CN116205320A

CN116205320A - Double-energy-storage system layered optimization configuration method based on multi-main-body investment

Info

Publication number: CN116205320A
Application number: CN202210725123.6A
Authority: CN
Inventors: 谢丽蓉; 卞一帆; 王智勇; 路朋; 包洪印
Original assignee: Xinjiang University
Current assignee: Xinjiang University
Priority date: 2022-06-24
Filing date: 2022-06-24
Publication date: 2023-06-02

Abstract

The hierarchical optimal configuration method of the double energy storage systems based on multi-main investment combines two energy storage A, B with equal capacity and equal power: 1) The two groups of energy storage adopt an alternate working mode to respectively bear charging and discharging work, and only one group of energy storage works in the same period; when one group of energy storage is in a charging or discharging state, the other group of energy storage is in a floating charge-to-discharge state or a floating charge-to-charge state; respectively classifying the floating charge state and the floating charge state into charge and discharge states; 2) The two groups of energy storage adopt a synchronous switching mode, and when the switching condition is reached, the two groups of energy storage are synchronously switched at the juncture of the operation time periods; 3) Repeating the process 2) until the simulation is finished. The double energy storage operation strategy of an alternate working and synchronous switching mode is adopted to reduce the service life loss caused by frequent switching of energy storage. And then, providing an energy storage layering optimization economical model to realize balance of investment body interests and maximize and maintain strong charge and discharge capacity of energy storage.

Description

Double-energy-storage system layered optimization configuration method based on multi-main-body investment

Technical Field

The invention relates to the field of double energy storage system optimization, in particular to a double energy storage system layering optimization configuration method based on multi-main investment.

Background

In recent years, energy storage is widely applied to power generation links of power systems, and the energy storage in a wind power plant mainly plays roles of peak regulation and frequency modulation, stabilizing wind power fluctuation, absorbing wind abandoning, compensating wind power prediction errors and the like. The flexible energy storage regulation capability can effectively balance the randomness, intermittence and anti-peak regulation characteristics of wind power output, thereby promoting wind power consumption. In the related research of wind energy storage systems, the problems of optimal configuration of energy storage, how to reduce energy storage cost and the like are hot spots of current research.

Aiming at the problem of optimal configuration of energy storage, xu Guodong, cheng Haozhong, fang Sidu, and the like, a battery energy storage configuration optimization model [ J ] for improving the operation benefit of a wind farm is used for power system automation, 2016,40 (5): 62-70; cheng Xin, xu Liang, zhou Shucan, etc. an energy storage configuration method based on new energy output guarantee rate track sensitivity analysis [ J ]. Electric power system automation 2020,44 (13): 25-31, sang Bingyu, wang Deshun, yang Bo, etc. an energy storage optimization configuration method for smoothing new energy output fluctuation [ J ]. Chinese motor engineering journal, 2014,34 (22): 3700-3706.

The energy storage optimizing configuration method under the traditional single investment main body is analyzed, wherein a grid structure is partially considered to provide a configuration method aiming at the maximum wind-storage combined operation benefit; model-based predictive control also proposes power generation side energy storage configuration schemes under different stabilizing strategies; or the loss of the stored energy charge and discharge is considered, and the wind energy optimal configuration result is solved by utilizing spectrum analysis and low-pass filtering. However, the single investment in the above documents will bear huge energy storage configuration cost, which is disadvantageous to the economy.

In order to improve the economic benefit of investment principals, the energy storage operation strategy is optimized in the prior art, wherein a double energy storage operation strategy considering the cycle life is adopted, and the result shows that the provided strategy can improve the energy storage economy, but only after single energy storage is configured, the energy storage with the same scale is artificially added, and the unified optimization of the two groups of energy storage is not performed; the dynamic switching of the charging and discharging functions of the double retired battery pack is realized by utilizing hierarchical control, so that the economy is improved, but the recycling safety of the retired battery is still to be further examined. Or research on project investment patterns, which indicates that multi-subject investment power projects can effectively disperse the cost to each investor, thereby improving project economy. The method for planning the micro-grid comprises the steps of providing a micro-grid planning method for joint investment of a power distribution network and a micro-grid operator based on evolution games in the prior art; the comprehensive energy multi-main-body investment benefit balanced optimization scheduling method based on the improved non-dominant sorting genetic algorithm is provided; there have also been proposed a multi-subject investment virtual power plant capacity allocation model based on cost-benefit analysis and non-dominant ranking genetic algorithm solution. However, the research on multi-main investment is mainly focused on the planning problem of micro-grids, comprehensive energy systems and virtual power plants, and the energy storage planning problem is less studied in literature.

Aiming at the problems, the application firstly adopts a double-energy-storage operation strategy in an alternating work and synchronous switching mode to reduce the service life loss caused by frequent switching of energy storage. And then, providing an energy storage layering optimization economical model to realize balance of investment body interests and maximize and maintain strong charge and discharge capacity of energy storage. Further, the layering model is solved by adopting a single-target and multi-target mucosae algorithm and a fuzzy membership function to obtain an optimal configuration result of double energy storage. Finally, the scheme of the invention is verified by using actual data, and the operation effects of a sodium-sulfur battery (sodium sulful battery, NAS), an all-vanadium redox flow battery (vanadium redox battery, VRB), a polysulfide/bromine redox flow battery (polysulfide bromine battery, PSB), a lead-acid battery (VRLA) and a lithium iron phosphate battery (lithium iron phosphatebattery, LFP) are compared and analyzed.

Disclosure of Invention

In order to solve the problems in the prior art, the invention aims to provide a double-energy-storage system layering optimization configuration method based on multi-main-body investment.

In order to achieve the above purpose, the technical scheme of the invention is as follows:

double-energy-storage system layering optimal configuration method based on multi-main-body investment

Two energy storages A, B with equal capacity and equal power are combined, and the specific steps are as follows:

1) The two groups of energy storage adopt an alternate working mode to respectively bear charging and discharging work, and only one group of energy storage works in the same period; when one group of energy storage is in a charging or discharging state, the other group of energy storage is in a floating charge-to-discharge state or a floating charge-to-charge state; respectively classifying the floating charge state and the floating charge state into charge and discharge states;

2) The two groups of energy storage adopt a synchronous switching mode, and when the switching condition is reached, the two groups of energy storage are synchronously switched at the juncture of the operation time periods;

the switching strategy is:

according to the charge state of the energy storage A, B at the end of the t period and the wind power output and load size of the t+1 period, determining the switching condition of the charge and discharge states of the double-energy storage system at the end of the t period, and assuming that the energy storage A is in the charge state and the energy storage B is in the discharge state at the end of the t period, the charge state value of the energy storage A at the end of the t period is

The state of charge value of the energy storage B at the end of the period t is +.>

The charge-discharge state switching strategy of the energy storage a and the energy storage B is as follows:

(1) When the wind power output of the t+1 period is larger than the load, the energy storage system is required to charge in the t+1 period:

scheme 1: when (when)

And->

When the charging and discharging states of the energy storage A and the energy storage B do not need to be switched at the end of the t period, the energy storage A is in a charging state and performs charging action, and the energy storage B is in a discharging state but does not act in the t+1 period;

scheme 2: when S is _oc.A ＝S _oc.B ＝S _oc.max When the energy storage system is in a charging state, the charging and discharging states of the energy storage A and the energy storage B do not need to be switched at the end of the t period, and in the t+1 period, the energy storage A is in a charging state and does not act, and the energy storage B is in a discharging state and does not act;

scheme 3: when S is _oc.A ＝S _oc.max And S is _oc.min ≤S _oc.B <S _oc.max When the charging and discharging states of the energy storage A and the energy storage B need to be switched at the end of the t period, the energy storage A is in a discharging state but does not act, and the energy storage B is in a charging state and performs charging action in the t+1 period;

