A method for determining a multi-service coordinated operation strategy of a liquid flow battery energy storage power station

Abstract

The application provides a determination method of a multiple service coordinated operation strategy of a liquid flow battery energy storage power station, and belongs to the field of power station operation. The method comprises the following steps: clustering energy storage fluctuation flattening power curves to obtain energy storage fluctuation flattening power demand and capacity demand; obtaining secondary frequency modulation standby power according to historical frequency modulation signal total values, and obtaining energy storage participation secondary frequency modulation auxiliary service income; calculating energy storage peak shaving capacity demand and energy storage participation peak shaving auxiliary service income; setting energy storage primary frequency modulation droop coefficients by using energy storage overload capacity, and setting energy storage operation power and energy storage state of charge intervals according to the calculation results. The application considers primary frequency modulation, secondary frequency modulation and fluctuation flattening wind power functions on the basis of executing a single peak shaving working condition, and compared with the multiple service operation strategy of the current energy storage power station, the fluctuation flattening wind power function is considered, so that the liquid flow energy storage power station has better economy.

Description

Technical Field

[0001] This invention belongs to the field of power plant operation, and more specifically, relates to a method for determining a multi-service coordinated operation strategy for a flow battery energy storage power plant. Background Technology

[0002] The output power of a flow battery energy storage system is determined by the size and number of fuel cells, while the storage capacity is determined by the volume of the electrolyte. Increasing output power can be achieved by increasing the electrode area and the number of fuel cells; increasing storage capacity can be achieved by increasing the volume of the electrolyte. Vanadium redox flow batteries, in addition to their high safety and long service life, also offer advantages such as high cost-effectiveness, economic efficiency, and low environmental impact. Therefore, vanadium redox flow batteries are the optimal choice for high-power, high-capacity energy storage technology. Currently, flow battery energy storage power stations are generally used in single, long-term peak-shaving scenarios. The decoupling of power and capacity in flow battery energy storage power stations makes them more flexible in configuration compared to lithium battery energy storage power stations, enabling them to meet various ancillary service functions. If a flow battery energy storage power station operates under a single condition, its potential value cannot be fully realized. A multi-service operation strategy for power-type flow battery energy storage power stations is needed to improve their economic efficiency.

[0003] There is a lack of research on multi-ancillary service operation strategies for flow battery energy storage power stations. Most studies involving multi-service operation focus on lithium battery energy storage power stations, and these studies are generally limited to peak shaving, frequency regulation mode switching, or coordinated operation. However, because flow battery energy storage power stations have a longer deployment time than lithium battery energy storage power stations, the constraints on the state of charge (SOC) of flow battery energy storage power stations during multi-service operation are less stringent. Therefore, flow battery energy storage power stations can simultaneously support more ancillary services and meet more operating conditions compared to lithium battery energy storage power stations. Summary of the Invention

[0004] In view of the shortcomings of the prior art, the purpose of this invention is to provide a method for determining the multi-service coordinated operation strategy of a flow battery energy storage power station, which aims to solve the problem that existing flow battery energy storage power stations only operate under a single peak-shaving condition, resulting in a waste of power and capacity of the energy storage power station.

[0005] To achieve the above objectives, the present invention provides a method for determining a multi-service coordinated operation strategy for a flow battery energy storage power station, the method comprising the following steps:

[0006] S1 clusters the power curves for energy storage to mitigate fluctuations, obtaining the power and capacity requirements for energy storage to mitigate fluctuations.

[0007] S2 obtains the secondary frequency regulation backup power based on the total historical frequency regulation signal value, and gains revenue from energy storage participating in secondary frequency regulation ancillary services;

[0008] S3 calculates the energy storage peak-shaving capacity requirement and the revenue from energy storage participating in peak-shaving ancillary services.

[0009] S4 uses the energy storage overload capacity to set the primary frequency regulation droop coefficient of the energy storage;

[0010] S5 sets the energy storage operating power strategy. Mode 1 is when the energy storage operates in peak shaving and frequency regulation conditions. If the revenue of the energy storage participating in secondary frequency regulation ancillary services is lower than or equal to the revenue of the energy storage peak shaving ancillary services, then no frequency regulation power is reserved. If the revenue of the energy storage participating in secondary frequency regulation ancillary services is higher than the revenue of the energy storage peak shaving ancillary services, then the peak shaving power is reduced for frequency regulation reserve based on the secondary frequency regulation upper and lower reserve power obtained in step S2. Mode 2 is when the energy storage operates in frequency regulation and fluctuation smoothing conditions. The energy storage is prioritized for fluctuation smoothing, and the frequency regulation reserve power is adjusted based on the energy storage fluctuation smoothing power demand obtained in step S1.

