A soc adaptive adjustment hybrid energy storage optimization configuration method

Abstract

This invention discloses a hybrid energy storage optimization configuration method with adaptive SOC adjustment, comprising the following steps: calculating the system's frequency response under maximum disturbance; determining the minimum support power of energy storage based on frequency stability indicators; smoothing new energy fluctuations according to new energy fluctuation smoothing standards to obtain new energy grid-connected power and energy storage action commands; adaptively adjusting the lower limit of SOC based on power support requirements to establish a hybrid energy storage optimization configuration model; and allocating the internal power of the hybrid energy storage according to the energy storage action commands and the optimization configuration model to obtain the optimization configuration result. This invention comprehensively considers the needs for emergency power support from energy storage and new energy fluctuation smoothing, and optimizes the energy storage capacity configuration based on economic efficiency. The energy storage configuration result, which comprehensively considers multiple factors, is more in line with actual engineering needs. This invention allocates energy storage action commands to different energy storage systems based on the hybrid energy storage optimization configuration model, fully leveraging the characteristics of different energy storage systems.

Description

Technical Field

[0001] This invention belongs to the field of power system energy storage optimization configuration, specifically involving a hybrid energy storage optimization configuration method with SOC adaptive adjustment. Background Technology

[0002] In recent years, the installed capacity of pollution-free and sustainable renewable energy power generation connected to the grid has grown rapidly. Renewable energy power generation, such as wind power and photovoltaic power, is characterized by volatility, intermittency, and uncertainty, and exhibits dynamic response characteristics different from conventional power generation. When a high proportion of these renewable energy sources are integrated into the power system, it can easily lead to frequency stability issues, posing new challenges to the safe and stable operation of the power system and the adequacy of power supply. How to maintain the stability of the power system while increasing the penetration rate of renewable energy is a pressing issue.

[0003] Energy storage features energy time-shifting, rapid response, and flexible deployment, enabling it to improve the volatility of renewable energy generation, enhance the dispatchability of renewable energy, and supplement grid frequency. This provides technical support for improving the operational stability of new power systems under both conventional and extreme scenarios. Based on their output characteristics, power-type energy storage such as supercapacitors can provide instantaneous high power to meet the short-term power needs of high-proportion renewable energy systems; while energy-type energy storage such as pumped hydro storage can maintain continuous power output for extended periods, achieving long-term energy supply.

[0004] Hybrid energy storage combines the advantages of single-type energy storage, possessing multi-timescale response and dual power-energy regulation characteristics. It can improve the output characteristics of new energy sources, enhance their absorption capacity, ensure and improve power quality, and strengthen the system's anti-interference and stable operation capabilities. Regarding energy storage for mitigating new energy fluctuations, some scholars have obtained power references for hybrid energy storage based on wavelet packet decomposition and used fuzzy control to adaptively optimize the power commands of battery energy storage and supercapacitors. Others have decomposed the original signal using empirical mode decomposition (EMD) to obtain active power references for the electric-hydrogen hybrid energy storage system and formulated energy management strategies considering the operating characteristics of alkaline electrolyzers. Still others have considered dual photovoltaic evaluation indicators and used the grey relational analysis method to reconstruct the decomposed photovoltaic power into high-frequency and low-frequency components to obtain HESS active power references, improving the smoothness of photovoltaic grid connection.

[0005] Regarding energy storage participation in primary and secondary frequency regulation, some scholars have proposed energy storage capacity configuration strategies based on minimum inertia control and considering the primary frequency regulation capability of energy storage; others have proposed an adaptive strategy for battery energy storage to participate in primary frequency regulation by combining virtual droop and virtual inertia control strategies; and still others have proposed a power command allocation strategy and capacity configuration strategy for energy storage to participate in secondary frequency regulation, considering the ramp rate limitation of conventional generating units.

[0006] However, the above methods are all based on the static frequency regulation capability of energy storage and do not consider the dynamic frequency response of energy storage in emergency power support. To address this issue, it is necessary to invent an energy storage optimization configuration model that simultaneously considers emergency power support and the demand for smoothing out new energy fluctuations. Summary of the Invention

[0007] This invention is proposed to address the problems existing in the prior art, and its purpose is to provide a hybrid energy storage optimization configuration method with SOC adaptive adjustment.

[0008] The technical solution of this invention is: a hybrid energy storage optimization configuration method with adaptive SOC adjustment, comprising the following steps:

[0009] A. Calculate the frequency response of the system under maximum disturbance, and determine the minimum support power of energy storage based on the frequency stability index;

[0010] B. Based on the new energy fluctuation smoothing standard, the new energy fluctuation is smoothed to obtain the new energy grid-connected power and energy storage action instructions;

[0011] C. Adaptively adjust the lower limit of SOC according to power support requirements and establish a hybrid energy storage optimization configuration model;

[0012] D. Based on the energy storage action command and the optimization configuration model, the internal power of the hybrid energy storage is allocated, and the optimization configuration result is obtained.

[0013] Furthermore, step A calculates the system's frequency response under maximum disturbance, determines the minimum support power for energy storage based on frequency stability indicators, and utilizes the energy storage device to provide emergency power support when the system frequency decreases until the system frequency returns to a safe level.

[0014] Furthermore, step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage, including determining the transfer function expression of the system frequency and the disturbance. The specific process is as follows:

[0015] First, assuming that the speed and frequency of a traditional generator are equal, without considering the load frequency regulation effect and damping coefficient, and that the automatic generation control (AGC) does not operate on the primary frequency regulation time scale;

[0016] Then, according to the synchronous motor rotor motion equation, the system frequency and the transfer function of the disturbance satisfy the following equation:

[0017]

[0018] In the formula, Δf * This is the per-unit value of the frequency deviation;

[0019] s is a complex frequency;

[0020] J eq This is the system's equivalent inertial time constant;

[0021] K * This is the per-unit value of the primary frequency modulation coefficient;

[0022] T s The inertial time constant of the unit;

[0023] This is the per-unit value for the incremental electromagnetic power of the unit.

