A Two-Layer Optimal Method for Location and Capacity Determination of Hybrid Energy Storage Considering Node Frequency Distribution

Abstract

The present invention relates to the technical field of energy storage frequency modulation, and discloses a hybrid energy storage site selection and capacity determination double-layer optimization method considering node frequency distribution, including the following steps: S1. Establish a hybrid energy storage optimization site selection and capacity determination double-layer model; S2. Solve the hybrid energy storage site selection and capacity determination double-layer model; S3. Analyze the effectiveness of different schemes. The S1 step specifically includes the following steps: S1.1. Establish a hybrid energy storage site selection layer model; S1.2. Establish a hybrid energy storage capacity determination layer model. The S1.1 step specifically includes the following steps: S1.11. Establish a network node frequency fluctuation calculation model; 12. Determine the objective function of the site selection layer. By establishing a hybrid energy storage site selection and capacity determination double-layer model; constructing a hybrid energy storage double-layer configuration model at the capacity determination level, the problems of system frequency instability, unbalanced fluctuation distribution caused by large-scale grid connection of new energy and the application of high-proportion power electronic devices, and the impact of large-scale centralized hybrid energy storage grid connection on system power flow are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage frequency modulation, and particularly to a double-layer optimization method for hybrid energy storage site selection and capacity determination considering node frequency distribution. Background Technique

[0002] In the past few decades, the global energy structure has changed significantly. Traditional fossil fuel energy is gradually being replaced by clean and renewable energy. The rapid development of renewable energy such as wind energy and solar energy not only reduces carbon emissions but also plays a positive role in environmental protection. However, the high volatility and uncertainty of renewable energy have brought new challenges to the stable operation of the power system.

[0003] At the same time, the extensive application of power electronics technology has further promoted the development of the power system but also caused new problems. The large-scale grid connection of renewable energy and the high proportion of power electronic devices in the power system have led to a decrease in the system inertia level. The volatility and uncertainty of new energy have brought frequency stability impacts and uneven power flow distribution to the system. Large-scale centralized hybrid energy storage, as an excellent frequency modulation resource, can buffer the adverse effects caused by renewable energy in the system. However, after the energy storage is incorporated into the system, its high charging and discharging power changes the nearby power flow distribution, resulting in node voltage over-limit and increased line load rate. In addition, the large-scale grid connection of new energy and the high proportion of power electronic devices in the system also cause uneven system inertia distribution, leading to uneven node frequency distribution. There is little consideration of the spatial inhomogeneity of node frequency changes in existing hybrid energy storage planning methods.

[0004] Therefore, aiming at the deficiencies of the existing technology, the present invention proposes a double-layer optimization method for hybrid energy storage site selection and capacity determination considering node frequency distribution. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the present invention provides a double-layer optimization method for hybrid energy storage site selection and capacity determination considering node frequency distribution, which solves the problems of system frequency stability and uneven power flow distribution after the centralized hybrid energy storage is connected to the grid.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A double-layer optimization method for hybrid energy storage site selection and capacity determination considering node frequency distribution, comprising the following steps:

[0007] S1. Establish a double-layer model for hybrid energy storage optimization site selection and capacity determination;

[0008] S2. Solve the double-layer model for hybrid energy storage site selection and capacity determination;

[0009] S3. Analyze the effectiveness of different schemes.

[0010] Preferably, the S1 step specifically includes the following steps:

[0011] S1.1. Establish a model for the location layer of hybrid energy storage;

[0012] S1.2. Establish a model for the capacity determination layer of hybrid energy storage.

[0013] Preferably, the step S1.1 specifically includes the following steps:

[0014] S1.11. Establish a calculation model for the frequency fluctuation of network nodes;

[0015] S1.12. Determine the objective function of the location layer;

[0016] S1.13. Determine the constraint conditions of the location layer;

[0017] S1.14. Calculate the line power and the Theil entropy of the node voltage frequency.

[0018] Preferably, the step S.12 specifically includes the following steps:

[0019] S1.21. Determine the objective function of the capacity determination layer;

[0020] S1.22. Determine the constraint conditions of the capacity determination layer;

[0021] S1.23. Establish a control strategy model for hybrid energy storage.

[0022] Preferably, the step S2 specifically includes the following steps:

[0023] S2.1. Initialize the configured capacity of hybrid energy storage;

[0024] S2.2. Substitute the initialized configured capacity of hybrid energy storage into the power flow calculation model and the node frequency calculation model to calculate the system state quantities required for location;

[0025] S2.3. Use the multi-objective particle swarm optimization algorithm to solve the optimal grid-connected node of hybrid energy storage;

[0026] S2.4. Substitute the result of the optimal energy storage grid-connected node into the capacity determination layer of energy storage;

[0027] S2.5. Use the single-objective particle swarm optimization algorithm to solve the optimal configuration of hybrid energy storage;

[0028] S2.6. Substitute the result of the optimal configuration of hybrid energy storage back into the hybrid energy storage location layer model to calculate the optimal location result;

[0029] S2.7. Judge whether the latest optimized location result is the same as the previous optimization result. If they are the same, output the latest location and capacity determination result. If they are not the same, substitute the new optimized configuration node into the capacity determination layer to perform optimization and capacity determination again until the location results are the same.

[0030] Preferably, the step S3 specifically includes the following steps:

[0031] S3.1. Verification of the timing of energy storage site selection nodes:

[0032] S3.2. Verification of the deviation value and spatial uniformity of energy storage site selection.

