Converter station capacity planning method and device

Abstract

The invention provides a converter station capacity planning method and device, and the method comprises the steps: obtaining historical wind power output data and historical photovoltaic output data, and building a wind-solar output model and a multivariate frequency response model; initializing hyper-parameters of a particle swarm algorithm, wherein the positions of particles in the particle swarm algorithm represent the direct current quantity fed into the receiving end power grid by the target converter station; solving to obtain a first position by taking the maximum direct current quantity fed into the receiving end power grid by the target converter station as a target and combining a wind-light output model and a multivariate frequency response model; solving to obtain a first result by taking the minimum comprehensive cost of the target converter station as a target and combining a wind-light output model and a multivariate frequency response model; and if the first result meets the convergence condition, outputting the direct current amount corresponding to the first position and the first result, and obtaining the direct current amount and cost of the target converter station fed into the receiving end power grid. The problem that a traditional power grid capacity planning method is difficult to meet large-scale new energy grid connection requirements can be solved.

Description

Technical Field

[0001] This application belongs to the technical field of converter station design, and particularly relates to a method and device for improving the capacity planning of converter stations. Background Art

[0002] With the increasing scarcity of traditional energy and the continuous attention to environmental issues, the sustainable development of energy requires us to shift from relying on traditional energy to green renewable energy represented by wind energy and solar energy. However, different from traditional power generation methods such as thermal power and hydropower, the output of new energy sources such as wind energy and solar energy is intermittent and random. Considering the acceptance capacity of the power system for new energy and the need to ensure the safe and stable operation of the system, the phenomena of "wind curtailment" and "solar curtailment" are very common. Traditional operation technologies severely limit the development and application of renewable energy in China.

[0003] As a key development direction for future power transmission, the DC power grid has shown unique advantages in large-scale new energy access with its lower energy loss and higher power transmission capacity. However, with the continuous emergence of new power sources and loads, the capacity planning of the DC power grid faces complex optimization challenges, and many factors such as the power generation characteristics of new energy, load demand, system stability, and dispatching strategies need to be comprehensively considered. Traditional DC power grid capacity planning methods are mainly based on fixed load characteristics and the production scheduling rules of traditional generator sets, and it is difficult to meet the requirements of large-scale new energy grid connection. This is because the power generation characteristics of new energy are uncertain and random, especially in different climate conditions and time periods, there are significant fluctuations in power generation capacity. This uncertainty requires the power system to have higher dispatching capabilities and flexibility to ensure the reliability, economy, and stability of power supply. Summary of the Invention

[0004] The embodiments of this application provide a method and device for converter station capacity planning to solve the problem that traditional DC power grid capacity planning methods are difficult to meet the requirements of large-scale new energy grid connection.

[0005] This application is implemented through the following technical solutions:

[0006] In the first aspect, the embodiments of this application provide a method for converter station capacity planning, including:

[0007] Obtain historical wind power output data and historical photovoltaic power output data, and establish a wind-solar power output model and a multi-frequency response model.

[0008] Initialize the hyperparameters of the particle swarm optimization algorithm. The position of the particle in the particle swarm optimization algorithm represents the DC current fed by the target converter station into the receiving-end power grid. With the goal of maximizing the DC current fed by the target converter station into the receiving-end power grid and using the first constraint, the wind-solar power output model, and the multi-frequency response model as conditional constraints, use the particle swarm optimization algorithm to solve the optimal position of the particle and obtain the first position.

[0009] Taking the minimum comprehensive cost of the target converter station as the goal, and taking the direct current quantity corresponding to the first position, the second constraint, the wind-solar power output model, and the multi-frequency response model as conditional constraints, solve using the second-order cone relaxation algorithm to obtain the first result.

[0010] If the first result meets the convergence condition, output the direct current quantity corresponding to the first position and the first result to obtain the direct current quantity and cost fed by the target converter station into the receiving-end power grid.

[0011] Combined with the first aspect, in some possible implementation manners, the method further includes:

[0012] If the first result does not meet the convergence condition, return to the step of solving the optimal position of the particle using the particle swarm algorithm to obtain the sub-optimal position, and use the direct current quantity fed by the target converter station corresponding to the sub-optimal position into the receiving-end power grid as the second result.

[0013] Taking the minimum comprehensive cost as the goal, and taking the second result, the second constraint, the wind-solar power output model, and the multi-frequency response model as conditional constraints, solve using the second-order cone relaxation algorithm to obtain the third result.

[0014] If the third result meets the convergence condition, output the second result and the third result to obtain the direct current quantity and cost fed by the target converter station into the receiving-end power grid.

[0015] If the third result does not meet the convergence condition, return to the step of solving the optimal position of the particle using the particle swarm algorithm until the convergence condition is met.

[0016] Combined with the first aspect, in some possible implementation manners, the first constraint includes: DC capacity expansion constraint, active power margin constraint, inertia constraint, and DC power constraint.

[0017] The DC capacity expansion constraint is:

[0018] P DC = P′ DC x + P″ DC y

[0019] 0 ≤ x + y ≤ 1

[0020] Wherein, P′ DC and P″ DC represent the DC capacities in different construction cases, x ∈ {0, 1}, y ∈ {0, 1}, when x takes 1, it means the DC capacity built at x is P′ DC , when y takes 1, it means the DC capacity built at y is P″ DC , when x = 0 or y = 0, it means no construction.

[0021] The active power margin constraint is:

[0022]

[0023] Among them, represents the maximum DC power fed into the receiving-end power grid, represents the maximum active power provided by wind power, represents the maximum active power provided by photovoltaic power, represents the maximum active power provided by energy storage, represents the maximum active power provided by thermal power units. Re represents the reserve capacity factor, which is taken as 10% here, represents the maximum active power required by the load, N W represents the number of wind turbines, N S represents the number of photovoltaic units, N E represents the number of energy storage units, N L represents the number of loads, N G represents the number of thermal power units.

[0024] The inertia constraint is:

[0025]

[0026] H DCmin ≤H DC ≤H DCmax

[0027] Among them, H represents the equivalent inertia constant of the combined frequency regulation of DC, wind power, photovoltaic power, and energy storage in the receiving-end power grid, J AC represents the rotational inertia of the receiving-end power grid AC system. The system rotational inertia should be ensured to be greater than the minimum inertia value for system safety and stability, H DC represents the inertia support provided by DC; H W represents the inertia support provided by wind power; H S represents the inertia support provided by photovoltaic power; H E represents the inertia support provided by energy storage.

[0028] The DC power constraint is:

[0029] P DCmin ≤P DC ≤P DCmax

[0030] Among them, P DCmin represents the minimum value of the DC inertia support power, P DCmax represents the maximum value of the DC inertia support power, which are the minimum and maximum powers of DC transmission respectively.

[0031] In combination with the first aspect, in some possible implementation manners, the specific range of the active power margin constraint is calculated based on the wind-solar power output model; the specific range of the inertia constraint is calculated based on the wind-solar power output model and the multi-source frequency response model.

[0032] In combination with the first aspect, in some possible implementation manners, the second constraint includes: power balance constraint, unit output constraint, reserve constraint, transmission limit power constraint of the receiving-end power grid line, and frequency change rate constraint under DC bipolar blocking.

[0033] The power balance constraint is:

[0034]

[0035] Wherein, P DC represents the DC power fed into the receiving-end power grid, P W represents the active power provided by the wind power, P S represents the active power provided by the photovoltaic power, P E represents the active power provided by the energy storage, P G represents the active power provided by the thermal power unit, P L represents the active power required by the load, N W represents the number of wind turbines, N S represents the number of photovoltaic panels, N E represents the number of energy storage units, N L represents the number of loads, N G represents the number of thermal power units.

