown 0-1 indicator variables, wind power generation, photovoltaic power generation, total energy storage charging and discharging power, CCS input power, and P2G input power at scene s and time t. fuel , κ buy They are the generator fuel price, electricity purchase price, and CO2 purchase cost; Consume fuel for the generator, For the purchased power, Producing fuel for P2G, is the amount of carbon dioxide purchased. ER is the carbon quota price, is the carbon emission cost of the distribution network that should be included in carbon accounting; seq is the unit carbon dioxide carbon sequestration cost, is the carbon storage amount. CUR Punishment for unit power abandonment, It is the amount of photovoltaic and wind power abandoned. UN For the unit's failure to provide punishment, is the unsupplied load of node i; are respectively a collection of scenes and a collection of scheduling times, It is a collection of distribution network nodes.

[0061] 1.2 Constraints

[0062] 1.2.1 Planning constraints

[0063] Capacity planning should meet some upper and lower constraints, such as

[0064]

[0065] In the formula, u eq 0-1 variables planned for the device, and Cap eq The upper and lower limits of the planned capacity of the equipment are respectively.

[0066] 1.2.2 Operational constraints

[0067] The operational constraints considered in the planning include equipment operation and energy conversion model, user load model, network and balance model.

[0068] (1) Generator

[0069] Generator constraints include:

[0070]

[0071] in, It is the 0-1 variable of the start and stop status of the generator set. P GEN , They are the upper and lower limits of the active output of the generator, Q GEN , are the upper and lower limits of the reactive power output of the generator, Carbon emissions from generators, m C is the relative molecular mass of carbon dioxide molecules and carbon atoms, OF is the oxidation rate, To consume fuel, is the upper limit of climbing power, T GEN,on 、T GEN,off Start and stop time respectively.

[0072] (2) New energy units

[0073] The constraints of new energy units are

[0074]

[0075] in, are the upper limit of the output of the new energy unit, the actual output and the abandoned power, w and r are the wind speed and sunlight, respectively. It is the adjustable range coefficient of reactive power output, and its value is cosarctan0.95 according to the manual. Cap rg Planning capacity for distributed clean energy.

[0076] (3)CCS

[0077]

[0078] in, It is the CCS start / stop status 0-1 variable. is the basic power of CCS startup, η CCS is the carbon capture efficiency, is the amount of captured carbon dioxide; M is a larger number, generally 1e6.

[0079] (4) P2G

[0080]

[0081] in, It is the P2G start / stop status 0-1 variable. is the basic power for P2G startup, η P2G is the P2G efficiency, H G The low calorific value of natural gas. The amount of natural gas produced by P2G. Consumption of carbon dioxide for P2G,υ G is the density of natural gas.

[0082] (5)ES

[0083]

[0084] in, is the current capacity of energy storage, are the energy storage charging and discharging power, Δt is a scheduling time, η ES,ch , η ES,dis is the charge and discharge efficiency, is the charge and discharge state 0-1 variable, k ES,φ It is the adjustable range coefficient of the reactive output of energy storage, and its value is tan arccos0.95 according to the manual. Provide reactive power for energy storage.

[0085] (6) Carbon storage (CS)

[0086]

[0087] in, is the carbon storage capacity, η CS To improve the efficiency of carbon dioxide charging and discharging, is the upper limit of carbon storage capacity, are the amount of carbon dioxide charged and discharged, They are 0-1 variables for charging and discharging carbon dioxide status, The upper and lower limits of carbon storage and release.

[0088] (7) Load

[0089] For nodes The load model is

[0090]

[0091] in, is the user load demand, To actually dispatch load power for users, Reactive power for user load, is the unsupplied load of node i.

[0092] (8) AC linear power flow

[0093]

[0094] is the mapping of the device to the nodes in the network, ε is the set of branches in the network, is the active and reactive power passing through the line, VS i,s,t It is the square of the node voltage per unit value. ij , X ij are the line resistance and reactance parameters, is the upper limit of the line transmission capacity, P is the upper limit of the purchased power; i,s,t , Q i,s,t are the node active load and node reactive load, VS 1,s,t is the square of the per-unit value of the relaxed node voltage, VS are the upper and lower limits of the square of the voltage per unit value, For outsourced power.

