Double-layer optimization method and system for secondary frequency modulation power of energy storage system and medium

Abstract

The invention discloses an energy storage system secondary frequency modulation power double-layer optimization method and system and a medium, and the method comprises the steps: obtaining a regional power grid AGC instruction and a thermal power generating unit frequency modulation output, and calculating a regional energy storage AGC instruction; according to the regional energy storage AGC instruction, combining the frequency modulation cost and SOC recovery to construct a multi-energy storage system frequency modulation power double-layer optimization model; solving the frequency modulation power double-layer optimization model of the multi-energy storage system, and outputting an optimization result; wherein the multi-energy storage system frequency modulation power double-layer optimization model comprises a first optimization sub-model and a second optimization sub-model, and the first optimization sub-model is a model used for energy storage power station power optimization distribution; the second optimization sub-model is a model used for SOC optimization of the energy storage unit, and the first optimization result of the first optimization sub-model is used as the input of the second optimization sub-model. The problem that the frequency modulation cost and the residual frequency modulation capacity of the energy storage power station are different is solved, the SOC of the energy storage unit is too high / too low, and regional frequency modulation optimization can be achieved.

Description

Technical Field

[0001] This invention relates to the field of secondary frequency regulation technology in power systems, specifically to a two-layer optimization method, system, and medium for secondary frequency regulation power in energy storage systems. Background Technology

[0002] For a long time, the secondary frequency regulation task of my country's power system has been mainly undertaken by thermal power units. However, this model has exposed multiple limitations under the current energy landscape: First, the response characteristics are insufficient. Thermal power units have large mechanical inertia, low ramp rate, and slow response speed, making it difficult to track rapidly changing automatic generation control (AGC) commands, and the frequency regulation accuracy cannot meet the needs of refined grid control. Second, there is a contradiction between economy and safety. Frequent increases and decreases in output lead to increased wear and tear on the units and excessive emissions, which increases operating costs and reduces combustion stability, posing safety hazards. Third, there is insufficient capacity support. With the increasing fluctuations in new energy sources, the single thermal power unit frequency regulation mode can no longer cover the grid frequency regulation needs in terms of capacity reserves and response speed, and the risk of frequency stability continues to accumulate.

[0003] Current frequency regulation strategies for energy storage converters, both domestically and internationally, are characterized by decentralized and independent control. Frequency regulation typically involves a single energy storage power station. In regional power grids containing multiple energy storage power stations, the following problems exist: First, it's impossible to coordinate the charging and discharging power of all energy storage power stations within the region, potentially leading to mutual adjustments between stations and unnecessary charging and discharging. Second, it's impossible to compare the frequency regulation costs of different energy storage power stations, making it difficult to formulate the most economically efficient frequency regulation scheme at the regional level. Therefore, how to coordinate the output of all energy storage power stations within a region to achieve optimal regional frequency regulation economics has become an urgent problem to solve.

[0004] Some scholars have studied power coordination and allocation strategies among energy storage units from the perspective of energy storage clusters, but research in the field of frequency regulation is currently limited. In practical engineering, large-scale energy storage power stations consist of multiple energy storage units. The State of Charge (SOC) state of each energy storage unit directly determines the frequency regulation capability of the power station. Excessively high or low SOC reduces the bidirectional regulation capability of the power station, while restoring the SOC of energy storage units helps extend the time for energy storage to participate in frequency regulation. Therefore, in-depth research on the frequency regulation power redistribution problem among multiple sub-units within an energy storage power station based on SOC state has significant engineering value.

[0005] Therefore, existing methods suffer from issues such as varying frequency regulation costs at energy storage power stations and excessively high or low State of Charge (SOC) for each energy storage unit within the power station, thus failing to achieve optimal regional frequency regulation. Summary of the Invention

[0006] To address the issues of discrepancies in frequency regulation costs and remaining frequency regulation capacity of energy storage power stations, as well as excessively high / low State of Charge (SOC) of energy storage units, which prevent the achievement of optimal regional frequency regulation, this invention provides a two-layer optimization method, system, and medium for secondary frequency regulation power in energy storage systems. This invention designs a two-layer optimization strategy for secondary frequency regulation power in energy storage systems that considers both frequency regulation costs and state of charge recovery. In the energy storage power regulation optimization layer, an resistance coefficient related to remaining frequency regulation capacity is introduced. This resistance coefficient, together with the regional frequency regulation cost, constitutes the first objective function, thereby determining the required frequency regulation power for each energy storage station based on economic optimization while considering the remaining frequency regulation capacity of each station. The energy storage unit SOC optimization layer addresses the issue of excessively high / low SOC of energy storage units within the energy storage station by introducing weighted coefficients to restore the SOC of energy storage units and extend the frequency regulation time of energy storage units. Finally, the proposed model is validated in software based on actual AGC data. This invention achieves a two-layer optimization strategy that utilizes energy storage frequency regulation more economically and fully, and has promising application prospects.

[0007] This invention is achieved through the following technical solution:

[0008] In a first aspect, the present invention provides a two-layer optimization method for secondary frequency regulation power of an energy storage system, the method comprising:

[0009] Obtain AGC commands from the regional power grid and frequency regulation output from thermal power units;

[0010] Calculate the regional energy storage AGC command based on the regional power grid AGC command and the frequency regulation output of thermal power units;

[0011] Based on the regional energy storage AGC instructions, a two-layer optimization model for the frequency regulation power of a multi-energy storage system is constructed by combining frequency regulation cost and SOC recovery.

[0012] Solve the two-level optimization model of frequency regulation power for a multi-energy storage system and output the optimization results;

[0013] The multi-energy storage system frequency regulation power dual-layer optimization model includes a first optimization sub-model and a second optimization sub-model. The first optimization sub-model is a model for power optimization allocation of energy storage power stations; the second optimization sub-model is a model for SOC optimization of energy storage units, and the first optimization result of the first optimization sub-model is used as the input of the second optimization sub-model.