[0036] The unit output constraint is:

[0037] P i +ΔP U ≤P i max

[0038] P i -ΔP D ≤P i min

[0039] |P i -P i-1 |≤ΔP

[0040]

[0041] Wherein, P i represents any one of the DC power fed into the receiving-end power grid at the i-th moment, the active power provided by the wind power at the i-th moment, and the active power provided by the energy storage at the i-th moment, P i max represents the maximum value of P i the maximum value of P imax Represents the minimum value of P i , and ΔP U represents the upward regulation reserve, and ΔP D represents the downward regulation reserve.

[0042] The reserve constraint is:

[0043]

[0044] where ΔP U,i represents the active power deficit at time i, and PΔ represents the active power required for system frequency regulation when power deviation occurs in the system. DC, wind power, photovoltaic power, and energy storage can provide sufficient active power support at any time.

[0045] The transmission limit power constraint of the receiving-end grid line is:

[0046]

[0047] where PT is the power transmission transfer factor, C = G + W + E + S + L, SD represents the DC connection factor, SG represents the thermal power unit connection factor, SW represents the wind turbine connection factor, SE represents the energy storage connection factor, SS represents the photovoltaic connection factor, SL represents the connection factor of the load accessing the receiving-end grid; G represents the number of thermal power units connected, W represents the number of wind turbines connected, E represents the number of energy storage connected, S represents the number of photovoltaic connected, and L represents the number of loads connected; represents the active power transmission limit value of the receiving-end grid AC grid.

[0048] The constraint on the rate of change of frequency under DC bipolar blocking is:

[0049] P DCmin ≤P DC ≤P DCmax

[0050]

[0051]

[0052] where f ref represents the reference frequency, H' represents the inertia constant of the combined frequency regulation of DC, wind power, photovoltaic power, and energy storage, and RoCoF max represents the maximum allowable rate of change of frequency of the system.

[0053] Combined with the first aspect, in some possible implementation manners, the specific ranges of the power balance constraint and the transmission limit power constraint of the receiving-end grid line are calculated based on the wind-solar output model; the specific range of the constraint on the rate of change of frequency under DC bipolar blocking is calculated based on the wind-solar output model and the multi-source frequency response model.

[0054] In combination with the first aspect, in some possible implementation manners, the comprehensive cost of the target converter station is:

[0055]

[0056] wherein, F2 represents the comprehensive cost of the target converter station, c represents the cash conversion coefficient, Y E represents the operating cost of the energy storage device per unit power, Y G represents the operating cost of the thermal power device per unit power, W E represents the maintenance cost of the energy storage device per unit power, W G represents the maintenance cost of the thermal power device per unit power, P t E represents the energy storage output at time t, P t G represents the thermal power output at time t, and N represents the total number of thermal power devices and energy storage devices.

[0057] In combination with the first aspect, in some possible implementation manners, historical wind power output data and historical photovoltaic power output data are obtained, and a wind-solar power output model and a multi-source frequency response model are established, including:

[0058] Obtain historical wind power output data and historical photovoltaic power output data.

[0059] Based on the historical wind power output data and historical photovoltaic power output data, a wind-solar power output model is established.

[0060] Based on the support principles of wind power generation, photovoltaic power generation, energy storage system power supply, and DC power supply for the grid frequency stability, a multi-source frequency response model is obtained.

[0061] In combination with the first aspect, in some possible implementation manners, the wind-solar power output model is:

[0062] A = {(W, S), L(t) ~ N(μ L , σ L 2 )}

[0063] wherein, A represents the wind-solar power output model, (W, S) represents the combined output scenario of wind power output and photovoltaic power output, a typical scenario set generated by the joint probability distribution C(u, v) and the sampling method, W represents the wind power output, S represents the photovoltaic power output, u represents the marginal distribution function of the wind power output, v represents the marginal distribution function of the photovoltaic power output; L(t) represents the power demand required by the power system, L(t) ~ N(μ L , σ L 2 ) represents that the power demand required by the power system satisfies the normal distribution, μ Lrepresents the mean value of the power demand required by the power system, and σ L 2 represents the variance of the power demand required by the power system.

[0064] The multi - variable frequency response model is:

[0065]

[0066] Among them, ΔP W represents the instantaneous change in the wind power inertia support power, H represents the inertia system constant, ω represents the change in the system frequency; P S represents the photovoltaic inertia support power, P sref represents the rated output power of the photovoltaic, f(t) represents the instantaneous power of the power grid, f nom = 50Hz, Δf represents the acceptable frequency fluctuation range; P E represents the energy storage inertia support power, P max represents the maximum power that the energy storage system can provide instantaneously, v i represents the instantaneous voltage of the battery, v ref represents the reference voltage or the rated voltage of the power grid, R represents the internal resistance of the battery; P DC represents the DC inertia support power, represents the rate of change of frequency.

[0067] In a second aspect, an embodiment of the present application provides a converter station capacity planning device, including:

[0068] A data acquisition module, configured to acquire historical wind power output data and historical photovoltaic power output data, and establish a wind - solar power output model and a multi - variable frequency response model.

[0069] A first operation module, configured to initialize the hyperparameters of the particle swarm algorithm, where the position of the particle in the particle swarm algorithm represents the DC quantity fed from the target converter station to the receiving - end power grid; aiming at maximizing the DC quantity fed from the target converter station to the receiving - end power grid, and taking the first constraint, the wind - solar power output model and the multi - variable frequency response model as conditional constraints, use the particle swarm algorithm to solve the optimal position of the particle and obtain the first position.

[0070] A second operation module, configured to aim at minimizing the comprehensive cost of the target converter station, and taking the DC quantity corresponding to the first position, the second constraint, the wind - solar power output model and the multi - variable frequency response model as conditional constraints, use the second - order cone relaxation algorithm to solve and obtain the first result.

[0071] A result output module, configured to output the DC quantity corresponding to the first position and the first result if the first result meets the convergence condition, so as to obtain the DC quantity fed from the target converter station to the receiving - end power grid and the cost.

[0072] It is understandable that the beneficial effects of the above second aspect can be referred to the relevant descriptions in the above first aspect, and will not be elaborated here.

[0073] The beneficial effects of the embodiments of the present application compared with the prior art are as follows:

[0074] First, the present application establishes a wind-solar output model and a multi-frequency response model through historical wind power output data and historical photovoltaic output data. Then, with the goal of maximizing the direct current fed into the receiving-end power grid by the target converter station, and with the first constraint, the wind-solar output model and the multi-frequency response model as conditional constraints, the particle swarm algorithm is used to solve the optimal position of the particles to obtain the first position. After that, with the goal of minimizing the comprehensive cost of the target converter station, and with the direct current corresponding to the first position, the second constraint, the wind-solar output model and the multi-frequency response model as conditional constraints, the second-order cone relaxation algorithm is used to solve and obtain the first result. Finally, if the first result meets the convergence condition, the direct current corresponding to the first position and the first result are output to obtain the direct current and cost of the target converter station fed into the receiving-end power grid. This solution comprehensively considers the influence of the wind-solar output model and the multi-frequency response model on the converter station. At the same time, with the goal of maximizing the direct current fed into the receiving-end power grid by the target converter station and minimizing the comprehensive cost, it can improve the cost performance of building the converter station. The converter station built according to the results of this solution has higher dispatching capabilities and flexibility, can ensure the reliability, economy and stability of power supply, and meet the needs of large-scale new energy grid connection.

