Collaborative optimization method and system for energy storage regulation and control based on source-load fluctuation quantization

Abstract

The invention provides a collaborative optimization method and system for energy storage regulation and control based on source-load fluctuation quantization, and relates to the technical field of electric power. The method comprises the steps of performing discretization division on a total load fluctuation interval and a total output fluctuation interval of a power system, and constructing a source-load fluctuation matching matrix; based on the source-load fluctuation matching matrix, establishing a mapping relation from a source-load fluctuation state to a system regulation and control demand, and generating a regulation and control demand signal interval; constructing an energy storage scheduling model based on dynamic programming, taking a regulation and control demand signal interval as an excitation signal, taking maximization of the sum of the normalized system source-load matching degree improvement quantity and the relative achievement rate of the energy storage scheduling comprehensive benefit index as an optimization target, and carrying out collaborative optimization solution, and outputting an optimal energy storage scheduling strategy corresponding to the optimization target. The problems that in the prior art, energy storage economic benefits and power grid dispatching requirements cannot be considered at the same time, so that the energy storage participation degree is low, and the system adjusting capacity is limited are solved.

Description

Technical Field

[0001] This application relates to the field of power technology, and in particular to a collaborative optimization method and system for energy storage regulation based on source-load fluctuation quantization. Background Technology

[0002] As the proportion of installed capacity from renewable energy sources gradually increases, the difference between source and load fluctuations continues to widen. The system faces the challenge of ensuring power supply during peak hours and absorbing renewable energy during off-peak hours. For example, at the power system operation level, the seasonal fluctuations of a high proportion of renewable energy lead to a dual challenge of supply-demand balance and absorption contradictions. Taking a certain provincial power grid as an example, when the proportion of renewable energy generation reaches 30%, the renewable energy curtailment rate is as high as 23%, highlighting the inadequacy of the system's regulation capacity.

[0003] Currently, some research has been conducted on the participation of independent energy storage in dispatching, but energy storage participation in the electricity market also faces multiple challenges: First, the lack of a market-based pricing mechanism limits profit margins; second, participating in trading as a power generation company during discharge may reduce the interests of traditional power generation entities; third, treating energy storage as a power user during charging may lead to energy storage only charging during off-peak hours, restricting its flexibility. This provides a reference for the optimal allocation and market participation of flexible resources such as energy storage.

[0004] To address the three types of problems mentioned above by guiding energy storage to participate in the electricity market, the following key aspects need to be considered: First, a method for quantifying source-load fluctuation differences needs to be constructed. Second, a control signal adapted to energy storage needs to be constructed using the source-load supply and demand differences. Third, a collaborative optimization algorithm that balances revenue and source-load matching needs to be developed, ensuring energy storage revenue while continuously participating in system scheduling and operation, playing a "ballast" role in medium- and long-term transactions, and reducing the structural price difference between the medium- and long-term market and the spot market.

[0005] To address the issue of source-load fluctuation quantification, the paper "Wu Wenchuan, Xu Shuwei, Yang Yue, et al. Risk quantification probabilistic scheduling for high-proportion new energy power systems [J]. Automation of Electric Power Systems, 2023, 47(15):3-11" first proposed the definition and framework of risk quantification probabilistic scheduling, providing a systematic solution for promoting the safe operation of new power systems. The paper "Ye J, Xie L, Ma L, et al. Low-carbon optimal scheduling for multi-source power systems based on source-load matching under active demand response [J]. SolarEnergy, 2024, 267: 112241" introduces the source-load difference index to measure the degree of source-load matching, balancing the economic cost and safety performance of system scheduling. However, the paper relies on traditional peak-valley electricity pricing, which is slightly insufficient to adapt to the fluctuation characteristics of new energy. The paper "Li J, Zhao J, Chen Y, et al. Optimal sizing for a wind-photovoltaic-hydrogen hybrid system considering levelized cost of storage and source-load interaction[J]. International Journal of Hydrogen Energy, 2023, 48(11): 4129-4142" proposes two matching indices to motivate source-load interaction, but the proposed model focuses too much on economic efficiency and ignores the impact of source-load fluctuations on system operation. The paper "Gu Guangrong, Yang Peng, Tang Bo, et al. A method for improving the balance capacity of distribution networks through source-load-storage synergistic optimization[J]. Proceedings of the CSEE, 2024, 44(13): 5097-5109" proposes a quantitative evaluation method for source-load power time-series balance based on information entropy theory, which promotes the consumption of new energy sources while maximizing the balance capacity of the power grid.

[0006] To address the issue of constructing a market-based electricity pricing mechanism to guide energy storage participation, the paper "Zhao D, Jafari M, Botterud A, et al. Strategic energy storage investments: A case study of the CAISO electricity market[J]. Applied Energy, 2022, 325: 119909" analyzes CAISO (California Independent System Operator) data and points out that energy arbitrage had the highest profit share among 11 energy storage application scenarios surveyed in 2019. The paper "Xiao J, He G, Fan S, et al. Substitute energy price market mechanism for renewable energy power system with generalized energy storage[J]. Applied Energy, 2022, 328: 120219" proposes an innovative market mechanism adapted to high-proportion renewable energy systems. By exploring the energy and regulation value of energy storage and combining it with the Vickrey-Clarke-Groves (VCG) mechanism to optimize the trading design, it demonstrates that this mechanism is more effective than traditional models in incentivizing energy storage participation in market transactions and system regulation, effectively guiding energy storage to achieve energy time shifting.

[0007] To address the issue of independent energy storage participating in the electricity market to improve system dispatch and operation, the paper "Ju L, Lv S, Zhang Z, et al. Data-driven two-stage robust optimization dispatching model and benefit allocation strategy for a novel virtual power plant considering carbon-green certificate equivalence conversion mechanism[J]. Applied Energy,2024, 362: 122974" proposes a data-driven two-stage robust optimization dispatching model that considers the carbon-green certificate mechanism. This model maximizes the trading revenue of the virtual power plant in the medium and long term and the spot market through a benefit allocation strategy. However, this model mainly focuses on the profit distribution among aggregators. Among them, the literature "Zhang Xueqiu, Liu Dunnan, Wang Xiaolu, et al. Research on intra-month fusion trading model and unified price limit considering the connection between medium- and long-term electricity and spot markets [J]. Power System Technology, 2024, 48(4):1418-1430" analyzes the key issues in market connection. The literature "Wang Xiaoang, Zou Peng, Ren Yuan, et al. Problems and countermeasures of connection between medium- and long-term electricity and spot markets in Shanxi [J]. Power System Technology, 2022, 46(1):20-27" proposes a continuous operation scheme based on standard energy blocks. The literature "Liu Dunnan, Li Zhu, Dong Zhixin, et al. Design of continuous operation scheme for medium- and long-term electricity market based on standard energy block contracts [J]. Power System Technology, 2023, 47(1):129-144" studies the intra-month fusion trading model. Although these studies clarify the energy relationship in multi-stage transactions, the large difference in the connection time scale limits their practical application effect in market coordination. Furthermore, the paper "Li Junzhou, Zhao Jinbin, Chen Yiwen, et al. Optimization of capacity configuration for electro-hydrogen (P2H) equipment considering dynamic power range and hydrogen production efficiency [J]. Journal of Electrical Engineering, 2023, 38(18): 4864-4874+4920" proposes that the dynamic operating characteristics and efficiency of energy conversion equipment are crucial to the overall economy and energy utilization efficiency of the system. The paper "Chen Minghao, Zhu Yueyao, Sun Yi, et al. Predictive regulation of integrated energy system considering high-penetration photovoltaic absorption and deep reinforcement learning [J]. Journal of Electrical Engineering, 2024, 39(19): 6054-6071+6103" explores how to achieve a balance between system economy and energy utilization efficiency through optimized scheduling while considering the uncertainty of new energy sources.