[0014] Furthermore, the regional energy storage AGC command is the difference between the regional power grid AGC command and the frequency regulation output of the thermal power unit.

[0015] Furthermore, the construction process of the first optimization sub-model is as follows:

[0016] The resistance coefficient of the energy storage power station is calculated based on the upper limit of SOC, lower limit of SOC, actual SOC value, rated capacity, and total number of energy storage units in the energy storage power station.

[0017] The resistance coefficient is normalized to obtain the normalized resistance coefficient;

[0018] Based on the normalized resistance coefficient and frequency regulation cost, a first objective function is constructed; the frequency regulation cost includes initial investment cost, energy loss cost, and lifetime depreciation cost.

[0019] The total remaining frequency regulation capacity of the regional energy storage power station is calculated based on the normalized resistance coefficient.

[0020] Based on the relationship between the total remaining frequency regulation capacity and the regional energy storage AGC command, the first constraint condition for the regional energy storage power station is determined.

[0021] Furthermore, the formula for calculating the resistance coefficient is:

[0022] ;

[0023] in, , Let be the charging resistance coefficient and the discharging resistance coefficient of energy storage station i at time t, respectively. , S i,j (t-1) represents the upper limit of SOC, the lower limit of SOC, and the actual SOC value at time (t-1) of energy storage unit j within energy storage power station i; E rate,i,j denoted as , where is the rated capacity of energy storage unit j within energy storage power station i; J is the total number of energy storage units within energy storage power station i.

[0024] Furthermore, the expression for the first objective function is: Where I represents the total number of energy storage power stations in the regional power grid; T represents the total number of dispatches; and C represents the total number of dispatches. i (t) represents the frequency regulation cost of power station i at time t, R EF-i (t) is the resistance coefficient.

[0025] Furthermore, based on the relationship between the total remaining frequency regulation capacity and the regional energy storage AGC instructions, the first constraint condition for the regional energy storage power station is determined, including:

[0026] If the total remaining frequency regulation capacity meets the current regional energy storage AGC command, then the first constraint condition is determined as the first constraint condition of the regional energy storage power station; the first constraint condition includes frequency security constraints and frequency regulation output constraints.

[0027] If the total remaining frequency regulation capacity does not meet the current regional energy storage AGC command, then the first and second constraints will be determined as the second constraints of the regional energy storage power station; the first and second constraints include the regional energy storage power station outputting power according to the remaining frequency regulation capacity.

[0028] Furthermore, the construction process of the second optimization sub-model is as follows:

[0029] Calculate the SOC weight of energy storage units within the regional energy storage power station based on the frequency regulation power allocated to the energy storage power station.

[0030] The SOC weights are normalized to obtain the normalized SOC weights.

[0031] Based on the normalized SOC weights, a second objective function is constructed by minimizing the absolute value of the difference between the SOC of all energy storage units in the regional energy storage power station and the ideal state.

[0032] The energy storage unit output constraint, rated power constraint, and SOC upper and lower limit constraints are used as the second constraint conditions of the second objective function.

[0033] Furthermore, the SOC weight is used to allocate frequency regulation power among the energy storage units within a regional energy storage power station according to their respective SOC values.

[0034] Furthermore, the expression for the second objective function is:

[0035]

[0036] in, The normalized SOC weight of energy storage unit j within power station i at time t. For all energy storage units in energy storage power station i, SOC In an ideal state, J represents the total number of energy storage units within energy storage power station i.

[0037] Secondly, the present invention provides a dual-layer optimization system for secondary frequency regulation power of an energy storage system, the system comprising:

[0038] The acquisition unit is used to acquire AGC commands from the regional power grid and frequency regulation output from thermal power units;

[0039] The calculation unit is used to calculate the regional energy storage AGC command based on the regional power grid AGC command and the frequency regulation output of the thermal power unit.

[0040] The model building unit is used to construct a two-layer optimization model for the frequency regulation power of a multi-energy storage system based on the regional energy storage AGC instructions, combined with frequency regulation costs and SOC recovery. The two-layer optimization model for the frequency regulation power of a multi-energy storage system includes a first optimization sub-model and a second optimization sub-model. The first optimization sub-model is a model for power optimization allocation of energy storage power stations. The second optimization sub-model is a model for SOC optimization of energy storage units, and the first optimization result of the first optimization sub-model is used as the input of the second optimization sub-model.

[0041] The optimization solution unit is used to solve the two-layer optimization model of frequency regulation power of multi-energy storage system and output the optimization results.

[0042] Thirdly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for dual-layer optimization of secondary frequency regulation power in an energy storage system.

[0043] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0044] This invention discloses a two-layer optimization method, system, and medium for secondary frequency regulation power in energy storage systems. The invention designs a two-layer optimization model for the frequency regulation power of multiple energy storage systems, considering both frequency regulation cost and SOC recovery. In the frequency regulation power optimization layer of the energy storage power station, an resistance coefficient related to the remaining frequency regulation capacity is introduced. This resistance coefficient, together with the regional frequency regulation cost, constitutes the first objective function. This determines the required frequency regulation power for each energy storage power station based on economic optimization, considering the remaining frequency regulation capacity of each station. The energy storage unit SOC optimization layer addresses the issue of excessively high / low SOC of energy storage units within the power station. A weighted coefficient is introduced to recover the SOC of energy storage units, extending the frequency regulation time. Finally, the proposed model is validated in software based on actual AGC data. This invention achieves a two-layer optimization strategy that utilizes energy storage frequency regulation more economically and fully, demonstrating promising application prospects. This invention solves the problems of differences in frequency regulation cost and remaining frequency regulation capacity of energy storage power stations, as well as excessively high / low SOC of energy storage units, enabling optimal regional frequency regulation. Attached Figure Description

[0045] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0046] Figure 1 This is a flowchart of a two-layer optimization method for secondary frequency regulation power in an energy storage system according to the present invention;

[0047] Figure 2 This is a schematic diagram of a regional power grid containing various types of energy storage power stations according to the present invention;

[0048] Figure 3 This is a block diagram of the dual-layer optimization control strategy (dual-layer optimization model for frequency regulation power of multi-energy storage system) of the present invention;

[0049] Figure 4 This is a detailed flowchart of the dual-layer optimization control strategy (dual-layer optimization model for frequency regulation power of multi-energy storage system) of the present invention;

[0050] Figure 5 This is a schematic diagram illustrating the AGC command response of an energy storage power station under different strategies according to the present invention;

[0051] Figure 6 This is a schematic diagram comparing the output of various energy storage power stations under different strategies of the present invention;

[0052] Figure 7 This invention illustrates the composition of regional energy storage frequency regulation output under different strategies.