[0075] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit this specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0077] Figure 1 is a flowchart of a converter station capacity planning method provided by an embodiment of the present application;

[0078] Figure 2 is a photovoltaic output scenario provided by an embodiment of the present application;

[0079] Figure 3 is a wind power output scenario provided by an embodiment of the present application;

[0080] Figure 4 is a solution flow of the solution of the present application provided by an embodiment of the present application;

[0081] Figure 5 It is the voltage simulation curve under the AC line fault provided by an embodiment of the present application;

[0082] Figure 6 It is the frequency change situation under 6 scenarios provided by an embodiment of the present application;

[0083] Figure 7 It is the structural schematic diagram of the converter station capacity planning device provided by an embodiment of the present application. Detailed implementation manners

[0084] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system structures and technologies are presented in order to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0085] It should be understood that when used in the specification and the appended claims of the present application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0086] It should also be understood that the term "and / or" used in the specification and the appended claims of the present application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0087] As used in the specification and the appended claims of the present application, the term "if" can be interpreted as "when", "once", "in response to determining", or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if detecting [the described condition or event]" can be interpreted as meaning "once determined", "in response to determining", "once detecting [the described condition or event]", or "in response to detecting [the described condition or event]" according to the context.

[0088] In addition, in the description of the specification and the appended claims of the present application, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0089] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that specific features, structures, or characteristics described in connection with that embodiment are included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized.

[0090] An embodiment of this application provides a method for planning the capacity of a converter station. Figure 1 It is a schematic flowchart of the method for planning the capacity of a converter station provided by an embodiment of this application. Referring to Figure 1 , the detailed description of this method for planning the capacity of a converter station is as follows:

[0091] Step 101: Obtain historical wind power output data and historical photovoltaic power output data, and establish a wind-solar power output model and a multi-source frequency response model.

[0092] Exemplarily, the historical wind power output data and historical photovoltaic power output data are historical data on both the sending end and the receiving end of the target converter station.

[0093] Exemplarily, step 101 may include:

[0094] Obtain historical wind power output data and historical photovoltaic power output data.

[0095] Based on the historical wind power output data and historical photovoltaic power output data, establish a wind-solar power output model.

[0096] Based on the principles of the support of wind power generation, photovoltaic power generation, energy storage system power supply, and DC power supply for the frequency stability of the power grid, obtain a multi-source frequency response model.

[0097] Exemplarily, the wind-solar power output model may be:

[0098] A = {(W,S), L(t) ~ N(μ L ,σ L 2 )}

[0099] where A represents the wind-solar power output model, (W,S) represents the combined power output scenario of wind power output and photovoltaic power output, a set of typical scenarios generated through the joint probability distribution C(u,v) and sampling method, W represents the wind power output, S represents the photovoltaic power output, u represents the marginal distribution function of the wind power output, v represents the marginal distribution function of the photovoltaic power output; L(t) represents the power demand required by the power system, L(t) ~ N(μL , σ L 2 ) indicates that the power demand required by the power system follows a normal distribution, μ L represents the mean value of the power demand required by the power system, and σ L 2 represents the variance of the power demand required by the power system.

[0100] Exemplarily, the establishment process of the wind-solar output model may include:

[0101] 1) Use FrankCopula to define the joint distribution function of wind power and photovoltaic power:

[0102]

[0103] In the formula, u = F W (w) and v = F S (s) are the marginal distribution functions of wind power and photovoltaic power respectively, and θ is the Copula parameter, which controls the dependence of the output.

[0104] 2) Use Latin hypercube sampling to generate samples, and randomly extract N samples (W W (w) and F S (s) to form the wind-solar output curve set X = {(W1, S1), (W2, S2),..., (W i , S i )}. N , S N )}.

[0105] 3) Calculate the Spearman correlation coefficient to quantify the correlation between wind power and photovoltaic power:

[0106]

[0107] where d i is the rank difference between each pair of samples, and n is the number of samples.

[0108] 4) Perform K-means clustering on the output curve set X so that we can obtain typical wind-solar output scenarios C = {C1, C2,..., C k}, assignment stage:

[0109] arg min j ||(W i , S i ) - C j || 2

[0110] Update stage:

[0111]

[0112] Among them, s j is the sample set belonging to the clustering center C j and N j is the number of samples.

[0113] 5) Assume that the electrical load L(t) satisfies a normal distribution at different time periods within a day. Based on the above content, the final wind-solar power output model can be expressed as:

[0114] A = {(W, S), L(t) ~ N(μ L , σ L 2 )}

[0115] The multi-frequency response model can be:

[0116]

[0117] Among them, ΔP W represents the instantaneous change in the wind inertia support power, H represents the inertia system constant, ω represents the change in the system frequency; P S represents the photovoltaic inertia support power, P sref represents the rated output power of the photovoltaic, f(t) represents the instantaneous power of the power grid, f nom = 50Hz, Δf represents the acceptable frequency fluctuation range; P E represents the energy storage inertia support power, P max represents the maximum power that the energy storage system can provide instantaneously, v i represents the instantaneous voltage of the battery, v ref represents the reference voltage or the rated voltage of the power grid, R represents the internal resistance of the battery; P DC represents the DC inertia support power, represents the frequency change rate.

[0118] Exemplarily, since wind power generation cannot adapt to the change in the system frequency, it is necessary to control the virtual synchronous machine to make the inverter simulate the behavior of the synchronous generator. The virtual synchronous machine provides inertia support for the power grid and participates in frequency regulation. When the power grid frequency fluctuates, the virtual synchronous machine can quickly adjust its output power to assist in maintaining the stability of the power grid. Its inertia support is mainly manifested in being able to quickly adjust the rotational speed or output power of the generator according to the change in the power grid frequency.

[0119] The behavior of the virtual synchronous machine participating in the primary frequency modulation response can be expressed as:

[0120] ΔP f = K f ·Δf

[0121] Among them, ΔP fTo control the adjustment amount of power, K f is the adjustment sensitivity coefficient, which is determined according to the amplitude-frequency margin and phase-frequency margin, characterizes the response ability of the controller, and increases or decreases the proportional relationship between power and frequency deviation. Δf is the change in the grid frequency.

[0122] The transient power support ability of the virtual synchronous machine (according to its corresponding control strategy) can be expressed by the following equation:

[0123] P gen = P ref + ΔP W + ΔP f

[0124] where P gen is the generated power of the virtual synchronous machine, and P ref is the rated power of the unit.

[0125] Exemplarily, the transient support power it provides during power system frequency fluctuations is crucial for ensuring the stability of the power grid. Its core principle and calculation method mainly explore the interaction between the regulation of the photovoltaic system output power and frequency changes.

[0126] Exemplarily, the energy storage system (such as a lithium-ion battery) can provide transient support power for the power grid through rapid charge and discharge when the power system frequency fluctuates.

[0127] Exemplarily, the power response of the energy storage system is related to its charge and discharge rate, characterized as:

[0128] P(t) = k(f nom - f(t))

[0129] where P(t) is the instantaneous power output at time t, and k is the adjustment coefficient, representing the sensitivity of power to frequency changes. The energy storage system will quickly adjust its charge and discharge state according to the change of the grid frequency to provide the necessary support power, thus helping to maintain the stability of the power grid.

[0130] The advantage of the battery energy storage system lies in its fast response ability. Generally, the dynamic response time of the battery system is between dozens of milliseconds and several seconds, which enables them to provide transient support relatively quickly. This response ability can be quantified by the following formula:

[0131]

[0132] where ΔP is the power change amount that the energy storage system can provide within a small time interval Δt, E is the available capacity of the battery, and Δt is the response time.