[0008] Therefore, there is an urgent need for a comprehensive method that starts from the grid side, can accurately quantify source-load fluctuations, generate dynamic guidance signals accordingly, and verify their synergistic effectiveness in improving system performance and ensuring energy storage benefits, so as to guide flexible resources such as independent energy storage to participate in power system dispatch efficiently and sustainably. Summary of the Invention

[0009] This application provides a collaborative optimization method for energy storage regulation based on source-load fluctuation quantification to solve the following technical problems: how to construct a source-load fluctuation difference quantification scheme, how to design system regulation requirements that meet the characteristics of energy storage, and how to enable energy storage to fully play its regulatory role based on system regulation requirements.

[0010] Firstly, this application provides a collaborative optimization method for energy storage regulation based on source-load fluctuation quantization, including: The total load fluctuation range and total output fluctuation range of the power system are discretized and a source-load fluctuation matching matrix is ​​constructed to quantify the source-load fluctuation differences in each time period. Based on the source-load fluctuation matching matrix, a mapping relationship from the source-load fluctuation state to the system control requirements is established, and a control requirements signal range is generated. A dynamic programming-based energy storage scheduling model is constructed, using the aforementioned regulation demand signal range as the excitation signal. The optimization objective is to maximize the sum of the normalized system source-load matching degree improvement and the relative achievement rate of the comprehensive benefit index of energy storage scheduling. The model is then used for collaborative optimization and the optimal energy storage scheduling strategy corresponding to the optimization objective is output.

[0011] Secondly, this application provides a collaborative optimization system for energy storage regulation based on source-load fluctuation quantization, used to implement the method described above, the system comprising: The source-load fluctuation quantification module is configured to discretize the total load fluctuation range and the total output fluctuation range of the power system, and construct a source-load fluctuation matching matrix to quantify the source-load fluctuation differences in each time period. The regulation demand determination module is configured to establish a mapping relationship from the source-load fluctuation state to the system regulation demand based on the source-load fluctuation matching matrix, and generate a regulation demand signal range. The dynamic programming module is configured to construct an energy storage scheduling model based on dynamic programming. It uses the regulation demand signal range as the excitation signal, and takes maximizing the sum of the normalized system source-load matching degree improvement and the relative achievement rate of the comprehensive benefit index of energy storage scheduling as the optimization objective. It performs collaborative optimization and outputs the optimal energy storage scheduling strategy corresponding to the optimization objective.

[0012] The collaborative optimization method for energy storage regulation based on source-load fluctuation quantization provided in this application has at least the following beneficial effects: 1) Achieved precise quantification of source-load fluctuation differences: Clustering of source-load fluctuations caused by random factors such as fluctuations in new energy output, and forming full-state combinations from the divided sub-intervals. Then, the full-state combinations are scored to construct a source-load fluctuation matching matrix, quantifying the source-load fluctuation differences in each time period.

[0013] 2) An effective market incentive mechanism has been established: The dynamic control demand range proposed in this application can reflect the flexibility demand of the power grid in real time, form an effective control demand signal, guide energy storage to charge and discharge when the system needs it most, and significantly improve resource utilization efficiency.

[0014] 3) Achieved synergistic win-win between power grid and energy storage: Through a multi-objective optimization framework based on relative achievement rate, this application can maximize the supporting role of energy storage in the power grid while ensuring reasonable economic benefits for energy storage, and achieve synergistic improvement of source-load matching degree and energy storage benefits, providing a feasible technical path for the stable and economical operation of new power systems. Attached Figure Description

[0015] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0016] Figure 1 A technical roadmap for a collaborative optimization method for energy storage regulation based on source-load fluctuation quantization, provided for embodiments of this application; Figure 2 A flowchart illustrating a collaborative optimization method for energy storage regulation based on source-load fluctuation quantization, provided in an embodiment of this application; Figure 3 This application provides a schematic diagram illustrating the evolution of electricity market transaction curves and transaction intervals in an embodiment. Figure 4 A comparison diagram of source-load states before and after source-side / independent energy storage regulation is provided for embodiments of this application; Figure 5 A comparison diagram of source-load matching degree before and after source-side / independent energy storage scheduling provided in the embodiments of this application; Figure 6 A comparison chart of MPRC indicators for different energy storage configurations guided by FPRI, provided for embodiments of this application; Figure 7 MPRC index evaluation diagram for different energy storage configurations under dynamic regulation demand guidance provided in the embodiments of this application; Figure 8 MPRC index evaluation diagram for different energy storage configurations guided by FPRI, provided for embodiments of this application; Figure 9A radar chart comparing multiple indicators of L6C8 energy storage based on relative achievement rate is provided for embodiments of this application. Figure 10 Time utilization rate with source-load matching degree / energy storage revenue as the target under different guidance methods provided in the embodiments of this application; Figure 11 A comparison chart of energy storage revenue and matching degree under different energy storage configurations provided in the embodiments of this application; Figure 12 This application provides a comparison of the trends in improved curtailment rates under different energy storage configurations in the embodiments of this application. Figure 13 A comparison chart of time / capacity / overall utilization rate under different energy storage configurations provided in the embodiments of this application; Figure 14 A comparison chart of energy storage revenue and matching degree under different energy storage configurations provided in the embodiments of this application; Figure 15 This is a comparison trend diagram of the improved curtailment rate under different energy storage configurations provided in the embodiments of this application; Figure 16 A comparison chart of time / capacity / overall utilization rate under different energy storage configurations provided in the embodiments of this application; Figure 17 Energy storage based on relative achievement rate provided in the embodiments of this application L 2 C 8. Multi-indicator comparison radar chart; Figure 18 Energy storage based on relative achievement rate provided in the embodiments of this application L 6 C 8. Multi-indicator comparison radar chart; Figure 19 This is a load fluctuation clustering inflection point diagram provided in an embodiment of this application; Figure 20 This is a structural diagram of a collaborative optimization system for energy storage regulation based on source-load fluctuation quantization, provided in an embodiment of this application.

[0017] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0018] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0019] The collection, storage, use, processing, transmission, provision, and disclosure of financial data or user data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0020] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0021] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0022] The new power system is undergoing profound changes to address the challenges posed by the high proportion of renewable energy integration. During this transformation, the system's source-load fluctuation characteristics exhibit three significant features: (1) The differences and uncertainties in system source-load fluctuations have increased significantly. With the continuous increase in the proportion of new energy installed capacity, the system operation scenarios are becoming increasingly diversified. Taking a typical province as an example, the daily fluctuations of photovoltaic and wind power can reach 82% and 88% of the installed capacity, respectively. This drastic fluctuation characteristic makes it difficult for traditional deterministic optimization methods to meet the system regulation requirements.

[0023] (2) The system regulation demand exhibits significant spatiotemporal distribution characteristics. Due to the seasonal fluctuations of a high proportion of new energy sources, provincial power systems face differentiated regulation pressures at different times, with peak-shaving demand in some areas increasing by 30%-40% in winter compared to summer. This seasonal difference exacerbates the system's peak-shaving pressure and also highlights the contradiction between ensuring power supply during peak hours and absorbing new energy sources during off-peak hours.

[0024] (3) The increased proportion of new energy sources makes it difficult for energy storage to fully play its regulatory role.