[0095] (9) Carbon balance constraints

[0096]

[0097] in, Net CO2 emissions from the distribution network, For the purchased carbon dioxide, For the stored carbon dioxide, Indirect carbon emissions introduced by purchasing electricity; up is the carbon emission factor of the upper power source, System carbon emissions that should be included in the carbon quota assessment.

[0098] 1.3 Uncertainty analysis

[0099] 1.3.1 Modeling long-term uncertainty

[0100] The price of carbon emission quotas and the unit deployment cost of new emission reduction equipment, including WT, PV, ES, CCS, P2G, etc., are long-term uncertain factors. Since they lack historical data or are easily affected by social and economic exogenous factors, it is difficult to generate scenarios similar to the fluctuations of renewable energy and energy prices on a daily time scale. This embodiment proposes an information gap decision model based on envelope constraints to model this type of uncertainty. Its specific form is

[0101]

[0102] in, The uncertainty set of equipment planning cost for the information gap decision model based on envelope constraints, is the predicted value of equipment planning cost, α eq is the uncertainty margin, is the uncertainty set of carbon quota prices in the information gap decision model based on envelope constraints, π ER is the carbon quota price, Predicting prices for carbon allowances.

[0103] 1.3.2 Modeling short-term uncertainty

[0104] Short-term uncertainties include available renewable resources, energy prices, and load demand. These parameters have the following characteristics: (1) they fluctuate frequently within or between days; (2) they have accumulated a wealth of historical data, and patterns can be analyzed through data mining; (3) they are mainly affected by seasonal factors and are limited by economic, policy, and other factors. Therefore, compared with long-term uncertainties, a complex method is needed to model such uncertainties to strike a balance between optimality and conservatism, rather than simply specifying upper and lower limits. The present invention adopts an uncertainty budget model to model short-term uncertainties:

[0105]

[0106] in, rs s,t , The predicted value, Δ is the preset deviation size, is a positive deviation 0-1 variable, is a negative deviation 0-1 variable, Γ RS , Γ L , Γ E , Γ G is the uncertainty budget parameter.

[0107] 2 Two-level robust programming model

[0108]

[0109] in is the user's expected planning cost, RB is the parameter for setting system uncertainty, C is the total cost, C = C INV +σ year C OP , σ year The number of planned operating years.

[0110] 3 Solution Strategies

[0111] 3.1 Nested Generated Column Constraint Algorithm

[0112] This section describes the main steps of the Nested Column & Constraint Generation Algorithm (Nested C&CG). The lower-level problem RO in (24) can be written in the following general form:

[0113]

[0114] Among them, c, d, g are decision variable cost coefficients, y is the first-stage optimization variable, x and z are the second-stage optimization variables, and Ay≤b is the first-stage constraint. The second stage optimizes the variable value range considering uncertainty. Then the main problem can be expressed as

[0115]

[0116] Assume y * It is known that the subproblem can be written as follows

[0117]

[0118] Assuming feasible domain can be expressed as Then the sub-master problem (InMP) and sub-slave problem (InSP) can be expressed as

[0119]

[0120] The sub-problem can be embedded into the main problem through KKT condition transformation and solved. The pseudo code of this method is as follows:

[0121] Algorithm 1: Nested column constraint generation algorithm Input: cost parameters ρ, π ER

[0122] Output: planning result u, Cap

[0123] Initialization: OutUB←+∞, OutLB←-∞, ∈←+∞, m←0;

[0124] While do

[0125] m←m+1;

[0126] Solve MP to obtain y (m),* ;

[0127]

[0128] InUB←+∞,InLB←-∞,∈′←+∞,n←0;

[0129] While do

[0130] n←n+1;

[0131] Derive the dual multiplier λ of InSP (n) and KKT conditions and added to InMP;

[0132] Solve InMP and obtain γ (n),* ;

[0133] InUB←θ (n),* ;

[0134] Solve InSP and obtain (x (n),* ,z (n),* );

[0135] InLB←max(InLB,dx (n),* +gz (n),* );

[0136]

[0137] end

[0138] OutUB←min(OutUB,cy (n),* +InLB);

[0139] Create a variable (x (m+1) ,z (m+1 ))

[0140] Add the following constraints to MP:

[0141] η≥dx (m+1) +gz (m+1) ,Ex (m+1) +Gz (m+1) ≤f-Rγ (m+1),* -Dy

[0142]

[0143] end

[0144] 3.2 Multi-objective Harris Hawk Optimization

[0145] The main steps of the Harris Hawk Optimization (HHO) algorithm include two phases: exploration phase and development phase.