[0133] Exemplarily, the principle of HVDC frequency control considering inertial response is that the HVDC system measures the frequency deviation Δf of the AC power grid and adjusts the HVDC output power ΔP by combining droop control and inertial control links to support the stability of the grid frequency.

[0134] In a DC power grid, the inertia support power is crucial for frequency stability. When the grid frequency fluctuates, the DC system can quickly respond with the help of a DC frequency modulation controller to provide the necessary power support to maintain the stability of the system. This function becomes particularly critical after large-scale grid connection of renewable energy sources such as wind power and solar energy.

[0135] Step 102: Initialize the hyperparameters of the particle swarm algorithm. The position of the particles in the particle swarm algorithm represents the DC current fed from the target converter station into the receiving-end power grid. With the goal of maximizing the DC current fed from the target converter station into the receiving-end power grid and under the conditional constraints of the first constraint, the wind-solar power output model, and the multi-frequency response model, use the particle swarm algorithm to solve for the optimal position of the particles and obtain the first position.

[0136] Exemplarily, the first constraint may include: DC capacity expansion constraint, active power margin constraint, inertia constraint, and DC power constraint.

[0137] The DC capacity expansion constraint can be:

[0138] P DC =P′ DC x + P″ DC y, 0 ≤ x + y ≤ 1

[0139] Where, P′ DC and P″ DC represent the DC capacities in different construction scenarios, x ∈ {0, 1}, y ∈ {0, 1}. When x takes 1, it means the DC capacity built at x is P′ DC , when y takes 1, it means the DC capacity built at y is P″ DC , when x = 0 or y = 0, it means no construction.

[0140] The active power margin constraint can be:

[0141]

[0142] Where, represents the maximum DC power fed into the receiving-end power grid, represents the maximum active power provided by wind power, represents the maximum active power provided by photovoltaic power, represents the maximum active power provided by energy storage, represents the maximum active power provided by thermal power units, and Re represents the reserve capacity coefficient, which is taken as 10% here, Represents the maximum active power required by the load, N W Represents the number of wind turbines, N S Represents the number of PVs, N E Represents the number of energy storages, N L Represents the number of loads, N G Represents the number of thermal power units.

[0143] The inertia constraint can be:

[0144]

[0145] H DCmin ≤H DC ≤H DCmax

[0146] Wherein, H represents the equivalent inertia constant of the combined frequency regulation of DC, wind power, PV and energy storage in the receiving-end power grid, J AC Represents the rotational inertia of the receiving-end power grid AC system. The system rotational inertia should ensure that it is greater than the minimum inertia value for system safety and stability, H DC Represents the inertia support provided by DC; H W Represents the inertia support provided by wind power; H S Represents the inertia support provided by PV; H E Represents the inertia support provided by the energy storage.

[0147] The DC power constraint can be:

[0148] P DCmin ≤P DC ≤P DCmax

[0149] Wherein, P DCmin Represents the minimum value of the DC inertia support power, P DCmax Represents that the maximum value of the DC inertia support power is the minimum and maximum power of DC transmission respectively.

[0150] Exemplarily, the specific range of the active power margin constraint is calculated based on the wind-solar output model; the specific range of the inertia constraint is calculated based on the wind-solar output model and the multi-frequency response model.

[0151] Exemplarily, the annealing algorithm can be used to optimize the particle swarm algorithm, reduce the time for the particle swarm algorithm to obtain the first position, and improve the calculation efficiency.

[0152] Step 103, with the goal of minimizing the comprehensive cost of the target converter station, using the DC quantity corresponding to the first position, the second constraint, the wind-solar output model and the multi-frequency response model as conditional constraints, and solving using the second-order cone relaxation algorithm to obtain the first result.

[0153] Exemplarily, the comprehensive cost of the target converter station can be:

[0154]

[0155] Among them, F2 represents the comprehensive cost of the target converter station, c represents the cash conversion coefficient, and Y E represents the operating cost of the energy storage device per unit power, and Y G represents the operating cost of the thermal power device per unit power, and W E represents the maintenance cost of the energy storage device per unit power, and W G represents the maintenance cost of the thermal power device per unit power, and P t E represents the energy storage output at time t, and P t G represents the thermal power output at time t, and N represents the total number of thermal power devices and energy storage devices.

[0156] Exemplarily, the second constraint may include: power balance constraint, unit output constraint, reserve constraint, transmission limit power constraint of the receiving-end grid line, and frequency change rate constraint under DC bipolar blocking.

[0157] The power balance constraint may be:

[0158]

[0159] Among them, P DC represents the DC power fed into the receiving-end grid, P W represents the active power provided by wind power, P S represents the active power provided by photovoltaic power, P E represents the active power provided by the energy storage, P G represents the active power provided by the thermal power unit, P L represents the active power required by the load, N W represents the number of wind turbines, N S represents the number of photovoltaic units, N E represents the number of energy storage devices, N L represents the number of loads, N G represents the number of thermal power units.

[0160] The unit output constraint may be:

[0161]

[0162] Among them, P i represents any one of the DC power fed into the receiving-end grid at time i, the active power provided by wind power at time i, and the active power provided by the energy storage at time i, P i max represents P i the maximum value of, P imax Denote the minimum value of P i , and ΔP U denotes the upward regulation reserve, and ΔP D denotes the downward regulation reserve.

[0163] The reserve constraint can be:

[0164]

[0165] where ΔP U,i denotes the active power deficit at time i, PΔ denotes the active power required for system frequency regulation when power deviation occurs in the system, and DC, wind power, photovoltaic power, and energy storage can provide sufficient active power support at any time.

[0166] The transmission limit power constraint of the receiving-end grid line can be:

[0167]

[0168] where PT is the power transmission transfer factor, C = G + W + E + S + L, SD denotes the DC connection factor, SG denotes the thermal power unit connection factor, SW denotes the wind turbine connection factor, SE denotes the energy storage connection factor, SS denotes the photovoltaic connection factor, SL denotes the connection factor of the load accessing the receiving-end grid; G denotes the number of thermal power units connected, W denotes the number of wind turbines connected, E denotes the number of energy storage connected, S denotes the number of photovoltaic connected, and L denotes the number of loads connected; denotes the active power transmission limit value of the receiving-end grid AC grid.

[0169] The frequency change rate constraint under DC bipolar blocking can be:

[0170]

[0171] where f ref denotes the reference frequency, H' denotes the inertia constant of the combined frequency regulation of DC, wind power, photovoltaic power, and energy storage, and RoCoF max denotes the maximum allowable frequency change rate of the system.

[0172] Exemplarily, the specific ranges of the power balance constraint and the transmission limit power constraint of the receiving-end grid line are calculated based on the wind-solar output model; the specific range of the frequency change rate constraint under DC bipolar blocking is calculated based on the wind-solar output model and the multi-source frequency response model.

[0173] Step 104, if the first result meets the convergence condition, output the DC quantity corresponding to the first position and the first result to obtain the DC quantity and cost fed into the receiving-end grid by the target converter station.

[0174] Exemplarily, the method may further include:

[0175] If the first result does not meet the convergence condition, return to the step of using the particle swarm optimization algorithm to solve the optimal position of the particles, obtain the sub-optimal position, and use the direct current amount fed by the target converter station corresponding to the sub-optimal position into the receiving-end power grid as the second result.

[0176] With the goal of minimizing the comprehensive cost and using the second result, the second constraint, the wind-solar power output model, and the multi-frequency response model as conditional constraints, use the second-order cone relaxation algorithm to solve and obtain the third result.

[0177] If the third result meets the convergence condition, output the second result and the third result to obtain the direct current amount fed by the target converter station into the receiving-end power grid and the cost.