[0053] Figure 8 This is a block diagram of a dual-layer optimization system for secondary frequency regulation power in an energy storage system according to the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0055] Although energy storage-assisted frequency regulation has become a research hotspot, current technical solutions still have significant bottlenecks:

[0056] (1) Challenges in State of Charge (SOC) control: During frequency regulation, the energy storage is frequently charged and discharged, which can easily lead to overcharging and over-discharging, resulting in battery life degradation and increased safety risks. At the same time, if the SOC deviates from the optimal range (about 50%), it will reduce the energy storage's continuous frequency regulation capability, and existing strategies are difficult to achieve a dynamic balance between frequency regulation requirements and SOC recovery.

[0057] (2) Lack of optimization of frequency regulation economy: The cost of energy storage participating in frequency regulation includes equipment depreciation, operation and maintenance expenses and battery life loss. Existing power allocation strategies (such as proportional allocation and differential allocation) focus on frequency regulation performance and do not fully consider frequency regulation costs, resulting in some schemes having excellent frequency regulation effects but frequent battery replacement and low economic benefits.

[0058] (3) Insufficient control strategy hierarchy: A single control mode is difficult to take into account multiple objectives such as "frequency response speed, SOC stability and cost optimization". Existing dual-layer control strategies focus on power distribution and equipment protection, and have not formed a collaborative architecture of "upper-layer cost-performance optimization + lower-layer SOC accurate recovery", resulting in problems such as insufficient control accuracy and limited adaptability.

[0059] In summary, given the increasing pressure on power grid frequency regulation due to energy transition, the unsustainability of traditional frequency regulation modes, and the application potential of energy storage technology despite multiple shortcomings in existing strategies, this invention designs a dual-layer optimization of secondary frequency regulation power for energy storage systems that considers both frequency regulation costs and state of charge recovery. This is not only a practical requirement for solving the problem of power grid frequency stability, but also a key path to improve the economy and reliability of energy storage frequency regulation.

[0060] Example 1

[0061] like Figure 1 As shown, the present invention provides a two-layer optimization method for secondary frequency regulation power of an energy storage system, the method comprising:

[0062] Step 1: Obtain the regional power grid AGC command and the frequency regulation output of the thermal power unit; Calculate the regional energy storage AGC command based on the regional power grid AGC command and the frequency regulation output of the thermal power unit.

[0063] In this embodiment, the present invention takes a regional power grid containing i different types of battery energy storage power stations (BESPSs) as the research object, wherein the i battery energy storage power station is composed of j battery energy storage units (BESUs), and its structure is as follows. Figure 2 As shown.

[0064] Figure 2 In the middle, P G (t), P W (t), P V (t), P L (t) represents the power of the thermal power unit, wind farm, photovoltaic power station, and load at time t, respectively; P b,i,j (t) represents the secondary frequency regulation power (hereinafter referred to as frequency regulation power) of energy storage unit j within energy storage power station i at time t; P b,i (t) represents the frequency regulation power of energy storage station i at time t, and their expressions are as follows:

[0065]

[0066] in, and These represent the charging power (downward frequency modulation) and discharging power (upward frequency modulation) of energy storage unit j within energy storage station i at time t. and Let be the charging power and discharging power of energy storage station i at time t, respectively. , The method for determining the charge / discharge indicator is as follows:

[0067]

[0068] Among them, PAGC BESS(t) is the total automatic generation control (AGC) command for the energy storage power station in this area, which is the regional energy storage AGC command.

[0069] Considering that the focus of this invention is on how AGC commands are distributed among energy storage power stations, the following simplification is made: the difference between the regional power grid AGC command and the output of the thermal power unit is taken as the AGC command of the regional energy storage power station (i.e., the regional energy storage AGC command), as shown below:

[0070]

[0071] in, , These represent the AGC command of the power grid and the AGC output of the thermal power unit at time t, respectively.

[0072] Step 2: Based on the regional energy storage AGC instructions, and combined with frequency regulation costs and SOC recovery, construct a two-layer optimization model for the frequency regulation power of the multi-energy storage system;

[0073] This invention addresses the discrepancies between frequency regulation costs and remaining frequency regulation capabilities in energy storage power stations, as well as the issues of excessively high / low State of Charge (SOC) for energy storage units. It proposes a two-layer optimization model for the frequency regulation power of multi-energy storage systems, comprising an energy storage power station frequency regulation power optimization layer (hereinafter referred to as the frequency regulation power optimization layer) and an energy storage unit SOC optimization layer (hereinafter referred to as the SOC optimization layer). Its framework is as follows: Figure 3 As shown.

[0074] Specifically, the frequency regulation power optimization layer proposes a first optimization sub-model, namely the energy storage power station power optimization allocation model, which comprehensively considers the frequency regulation cost and remaining frequency regulation capacity of each energy storage power station and initially allocates the total AGC commands among the energy storage power stations; the first optimization result of the first optimization sub-model is used as the input of the second optimization sub-model.

[0075] The SOC optimization layer proposes a second optimization sub-model, namely the energy storage unit SOC optimization model, which redistributes the results of the frequency regulation power optimization layer among the energy storage units within each energy storage power station, aiming to restore the SOC of the energy storage unit to the optimal state.