[0033] Preferably, the objective function of the site selection layer in the step S1.12 is specifically:

[0034] Among them, f1, f2, f3, and f4 respectively correspond to the comprehensive evaluation index of power fluctuation and line load rate, the comprehensive evaluation index of node voltage fluctuation deviation degree and uniformity, network loss, and the comprehensive evaluation index of node frequency deviation degree and uniformity, λ i is the load rate of line i at time t, ΔU i is the node voltage deviation, Δf i is the node frequency deviation, R j is the resistance of the j-th branch, P j , Q j are the active and reactive powers of branch j respectively, U j is the voltage amplitude at the end of branch j, N is the number of system nodes, N br is the number of branches in the system, T power , T voltage , T frequency are the Theil entropy of power fluctuation, the Theil entropy of voltage fluctuation, and the Theil entropy of frequency fluctuation at time t respectively, and ω is the weight between different indexes in the objective function;

[0035] The constraint conditions of the site selection layer in the step S1.13 are specifically:

[0036] ΔP Li is the load disturbance of node i, ΔP b is the output of battery energy storage, ΔP fw is the output of flywheel energy storage, ΔP gi is the output of the i-th thermal power unit, P B.rated is the rated power of battery energy storage, P FW.rated is the rated power of flywheel energy storage, P ij (t) is the active power of the line between node i and node j, is the maximum power allowed for this line, are the maximum and minimum values of node i voltage respectively, U i (t) is the voltage of node i at time t, N site is the node for hybrid energy storage configuration, and N is the number of system nodes.

[0037] Preferably, the objective function of the fixed-capacity layer in the step S1.21 is specifically:

[0038] F = C inv + C OM + C scr + C F - B FR - B upgrade - B netloss , where C inv is the investment cost, and C scr is the battery scrapping treatment cost; C F is the energy storage system fault treatment cost; B FR is the energy storage frequency regulation revenue; B upgrade is the benefit of the energy storage system delaying the grid upgrade; B netloss The benefit of the energy storage reducing the system network loss.

[0039] Preferably, the constant volume layer constraint conditions in the step S1.22 include:

[0040] Energy storage configuration constraint:

[0041] P B.rated ≥ 0,

[0042] Energy storage output constraint:

[0043]

[0044] Hybrid energy storage output constraint:

[0045] where P B·rated and P FW.rated are the rated powers of the battery energy storage and the flywheel energy storage respectively.

[0046] The present invention provides a two - layer optimization method for the location and capacity determination of hybrid energy storage considering the node frequency distribution. It has the following beneficial effects:

[0047] 1. By considering the node frequency distribution, the proposed two - layer optimization method for the location and capacity determination of hybrid energy storage can effectively reduce the uneven distribution of the system power flow after the energy storage is connected to the grid and improve the stability of the system.

[0048] 2. In the optimization process, the present invention comprehensively considers the deviation degree and uniformity of the node voltage and frequency, ensures that the voltages and frequencies of each node are more balanced, and reduces the problems of voltage over - limit and frequency fluctuation caused by the connection of the energy storage to the grid.

[0049] 3. Through reasonable energy storage location and capacity determination configuration, the present invention reduces the load rate of the line, reduces the risk of line overload, and improves the operation safety and reliability of the system.

[0050] 4. The present invention adopts a double - layer model and optimizes from two levels of site selection and capacity determination respectively, making the configuration of the energy storage system more reasonable, taking into account both the investment cost and the operation benefit, and ensuring the economy and sustainability of the energy storage system.

[0051] 5. By considering the spatial distribution characteristics of node frequency fluctuations, the method proposed in the present invention improves the ability of the system to respond to frequency, effectively coping with the frequency stability challenges brought about by the large - scale grid connection of renewable energy and the high - proportion application of power electronic devices.

[0052] 6. During the process of site selection and capacity determination, the present invention comprehensively considers multiple indexes such as power fluctuation, line load rate, node voltage fluctuation and frequency fluctuation, making the optimization results have good effects in multiple aspects and improving the overall performance of the energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is the flowchart of the method of the present invention;

[0054] Figure 2 is the schematic diagram of the double - layer model for hybrid energy storage site selection and capacity determination of the present invention;

[0055] Figure 3 is the schematic diagram of the control strategy model of the hybrid energy storage of the present invention;

[0056] Figure 4 is the schematic diagram of the operation state of the hybrid energy storage system of the present invention;

[0057] Figure 5 is the flowchart for solving the double - layer model of hybrid energy storage site selection and capacity determination of the present invention;

[0058] Figure 6 is the IEEE 39 - node system diagram of the present invention;

[0059] Figure 7 is the schematic diagram of the per - unit value of the load disturbance at node 16 of the present invention;

[0060] Figure 8 is the schematic diagram of the per - unit value of the load disturbance at node 25 of the present invention;

[0061] Figure 9 is the schematic diagram of the per - unit value of the load disturbance at node 32 of the present invention;

[0062] Figure 10 is the schematic diagram of the node frequency before configuring the energy storage of the present invention;

[0063] Figure 11 is the schematic diagram of the node frequency after configuring the energy storage of the present invention;

[0064] Figure 12 is the schematic diagram of the line load rate before configuring the energy storage of the present invention;

[0065] Figure 13 Schematic diagram of the line load rate after configuring energy storage for the present invention;

[0066] Figure 14 Schematic diagram of the line load rate of the 9th line of the present invention;

[0067] Figure 15 Schematic diagram of the line load rate of the 13th line of the present invention;

[0068] Figure 16 Schematic diagram of the line load rate of the 25th line of the present invention;

[0069] Figure 17 Schematic diagram of the line power fluctuation before configuring energy storage for the present invention;

[0070] Figure 18 Schematic diagram of the line power fluctuation after configuring energy storage for the present invention;

[0071] Figure 19 Schematic diagram of the line power fluctuation of the 14th line of the present invention;

[0072] Figure 20 Schematic diagram of the node voltage before configuring energy storage for the present invention;

[0073] Figure 21 Schematic diagram of the node voltage after configuring energy storage for the present invention;

[0074] Figure 22 Schematic diagram of the node voltage fluctuation of the 12th node of the present invention;

[0075] Figure 23 Schematic diagram of the node voltage fluctuation of the 25th node of the present invention;

[0076] Figure 24 Schematic diagram of the uniformity of the line power fluctuation of the present invention;

[0077] Figure 25 Schematic diagram of the node voltage fluctuation of the present invention;

[0078] Figure 26 Schematic diagram of the node voltage uniformity of the present invention;

[0079] Figure 27 Schematic diagram of the node frequency fluctuation of the energy storage calculation example of the present invention;

[0080] Figure 28 Schematic diagram of the relative deviation of the frequency fluctuation of the present invention. Detailed implementation manner

[0081] Next, in combination with the accompanying drawings of the present invention, the technical solutions of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0082] Embodiment:

[0083] Please refer to the attached Figure 1 , the embodiment of the present invention provides a two-layer optimization method for the siting and sizing of a hybrid energy storage system considering node frequency distribution, including the following steps:

[0084] S1. Establish a two-layer model for the optimal siting and sizing of a hybrid energy storage system;

[0085] S2. Solve the two-layer model for the siting and sizing of a hybrid energy storage system;

[0086] S3. Analyze the effectiveness of different schemes.