[0178] If the third result does not meet the convergence condition, return to the step of using the particle swarm optimization algorithm to solve the optimal position of the particles until the convergence condition is met.

[0179] For the above converter station capacity planning method, first, establish the wind-solar power output model and the multi-frequency response model through historical wind power output data and historical photovoltaic power output data. Then, with the goal of maximizing the direct current amount fed by the target converter station into the receiving-end power grid and using the first constraint, the wind-solar power output model, and the multi-frequency response model as conditional constraints, use the particle swarm optimization algorithm to solve the optimal position of the particles to obtain the first position. After that, with the goal of minimizing the comprehensive cost of the target converter station and using the direct current amount corresponding to the first position, the second constraint, the wind-solar power output model, and the multi-frequency response model as conditional constraints, use the second-order cone relaxation algorithm to solve and obtain the first result. Finally, if the first result meets the convergence condition, output the direct current amount corresponding to the first position and the first result to obtain the direct current amount fed by the target converter station into the receiving-end power grid and the cost. This solution comprehensively considers the influence of the wind-solar power output model and the multi-frequency response model on the converter station. At the same time, with the goal of maximizing the direct current amount fed by the target converter station into the receiving-end power grid and minimizing the comprehensive cost, it can improve the cost performance of building the converter station. The converter station built according to the results of this solution has higher dispatching capabilities and flexibility, can ensure the reliability, economy, and stability of power supply, and meet the needs of large-scale new energy grid connection.

[0180] Exemplarily, in some embodiments, when conducting system planning, it is crucial to depict the uncertainty of new energy output through scenario analysis methods, which include methods such as time series simulation method, typical day method, and scenario clustering method. The scenario clustering method is particularly worth mentioning. It accurately depicts the output characteristics through scenario probabilities and is very suitable for medium- and long-term planning of power systems. When applying the scenario clustering method, we considered the complementarity and correlation of wind energy and solar energy output and selected the Frank Copula function as the connection function for the joint probability distribution of wind power and photovoltaic power. Through the Latin hypercube sampling technique, we generated correlated wind-solar output curves and used the Spearman coefficient to quantitatively analyze the correlation magnitude of wind-solar output. Based on the K-means method, we further carried out scenario reduction to obtain typical wind-solar output scenarios. Similarly, the electricity load also exhibits volatility and randomness, and the electricity load at different times of the day generally follows the normal distribution law. The scenario generation steps for wind power and photovoltaic power generation are as described below. The wind power and photovoltaic output curves after reduction to 500 scenarios are respectively shown in Figure 2 and Figure 3 .

[0181] The solution process is as shown in Figure 4 . First, collect historical wind-solar output data and perform fitting to determine the scenarios for simulation operation. Then, initialize the calculation parameters, set the iteration number k = 0, and given the initial upper limit U = ∞ and lower limit L = -∞. The optimization objective is to minimize the total cost of building AC and DC lines (including construction cost and operation cost), and the decision variables are whether to newly build the k-th AC line and the branch feeding or DC single-end feeding of the DC line between nodes. Use the improved particle swarm algorithm to solve the upper-layer optimization problem. The objective function of the lower-layer optimization problem is to minimize the load shedding amount after the DC bipolar blocking fault, and the decision variable is the load shedding amount after the n-th DC bipolar blocking. Use the second-order cone relaxation optimization problem to solve. Set the convergence limit as ξ, and judge whether the upper bound solution U and the lower bound solution L satisfy the convergence condition U - L < ξ. If satisfied, end the calculation to obtain the DC landing point planning scheme; if not satisfied, return to the upper-layer optimization problem.

[0182] Conduct three-phase short-circuit fault verification on the transmission lines in the IEEE-39 bus system. The verification results show that after the three-phase short-circuit fault of the line, the wind turbines are disconnected from the grid due to transient overvoltage. The risk of wind power disconnection (wind power disconnection probability is 10.42%) after the three-phase short-circuit fault of the line in the comparison model is much higher than that of the proposed model (wind power disconnection probability is 2.04%). In order to further verify the transient voltage stability of the power grid after the AC line fault, set a three-phase short-circuit fault on the AC line near the DC landing point. The fault occurs at 4 seconds and is cleared 0.4 seconds later. The transient voltage simulation curves of the proposed model and the comparison model are as shown in Figure 5 .

[0183] Set scenarios 1 to 6. For scenarios 1 to 3, the proposed scheme is adopted. The fault is the bipolar blocking of DC 3, and the load shedding amounts are 415.2 MW, 445.2 MW, and 475.2 MW respectively. For scenarios 4 to 6, the comparative scheme is adopted. The fault is also the bipolar blocking of DC 3, and the load shedding amounts are 415.2 MW, 445.2 MW, and 475.2 MW respectively. From Figure 6 It can be seen that the frequency does not drop below 49.5 Hz after the bipolar blocking fault of DC 3 in scenarios 1 to 5, and can recover to a relatively high level; the frequency becomes unstable after the bipolar blocking of DC 3 in scenario 6. This phenomenon indicates that considering the frequency stability constraint in the model can improve the frequency support ability of the AC power grid for DC after the implementation of the DC landing point planning scheme.

[0184] The IEEE 33-node distribution system is used as an example for analysis. The system includes 39 nodes, 46 transmission lines, and one UHV DC. The generator at node 31 is a balancing machine. The active power gap is set to 2000 MW, the single-terminal DC capacity is 1000 MW, and the capacity of each end of the split-terminal DC is 500 MW.

[0185] Step 1: Model calculation results

[0186] By solving the DC landing point planning model and the comparative model, the following two DC landing point planning schemes are obtained. The details of the first scheme are: the access mode of DC 2 is single-terminal access, and the landing point is node 3; the access mode of DC 3 is split-terminal access, and the landing points are node 16 and node 20; the lines to be built are line 3-4, line 15-16, and line 16-17; the total planned cost is 277.113 million yuan; the minimum load shedding amount after the bipolar blocking of the DC is 415.2 MW. The details of the second scheme are: the access mode of DC 2 is single-terminal access, and the landing point is node 3; the access mode of DC 3 is single-terminal access, and the landing point is node 4; the lines to be built are line 3-4 and line 4-14; the total planned cost is 255.6885 million yuan; the minimum load shedding amount after the bipolar blocking of the DC is 0 MW (the frequency stability constraint is not considered in the model). Perform stability verification and comparative analysis on the two planning schemes obtained from the DC landing point planning model and the comparative model.

[0187] Step 2: Static voltage stability simulation verification

[0188] The multi-infeed short-circuit ratio is adopted as the verification index for the static voltage stability of the system. A multi-infeed short-circuit ratio greater than 3 is the basis for judging that the AC power grid has strong voltage support ability. According to the calculation results of the DC landing point planning model and the comparison model, the corresponding IEEE-39 bus system is built and the multi-infeed short-circuit ratios of each DC are calculated. The multi-infeed short-circuit ratios of the proposed models for DC1, DC2, DC3, and DC3 (terminal 2) are 3.1952, 3.1572, 3.2627, and 3.5742 respectively. The multi-infeed short-circuit ratios of the comparison models for DC1, DC2, and DC3 are 3.3732, 2.9679, and 3.7265 respectively.

[0189] It can be seen from the above results that the multi-infeed short-circuit ratios of each DC in the IEEE-39 bus system built according to the proposed model are all greater than 3, indicating that the AC power grid has strong voltage support ability; while the multi-infeed short-circuit ratio of DC2 in the comparison model is less than 3, not meeting the requirement that the AC power grid has strong voltage support ability. The reason for this difference is that the multi-infeed short-circuit ratio constraint is not included in the comparison model. Therefore, incorporating the multi-infeed short-circuit ratio constraint into the model can enhance the voltage support ability of the receiving-end power grid and improve the static voltage stability of the power grid.