[0146] (1) Exploration: The exploration phase simulates the behavior of an eagle when detecting prey. Their positions indicate the target solution for the (τ+1)th iteration, and their positions are randomly generated by

[0147]

[0148] Among them, X rabbit (τ) is the position of the rabbit, X rand (τ) is the randomly selected eagle position, X m (τ) is the average position of the current population. UB and LB define the upper and lower limits of the variables to be determined, which are the optimization variables of the upper problem in equation (24). r1~r4 are random parameters between (0,1).

[0149] (2) Transition from exploration to exploitation:

[0150] By simulating a real hunting scenario, the eagle's behavior changes from exploration to exploitation when the prey's escape energy decreases. The remaining escape energy EN of the prey is modeled as

[0151]

[0152] EN0 is the randomly generated prey energy level.

[0153] (3) Exploitation stage: In this stage, the eagle forms a siege around the prey and adjusts its hunting behavior according to the state of the prey. In the development stage, the eagle will develop four strategies, and the specific strategy to be adopted depends on the escape energy EN and the chance of successful escape of the prey.

[0154] Strategy 1: Soft Encirclement

[0155]

[0156] Among them, J=2(1-r5) is the length of the bunny's random jump, and r5 is a random parameter between (0,1).

[0157] Strategy 2: Hard Encirclement

[0158] X(τ+1)=X rabbit (τ)-EN|ΔX(τ)| (32)

[0159] Strategy 3: Soft siege and gradual rapid dive

[0160] At this stage, the prey has a chance to escape the siege, so the hawks need to form a smarter siege route before capturing the prey. In this strategy, two types of next actions are considered, defined as

[0161]

[0162] D is the dimension of the problem, and S is a 1×D random vector.

[0163]

[0164] Among them, β = 1.5, μ, σ are (0,1) random variables, so in the next iteration, the position of the eagle is

[0165]

[0166] F(·) is the adaptation function value, that is, the value of the objective function of the upper model in formula (24).

[0167] Strategy 4: Hard Siege and Progressive Fast Dive

[0168] Similar to strategy 3, two movement strategies are defined

[0169]

[0170] The judgment conditions are the same as those of strategy 3.

[0171] The pseudo code of multi-objective HHO is as follows:

[0172] Algorithm 2: Multi-objective Harris Hawk optimization algorithm

[0173] Input: population size nPop, storage size nRep, maximum iteration MaxIt, upper and lower limits of the optimization variables in (24).

[0174] Output: Pareto solution set X rep

[0175] Initialization: it←0, Generate the initial population Xi ,i=1,…,nPop

[0176] Calculate the fitness of the initial population. The fitness function is the objective function of the upper model in (24)

[0177] Let X rabbit Position of prey

[0178] Determine whether it is a Pareto solution and put it in X rep middle

[0179] for it=1:MaxIt do

[0180]

[0181] for i=1:nPop do

[0182] From X rep Choose a solution as the leader

[0183] EN0←2RAND-1,r←RAND

[0184] EN←EN0×EN1

[0185] If EN≥1, execute the exploration phase end

[0186] if EN<1

[0187] if EN≥0.5&r≥0.5 execute strategy 1end

[0188] if EN<0.5&r≥0.5Execute strategy 2end

[0189] if EN≥0.5&r<0.5Execute strategy 3end

[0190] if EN<0.5&r<0.5Execute strategy 4end

[0191] end

[0192] Mutations:

[0193] Mu←RAND

[0194] if Mu<pm mutates current X rabbit end

[0195] If Mu ≥ pm, update X with the previous mutation rabbit end

[0196] Add non-dominated solutions to X rep

[0197] Determine the current X rep Is there a dominated solution in the solution and only keep the non-dominated solutions

[0198] Check X rep Is it full?

[0199] end

[0200] 3.3 Multi-objective Harris Hawk Algorithm with Constraint Strengthening of Nested Generated Columns

[0201] The present invention adopts the Nested Column & Constraint Enhanced Multi-objective Harris Hawk Optimization (Nested C&CGenhanced MOHHO) algorithm to solve the double-layer multi-objective mixed integer optimization problem such as (24). The basic process is as follows: Figure 1 shown.