[0087] Specifically, in the method steps of the present invention, by considering the node frequency distribution, the uneven distribution of the system power flow after the energy storage is connected to the grid is reduced, the node voltage and frequency are balanced, the line load rate is reduced, and the operation safety and reliability of the system are improved.

[0088] Adopting a two-layer model, optimization is carried out from two levels of siting and sizing, making the configuration of the energy storage system more reasonable, taking into account both the investment cost and the operation benefit, and ensuring the economy and sustainability of the energy storage system.

[0089] Improve the ability of the system to respond to frequency, and effectively cope with the frequency stability challenges brought by the large-scale grid connection of renewable energy and the high proportion application of power electronic devices.

[0090] The method is applicable to power systems of different scales and types, and has strong adaptability and application prospects.

[0091] In step S1, by establishing a two-layer model for the optimal siting and sizing of a hybrid energy storage system, a theoretical basis and a model foundation are provided for the siting and sizing of the energy storage system.

[0092] The problem that the spatial inhomogeneity of node frequency changes is rarely considered in traditional energy storage planning methods is solved.

[0093] In step S2, the optimal siting and configuration scheme of the energy storage system is determined by solving the model.

[0094] The problem of how to find the optimal energy storage siting and configuration scheme of the system under the comprehensive consideration of various indicators is solved.

[0095] Step S3 ensures the feasibility and advantages of the proposed method in practical applications by verifying and comparing the effectiveness of different solutions.

[0096] Please refer to the appendix Figure 2 - appendix Figure 4 , and Step S1 specifically includes the following steps:

[0097] S1.1. Establish a model for the location layer of hybrid energy storage;

[0098] S1.2. Establish a model for the capacity determination layer of hybrid energy storage.

[0099] Step S1.1 specifically includes the following steps:

[0100] S1.11. Establish a calculation model for the frequency fluctuation of network nodes;

[0101] S1.12. Determine the objective function of the location layer;

[0102] S1.13. Determine the constraint conditions of the location layer;

[0103] S1.14. Calculate the line power and Theil entropy of node voltage frequency.

[0104] Specifically, in Step S1.11, the relationship between the frequencies of each node in the system and the frequency of the generator node can be derived through the node voltage equation:

[0105]

[0106] In the formula, Δf k is the frequency change at network node k; Δf j is the frequency change at generator j; r kj is the impedance between node k and generator j; U j is the amplitude of the internal electromotive force of generator j; it can be seen from this formula that when the frequencies at each generator node are known, the frequencies of each node in the system can be obtained.

[0107] When there is a disturbance power ΔP at node k, the disturbance power components allocated to each generator are determined by the synchronizing power coefficient, and the synchronizing power coefficient is shown in the following formula:

[0108] D jk = U j U k B j k cosδ jko

[0109] In the formula: U k is the voltage amplitude at the disturbance node k; B jk is the susceptance between the internal electromotive force node j of the generator and the disturbance node k; δ jkois the difference between the initial phase angle of the voltage at the internal potential node j of the generator and the initial phase angle of the voltage at the disturbance node k. It can be seen from the above formula that the key to calculating the synchronous power coefficient is to calculate the susceptance between the disturbance node and the internal potential node of the generator.

[0110] For the port formed by the internal potential node j of the generator and the disturbance node k, its corresponding node-port incidence vector is:

[0111]

[0112] The equivalent impedance of the port formed by the internal potential node i of the generator and the node m is:

[0113]

[0114] Where: Z is the node impedance matrix.

[0115] Therefore, the equivalent admittance of the port is:

[0116]

[0117] Ignoring the resistance, the obtained equivalent admittance Y eq is the susceptance B between the internal potential node j of the generator and the node k jk .

[0118] After calculating the susceptance between the disturbance node and the internal potential node of the generator, substituting it into the synchronous power coefficient formula to calculate the synchronous power coefficients of each generator, so the power distribution of each generator is:

[0119]

[0120] Where, D jk and ΔP j are the synchronous power coefficient corresponding to the generator j and the unbalanced power distributed respectively.

[0121] From the generator rotor motion equation, the frequency change rate of the generator j is:

[0122]

[0123] Where, 2H j is the inertia time constant of the generator j. Substituting the above formula into the relationship between the frequencies of each node in the system and the frequency of the generator node can obtain the frequency fluctuations of each node.

[0124] In step S1.12, the objective function of the site selection layer is specifically:

[0125] Among them, f1, f2, f3, and f4 respectively correspond to the comprehensive evaluation index of power fluctuation and line load rate, the comprehensive evaluation index of the deviation degree and uniformity of node voltage fluctuation, network loss, and the comprehensive evaluation index of the deviation degree and uniformity of node frequency, and λ i is the load rate of line i at time t, ΔU i is the node voltage deviation, Δf i is the node frequency deviation, R j is the resistance of the j-th branch, P j , Q j are the active and reactive powers of branch j respectively, U j is the voltage amplitude at the end of branch j, N is the number of system nodes, N br is the number of branches in the system, T power , T voltage , T frequency are the Theil entropy of power fluctuation, the Theil entropy of voltage fluctuation, and the Theil entropy of frequency fluctuation at time t respectively, ω is the weight between different indexes in the objective function, taking 0.5.