[0076] The proposed dual-layer optimization model for frequency regulation power of multi-energy storage systems can achieve hierarchical optimization of frequency regulation power. It can take into account both the economic efficiency of regional power grid frequency regulation and the remaining frequency regulation capacity of each energy storage power station, and can also restore the SOC of energy storage units, thus ensuring the frequency regulation reliability and sustainability of energy storage power stations over long time scales.

[0077] First, the construction of the first optimization sub-model is as follows:

[0078] The frequency regulation power optimization layer of the energy storage power station aims to coordinate the output of all energy storage power stations in the region, and its first optimization result serves as the input of the SOC optimization layer.

[0079] To fully explore the frequency regulation potential of energy storage, this invention does not currently consider the distribution of frequency regulation profits among energy storage power stations, focusing instead on reducing frequency regulation costs from an economic perspective. The basic idea is to construct frequency regulation cost calculation models for different battery types. Constrained by frequency safety and rated power, and considering the remaining frequency regulation capacity of energy storage power stations, AGC commands are allocated among them with the objective of minimizing the total regional frequency regulation cost. This prioritizes power stations with low frequency regulation costs and large remaining frequency regulation capacity, while minimizing the frequency regulation responsibilities of power stations with high costs and small remaining capacity. The method for constructing the energy storage power station frequency regulation power optimization layer model is as follows.

[0080] First, an objective function is established. To facilitate the calculation of energy storage frequency regulation costs and reduce model complexity, this invention adopts a local balancing method to regulate power fluctuations, neglecting line losses caused by long-distance power transmission. Therefore, the frequency regulation cost can be simplified into three main costs that reflect the differences in frequency regulation costs among different batteries: initial investment cost, energy loss cost, and lifespan depreciation cost, as shown below:

[0081]

[0082] in: , , These represent the initial investment cost, charging and discharging energy loss cost, and lifespan depreciation cost of the i-th energy storage power station, respectively. , These represent the unit capacity cost and rated capacity of power station i, respectively. The float charging life of energy storage station i is related to the battery type; r is the discount rate, taken as 8%; s is the maximum number of dispatches in one day, taken as 1440; c e The feed-in tariff is set at 520 yuan / (MWh), and the impact of time-of-use tariffs on frequency regulation costs is not considered; Δt is the dispatch time step, set at 1 minute. , These are the charging and discharging efficiencies of energy storage station i, respectively. , These represent the unit power cost and rated power of energy storage power station i, respectively. k represents the equivalent number of cycles for energy storage power station i at 100% charge / discharge depth. p This is a constant, which can be obtained by fitting the relationship between the number of energy storage cycles and the depth of discharge using actual operating data provided by the battery manufacturer. It is generally between 0.8 and 2.1, and this invention uses 1. It should be noted that the cost of life loss includes the cost of battery life degradation and other costs caused by life degradation, such as replacement costs.

[0083] The frequency regulation cost C of power station i at time t i (t) is shown below:

[0084]

[0085] Simply taking the minimum frequency regulation cost as the primary objective function may result in energy storage power stations with low frequency regulation costs but small remaining frequency regulation capacity bearing the responsibility of frequency regulation, while energy storage power stations with high frequency regulation costs but large remaining frequency regulation capacity remain inactive. This is very likely to dampen the enthusiasm of energy storage power stations with high frequency regulation costs for frequency regulation and is not conducive to improving the overall frequency regulation sustainability of energy storage.

[0086] Therefore, this invention introduces a resistance coefficient. The first objective function is modified. The resistance coefficient represents the degree of resistance of the energy storage power station to charging / discharging. For example, the larger the charging resistance coefficient, the more resistant the energy storage power station is to charging. The idea behind the formation of the resistance coefficient is that the higher the remaining frequency regulation capability during charging / discharging, the smaller the charging / discharging resistance coefficient, and the larger the discharging / charging resistance coefficient. Taking an energy storage power station with high frequency regulation cost but low SOC (larger remaining frequency regulation capability during charging and smaller remaining frequency regulation capability during discharging) as an example, the charging resistance coefficient should be small and the discharging resistance coefficient should be large, thereby increasing the charging willingness and decreasing the discharging willingness of the energy storage power station. Conversely, for an energy storage power station with high frequency regulation cost but high SOC (larger remaining frequency regulation capability during discharging and smaller remaining frequency regulation capability during charging), the discharging resistance coefficient should be small and the charging resistance coefficient should be large, thereby increasing the discharging willingness and decreasing the charging willingness. The formula for the resistance coefficient is shown below:

[0087] ;

[0088] in, , Let be the charging resistance coefficient and the discharging resistance coefficient of energy storage station i at time t, respectively. , , These represent the upper limit of SOC, the lower limit of SOC, and the actual SOC value at time (t-1) for energy storage unit j within energy storage power station i. denoted as , where is the rated capacity of energy storage unit j within energy storage power station i; J is the total number of energy storage units within energy storage power station i.

[0089] To avoid the resistance coefficient calculated by the above formula being too large, it is normalized as follows:

[0090]

[0091] in, , This represents the normalized resistance coefficients for charging and discharging.

[0092] In summary, to achieve optimal allocation of frequency regulation power while balancing the remaining frequency regulation capacity of energy storage power stations with minimizing the frequency regulation cost of the regional power grid, a first objective function is constructed based on the normalized resistance coefficient and frequency regulation cost; the expression of the first objective function is:

[0093]

[0094] Where I represents the total number of energy storage power stations in the regional power grid; T represents the total number of dispatches; and C represents the total number of dispatches. i (t) represents the frequency regulation cost of power station i at time t. The resistance coefficient.