[0011] S6 sets the energy storage state of charge range strategy, wherein the fluctuation smoothing SOC working range is set according to the energy storage fluctuation smoothing capacity demand obtained in step S1, the secondary frequency regulation upper and lower reserve power of mode 1 and mode 2 secondary frequency regulation SOC working range is set according to the secondary frequency regulation upper and lower reserve power obtained in step S2, and the peak shaving SOC working range is set according to the energy storage peak shaving capacity demand obtained in step S3, thereby realizing the coordinated operation of multiple services of the power type flow battery energy storage power station.

[0012] As a further preferred embodiment, step S1 specifically comprises:

[0013] S11 selects typical features of m curves from w energy storage power fluctuation curves in a certain year, a certain quarter, and a certain weather feature within the (r+1)T-(r+2)T period, and generates a vector set X composed of w vectors in m-dimensional Euclidean space.

[0014] S12 randomly selects K sample points from the vector set X in step S11 as the initial center points of K clusters, calculates the Euclidean distance from each sample in the vector set X to each initial center point, and assigns them to the cluster with the smallest Euclidean distance.

[0015] S13 calculates the sum of distances between a sample in each cluster and other samples in the cluster, and takes the point with the smallest sum of distances as the update center point of that cluster.

[0016] S14 Substitute the updated center point into step S12 and repeat steps S12 and S13 until the center point of the cluster no longer changes, thereby completing one cluster analysis;

[0017] In step S15, K is assigned values ​​from 1 to w, and steps S11 to S14 are repeated to complete w cluster analyses. The average silhouette coefficient of each cluster analysis is calculated using the following formula, and the K value with the largest average silhouette coefficient is selected as the final cluster analysis result.

[0018]

[0019] In the formula, Let a(j) be the average silhouette coefficient, and a(j) be the sample X. j The average Euclidean distance to samples within the same cluster, b(j) is the sample X. j The minimum Euclidean distance to other samples in other clusters;

[0020] S16 Based on the final cluster analysis results [P] clam1 ,P clam2 ,...P clamu ,...,P clamk ], 1≤u≤k and their corresponding occurrence probabilities [ρ1,ρ2,...,ρ k The following formula is used to calculate the power and capacity demand for energy storage to mitigate fluctuations.

[0021]

[0022] In the formula, E clam To mitigate fluctuating capacity demand for energy storage, P clam_in To mitigate fluctuating charging power demand, P clam_out To mitigate fluctuating discharge power demand for energy storage, t represents a point in time within a period T.

[0023] As a further preferred embodiment, in step S11, the typical characteristics of the curve include maximum charging power, maximum discharging power, average charging power, and average discharging power.

[0024] As a further preferred embodiment, step S2 specifically includes:

[0025] S21 uses the positive frequency modulation signal as the frequency modulation for energy storage discharge and the negative frequency modulation signal as the frequency modulation for energy storage charging. It statistically analyzes the power values ​​and durations of the up and down frequency modulation signals within the (r+1)T-(r+2)T period of a certain year, quarter, and weather characteristic, and calculates the total value L of the up and down frequency modulation signals within that T-period using the following formula. total ,

[0026]

[0027] In the formula, L total1 L is the total up-modulated signal value within period T. total2 L is the total value of the down-modulated signal within period T. AGC (q) represents the FM mileage of the qth FM signal, Δ(t) represents the duration of the qth FM signal, d represents the FM signal of a certain day, and s represents the total number of days of a certain year, a certain quarter, and a certain weather feature;

[0028] S22 calculates the total value L of the up and down frequency modulation signals obtained in step S21. total The upper and lower reserve power of the secondary frequency modulation is calculated using the following formula.

[0029]

[0030] In the formula, The energy storage is used for frequency regulation reserve power during cycle T; N1 represents the reserve power for down-frequency modulation of energy storage within cycle T, N2 represents the total number of up- and down-frequency modulation signals, and N2 represents the total number of down-frequency modulation signals.