[0126] In step S1.13, the constraint conditions of the site selection layer are specifically as follows:

[0127] ΔP Li is the load disturbance of node i, ΔP b is the output of battery energy storage, ΔP fw is the output of flywheel energy storage, ΔP gi is the output of the i-th thermal power unit, P B.rated is the rated power of battery energy storage, P FW.rated is the rated power of flywheel energy storage, P ij (t) is the active power of the line between node i and node j, is the maximum power allowed for this line, are the maximum and minimum values of node i voltage respectively, U i (t) is the voltage of node i at time t, N site is the node for hybrid energy storage configuration, N is the number of system nodes.

[0128] In step S1.14, the calculation of the Theil entropy of line power, node voltage and frequency is specifically as follows:

[0129] ① Theil entropy of power fluctuation:

[0130] When the line power flow changes, the overload of heavy - loaded lines is more serious than that of light - loaded lines. Therefore, the lines are first divided into three cases: light - loaded, medium - loaded, and heavy - loaded. When the line load rate λ ∈ (0, 0.33], it is a light - loaded line; when the line load rate λ ∈ (0.33, 0.66], it is a medium - loaded line; when the line load rate λ ∈ (0.66, 1], it is a heavy - loaded line. The above three types of lines correspond to the set Γ m , (m = 1 for light - loaded; m = 2 for medium - loaded; m = 3 for heavy - loaded).

[0131] When the active power at node k changes at time t, the power flow change of line ij is:

[0132] ΔP ij (t)=|P ij (t)-P ij (t - 1)|

[0133] In the formula: P ij (t) is the active power flowing through line ij at time t.

[0134] Calculate the total power flow changes of light - loaded lines, medium - loaded lines, and heavy - loaded lines at time t respectively:

[0135]

[0136] In the formula: ΔP m (t) is the total sum of the active power flow changes of each line in set m at time t; is the active power flow change of line ij in set m at time t.

[0137] The Theil index can measure the contributions of the within - group gap and the between - group gap to the total gap respectively. According to the definition, the Theil entropy of power fluctuation within the interval and the Theil entropy of power fluctuation outside the interval can be calculated.

[0138] The Theil entropy of power fluctuation within the interval at time t

[0139]

[0140] In the formula: N br is the total number of lines in set m

[0141] The Theil entropy of power fluctuation outside the interval at time t

[0142]

[0143] In the formula: N br is the total number of lines in the system

[0144] Finally, the Theil entropy of power fluctuation at time t is:

[0145]

[0146] ② Theil entropy of node voltage fluctuation

[0147] When the active power changes at node k at time t, the voltage change rate of node i is as follows:

[0148]

[0149] In the formula: ΔV i (t) is the amplitude of the voltage change of node i at time t; V i (t) is the voltage of node i at time t; is the initial voltage amplitude of node i before the energy storage is connected; is the rated voltage of node i.

[0150] Then the Theil entropy of voltage fluctuation can be expressed as:

[0151]

[0152] In the formula: γ(t) is the sum of voltage fluctuations within voltage level v; γ(t) is the sum of overall voltage fluctuations; N v is the number of nodes within voltage level v; N is the total number of nodes; is the Theil entropy of voltage fluctuation within voltage level v, and the calculation formula is as follows:

[0153]

[0154] In the formula: is the voltage change rate of node i within voltage level v.

[0155] ③ Theil entropy of node frequency fluctuation

[0156] When the active power changes at node k at time t, the frequency fluctuation value of node i is:

[0157] Δf i (t) = |f i (t) - f0|

[0158] In the formula: Δf i (t) is the frequency of node i at time t, and f0 is the system rated frequency (50Hz).

[0159] According to the magnitude of frequency fluctuation, the nodes are divided into three categories. When Δf i (t) ∈ [0, 0.2), it is not out of limit; when Δf i (t) ∈ [0.2, 0.5), it is out of limit; when Δf i (t) ∈ [0.5, +∞), it is severely out of limit, corresponding to the sets ∧m (m = 1 for not out of limit; m = 2 for out of limit; m = 3 for severely out of limit) respectively.

[0160] Calculate the sum of the frequency fluctuation values within each set:

[0161]

[0162] According to the definition of Theil entropy, calculate the Theil entropy of frequency fluctuation within the interval and the Theil entropy of frequency fluctuation outside the interval.

[0163] Theil entropy of frequency fluctuation within the interval:

[0164]

[0165] In the formula: Nm is the number of nodes within the interval

[0166] Theil entropy of frequency fluctuation outside the interval:

[0167]

[0168] In the formula: N is the number of nodes within the interval

[0169] The final Theil entropy of frequency fluctuation is:

[0170]

[0171] Please refer to Appendix Figure 3 , and the specific steps of S.12 include the following steps:

[0172] S1.21. Determine the objective function of the constant volume layer;

[0173] S1.22. Determine the constraint conditions of the constant volume layer;

[0174] S1.23. Establish a hybrid energy storage control strategy model.

[0175] Specifically, in step S1.21, the objective function of the constant volume layer is specifically:

[0176] F = C inv + C OM + C scr + C F - B FR - B upgrade - B netloss , where C inv is the investment cost, C scr is the battery scrapping and disposal cost; C F is the energy storage system fault handling cost; B FR is the energy storage frequency modulation revenue; B upgrade is the benefit of the energy storage system in delaying the grid upgrade; B netloss is the benefit of the energy storage in reducing the system network loss. The calculation formulas for each cost and revenue are as follows:

[0177]