(2) When the wind power output of the t+1 period is smaller than the load, the energy storage system is required to discharge in the t+1 period:

scheme 4: when S is _oc.min ≤S _oc.A ≤S _oc.max And S is _oc.min <S _oc.B ≤S _oc.max When the energy storage A and the energy storage B are in a charging state but not in a discharging state, and the energy storage B is in a discharging state and performs a discharging action in a t+1 period;

scheme 5: when S is _oc.A ＝S _oc.B ＝S _oc.min When the energy storage system is in a charging state, the charging and discharging states of the energy storage A and the energy storage B do not need to be switched at the end of the t period, and in the t+1 period, the energy storage A is in the charging state and does not act, and the energy storage B is in the discharging state and does not act;

Scheme 6: when S is _oc.min <S _oc.A ≤S _oc.max And S is _oc.B ＝S _oc.min When (1): the charge and discharge states of the energy storage A and the energy storage B need to be switched at the end of the t period, and in the t+1 period, the energy storage A is in a discharge state and performs a discharge action, and the energy storage B is in a charge state but does not act;

3) Repeating the process 2) until the simulation is finished.

Further, an energy storage layered optimization model is established by the energy storage charging and discharging capacity index and the multi-investment main body income, and an upper optimization model is responsible for distributing investment operation cost and income of the main bodies of the two sides to the energy storage, so that the benefit maximization of the main bodies of the two sides is achieved; the lower optimization model is responsible for optimizing the energy storage charging and discharging capacity, so that the double energy storage can keep stronger charging and discharging capacity;

wherein the upper layer objective function is

When the energy storage power station is planned, the main targets of both parties are that the total annual income is the largest after the energy storage is configured, namely:

wherein: f (F) _cur The income of the wind disposal on the internet is increased; f (F) _sub,w Repairing the abandoned wind on the internet; alpha is the investment duty ratio and dividing coefficient of wind farm operators, and the value is between 0 and 1; f (F) _evn Is environmental benefit, element; f (F) _sub,s The method comprises the steps of supplementing income for energy storage operation and repairing the energy storage operation; c (C) _run,year The maintenance cost is for the double energy storage annual operation; c (C) _inv,year To account for the double energy storage annual investment costs of life loss; f (F) _tgc Annual income for power grid green evidence trade; f (F) _sel The electricity selling income is newly increased for the power grid year;

1) Annual investment costs accounting for life loss

The energy storage investment construction cost comprises hardware cost and software cost, wherein the hardware cost refers to the cost of energy storage with a certain capacity, and the software cost refers to the cost of equipment such as a power conversion system (power conversion system, PCS), a battery management system (battery management system, BMS) and the like; the cost function is as follows:

C _inv ＝C _E (E _b,A +E _b,B )+C _P (P _b,A +P _b,B ) (3)

wherein: c (C) _inv Initial investment cost for energy storage is primary; c (C) _E The cost is per energy storage unit capacity, yuan/kWh; e (E) _b,A 、E _b,B Rated capacity of the stored energy A, B and kWh respectively; c (C) _P The power cost per unit of energy storage PCS is Yuan/kW; p (P) _b,A 、P _b,B Rated power of the stored energy A, B and kW respectively; r is the discount rate, 6%; τ _bat The service life of the double-energy storage device is prolonged;

the cycle life of the double energy storage is influenced by factors such as working temperature, charge-discharge switching times, discharge depth and the like, and the running of the energy storage can cause slow degradation of the performance and generate cycle life loss; to accurately calculate the service life tau of double energy storage _bat The invention mainly considers the influence of the charge and discharge times and the discharge depth of the stored energy A, B on the cycle life;

τ _bat ＝min(T _s,A ,T _s,B ,τ _bat,b ) (5)

wherein: t (T) _s,A 、T _s,B The service life of the energy storage A and the service life of the energy storage B are respectively the service life and the service life of the energy storage B; τ _bat.b Representing the warranty period and year of the energy storage battery;

equivalent cycle life corresponding to the discharge depth of the energy storage A, B of 1 is determined by the self-characteristics of the energy storage batterySetting; d (D) _od,A,u 、D _od,B,u The depth of discharge of the stored energy A, B in the u-th cycle; n (N) _ctf,A (D _od,A,u )、N _ctf,B (D _od,B,u ) The discharge depth of the stored energy A, B is D _od,A,u 、D _od,B,u Equivalent cycle life corresponding to the time; h _A 、H _B The total charge-discharge switching times of the stored energy A, B in one year are respectively; t is the number of scheduling periods in one year; v (V) _A,ch (t)、V _B,ch (t)、V _A,dis (t)、V _B,dis (t) binary variable representing the switching of the charge and discharge states of the t-period end energy storage A, B, V _A,ch (t)、V _B,ch (t) when taking "1" it means that the energy storage A, B is switched from the charged state to the discharged state at the end of the period t, V _A,dis (t)、V _B,dis (t) taking a "1" to indicate that the stored energy A, B is switched from a discharged state to a charged state at the end of the t period;

2) Annual operation maintenance cost

The operation and maintenance cost of the energy storage system is mainly related to the size of the energy storage battery, and comprises a fixed part determined by the power conversion subsystem and a variable part determined by the charge and discharge electric quantity of the energy storage, and the cost function is as follows:

wherein: c (C) _run The cost is maintained for the operation of the whole life cycle of the energy storage system; c (C) _run,P The operation and maintenance cost is the energy storage unit power, and the energy storage unit power is yuan/kW; p (P) _b,A 、P _b,B Rated power of the stored energy A, B and kW respectively; c (C) _run,E The operation and maintenance cost is per unit capacity, and the unit/kWh is calculated; w (W) _1,i The energy is stored as the total charge and discharge electric quantity in the ith year, kWh; i is the number of years since the construction of the double energy storage system;

3) Annual wind-abandoning internet surfing benefits

After the energy storage system is built, the abandoned wind energy can be stored and is integrated into a power grid at the time of load peak, and the abandoned wind surfing income is obtained, and the income function is as follows:

wherein: w (W) ₂ (t) is the amount of abandoned wind surfing in t time period, kWh; c (C) ₁ Guiding unit price for wind power surfing, unit cell/kWh;

4) Annual wind power subsidy benefit

According to the notification about perfecting the policy of wind power online price, if the wind farm meets the policy requirement, the newly increased online electric quantity will get subsidy income ^[21] The benefit function is as follows:

wherein: c (C) ₂ The unit price, unit cell/kWh are subsidized for wind power on-line;

5) Annual environmental benefit

The environmental benefit of energy storage mainly comprises two parts, wherein one part is to network part of abandoned wind power so as to reduce grid-connected power of the traditional thermal power generating unit and realize greenhouse gas and pollutants (mainly comprising CO) ₂ 、SO ₂ 、NO _x Carbon dust, suspended particulate matters, etc.) and reducing the emission; another part is recovery benefit of extracting metallic material from the cell after the end of the energy storage life, the benefit function is as follows:

F _env ＝F _emi +F _rec (14)

wherein: f (F) _emi To reduce the pollution annual income of the traditional unit; f (F) _rec Recovering annual values such as income for the energy storage battery; n is the total number of discharged pollutants; lambda (lambda) _j Cost per unit of environmental load for the jth pollutant, yuan/kg; q (Q) _j The j pollutant emission amount is kg/(kWh) for the power generation of the traditional thermal power generating unit; k is the total number of metal categories contained in the battery; r is R _met,k Is the unit price of metal k, yuan/ton; beta _met.k The content of the metal k in the energy storage battery per unit weight is ton; c (C) ₃ To treat the waste battery of unit weight and need productive expenditure, yuan/ton; zeta type toy _enery The energy ratio of the energy storage battery is in kg/kWh;