[0031] S23 Based on the total value L of the up and down frequency modulation signals obtained in step S21 total The following formula is used to calculate the income from energy storage participating in secondary frequency regulation ancillary services. AGC ,

[0032] Income AGC =K AGC F AGC L total

[0033] In the formula, K AGC For the overall performance indicators of frequency modulation, F AGC To clear out prices for paid frequency regulation of energy storage.

[0034] As a further preferred embodiment, in step S3, the energy storage peak-shaving capacity requirement is calculated using the following formula.

[0035]

[0036] In the formula, E Peak E valley To meet the capacity required for energy storage and valley filling, P Peak For the peak-shaving charging power demand of energy storage within cycle T, P valley T1 is the energy storage valley filling discharge power within period T, T2 is the energy storage peak shaving duration within period T, T1 is the energy storage valley filling duration within period T, and η is the energy storage valley filling discharge power within period T. out This refers to the energy storage discharge power.

[0037] As a further preferred embodiment, in step S3, the revenue from energy storage participating in peak-shaving ancillary services is calculated using the following formula.

[0038] Income Peak-valley =F Peak-valley (P Peak T2+P valley T1η out )

[0039] In the formula, Income Peak-valley For revenue from energy storage participating in peak-shaving ancillary services, F Peak-valleyFor paid peak-shaving pricing of energy storage, P Peak For the peak-shaving charging power demand of energy storage within cycle T, P valley T1 is the energy storage valley filling discharge power within period T, T2 is the energy storage peak shaving duration within period T, T1 is the energy storage valley filling duration within period T, T1+T2=T, η out This refers to the energy storage discharge power.

[0040] As a further preferred embodiment, in step S4, the primary frequency regulation droop coefficient K of the energy storage is calculated using the following formula.

[0041]

[0042] In the formula, K1 is the up-modulation droop coefficient, K2 is the down-modulation droop coefficient; λ out λ is the overload factor for energy storage charge and discharge. in Overload factor for energy storage charging; f max1 f is the grid frequency corresponding to the maximum value of the up-modulated active power. max2 The grid frequency corresponding to the maximum value of down-regulation active power; f dead1 To increase the frequency dead zone, f dead2 To reduce the frequency dead zone, P max_out For the maximum discharge power of energy storage, P max_in This represents the maximum charging power for energy storage.

[0043] As a further preferred embodiment, in step S5, the calculation formula for the peak-shaving power reduced by mode 1 is as follows:

[0044]

[0045] In the formula, P Peak-valley (t) represents the reduced peak-shaving power, P max_out For the maximum discharge power of energy storage, P max_in This is the maximum charging power for energy storage. The energy storage is used for frequency regulation reserve power during cycle T; This refers to the backup power for energy storage during cycle T, which is adjusted downwards.

[0046] As a further preferred embodiment, in step S5, the calculation formula for the mode 2 corrected frequency modulation reserve power is as follows:

[0047]

[0048] In the formula, To correct the up-modulation reserve power, To correct the down-modulation reserve power, P clam_in To mitigate fluctuating charging power demand, P clam_out To mitigate fluctuating discharge power demand, P max_outFor the maximum discharge power of energy storage, P max_in This represents the maximum charging power for energy storage.

[0049] As a further preferred embodiment, in step S6, the SOC working range for smoothing fluctuations is:

[0050]

[0051] In the formula, SOC1 is the SOC working range for smoothing fluctuations, and E min_clam To minimize fluctuating capacity demand for energy storage, E max_clam To mitigate the maximum fluctuating capacity demand for energy storage, E N This refers to the rated capacity of the energy storage.

[0052] The operating range of the Mode 1 secondary frequency modulation SOC is:

[0053]

[0054] In the formula, SOC2 is the operating range of the secondary frequency modulation SOC. The energy storage is used for frequency regulation reserve power during cycle T; The energy storage power is used for downward frequency regulation during cycle T;

[0055] The operating range of the Mode 2 secondary frequency modulation SOC is:

[0056]

[0057] In the formula, SOC3 is the operating range of the secondary frequency modulation SOC, and E clam To mitigate fluctuating capacity demand for energy storage;

[0058] The peak-shaving SOC operating range is:

[0059]

[0060] In the formula, SOC4 is the peak-shaving SOC working range, and L total1 L is the total up-modulated signal value within period T. total2 E is the total value of the down-modulated signal within period T. Peak E valley The capacity required for energy storage and valley filling.