[0190] Step 3: Transient voltage stability simulation verification

[0191] The planning schemes obtained from the DC landing point planning model and the comparison model are subjected to transient voltage stability verification. The verification method is the impact of a three-phase short-circuit fault on the transient voltage stability of the AC system. The transmission lines in the IEEE-39 bus system are subjected to three-phase short-circuit fault verification. The verification results show that after the three-phase short-circuit fault of the line, wind turbines trip due to transient overvoltage. The risk of wind power tripping (wind power tripping probability is 10.42%) after the three-phase short-circuit fault of the line in the DC planning scheme obtained from the comparison model is much higher than that of the proposed model (wind power tripping probability is 2.04%). In order to further verify the transient voltage stability of the power grid after the AC line fault, a three-phase short-circuit fault is set for the AC line in the near area of the DC landing point. The fault occurs at 4 seconds and is cleared 0.4 seconds later. The transient voltage simulation curves of the proposed model and the comparison model are as Figure 5 shown (the solid line in the figure represents the comparison model, and the dashed line represents the DC landing point planning model).

[0192] Step 4: Frequency stability simulation verification

[0193] In the context of the increasing number and capacity of DC feed-ins, DC bipolar blocking has become the N-2 fault that causes the largest active power loss in the receiving-end power grid, and the resulting frequency problem is the most serious. Therefore, DC bipolar blocking is used as a test fault to verify the frequency support ability of the receiving-end power grid. Simulate the frequency curve of the DC landing point planning scheme after the test fault. Set scenarios 1 to 6 for frequency simulation. Scenarios 1 to 3 all adopt the proposed scheme, the fault is DC 3 bipolar blocking, and the load shedding amounts are 415.2 MW, 445.2 MW, and 475.2 MW respectively. Scenarios 4 to 6 all adopt the comparative scheme, the fault is DC 3 bipolar blocking, and the load shedding amounts are 415.2 MW, 445.2 MW, and 475.2 MW respectively. From Figure 6 It can be seen that the frequencies in Scenarios 1 to 5 do not drop below 49.5 Hz after the DC 3 bipolar blocking fault and can recover to a relatively high level; the frequency in Scenario 6 becomes unstable after the DC 3 bipolar blocking. This phenomenon indicates that considering the frequency stability constraint in the model can improve the frequency support ability of the AC power grid for DC after the implementation of the DC landing point planning scheme.

[0194] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0195] Corresponding to the converter station capacity planning method described in the above embodiments, Figure 7 The structural block diagram of the converter station capacity planning device provided by the embodiment of the present application is shown. For the sake of convenience of description, only the parts related to the embodiment of the present application are shown.

[0196] See Figure 7 , the converter station capacity planning device in the embodiment of the present application may include:

[0197] A data acquisition module 201, configured to acquire historical wind power output data and historical photovoltaic power output data, and establish a wind-solar power output model and a multi-source frequency response model.

[0198] A first operation module 202, configured to initialize the hyperparameters of the particle swarm algorithm. The position of the particle in the particle swarm algorithm represents the DC quantity fed into the receiving-end power grid by the target converter station; aiming at maximizing the DC quantity fed into the receiving-end power grid by the target converter station, and taking the first constraint, the wind-solar power output model and the multi-source frequency response model as conditional constraints, use the particle swarm algorithm to solve the optimal position of the particle to obtain the first position.

[0199] A second operation module 203, configured to aim at minimizing the comprehensive cost of the target converter station, and taking the DC quantity corresponding to the first position, the second constraint, the wind-solar power output model and the multi-source frequency response model as conditional constraints, and use the second-order cone relaxation algorithm to solve to obtain a first result.

[0200] The result output module 204 is configured to output the direct current quantity corresponding to the first position and the first result if the first result meets the convergence condition, so as to obtain the direct current quantity and cost fed by the target converter station into the receiving-end power grid.

[0201] Exemplarily, the result output module 204 can also be configured to:

[0202] if the first result does not meet the convergence condition, return to the step of solving the optimal position of the particle using the particle swarm algorithm to obtain the sub-optimal position, and use the direct current quantity fed by the target converter station corresponding to the sub-optimal position into the receiving-end power grid as the second result.

[0203] With the goal of minimizing the comprehensive cost, and with the second result, the second constraint, the wind-solar power output model, and the multi-frequency response model as conditional constraints, solve using the second-order cone relaxation algorithm to obtain the third result.

[0204] if the third result meets the convergence condition, output the second result and the third result to obtain the direct current quantity and cost fed by the target converter station into the receiving-end power grid.

[0205] if the third result does not meet the convergence condition, re-return to the step of solving the optimal position of the particle using the particle swarm algorithm until the convergence condition is met.

[0206] Exemplarily, the first constraint may include: DC capacity expansion constraint, active power margin constraint, inertia constraint, and DC power constraint.

[0207] The DC capacity expansion constraint can be:

[0208] P DC =P′ DC x+P″ DC y

[0209] 0≤x+y≤1

[0210] where, P′ DC and P″ DC represent the DC capacities under different construction scenarios, x∈{0,1}, y∈{0,1}, when x takes 1, it means the DC capacity built at x is P′ DC , when y takes 1, it means the DC capacity built at y is P″ DC , when x = 0, or y = 0, it means no construction.

[0211] The active power margin constraint can be:

[0212]

[0213] where, represents the maximum DC power fed into the receiving-end power grid, represents the maximum active power provided by the wind power, represents the maximum active power provided by the photovoltaic represents the maximum active power provided by the energy storage represents the maximum active power provided by the thermal power unit, and Re represents the reserve capacity factor, here taking 10% represents the maximum active power required by the load, N W represents the number of wind turbines, N S represents the number of photovoltaics, N E represents the number of energy storages, N L represents the number of loads, N G represents the number of thermal power units.

[0214] The inertia constraint can be:

[0215]

[0216] H DCmin ≤H DC ≤H DCmax

[0217] wherein, H represents the equivalent inertia constant of the combined frequency regulation of DC, wind power, photovoltaic and energy storage in the receiving-end power grid, J AC represents the rotational inertia of the receiving-end power grid AC system. The system rotational inertia should ensure that it is greater than the minimum inertia value for system safety and stability, H DC represents the inertia support provided by the DC; H W represents the inertia support provided by the wind power; H S represents the inertia support provided by the photovoltaic; H E represents the inertia support provided by the energy storage.

[0218] The DC power constraint can be:

[0219] P DCmin ≤P DC ≤P DCmax

[0220] wherein, P DCmin represents the minimum value of the DC inertia support power, P DCmax represents the maximum value of the DC inertia support power, which are the minimum and maximum powers of DC transmission respectively.

[0221] Exemplarily, the specific range of the active power margin constraint is calculated based on the wind-solar output model; the specific range of the inertia constraint is calculated based on the wind-solar output model and the multi-frequency response model.

[0222] Exemplarily, the second constraint may include: power balance constraint, unit output constraint, reserve constraint, transmission limit power constraint of the receiving-end power grid line, and frequency change rate constraint under DC bipolar blocking.

[0223] The power balance constraint can be:

[0224]

[0225] Among them, P DC represents the DC power fed into the receiving-end power grid, P W represents the active power provided by wind power, P S represents the active power provided by photovoltaic power, P E represents the active power provided by energy storage, P G represents the active power provided by thermal power units, P L represents the active power required by the load, N W represents the number of wind turbines, N S represents the number of photovoltaic units, N E represents the number of energy storage units, N L represents the number of loads, N G represents the number of thermal power units.