6) Annual energy storage benefit

For the electricity quantity of the power grid in the province sold by the self-storing facility in the 'new energy and energy storage' project, the Qinghai province gives 0.10 yuan of operation subsidy per kilowatt hour, the policies of different areas are different,

if there is a relevant subsidy policy, the stored energy will receive subsidy revenue, and the benefit function is as follows:

wherein: c (C) ₄ The unit price is complemented for the stored and sold electric energy, and the unit price is per kWh;

7) Trade income for annual green license of power grid

The transactable green certificate system (tradable green certificates, TGC) is a more common quota system, and the number of certificates of the power grid company represents the situation that the requirements of the power grid company on the quota system are completed; if the grid company cannot meet the quota system requirement, the grid company can be punished by related departments, so that the grid company can purchase renewable energy from a power supply side to obtain a certificate or purchase redundant certificates of other grid companies from a market side to meet the quota requirement;

The abandoned wind on-line electric quantity is renewable energy electric quantity newly added into a power grid, and can be converted into income of a power grid operator by using a green certificate transaction system, and the income function is as follows:

in the middle of：C ₅ Unit cell/kWh for green certificate trade unit price;

8) Annual electricity selling income of power grid

The power grid operator transmits and sells the electric energy consumed by the energy storage to the user through the power transmission grid, and corresponding electricity selling income can be obtained, and the income function is as follows:

wherein: c (C) ₆ Average electricity selling price for the power grid company, yuan/kWh; c (C) ₇ The unit electricity quantity is the network cost, the unit element/kWh;

from the above, it can be seen that the annual income f of wind farm operators ₁ And annual revenue f for grid operators ₂ Respectively, can be expressed as:

f ₁ ＝F _cur +F _sub,w +α(F _evn +F _sub,s ) (20)

f ₂ ＝F _tgc +F _sel +(1-α)(F _evn +F _sub,s ) (21)

the underlying objective function is:

when participating in the wind-abandoning and absorbing, the charge states of the two stored energy should be maintained in an ideal interval as much as possible so as to ensure enough charge-discharge capacity storage; by using the battery charge-discharge capacity index f ₃ The degree of the deviation of the SOC of the battery from the ideal interval is measured, and the larger the value is, the larger the degree of the deviation of the SOC of the energy storage battery from the ideal interval is; when the SOC of the battery was 0.5, it was demonstrated that the battery had good charge-discharge capacity reserve [25]Set ideal interval of battery SOC as [0.4,0.6 ] ]Selecting a stricter interval of [0.45,0.55 ]]The method comprises the steps of carrying out a first treatment on the surface of the The lower layer optimization goal is that the sum of the battery charge and discharge capability index values of the stored energy A, B is minimum in the annual schedule period:

minf ₃ ＝f _3,A +f _3,B (22)

wherein: f (f) _3,A 、f _3,B Respectively storing energy storage A, B as battery charging and discharging capability index values in a annual scheduling period; s is S _oc,A,avg The average value of the SOC of the energy storage A in the annual scheduling period is obtained; s is S _oc,A (t) is the SOC of the energy storage A at the end of the period t; the charge and discharge capacity index value calculation method of the energy storage B is the same as that of the energy storage A, and is not repeated here;

further, the constraint conditions in the invention are as follows:

the energy storage constraint comprises energy storage charge state constraint, charge and discharge power constraint and power balance constraint, wherein the constraint of the energy storage A is consistent with the constraint of the energy storage B, and only the constraint condition of the energy storage A is described herein;

1) Energy storage A charge-discharge power constraint

Assuming that the charge and discharge power of the energy storage battery is constant in the t period, the charge and discharge power of the energy storage battery is related to rated power and abandoned wind power and the residual charge and discharge capacity of the energy storage battery;

wherein: p (P) _cha,A (t) is the charging power of the energy storage A in the t period, kW; p (P) _dis,A (t) is the discharge power of the energy storage A in the t period, kW; p (P) _net (t) is grid-connected power in a period t of the wind storage system, and kW; p (P) _win (t) is the output power of the wind farm in the period t, kW; s is S _oc,A,max 、S _oc,A,min The upper limit value and the lower limit value of the SOC of the energy storage A are respectively; s is S _oc,A (t-1) is the SOC of the energy storage A at the end of the t-1 period; η is the charge and discharge efficiency of the energy storage battery;

2) Energy storage a state of charge constraints

S _oc,A,min ≤S _oc,A (t+1)≤S _oc,A,max (28)

0≤L _cha,A (t)+L _dis,A (t)≤1 (29)

S _oc,A (0)＝S _oc,A (T) (30)

Wherein: s is S _oc,A (t+1)、S _oc,A (t) is the state of charge of the stored energy at times t and t+1 respectively; τ is the self-discharge rate of the stored energy; l (L) _cha,A (t) is the charging state of energy storage in the period of t, and the value is 0 or 1, wherein 0 represents floating charge and waiting for discharging, and 1 represents charging; l (L) _dis,A (t) is a discharge state of energy storage at the moment t, and the value is 0 or 1, wherein 0 represents floating charge waiting for charging, and 1 represents discharging; s is S _oc,A (0) Initial SOC for energy storage a; s is S _oc,A (T) is the SOC of the energy storage A at the end of the scheduling period;

3) System power balance constraint

The grid-connected power is the sum of wind power output power and discharge power of the energy storage power station, and the system power balance constraint is as follows:

wherein: p (P) _w (t) is the wind power grid-connected power at the moment t, and kW; l of energy storage B _cha,B (t)、L _dis,B (t)、P _cha,B (t)、P _dis,B The parameters (t) and the like have the same meaning as the energy storage A parameter.

Further, a classical investment evaluation index investment recovery period and an investment yield are selected to scientifically evaluate the economic benefit of the double energy storage;

1) Investment recovery period

The investment recovery period is an important index for measuring the project investment risk degree from the time angle, the invention selects the investment recovery period as an evaluation index, and the calculation formula is as follows:

Wherein: m is M ₁ The investment recovery period for the investment energy storage of the wind power plant is annual; m is M ₂ The method is an investment recovery period for energy storage of the investment of the power grid, and the year; the smaller the investment recovery period, the more secure the investment in the energy storage;

2) Investment yield

The investment yield is an economic index for measuring the profitability level of an investment project, and can be expressed by the ratio of annual average total yield to total investment operation cost in the whole life cycle of the system, and the calculation formula is as follows:

wherein: m is M ₃ Is the investment yield for the wind power plant; m is M ₄ Is the investment yield for the power grid; the larger the investment yield, the better the profitability level of the investment project.

Further, the objective function of the upper layer optimization model and the objective function of the lower layer optimization model have different dimensions, and the satisfaction degree of each objective in each group of configuration results is determined according to the fuzzy set theory, and can be represented by a fuzzy membership function:

wherein: f (f) _d Is the d-th objective function value (d=1, 2, 3); f (f) _d,min 、f _d,max Minimum and maximum values of the d-th objective function; h is a _d When 0 or 1 is the total dissatisfaction or the total satisfaction of the d-th objective function, the normalized satisfaction of all configuration results is defined as:

wherein: h is the standardized satisfaction degree of each group of configuration results, and finally the configuration result with the largest standard satisfaction degree is selected as the optimal configuration result.