[0025] To address the aforementioned challenges, embodiments of this application provide a collaborative optimization method for energy storage regulation based on source-load fluctuation quantization, such as... Figure 1 As shown, this method first integrates historical load / output curves and, based on statistical analysis of historical curves and extreme values, determines the total output fluctuation range and the total load fluctuation range. Second, it constructs a source-load state matrix to accurately quantify source-load fluctuation differences. Specifically, it first discretizes the total load and output fluctuation range of the regional power grid, dividing it into several sub-ranges; by combining the source-load sub-ranges, it generates a fluctuation matching matrix, completing the quantitative characterization of source-load fluctuation characteristics. Finally, it combines the quantified fluctuation range with the regulation demand signal range to construct a verification framework: first, based on the matching matrix, it transforms the source-load fluctuation difference into a regulation demand signal range; second, it uses the regulation demand signal range to guide energy storage participation in the medium- and long-term dispatch of the power market, solving for source-load matching degree and energy storage revenue as single objectives respectively; finally, using the maximum value of the single objective as the benchmark, it simulates the response behavior of energy storage under the guidance of the regulation demand signal range through the source-load-storage matching matrix, aiming to maximize the sum of the matching degree and the relative achievement rate of energy storage economics, thus obtaining the optimal energy storage configuration.

[0026] Specifically, such as Figure 2 As shown, the collaborative optimization method for energy storage regulation based on source-load fluctuation quantization includes the following steps S10-S30.

[0027] S10: Discretize the total load fluctuation range and total output fluctuation range of the power system, and construct a source-load fluctuation matching matrix to quantify the source-load fluctuation differences in each time period.

[0028] To achieve accurate quantification of source-load fluctuations, this embodiment employs a discrete fluctuation range division method based on cluster analysis. Considering that output fluctuations far exceed load fluctuations, to improve division accuracy, the Total Load Fluctuation Range (TLFR), which has a smaller fluctuation range, is selected for division, and then the Total Output Fluctuation Range (TOFR) is divided based on the division results.

[0029] In this embodiment, step S10 specifically includes the following steps S101-S107.

[0030] S101: Based on different numbers of clusters k The intra-group moment of inertia (IW) is used to establish the IW curve, and the inflection point of the IW curve is analyzed to determine the optimal number of clusters.

[0031] To determine the optimal number of fluctuation intervals, this embodiment introduces the Inertia Within (IW) method for cluster analysis. The IW value is calculated as follows: (1) In the formula: It is the number of clusters. i For cluster index, It is the first i A sample set of clusters, For any sample within the cluster, It is a cluster center. Represents sample points To its cluster center The square of the Euclidean distance. This is calculated by considering different numbers of clusters. k The IW value is calculated, and the inflection point of the IW curve is analyzed to determine the optimal number of clusters, i.e., the optimal number of interval divisions. k Optimal number of clusters k This corresponds to the inflection point on the IW curve. This inflection point is usually the point where the rate of decrease in the IW value changes significantly (e.g., from a sharp decrease to a gradual decrease), and on the graph, it resembles an "elbow" (see reference). Figure 19 By identifying this inflection point, we can avoid over-subdivision while ensuring the accuracy of the division, thereby determining the most reasonable number of interval divisions.

[0032] S102: Based on the optimal number of clusters, calculate the power variation amplitude and minimum energy block of a single wavelet interval.

[0033] To characterize the energy characteristics of power system source-load fluctuations, this embodiment introduces two quantitative indicators: one is the unit power difference within the fluctuation range. (MW) is used to measure power variation over a range; the second is the minimum energy block. (MWh) is used to characterize the energy level of different ranges of regulation.

[0034] (2) In the formula: The power variation amplitude (MW) is the power variation amplitude of a single fluctuation sub-interval. and These are the minimum and maximum values ​​(MW) of TLFR, respectively. The standard duration (h) for each time step is set to 1h in this paper to match the time-sharing trading period of the medium- and long-term electricity market; The minimum energy block (MWh) is used to represent the minimum energy required for a quantization transition between states.

[0035] S103: Calculate the load state based on the power change amplitude of a single fluctuation sub-interval. The calculation formula is as follows: (3) In the formula: For the firsti Each load condition for t Any value (MW) of the load fluctuation range at any given time. and These represent the minimum and maximum values ​​(MW) of the load fluctuation range.

[0036] S104: Based on the power change amplitude and load status of a single fluctuation sub-interval, the total output fluctuation interval is expanded and divided into a sequence of output status intervals.

[0037] To construct a unified source-load fluctuation quantization framework, the sub-interval width of TLFR is... Mapping to TOFR, and by covering the full fluctuation range of TOFR, a sequence of power fluctuation state intervals is constructed. : (4) In the formula: For the first j The output state range, in formulas involving algebraic operations, , They respectively refer to the midpoint of the corresponding interval; for t Total output (MW) at any given moment and These are the minimum and maximum values ​​(MW) of TOFR, respectively. n This indicates the number of subintervals extended between the minimum TOFR and the minimum TLFR. m This indicates the total number of subintervals that TOFR adds compared to TLFR.

[0038] S105: The first i The load state and the first j A combination of output state intervals is used to construct a discrete source-load full-fluctuation state combination. S ij .

[0039] In this embodiment, a state label (such as L, M, H, etc.) is assigned to each discrete interval to distinguish different states, and then the Cartesian product operation is used to convert each possible discrete state. and By combining these elements, a discrete source-load fully fluctuating state combination can be constructed. : (5) S106: Based on the first i Load status L i and the j Combination of output state intervals Q jCorresponding discrete level index and Calculate the source-load matching degree.

[0040] To achieve accurate quantification of each source-load discrete state combination, this embodiment first defines the load state. and output status The corresponding discrete level indices are respectively and Based on the hierarchical differences between the two, a source-load matching degree score function is constructed to calculate the source-load matching degree: (6) In the formula, It is a non-positive integer, and its value quantifies the degree of mismatch between the source and load fluctuation states: when Time: Indicates that the system is in a non-equilibrium state; the smaller the value, the greater the deviation between the source and load levels; when =0 (equilibrium state): indicates that the source and load states are matched.

[0041] S107: Establish a source-load fluctuation matching matrix based on source-load matching degree. M .

[0042] The calculated source-load matching degree Fill in the matching matrix Construct a source-load fluctuation matching matrix to comprehensively quantify the supply and demand matching characteristics of the system: (7) S20: Based on the source-load fluctuation matching matrix, establish the mapping relationship from the source-load fluctuation state to the system control demand, and generate the control demand signal range.

[0043] The forecasting uncertainty brought about by a high proportion of renewable energy output is one of the increasingly prominent factors in the current medium- and long-term electricity market operation. To address this issue, this embodiment proposes a method for constructing system regulation demand based on forecast curves. The core of this method consists of two steps: First, it constructs the fluctuation range of source-load power based on the power forecast curve, quantifying the forecast uncertainty; second, it maps the difference between the source-load fluctuation ranges to system regulation demand. This method expands the expected error range of the power forecast curve, establishing a link between fluctuation range changes and regulation demand changes, replacing the traditional fixed price range interval (FPRI). Simultaneously, energy storage systems with different capacity configurations have the ability to cover the difference between predicted and actual power within one or more fluctuation sub-intervals in any time period. This characteristic allows the system to use interval power forecasting instead of power curve forecasting when dealing with power fluctuations on medium- and long-term time scales. This not only simplifies the decision-making difficulty of energy storage on medium- and long-term scales but also transforms the reliance on accurate forecasting into the management of source-load interval states, thereby effectively smoothing source-load fluctuations in the medium- and long-term market and ensuring the long-term reliable operation of the system.

[0044] like Figure 3 As shown, this method is applied to the medium- and long-term energy decomposition piecewise linear curve (…). Figure 3 (b) and the power forecast curve ( Figure 3 Introducing system regulation requirements between (d) and (middle) Figure 3 (c)). By evaluating the source-load fluctuation state at each time period, combined with the matching matrix. It can quantify the source-load fluctuation differences in different time periods, alleviating the problem of the broken line in medium- and long-term electricity trading. Figure 3 (b) and the ten-day electricity trading curve ( Figure 3 (d) Figure 3 Middle (e) Figure 3 The connection between (f) and (m)).