[0226] The unit output constraint can be:

[0227] P i +ΔP U ≤P i max

[0228] P i -ΔP D ≤P i min

[0229] |P i -P i-1 |≤ΔP

[0230]

[0231] Among them, P i represents any one of the DC power fed into the receiving-end power grid at time i, the active power provided by wind power at time i, and the active power provided by energy storage at time i, P i max represents P i the maximum value of, P i max represents P i the minimum value of, ΔP U represents the upward reserve, ΔP D represents the downward reserve.

[0232] The reserve constraint can be:

[0233]

[0234] Among them, ΔP U,irepresents the active power deficit at time i, and PΔ represents the active power required for system frequency regulation when power deviation occurs in the system. DC, wind power, photovoltaic power, and energy storage can provide sufficient active power support at any time.

[0235] The transmission limit power constraint of the receiving-end grid line can be:

[0236]

[0237] Among them, PT is the power transmission transfer factor, C = G + W + E + S + L, SD represents the DC connection factor, SG represents the thermal power unit connection factor, SW represents the wind turbine connection factor, SE represents the energy storage connection factor, SS represents the photovoltaic connection factor, and SL represents the connection factor of the load accessing the receiving-end grid; G represents the number of thermal power units connected, W represents the number of wind turbines connected, E represents the number of energy storage units connected, S represents the number of photovoltaics connected, and L represents the number of loads connected. represents the active power transmission limit value of the receiving-end grid AC grid.

[0238] The frequency change rate constraint under DC bipolar blocking can be:

[0239] P DCmin ≤ P DC ≤ P DCmax

[0240]

[0241] Among them, f ref represents the reference frequency, H' represents the inertia constant of the combined frequency regulation of DC, wind power, photovoltaic power, and energy storage, and RoCoF max represents the maximum allowable frequency change rate of the system.

[0242] Exemplarily, the specific ranges of the power balance constraint and the transmission limit power constraint of the receiving-end grid line are calculated based on the wind-solar output model; the specific range of the frequency change rate constraint under DC bipolar blocking is calculated based on the wind-solar output model and the multi-source frequency response model.

[0243] Exemplarily, the comprehensive cost of the target converter station can be:

[0244]

[0245] Among them, F2 represents the comprehensive cost of the target converter station, c represents the cash conversion coefficient, Y E represents the operating cost per unit power of the energy storage device, Y G represents the operating cost per unit power of the thermal power device, W E represents the maintenance cost per unit power of the energy storage device, W G represents the maintenance cost per unit power of the thermal power device, Pt E represents the energy storage output at time t, P t G represents the thermal power output at time t, and N represents the total number of thermal power equipment and energy storage equipment.

[0246] Exemplarily, the data acquisition module 201 can be used to:

[0247] Obtain historical wind power output data and historical photovoltaic power output data.

[0248] Based on the historical wind power output data and historical photovoltaic power output data, establish a wind-solar power output model.

[0249] Based on the support principles of wind power generation, photovoltaic power generation, energy storage system power supply, and DC power supply for the grid frequency stability, obtain a multi-source frequency response model.

[0250] Exemplarily, the wind-solar power output model can be:

[0251] A = {(W, S), L(t) ~ N(μ L , σ L 2 )}

[0252] where A represents the wind-solar power output model, (W, S) represents the combined output scenario of wind power output and photovoltaic power output, a set of typical scenarios generated by the joint probability distribution C(u, v) and sampling method, W represents the wind power output, S represents the photovoltaic power output, u represents the marginal distribution function of wind power output, v represents the marginal distribution function of photovoltaic power output; L(t) represents the power demand required by the power system, L(t) ~ N(μ L , σ L 2 ) means that the power demand required by the power system satisfies a normal distribution, μ L represents the mean value of the power demand required by the power system, and σ L 2 represents the variance of the power demand required by the power system.

[0253] The multi-source frequency response model can be:

[0254]

[0255] where ΔP W represents the instantaneous change in the wind inertia support power, H represents the inertia system constant, and ω represents the change in the system frequency; P S represents the photovoltaic inertia support power, P sref represents the rated output power of the photovoltaic, f(t) represents the instantaneous power of the power grid, f nom = 50Hz, and Δf represents the acceptable frequency fluctuation range; PE Indicates the energy storage inertia support power, P max Indicates the maximum power that the energy storage system can instantaneously provide, v i Indicates the instantaneous voltage of the battery, v ref Indicates the reference voltage or the rated voltage of the power grid, and R represents the internal resistance of the battery; P DC Indicates the DC inertia support power, Indicates the rate of change of frequency.

[0256] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0257] Those of ordinary skill in the art can realize that the templates, units, and algorithm steps of the various examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0258] If the module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes of the above method embodiments of the present invention, it can also be completed by a computer program instructing the relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments for capacity planning of each converter station can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0259] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for capacity planning of a converter station, characterized in that Including: Obtain historical wind power output data and historical photovoltaic power output data, and establish a wind-solar power output model and a multi-frequency response model; Initialize the hyperparameters of the particle swarm optimization algorithm. The position of the particle in the particle swarm optimization algorithm represents the direct current quantity fed by the target converter station into the receiving-end power grid. Taking the maximum direct current quantity fed by the target converter station into the receiving-end power grid as the objective, and using the first constraint, the wind-solar power output model and the multi-frequency response model as conditional constraints, use the particle swarm optimization algorithm to solve the optimal position of the particle to obtain the first position; Taking the minimum comprehensive cost of the target converter station as the objective, and using the direct current quantity corresponding to the first position, the second constraint, the wind-solar power output model and the multi-frequency response model as conditional constraints, use the second-order cone relaxation algorithm to solve and obtain the first result; If the first result meets the convergence condition, output the direct current quantity corresponding to the first position and the first result to obtain the direct current quantity and cost fed by the target converter station into the receiving-end power grid.

2. The method for planning the capacity of a converter station according to claim 1, characterized in that The method further includes: If the first result does not meet the convergence condition, return to the step of using the particle swarm optimization algorithm to solve the optimal position of the particle to obtain the sub-optimal position, and use the direct current quantity fed by the target converter station into the receiving-end power grid corresponding to the sub-optimal position as the second result; Taking the minimum comprehensive cost as the objective, and using the second result, the second constraint, the wind-solar power output model and the multi-frequency response model as conditional constraints, use the second-order cone relaxation algorithm to solve and obtain the third result; If the third result meets the convergence condition, output the second result and the third result to obtain the direct current quantity and cost fed by the target converter station into the receiving-end power grid; If the third result does not meet the convergence condition, return to the step of using the particle swarm optimization algorithm to solve the optimal position of the particle again until the convergence condition is met.