Compared with the prior art, the invention has the beneficial effects that:

the double-energy-storage system layered optimization configuration method based on multi-main investment 1) is from the aspect of mechanism modeling: the double energy storage mode is explored from the theoretical mechanism, two battery packs with equal capacity and power are combined to form the double energy storage system, and the working mode of alternately working and synchronously switching is used, so that the energy storage system acts in real time according to the instruction of the energy management system to finish the abandoned wind absorption task. The single energy storage system does not consider the charge-discharge switching strategy, and directly performs charge-discharge work according to the instruction of the energy management system;

2) From the aspect of working efficiency: when the battery needs to be charged in the t period and discharged in the t+1 period, the energy storage needs to respond rapidly, the charge and discharge states of the single energy storage need to be switched relative to the double energy storage batteries, the response time is relatively long, and a certain group of batteries in the double energy storage system are in a floating charge and discharge state, so that the power grid requirements can be met through rapid discharge. Meanwhile, the digestion task which needs to be completed by the single energy storage can be completed by the double energy storage systems, and the two energy storage can improve the working efficiency of the systems;

3) From the aspect of potential risk: when a single energy storage system fails, the effect of the system for absorbing the abandoned wind is affected, the risk is high, the potential risk can be reduced by using the double energy storage system, and when one battery pack fails, the other battery pack fully bears charge and discharge work, so that the running safety of the system is improved;

4) From the aspect of system lifetime: when the total capacity and the total power of the single energy storage system and the double energy storage system are consistent, simulation analysis shows that the double energy storage system has longer service life;

5) From the aspect of economy: on the basis of a cost model considering energy storage life loss, the double energy storage system has longer service life and can absorb more waste wind electric quantity compared with single energy storage, and can bring more benefits to operators compared with the single energy storage system, so that the double energy storage system has better economical efficiency.

Drawings

FIG. 1 is a hierarchical optimization scheme block diagram of an economic optimization configuration;

FIG. 2 illustrates economic index values of different batteries under different conditions;

FIG. 3 is a graph of the impact of reduction in energy storage costs on a system, wherein: FIG. 3 (a) is a graph of optimal configuration results; FIG. 3 (b) economic impact graph;

FIG. 4 effect of increase in ideal interval of battery SOC on system; wherein FIG. 4A best-fit results; FIG. 4B economic impact

FIG. 5 effects different investment ratios on the system, wherein FIG. 5A is the optimal configuration results; FIG. 5B economic impact;

fig. 6 is a diagram of a dual energy storage system charge-discharge system: FIG. 6A is a flow chart of a dual energy storage charge-discharge state switching strategy; FIG. 6B is a schematic diagram of a charge-discharge switching of the dual energy storage system;

FIG. 7 is a typical daily load, wind power data and time-of-use electricity price graph of a certain Hami district year;

FIG. 8 is a graph of change in charge and discharge power of a lithium iron phosphate energy storage battery;

fig. 9 is a diagram of five battery parameter radar comparisons.

Detailed Description

The technical scheme of the invention is further described in detail below with reference to the attached drawings and the detailed description:

as shown in fig. 1, the present invention combines two equal capacity, equal power energy stores A, B to provide a dual energy store operation strategy. The specific strategy is as follows:

1) The two groups of energy storage adopt an alternate working mode, charge and discharge work is respectively carried out, and only one group of energy storage work is carried out in the same period. When one group of energy storage is in a charging or discharging state, the other group of energy storage is in a floating charge-to-discharge state or a floating charge-to-charge state. In order to simplify analysis, the floating charge state and the floating charge state are respectively classified into a charge state and a discharge state.

2) The two groups of energy storage adopt a synchronous switching mode, and when the switching condition is reached, the two groups of energy storage are synchronously switched at the juncture of the operation time periods.

3) Repeating the process 2) until the simulation is finished.

Layered optimization configuration model of 2 multi-main-body investment wind storage system

How to reasonably configure energy storage capacity and power, improve the capacity of the system to absorb the abandoned wind and the earnings of operators is an important point of planning energy storage, and compared with a single-main investment mode, the multi-main investment mode has more investment operators and can effectively apportion project investment cost. The chapter establishes an energy storage optimal configuration model under a multi-main-body investment mode.

2.1 optimization model Integrated framework

An energy storage layered optimization model is established by comprehensively considering energy storage charging and discharging capacity indexes and multi-investment main body profits, and an upper optimization model is responsible for distributing investment operation and maintenance costs and profits of the main bodies of the two sides for energy storage, so that the benefit maximization of the main bodies of the two sides is achieved; the lower optimization model is responsible for optimizing the energy storage charge and discharge capacity, so that the double energy storage can keep stronger charge and discharge capacity, and the layered optimization scheme structure is shown in figure 1.

Upper layer objective function

wherein: f (F) _cur The income of the wind disposal on the internet is increased; f (F) _sub,w Repairing the abandoned wind on the internet; alpha is the investment duty ratio and dividing coefficient of wind farm operators, and the value is between 0 and 1; f (F) _evn Is environmental benefit, element; f (F) _sub,s The method comprises the steps of supplementing income for energy storage operation and repairing the energy storage operation; c (C) _run,year The maintenance cost is for the double energy storage annual operation; c (C) _inv,year To account for life lossInvestment cost of energy storage year, yuan; f (F) _tgc Annual income for power grid green evidence trade is acquired; f (F) _sel And the method increases electricity selling income for the power grid year.

1) Annual investment costs accounting for life loss

The energy storage investment construction cost comprises hardware cost and software cost, wherein the hardware cost refers to the cost of energy storage with a certain capacity, and the software cost refers to the cost of equipment such as a power conversion system (power conversion system, PCS), a battery management system (battery management system, BMS) and the like. The cost function is as follows:

C _inv ＝C _E (E _b,A +E _b,B )+C _P (P _b,A +P _b,B ) (3)

Wherein: c (C) _inv Initial investment cost for energy storage is primary; c (C) _E The cost is per energy storage unit capacity, yuan/kWh; e (E) _b,A 、E _b,B Rated capacity of the stored energy A, B and kWh respectively; c (C) _P The power cost per unit of energy storage PCS is Yuan/kW; p (P) _b,A 、P _b,B Rated power of the stored energy A, B and kW respectively; r is the discount rate, 6%; τ _bat The service life of the double-energy storage is as long as the service life of the double-energy storage.

The cycle life of the double energy storage is influenced by factors such as working temperature, charge-discharge switching times, discharge depth and the like, and the operation of the energy storage can cause slow degradation of the performance and generate cycle life loss. To accurately calculate the service life tau of double energy storage _bat The invention mainly considers the influence of the charge and discharge times and the discharge depth of the stored energy A, B on the cycle life.

τ _bat ＝min(T _s,A ,T _s,B ,τ _bat,b ) (5)

the equivalent cycle life corresponding to the discharge depth of the energy storage A, B is 1 respectively, and is determined by the characteristics of the energy storage battery; d (D) _od,A,u 、D _od,B,u The depth of discharge of the stored energy A, B in the u-th cycle; n (N) _ctf,A (D _od,A,u )、N _ctf,B (D _od,B,u ) The discharge depth of the stored energy A, B is D _od,A,u 、D _od,B,u Equivalent cycle life corresponding to the time; h _A 、H _B The total charge-discharge switching times of the stored energy A, B in one year are respectively; t is the number of scheduling periods in one year; v (V) _A,ch (t)、V _B,ch (t)、V _A,dis (t)、V _B,dis (t) binary variable representing the switching of the charge and discharge states of the t-period end energy storage A, B, V _A,ch (t)、V _B,ch (t) when taking "1" it means that the energy storage A, B is switched from the charged state to the discharged state at the end of the period t, V _A,dis (t)、V _B,dis (t) taking a "1" indicates that the stored energy A, B was switched from a discharged state to a charged state at the end of the t period. Wherein the depth of discharge and equivalent cycle life of the energy storage cell are according to document [19]The method in (2) is calculated and will not be described in detail herein.