[0178] In the formula, C B.P and C FW.P are the unit power costs of battery energy storage and flywheel energy storage respectively; C B.E and C FW.E are the unit capacity costs of battery energy storage and flywheel energy storage respectively; P B.rated and P FW.rated are the rated powers of battery energy storage and flywheel energy storage respectively, in MW; EB.rated and E FW.rated are the rated capacities of battery energy storage and flywheel energy storage respectively, in MWh; r is the discount rate; TLCC is the total life cycle, taking 20 years; n is the replacement times, n = T LCC / T life , plus the initial installation, a total of (n + 1) energy storage devices are invested; T life is the service life of lithium-ion batteries, calculated using the rain flow counting method; C B.PO is the operation and maintenance cost per unit power of the battery; C FW.PO is the operation and maintenance cost per unit power of the flywheel; C B.EO is the operation and maintenance cost per unit energy of the battery; C FW.EO is the operation and maintenance cost per unit energy of the flywheel; W B (t) and W FW (t) are the annual charge and discharge amounts of the battery and the flywheel respectively; C B.Pscr and C B.Escr are the battery unit power scrap disposal cost and the battery unit energy scrap disposal cost respectively; N F is the average annual failure times of the energy storage device; C F is the average failure handling cost; T of is the average annual power outage time of the energy storage device; e(i) is the electricity price of the energy storage auxiliary service market in the i-th year; W(i) is the charge and discharge electricity amount of the energy storage in the i-th year; λfr is the unit revenue of frequency modulation, 0.6 yuan / kWh, W fr is the frequency modulation electricity amount; ρ inf is the unit capacity expansion cost of the power grid transmission and transformation equipment, 2 million yuan / MW; and are the active powers flowing through branch k at time t before and after the energy storage is added respectively; i0 is the expected rate of return, 0.06; Δn is the number of years that the energy storage system delays the grid expansion; γ is the peak shaving rate of 1.64%; λ is the annual load growth rate of 0.015; ρ is the network loss electricity price, 0.3 yuan / kWh; ΔP t and ΔP t0 are the network loss powers of the system before and after the energy storage is added respectively.

[0179] In step S1.22, determine the constraints of the constant volume layer:

[0180] The constraints of the constant volume layer are specifically:

[0181] ① Energy storage configuration constraint

[0182] The discharge time scale of lithium-ion batteries ranges from minutes to hours. Therefore, the constraint conditions for capacity configuration are as follows.

[0183] P B.rated ≥0,

[0184] ② Energy storage output constraint

[0185] To avoid the life attenuation caused by overcharge and over-discharge of battery energy storage, a double Logistic function is used to reflect the output constraint of the battery under different S oc The expression is as follows:

[0186]

[0187] When S oc <0.1 or S oc ≥0.9, K(S oc ) = 0. In the formula, P0 takes 0.01; K max takes 1; n describing the speed of curve change takes 15;

[0188] S oc is the state of charge of the energy storage, defined as follows:

[0189]

[0190] In the formula: E rated is the rated capacity of the energy storage; P b is the discharge power of the lithium battery energy storage; S oc ranges from 0 to 1.

[0191] To avoid the over-limit of the energy storage S oc its constraint is as follows:

[0192]

[0193] In the formula: S OCL and S OCH are the maximum and minimum values of S OC respectively, taking the values of 0.1 and 0.9; S OCb (t) and S OCf (t) are the S OC of the battery energy storage and the S OC of the flywheel energy storage at time t respectively.

[0194] Hybrid energy storage output constraint:

[0195] Among them, P B·rated and P FW.ratedThey are the rated powers of battery energy storage and flywheel energy storage respectively.

[0196] In step S1.13, establish a hybrid energy storage control strategy model:

[0197] The hybrid energy storage control strategy model is specifically as follows:

[0198] First, according to the S of the hybrid energy storage OC Judge the operating condition S of the hybrid energy storage n , the judgment method is shown in Table 1, and then establish a hybrid energy storage state transition equation according to the operating condition of the hybrid energy storage combined with the active power fluctuation ΔP of the system.

[0199] Table 1: Judgment of the operating condition of the hybrid energy storage

[0200]

[0201] The specific hybrid energy storage state transition equation is as follows:

[0202] ① Energy storage discharge state

[0203] When the operating condition of the hybrid energy storage system is S5, S6, S8, S g , both the battery energy storage and the flywheel energy storage have a large discharge capacity. At this time, the flywheel energy storage responds to the command first. If the power of the flywheel energy storage is insufficient, the battery responds. If the powers of both the flywheel and the battery are insufficient, the thermal power unit adjusts, that is:

[0204]

[0205] In the formula: P f_c , P b_c , P g_s are the output commands of the flywheel energy storage, the battery output command, and the supplementary output command of the thermal power unit respectively; P re_f , P re_b are the SOC recovery power commands of the flywheel energy storage and the battery respectively; P Hy_c is the power command of the hybrid energy storage system. The above formula corresponds to B1 in Appendix Figure 4 .

[0206] When the operating condition of the hybrid energy storage system is S2, S3, the battery energy storage has a large discharge capacity, while the flywheel energy storage S OC is low. At this time, the battery discharges first, the flywheel does not respond, and the thermal power unit restores at the rated power S OC , and the thermal power unit participates in the regulation when the battery discharge power is insufficient, that is:

[0207]

[0208] The above formula corresponds to B2 in Appendix Figure 4 .

[0209] When the operating conditions of the hybrid energy storage system are S4 and S7, the flywheel energy storage has a large discharge capacity, while the battery S OC is relatively low. At this time, the flywheel energy storage discharges first, the battery does not respond, and the thermal power unit restores S at the rated power OC , and the thermal power unit participates in the regulation when the discharge power of the flywheel is insufficient, that is:

[0210]

[0211] The above formula corresponds to B3 in the appendix Figure 4 .

[0212] When the operating condition of the hybrid energy storage is S1, both the battery and the flywheel S OC are relatively low and cannot respond to the discharge command. The hybrid energy storage is restored by the thermal power unit OC , and at this time, the thermal power unit increases the power for regulation, that is:

[0213]

[0214] The above formula corresponds to B4 in the appendix Figure 4 .

[0215] ② Energy storage charging state

[0216] When the operating conditions of the hybrid energy storage system are S1, S2, S4, and S5, both the battery energy storage and the flywheel energy storage have a large charging capacity. At this time, the flywheel energy storage responds to the command first. If the charging power of the flywheel energy storage is insufficient, the battery responds. If the charging capacities of both the flywheel and the battery are insufficient, the thermal power unit regulates, that is:

[0217]

[0218] The above formula corresponds to B5 in the appendix Figure 4 .