[0202] An application example of the present invention is provided below.

[0203] 3.1 Operating parameters

[0204] The simulations in this section consider a ten-year planning and operation period. Figure 2 As shown in Figure 1, the unit planning costs of PV, WT, ES, CCS and P2G are 6.10, 8.78, 1.77, 0.55 and 10.75 million RMB per MW or MWh respectively. We assume that planning starts in 2026. The carbon right price is 116.18￥ / tCO2. The annual operation includes four scenarios to describe the potential possibilities. The electricity purchase price and natural gas sales price adopted in each scenario are as follows Figure 3 As shown. The carbon emission factor of the upstream power grid is fixed at 0.4872. The short-term uncertainty offset is set to 10%, and the long-term uncertainty offset variable ranges from [0, 0.5]. The storage size is 10, the population size is 10, and the maximum iteration is 20. RB is 0.2.

[0205] Table 1 Operation parameters

[0206]

[0207] Table 2 Planning parameters

[0208]

[0209] 3.2 Solution results

[0210] The example solves 10 solutions such as Figure 4As shown, the planning values ​​corresponding to Sol.1 are PV: 0.289, WT: 1.005, ES: 5.982, CCS: 0.859, P2G: 0.100, and the total cost is 130.439 million yuan.

[0211] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0212] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0213] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0214] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0215] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0216] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A distribution network low-carbon transformation planning method considering multi-source uncertainty, characterized by The steps include: Step 1: construct a deterministic model of the low-carbon transformation planning model based on the grid parameters to be planned and the parameters involved in the planning operation; Step 2: construct an uncertainty model to obtain the predicted values ​​of long-term uncertainty variables and short-term uncertainty variables; Step 3: Based on Step 1 and Step 2, a low-carbon transformation planning model is constructed; Step 4, using the multi-objective Harris Eagle algorithm strengthened by nested generated column constraints to solve the planning problem; Step 5: Output all planning solutions and select the preferred solution based on the risk management objectives.

2. The method according to claim 1, characterized in that: In step 1, the goal of the low-carbon transformation planning model is to minimize the investment cost of new equipment and the total operating cost of the energy supply system during the target period, and its constraints include capacity planning constraints and operation constraints.

3. The method according to claim 2, characterized in that: The equipment with operation constraints includes generators, new energy units, carbon capture equipment, electric gas, energy storage equipment, carbon storage equipment and loads.

4. The method according to claim 1, characterized in that: In step 2, the information gap decision model based on envelope constraints is used to model long-term uncertainty, which is expressed as: in, The uncertainty set of equipment planning cost for the information gap decision model based on envelope constraints, is the predicted value of equipment planning cost, α eq is the uncertainty margin, is the uncertainty set of carbon quota prices in the information gap decision model based on envelope constraints, π ER is the carbon quota price, Predict prices for carbon allowances.

5. The method according to claim 1, characterized in that: In step 2, an uncertainty budget model is used to model short-term uncertainty, which is expressed as: in, rs s,t , The predicted value, Δ is the preset deviation size, is a positive deviation 0-1 variable, is a negative deviation 0-1 variable, Γ RS , Γ L , Γ E , Γ G is the uncertainty budget parameter.

6. The method according to claim 1, characterized in that: In step 3, the expression of the low-carbon transformation planning model is: in, is the user's expected planning cost, RB is the parameter for setting system uncertainty, C is the total cost, The number of planned operating years.

7. The method according to claim 6, characterized in that: The specific process of step 4 is: Step 41, obtaining a long-term uncertainty variable prediction value; Step 42, based on the upper limit value of the long-term uncertainty variable, solving the lower problem RO to obtain a first worst scenario; based on the lower limit value of the long-term uncertainty variable, solving the lower problem RO to obtain a second worst scenario; Step 43: Substitute the first worst scenario and the second worst scenario into the lower problem RO, and solve the planning model to obtain the Pareto solution. re=1,…,nRep; Step 44, let re take the values ​​of 1, ..., nRep respectively, and perform the following operations: Based on the re-th solution of the long-term uncertainty variable, solve the lower-level problem RO and determine whether the constraint is established. If the constraint is established, output the corresponding planning result, otherwise cancel the solution; Step 45, obtain all planning solutions.

Images