2) Annual operation maintenance cost

wherein: c (C) _run The cost is maintained for the operation of the whole life cycle of the energy storage system; c (C) _run,P The operation and maintenance cost is the energy storage unit power, and the energy storage unit power is yuan/kW; p (P) _b,A 、P _b,B Rated power of the stored energy A, B and kW respectively; c (C) _run,E The operation and maintenance cost is per unit capacity, and the unit/kWh is calculated; w (W) _1,i The energy is stored as the total charge and discharge electric quantity in the ith year, kWh; i is the number of years since the construction of the dual energy storage system.

3) Annual wind-abandoning internet surfing benefits

Wherein: w (W) ₂ (t) is the amount of abandoned wind surfing in t time period, kWh; c (C) ₁ And guiding unit price for wind power surfing, and guiding yuan/kWh.

4) Annual wind power subsidy benefit

wherein: c (C) ₂ And supplementing unit price, yuan/kWh for wind power on-line.

5) Annual environmental benefit

The energy storage environmental benefit mainly comprises two parts, one partThe method is characterized in that partial abandoned wind electricity quantity is connected with the network so as to reduce the grid-connected electricity quantity of the traditional thermal power generating unit, and realize greenhouse gas and pollutants (mainly comprising CO) ₂ 、SO ₂ 、NO _x Carbon dust, suspended particulate matters, etc.) and reducing the emission; another part is recovery benefit of extracting metallic material from the cell after the end of the energy storage life, the benefit function is as follows:

F _env ＝F _emi +F _rec (14)

wherein: f (F) _emi To reduce the pollution annual income of the traditional unit; f (F) _rec Recovering annual values such as income for the energy storage battery; n is the total number of discharged pollutants; lambda (lambda) _j Cost per unit of environmental load for the jth pollutant, yuan/kg; q (Q) _j The j pollutant emission amount is kg/(kWh) for the power generation of the traditional thermal power generating unit; k is the total number of metal categories contained in the battery; r is R _met,k Is the unit price of metal k, yuan/ton; beta _met.k The content of the metal k in the energy storage battery per unit weight is ton; c (C) ₃ To treat the waste battery of unit weight and need productive expenditure, yuan/ton; zeta type toy _enery The energy storage battery is of energy-to-weight ratio, kg/kWh.

6) Annual energy storage benefit

wherein: c (C) ₄ Selling electric energy for energy storagePatch unit price, yuan/kWh.

7) Trade income for annual green license of power grid

A tradable green certification system (tradable green certificates, TGC) is a more common quota system, and the number of certificates of the grid company represents the situation where the requirements for the quota system are fulfilled. If the grid company cannot meet the quota system requirement, the grid company can be punished by related departments, so that the grid company can purchase renewable energy from a power source side to obtain a certificate or purchase redundant certificates of other grid companies from a market side to meet the quota requirement.

Wherein: c (C) ₅ Unit price for green certificate transaction, meta/kWh.

8) Annual electricity selling income of power grid

wherein: c (C) ₆ Average electricity selling price for the power grid company, yuan/kWh; c (C) ₇ The unit electricity quantity is the network charge and the unit/kWh.

f ₁ ＝F _cur +F _sub,w +α(F _evn +F _sub,s ) (20)

f ₂ ＝F _tgc +F _sel +(1-α)(F _evn +F _sub,s ) (21)

2.2 lower layer objective function

When participating in the wind-abandoning and absorbing, the state of charge (SOC) of the two stored energy should be maintained as much as possible within an ideal interval to ensure sufficient reserve of charge and discharge capacity. By using the battery charge-discharge capacity index f ₃ The degree of the deviation of the battery SOC from the ideal interval is measured, and the larger the value is, the larger the degree of the deviation of the battery SOC from the ideal interval is indicated. When the SOC of the battery was 0.5, it was demonstrated that the battery had good charge-discharge capacity reserve [25]Set ideal interval of battery SOC as [0.4,0.6 ]]The invention selects a stricter interval of [0.45,0.55 ]]. The lower layer optimization goal is that the sum of the battery charge and discharge capability index values of the stored energy A, B is minimum in the annual schedule period:

minf ₃ ＝f _3,A +f _3,B (22)

Wherein: f (f) _3,A 、f _3,B Respectively storing energy storage A, B as battery charging and discharging capability index values in a annual scheduling period; s is S _oc,A,avg The average value of the SOC of the energy storage A in the annual scheduling period is obtained; s is S _oc,A (t) is the SOC of the energy storage A at the end of the period t; the charge-discharge capacity index value calculation method of the energy storage B is the same as that of the energy storage a, and will not be described here again.

2.3 constraint

The energy storage constraint comprises energy storage charge state constraint, charge and discharge power constraint and power balance constraint, wherein the constraint of the energy storage A is consistent with the constraint of the energy storage B, and only the constraint condition of the energy storage A is described herein.

1) Energy storage A charge-discharge power constraint

Assuming that the charge and discharge power of the energy storage battery is constant in the t period, the charge and discharge power of the energy storage battery is related to not only rated power and waste wind power, but also the residual charge and discharge capacity of the energy storage battery.

Wherein: p (P) _cha,A (t) is the charging power of the energy storage A in the t period, kW; p (P) _dis,A (t) is the discharge power of the energy storage A in the t period, kW; p (P) _net (t) is grid-connected power in a period t of the wind storage system, and kW; p (P) _win (t) is the output power of the wind farm in the period t, kW; s is S _oc,A,max 、S _oc,A,min The upper limit value and the lower limit value of the SOC of the energy storage A are respectively; s is S _oc,A (t-1) is the SOC of the energy storage A at the end of the t-1 period; η is the charge and discharge efficiency of the energy storage battery.

2) Energy storage a state of charge constraints

S _oc,A,min ≤S _oc,A (t+1)≤S _oc,A,max (28)

0≤L _cha,A (t)+L _dis,A (t)≤1 (29)

S _oc,A (0)＝S _oc,A (T) (30)

Wherein: s is S _oc,A (t+1)、S _oc,A (t) is the state of charge of the stored energy at times t and t+1 respectively; τ is the self-discharge rate of the stored energy; l (L) _cha,A (t) is the charging state of energy storage in the period of t, and the value is 0 or 1, wherein 0 represents floating charge and waiting for discharging, and 1 represents charging; l (L) _dis,A (t) is a discharge state of energy storage at the moment t, and the value is 0 or 1, wherein 0 represents floating charge waiting for charging, and 1 represents discharging; s is S _oc,A (0) Initial SOC for energy storage a; s is S _oc,A And (T) is the SOC of the energy storage A at the end of the scheduling period.

3) System power balance constraint

3 multiple-main-body investment economy evaluation index of double energy storage system

And selecting a classical investment evaluation index investment recovery period and an investment yield rate to scientifically evaluate the economic benefit of the double energy storage.

1) Investment recovery period

wherein: m is M ₁ The investment recovery period for the investment energy storage of the wind power plant is annual; m is M ₂ The method is an investment recovery period for energy storage of the investment of the power grid, and the year; the smaller the investment recovery period, the safer the investment in the energy storage.

2) Investment yield

Model 4 solution

The intelligent algorithm has lower requirement on a mathematical model and convenient application, and has more mature application on the power system optimization problem. The configuration model is nonlinear and multi-constraint mixed integer optimization, and can be solved by adopting an improved multi-objective myxobacteria algorithm, so that a plurality of groups of configuration results are obtained, and then an optimal compromise solution is obtained by utilizing a fuzzy membership theory.