[0024] Furthermore, step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage, including the expression for frequency deviation. The specific process is as follows:

[0025] First, assume the power of the energy storage device configured in the system is P. s When the system frequency is below the critical point, the energy storage discharges at its maximum power to support the system frequency.

[0026] Then, let the above time be t1, then the frequency deviation Δf(t1) at time t1 relative to the rated value satisfies the following equation:

[0027] Δf(t1)=f t -f n

[0028] In the formula, f t This is the critical frequency for energy storage operation;

[0029] f n This is the rated value for the power grid frequency.

[0030] Furthermore, step A calculates the system's frequency response under maximum disturbance and determines the minimum support power for energy storage based on frequency stability indicators, including establishing the system's frequency response expression. The specific process is as follows:

[0031] After energy storage is put into operation, the system frequency response satisfies the following equation:

[0032]

[0033] In the formula, This refers to the per-unit value of the incremental mechanical power of a traditional generator set;

[0034] The per-unit value of the power output to provide emergency power support for energy storage;

[0035] Furthermore, step A calculates the system's frequency response under maximum disturbance and determines the minimum support power for energy storage based on frequency stability indicators, including determining system frequency constraints. The specific process is as follows:

[0036] First, based on the system's operational limitations, frequencies below f... min This may cause the system to activate safety measures such as low-frequency load reduction, affecting normal production;

[0037] Then, the energy storage is configured such that, in the event of the worst-case power disturbance, the lowest point of the power system frequency is above f. min It satisfies the following equation:

[0038] f nadir ≥f min

[0039] In the formula, f nadir This is the point of lowest frequency during the disturbance process;

[0040] f min The lowest point at which the system frequency is allowed to drop, in order to ensure system frequency stability;

[0041] Finally, by solving the above equations, the minimum supporting power P for energy storage is obtained. smin .

[0042] Furthermore, step B, based on the renewable energy fluctuation mitigation standard, mitigates renewable energy fluctuations, resulting in renewable energy grid-connected power and energy storage action commands. This includes establishing a correlation between renewable energy grid-connected power and hybrid energy storage mitigation power. The specific process is as follows:

[0043] First, the original power output signal of the new energy source is decomposed into P0 to extract the grid-connected power P of the new energy source. grid With hybrid energy storage to mitigate power P HESS ;

[0044] Then, establish the grid-connected power P of new energy sources. grid With hybrid energy storage to mitigate power P HESS Related expressions:

[0045] (1) Construct the sequence P(i)=P0+ε1E1(W(i)) to obtain the first set of residuals R1=ε1N(P(i));

[0046] (2) The first modal component imf1 = P0 - R1 is obtained.

[0047] (3) Continue adding white noise and calculate the residual R of the kth group in sequence. k =N(R) k-1 +ε k-1 E(W(i))) and the k-th modal component imf k=R k-1 -R k until all modes and residuals are obtained.

[0048] Wherein, P0 is the signal to be decomposed;

[0049] E k (·) represents the k-th order modal component generated by EMD decomposition;

[0050] N(·) is the local mean of the generated signal;

[0051] R k For the k-th group of residuals;

[0052] ε k These are the weighting coefficients for the k-th white noise element;

[0053] imf k This is the k-th modal component;

[0054] W(i) is a Gaussian white noise with a mean of 0 and a variance of 1 for group i.

[0055] Furthermore, step B smooths out the fluctuations in renewable energy based on the renewable energy fluctuation smoothing standard, resulting in renewable energy grid-connected power and energy storage action commands, including reconfiguration correlation expressions. The specific process is as follows:

[0056] The decomposition results are reconstructed to satisfy the following equation:

[0057]

[0058] In the formula, k1 is the number of IMF components obtained from the decomposition;

[0059] res represents the residual;

[0060] P grid Contribute grid-connected power to new energy sources;

[0061] P HESS To smooth out the power of HESS;

[0062] n1 is the number of the boundary layer between grid-connected power and power smoothing;

[0063] i and j are the component numbers, respectively.

[0064] Furthermore, step B, based on the renewable energy fluctuation smoothing standard, smooths out renewable energy fluctuations, resulting in renewable energy grid-connected power and energy storage action commands. This includes establishing power-based constraints, the specific process of which is as follows:

[0065] The difference between the maximum and minimum power of actual grid-connected renewable energy within a certain time interval shall not exceed the fluctuation limit of renewable energy output, and shall satisfy the following equation:

[0066]

[0067] In the formula, t represents time, indicating the moment when the new energy source outputs power;

[0068] Δt is the time interval;

[0069] P gmax and P gmin These represent the lower limit of SOC, the maximum grid-connected power, and the minimum grid-connected power during this period, respectively.

[0070] P N For new energy installed capacity;

[0071] σ max This represents the maximum output fluctuation rate.

[0072] Furthermore, step B, based on the renewable energy fluctuation smoothing standard, smooths out renewable energy fluctuations, obtaining renewable energy grid-connected power and energy storage action commands, including determining the optimal number of boundary layers. The specific process is as follows:

[0073] First, calculate whether the low-frequency fluctuation components under each boundary layer number n1 meet the smoothing requirements. If they do, stop the loop.

[0074] Then, if the smoothness requirement is not met, the number of boundary layers is increased until the optimal n1 is found, and the new energy grid-connected power signal P is obtained. grid and HESS power signal P HESS .

[0075] The beneficial effects of this invention are as follows:

[0076] This invention comprehensively considers the needs for emergency power support from energy storage and the need to mitigate fluctuations in new energy sources. Furthermore, it optimizes the configuration of energy storage capacity based on economic efficiency. The energy storage configuration result, which takes into account multiple factors, is more in line with the actual needs of engineering projects.