[0045] Step S20 generates a system regulation demand signal range by quantifying the real-time source-load fluctuation difference, thereby guiding the independent energy storage system. This embodiment establishes a correlation mechanism between fluctuation states and regulation demand signals based on the precise quantification of source-load fluctuation differences. Addressing the applicability issue of the original source-load matching degree matrix in system regulation demand mapping, this embodiment improves the definition of the matching degree: when the load state is lower than the output state, i.e., the system faces pressure to absorb new energy, a specific mapping rule is used to convert the source-load matching degree into an excitation signal for a low regulation demand range, guiding the energy storage to charge and absorb excess power; conversely, when supply falls short of demand, a signal for a high regulation demand range is generated to incentivize energy storage to discharge.

[0046] Based on the above technical concept, in some embodiments, step S20 can be specifically implemented through the following steps S201-S204.

[0047] S201: Based on the source-load fluctuation matching matrix, determine the quantitative value of the regulation demand and construct the quantitative matrix of the regulation demand.

[0048] This embodiment constructs a regulation demand quantification matrix suitable for electricity price mapping. , is represented as: (8) (9) In the formula: To regulate the quantitative value of demand; A quantitative matrix for regulating demand; The correspondence between the numerical characteristics and the supply and demand states is as follows: Sur. (Surplus, surplus state): when... Indicates supply exceeding demand; Def. (Deficit, a state of shortage): when Indicates supply falling short of demand; Bal. (Balance, equilibrium state): when... This indicates that the system is in a supply-demand balance.

[0049] S202: Extract the quantitative values ​​of all control requirements in the system and construct a set of system control requirement matching degrees.

[0050] Based on the quantitative matrix of regulation demand Extract the quantitative values ​​of all control requirements in the system. The following set of system regulation demand matching degrees is constructed: (10) Where: the set of system regulation demand matching degree It includes all possible source-load matching degrees in the system; set The unique match score is obtained by deduplicating, sorting, and arranging all matches in Ξ, while defining the total number of unique match scores in the system. .

[0051] S203: Construct the control demand signal interval based on the sorting position of the control demand quantification value in the system control demand matching degree set.

[0052] by As the basis for dividing the overall system regulation demand range, this ensures that the same source-load state combination receives the same price incentive, achieving a precise match between energy storage revenue and system regulation demand. A mapping relationship is constructed from source-load fluctuation states to system regulation demand. Specifically, this embodiment uses a descending order indexing method. First, all possible... Sort the values ​​from largest to smallest: >0 corresponds to system surplus, is sorted at the top of the index, and is mapped to the low regulation demand range to guide energy storage charging. A value less than 0 corresponds to a system shortage, and its sorting index is lower, mapping it to a high regulation demand range to guide energy storage discharge. A regulation demand signal range is constructed. as follows: (11) In the formula: To regulate the demand signal range, and These are the quantitative values ​​for regulating demand. The corresponding upper and lower limits of the control demand signal sub-interval; To regulate the overall fluctuation range of the demand signal, a preset constant is used; This represents the step size for the signal level. Indicates the quantitative value of demand for regulation In the set The sorting order within.

[0053] S204: Based on the range of control demand signals, a control demand signal matrix is ​​formed.

[0054] Finally, the formula for regulating the demand signal matrix is ​​expressed as follows: (12) In the formula: express t Moment State Combination The corresponding range of regulatory demand signals, To regulate the demand signal matrix, the matrix dimension is... .

[0055] S30: Construct a dynamic programming-based energy storage scheduling model, using the demand signal range as the excitation signal, and taking the maximization of the sum of the normalized system source-load matching degree improvement and the relative achievement rate of the comprehensive benefit index of energy storage scheduling as the optimization objective. Perform collaborative optimization to solve the problem and output the optimal energy storage scheduling strategy corresponding to the optimization objective.

[0056] Dynamic programming is an effective framework for solving multi-stage stochastic control problems. The dynamic programming model constructed in this embodiment includes a state space, decision variables, state transition equations, constraints, objective function and stage rewards, as well as the Bellman recurrence equation.

[0057] First, the mapping relationship between energy storage capacity parameters and fluctuation ranges is clarified. That is, the capacity configuration of the energy storage system is based on the source-load fluctuation range and is independent of the total system load or total installed capacity. According to the definition of the fluctuation range in formula (2), different energy storage capacity configurations can be used... The format is as follows: The maximum charge and discharge limit configuration of energy storage is expressed as (Limited Power Capacity Configuration): Indicates standard power. of x Times; Energy storage capacity configuration (CapacityConfiguration): Represents an energy block. (MWh) times y: (13) In the formula: The rated capacity (MWh) of the energy storage system. This represents the maximum charge / discharge power (MW) of the energy storage.

[0058] Based on the above mapping relationship, the adjustable state set Θ of the energy storage system can be established as follows: (14) In the formula: For a source-load full-wave state combination, and They are respectively t Real-time load and output status values. (Set) The source-load fluctuation range that an energy storage system can fully cover is defined: when the source-load fluctuation difference lies within the set Θ, the energy storage system can... t The system can achieve a balanced state at any time.

[0059] After independent energy storage participates in system regulation, it not only regulates both the source and load sides but also increases the number of adjustable ranges on both sides. The source-load-storage fluctuation matrix and the source-load-storage demand matrix are shown below: (15) (16) In the formula: The expanded matrix dimension represents the number of adjustable ranges added on both the source and load sides after energy storage participation. Increase; It is the range adjustment coefficient. This represents the upper limit of the energy storage's single charge / discharge power. The system's source-load combined state will transition based on decisions, and then according to... and The changed matching degree and the range of control demand signals are obtained.

[0060] The state space of a dynamic programming scheduling model describes the state of the system at each decision moment. First, a two-dimensional time-series state space is defined. :in This represents the discrete time step within the optimization cycle, where T is the total number of time steps in the entire optimization cycle. This represents the amount of energy stored at time t. Based on discretized representation, the basic unit is the smallest energy block. Its value range is .

[0061] The decision variable is for each time step. t The decision-making process determines the energy storage operation based on the current state. First, a three-dimensional decision variable is defined. :in Indicates the type of energy storage operation. These represent charging, discharging, and inactivity, respectively. Indicates the regulation target of energy storage. These represent the effects of energy storage regulation on the source-side state or the load-side state, respectively, which directly affect the combined state of the system after the state transition and its corresponding matching score and dynamic electricity price. This is the power regulation level, with a value range of [value range missing]. Used to characterize the energy storage charging and discharging power relative to a reference power. The multiplier, of which x This is the maximum adjustment factor.

[0062] Based on decision variables, t Actual charge and discharge power of energy storage at all times P ( t This can be represented as: (17) The state transition equations and constraints are as follows: Status of energy storage system SOC ( t The transition is determined by the current state and the decisions made: (18) In the formula: It is the state at the next moment calculated based on the state transition equation.

[0063] Meanwhile, the operation of energy storage systems must meet the following constraints: (19) Based on the source-load fluctuation matching matrix constructed above, the absolute value of the matching score This directly reflects the degree of imbalance between supply and demand in the system: the larger the absolute value, the more extreme the state of the system, and the higher the risk of power grid operation. Therefore, the primary goal of collaborative optimization is to reduce the magnitude of these extreme states as much as possible.

[0064] (20) In the formula: It represents the total cumulative improvement in source-load matching throughout the entire scheduling cycle; yes t The improvement in matching degree at any given time; and for t The degree of matching before and after energy storage participation.