4.1 Multi-target mucoid algorithm

The mucosae algorithm (slime mould algorithm, SMA) simulates positive and negative feedback during foraging by using weights and adjusts the mucosae search path according to the quality of the food. M. Premkuma et al, based on this, developed a multi-objective myxobacteria algorithm (multi-objective slime mould algorithm, MOSMA) and demonstrated the effectiveness of the algorithm.

4.2 Multi-objective myxobacteria algorithm based on generalized reverse learning

The invention applies a generalized inverse strategy to the initialization stage of the multi-objective mucosae algorithm. Assume that

Is a random individual in the initial population, and x _i ∈[lb _i ,ub _i ](i=1, 2,.,. V-1, v), the weight factor a is a random number between 0 and 1. The subject is a reversed subject->

Can be obtained by solving in the formula (36).

4.3 determining optimal configuration results

The objective function of the upper layer optimization model and the objective function of the lower layer optimization model have different dimensions, the satisfaction degree of each objective in each group of configuration results is determined according to the fuzzy set theory, and the satisfaction degree can be represented by a fuzzy membership degree function:

5 example analysis

5.1 example parameters

To test the solving performance of the improved algorithm in the multi-objective solving problem of the electric power system, an improved multi-objective myxobacteria algorithm (OLMSMA) is compared with four classical multi-objective intelligent algorithms. And selecting a classical double-target ZDT1-6 series function as a test function.

And selecting certain 200MW wind farm data, load data and time-of-use electricity price data in Hami areas of Xinjiang to establish an example system, wherein example parameter settings are shown in a table 4. Wind power data and load data of typical days in four seasons of spring, summer, autumn and winter are obtained by using a K-means clustering algorithm, and weighted average values of the wind power data and the load data of a certain typical day in one year are obtained, and the wind power data and the load data are shown in fig. 7. Five different storage batteries are configured by applying the proposal provided by the invention, and the battery parameters are shown in tables 5 and C3.

5.2 optimizing configuration results

Dividing four scenes to perform hierarchical optimization configuration on the capacity of the energy storage battery, wherein the four scenes are respectively:

scenario 1: single-body investment bill energy storage;

scenario 2: single-body investment double energy storage;

scenario 3: multi-body investment bill energy storage;

scenario 4: multiple bodies invest in double energy storage.

The five battery optimization configuration results are shown in table 7, and the situation 1 and the situation 3, the situation 2 and the situation 4 are respectively compared, so that the energy storage capacity and the power under the multi-main investment situation are larger. Assuming that the stored energy is charged positive and the stored energy is discharged negative, taking the configuration result of the lithium iron phosphate battery as an example, the change situation of the charge and discharge power of the stored energy A and B in the

multi-main investment scenes

2 and 4 along with time is drawn, and the situation is shown in fig. 8. It can be seen from fig. 8 that the stored energy A, B is in the "alternate on, synchronous switch" mode.

5.3 comparison analysis of optimal energy storage configuration Effect

Specific operating effect pairs for the optimal energy storage configurations for five cells under four different investment scenarios are shown in table 1.

Table 1 comparison of optimal energy storage configuration effects

As can be seen from comparing scenario 1 with scenario 2, scenario 3 with scenario 4, the dual energy storage can bring higher benefit to investors, and can absorb more waste wind, so that the increased benefit brought by the dual energy storage switching strategy can compensate for the high-capacity and high-power increased cost. As can be seen by comparing scenario 1 with scenario 3, scenario 2 with scenario 4, the multi-main investment is compared with the single main investment, and the multi-main investment brings benefits to the power grid operators while bringing greater benefits to the wind farm operators and improving the wind disposal capacity of the system, so as to realize the mutual benefits and win-win of the multi-investment main investment.

The battery parameter radar contrast diagram is shown in fig. 9, five battery configuration effects in the scenario 4 are analyzed, the LFP is as high as 753.73 ten thousand yuan and 427.45 ten thousand yuan for the benefit brought by both operators, the double energy storage charge and discharge index is as low as 0.005, and the consumed wind power amount is as high as 14019MWh, so that the battery parameter radar contrast diagram is an ideal energy storage type. The current popularization is smooth, namely the VRLA has the main reason that the technical maturity is high, but the VRLA has the defects of short cycle life and low recovery value, so that the VRLA has lower income for operators. In contrast, LFP has high energy conversion efficiency and long cycle life, but its high investment cost hinders its popularization and application. The economic advantages of NAS, PSB and VRB are moderate compared to other energy storage batteries, and are expected to be applied in a large number in power systems.

5.4 economic analysis

The investment recovery period and the investment return rate of the operators are shown in fig. 2.

The shorter the investment recovery period, the less the investment risk, and the more advantageous to the project investor. By analyzing fig. 2, the investment risk of the PSB is small, the investment risk of the LFP and the VRB is large, but the investment recovery period of different batteries under different situations is within the design service life of 15 years or 10 years, so that the PSB has economy.

The larger the return on investment, the better the profitability level of the investment project. In the figure, the investment yield of the VRLA is larger, the investment yields of the VRB and the LFP are smaller, the maximum investment yield of the VRLA is one of the reasons which are still widely used at present, and the life characteristics of the LFP and the VRB are more advantageous than those of the VRLA, but the investment yield is smaller due to the high cost.

In the multi-main investment mode, different investment ratios and different ideal SOC intervals can affect the energy storage optimal configuration result, and for this purpose, the multi-main investment LFP will be taken as an example to further analyze the influence of the energy storage battery cost, the ideal battery SOC interval and the operator investment ratio on the dual energy storage system.

5.4.1 analysis of the impact of Battery cost variation

As the energy storage technology is developed, the energy storage unit cost is also reduced, and the optimal configuration results under different energy storage costs are shown in fig. 3 (a). In the graph, the energy storage cost is reduced according to 10%, the capacity initially has a remarkable increasing trend, and then the rising trend is flattened. This is because when the energy storage capacity is large enough, choosing the proper power can consume more waste wind, and continuing to increase the capacity only increases the cost. When the cost is reduced by 70%, the capacity ratio of the stored energy is now increased by about 9.95MWh, and the power ratio is now increased by about 2.96MW.

Analysis of the economic impact of fig. 3 (b) shows that the increase in energy storage capacity power increases the amount of waste wind consumed, and the increased gain of the energy storage capacity power can compensate for the increased cost of high capacity and high power. However, as costs continue to decrease, investment recovery periods all tend to decrease, with a substantial increase in investment return, indicating that lower energy storage costs reduce the investment risk of the project investor, increasing the profitability level of the project investor.

5.4.2 impact analysis of ideal Battery SOC intervals

Based on [0.45,0.55], the left and right boundaries of the ideal interval of the battery SOC decrease and increase at 0.025 intervals, respectively, and the optimal configuration results under different ideal intervals are shown in FIG. 4A. Analysis shows that when the interval becomes larger, the capacity and the power size firstly show a trend of decreasing, because the small-capacity battery can obtain better charge and discharge capacity indexes at the end of the running period, and the larger ideal interval of the SOC weakens the influence of the charge and discharge capacity indexes on the configuration of the large-capacity battery. However, as the interval continues to increase, there is a growing trend in capacity and power, as the potential cost benefits of smaller capacity power have not been able to compensate for the loss of capacity to absorb the amount of the wind curtailment.

As can be seen from the economic impact of fig. 4B, the interval size has a small influence on annual income of the system and a large influence on the amount of waste wind. Both the investment recovery period and the investment yield of the operators of both sides show a trend of decreasing before increasing, but the change is not obvious, which indicates that the size of the ideal interval of the battery SOC has less influence on the investment risk and the profit level of the project.

