阅读说明：本技术 基于ga的储能系统优化配置方法 (Energy storage system optimal configuration method based on GA ) 是由 付林 廖孟柯 荆世博 李海峰 贾政豪 于 2021-07-19 设计创作，主要内容包括：本发明涉及储能系统优化方法技术领域,是一种基于GA的储能系统优化配置方法,本发明所述方法依据风电并网功率波动要求,采用变分模态分解方法将需要平抑的功率分为低频和高频2部分,分别使用蓄电池和超级电容进行平抑,建立储能充放电控制模型；考虑储能装置建设成本及弃风弃光和平抑不足带来的惩罚性费用,建立混合储能优化配置模型；以典型日数据作为基础,采用遗传算法对混合储能优化配置模型进行求解,得出全年最低成本及对应的储能额定功率与额定容量。从而对储能系统优化配置,能够实现储能设备的经济性充分利用。(The invention relates to the technical field of energy storage system optimization methods, in particular to a GA-based energy storage system optimization configuration method, which is characterized in that according to the wind power grid-connected power fluctuation requirement, a variational modal decomposition method is adopted to divide power to be stabilized into a low-frequency part and a high-frequency part 2, a storage battery and a super capacitor are respectively used for stabilizing, and an energy storage charging and discharging control model is established; considering the construction cost of the energy storage device and punitive cost brought by wind abandoning and light abandoning and insufficient stabilization, establishing a hybrid energy storage optimization configuration model; and (3) solving the hybrid energy storage optimal configuration model by adopting a genetic algorithm on the basis of typical daily data to obtain the lowest annual cost and the corresponding energy storage rated power and rated capacity. Therefore, the energy storage system is optimally configured, and the economical efficiency of the energy storage equipment can be fully utilized.)

1. A GA-based energy storage system optimal configuration method is characterized in that a variational modal decomposition method is adopted to divide wind power to be stabilized into a low-frequency part and a high-frequency part 2, a storage battery and a super capacitor are respectively used for stabilizing, and an energy storage charging and discharging control model is established; considering the construction cost of the energy storage device and punitive cost brought by wind abandoning and light abandoning and insufficient stabilization, establishing a hybrid energy storage optimization configuration model; and (3) solving the hybrid energy storage optimal configuration model by adopting a genetic algorithm on the basis of typical daily data to obtain the lowest annual cost and the corresponding energy storage rated power and rated capacity.

2. The GA-based energy storage system optimal configuration method according to claim 1, wherein the method for establishing the energy storage charging and discharging control model comprises the following steps:

for a wind-solar energy storage power plant, the charging and discharging mathematical model of the corresponding wind-solar energy storage and storage system is as follows:

wherein P is_nRepresentative of the power demand, P_sRepresentational sunPower generated by solar energy, P_wRepresenting the power generated by wind energy;

when the sum of the power generated by wind energy and solar energy is greater than the required power, the redundant power is dumped to the energy storage system;

P_in＝P_s+P_w-P_n (2)

P_inrepresenting the reasonable distribution of the input power of the energy storage system to the load interval, the division method is as follows:

the trough interval:

[P_min,P_min+(P_s+P_w)] (3)

peak interval:

[P_max-γ(P_s+P_w),P_max] (4)

the general interval:

[P_min+γ(P_s+P_w),P_max-γ(P_s+P_w)] (5)

wherein P is_minIs the minimum output load, P, of the energy storage system_maxGamma is a redundancy coefficient generated in the conversion or transmission process; in order to distinguish the purposes of two energy storage devices, namely a super capacitor and a storage battery, for reasonable distribution, energy storage power needs to be decomposed and redistributed, and VMD is adopted for operation;

namely, decomposing the reference power of the energy storage system into k components u_k(t) for each u_k(t) performing a Hilbert transform to obtain a single-sided spectrum,

is modulated to the corresponding base frequency band and,

and then calculates the norm of the gradient of the computational signal, and estimates each bandwidth,

wherein the content of the first and second substances,

{u_k}:＝{u₁Λu_k}{ω_k}:＝{{ω₁Λ{ω_k}

converting equation (8) to an unconstrained variant:

in the formula (10), α is a secondary penalty factor; λ is the Lagrangian multiplier;

{u_k},{ω_kthe updates of λ are as follows:

replacement of omega by omega-omega_k，

The integration form is changed into a non-negative frequency interval integration form,

the solution is further changed into the solution,

the center frequency is converted to the frequency domain,

the updating method comprises the following steps:

the lambda updating method comprises the following steps:

the iteration is terminated satisfying the following relationship:

thus obtaining the low-frequency component suitable for the storage battery and the high-frequency component suitable for the super capacitor.