[0095] The first constraint is set below. The frequency regulation power undertaken by the energy storage power station must be subject to the remaining frequency regulation capacity. Restrictions, The total remaining frequency regulation capacity of the regional energy storage power station is calculated based on the SOC state and rated power of the energy storage units within power station i, using the normalized immunity coefficient. The calculation method is as follows:

[0096]

[0097] in, , These represent the remaining frequency regulation capability for charging and the remaining frequency regulation capability for discharging of energy storage station i at time t, respectively.

[0098] After obtaining the remaining frequency regulation capacity of energy storage power station i, the first constraint condition of the regional energy storage power station is determined based on the relationship between the total remaining frequency regulation capacity and the regional energy storage AGC command; the details are as follows:

[0099] If the total remaining frequency regulation capacity meets the current regional energy storage AGC command, then the first constraint condition is determined as the first constraint condition for the regional energy storage power station; the first constraint condition includes frequency security constraints and frequency regulation output constraints, as follows:

[0100]

[0101]

[0102] If the total remaining frequency regulation capacity does not meet the current regional energy storage AGC command, then the first and second constraints will be determined as the second constraint of the regional energy storage power station; the first and second constraints include the regional energy storage power station outputting according to the remaining frequency regulation capacity, as follows:

[0103]

[0104] Second, the construction of the second optimization sub-model is as follows:

[0105] The SOC optimization layer controls the charging and discharging power of each energy storage unit in energy storage power station i based on the frequency regulation power allocated to i and the SOC state of its internal energy storage units, so that the SOC of each energy storage unit moves towards the ideal state S. ideal To restore and improve the sustainability and reliability of energy storage in frequency regulation.

[0106] Introducing SOC weights The frequency regulation power among the energy storage units within energy storage power station i is allocated according to their respective SOC values, as shown below:

[0107]

[0108] in, , These are the charging SOC weight and discharging SOC weight of energy storage unit j in power station i at time t, respectively.

[0109] This formula enables energy storage units with higher SOC values ​​to have a higher discharge weight and a lower charging weight, while energy storage units with lower SOC values ​​have a higher charging weight and a lower discharge weight. This allows for priority charging of energy storage units with low charge levels and priority discharging of energy storage units with high charge levels, thereby restoring the SOC value of the energy storage units.

[0110] The SOC weights are normalized as follows:

[0111]

[0112] in, , These are the charging SOC weights and discharging SOC weights of energy storage unit j within power station i at time t after normalization.

[0113] The SOC optimization layer takes the minimum absolute value of the difference between the SOC of all energy storage units in energy storage power station i and the ideal state as its second objective function. The expression of the second objective function is:

[0114]

[0115] in, The normalized SOC weight of energy storage unit j within power station i at time t. For all energy storage units in energy storage power station i, SOC In an ideal state, J represents the total number of energy storage units within energy storage power station i.

[0116] And the following constraints should be met:

[0117]

[0118]

[0119]

[0120] The above equations represent the energy storage unit output constraints, rated power constraints, and SOC upper and lower limits constraints, respectively.

[0121] Step 3: Solve the two-layer optimization model of frequency regulation power of the multi-energy storage system and output the optimization results.

[0122] The detailed flowcharts of steps 1 to 3 of the present invention are as follows: Figure 4 As shown, firstly, a regional energy storage AGC command is generated. The dispatch center and the control centers of each energy storage power station calculate the remaining frequency regulation capacity of energy storage power station i at time t based on the SOC of the energy storage unit, and generate the resistance coefficient of the frequency regulation power optimization layer and the SOC weight of the SOC optimization layer. Then, a two-layer optimization of frequency regulation power is initiated: In the frequency regulation power optimization layer, when the total remaining frequency regulation capacity of the regional energy storage power station is greater than the command, the frequency regulation power allocated to the energy storage power station is optimized by taking into account both the frequency regulation cost and the remaining frequency regulation capacity; when the total remaining frequency regulation capacity of the regional energy storage power station is less than the command, each energy storage power station outputs power according to its remaining frequency regulation capacity to meet the AGC requirements as much as possible. In the SOC optimization layer, the SOC recovery of the energy storage unit is achieved according to the principle of "prioritizing discharge when SOC is high and prioritizing charging when SOC is low".

[0123] The decision variables of the frequency regulation power optimization layer can be represented by the SOC optimization layer decision variables, ensuring that the decision variables of the entire bi-level programming model consist solely of the SOC optimization layer decision variables. This allows the bi-level optimization model of frequency regulation power in a multi-energy storage system to be transformed into a multi-objective mixed-integer nonlinear programming model, which can then be solved using YALMIP in MATLAB.

[0124] Based on the functionality of the proposed dual-layer optimization model for frequency regulation power of multi-energy storage systems, the effectiveness of the control strategy is evaluated from three aspects: frequency regulation output deviation of energy storage power stations, frequency regulation economy of energy storage power stations, and SOC recovery effect of energy storage units. Specifically:

[0125] 1) Frequency regulation output deviation of energy storage power station

[0126]

[0127] P index P is used to evaluate the deviation between the frequency regulation output of energy storage power stations and AGC commands under different strategies. index The smaller the value, the smaller the deviation.

[0128] 2) Frequency regulation mileage electricity price

[0129] Considering that the cumulative frequency regulation mileage provided by each energy storage power station may vary, a frequency regulation mileage electricity price is defined to evaluate the economic viability, and the calculation formula is as follows:

[0130]

[0131] CFR index is the frequency regulation mileage price, which represents the unit price of frequency regulation power. The lower the CFR index, the better the economic efficiency.

[0132] 3) SOC status

[0133]

[0134] in, Used to evaluate the State of Charge (SOC) of each energy storage unit within energy storage power station i. The smaller the value, the closer the State of Charge (SOC) is to the ideal state, and the better the state. To improve the bidirectional regulation capability of energy storage, S... ideal Take 0.5.