[0065] The overall goal is to maximize the net revenue obtained by energy storage through dynamic electricity price arbitrage throughout the entire dispatch cycle. Under the influence of system regulation demands, electricity price ranges are highly coupled with source-load matching: the more extreme the system state, the greater the price difference within the range. Regulation demand signals guide energy storage to respond to system demand during charge-discharge arbitrage, thereby maximizing the economic value of energy storage while mitigating fluctuations.

[0066] (twenty one) In the formula: The maximum total revenue (in yuan) from energy storage within a given time period. for t Instant earnings per moment (in yuan), for t The median value (yuan / MWh) of the quantitative range of demand is adjusted at all times.

[0067] After solving the problem separately with the goal of maximizing matching degree / profit, the maximum improvement in matching degree is obtained. Total revenue Then, using the maximum value of both as the base value, normalization is performed, with the goal of maximizing the sum of the relative achievement rates of the two objectives. .

[0068] (twenty two) In the formula: and These represent the relative achievement rates (%) of source-load matching degree and returns over the cumulative time period after energy storage participation. This represents the sum of relative achievement rates (%).

[0069] Dynamic programming uses the Bellman equation for backward recursion. This embodiment constructs the value function V( t SOC( t ))express t Always in state SOC ( t The optimal cumulative return obtained to reach the planning endpoint T: (twenty three) In the formula: for t The instantaneous revenue function. Its specific form depends on the optimization objective.

[0070] When the source-load matching degree is the objective, When the goal is energy storage revenue: In collaborative optimization: the direct goal of optimization is to maximize the sum of the relative achievement rates of source-load matching degree and energy storage benefits. (twenty four) The boundary condition is set to zero for the state value at the end of the planning period: (25) from t =T-1 Reverse recursion to... t =0, the optimal value function for all states at all time steps can be calculated. Then, by forward tracing, the optimal decision sequence starting from the initial state SOC(0) is found. D (0), D (1)... D (T-1).

[0071] The feasibility and progressiveness of the method in this application will be explained in detail below through specific calculation examples.

[0072] This embodiment constructs a computational example based on monthly data from a typical power grid region with high renewable energy penetration. Information such as the installed power generation capacity and load characteristics of this region's power grid is shown in Table 1. Through IW cluster analysis of historical load data, the optimal clustering inflection point is determined to be 3 (…). Figure 19 Based on this, the total load fluctuation range is divided into [ Low, Mid, High There are a total of 3 load sub-intervals. Based on the results of the total load fluctuation interval division, the total output fluctuation interval is further expanded and divided to obtain... Shortage 7 to Surplus There are 18 output sub-intervals in total, and the specific results are shown in Table 2. This embodiment selects a fixed electricity price interval, as shown in Table 3. Data analysis shows that the initial source-load total matching degree of the power grid in this region is -2613 in a typical month, and the system's average monthly curtailment rate is 5.39%. The urgent need for system regulation provides an ideal application scenario for verifying the method of this application.

[0073] Table 1. Statistics on Installed Capacity and Annual Power Generation of Regional Power Systems

[0074] Table 2 Source-Load State Interval Division Table

[0075] Table 3 Fixed Electricity Price Range

[0076] New energy storage technologies can be categorized into three types based on market participants: load-side energy storage, output-side energy storage, and grid-side independent energy storage. Given that the source-side fluctuation range is larger than the load-side fluctuation range, to analyze the differences in the guiding effect of the proposed method on different types of energy storage, this application compares grid-side independent energy storage and output-side energy storage, both with similar regulation ranges, under the same typical day. The energy storage configuration is uniformly set as follows: L 1 C 4 (Energy storage charging and discharging power limit is) Energy storage capacity is 4 times ).

[0077] based on Figure 4 The comparative analysis revealed two significant limitations of source-side energy storage regulation: first, the matching degree curve exhibits continuous oscillation characteristics, especially with frequent fluctuations in the 0-5 and 18-23 time periods; second, there is a local "reverse regulation" phenomenon, for example, after continuous discharge in the 10-14 time period, it is forced to switch to charging, which worsens the source-load matching degree. The root cause of the above problems lies in the inherent constraint of single-side energy storage on continuous regulation capability, which leads to overcompensation behavior in order to meet constraints when optimizing within a limited solution space.

[0078] In contrast, independent energy storage effectively solves the above problems through a collaborative optimization mechanism on both the source and load sides. For example... Figure 5 As shown, the broad optimization space brought about by dual-sided regulation enables energy storage to maintain a stable improvement in the source-load matching degree, which not only fundamentally avoids the oscillation risk associated with unilateral regulation, but also achieves the continuous and maximization of energy storage regulation benefits.

[0079] Table 4 Comparison of regulation capabilities between source-side energy storage and independent energy storage

[0080] Based on the quantitative indicators in Table 4, the method proposed in this application is more suitable for independent energy storage mode: 1) The source-load dual-side regulation mechanism provides a larger optimization space and realizes more efficient energy allocation. The multi-dimensional collaborative optimization naturally suppresses the oscillation problem that may occur in single-dimensional regulation; 2) Data shows that the daily average matching degree of independent energy storage is 23.49% higher than that of power source-side energy storage, and the energy storage utilization rate is 7.58% higher, which verifies the high adaptability of this method to independent energy storage participating in market transactions.

[0081] To comprehensively and quantitatively evaluate the guiding effects of different electricity pricing mechanisms, this embodiment constructs a five-dimensional evaluation index system called "MPRCT," which includes source-load matching degree, energy storage benefit, rate of curtailment, capacity utilization, and time utilization. This embodiment will analyze the differences in guiding effects of the proposed method and the fixed electricity price range (FPRI) on different energy storage configurations.

[0082] This embodiment is based on the quantized reference power and reference capacity. =1139.35MW, =1139.35 MWh), constructing differentiated energy storage configuration schemes: a total of 8 charging and discharging power limits are set ( L 1- L 8, covering a range of 1.1-9.1GW) and 6 types of energy storage installed capacity ( C 3- C 8. A combined configuration scheme covering the range of 3.4-9.1 GWh, considering that in practical applications the installed energy storage capacity must be greater than or equal to the upper limit of charge and discharge power (i.e. L ≤ C Under the constraints of ), 33 effective configuration combinations were finally selected.

[0083] like Figure 6 and Figure 7 As shown, when maximizing source-load matching is the optimization objective, multiple technical indicators exhibit a high degree of consistency under the guidance of system regulation requirements and fixed electricity prices. Specifically, both achieved the maximum improvement in source-load matching from -2615.00 to -881.00, and the system curtailment rate decreased from 5.39% to 1.7%. Simultaneously, the system capacity utilization rate remained stably maintained within the range of 57%-93% at different charge / discharge rates.

[0084] However, the two mechanisms show significant differences in terms of the economic feasibility of energy storage: by Figure 7 It is evident that, guided by DPRI, energy storage revenue maintains a growth trend as capacity configuration increases, and... L 8 C The maximum profit reached 60.804 million yuan with configuration 8; on the contrary... Figure 6 Under the guidance of FPRI, the profitability of energy storage deteriorates rapidly with the increase in capacity configuration, ultimately reaching its peak. L 3 CThe maximum loss of RMB 26.775 million occurred under configuration 8, indicating that FPRI cannot guarantee the economic viability of energy storage. The results fully demonstrate the mechanism advantage of the proposed method in guiding energy storage to play a system regulation role while ensuring its economic viability.

[0085] like Figure 8 and Figure 9 Quantitative results show that, C 3- C Within the configuration range of 8, the benefit advantage of this application method compared to FPRI remains between 17.4% and 22.4%, and the benefit advantage of this application method is further enhanced when configured as follows: L 8 C The maximum return of RMB 111.759 million was reached at 8:00 AM, demonstrating that the proposed method can continuously provide high economic returns at all stages of energy storage configuration.