[0061] In summary, the technical solutions conceived by this invention have the following beneficial effects compared with the prior art:

[0062] 1. This invention proposes a method for determining a multi-service coordinated operation strategy for a flow battery energy storage power station. Based on a single peak-shaving operation, it considers primary frequency regulation, secondary frequency regulation, and wind power fluctuation mitigation functions. A clustering method is used to perform feature clustering on the wind power fluctuation mitigation curve of the energy storage system to obtain the power and capacity required for the fluctuation mitigation scenario. Historical frequency regulation signals are used to obtain the upper and lower reserve power for secondary frequency regulation. Finally, based on whether the grid has peak-shaving needs, the energy storage system is operated in mode 1 and mode 2. The operating power, state-of-charge range, and primary frequency regulation droop coefficient of the energy storage system are optimized. Compared with current multi-service operation parameter optimization methods for energy storage power stations, this method considers the wind power fluctuation mitigation function, making the flow battery energy storage power station more economical.

[0063] 2. At the same time, the present invention optimizes the clustering method of energy storage fluctuation power curve, which can reduce the scenario by extracting typical features of the energy storage fluctuation curve with randomness, thereby obtaining the energy storage fluctuation power demand and capacity demand.

[0064] 3. In addition, the present invention optimizes the method for dividing the energy storage state of charge range, which enables energy storage to operate in different functional scenarios in different working ranges, fully explores the potential value of energy storage, and improves the economic efficiency of energy storage. Attached Figure Description

[0065] Figure 1 This is a flowchart of the method for determining the multi-service coordinated operation strategy of a flow battery energy storage power station provided in an embodiment of the present invention;

[0066] Figure 2 This is a schematic diagram of the setting of the primary frequency regulation droop coefficient for energy storage provided in a preferred embodiment of the present invention. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0068] like Figure 1 As shown, this invention provides a method for determining a multi-service coordinated operation strategy for a flow battery energy storage power station, the method comprising the following steps:

[0069] S1 uses the K-Means method to cluster the energy storage fluctuation power curve to obtain the energy storage fluctuation power demand and capacity demand.

[0070] S2 obtains the secondary frequency regulation backup power based on the total historical frequency regulation signal value, and gains revenue from energy storage participating in secondary frequency regulation ancillary services;

[0071] S3 calculates the energy storage peak-shaving capacity requirement and the revenue from energy storage participating in peak-shaving ancillary services.

[0072] The S4 thermal power unit only requires 60 seconds of primary frequency regulation, while some studies on primary frequency regulation of wind power units only require 30 seconds. Even if large disturbances occur continuously for a short period of time, energy storage only needs to operate for 30 seconds to 1 minute. Therefore, this invention considers the energy storage's 1 minute continuous power overload capability and uses the energy storage overload capability to set the energy storage primary frequency regulation droop coefficient.

[0073] S5 sets the energy storage operation power strategy. Since the peak shaving command is a constant power value, the combined wind and storage power generation of the energy storage station facing the wind farm is constant power when the peak shaving command is in effect. Therefore, the energy storage station cannot operate in both peak shaving and frequency regulation conditions simultaneously. For this reason, the flow battery energy storage station operates in two modes. Mode 1 is when the grid has peak shaving demand, and the energy storage operates in both peak shaving and frequency regulation conditions. The actual power P of the energy storage at time t is... bess (t) is:

[0074] P bess (t)=L AGC (t)+K(f(t)-f dead )+P Peak-valley (t) (1)

[0075] In the formula, f(t) is the power grid frequency at time t, and L AGC (t) represents the secondary frequency regulation mileage at time t, K is the primary frequency regulation droop coefficient for energy storage, and f is the mileage for the secondary frequency regulation mileage at time t. dead For up and down frequency modulation dead zones, P Peak-valley (t) represents the energy storage peak-shaving power;

[0076] If the revenue from energy storage participating in secondary frequency regulation ancillary services is lower than or equal to the revenue from energy storage peak shaving ancillary services, then no standby frequency regulation power will be used. If the revenue from energy storage participating in secondary frequency regulation ancillary services is higher than the revenue from energy storage peak shaving ancillary services, then the peak shaving power will be reduced for frequency regulation standby based on the secondary frequency regulation upper and lower standby power obtained in step S2.