3. The converter station capacity planning method according to claim 1, wherein The first constraint includes: DC capacity expansion constraint, active power margin constraint, inertia constraint and DC power constraint; The DC capacity expansion constraint is: P DC = P D ' C x + P D ” C y 0 ≤ x + y ≤ 1 Among them, P D ' C and P D ” C represent the DC capacities under different investment and construction scenarios, where x ∈ {0, 1}, y ∈ {0, 1}. When x takes 1, it means the DC capacity invested and constructed at x is P D ' C ; when y takes 1, it means the DC capacity invested and constructed at y is P D ” C ; when x = 0 or y = 0, it means no investment and construction The active power margin constraint is: Among them, represents the maximum DC power fed into the receiving-end power grid, represents the maximum active power provided by wind power, represents the maximum active power provided by photovoltaic power, represents the maximum active power provided by energy storage, represents the maximum active power provided by thermal power units. Re represents the reserve capacity factor, which is taken as 10% here, represents the maximum active power required by the load, N W represents the number of wind turbines, N S represents the number of photovoltaic panels, N E represents the number of energy storage units, N L represents the number of loads, N G represents the number of thermal power units; The inertia constraint is: H DCmin ≤H DC ≤H DCmax Among them, H represents the equivalent inertia constant of the combined frequency regulation of DC, wind power, photovoltaic, and energy storage in the receiving-end power grid, and J AC represents the rotational inertia of the AC system in the receiving-end power grid. The system's rotational inertia should ensure that it is greater than the minimum inertia value for system safety and stability. H DC represents the inertia support provided by the DC; H W represents the inertia support provided by the wind power; H S represents the inertia support provided by the photovoltaic; H E represents the inertia support provided by the energy storage; The DC power constraint is: P DCmin ≤P DC ≤P DCmax Among them, P DCmin represents the minimum value of the DC inertia support power, and P DCmax represents the maximum value of the DC inertia support power, which are the minimum and maximum powers of the DC transmission respectively.

4. The method for planning the capacity of a converter station according to claim 3, wherein The specific range of the active power margin constraint is calculated based on the wind-solar power output model; the specific range of the inertia constraint is calculated based on the wind-solar power output model and the multi-frequency response model.

5. The method for planning the capacity of a converter station according to claim 1, wherein The second constraint includes: power balance constraint, unit output constraint, reserve constraint, transmission limit power constraint of the receiving-end power grid line and frequency change rate constraint under DC bipolar blocking; The power balance constraint is: Among them, P DC represents the DC power fed into the receiving-end power grid, P W represents the active power provided by wind power, P S represents the active power provided by photovoltaic power, P E represents the active power provided by energy storage, P G represents the active power provided by thermal power units, P L represents the active power required by the load, N W represents the number of wind turbines, N S represents the number of photovoltaic units, N E represents the number of energy storage units, N L represents the number of loads, N G represents the number of thermal power units; The unit output constraint is: P i +ΔP U ≤P i max P i -ΔP D ≤P i min |P i -P i-1 |≤ΔP Among them, P i represents any one of the DC power fed into the receiving-end power grid at time i, the active power provided by the wind power at time i, and the active power provided by the energy storage at time i. P i max represents the maximum value of P i , and P i max represents the minimum value of P i . ΔP U represents the upward reserve, and ΔP D represents the downward reserve; The reserve constraint is: Among them, ΔP U,i represents the active power deficit at the i-th moment, PΔ represents the active power required for system frequency regulation when power deviation occurs in the system, and DC, wind power, photovoltaic, and energy storage can provide sufficient active power support at any moment; The transmission limit power constraint of the receiving-end power grid line is: Among them, PT is the power transmission transfer factor, C = G + W + E + S + L, SD represents the DC connection factor, SG represents the thermal power unit connection factor, SW represents the wind turbine connection factor, SE represents the energy storage connection factor, SS represents the PV connection factor, and SL represents the connection factor of the load accessing the receiving-end power grid; G represents the number of thermal power units connected, W represents the number of wind turbines connected, E represents the number of energy storage units connected, S represents the number of PVs connected, and L represents the number of loads connected. represents the active power transmission limit value of the receiving-end AC power grid; The frequency change rate constraint under DC bipolar blocking is: P DCmin ≤P DC ≤P DCmax Among them, f ref represents the reference frequency, H' represents the inertia constant of the combined frequency regulation of DC, wind power, photovoltaic and energy storage, and RoCoF max represents the maximum allowable frequency change rate of the system.

6. The converter station capacity planning method according to claim 5, characterized in that The specific ranges of the power balance constraint and the transmission limit power constraint of the receiving-end power grid line are calculated based on the wind-solar power output model; the specific range of the frequency change rate constraint under DC bipolar blocking is calculated based on the wind-solar power output model and the multi-frequency response model.

7. The method for planning the capacity of a converter station according to claim 1, characterized in that, The comprehensive cost of the target converter station is: Among them, F2 represents the comprehensive cost of the target converter station, c represents the cash conversion coefficient, Y E represents the operating cost of the energy storage device per unit power, Y G represents the operating cost of the thermal power device per unit power, W E represents the maintenance cost of the energy storage device per unit power, W G represents the maintenance cost of the thermal power device per unit power, P t E represents the energy storage output at time t, P t G represents the thermal power output at time t, and N represents the total number of thermal power devices and energy storage devices.

8. The method for capacity planning of a converter station according to claim 1, wherein The obtaining of the historical wind power output data and historical photovoltaic power output data, and the establishment of the wind-solar power output model and the multi-frequency response model include: Obtain historical wind power output data and historical photovoltaic power output data; Based on the historical wind power output data and the historical photovoltaic power output data, a wind-solar power output model is established; Based on the support principles of wind power generation, photovoltaic power generation, energy storage system power supply, and DC power supply for the grid frequency stability, the multi-source frequency response model is obtained.

9. The converter station capacity planning method according to claim 8, wherein The wind-solar power output model is: A = {(W, S), L(t) ~ N(μ L , σ L 2 )} Among them, A represents the wind-solar power output model, (W, S) represents the combined output scenario of wind power output and photovoltaic power output, a set of typical scenarios generated by the joint probability distribution C(u, v) and sampling method, W represents wind power output, S represents photovoltaic power output, u represents the marginal distribution function of wind power output, and v represents the marginal distribution function of photovoltaic power output; L(t) represents the power demand required by the power system, L(t) ~ N(μ L , σ L 2 ) indicates that the power demand required by the power system satisfies a normal distribution, μ L represents the mean value of the power demand required by the power system, and σ L 2 represents the variance of the power demand required by the power system; The multi-source frequency response model is: Among them, ΔP W represents the instantaneous change in wind power inertial support power, H represents the inertial system constant, and ω represents the change in system frequency; P S represents the photovoltaic inertial support power, P sref represents the rated output power of the photovoltaic, f(t) represents the instantaneous power of the power grid, f nom = 50Hz, Δf represents the acceptable frequency fluctuation range; P E represents the energy storage inertial support power, P max represents the maximum power that the energy storage system can provide instantaneously, v i represents the instantaneous voltage of the battery, v ref represents the reference voltage or the rated voltage of the power grid, and R represents the internal resistance of the battery; P DC represents the DC inertial support power, represents the rate of change of frequency.

10. A converter station capacity planning device, characterized in that, It includes: A data acquisition module, configured to acquire historical wind power output data and historical photovoltaic power output data, and establish a wind-solar power output model and a multi-source frequency response model; A first operation module, configured to initialize the hyperparameters of the particle swarm algorithm, where the position of the particle in the particle swarm algorithm represents the DC current fed by the target converter station into the receiving-end grid; aiming at the maximum DC current fed by the target converter station into the receiving-end grid, and taking the first constraint, the wind-solar power output model, and the multi-source frequency response model as conditional constraints, using the particle swarm algorithm to solve the optimal position of the particle, and obtaining the first position; A second operation module, configured to aim at minimizing the comprehensive cost of the target converter station, and taking the DC current corresponding to the first position, the second constraint, the wind-solar power output model, and the multi-source frequency response model as conditional constraints, and using the second-order cone relaxation algorithm to solve, and obtaining the first result; A result output module, configured to output the DC current corresponding to the first position and the first result if the first result meets the convergence condition, and obtain the DC current and cost fed by the target converter station into the receiving-end grid.

Images