[0135] In practical implementation, to verify the effectiveness and superiority of the proposed dual-layer optimization model for frequency regulation power of multi-energy storage systems, a simulation analysis of the regional power grid was conducted in MATLAB. It is assumed that the total installed capacity of the region is 2800MW, with new energy accounting for 20%, including three energy storage power stations with different battery types suitable for frequency regulation applications. The total installed capacity of energy storage is 56MW / 56MWh, accounting for 10% of the new energy installed capacity. Each energy storage power station consists of eight energy storage units. The battery type, some technical parameters, and construction costs of each energy storage power station are shown in Table 1, and the initial SOC values ​​of the energy storage units in each power station are shown in Table 2.

[0136] Table 1. System parameters for the example

[0137]

[0138] Table 2 Initial SOC values ​​of energy storage units in each energy storage power station

[0139]

[0140] Using the above example system, the effects of four different strategies on controlling energy storage power stations to participate in frequency regulation are compared. Strategy 1 is a dynamic proportional allocation strategy, which improves upon the equal proportional allocation method by allocating frequency regulation tasks according to the proportion of remaining frequency regulation capacity. Strategy 2 only takes the minimum frequency regulation cost as the objective function, i.e., optimal economic allocation, without considering the remaining frequency regulation capacity of energy storage power stations and the SOC optimization layer. Strategy 3 only has a frequency regulation power optimization layer for energy storage power stations, without a SOC optimization layer. Strategy 4 is the strategy of this invention, i.e., a two-layer optimization model for frequency regulation power of multi-energy storage systems.

[0141] Simulations were performed using actual AGC data from a certain region, showing the response of energy storage power stations to AGC commands under four different strategies. Figure 5 As shown.

[0142] Depend on Figure 5It can be seen that strategies 1 and 2 were unable to respond to AGC commands at 167 minutes due to insufficient energy storage for frequency regulation; strategies 3 and 4 were limited by the total installed capacity of energy storage and experienced power deficits at 169 minutes; after 172 minutes, strategy 4 provided more frequency regulation power than strategy 3. The cumulative frequency regulation mileage, frequency regulation power deficit, and output deviation indicators of each energy storage power station are shown in Table 3.

[0143] Table 8. Cumulative FM mileage and output deviation under different strategies

[0144]

[0145] It can be seen that the cumulative frequency regulation mileage provided by strategies 3 and 4 is higher than that of strategies 1 and 2; strategy 4 has the smallest power deficit, only 108.316MW, followed by strategy 3, and then strategies 1 and 2; the output deviation of strategy 4 is only 0.314, which is much smaller than the other three comparison strategies. The reasons will be analyzed in detail below.

[0146] The composition of power output and regional frequency regulation output of each energy storage power station under different strategies are as follows: Figure 6 , 7 As shown, discharging is positive and charging is negative. Figure 6 (a), (b), (c), and (d) show the output comparison of each energy storage power station under strategy 1, strategy 2, strategy 3, and strategy 4, respectively. Figure 7 (a), (b), (c), and (d) show the composition of regional energy storage frequency regulation output under strategies 1, 2, 3, and 4, respectively. Calculations show that, under the parameters of this invention's example, the frequency regulation power costs of power plants 1-3 are RMB 15.801 / MW, RMB 13.198 / MW, and RMB 7.784 / MW, respectively. Power plant 3 has the lowest unit frequency regulation cost, while power plant 1 has the highest.

[0147] The reason why the output deviation of strategies 3 and 4 is smaller than that of strategies 1 and 2 is that strategies 3 and 4 take into account the remaining frequency regulation capability in the objective function of the frequency regulation power optimization layer. This makes power station 1, which has a larger remaining frequency regulation capability for discharging, bear more discharge power than strategy 2 in the first 100 minutes, and power station 2, which has a larger remaining frequency regulation capability for charging, bear more charging power than strategy 2 in the first 40 minutes. This allows the overall energy storage capacity to be restored to a certain extent, and more energy can be released to track AGC commands at 167 minutes.

[0148] Conversely, Strategy 2, by strictly adhering to the goal of minimizing frequency regulation cost when allocating frequency regulation power, resulted in Power Station 3, with the lowest frequency regulation cost, bearing the majority of the frequency regulation power in the first 160 minutes. Power Stations 1 and 2, with higher frequency regulation costs, mostly only contributed power when Power Station 3's frequency regulation power was insufficient. This was detrimental to improving the frequency regulation capabilities of Power Stations 1 and 2, which had poor State of Charge (SOC), leading to insufficient frequency regulation power after 167 minutes. Strategy 1, which allocated frequency regulation power dynamically, resulted in a relatively balanced distribution of frequency regulation power among the three power stations, but this also hindered the recovery of energy storage power. At 167 minutes, insufficient energy storage power also caused a power output deviation.

[0149] The reason why the output deviation of Strategy 4 is smaller than that of Strategy 3 is that the SOC optimization layer in Strategy 4 will distribute the power regulation of the power plant to each energy storage unit according to the SOC state, thereby realizing the priority recovery of the SOC of the poor state. In contrast, Strategy 3 does not have a SOC optimization layer and distributes the power regulation of the energy storage unit evenly. The recovery effect of the energy storage unit with poor SOC is not as good as that of Strategy 4. Therefore, Strategy 4 provides more power regulation than Strategy 3.

[0150] Table 4 shows the frequency regulation costs and frequency regulation mileage electricity prices for each power station under the four strategies. As shown in Table 4, Strategy 2 has the lowest total frequency regulation cost, 9.67% lower than Strategy 1, 11.31% lower than Strategy 3, and 12.24% lower than Strategy 4. This is because, on the one hand, Strategy 2 aims to minimize frequency regulation costs, consistently allocating frequency regulation power from power station 3, which has the lowest frequency regulation cost, while power stations 1 and 2, which have higher costs, receive relatively less frequency regulation power. On the other hand, Strategy 2 provides the least frequency regulation mileage; while the other three strategies are still incurring energy losses and lifetime depreciation costs due to frequency regulation, Strategy 2's costs are already zero. The higher total frequency regulation costs of Strategies 3 and 4 are due to consideration of remaining frequency regulation capacity; when allocating frequency regulation power, a portion of the power is allocated to power stations 1 and 2, which have higher frequency regulation costs. Additionally, the higher accumulated frequency regulation mileage results in higher energy losses and lifetime depreciation costs.