[0219] When the operating conditions of the hybrid energy storage system are S7 and S8, the battery energy storage has a large charging capacity, while the flywheel energy storage S oc is relatively high. At this time, the battery charges first, the flywheel does not respond, and the discharge resistor restores S at the rated power oc , and the thermal power unit participates in the regulation when the charging power of the battery is insufficient, that is:

[0220]

[0221] The above formula corresponds to B6 in the appendix Figure 4 .

[0222] When the operating conditions of the hybrid energy storage system are S3 and S6, the flywheel energy storage has a large charging capacity, while the battery S ocOn the high side, at this time, the flywheel energy storage is preferentially charged, and the battery does not respond and restores S at the rated power through the discharge resistor oc , and the thermal power unit participates in the regulation when the charging power of the flywheel is insufficient, that is:

[0223]

[0224] The above formula corresponds to B7 in the appendix Figure 4 .

[0225] When the operating condition of the hybrid energy storage system is S g , both the battery and the flywheel S oc are on the high side and cannot continue to respond to the charging instruction. The hybrid energy storage restores S through the discharge resistor oc , and at this time, the thermal power unit reduces the power for regulation, that is:

[0226]

[0227] The above formula corresponds to B8 in the appendix Figure 4 .

[0228] Step S2 specifically includes the following steps:

[0229] S2.1. Initialize the configured capacity of the hybrid energy storage;

[0230] S2.2. Substitute the initialized configured capacity of the hybrid energy storage into the power flow calculation model and the node frequency calculation model to calculate the system state quantities required for site selection;

[0231] S2.3. Use the multi-objective particle swarm optimization algorithm to solve the optimal grid-connected node of the hybrid energy storage;

[0232] S2.4. Substitute the result of the optimal energy storage grid-connected node into the energy storage sizing layer;

[0233] S2.5. Use the single-objective particle swarm optimization algorithm to solve the optimal configuration of the hybrid energy storage;

[0234] S2.6. Substitute the result of the optimal configuration of the hybrid energy storage back into the hybrid energy storage site selection layer model to calculate the optimal site selection result;

[0235] S2.7. Judge whether the latest optimized site selection result is the same as the previous optimization result. If they are the same, output the latest site selection and sizing result. If they are different, substitute the new optimized configuration node into the sizing layer for re-optimized sizing until the site selection results are the same.

[0236] Please refer to the appendix Figure 5 -Appendix Figure 9 , Step S3 specifically includes the following steps:

[0237] S3.1. Verification of the timing of the energy storage site selection node:

[0238] S3.2, Verification of energy storage site selection deviation value and spatial uniformity.

[0239] Specifically, the IEEE 39 - node system is used for simulation. As Figure 6 shown, the base load of each node remains unchanged, and active power load fluctuations are superimposed on the base loads of the 25 - node, 16 - node, and 32 - node respectively, as Figure 6 shown by the red dots. The per - unit value of the superimposed active power load fluctuations is as follows Figures 7 - 9 shown; the data length is 1440, and the sampling interval is 1 min. The total capacity of the system generators is 7367 MW. The simulation tool uses the MTALAB & MATPOWER power flow calculation toolbox.

[0240] The objective function of the site selection layer in step S1.12 is specifically:

[0241] Among them, f1, f2, f3, and f4 respectively correspond to the comprehensive evaluation index of active power fluctuation - line load rate, the comprehensive evaluation index of node voltage fluctuation deviation degree and uniformity, network loss, and the comprehensive evaluation index of node frequency deviation degree and uniformity. λ i is the load rate of line i at time t, ΔU i is the node voltage deviation, Δf i is the node frequency deviation, R j is the resistance of the j - th branch, P j , Q j are the active and reactive powers of branch j respectively, U j is the voltage amplitude at the end of branch j, N is the number of system nodes, N br is the number of branches in the system, T power , T voltage , T frequency are the Theil entropy of power fluctuation, the Theil entropy of voltage fluctuation, and the Theil entropy of frequency fluctuation at time t respectively, and ω is the weight between different indexes in the objective function;

[0242] The constraint conditions of the site selection layer in step S1.13 are specifically:

[0243] ΔP Li is the load disturbance of node i, ΔP b is the output of battery energy storage, ΔP fw is the output of flywheel energy storage, ΔP gi is the output of the i - th thermal power unit, P B.rated is the rated power of battery energy storage, P FW.rated is the rated power of flywheel energy storage, P ij (t) is the active power of the line between node i and node j, is the maximum power allowed for this line, are the maximum and minimum values of the voltage of node i, respectively, and U i (t) is the voltage of node i at time t, and N site is the node for hybrid energy storage configuration, and N is the number of system nodes.

[0244] The specific objective function of the constant volume layer in step S1.21 is as follows:

[0245] F = C inv + C OM + C scr + C F - B FR - B upgrade - B netloss , where C inv is the investment cost, and C scr is the battery scrapping treatment cost; C F is the energy storage system fault treatment cost; B FR is the energy storage frequency regulation benefit; B upgrade is the benefit of the energy storage system delaying the grid upgrade; B netloss The benefit of the energy storage reducing the system network loss.

[0246] The constraint conditions of the constant volume layer in step S1.22 include:

[0247] Energy storage configuration constraint:

[0248]

[0249] Energy storage output constraint:

[0250]

[0251] Hybrid energy storage output constraint:

[0252] where P B·rated and P FW.rated are the rated powers of the battery energy storage and the flywheel energy storage, respectively.

[0253] Please refer to Appendix Figure 10 - Appendix Figure 28 , to verify the effectiveness of the hybrid energy storage siting and sizing double-layer planning model proposed by the present invention, the following experiments were designed:

[0254] Purpose of the experiment

[0255] The purpose of this experiment is to verify the effectiveness of the hybrid energy storage siting and sizing double-layer planning model proposed by the present invention. Specifically, it includes two aspects of verification:

[0256] 1. Verification of the time series of the energy storage siting nodes: By comparing the power flow and node frequency changes of the system before and after configuring the energy storage, the impact of the energy storage grid connection is evaluated.

[0257] 2. Verification of energy storage site selection deviation value and spatial uniformity: Analyze the spatial deviation value and uniformity of system indicators under different schemes to prove the effectiveness of this method in this regard.