[0086] Depend on Figure 10 As can be seen, the method in this application also shows significant advantages in terms of time utilization: First, the maximum improvement of energy storage to source-load matching degree under the guidance of system regulation demand reaches 462, while FPRI leads to a decrease of 378 for energy storage; Second, in terms of system utilization efficiency, the time utilization rate of system regulation demand reaches 71.10%, and the capacity dimension remains in a stable range of 54%-59%, while FPRI, although achieving a time utilization rate of 66.67%, fluctuates between 41%-62% in terms of capacity dimension; In terms of curtailment rate, the method in this application achieves the best improvement rate of 1.87%, while energy storage dispatch under the guidance of FPRI leads to a reverse increase in curtailment rate, with a maximum curtailment rate of 6.76%.

[0087] In summary, the multidimensional evaluation based on the MPRCT index system demonstrates that the proposed method exhibits excellent mechanism robustness. Compared with the capacity fixed price range (FPRI), the proposed method can maintain a highly consistent guiding effect under different energy storage configurations and optimization objectives through flexible demand signals, effectively overcoming the regulation oscillation and failure risks under traditional mechanisms, and proving its broad adaptability in the electricity market environment.

[0088] Based on the above demonstration of the advantages of guiding energy storage to participate in the electricity market in accordance with system regulation and control requirements, this embodiment will analyze the guiding effect of the dynamic electricity price mechanism on different energy storage capacity configurations by comparing the five technical indicators of MPRCT.

[0089] This embodiment will continue to analyze 33 energy storage configuration schemes ( L 1 C 3- L 8 C 8) A comparative analysis of MPRCT indicators was conducted to analyze the guiding effect of the DPRI mechanism on different configurations under the guidance of energy storage economics: Figures 11 to 13This study reveals the differentiated guiding effect of the proposed method on different energy storage configurations. The results show that, under fixed installed capacity conditions, the energy storage revenue exhibits a monotonically increasing trend as the upper limit of a single charge-discharge cycle increases. In terms of system regulation effect, both the curtailment rate and the matching degree show a trend of first increasing and then decreasing. Regarding system utilization efficiency, the capacity utilization rate fluctuates less, but the time utilization rate shows a significant downward trend.

[0090] Based on comprehensive multi-dimensional indicators, configuration L 2 C 8. Under the guidance of the dynamic electricity pricing mechanism, it exhibits the best overall balance of effects: among which, configuration L 2 C Project 8 generated a revenue of 89.297 million yuan. While this represents only 83.97% of the optimal revenue, it is still 20.8% higher than the average revenue. Furthermore, this configuration excelled in two key technical indicators: it not only achieved maximum matching accuracy but also improved the curtailment rate by 42.3%, reaching the optimal level. Figure 13 This indicates that its utilization rates in both the time dimension (52.28%) and the capacity dimension (57.75%) are significantly higher than the average level (45%). Therefore, configuration L 2 C 8 achieves an optimal balance between economic and technical indicators, and is the optimal solution based on demand-driven demand under the goal of energy storage revenue.

[0091] Depend on Figures 14 to 16 It can be seen that, with a fixed installed capacity C Under the premise of increasing the upper limit of single charge and discharge power. L While it can significantly drive the synchronous growth of source-load matching and energy storage revenue, it also faces the problem of diminishing marginal efficiency of system utilization, that is, the improvement of curtailment rate gradually approaches saturation, while time and capacity utilization show a downward trend.

[0092] Based on comprehensive multi-dimensional indicators, configuration L 6 C 8 demonstrates the best guiding effect of the mechanism. First, from the perspective of economic incentives, L 6 C Configuration 8 achieved the maximum benefit of 61.159 million yuan, establishing its market incentive advantage; secondly, the source-load matching degree improvement reached 1,729, ranking second among all configuration schemes, but it already demonstrated a significant adjustment effect; in terms of energy utilization efficiency, L 6 C Configuration 8 significantly improved the system's curtailment rate, reducing it dramatically from 5.39% to 1.71%, a performance essentially equivalent to the optimal configuration. Furthermore, this configuration achieved utilization rates of 38.73% in the time dimension and 58.28% in the capacity dimension, stabilizing the overall system utilization rate at a reasonable level of 48.51%. In summary, L 6C The configuration 8 is the globally optimal configuration scheme determined by the method of this application, which takes into account both system adjustment effect and economic benefits.

[0093] The examples above are all based on optimization objectives for a single indicator, but the synergistic optimization of source-load matching degree and energy storage revenue is more practically significant. Research has found a significant difference between the matching degree and revenue indicators, not only in terms of magnitude but also in their fundamental differences in physical meaning and computational units. This makes it difficult to directly apply traditional normalization methods and multi-objective optimization methods.

[0094] To address this issue, this embodiment proposes a relative achievement rate evaluation system. Based on the maximum improvement in matching degree and the maximum benefit obtained under different single objectives, the system solves for the sum of the maximum achievement rate of matching degree and energy storage benefit, thereby converting optimization objectives of different dimensions into a unified relative index and establishing a quantitative evaluation framework for collaborative optimization.

[0095] like Figure 17 As shown, the collaborative optimization strategy achieves this by dynamically balancing the coupling relationship between source-load matching degree and economic benefits. L 2 C The 8-configuration system achieved a comprehensive improvement in system performance. Regarding source-load matching, the optimized system improved the matching by 1,202, reaching 95.40% of the optimal value, significantly outperforming single-objective optimization strategies. In terms of economic benefits, energy storage revenue reached 81.331 million yuan, achieving a relative success rate of 91.08%, a 96% improvement compared to single-matching strategies. Even more significant is the synergistic effect of the system's technical indicators: time utilization exceeded 162.90%, while maintaining a stable capacity utilization rate of 57.78%, maximizing charging and discharging efficiency while ensuring the safe operation of the energy storage system. Simultaneously, the system reduced the average curtailment rate from 5.39% to 2.33%, significantly improving the absorption capacity of renewable energy.

[0096] like Figure 18 As shown, through configuration L 6 CAn analysis of the optimization strategy under option 8 reveals that the dual-objective optimization method demonstrates superior system performance. In terms of economic benefits, this strategy achieved a significant energy storage revenue of 88.086 million yuan, reaching 84.02% of the optimal value, a 44.02% improvement compared to the single-matching-degree scheme. At the source-load matching degree level, the optimized system achieved a matching degree improvement of 1,602, 13.35 times higher than the single-revenue-oriented approach. Particularly significant is the synergistic breakthrough in system operation indicators: time utilization reached 220.70%, more than 5.7 times higher than the other two schemes, while maintaining a stable capacity utilization rate of 55.66%, achieving a balance between system scheduling flexibility and equipment safety. Simultaneously, the system curtailment rate improved by 3.54%, matching the matching degree target and significantly higher than the single-revenue-oriented approach.

[0097] In summary, this embodiment, from the perspective of grid-side mechanism design, addresses the problem of aggravated source-load fluctuations in new power systems. Through numerical examples, it verifies the rationality and effectiveness of the proposed method, aiming to provide a theoretical basis for guiding independent energy storage to efficiently participate in medium- and long-term market dispatch on the grid side, achieving synergistic optimization of both source-load matching and energy storage revenue objectives. Through systematic theoretical analysis and numerical verification, the conclusions are as follows: 1) A method for accurately quantifying source-load fluctuation differences was constructed. This method first divides the fluctuation ranges of total load and total output, constructs a source-load fluctuation matching matrix, and achieves accurate quantification of source-load fluctuation differences in each time period. This provides a reliable theoretical basis for the subsequent construction of an electricity price range mechanism.