[0077] Mode 2 is when the power grid has no peak-shaving demand, and the energy storage operates in frequency regulation and fluctuation smoothing conditions. The actual power P of the energy storage at time t is... bess (t) is:

[0078] P bess (t)=L AGC (t)+P clam (t)+K(f(t)-f dead (2)

[0079] In the formula, P clam(t) represents the power of energy storage at time t to smooth out fluctuations. At this time, the energy storage power station should take smoothing out fluctuations as its basic function. If the energy storage power reserve for smoothing out fluctuations and the power required for frequency regulation are greater than the rated power of the energy storage power station, the energy storage should prioritize the power reserve for smoothing out fluctuations, and the remaining power should be used for frequency regulation reserve. The power reserve for frequency regulation should be adjusted according to the power requirement for smoothing out fluctuations obtained in step S1.

[0080] S6 sets the energy storage state of charge (SOC) range strategy, where primary frequency regulation is generally between 0.5 min and 1 min. Due to the random fluctuation of frequency, even if multiple large disturbances in the same direction occur consecutively in a short period of time, the energy storage of a battery with 12 minutes of full-power discharge can meet the requirements of primary frequency regulation. The primary frequency regulation working range is set to SOC ∈ [5%, 95%]. The fluctuation smoothing SOC working range is set according to the energy storage fluctuation smoothing capacity requirement obtained in step S1. The secondary frequency regulation upper and lower reserve power is set according to the secondary frequency regulation upper and lower reserve power obtained in step S2. The peak shaving SOC working range is set according to the energy storage peak shaving capacity requirement obtained in step S3, thereby realizing the coordinated operation of multiple services of the power-type flow battery energy storage power station.

[0081] Applying the method provided by this invention to the coordinated operation of power-type flow battery energy storage power stations can fully tap the potential value of flow battery energy storage power stations and takes into account the function of smoothing wind power fluctuations, making flow battery energy storage power stations more economical.

[0082] Furthermore, step S1 specifically includes:

[0083] S11 If there are w energy storage power fluctuation smoothing curves within a certain period of a certain year and a certain weather feature (r+1)T-(r+2)T, where T is the time period, and in the preferred embodiment T=1h, select m typical features of these curves (in the preferred embodiment, maximum charging power, maximum discharging power, average charging power, and average discharging power) to generate a vector set X = (X1, X2, ... X...) consisting of w vectors in an m-dimensional Euclidean space. j ,…,X w ), where X j =(x j1 ,x j2 ,…,x jm ), j = 1, 2, ..., w;

[0084] S12 randomly selects K sample points (V1, V2, V3, ..., V3) from the vector set X in step S11. k Using each sample in the vector set X as the initial centroid of K clusters, calculate the Euclidean distance from each sample in the vector set X to each initial centroid, and assign each sample to the cluster with the smallest Euclidean distance.

[0085] C k ={X j||X j -V k || 2 ≤||X j -V λ || 2} (3)

[0086] In the formula, C k For the k-th cluster, V k V is the center point of the k-th cluster. λ It serves as the center point for other clusters;

[0087] S13 calculates the sum of distances from a sample in each cluster to all other samples in the cluster, and takes the point with the smallest sum of distances as the updated center point of that cluster. The updated cluster center points (m1, m2, m3, ..., m...) are then set up. k ) replaces (V1,V2,V3,...,V k ),

[0088]

[0089] In the formula, X is... c and X j These are different samples;

[0090] S14 will update the centroids and substitute them into step S12, and repeat steps S12 and S13 until the centroids of the clusters no longer change, thus completing one cluster analysis.

[0091] In step S15, K is assigned a value from 1 to w, and steps S11 to S14 are repeated to complete w cluster analyses. The number of clusters K plays a decisive role in the clustering quality. The average silhouette coefficient is used to measure the quality of clustering; the closer the average silhouette coefficient is to 1, the better the clustering effect. The average silhouette coefficient for each cluster analysis is calculated using the following formula, and the K value with the largest average silhouette coefficient is selected as the final cluster analysis result.

[0092]

[0093] In the formula, Let a(j) be the average silhouette coefficient, and a(j) be the sample X. j The average Euclidean distance to samples within the same cluster, b(j) is the sample X. j The minimum Euclidean distance to other samples in other clusters;

[0094] S16 Based on the final cluster analysis results [P] clam1 ,P clam2 ,...P clamu ,...,P clamk ], 1≤u≤k and their corresponding occurrence probabilities [ρ1,ρ2,...,ρk The following formula is used to calculate the power and capacity demand for energy storage to mitigate fluctuations, defining charging power as positive and discharging power as negative.