3. A GA-based energy storage system optimal configuration method according to claim 1 or 2, wherein the method for establishing the hybrid energy storage optimal configuration model comprises the following steps:

photovoltaic array M considering purchase cost required by energy storage system_sWind turbine generator system M_sThe purchase cost of the energy storage system is mainly the storage battery M_baSuper capacitor M_scPunishment cost M brought by insufficient stabilizing fluctuation of power generation system_laAnd punishment cost M caused by light and wind abandonment_abThe objective function of the acquisition cost is:

M_min＝M_ba+M_sc+M_la+M_ab (19)

because of the unequal life of the energy storage devices, the acquisition cost is converted here to the equivalent in years as follows:

wherein r is the conversion ratio, Y_scFor the lifetime of the supercapacitor, Y_baFor the life of the accumulator, Y_p-scFor the unit power cost of the supercapacitor, f_p-baFor the cost per unit power of the accumulator, f_e-scCost per unit capacity of supercapacitor, f_e-baFor the cost per unit capacity of the accumulator, P_baFor the rated power of the accumulator, P_scRated power for super capacitor, E_baFor rated capacity of the battery, E_scThe rated capacity of the super capacitor is obtained;

punishment cost M brought by insufficient stabilizing fluctuation of power generation system_laAnd punishment cost M caused by light and wind abandonment_abThe expression is as follows:

M_la＝αE_la (22)

M_la＝αE_la (23)

wherein E_laTo balance the power caused by the deficiency, E_abIn order to discard the electric quantity wasted by the light and wind, alpha is unit loss cost caused by insufficient stabilization, and beta is unit cost wasted by the light and wind discarding;

the constraint conditions are as follows:

|P_ba|≤P_ba·max (24)

|P_sc|≤P_sc·max (25)

C_ba·min≤C_ba≤C_ba·max (26)

C_sc·min≤C_sc≤C_sc·max (27)

wherein P is_baIs the power of the accumulator, P_scIs the power of a super capacitor, C_baFor the stored charge of the battery, C_scThe storage capacity of the super capacitor;

s^ch+s^di≤1 (28)

s^chindicating a charge indicating variable, s, of the energy storage device^diAnd (2) a discharge indicating variable of the energy storage device, wherein the formula (28) indicates that the energy storage device cannot simultaneously operate in a charging and discharging state.

4. A GA-based energy storage system optimal configuration method according to claim 1 or 2, wherein the method for solving the hybrid energy storage optimal configuration model by using a genetic algorithm comprises the following steps:

step 1: determining expected grid-connected power of wind power and photovoltaic and stabilizing required power fluctuation according to actual data, and initializing power and capacity of an energy storage system;

step 2: decomposing the desired flat power fluctuation into a series of subcomponents using VMD (i.e., variational modal decomposition);

and step 3: the decomposed series of subcomponents are assigned to high-frequency and low-frequency components;

and 4, step 4: randomly generating a population and updating the population;

and 5: cross-swapping to produce children;

step 6: carrying out gene variation;

and 7: expressing the characters;

and 8: natural selection is carried out by utilizing a gambling wheel;

and step 9: searching the optimal individual and returning to the step 4;

step 10: and exiting after the iteration times are finished.