[0077] This invention determines the minimum support power for energy storage based on the system's dynamic frequency response, taking into account worst-case conditions to improve the applicability of the minimum support power; considering the power demand for emergency power support, it optimizes the lower limit of the state of charge (SOC) of the energy storage device during normal operation to achieve adaptive adjustment.

[0078] This invention employs an adaptive improved complete set empirical mode algorithm to decompose the original output of new energy sources, achieving a smoothing effect slightly better than the traditional EMD algorithm. It also uses hybrid energy storage to smooth fluctuations and allocates energy storage action commands to different energy storage systems based on a hybrid energy storage optimization configuration model, thus fully leveraging the characteristics of different energy storage systems. Attached Figure Description

[0079] Figure 1 This is a flowchart of the present invention;

[0080] Figure 2 These are the frequency response curves before and after storage in this invention;

[0081] Figure 3 This invention demonstrates the effect of smoothing fluctuations before and after the adoption of hybrid energy storage.

[0082] (a) Wind power curves before and after energy storage allocation; (b) Power fluctuations within 1 minute before and after energy storage allocation; (c) Power fluctuations within 10 minutes before and after energy storage allocation.

[0083] Figure 4 This is the configuration result of Scheme 1 in this invention;

[0084] (a) Supercapacitor operating curve; (b) Lithium iron phosphate battery operating curve; (c) Supercapacitor SOC curve; (d) Lithium iron phosphate battery SOC curve;

[0085] Figure 5 This is the configuration result of Scheme 1 in this invention;

[0086] (a) Operating curve of lithium iron phosphate battery; (b) SOC curve of lithium iron phosphate battery;

[0087] Figure 6 This is the configuration result of Scheme 1 in this invention;

[0088] (a) Supercapacitor operating curve; (b) Supercapacitor SOC curve. Detailed Implementation

[0089] The present invention will now be described in detail with reference to the accompanying drawings and embodiments:

[0090] like Figures 1 to 6 As shown, a hybrid energy storage optimization configuration method with SOC adaptive adjustment includes the following steps:

[0091] A. Calculate the frequency response of the system under maximum disturbance, and determine the minimum support power of energy storage based on the frequency stability index;

[0092] B. Based on the new energy fluctuation smoothing standard, the new energy fluctuation is smoothed to obtain the new energy grid-connected power and energy storage action instructions;

[0093] C. Adaptively adjust the lower limit of SOC according to power support requirements and establish a hybrid energy storage optimization configuration model;

[0094] D. Based on the energy storage action command and the optimization configuration model, the internal power of the hybrid energy storage is allocated, and the optimization configuration result is obtained.

[0095] Step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage and utilizes the energy storage device to provide emergency power support when the system frequency decreases until the system frequency returns to a safe level.

[0096] Step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage, including determining the transfer function expression of the system frequency and the disturbance. The specific process is as follows:

[0097] First, assuming that the speed and frequency of a traditional generator are equal, without considering the load frequency regulation effect and damping coefficient, and that the automatic generation control (AGC) does not operate on the primary frequency regulation time scale;

[0098] Then, according to the synchronous motor rotor motion equation, the system frequency and the transfer function of the disturbance satisfy the following equation:

[0099]

[0100] In the formula, Δf * This is the per-unit value of the frequency deviation;

[0101] s is a complex frequency;

[0102] J eq The system's equivalent inertial time constant;

[0103] K * This is the per-unit value of the primary frequency modulation coefficient;

[0104] T s The inertial time constant of the unit;

[0105] This is the per-unit value for the incremental electromagnetic power of the unit.

[0106] Step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage, including the expression for frequency deviation. The specific process is as follows:

[0107] First, assume the power of the energy storage device configured in the system is P. s When the system frequency is below the critical point, the energy storage discharges at its maximum power to support the system frequency.

[0108] Then, let the above time be t1, then the frequency deviation Δf(t1) at time t1 relative to the rated value satisfies the following equation:

[0109] Δf(t1)=f t -f n

[0110] In the formula, f tThis is the critical frequency for energy storage operation;

[0111] f n This is the rated value for the power grid frequency.

[0112] Step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage, including establishing the system's frequency response expression. The specific process is as follows:

[0113] After energy storage is put into operation, the system frequency response satisfies the following equation:

[0114]

[0115] In the formula, This refers to the per-unit value of the incremental mechanical power of traditional generator sets;

[0116] The per-unit value of the output power used to provide emergency power support for energy storage.

[0117] Step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage, including determining the system frequency constraints. The specific process is as follows:

[0118] First, based on the system's operational limitations, frequencies below f... min This may cause the system to activate safety measures such as low-frequency load reduction, affecting normal production;

[0119] Then, the energy storage is configured such that, in the event of the worst-case power disturbance, the lowest point of the power system frequency is above f. min It satisfies the following equation:

[0120] f nadir ≥f min

[0121] In the formula, f nadir This is the point of lowest frequency during the disturbance process;

[0122] f min The lowest point at which the system frequency is allowed to drop in order to ensure system frequency stability.

[0123] Finally, by solving the above equations, the minimum supporting power P for energy storage is obtained. smin .

[0124] Step B, based on the renewable energy fluctuation mitigation standard, mitigates renewable energy fluctuations, resulting in renewable energy grid-connected power and energy storage action commands. This includes establishing the correlation between renewable energy grid-connected power and hybrid energy storage mitigation power. The specific process is as follows:

[0125] First, the original power output signal of the new energy source is decomposed into P0 to extract the grid-connected power P of the new energy source. grid With hybrid energy storage to mitigate power P HESS ;

[0126] Then, establish the grid-connected power P of new energy sources. grid With hybrid energy storage to mitigate power P HESS Related expressions:

[0127] (1) Construct the sequence P(i)=P0+ε1E1(W(i)) to obtain the first set of residuals R1=ε1N(P(i));

[0128] (2) The first modal component imf1 = P0 - R1 is obtained.