[0095]

[0096] In the formula, E clam To mitigate fluctuating capacity demand for energy storage, P clam_in To mitigate fluctuating charging power demand, P clam_out To mitigate fluctuating discharge power demand for energy storage, t represents a point in time within a period T.

[0097] Furthermore, step S2 specifically includes:

[0098] S21 uses the positive frequency modulation signal as the frequency modulation for energy storage discharge and the negative frequency modulation signal as the frequency modulation for energy storage charging. It statistically analyzes the power values ​​and durations of the up and down frequency modulation signals within the (r+1)T-(r+2)T period of a certain year, quarter, and weather characteristic, and calculates the total value L of the up and down frequency modulation signals within that T-period using the following formula. total ,

[0099]

[0100] In the formula, L total1 L represents the total up-modulated signal value within period T, in MW·h. total2 L represents the total down-modulated signal value within period T, in MW·h. AGC (q) represents the frequency modulation mileage of the qth frequency modulation signal, which refers to the absolute value of the difference between the actual output value at the end after responding to the AGC control command and the output value at the time of responding to the command; Δ(t) represents the duration of the qth frequency modulation signal; d represents the frequency modulation signal of a certain day; s represents the total number of days of a certain year, a certain quarter, and a certain weather feature, and the frequency modulation signals of a certain year, a certain quarter, and a certain weather feature for s days are added together;

[0101] S22 calculates the total value L of the up and down frequency modulation signals obtained in step S21. total The upper and lower reserve power of the secondary frequency modulation is calculated using the following formula.

[0102]

[0103] In the formula, The energy storage is used for frequency regulation reserve power during cycle T; N1 represents the reserve power for down-frequency modulation of energy storage within cycle T, N2 represents the total number of up-frequency modulation signals, and N2 represents the total number of down-frequency modulation signals.

[0104] S23 Based on the total value L of the up and down frequency modulation signals obtained in step S21 total The following formula is used to calculate the income from energy storage participating in secondary frequency regulation ancillary services. AGC ,

[0105] Income AGC =K AGC F AGC L total (9)

[0106] In the formula, K AGC This is a comprehensive performance indicator for frequency modulation, and its value is related to response time, response accuracy, and adjustment rate. F AGC To clear out prices for paid frequency regulation of energy storage.

[0107] Furthermore, in step S3, the energy storage peak-shaving capacity requirement is calculated using the following formula:

[0108]

[0109] In the formula, E Peak E valley To meet the capacity required for energy storage and valley filling, P Peak For the peak-shaving charging power demand of energy storage within cycle T, P valley T1 is the energy storage valley filling discharge power within period T, T2 is the energy storage peak shaving duration within period T, T1 is the energy storage valley filling duration within period T, T1+T2=T, η out This refers to the energy storage discharge power;

[0110] The revenue from energy storage participating in peak-shaving ancillary services can be calculated using the following formula.

[0111] Income Peak-valley =F Peak-valley (P Peak T2+P valley T1η out (11)

[0112] In the formula, Income Peak-valley For revenue from energy storage participating in peak-shaving ancillary services, F Peak-valley For paid peak-shaving pricing of energy storage, P Peak For the peak-shaving charging power demand of energy storage within cycle T, P valley T1 is the energy storage discharge power during the valley filling period in period T, T2 is the energy storage peak shaving duration during period T, T1 is the energy storage valley filling duration during period T, and T1+T2=T.

[0113] Furthermore, in step S4, the primary frequency regulation droop coefficient is set using the energy storage overload capacity.

[0114]

[0115] In the formula, K1 is the up-modulation droop coefficient, K2 is the down-modulation droop coefficient; λ out λ is the overload factor for energy storage charge and discharge.in Overload factor for energy storage charging; f max1 f is the grid frequency corresponding to the maximum value of the up-modulated active power. max2 The grid frequency corresponding to the maximum value of down-regulation active power; f dead1 To increase the frequency dead zone, f dead2 To reduce the frequency dead zone, P max_out For the maximum discharge power of energy storage, P max_in This represents the maximum charging power for energy storage.

[0116] Furthermore, in step S5, the calculation formula for the peak-shaving power reduced by Mode 1 is as follows:

[0117]

[0118] In the formula, P Peak-valley (t) represents the reduced peak-shaving power, P max_out For the maximum discharge power of energy storage, P max_in This represents the maximum charging power for energy storage.