5. A GA-based energy storage system optimal configuration method according to claim 3, wherein the solution method for the hybrid energy storage optimal configuration model using the genetic algorithm comprises the steps of:

step 1: determining expected grid-connected power of wind power and photovoltaic and stabilizing required power fluctuation according to actual data, and initializing power and capacity of an energy storage system;

step 2: decomposing the desired flat power fluctuation into a series of subcomponents using VMD (i.e., variational modal decomposition);

and step 3: the decomposed series of subcomponents are assigned to high-frequency and low-frequency components;

and 4, step 4: randomly generating a population and updating the population;

and 5: cross-swapping to produce children;

step 6: carrying out gene variation;

and 7: expressing the characters;

and 8: natural selection is carried out by utilizing a gambling wheel;

and step 9: searching the optimal individual and returning to the step 4;

step 10: and exiting after the iteration times are finished.

Technical Field

The invention relates to the technical field of energy storage system optimization methods, in particular to a GA-based energy storage system optimization configuration method.

Background

According to the power generation characteristics of new energy, an independent power grid formed by renewable energy is constructed, and the method has important significance for full utilization of the new energy. At present, most researches are discussed and analyzed aiming at an island operation environment, the renewable energy source networking is indispensable in consideration of remote transportation and full utilization, the scheduling requirement of a large power grid on an independent power grid can be met in a networking mode, and the operation interference on the large power grid is reduced. And the traditional scheduling strategy can not be realized for the fluctuation of the renewable energy power generation, so that the utilization rate of the renewable energy can be maximized by combining the energy storage configuration and the operation strategy, the further application of the new energy power generation technology is promoted, and the intellectualization of the power grid is improved, thereby having important value.

The main methods for optimizing energy storage systems at home and abroad are as follows:

the application of a super capacitor energy storage system in a micro-grid [ J ]. Shaanxi electric power, 2010,38(08):12-6 ] in document 1 (Lele, Huangwei, Marcel.) uses simulation to prove the advantages of super capacitors in improving voltage sag and impact load and provide a method for selecting super capacitor parameters. In document 2 (wangliang, george, and Bingqiang, strong, energy storage type direct-drive permanent magnet synchronous wind power generation control system [ J ]. power system protection and control, 2010,38(14):43-8+78.) to solve the balance of low voltage ride through and cost, vanadium battery energy storage is used as a solution, and feasibility of stabilizing power fluctuation, dynamic response and low voltage ride through of a wind power system is verified through simulation. Document 3 (Dingming, Wangweisen, Wangxulie, etc.. Large-scale photovoltaic power generation influences on an electric power system are reviewed [ J ]. Chinese Motor engineering Proc., 2014,34(01):1-14.) the influences of reactive voltage, active frequency, electric energy quality, attack angle stability and power distribution system protection are analyzed by utilizing a modeling simulation method.

Disclosure of Invention

The invention provides a GA-based energy storage system optimal configuration method, which overcomes the defects of the prior art, optimizes the energy storage system, and can realize the full utilization of the economy of energy storage equipment.

The technical scheme of the invention is realized by the following measures: a GA-based energy storage system optimal configuration method comprises the steps of dividing wind power to be stabilized into a low-frequency part and a high-frequency part 2 by adopting a variational modal decomposition method, stabilizing by using a storage battery and a super capacitor respectively, and establishing an energy storage charging and discharging control model; considering the construction cost of the energy storage device and punitive cost brought by wind abandoning and light abandoning and insufficient stabilization, establishing a hybrid energy storage optimization configuration model; and (3) solving the hybrid energy storage optimal configuration model by adopting a genetic algorithm on the basis of typical daily data to obtain the lowest annual cost and the corresponding energy storage rated power and rated capacity.

The following is further optimization or/and improvement of the technical scheme of the invention:

the method for establishing the energy storage charging and discharging control model comprises the following steps:

for a wind-solar energy storage power plant, the lowest discharge power of an energy storage system cannot be lower than the maximum required power, so that the charging and discharging mathematical model of the corresponding wind-solar energy storage system is as follows:

wherein P is_nRepresentative of the power demand, P_sRepresenting the power generated by solar energy, P_wRepresenting the power generated by wind energy;

when the sum of the power generated by wind energy and solar energy is greater than the required power, the redundant power is dumped to the energy storage system;

P_in＝P_s+P_w-P_n (2)

P_inrepresents a rational distribution of the input power of the energy storage system to the load intervals, which isThe primary goal of the wind, light and storage integrated charging and discharging goal is achieved through control. The dividing method comprises the following steps:

the trough interval:

[P_min,P_min+(P_s+P_w)] (3)

peak interval:

[P_max-γ(P_s+P_w),P_max] (4)

the general interval:

[P_min+γ(P_s+P_w),P_max-γ(P_s+P_w)] (5)

wherein P is_minIs the minimum output load, P, of the energy storage system_maxFor the maximum output load of the energy storage system, γ is a redundancy coefficient generated during conversion or transmission, and may be set to 1.2.