[0129] (3) Continue adding white noise and calculate the residual R of the kth group in sequence. k =N(R) k-1 +ε k-1 E(W(i))) and the k-th modal component imf k =R k-1 -R k until all modes and residuals are obtained.

[0130] Wherein, P0 is the signal to be decomposed;

[0131] E k (·) represents the k-th order modal component generated by EMD decomposition;

[0132] N(·) is the local mean of the generated signal;

[0133] R k For the k-th group of residuals;

[0134] ε k These are the weighting coefficients for the k-th white noise element;

[0135] imf k This is the k-th modal component;

[0136] W(i) is a Gaussian white noise with a mean of 0 and a variance of 1 for group i.

[0137] Step B, based on the renewable energy fluctuation smoothing standard, smooths out renewable energy fluctuations, obtaining renewable energy grid-connected power and energy storage action commands, including reconfiguration correlation expression. The specific process is as follows:

[0138] The decomposition results are reconstructed to satisfy the following equation:

[0139]

[0140] In the formula, k1 is the number of IMF components obtained from the decomposition;

[0141] res represents the residual;

[0142] P grid Contribute grid-connected power to new energy sources;

[0143] P HESS To smooth out the power of HESS;

[0144] n1 is the number of the boundary layer between grid-connected power and power smoothing;

[0145] i and j are the component numbers, respectively.

[0146] Step B, based on the renewable energy fluctuation mitigation standard, smooths out renewable energy fluctuations, resulting in renewable energy grid-connected power and energy storage action commands. This includes establishing power-based constraints, and the specific process is as follows:

[0147] The difference between the maximum and minimum power of actual grid-connected renewable energy within a certain time interval shall not exceed the fluctuation limit of renewable energy output, and shall satisfy the following equation:

[0148]

[0149] In the formula, t represents time, indicating the moment when the new energy source outputs power;

[0150] Δt is the time interval;

[0151] P gmax and P gmin These represent the lower limit of SOC, the maximum grid-connected power, and the minimum grid-connected power during this period, respectively.

[0152] P N For new energy installed capacity;

[0153] σ max This represents the maximum output fluctuation rate.

[0154] Step B involves smoothing out renewable energy fluctuations according to the renewable energy fluctuation smoothing standard, obtaining renewable energy grid-connected power and energy storage action commands, including determining the optimal number of boundary layers. The specific process is as follows:

[0155] First, calculate whether the low-frequency fluctuation components under each boundary layer number n1 meet the smoothing requirements. If they do, stop the loop.

[0156] Then, if the smoothness requirement is not met, the number of boundary layers is increased until the optimal n1 is found, and the new energy grid-connected power signal P is obtained. grid and HESS power signal P HESS .

[0157] Specifically, step C adaptively adjusts the lower limit of SOC based on power support requirements and establishes a hybrid energy storage optimization configuration model, including establishing the objective function and establishing constraints.

[0158] Specifically, the objective function is established, and the specific process is as follows:

[0159] Considering the initial investment cost, operation and maintenance cost, scrapping and recycling cost, and power deficit cost of HESS, a full life cycle model of HESS is established, with the minimum investment over the entire life cycle as the objective function for optimal configuration, satisfying the following equations:

[0160] minC=C i +C m +C d +C r +C q

[0161] In the formula, C i This is the initial investment cost, C m It is the operation and maintenance cost, C d It is the replacement cost, C r It is the cost of scrapping and recycling, C q It is the cost of power deficit.

[0162] Specifically, the initial investment cost satisfies the following equation:

[0163] C i =C eb E bm +C ec E cm +C pb P bm +C pc P cm

[0164] In the formula, C eb and C ec The initial investment cost per unit capacity for lithium iron phosphate batteries and supercapacitors are respectively, C. pb and C pc The initial investment unit power cost for lithium iron phosphate batteries and supercapacitors are respectively E bm and E cm The respective configuration capacities of lithium iron phosphate batteries and supercapacitors, P bm and P cm These are the power ratings for lithium iron phosphate batteries and supercapacitors, respectively.

[0165] Specifically, the operation and maintenance costs are calculated as a certain percentage of the initial capacity investment cost of HESS, satisfying the following equation:

[0166] C m =kmb C eb E bm +k mc C ec E cm

[0167] In the formula, k mb and k mc These are the estimated operating and maintenance coefficients for lithium iron phosphate batteries and supercapacitors, respectively.

[0168] Specifically, the replacement cost satisfies the following equation:

[0169]

[0170] In the formula, N is the number of times the lithium iron phosphate battery is replaced within the HESS operating cycle, T is the operating cycle of the lithium iron phosphate battery, r is the discount rate, and i represents the i-th replacement of the lithium iron phosphate battery.

[0171] Specifically, the cost of scrapping and recycling satisfies the following equation:

[0172]

[0173] In the formula, C ebs and C pbs α represents the disposal cost per unit capacity and per unit power of lithium iron phosphate batteries, respectively, and α is the HESS recycling residual value rate.

[0174] Specifically, the power deficit cost refers to the cost incurred due to the limited acceptance capacity of the HESS, which cannot fully follow the operation command after charging to the maximum SOC or discharging to the minimum SOC, resulting in a power deficit. It satisfies the following equation:

[0175]

[0176] In the formula, C qe E is the power deficit penalty factor. qe (t) represents the cumulative power deficit in year t.

[0177] Specifically, the constraints are established, and the specific process is as follows:

[0178] First, the charging and discharging power of the energy storage device cannot exceed its configured power. The charging and discharging power constraint of HESS satisfies the following equation:

[0179] -P bm ≤P b (t)≤P bm

[0180] -P cm ≤P c (t)≤P cm

[0181] In the formula, P b (t) and P c (t) represents the charging and discharging power of the lithium iron phosphate battery and the supercapacitor at time t, respectively. When the value is positive, the energy storage device is in a charging state, and when the value is negative, the energy storage device is in a discharging state.