[0098] 2) A dynamic demand mechanism based on source-load fluctuation characteristics was designed. A mapping from source-load fluctuation state to regulation demand range was established, transforming the quantified source-load fluctuation differences into flexible regulation signals, laying the foundation for guiding energy storage to participate in the electricity market in the future.

[0099] 3) The advantages of the dynamic demand mechanism in guiding independent energy storage were verified. Comparative studies showed that, compared with output-side energy storage, independent energy storage has stronger system regulation capabilities and greater optimization potential under the guidance of the dynamic demand mechanism.

[0100] 4) This demonstrates that the dynamic demand mechanism has significant advantages over the fixed electricity price mechanism. The numerical example shows the advantages of energy storage configuration. L 2 C Under 8 conditions, the method in this application improves the system source-load matching degree by 66.3%; in terms of economic benefits, the DPRI scheme under the L8C8 energy storage configuration improves by 22.4% compared with FPRI.

[0101] 5) A mechanism evaluation method based on relative achievement rate was constructed, and the effectiveness and rationality of the proposed method were demonstrated through numerical examples. In the case of energy storage configuration as... L 2 CAt 8:00 AM, the collaborative optimization strategy achieved a relative matching rate of 95.4% and a relative benefit rate of 91.1%, with a total achievement rate of 186.5%. (The last sentence appears to be incomplete and possibly refers to an energy storage configuration.) L 6 C At 8:00, the relative achievement rates of matching degree and return reached 92.7% and 84.0% respectively, with a total achievement rate of 176.7%. The calculation results fully demonstrate that the method proposed in this application can effectively guide energy storage to play a system regulation role while ensuring reasonable returns for energy storage.

[0102] This application also provides a collaborative optimization system for energy storage regulation based on source-load fluctuation quantization, used to implement the method described in any of the above embodiments, such as... Figure 20 As shown, the system includes: The source-load fluctuation quantification module 201 is configured to discretize the total load fluctuation range and the total output fluctuation range of the power system, and construct a source-load fluctuation matching matrix to quantify the source-load fluctuation differences in each time period. The regulation demand determination module 202 is configured to establish a mapping relationship from the source-load fluctuation state to the system regulation demand based on the source-load fluctuation matching matrix, and generate a regulation demand signal range. The dynamic programming module 203 is configured to construct an energy storage scheduling model based on dynamic programming. It uses the regulation demand signal range as the excitation signal, takes maximizing the sum of the normalized system source-load matching degree improvement and the relative achievement rate of the comprehensive benefit index of energy storage scheduling as the optimization objective, performs collaborative optimization, and outputs the optimal energy storage scheduling strategy corresponding to the optimization objective.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A collaborative optimization method for energy storage regulation based on source-load fluctuation quantization, characterized in that, include: The total load fluctuation range and total output fluctuation range of the power system are discretized and a source-load fluctuation matching matrix is ​​constructed to quantify the source-load fluctuation differences in each time period. Based on the source-load fluctuation matching matrix, a mapping relationship from the source-load fluctuation state to the system control requirements is established, and a control requirements signal range is generated. A dynamic programming-based energy storage scheduling model is constructed, using the aforementioned regulation demand signal range as the excitation signal. The optimization objective is to maximize the sum of the normalized system source-load matching degree improvement and the relative achievement rate of the comprehensive benefit index of energy storage scheduling. The model is then used for collaborative optimization and the optimal energy storage scheduling strategy corresponding to the optimization objective is output.

2. The collaborative optimization method for energy storage regulation based on source-load fluctuation quantization according to claim 1, characterized in that, The total load fluctuation range and total output fluctuation range of the power system are discretized, and a source-load fluctuation matching matrix is ​​constructed, including: Based on different numbers of clusters k The inertial moment IW within each group is used to establish the IW curve, and the inflection point of the IW curve is analyzed to determine the optimal number of clusters. Based on the optimal number of clusters, calculate the power variation amplitude and minimum energy block of a single fluctuation sub-interval; The load status is calculated based on the power change amplitude of the single fluctuation sub-interval; Based on the power change amplitude and load status of a single fluctuation sub-interval, the total output fluctuation interval is expanded and divided into a sequence of output status intervals. The first i The load state and the first j A combination of output state intervals is used to construct a discrete source-load full-fluctuation state combination. S ij ; Based on the i Load status L i and the j Combination of output state intervals Q j Corresponding discrete level index and Calculate the source-load matching degree; Based on the source-load matching degree, a source-load fluctuation matching matrix is ​​established. M .

3. The collaborative optimization method for energy storage regulation based on source-load fluctuation quantization according to claim 2, characterized in that, The formula for calculating the in-group moment of inertia IW is: （1） In the formula: k It is the number of clusters. i For cluster index, It is the first i A sample set of clusters, For any sample within the cluster, It is a cluster center. Represents sample points To its cluster center The square of the Euclidean distance; Based on the optimal number of clusters, the power variation amplitude and minimum energy block of a single fluctuation sub-interval are calculated using the following formula: （2） In the formula: The power variation amplitude of a single fluctuation sub-interval. and These are the minimum and maximum values ​​of TLFR, respectively. The standard duration for each time step; This is the minimum energy block, used to represent the minimum energy required for a quantized state transition; k To determine the optimal number of clusters; Based on the power variation amplitude of the single fluctuation sub-interval, the load state is calculated using the following formula: （3） In the formula: L i For the first i Each load condition for t Any value within the load fluctuation range at any given time. and These represent the minimum and maximum values ​​within the load fluctuation range; Based on the power change amplitude and load status of a single fluctuation sub-interval, the total output fluctuation interval is divided into a sequence of output status intervals using the following formula: （4） In the formula: Q j For the first j One output state range; for t Total output at all times and These are the minimum and maximum values ​​of TOFR, respectively. n This indicates the number of subintervals extended between the minimum TOFR and the minimum TLFR. m This indicates the total number of subintervals added by TOFR compared to TLFR; Q 1 and Q n This indicates the first and second points in the total output fluctuation range that expand towards the minimum value side. n One output state range, Q n+1 , With the first and the second k The output state interval is matched to the load state interval; and These represent the first and second extensions of TOFR towards the maximum value side, respectively. k One output state range; L 1 and They represent the 1st and the 2nd respectively. k One load condition interval; The discrete source-load full-wave state combination S ij Represented as: （5） In the formula: L It is a set of load state intervals; Q It is a set of power output state intervals; Based on the i Load status L i and the j Combination of output state intervals Q j Corresponding discrete level index and The source-load matching degree is calculated using the following formula: （6） In the formula: m i,j The source-load matching degree quantifies the degree of mismatch between the source and load fluctuation states: when m i,j When the value is ≠0, it indicates that the system is in a non-equilibrium state. The smaller the value, the greater the deviation between the source and load levels. when m i,j When =0, it indicates that the source and load states are matched; The source-load fluctuation matching matrix M Represented as: （7） In the formula: [ Indicated by m i,j The dimension formed by the elements is k OK m + k A matrix of columns.

4. The collaborative optimization method for energy storage regulation based on source-load fluctuation quantization according to claim 3, characterized in that, Based on the source-load fluctuation matching matrix, a mapping relationship from the source-load fluctuation state to the system control demand is established, generating a control demand signal range, including: Based on the source-load fluctuation matching matrix, the quantitative value of the regulation demand is determined, and the quantitative matrix of the regulation demand is constructed. Extract the quantitative values ​​of all control requirements in the system and construct a set of system control requirement matching degrees; Based on the sorting position of the quantitative values ​​of the control demand in the system control demand matching degree set, a control demand signal interval is constructed. Based on the aforementioned control demand signal range, a control demand signal matrix is ​​formed.