In order to distinguish the purposes of two energy storage devices, namely a super capacitor and a storage battery, for reasonable distribution, energy storage power needs to be decomposed and redistributed, and VMD is adopted for operation;

namely, decomposing the reference power of the energy storage system into k components u_k(t) for each u_k(t) performing a Hilbert transform to obtain a single-sided spectrum,

is modulated to the corresponding base frequency band and,

and then calculates the norm of the gradient of the computational signal, and estimates each bandwidth,

wherein the content of the first and second substances,

{u_k}:＝{u₁Λu_k}{ω_k}:＝{{ω₁Λ{ω_k}

converting equation (8) to an unconstrained variant:

in the formula (10), α is a secondary penalty factor; λ is the Lagrangian multiplier;

{u_k},{ω_kthe updates of λ are as follows:

replacement of omega by omega-omega_k，

The integration form is changed into a non-negative frequency interval integration form,

the solution is further changed into the solution,

the center frequency is converted to the frequency domain,

the updating method comprises the following steps:

the lambda updating method comprises the following steps:

the iteration is terminated satisfying the following relationship:

the mode center frequencies are ensured to be dissimilar, thereby preventing the over-decomposition phenomenon. And reasonably selecting and processing each decomposed mode to obtain a low-frequency component suitable for the storage battery and a high-frequency component suitable for the super capacitor.

The method for establishing the hybrid energy storage optimal configuration model comprises the following steps:

photovoltaic array M considering purchase cost required by energy storage system_sWind turbine generator system M_sThe purchase cost of the energy storage system is mainly the storage battery M_baSuper capacitor M_scPunishment cost M brought by insufficient stabilizing fluctuation of power generation system_laAnd punishment cost M caused by light and wind abandonment_abThe objective function of the acquisition cost is:

M_min＝M_ba+M_sc+M_la+M_ab (19)

because of the unequal life of the energy storage devices, the acquisition cost is converted here to the equivalent in years as follows:

whereinr is the conversion rate, Y_scFor the lifetime of the supercapacitor, Y_baFor the life of the accumulator, Y_p-scFor the unit power cost of the supercapacitor, f_p-baFor the cost per unit power of the accumulator, f_e-scCost per unit capacity of supercapacitor, f_e-baFor the cost per unit capacity of the accumulator, P_baFor the rated power of the accumulator, P_scRated power for super capacitor, E_baFor rated capacity of the battery, E_scThe rated capacity of the super capacitor is obtained;

punishment cost M brought by insufficient stabilizing fluctuation of power generation system_laAnd punishment cost M caused by light and wind abandonment_abThe expression is as follows:

M_la＝αE_la (22)

M_la＝αE_la (23)

wherein E_laTo balance the power caused by the deficiency, E_abIn order to discard the electric quantity wasted by the light and wind, alpha is unit loss cost caused by insufficient stabilization, and beta is unit cost wasted by the light and wind discarding;

the constraint conditions are as follows:

|P_ba|≤P_ba·max (24)

|P_sc|≤P_sc·max (25)

C_ba·min≤C_ba≤C_ba·max (26)

C_sc·min≤C_sc≤C_sc·max (27)

wherein P is_baIs the power of the accumulator, P_scIs the power of a super capacitor, C_baFor the stored charge of the battery, C_scThe storage capacity of the super capacitor;

s^ch+s^di≤1 (28)

s^chindicating a charge indicating variable, s, of the energy storage device^diAnd (2) a discharge indicating variable of the energy storage device, wherein the formula (28) indicates that the energy storage device cannot simultaneously operate in a charging and discharging state.