[0182] Then, the lithium iron phosphate battery takes on the task of emergency power support. When a power disturbance occurs and the system frequency falls below the threshold value, the lithium iron phosphate battery immediately discharges at its maximum power. Therefore, the configured power cannot be lower than the minimum support power. Hence, this constraint satisfies the following equation:

[0183] P bm ≥P smin

[0184] In the formula, P smin It is the minimum supporting power for energy storage.

[0185] Furthermore, when the energy storage device is in a charging state, its SOC at time t satisfies the following equation:

[0186]

[0187] In the formula, SOC b and SOC c The SOC of the lithium iron phosphate battery and the supercapacitor are respectively, Δt is the unit time, and η is the state of charge. b and η c The figures represent the charge and discharge efficiencies of lithium iron phosphate batteries and supercapacitors, respectively.

[0188] Furthermore, when the energy storage device is in a discharging state, its SOC at time t satisfies the following equation:

[0189]

[0190] Furthermore, to prevent HESS from being overcharged and over-discharged, the SOC must satisfy the following equation:

[0191]

[0192] In the formula, and These represent the upper and lower limits of the state of charge (SOC) for lithium iron phosphate batteries. and These represent the upper and lower limits of the SOC of the supercapacitor.

[0193] Furthermore, in this invention, the lithium iron phosphate battery needs to undertake additional emergency power support tasks, and its SOC lower limit is treated as an optimization variable, subject to the constraint of the minimum frequency-supported power, which satisfies the following equation:

[0194]

[0195] In the formula, T represents the lower limit of the state of charge (SOC) of a lithium iron phosphate battery at its rated depth of discharge. fsup The duration for which emergency power support is needed for energy storage devices.

[0196] Furthermore, this invention employs a rainflow counting method to calculate the number of cycles at different discharge depths relative to a reference discharge depth. The conversion factor is then recorded. Therefore, it satisfies the following equation:

[0197]

[0198] In the formula, D a This is the actual depth of discharge, and the corresponding number of cycles is N. a ;D r It is the rated depth of discharge, N r It represents the rated number of discharges, and e is a natural constant.

[0199] Finally, the service life of lithium iron phosphate batteries satisfies the following equation:

[0200]

[0201] In the formula, S bess It refers to the configuration capacity of lithium iron phosphate batteries, E n It is the daily discharge amount, calculated using the rain flow counting method.

[0202] Specifically, step D allocates the internal power of the hybrid energy storage system based on the energy storage action command and the optimized configuration model, and obtains the optimized configuration result. The specific process is as follows:

[0203] First, the power P of HESS was smoothed again using the ICEEMDAN decomposition method. HESS Decompose and reconstruct;

[0204] Then, the low-frequency and high-frequency components, i.e., the lithium iron phosphate battery action signal P, are obtained cyclically under different boundary layer numbers. bess and supercapacitor action signal P capa ;

[0205] Then, the above optimization configuration model is substituted sequentially, and the optimal number of boundary layers is adaptively selected according to the objective function to perform internal power allocation of HESS and obtain the energy storage optimization configuration result.

[0206] Specifically, in step B, according to the Chinese national standard "Technical Regulations for Wind Farm Access to Power Systems", the original power output signal of the new energy source is decomposed into P0 using the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) method.

[0207] Specifically, step B calculates whether the low-frequency fluctuation components under each boundary layer number n1 meet the smoothing requirements, and the verification requirements are in accordance with the "Technical Regulations for Wind Farm Access to Power System".

[0208] To consider the emergency power support from energy storage and the demand for smoothing fluctuations from new energy sources, as well as to meet economic indicators, this invention first analyzes the dynamic frequency characteristics of the system to determine the power support demand from energy storage. Second, it employs an improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) algorithm to decompose the original output of new energy sources, obtaining the HESS power smoothing demand. Then, considering the power demand for emergency power support, it optimizes the lower limit of the state of charge (SOC) of the energy storage device during normal operation to achieve adaptive adjustment, establishes a life-cycle cost model for the hybrid energy storage system, and obtains the economically optimal configuration result for hybrid energy storage.

[0209] Specifically, in step A, when a power disturbance occurs in the system, the generator units will suppress the change in grid frequency by performing primary frequency regulation. However, the decoupling of new energy units from the grid frequency reduces the equivalent inertia of the power system, leading to insufficient primary frequency regulation capability and even causing the system to perform low-frequency load shedding. Therefore, power support equipment should be provided for high-proportion new energy systems to minimize the impact of measures such as generator tripping and load shedding on the stable operation of the system.

[0210] This invention utilizes an energy storage device to provide emergency power support when the system frequency drops until the system frequency returns to a safe level. This process requires the energy storage device to continuously discharge, and it requires that the energy storage device have sufficient remaining power when providing power support. Therefore, a certain amount of power needs to be reserved during normal operation.

[0211] In the lithium iron phosphate battery-supercapacitor hybrid energy storage system of the present invention, the supercapacitor is a power-type energy storage device with high capacity configuration cost and is not suitable for providing energy for long-term discharge. Therefore, the present invention selects lithium iron phosphate battery energy storage to undertake the task of emergency power support.

[0212] During configuration, the reserved capacity of energy storage is determined by the lower limit of its SOC during normal operation. A lower value will significantly increase the configuration capacity. Therefore, this invention uses the lower limit of SOC during normal operation of energy storage as an optimization variable to reduce the configuration capacity while simultaneously meeting the needs of emergency power support and power stabilization.

[0213] Specifically, solving the system frequency response equation in step A yields the change in system frequency after energy storage is integrated. It can be seen that the power support provided by energy storage effectively reduces system power disturbances, thereby increasing the minimum point and steady-state value in the frequency response process.

[0214] Specifically, in step A, the present invention sets the minimum dynamic frequency point to 47.5Hz, that is, limits the dynamic frequency to drop by a maximum of 2.5Hz.