5.4.3 analysis of influence of operator investment ratio variations

There is economical game between wind power plant and electric network operators, and different investment ratios can also influence the economical efficiency of the system. Figure 5 thus shows the result of the optimal configuration of the energy storage and the economic impact of different investment ratios on the system. When the investment ratio is between 0.6 and 0.65, the investment return rates of the two operators are equal and the investment recovery periods of the two operators are not great, so that the energy storage investment ratio of the wind farm operators takes a value between 0.6 and 0.65, and the benefits of the two operators can be balanced better.

In the invention, in order to improve the utilization rate of new energy and the economical efficiency of the system under the background of an energy strategy target, a double-energy-storage system layering optimal configuration scheme based on multi-main investment is constructed on the basis of a double-energy-storage coordinated operation strategy, and the conflict of interests of two investment main bodies is considered, so that a complex game problem is converted into a multi-target solving problem, and the following conclusion is obtained through experimental verification:

1) The double-layer optimization model of the multi-main investment wind reservoir, which takes account of the return of the investment main body and the charge and discharge capability of the battery, is constructed, and the charge and discharge capability of the battery and the economic benefit of the system are improved.

2) The operation strategy of the double energy storage systems is provided, so that the double energy storage systems reduce the charge and discharge switching times of single-group energy storage in the double energy storage systems on the basis of completing the same digestion task, and the service life of the double energy storage systems is prolonged.

The next research work of the invention will increase the energy storage application scene, further perfect the operation strategy of the double energy storage system and the multi-main investment energy storage economical model, and apply the model to the energy storage configuration problems of the distribution network side and the user side.

According to the charge state of the energy storage A, B at the end of the t period and the wind power output and load size of the t+1 period, the invention determines the switching condition of the charge and discharge states of the double energy storage system at the end of the t period, and assumes that the energy storage A is in the charge state and the energy storage B is in the discharge state at the end of the t period, and the charge state value of the energy storage A at the end of the t period is

Scheme 1: when (when)

And->

In the time period t, the charge and discharge states of the energy storage A and the energy storage B do not need to be switched, and in the time period t+1, the energy storage A is in a charge state and performs charge operation, and the energy storage B is in a discharge state but does not operate.

Scheme 2: when S is _oc.A ＝S _oc.B ＝S _oc.max In the time period t, the charge and discharge states of the energy storage A and the energy storage B do not need to be switched, and in the time period t+1, the energy storage A is in a charge state but does not act, and the energy storage B is in a discharge state but does not act.

Scheme 3: when S is _oc.A ＝S _oc.max And S is _oc.min ≤S _oc.B <S _oc.max In the time period t, the charge and discharge states of the energy storage A and the energy storage B need to be switched, and in the time period t+1, the energy storage A is in a discharge state but does not act, and the energy storage B is in a charge state and carries out charging action.

scheme 4: when S is _oc.min ≤S _oc.A ≤S _oc.max And S is _oc.min <S _oc.B ≤S _oc.max In the time period t, the charge and discharge states of the energy storage A and the energy storage B do not need to be switched, and in the time period t+1, the energy storage A is in a charge state but does not act, and the energy storage B is in a discharge state and performs a discharge action.

Scheme 5: when S is _oc.A ＝S _oc.B ＝S _oc.min In the time period t, the charge and discharge states of the energy storage A and the energy storage B do not need to be switched, and in the time period t+1, the energy storage A is in a charge state but does not act, and the energy storage B is in a discharge state but does not act.

Scheme 6: when S is _oc.min <S _oc.A ≤S _oc.max And S is _oc.B ＝S _oc.min When (1): the charge and discharge states of the energy storage A and the energy storage B need to be switched at the end of the t period, and in the t+1 period, the energy storage A is in a discharge state and performs discharge action, and the energy storage B is in a charge state but does not act.

The flow chart of the switching strategy of the double energy storage charging and discharging states is shown in fig. 6A, and the switching schematic diagram of the double energy storage charging and discharging states is shown in fig. 6B.

The selected classical multi-objective algorithm is a non-dominant ordered genetic algorithm (non-dominated sorted genetic algorithm-II, NSGA-II), a multi-objective particle swarm algorithm (multi-objective particle swarm optimization, MOPSO), a multi-objective decomposition evolution algorithm (multi-objective evolutionary algorithm based on decomposition, MOEA/D) and a multi-objective myxobacterial algorithm, and simulation calculation is completed in MATLAB R2015b software in a computer with an Intel Core i7-7500 CPU, a main frequency of 2.7GHz and a memory of 8 GB. Classical double-target ZDT1-6 series functions are selected as test functions, and substitution distance (generational distance, GD), spacing and anti-generation distance evaluation indexes (inverted generational distance, IGD) are selected as evaluation indexes of a multi-target algorithm. The substitution distance represents the proximity between the pareto optimal solution and the real solution, and the smaller the value is, the closer the optimizing result is to the optimal solution, and the better the convergence of the algorithm is; the spacing index represents the standard deviation of the minimum distance from each non-inferior solution to other solutions, and the smaller the value is, the more uniform the non-inferior solution set is; the reverse generation distance represents the average value of the distance from each reference point to the nearest solution, and the smaller the value, the better the algorithm comprehensive performance including convergence and diversity.

To avoid randomness of the test results, each algorithm was run 30 times and the median and quartile differences of the 30 results were calculated for comparison, with the statistical results shown in table 2. As can be seen from Table 3, in the double-objective optimization test result, the GD value, the spa value and the IGD value obtained by the OLMSMA are relatively smaller, especially the GD value and the IGD value are smaller by 1-2 orders of magnitude than those of three classical algorithms, which shows that the proposed algorithm has better diversity and convergence when solving the double-objective optimization problem, and the obtained non-inferior solution set is more uniform. Comparing the results of each parameter of the OLMSMA and the MOSMA shows that the values of the four parameters of the OLMSMA are smaller than the MOSMA, which indicates that the algorithm added with the generalized reverse learning strategy further improves the performance of the algorithm solving problem, and the method has certain effectiveness in solving the double-target optimization problem.

Table 2 algorithm main parameter settings

Table 3 algorithm test results comparison

The improved multi-objective coliform algorithm is brought into an example simulation, and the specific simulation steps are as follows:

1) Inputting simulation parameters of an example model, and setting algorithm parameters such as population size, cycle termination condition and maximum iteration times;

2) Carrying out real number coding on the energy storage capacity and the PCS rated power, wherein the length of a coding chromosome is 2 because the capacity and the power of double energy storage are equal, an ith chromosome is shown as a formula (B1), a primary population x is formed, and a reverse solution of a corresponding mucoid individual is obtained by using a formula (36);

3) Comparing the fitness values of the initial population individuals and the corresponding reverse population individuals, selecting N individuals with better fitness values as the initial population of the whole solving process according to the formula (B2), and enabling the iteration times t to be 1;

4) Updating the positions of individuals in the initial population of the coliform bacteria, and calculating fitness values of N individuals in the updated population by using the formulas (1) and (2);

5) Non-dominant solutions in the population are determined and archived, and dominant solutions in the archive are deleted. Calculating crowding distances of all individuals in the file, and removing individuals with small crowding distances as many as possible;

6) Non-dominant sorting is carried out on non-inferior solutions in the files according to a crowding degree mechanism, and N individuals with the top sorting are selected to be used as initial populations in the next iteration process;

7) Judging whether a termination condition is met or the preset iteration times are reached, if not, enabling t=t+1 to be transferred to the step 4) for iteration continuing, otherwise, outputting a global optimal solution as an energy storage optimal configuration result.