The method for solving the hybrid energy storage optimization configuration model by adopting the genetic algorithm comprises the following steps:

step 1: determining expected grid-connected power of wind power and photovoltaic and stabilizing required power fluctuation according to actual data, and initializing power and capacity of an energy storage system;

step 2: decomposing the desired flat power fluctuation into a series of subcomponents using VMD (i.e., variational modal decomposition);

and step 3: the decomposed series of subcomponents are assigned to high-frequency and low-frequency components;

and 4, step 4: randomly generating a population and updating the population;

and 5: cross-swapping to produce children;

step 6: carrying out gene variation;

and 7: expressing the characters;

and 8: natural selection is carried out by utilizing a gambling wheel;

and step 9: searching the optimal individual and returning to the step 4;

step 10: and exiting after the iteration times are finished.

According to the method, according to the fluctuation requirement of the wind power grid-connected power, a variational modal decomposition method is adopted to divide the power to be stabilized into a low frequency part and a high frequency part 2, a storage battery and a super capacitor are respectively used for stabilizing, and an energy storage charging and discharging control model is established; considering the construction cost of the energy storage device and punitive cost brought by wind abandoning and light abandoning and insufficient stabilization, establishing a hybrid energy storage optimization configuration model; taking typical daily data as a basis, solving the hybrid energy storage optimal configuration model by adopting a genetic algorithm to obtain the lowest annual cost and the corresponding energy storage rated power and rated capacity; therefore, the energy storage system is optimally configured, and the economical efficiency of the energy storage equipment can be fully utilized.

Drawings

FIG. 1 is a control flow diagram of the method of the present invention.

Detailed Description

The present invention is not limited by the following examples, and specific embodiments may be determined according to the technical solutions and practical situations of the present invention.

The invention is further described below with reference to the following examples:

example (b): as shown in the attached figure 1, the GA-based energy storage system optimal configuration method comprises the steps of dividing wind power to be stabilized into a low-frequency part and a high-frequency part 2 by adopting a variational modal decomposition method, stabilizing by using a storage battery and a super capacitor respectively, and establishing an energy storage charging and discharging control model; considering the construction cost of the energy storage device and punitive cost brought by wind abandoning and light abandoning and insufficient stabilization, establishing a hybrid energy storage optimization configuration model; taking typical daily data as a basis, solving the hybrid energy storage optimal configuration model by adopting a genetic algorithm to obtain the lowest annual cost and the corresponding energy storage rated power and rated capacity; wherein the content of the first and second substances,

the method for establishing the energy storage charging and discharging control model comprises the following steps:

for a wind-solar energy storage power plant, the lowest discharge power of an energy storage system cannot be lower than the maximum required power, so that the charging and discharging mathematical model of the corresponding wind-solar energy storage system is as follows:

wherein P is_nRepresentative of the power demand, P_sRepresenting the power generated by solar energy, P_wRepresenting the power generated by wind energy;

when the sum of the power generated by wind energy and solar energy is greater than the required power, the redundant power is dumped to the energy storage system;

P_in＝P_s+P_w-P_n (2)

P_inthe method represents reasonable distribution of input power of an energy storage system to a load interval, and is a primary target for controlling and achieving the wind-solar-energy storage integrated charging and discharging target. The dividing method comprises the following steps:

the trough interval:

[P_min,P_min+(P_s+P_w)] (3)

peak interval:

[P_max-γ(P_s+P_w),P_max] (4)

the general interval:

[P_min+γ(P_s+P_w),P_max-γ(P_s+P_w)] (5)

wherein P is_minIs the minimum output load, P, of the energy storage system_maxFor the maximum output load of the energy storage system, γ is a redundancy coefficient generated during conversion or transmission, and may be set to 1.2.