[0215] Specifically, in step C, the power-type energy storage can be cycled 500,000 to 1,000,000 times, far exceeding that of energy-type energy storage. Therefore, the lifespan of the power-type medium can be set as a constant. However, under actual operating conditions, excessively long operating years of supercapacitors will reduce safety. Therefore, this invention sets the lifespan of supercapacitor energy storage to 20 years.

[0216] Unlike power storage devices with a relatively fixed lifespan, the lifespan of lithium iron phosphate batteries is significantly affected by factors such as operating temperature and depth of discharge. This invention primarily considers the calculation method for the equivalent cycle life of lithium iron phosphate batteries, taking only the depth of discharge D into account. a The impact on battery cycle life: For a given discharge cycle, the actual discharge capacity depends on the ratio of the actual depth of discharge to the rated depth of discharge. Under the same operating conditions, the number of cycles is a decreasing function of the depth of discharge; the greater the depth of discharge, the shorter the cycle life.

[0217] The technical features of this invention are as follows:

[0218] This invention comprehensively considers the needs for emergency power support from energy storage and the need to mitigate fluctuations in new energy sources, and optimizes the configuration of energy storage capacity based on economic efficiency; determines the minimum support power of energy storage based on the system's dynamic frequency response; optimizes the lower limit of the state of charge (SOC) of the energy storage device during normal operation, considering the power demand for emergency power support, to achieve adaptive adjustment; uses an adaptive improved fully ensemble empirical mode algorithm to decompose the original output of new energy sources; and employs hybrid energy storage to mitigate fluctuations.

[0219] Example 1

[0220] Using the unit data and actual daily power generation data of an independent power grid with a wind power installed capacity of 25MW and a new energy penetration rate of over 40%, with a wind power sampling interval of 1 minute, the hybrid energy storage optimization configuration method of the present invention is simulated and verified based on the MATLAB platform.

[0221] The equivalent frequency regulation coefficient of the system is K = 27.68 MW / Hz, and the generator's equivalent inertial time constant is J. eq = 3.83s, Unit inertial time constant T s =4s, when the system is in steady state, the system frequency is 50Hz.

[0222] Furthermore, considering the most severe fault scenario of the power grid, the maximum power disturbance is the shutdown of all 25MW wind turbine units. The energy storage is required to immediately discharge at maximum power when the system frequency is below 49.5Hz and maintain this discharge for at least 5 minutes, ensuring that the system frequency is not lower than 47.5Hz at any time.

[0223] Regarding fluctuation smoothing standards, the maximum limits for output power variation within 1 minute and 10 minutes of wind farms are specified in the Chinese national standard "Technical Regulations for Wind Farm Connection to Power System", as shown in Table 1.

[0224] Table 1 Maximum Limits of Wind Farm Power Variation

[0225]

[0226] Assuming a 25MW power disturbance occurs at t=0, the system frequency will drop to 49.5Hz after 0.113 seconds. At this point, the energy storage system must immediately activate and provide at least 8.1MW of power to maintain a system frequency above 47.5Hz. Simulations verify this by substituting the values, and the resulting dynamic frequency response curves before and after 8.1MW of energy storage are shown below. Figure 2 As shown, the lowest system frequency before storage allocation was below 47.5Hz, while after storage allocation, the lowest system frequency increased to above 47.5Hz. This effectively improves system stability and avoids triggering low-frequency load shedding, which is consistent with theoretical results.

[0227] The original wind power signal was decomposed using both traditional EMD and ICEEMDAN methods, and the effect after being smoothed by the lithium iron phosphate battery-supercapacitor hybrid energy storage system of this invention is as follows: Figure 3 As shown.

[0228] Depend on Figure 3It is evident that the original output power of wind power fluctuates significantly before energy storage is applied, with a maximum fluctuation rate of 66.49% in 1 minute and 71.34% in 10 minutes. After decomposition using the EMD and ICEEMDAN algorithms and smoothing by hybrid energy storage, the grid-connected power becomes smoother, and high-frequency fluctuations are significantly reduced. The maximum fluctuation rates in 1 minute are reduced to 3.51% and 3.29%, respectively, and in 10 minutes to 29.74% and 28.21%, respectively. This demonstrates that ICEEMDAN's smoothing effect is slightly better than the traditional EMD algorithm, and hybrid energy storage effectively reduces wind power fluctuations, improving system stability and the ability to absorb wind power.

[0229] Considering the need for energy storage to participate in emergency power support and smooth out fluctuations in new energy output, the capacity and power of the lithium iron phosphate battery-supercapacitor hybrid energy storage system are configured based on the HESS optimization configuration model proposed in this invention. The optimized configuration parameters are shown in Table 2.

[0230] Table 2 HESS Optimization Configuration Parameters

[0231]

[0232] To illustrate the rationality of the hybrid energy storage system, this invention sets up three schemes for simulation verification:

[0233] Option 1: Configure lithium iron phosphate battery energy storage and supercapacitor energy storage, with the lithium iron phosphate battery undertaking the task of emergency power support.

[0234] Option 2: Configure only lithium iron phosphate batteries for energy storage and use them for emergency power support.

[0235] Option 3: Configure only supercapacitor energy storage and assign it the task of emergency power support.

[0236] The configuration results for the three schemes are shown in Table 3, and the motion curves and SOC curves are as follows: Figure 4 , Figure 5 , Figure 6 As shown.

[0237] Table 3 Optimization results of the three schemes

[0238]

[0239]

[0240] It can be seen that among the three schemes, Scheme 1, which uses hybrid energy storage, is the most economical, while Scheme 3, which uses only supercapacitors, is the least economical. This is because supercapacitors have a high unit capacity cost and a large initial investment cost. Compared to the hybrid energy storage scheme, supercapacitors need to additionally undertake the task of low-frequency power smoothing and dynamic frequency support, which greatly increases their configuration capacity and power, resulting in higher costs. Scheme 2, which uses only lithium iron phosphate batteries, is more economical than Scheme 3, but less economical than Scheme 1. This is because the unit capacity and power cost of lithium iron phosphate batteries are significantly lower than those of supercapacitors, thus significantly reducing investment costs. However, because lithium iron phosphate batteries additionally undertake the task of smoothing high-frequency fluctuations, they require high-frequency charging and discharging, resulting in a significantly shortened lifespan and the need for frequent replacements, thereby increasing the overall configuration cost over the lifespan of the energy storage system.