[0258] Experimental process

[0259] The experiment uses the IEEE 39 - node system for simulation. The basic load of each node remains unchanged, and active power load fluctuations are superimposed on the basic loads of the 25 - node, 16 - node, and 32 - node respectively. The simulation tool uses the MATLAB & MATPOWER power flow calculation toolbox. The experimental process is divided into the following steps:

[0260] 1. Establish a double - layer model for optimal site selection and capacity determination of hybrid energy storage

[0261] Model of the site selection layer:

[0262] Establish a calculation model for the frequency fluctuation of network nodes.

[0263] Determine the objective function and constraint conditions of the site selection layer.

[0264] Calculate the line power and Theil entropy of node voltage frequency.

[0265] Model of the capacity determination layer:

[0266] Determine the objective function and constraint conditions of the capacity determination layer.

[0267] Establish a control strategy model for hybrid energy storage.

[0268] 2. Solve the double - layer model for optimal site selection and capacity determination of hybrid energy storage

[0269] Initialize the configured capacity of hybrid energy storage.

[0270] Substitute the initialized configured capacity of hybrid energy storage into the power flow calculation model and node frequency calculation model to calculate the system state variables required for site selection.

[0271] Use the multi - objective particle swarm optimization algorithm to solve the optimal grid - connected nodes of hybrid energy storage.

[0272] Substitute the result of the optimal energy storage grid - connected nodes into the energy storage capacity determination layer.

[0273] Use the single - objective particle swarm optimization algorithm to solve the optimal configuration of hybrid energy storage.

[0274] Return the optimal configuration result of hybrid energy storage to the site selection layer model to calculate the optimal site selection result.

[0275] Judge whether the latest optimized site selection result is consistent with the previous optimization result. If not, continue the iteration until the results are consistent.

[0276] 3. Verify the effectiveness of different schemes

[0277] Temporal verification of energy storage siting nodes:

[0278] Compare the power flow and node frequency changes of the system before and after configuring energy storage.

[0279] The analysis results are as Figures 10 - 21 , shown in Table 2 - Table 5.

[0280] Table 2 Statistics of the absolute values of node frequency fluctuations before and after energy storage configuration:

[0281]

[0282] Table 3 Statistics of line load rates before and after energy storage configuration:

[0283]

[0284] Table 4 Statistics of the absolute values of line power fluctuations before and after energy storage configuration:

[0285]

[0286] Table 5 Average voltage fluctuations of nodes 11 - 39:

[0287]

[0288]

[0289] It can be seen from the results that:

[0290] 1) In terms of node frequency:

[0291] Before configuring energy storage, the frequency change of the system is relatively uniform in spatial distribution and has large fluctuations in time series. In addition, due to the large active power fluctuations of the system, the frequency fluctuations of the system are large. The highest peak of frequency fluctuation is 0.31 Hz, the lowest value is -0.32 Hz, and the average value of the absolute value of fluctuation is 0.097 Hz. From Figure 11 and Table 2, it can be seen that after configuring energy storage, the frequency fluctuations of the system decrease significantly. The highest peak of frequency fluctuation is 0.0026 Hz, the lowest value is -0.0027 Hz, and the average value of the absolute value of frequency fluctuation is 0.00047 Hz. While the frequency fluctuations decrease, there is also a certain frequency spatial distribution.

[0292] 2) In terms of line load rate:

[0293] The spatial distribution trend of the line load rate of the system before and after the integration of energy storage is roughly the same. However, in terms of temporality, after the integration of energy storage, the load rate fluctuations of 19 lines are greater than those without energy storage, but the load rates are slightly smaller, such as Line 9 and Line 13. As Figure 14 ) and Figure 15) As shown in the figure: when there is no energy storage configuration, the average load rate of line 9 is 68.47%, and after configuring energy storage, the average load rate is 68.45%; when there is no energy storage configuration for line 13, the average load rate is 94.79%, and after configuring energy storage, the average load rate is 94.78%. Only line 25 (node 25 to node 26) close to the energy storage connection node has a smaller load rate fluctuation than the case without energy storage as shown in Figure 16 ) In this case, the load rate before and after connecting energy storage is 26.34%.

[0294] It can be seen from Table 3 that overall, the average value of the line load rate before and after configuring energy storage is equal, both being 38.87%. However, the maximum and minimum values of the load rate both occur after configuring energy storage, which proves that after configuring energy storage, the distribution of the line load rate becomes uneven.

[0295] 3) In terms of line power fluctuation

[0296] Before configuring energy storage, the highest absolute value of the power fluctuation of each line occurred at the 916th minute of line 14, which was 38.20 MW. After configuring energy storage, the highest value of the power fluctuation of each line occurred at the 931st minute of line 14, which was 75.94 MW. It can be seen that the line with the largest power fluctuation is line 14, and configuring energy storage exacerbates the power fluctuation of the line. The power fluctuation curve of line 14 is as shown in Figure 19 In general, after configuring energy storage, the average value of the absolute value of the line power fluctuation increases, from 1.65 kW to 3.63 kW.

[0297] 4) In terms of node voltage

[0298] The voltage of each node has a large difference in spatial distribution, a small difference in time series, and there is not much difference before and after connecting energy storage. However, according to the statistical results in Table 5, after configuring energy storage, the average value of the voltage fluctuation of this node is reduced.

[0299] By carefully comparing the time-series fluctuations of the voltage of each node before and after configuring energy storage, it is found that except for nodes 11, 12, and 13, the voltage fluctuations of the remaining nodes are reduced after configuring energy storage. Figure 22 ) shows the voltage fluctuation of node 12, Figure 23 ) shows the voltage fluctuation of node 25.

[0300] Verification of energy storage location selection deviation value and spatial uniformity:

[0301] Select the data at the moment of the largest active power fluctuation, that is, the 411th minute, and analyze and compare various indicators under different schemes.

[0302] Different schemes include:

[0303] Scheme 1: Configure energy storage without considering the index uniformity.

[0304] Scenario 2: Configure energy storage without considering the degree of index deviation.