5. The collaborative optimization method for energy storage regulation based on source-load fluctuation quantization according to claim 4, characterized in that, The quantitative value of the regulation demand is determined by the following formula: （8） In the formula: To regulate the quantitative value of demand; The correspondence between the numerical characteristics and the supply and demand status is as follows: When This indicates that supply exceeds demand; when This indicates that supply cannot meet demand; when This indicates that the system's supply and demand are in balance; The quantitative matrix of regulation demand is represented as follows: （9） In the formula, To regulate the demand quantification matrix, Indicated by The dimension formed by the elements is k OK m + k A matrix of columns; The system regulation demand matching degree set is represented as follows: （10） In the formula: Ξ represents the set of matching degrees between system regulation and demand; This is the set obtained by removing duplicates, sorting in descending order, and arranging all matching degrees in Ξ. st represents the constraint condition; , and It is the unique matching degree value after deduplication and ascending order of Ξ; The control demand signal range is represented as follows: （11） In the formula: To regulate the demand signal range, and These are the quantitative values ​​for regulating demand. The corresponding upper and lower limits of the control demand signal sub-interval; To regulate the overall fluctuation range of the demand signal, a preset constant is used; This represents the step size for the signal level. Indicates the quantitative value of demand for regulation In the set The sorting order; This represents the total number of unique matching values ​​in the system.

6. The collaborative optimization method for energy storage regulation based on source-load fluctuation quantization according to claim 4, characterized in that, The regulation demand signal matrix is ​​represented as follows: （12） In the formula: express t Moment State Combination S ij The corresponding range of regulatory demand signals, To regulate the demand signal matrix; To indicate with The dimension formed by the elements is k OK m + k A matrix of columns.

7. The collaborative optimization method for energy storage regulation based on source-load fluctuation quantization according to claim 1, characterized in that, A dynamic programming-based energy storage scheduling model is constructed as follows: The mapping relationship between energy storage capacity parameters and fluctuation range is established, expressed as: （13） In the formula: L x Configured to the maximum charge and discharge limit of energy storage. P max For maximum charge and discharge power of energy storage, x The power variation amplitude of a single oscillating sub-interval Multiples of, C y Configured for energy storage capacity, SOC max The rated capacity of the energy storage system, y Minimum energy block Multiples of; Based on the mapping relationship between the energy storage capacity parameters and the fluctuation range, the adjustable state set Θ of the energy storage system is established, represented as: （14） In the formula: S ij For a source-load full-wave state combination, K L (t) and K Q ( t ) are respectively t Real-time load and output status values; Determine the state space and decision variables of the energy storage scheduling model; wherein, the state space is represented as... S ( t , SOC ( t )), SOC ( t )for t At all times, with the smallest energy block The unit of energy storage capacity t ∈{0,1,..., T -1} represents the discrete time step within the optimization cycle. T It is the total number of time steps in the entire optimization cycle; the decision variable is expressed as D ( t )=( a ( t ), b ( t ), u ( t )), a ( t () indicates the energy storage operation type. a ( t )∈{ Charge , DisCharge , None }, Charge , DisCharge and None These represent charging, discharging, and no operation, respectively. b ( t () indicates the regulation target of energy storage. b ( t )∈{ Output , Load }, Output This indicates that energy storage regulation is applied to the source-side state. Load This indicates that energy storage regulation is applied to the load-side state. u ( t () represents the power regulation level. u ( t )∈{1,2,..., x }; Based on the decision variables, determine t The actual charge and discharge power of the stored energy at any given time is expressed as: （17） In the formula: P ( t )for t The actual charging and discharging power of the stored energy at all times; The state transition equations and constraints of the energy storage scheduling model are defined as follows: （18） （19） In the formula: SOC ( t+ 1) is the state at the next moment calculated based on the state transition equation.

8. The collaborative optimization method for energy storage regulation based on source-load fluctuation quantization according to claim 7, characterized in that, Using the aforementioned demand signal range as the excitation signal, and with the optimization objective of maximizing the sum of the normalized system source-load matching improvement and the relative achievement rate of the comprehensive benefit index of energy storage scheduling, a collaborative optimization solution is performed, and the optimal energy storage scheduling strategy corresponding to the optimization objective is output, including: After introducing energy storage, an extended source-load-storage fluctuation matrix is ​​constructed. M E Demand signal matrix for source-load-storage regulation E , respectively represented as: （15） （16） In the formula: The expanded matrix dimension represents the number of adjustable ranges added on both the source and load sides after energy storage participation. Increase; It is the interval adjustment coefficient. P max This refers to the upper limit of the single charge and discharge power of energy storage. b max This is the maximum number of energy storage regulating blocks; Indicated by The dimension formed by the elements is OK A matrix of columns; Indicated by The dimension formed by the elements is OK A matrix of columns; Calculate the total improvement in source-load matching degree and the maximum value of the comprehensive benefit index of energy storage dispatch; the formula for calculating the total improvement in source-load matching degree is: （20） In the formula: To improve the overall source-load matching degree; M ( t )for t Time-of-fact matching improvement value; m i,j ( t )and Let t be the degree of matching before and after energy storage participation; The formula for calculating the maximum value of the comprehensive benefit index of energy storage dispatch is as follows: （21） In the formula: To maximize the total revenue from energy storage within a given time period. E ( t ( ) represents the immediate profit (in yuan) at time t. for t The median of the quantitative range of demand should be adjusted at all times; The sum of the relative achievement rates is calculated by normalizing the total improvement in the source-load matching degree and the maximum value of the comprehensive benefit index of energy storage dispatch as the benchmark value. The formula for calculating the sum of the relative achievement rates is: （22） In the formula: R m and These represent the relative achievement rates of source-load matching degree and returns over the cumulative time period after energy storage participation. This represents the sum of relative achievement rates; Using the sum of relative achievement rates as the optimization objective, a dynamic programming method is employed to solve the problem. Through backward recursion and forward tracing, the optimal decision sequence that maximizes the sum of relative achievement rates is obtained. D (0), D (1)... D (T-1)} is the optimal energy storage scheduling strategy.

9. The collaborative optimization method for energy storage regulation based on source-load fluctuation quantization according to claim 8, characterized in that, When using dynamic programming to solve the problem, the Bellman equation is used for backward recursion to construct the value function. V ( t SOC) represents t Always in state SOC ( t The optimal cumulative return obtained to the planning endpoint: (23) In the formula: Let t be the immediate return function at time t; When the source-load matching degree is the objective, ; When the goal is energy storage revenue: In collaborative optimization, the direct goal is to maximize the sum of the relative achievement rates of source-load matching degree and energy storage benefits. (24) The boundary condition is set to zero at the end of the planning period; from t =T-1 Reverse recursion to... t =0, calculate the optimal value function of all states at all time steps, and find the optimal decision sequence starting from the initial state SOC(0) by forward tracing.

10. A collaborative optimization system for energy storage regulation based on source-load fluctuation quantization, used to implement the method described in any one of claims 1 to 9, characterized in that, The system includes: The source-load fluctuation quantification module is configured to discretize the total load fluctuation range and the total output fluctuation range of the power system, and construct a source-load fluctuation matching matrix to quantify the source-load fluctuation differences in each time period. The regulation demand determination module is configured to establish a mapping relationship from the source-load fluctuation state to the system regulation demand based on the source-load fluctuation matching matrix, and generate a regulation demand signal range. The dynamic programming module is configured to construct an energy storage scheduling model based on dynamic programming. It uses the regulation demand signal range as the excitation signal, and takes maximizing the sum of the normalized system source-load matching degree improvement and the relative achievement rate of the comprehensive benefit index of energy storage scheduling as the optimization objective. It performs collaborative optimization and outputs the optimal energy storage scheduling strategy corresponding to the optimization objective.

Images