[0241] In summary, a reasonable configuration of lithium iron phosphate batteries and supercapacitors allows for complementary advantages between the two energy storage methods, improving the system's economic efficiency throughout its entire lifecycle. Compared to configuring lithium iron phosphate batteries or supercapacitors alone, the hybrid energy storage system used in this invention reduces the total lifecycle cost by 0.09 billion yuan and 0.14 billion yuan, respectively, fully leveraging the advantages of both energy storage methods and achieving superior economic performance.

[0242] Thus, the task of the proposed method for optimizing hybrid energy storage configuration with SOC adaptive adjustment has been completed.

[0243] This invention comprehensively considers the needs for emergency power support from energy storage and the need to mitigate fluctuations in new energy sources. Furthermore, it optimizes the configuration of energy storage capacity based on economic efficiency. The energy storage configuration result, which takes into account multiple factors, is more in line with the actual needs of engineering projects.

[0244] This invention determines the minimum support power for energy storage based on the system's dynamic frequency response, taking into account worst-case conditions to improve the applicability of the minimum support power; considering the power demand for emergency power support, it optimizes the lower limit of the state of charge (SOC) of the energy storage device during normal operation to achieve adaptive adjustment.

[0245] This invention employs an adaptive improved complete set empirical mode algorithm to decompose the original output of new energy sources, achieving a smoothing effect slightly better than the traditional EMD algorithm. It also uses hybrid energy storage to smooth fluctuations and allocates energy storage action commands to different energy storage systems based on a hybrid energy storage optimization configuration model, thus fully leveraging the characteristics of different energy storage systems.

Claims

1. A method for optimizing the configuration of hybrid energy storage with adaptive SOC adjustment, characterized in that: Includes the following steps: A. Calculate the frequency response of the system under maximum disturbance, and determine the minimum support power of energy storage based on the frequency stability index; B. Based on the new energy fluctuation smoothing standard, the new energy fluctuation is smoothed to obtain the new energy grid-connected power and energy storage action instructions; C. Adaptively adjust the lower limit of SOC according to power support requirements and establish a hybrid energy storage optimization configuration model; D. Based on the energy storage action command and the optimization configuration model, allocate the internal power of the hybrid energy storage and obtain the optimization configuration result; Step C adaptively adjusts the lower limit of SOC according to power support requirements and establishes a hybrid energy storage optimization configuration model, including establishing the objective function and establishing constraints. The objective function is established as follows: Considering the initial investment cost, operation and maintenance cost, scrapping and recycling cost, and power deficit cost of HESS, a full life cycle model of HESS is established, with the minimum investment over the entire life cycle as the objective function for optimal configuration, satisfying the following equations: minC=C i +C m +C d +C r +C q In the formula, C i This is the initial investment cost, C m It is the operating and maintenance cost, C d It is the replacement cost, C r It is the cost of scrapping and recycling, C q It is the cost of power deficit; The specific process for establishing constraints is as follows: First, the charging and discharging power of the energy storage device cannot exceed its configured power. The charging and discharging power constraint of HESS satisfies the following equation: -P bm ≤P b (t)≤P bm -P cm ≤P c (t)≤P cm In the formula, P b (t) and P c (t) represents the charging and discharging power of the lithium iron phosphate battery and the supercapacitor at time t, respectively. When the value is positive, the energy storage device is in a charging state; when the value is negative, the energy storage device is in a discharging state. bm and P cm These are the power configurations for lithium iron phosphate batteries and supercapacitors, respectively. Then, the lithium iron phosphate battery takes on the task of emergency power support. When a power disturbance occurs and the system frequency falls below the threshold value, the lithium iron phosphate battery immediately discharges at its maximum power. Therefore, the configured power cannot be lower than the minimum support power. Hence, this constraint satisfies the following equation: P bm ≥P smin In the formula, P smin It is the minimum supporting power for energy storage; Furthermore, when the energy storage device is in a charging state, its SOC at time t satisfies the following equation: In the formula, SOC b and SOC c The SOC of the lithium iron phosphate battery and the supercapacitor are respectively, Δt is the unit time, and η is the state of charge. b and η c The charging and discharging efficiencies of lithium iron phosphate batteries and supercapacitors are respectively; E bm and E cm These are the configuration capacities of lithium iron phosphate batteries and supercapacitors, respectively. Furthermore, when the energy storage device is in a discharging state, its SOC at time t satisfies the following equation: Furthermore, to prevent HESS from being overcharged and over-discharged, the SOC must satisfy the following equation: In the formula, and These represent the upper and lower limits of the state of charge (SOC) for lithium iron phosphate batteries. and These are the upper and lower limits of the SOC of the supercapacitor; Furthermore, lithium iron phosphate batteries need to undertake additional emergency power support tasks, and their SOC lower limit is treated as an optimization variable, constrained by the minimum frequency-supported power, which satisfies the following equation: In the formula, T represents the lower limit of the state of charge (SOC) of a lithium iron phosphate battery at its rated depth of discharge. fsup The duration for which emergency power support is needed for energy storage devices.

2. The method for optimizing the configuration of hybrid energy storage with adaptive SOC adjustment according to claim 1, characterized in that: Step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage and utilizes the energy storage device to provide emergency power support when the system frequency decreases until the system frequency returns to a safe level.