[0151] Table 4 Frequency modulation costs under different strategies

[0152]

[0153] Therefore, to more accurately evaluate the economics of the four strategies, frequency regulation mileage pricing is required. Table 4 shows that Strategy 4 has the lowest frequency regulation mileage pricing and the best economics, outperforming Strategies 1, 2, and 3 by 4.57%, 0.35%, and 0.32%, respectively. The reasons are as follows: Frequency regulation mileage pricing is related to the cumulative frequency regulation mileage and total frequency regulation cost. Strategy 4's frequency regulation power optimization layer uses the lowest resistance coefficient and frequency regulation cost as its objective function, which not only makes its frequency regulation power allocation economical but also restores the power output of power plants 1 and 2, resulting in the highest cumulative frequency regulation mileage for Strategy 4, hence the lowest frequency regulation mileage pricing. Strategy 3, lacking a SOC optimization layer, has a lower cumulative frequency regulation mileage than Strategy 4, leading to an increased frequency regulation mileage pricing, but it still maintains good economics compared to Strategies 1 and 2.

[0154] It is easy to see that Strategy 1 has the highest frequency regulation mileage price and poor economic efficiency; Strategy 2 has a lower frequency regulation mileage price than Strategy 1, but the frequency regulation output deviation is the largest, and the improvement in economic efficiency comes at the cost of technical efficiency; while the resistance coefficients proposed by Strategies 3 and 4 can incorporate the remaining frequency regulation capacity into the optimization model, achieving better economic efficiency while reducing output deviation. In summary, the dual-layer optimization model for frequency regulation power of multi-energy storage systems proposed in this invention can effectively balance the remaining frequency regulation capacity and frequency regulation cost of energy storage power stations to achieve optimal allocation of frequency regulation power.

[0155] The SOC state is analyzed below. Table 8.5 shows the SOC indices under different strategies. As shown in Table 8.5, Strategy 4's SOC state is superior to the other strategies, being 19.05% better than Strategy 1, 33.29% better than Strategy 2, and 5.92% better than Strategy 3. Therefore, this invention can effectively restore the SOC state of the energy storage unit based on the initial SOC value.

[0156] Table 5 Comparison of SOC Indicators

[0157]

[0158] In summary, for the current power system facing rapid development of new energy sources and a shortage of frequency regulation resources, the dual-layer optimization model for frequency regulation power of multi-energy storage systems in this invention can make more economical and fuller use of energy storage for frequency regulation, and has good application prospects.

[0159] Furthermore, the existing energy storage capacity cannot meet all frequency regulation needs during the entire frequency regulation process. Therefore, the power plant capacity will be increased by 3 times, i.e., the total installed energy storage capacity will be 168MW / 168MWh. The impact of capacity configuration on control strategies will be explored. Key technical indicators under different strategies are shown in Table 6.

[0160] Table 6 Technical parameters for expanding energy storage capacity

[0161]

[0162] As shown in Table 6, Strategy 2 exhibits the best economic efficiency when the energy storage capacity is expanded, while the economic efficiency of Strategy 4 decreases. This is because Strategy 4 utilizes power plants 1 and 2, which have higher frequency regulation costs, to restore the State of Charge (SOC) of the energy storage units in power plants 1 and 2, thus increasing the total frequency regulation cost. Furthermore, due to the expanded power plant capacity, the restored electricity from Strategy 4 is not reused, resulting in a lack of advantage in cumulative frequency regulation mileage and consequently, an increase in the frequency regulation mileage price for Strategy 4. However, Strategy 4 possesses the best SOC index, with a superior SOC state compared to other strategies, thereby improving its subsequent bidirectional regulation capability.

[0163] It is not difficult to see that when the capacity ratio increases to 30%, although the output deviation of each strategy is improved, the electricity price of frequency regulation mileage increases significantly compared to the 10% capacity ratio. This is because the initial investment cost required for capacity expansion will increase, and this expanded capacity is often not fully utilized, resulting in the waste of valuable energy storage resources.

[0164] In summary, when the energy storage capacity is small, Strategy 4 can make full use of the existing energy storage capacity and complete as many frequency regulation tasks as possible with less configuration; when the energy storage capacity configuration is large, Strategy 2 has the best economy, while Strategy 4 has the best SOC state. As the scheduling time increases, Strategy 4 will inevitably be able to provide more frequency regulation power.

[0165] Example 2

[0166] like Figure 8 As shown, the difference between this embodiment and Embodiment 1 is that this embodiment provides a two-layer optimization system for secondary frequency regulation power of an energy storage system, which corresponds one-to-one with the two-layer optimization method for secondary frequency regulation power of an energy storage system in Embodiment 1; the system includes:

[0167] The acquisition unit is used to acquire AGC commands from the regional power grid and frequency regulation output from thermal power units;

[0168] The calculation unit is used to calculate the regional energy storage AGC command based on the regional power grid AGC command and the frequency regulation output of the thermal power unit.

[0169] The model building unit is used to construct a two-layer optimization model for the frequency regulation power of a multi-energy storage system based on the regional energy storage AGC instructions, combined with frequency regulation costs and SOC recovery. The two-layer optimization model for the frequency regulation power of a multi-energy storage system includes a first optimization sub-model and a second optimization sub-model. The first optimization sub-model is a model for power optimization allocation of energy storage power stations. The second optimization sub-model is a model for SOC optimization of energy storage units, and the first optimization result of the first optimization sub-model is used as the input of the second optimization sub-model.

[0170] The optimization solution unit is used to solve the two-layer optimization model of frequency regulation power of multi-energy storage system and output the optimization results.