In order to distinguish the purposes of two energy storage devices, namely a super capacitor and a storage battery, for reasonable distribution, energy storage power needs to be decomposed and redistributed, and VMD is adopted for operation;

namely, decomposing the reference power of the energy storage system into k components u_k(t) for each u_k(t) performing a Hilbert transform to obtain a single-sided spectrum,

is modulated to the corresponding base frequency band and,

and then calculates the norm of the gradient of the computational signal, and estimates each bandwidth,

wherein the content of the first and second substances,

{u_k}:＝{u₁Λu_k}{ω_k}:＝{{ω₁Λ{ω_k}

converting equation (8) to an unconstrained variant:

in the formula (10), α is a secondary penalty factor; λ is the Lagrangian multiplier;

{u_k},{ω_kthe updates of λ are as follows:

replacement of omega by omega-omega_k，

The integration form is changed into a non-negative frequency interval integration form,

the solution is further changed into the solution,

the center frequency is converted to the frequency domain,

the updating method comprises the following steps:

the lambda updating method comprises the following steps:

the iteration is terminated satisfying the following relationship:

the mode center frequencies are ensured to be dissimilar, thereby preventing the over-decomposition phenomenon. And reasonably selecting and processing each decomposed mode to obtain a low-frequency component suitable for the storage battery and a high-frequency component suitable for the super capacitor.

The method for establishing the hybrid energy storage optimization configuration model comprises the following steps:

photovoltaic array M considering purchase cost required by energy storage system_sWind turbine generator system M_sThe purchase cost of the energy storage system is mainly the storage battery M_baSuper capacitor M_scPunishment cost M brought by insufficient stabilizing fluctuation of power generation system_laAnd punishment cost M caused by light and wind abandonment_abThe objective function of the acquisition cost is:

M_min＝M_ba+M_sc+M_la+M_ab (19)

because of the unequal life of the energy storage devices, the acquisition cost is converted here to the equivalent in years as follows:

wherein r is the conversion ratio, Y_scFor the lifetime of the supercapacitor, Y_baFor the life of the accumulator, Y_p-scFor the unit power cost of the supercapacitor, f_p-baFor the cost per unit power of the accumulator, f_e-scCost per unit capacity of supercapacitor, f_e-baFor the cost per unit capacity of the accumulator, P_baFor the rated power of the accumulator, P_scRated power for super capacitor, E_baFor rated capacity of the battery, E_scThe rated capacity of the super capacitor is obtained;

punishment cost M brought by insufficient stabilizing fluctuation of power generation system_laAnd punishment cost M caused by light and wind abandonment_abThe expression is as follows:

M_la＝αE_la (22)

M_la＝αE_la (23)

wherein E_laTo balance the power caused by the deficiency, E_abIn order to discard the electric quantity wasted by the light and wind, alpha is unit loss cost caused by insufficient stabilization, and beta is unit cost wasted by the light and wind discarding;

the constraint conditions are as follows:

|P_ba|≤P_ba·max (24)

|P_sc|≤P_sc·max (25)

C_ba·min≤C_ba≤C_ba·max (26)

C_sc·min≤C_sc≤C_sc·max (27)

wherein P is_baIs the power of the accumulator, P_scIs the power of a super capacitor, C_baFor the stored charge of the battery, C_scThe storage capacity of the super capacitor;

s^ch+s^di≤1 (28)

s^chindicating a charge indicating variable, s, of the energy storage device^diAnd (2) a discharge indicating variable of the energy storage device, wherein the formula (28) indicates that the energy storage device cannot simultaneously operate in a charging and discharging state.

The method for solving the hybrid energy storage optimization configuration model by adopting the genetic algorithm comprises the following steps:

step 1: determining expected grid-connected power of wind power and photovoltaic and stabilizing required power fluctuation according to actual data, and initializing power and capacity of an energy storage system;

step 2: decomposing the desired flat power fluctuation into a series of subcomponents using VMD (i.e., variational modal decomposition);

and step 3: the decomposed series of subcomponents are assigned to high-frequency and low-frequency components;

and 4, step 4: randomly generating a population and updating the population;

and 5: cross-swapping to produce children;

step 6: carrying out gene variation;

and 7: expressing the characters;

and 8: natural selection is carried out by utilizing a gambling wheel;

and step 9: searching the optimal individual and returning to the step 4;

step 10: and exiting after the iteration times are finished.

The technical characteristics form an embodiment of the invention, which has strong adaptability and implementation effect, and unnecessary technical characteristics can be increased or decreased according to actual needs to meet the requirements of different situations.