[0305] Scenario 3: A method for determining the location and capacity of hybrid energy storage by comprehensively considering the degree of deviation and uniformity of various indicators.

[0306] The simulation results are as Figures 24 - 28 , shown in Tables 6 - 8.

[0307]

[0308]

[0309] Table 7 Average value of frequency deviation in space for each example

[0310]

[0311] Table 8 Comparison of the effects of each planned energy storage method

[0312]

[0313] It can be seen from the results that in Scenario 2, only the standard deviation of index deviation is considered in the objective function, which can reduce the average value of the system line load rate, the voltage deviation value of each node, and the average value of frequency fluctuation, but makes the power fluctuation distribution of each line, the voltage fluctuation distribution of each node, and the frequency distribution of each node uneven; in Scenario 3, only the Theil entropy of the index is considered in the objective function, which can make the power fluctuation distribution of each line, the voltage fluctuation distribution of each node, and the frequency distribution of each node uniform, but the load rate, the voltage fluctuation of each node, and the average value of frequency fluctuation are larger; this method (Scenario 4) comprehensively considers the standard deviation of each index deviation and its Theil entropy, making the degree of fluctuation and uniformity of each index in a better state.

[0314] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A two-layer optimization method for the location and capacity determination of hybrid energy storage considering node frequency distribution, characterized in that, It includes the following steps: S1. Establish a two - layer model for optimal siting and sizing of hybrid energy storage; The S1 step specifically includes the following steps: S1.

1. Establish a siting layer model of hybrid energy storage; The S1.1 step specifically includes the following steps: S1.

11. Establish a calculation model for network node frequency fluctuation; S1.

12. Determine the objective function of the siting layer; S1.

13. Determine the constraint conditions of the siting layer; S1.

14. Calculate the line power, node voltage frequency Theil entropy; S1.

2. Establish a sizing layer model of hybrid energy storage; S2. Solve the two - layer model for optimal siting and sizing of hybrid energy storage; S3. Analyze the effectiveness of different schemes; The S3 step specifically includes the following steps: S3.

1. Verification of the timing of energy storage siting nodes: S3.

2. Verification of the deviation value and spatial uniformity of energy storage siting.

2. A hybrid energy storage site selection and capacity determination double-layer optimization method considering node frequency distribution according to claim 1, characterized in that The S1.2 step specifically includes the following steps: S1.

21. Determine the objective function of the sizing layer; S1.

22. Determine the constraint conditions of the sizing layer; S1.

23. Establish a control strategy model of hybrid energy storage.

3. A hybrid energy storage site selection and capacity determination double-layer optimization method considering node frequency distribution according to claim 1, characterized in that The S2 step specifically includes the following steps: S2.

1. Initialize the configured capacity of hybrid energy storage; S2.

2. Substitute the initialized configured capacity of hybrid energy storage into the power flow calculation model and node frequency calculation model to calculate the system state variables required for siting; S2.

3. Use the multi - objective particle swarm optimization algorithm to solve the optimal grid - connected nodes of hybrid energy storage; S2.

4. Substitute the optimal energy storage grid - connected node result into the energy storage sizing layer; S2.

5. Use the single - objective particle swarm optimization algorithm to solve the optimal configuration of hybrid energy storage; S2.

6. Substitute the optimal configuration result of hybrid energy storage back into the hybrid energy storage siting layer model to calculate the optimal siting result; S2.

7. Judge whether the latest optimized siting result is consistent with the previous optimization result. If it is consistent, output the latest siting and sizing result. If it is not consistent, substitute the new optimized configuration node into the sizing layer to optimize the sizing again until the siting result is consistent.

4. A method for double-layer optimization of the location and capacity determination of a hybrid energy storage system considering the node frequency distribution according to claim 1, characterized in that The objective function of the siting layer in the S1.12 step is specifically: Among them, f1, f2, f3, and f4 respectively correspond to the comprehensive evaluation index of power fluctuation and line load rate, the comprehensive evaluation index of node voltage fluctuation deviation degree and uniformity, network loss, and the comprehensive evaluation index of node frequency deviation degree and uniformity, λ i is the load rate of line i at time t, ΔU i is the node voltage deviation, Δf i is the node frequency deviation, R j is the resistance of the jth branch, P j , Q j are the active and reactive powers of branch j respectively, U j is the voltage amplitude at the end of branch j, N is the number of system nodes, N br is the number of branches in the system, T power , T voltage , T frequency are the Theil entropy of power fluctuation, the Theil entropy of voltage fluctuation, and the Theil entropy of frequency fluctuation at time t respectively, and ω is the weight between different indexes in the objective function; The constraint conditions of the siting layer in the S1.13 step are specifically: ΔP Li is the load disturbance of node i, ΔP b is the output of battery energy storage, ΔP fw is the output of flywheel energy storage, ΔP gi is the output of the i-th thermal power unit, P B.rated is the rated power of battery energy storage, P FW.rated is the rated power of flywheel energy storage, P ij (t) is the active power of the line between node i and node j, is the maximum power allowed for this line, are the maximum and minimum values of the voltage of node i, U i (t) is the voltage of node i at time t, N site is the node with hybrid energy storage configuration, N is the number of system nodes.

5. A hybrid energy storage site selection and capacity determination double-layer optimization method considering node frequency distribution according to claim 1, characterized in that The objective function of the sizing layer in the S1.21 step is specifically: F = C inv + C OM + C scr + C F - B FR - B upgrade - B netloss , where C inv is the investment cost, C scr is the battery end - of - life treatment cost; C F is the energy storage system fault treatment cost; B FR is the energy storage frequency regulation revenue; B upgrade is the benefit of the energy storage system delaying the grid upgrade; B netloss The benefit of the energy storage reducing the system network loss.

6. A hybrid energy storage site selection and capacity determination double-layer optimization method considering node frequency distribution according to claim 1, characterized in that The constraint conditions of the sizing layer in the S1.22 step include: Energy storage configuration constraint: P B.rated ≥0， Energy storage output constraint: Hybrid energy storage output constraint: where P B·rated and P FW.rated are the rated powers of battery energy storage and flywheel energy storage respectively.

Images