[0171] The execution process of each unit can be carried out according to the process steps of the energy storage system secondary frequency regulation power dual-layer optimization method in Embodiment 1, and will not be described in detail in this embodiment.

[0172] Meanwhile, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned method for dual-layer optimization of secondary frequency regulation power in an energy storage system.

[0173] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0174] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0175] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0176] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0177] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A two-layer optimization method for secondary frequency regulation power in an energy storage system, characterized in that, The method includes: Obtain AGC commands from the regional power grid and frequency regulation output from thermal power units; Calculate the regional energy storage AGC command based on the regional power grid AGC command and the frequency regulation output of the thermal power unit. Based on the regional energy storage AGC instructions, a two-layer optimization model for the frequency regulation power of a multi-energy storage system is constructed by combining frequency regulation cost and SOC recovery. Solve the two-layer optimization model of frequency regulation power of the multi-energy storage system and output the optimization results; The multi-energy storage system frequency regulation power dual-layer optimization model includes a first optimization sub-model and a second optimization sub-model. The first optimization sub-model is a model for power optimization allocation of energy storage power stations; the second optimization sub-model is a model for SOC optimization of energy storage units, and the first optimization result of the first optimization sub-model is used as the input of the second optimization sub-model.

2. The method for dual-layer optimization of secondary frequency regulation power in an energy storage system according to claim 1, characterized in that, The regional energy storage AGC command is the difference between the regional power grid AGC command and the frequency regulation output of the thermal power unit.

3. The method for dual-layer optimization of secondary frequency regulation power in an energy storage system according to claim 1, characterized in that, The construction process of the first optimization sub-model is as follows: The resistance coefficient of the energy storage power station is calculated based on the upper limit of SOC, lower limit of SOC, actual SOC value, rated capacity, and total number of energy storage units in the energy storage power station. The resistance coefficient is normalized to obtain the normalized resistance coefficient; Based on the normalized resistance coefficient and frequency modulation cost, a first objective function is constructed; the frequency modulation cost includes initial investment cost, energy loss cost, and lifetime depreciation cost. The total remaining frequency regulation capacity of the regional energy storage power station is calculated based on the normalized resistance coefficient. Based on the relationship between the total remaining frequency regulation capacity and the regional energy storage AGC command, the first constraint condition of the regional energy storage power station is determined.

4. The method for dual-layer optimization of secondary frequency regulation power in an energy storage system according to claim 3, characterized in that, The formula for calculating the resistance coefficient is: ； in, , Let be the charging resistance coefficient and the discharging resistance coefficient of energy storage station i at time t, respectively. , S i,j (t-1) represents the upper limit of SOC, the lower limit of SOC, and the actual SOC value at time (t-1) of energy storage unit j within energy storage power station i; E rate,i,j denoted as , where is the rated capacity of energy storage unit j within energy storage power station i; J is the total number of energy storage units within energy storage power station i.

5. The method for dual-layer optimization of secondary frequency regulation power in an energy storage system according to claim 3, characterized in that, The expression for the first objective function is: Where I represents the total number of energy storage power stations in the regional power grid; T represents the total number of dispatches; and C represents the total number of dispatches. i (t) represents the frequency regulation cost of power station i at time t, R EF-i (t) is the resistance coefficient.

6. The method for dual-layer optimization of secondary frequency regulation power in an energy storage system according to claim 3, characterized in that, Based on the relationship between the total remaining frequency regulation capacity and the regional energy storage AGC command, the first constraint condition for the regional energy storage power station is determined, including: If the total remaining frequency regulation capacity meets the current regional energy storage AGC command, then the first constraint condition is determined as the first constraint condition of the regional energy storage power station; the first constraint condition includes frequency security constraints and frequency regulation output constraints. If the total remaining frequency regulation capacity does not meet the current regional energy storage AGC command, then the first and second constraints will be determined as the second constraints of the regional energy storage power station; the first and second constraints include the regional energy storage power station outputting power according to the remaining frequency regulation capacity.

7. The method for dual-layer optimization of secondary frequency regulation power in an energy storage system according to claim 1, characterized in that, The construction process of the second optimization sub-model is as follows: Calculate the SOC weight of energy storage units within the regional energy storage power station based on the frequency regulation power allocated to the energy storage power station. The SOC weights are normalized to obtain normalized SOC weights. Based on the normalized SOC weights, a second objective function is constructed by minimizing the absolute value of the difference between the SOC of all energy storage units in the regional energy storage power station and the ideal state. The energy storage unit output constraint, rated power constraint, and SOC upper and lower limit constraints are used as the second constraint conditions for the second objective function.

8. The method for dual-layer optimization of secondary frequency regulation power in an energy storage system according to claim 1, characterized in that, The SOC weight is used to allocate frequency regulation power among energy storage units within a regional energy storage power station according to their respective SOC values.

9. A dual-layer optimization system for secondary frequency regulation power in an energy storage system, characterized in that, The system includes: The acquisition unit is used to acquire AGC commands from the regional power grid and frequency regulation output from thermal power units; The calculation unit is used to calculate the regional energy storage AGC command based on the regional power grid AGC command and the frequency regulation output of the thermal power unit; The model building unit is used to construct a two-layer optimization model for the frequency regulation power of a multi-energy storage system based on the regional energy storage AGC instruction and in combination with frequency regulation cost and SOC recovery. The two-layer optimization model for the frequency regulation power of the multi-energy storage system includes a first optimization sub-model and a second optimization sub-model. The first optimization sub-model is a model for power optimization allocation of energy storage power stations. The second optimization sub-model is a model for SOC optimization of energy storage units, and the first optimization result of the first optimization sub-model is used as the input of the second optimization sub-model. The optimization solution unit is used to solve the two-layer optimization model of the frequency regulation power of the multi-energy storage system and output the optimization results.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements a two-layer optimization method for secondary frequency regulation power of an energy storage system as described in any one of claims 1 to 8.

Images