Table 4 simulation parameter settings

Table 5 five energy storage battery parameters tab.c2 Parameters ofthe five types ofenergy storage batteries

Table 6 recyclable materials Tab.C3 Recyclable materials ofthe five types ofenergy storage batteries for five energy storage batteries

Table 7 results of optimal configuration of five batteries under four investment scenarios

In order to further verify the superiority of the double energy storage system, the single energy storage scale parameter configured in the multi-main investment scene is assumed to be 20MWh/10MW, the scale parameters of the energy storage A and the energy storage B configured in the multi-main investment scene are assumed to be 10MWh/10MW, simulation analysis is carried out on two groups of configuration results, and the superiority of the double energy storage system is verified by comparing the aspects of annual energy waste, annual income of operators, service life of the energy storage system, charging and discharging capacity indexes and the like.

Table 8 comparison of multi-agent investment single energy storage and double energy storage simulation effects

As shown in Table 8, the total capacity and the power of the single energy storage system and the double energy storage system are consistent, and the simulation comparison shows that the double energy storage system can bring better benefits to wind farm operators and power grid operators, but because the capacity of the energy storage A and the energy storage B in the double energy storage system is half of the single energy storage capacity, the upper limit value and the lower limit value of the SOC are easy to achieve, so that the probability of deviating from an ideal interval of the SOC is larger, and the charge and discharge capacity index of the double energy storage system is larger than that of the single energy storage system. The investment recovery period and the investment yield in the scene of the double energy storage system are superior to those of the single energy storage system, the consumed wind-discarding electric quantity of the double energy storage system is also higher than that of the single energy storage system, the service life of the double energy storage system is longer than that of the single energy storage system, and the double energy storage system has certain advantages compared with the single energy storage system through comparison. The dual energy storage advantage can be further analyzed according to several aspects, the analysis results are briefly shown in table 9:

1) From the aspect of mechanism modeling: the double energy storage mode is explored from the theoretical mechanism, two battery packs with equal capacity and power are combined to form the double energy storage system, and the working mode of alternately working and synchronously switching is used, so that the energy storage system acts in real time according to the instruction of the energy management system to finish the abandoned wind absorption task. The single energy storage system does not consider the charge-discharge switching strategy, and directly performs charge-discharge work according to the instruction of the energy management system;

TABLE 9 comparison of advantages of different energy storage systems Tab.D2 Comparison of advantages ofdifferent energy storage systems

The foregoing is merely illustrative of specific embodiments of the present invention, and the scope of the invention is not limited thereto, but any changes or substitutions that do not undergo the inventive effort should be construed as falling within the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope defined by the claims.

Claims

1. The double energy storage system layering optimization configuration method based on multi-main investment is characterized in that,

the switching strategy is:

scheme 1: when (when)

And->

3) Repeating the process 2) until the simulation is finished.

2. The method of claim 1, wherein the energy storage charge and discharge capacity index and the multi-investment body income establish an energy storage layered optimization model, and an upper optimization model is responsible for distributing investment operation and maintenance cost and income of the two parties to the energy storage, so as to maximize benefits of the two parties; the lower optimization model is responsible for optimizing the energy storage charging and discharging capacity, so that the double energy storage can keep stronger charging and discharging capacity;

wherein the upper layer objective function is

1) Annual investment costs accounting for life loss

C _inv ＝C _E (E _b,A +E _b,B )+C _P (P _b,A +P _b,B ) (3)

to accurately calculate the service life tau of double energy storage _bat Calculating the influence of the charge and discharge times and the discharge depth of the stored energy A, B on the cycle life;

τ _bat ＝min(T _s,A ,T _s,B ,τ _bat,b ) (5)

the equivalent cycle life corresponding to the discharge depth of the energy storage A, B is 1 respectively, and is determined by the characteristics of the energy storage battery; d (D) _od,A,u 、D _od,B,u The depth of discharge of the stored energy A, B in the u-th cycle; n (N) _ctf,A (D _od,A,u )、N _ctf,B (D _od,B,u ) The discharge depth of the stored energy A, B is D _od,A,u 、D _od,B,u Equivalent cycle life corresponding to the time; h _A 、H _B The total charge-discharge switching times of the stored energy A, B in one year are respectively; t is the number of scheduling periods in one year; v (V) _A,ch (t)、V _B,ch (t)、V _A,dis (t)、V _B,dis (t) TableBinary variable for showing charge and discharge state switching of t period end energy storage A, B, V _A,ch (t)、V _B,ch (t) when taking "1" it means that the energy storage A, B is switched from the charged state to the discharged state at the end of the period t, V _A,dis (t)、V _B,dis (t) taking a "1" to indicate that the stored energy A, B is switched from a discharged state to a charged state at the end of the t period;

2) Annual operation maintenance cost

3) Annual wind-abandoning internet surfing benefits

4) Annual wind power subsidy benefit

5) Annual environmental benefit

F _env ＝F _emi +F _rec (14)

6) Annual energy storage benefit

7) Trade income for annual green license of power grid

wherein: c (C) ₅ Unit cell/kWh for green certificate trade unit price;

8) Annual electricity selling income of power grid

f ₁ ＝F _cur +F _sub,w +α(F _evn +F _sub,s ) (20)

f ₂ ＝F _tgc +F _sel +(1-α)(F _evn +F _sub,s ) (21)

the underlying objective function is:

when participating in the wind-abandoning and absorbing, the charge states of the two stored energy should be maintained in an ideal interval as much as possible so as to ensure enough charge-discharge capacity storage; by using the battery charge-discharge capacity index f ₃ The degree of the deviation of the SOC of the battery from the ideal interval is measured, and the larger the value is, the larger the degree of the deviation of the SOC of the energy storage battery from the ideal interval is; when the SOC of the battery was 0.5, it was demonstrated that the battery had good charge-discharge capacity reserve [25]Set ideal interval of battery SOC as [0.4,0.6 ]]Selecting a stricter interval of [0.45,0.55 ]]The method comprises the steps of carrying out a first treatment on the surface of the The lower layer optimization goal is that the sum of the battery charge and discharge capability index values of the stored energy A, B is minimum in the annual schedule period:

min f ₃ ＝f _3,A +f _3,B (22)

3. The method of claim 1, wherein the constraints are:

1) Energy storage A charge-discharge power constraint

2) Energy storage a state of charge constraints

S _oc,A,min ≤S _oc,A (t+1)≤S _oc,A,max (28)

0≤L _cha,A (t)+L _dis,A (t)≤1 (29)

S _oc,A (0)＝S _oc,A (T) (30)

Wherein: s is S _oc,A (t+1)、S _oc,A (t) is the state of charge of the stored energy at times t and t+1 respectively; τ is the self-discharge rate of the stored energy;

L _cha,A (t) is the charging state of energy storage in the period of t, and the value is 0 or 1, wherein 0 represents floating charge and waiting for discharging, and 1 represents charging;

L _dis,A (t) is a discharge state of energy storage at the moment t, and the value is 0 or 1, wherein 0 represents floating charge waiting for charging, and 1 represents discharging;

S _oc,A (0) Initial SOC for energy storage a; s is S _oc,A (T) is the SOC of the energy storage A at the end of the scheduling period;

3) System power balance constraint

4. The method of claim 1, wherein the classical investment evaluation index investment recovery period and the investment return rate are selected to scientifically evaluate the economic benefit of the double energy storage;

1) Investment recovery period

The investment recovery period is an important index for measuring the project investment risk degree from the time angle, the investment recovery period is selected as an evaluation index, and the calculation formula is as follows:

2) Investment yield