3. The method for optimizing the configuration of hybrid energy storage with adaptive SOC adjustment according to claim 1, characterized in that: Step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage, including determining the transfer function expression of the system frequency and the disturbance. The specific process is as follows: First, assuming that the speed and frequency of a traditional generator are equal, without considering the load frequency regulation effect and damping coefficient, and that the automatic generation control (AGC) does not operate on the primary frequency regulation time scale; Then, according to the synchronous motor rotor motion equation, the system frequency and the transfer function of the disturbance satisfy the following equation: In the formula, Δf * This is the per-unit value of the frequency deviation; s is a differential operator; J eq The system's equivalent inertial time constant; K * This is the per-unit value of the primary frequency modulation coefficient; T s The inertial time constant of the unit; This is the per-unit value for the incremental electromagnetic power of the unit.

4. The method for optimizing the configuration of hybrid energy storage with adaptive SOC adjustment according to claim 3, characterized in that: Step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage, including the expression for frequency deviation. The specific process is as follows: First, assume the power of the energy storage device configured in the system is P. s When the system frequency is below the critical point, the energy storage discharges at its maximum power to support the system frequency. Then, let the above time be t1, then the frequency deviation Δf(t1) at time t1 relative to the rated value satisfies the following equation: Δf(t1)=f t -f n In the formula, f t This is the critical frequency for energy storage operation; f n This is the rated value for the power grid frequency.

5. The method for optimizing the configuration of hybrid energy storage with adaptive SOC adjustment according to claim 4, characterized in that: Step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage, including establishing the system's frequency response expression. The specific process is as follows: After energy storage is put into operation, the system frequency response satisfies the following equation: In the formula, This refers to the per-unit value of the incremental mechanical power of traditional generator sets; The per-unit value of the output power used to provide emergency power support for energy storage.

6. The method for optimizing the configuration of hybrid energy storage with adaptive SOC adjustment according to claim 5, characterized in that: Step A calculates the system's frequency response under maximum disturbance. Based on the frequency stability index, it determines the minimum support power for energy storage, including determining the system frequency constraints. The specific process is as follows: First, based on the system's operational limitations, frequencies below f... min This may cause the system to activate safety measures such as low-frequency load reduction, affecting normal production; Then, the energy storage is configured such that, in the event of the worst-case power disturbance, the lowest point of the power system frequency is above f. min It satisfies the following equation: f nadir ≥f min In the formula, f nadir This is the point of lowest frequency during the disturbance process; f min The lowest point at which the system frequency is allowed to drop, in order to ensure system frequency stability; Finally, by solving the above equations, the minimum supporting power P for energy storage is obtained. smin .

7. The method for optimizing the configuration of hybrid energy storage with adaptive SOC adjustment according to claim 1, characterized in that: Step B, based on the renewable energy fluctuation mitigation standard, mitigates renewable energy fluctuations, resulting in renewable energy grid-connected power and energy storage action commands. This includes establishing the correlation between renewable energy grid-connected power and hybrid energy storage mitigation power. The specific process is as follows: First, the original power output signal of the new energy source is decomposed into P0 to extract the grid-connected power P of the new energy source. grid With hybrid energy storage to mitigate power P HESS ; Then Establish new energy grid-connected power P grid With hybrid energy storage to mitigate power P HESS Related expressions: (1) Construct the sequence P(i)=P0+ε1E1(W(i)) to obtain the first set of residuals R1=ε1N(P(i)); (2) The first modal component imf1 = P0 - R1 is obtained; (3) Continue adding white noise and calculate the residual R of the kth group in sequence. k =N(R) k-1 +ε k-1 E(W(i))) and the k-th modal component imf k =R k-1 -R k until all modes and residuals are obtained. Wherein, P0 is the signal to be decomposed; E k (·) represents the k-th order modal component generated by EMD decomposition; N(·) is the local mean of the generated signal; R k For the k-th group of residuals; ε k These are the weighting coefficients for the k-th white noise element; imf k This is the k-th modal component; W(i) is a Gaussian white noise with a mean of 0 and a variance of 1 for group i.

8. The method for optimizing the configuration of hybrid energy storage with adaptive SOC adjustment according to claim 1, characterized in that: Step B, based on the renewable energy fluctuation smoothing standard, smooths out renewable energy fluctuations, obtaining renewable energy grid-connected power and energy storage action commands, including reconfiguration correlation expression. The specific process is as follows: The decomposition results are reconstructed to satisfy the following equation: In the formula, k1 is the number of IMF components obtained from the decomposition; res represents the residual; P grid Contribute grid-connected power to new energy sources; P HESS To smooth out the power of HESS; n1 is the number of the boundary layer between grid-connected power and power smoothing; i and j are the component numbers, respectively.

9. The method for optimizing the configuration of hybrid energy storage with adaptive SOC adjustment according to claim 1, characterized in that: Step B, based on the renewable energy fluctuation mitigation standard, smooths out renewable energy fluctuations, resulting in renewable energy grid-connected power and energy storage action commands. This includes establishing power-based constraints, and the specific process is as follows: The difference between the maximum and minimum power of actual grid-connected renewable energy within a certain time interval shall not exceed the fluctuation limit of renewable energy output, and shall satisfy the following equation: In the formula, t represents time, indicating the moment when the new energy source outputs power; Δt is the time interval; P gmax and P gmin These represent the maximum and minimum grid-connected power during the lower limit period of SOC, respectively; P N For new energy installed capacity; σ max This represents the maximum output fluctuation rate.

10. The method for optimizing the configuration of hybrid energy storage with adaptive SOC adjustment according to claim 1, characterized in that: Step B involves smoothing out renewable energy fluctuations according to the renewable energy fluctuation smoothing standard, obtaining renewable energy grid-connected power and energy storage action commands, including determining the optimal number of boundary layers. The specific process is as follows: First, calculate whether the low-frequency fluctuation components under each boundary layer number n1 meet the smoothing requirements. If they do, stop the loop. Then, if the smoothness requirement is not met, the number of boundary layers is increased until the optimal n1 is found, and the new energy grid-connected power signal P is obtained. grid and HESS power signal P HESS .

Images