[0119] In step S5, the calculation formula for the modified frequency modulation reserve power in mode 2 is as follows:

[0120]

[0121] In the formula, To correct the up-modulation reserve power, To correct the down-modulation reserve power, P clam_in To mitigate fluctuating charging power demand, P clam_out To mitigate fluctuating discharge power demand, P max_out For the maximum discharge power of energy storage, P max_in This represents the maximum charging power for energy storage.

[0122] Furthermore, in step S6, the SOC working range for smoothing fluctuations is:

[0123]

[0124] In the formula, SOC1 is the SOC working range for smoothing fluctuations, and E min_clam To minimize fluctuating capacity demand for energy storage, E max_clam To mitigate the maximum fluctuating capacity demand for energy storage, E N This refers to the rated capacity of the energy storage.

[0125] The operating range of the Mode 1 secondary frequency modulation SOC is:

[0126]

[0127] In the formula, SOC2 is the operating range of the secondary frequency modulation SOC. The energy storage is used for frequency regulation reserve power during cycle T; The energy storage power is used for downward frequency regulation during cycle T;

[0128] The operating range of the Mode 2 secondary frequency modulation SOC is:

[0129]

[0130] In the formula, SOC3 is the operating range of the secondary frequency modulation SOC, and E clam To mitigate fluctuating capacity demand for energy storage;

[0131] The peak-shaving SOC operating range is:

[0132]

[0133] In the formula, SOC4 is the peak-shaving SOC working range, and L total1 L is the total up-modulated signal value within period T. total2 E is the total value of the down-modulated signal within period T. Peak E valley The capacity required for energy storage and valley filling.

[0134] The technical solution provided by the present invention will be further described below with reference to specific embodiments.

[0135] The selected flow battery energy storage power station has a power of 20MW, a configuration duration of 4h, and a charge and discharge efficiency of 0.85.

[0136] (1) Assume that there is a peak discharge demand of 40 minutes and a peak charging demand of 20 minutes during the period from 21:00 to 22:00 on a certain day, the peak compensation price is 0.6 kWh, and the daytime frequency regulation price during this period is 15 yuan per MW.

[0137] The average total value of the up and down frequency modulation signals during this period, L total1 L total2 Both are 30, the average frequency modulation performance index is 2, and the total number of up and down frequency modulation signals is 10. Therefore, the energy storage up-frequency modulation backup is... 3MW, frequency regulation reserve The capacity is 3MW. The revenue from this frequency regulation reserve power used for peak shaving is... The value is 1710 yuan. Based on historical data, the revenue from frequency regulation during this period is 1800 yuan. Since the revenue from frequency regulation is greater than that from peak regulation, the energy storage power station adjusts its peak regulation power during this period to 17MW of peak discharge for 40 minutes and 17MW of peak charging for 20 minutes. At this time, it is operating in mode 1, with the secondary frequency regulation SOC operating range being [8.75%, 91.25%] and the peak regulation SOC operating range being [19.17%, 80.83%].

[0138] (2) Taking the power data of a 100MW wind farm during a sunny weather period from 22:00 to 23:00 in spring of a certain year as an example, clustering was performed. The clustering result was the best when the total number of clusters reached 5. The maximum charging and discharging power of each cluster scenario and the probability values ​​of each scenario are shown in the table below. The weighted charging power demand is 4.62MW and the discharging power demand is 5.49MW. Based on historical data E clam It is 8.35 MW·h.

[0139] Table 1 Clustering results of energy storage power fluctuation curves

[0140]

[0141]

[0142] Assuming there is no peak-shaving demand between 22:00 and 23:00 on this day, the system operates in mode 2, with the average of the up and down frequency modulation signals being L. total1 L total2 Both are 40, and the total number of up and down frequency modulation signals is 10, so the energy storage up frequency modulation is used for standby. 4MW, frequency regulation reserve The capacity is 4MW. Since the sum of frequency regulation reserve and fluctuation suppression demand does not exceed the rated power of energy storage, energy storage can simultaneously act on wind power fluctuation suppression and frequency regulation reserve at 4MW. The secondary frequency regulation SOC operating range is [10%, 90%], and the fluctuation suppression SOC operating range is [20.44%, 79.56%].

[0143] (3) If the overload capacity of the energy storage power station is set to 120%, then the maximum charging and discharging power of the energy storage power station that can be used for primary frequency regulation is 2MW. Figure 2 As shown, f max1 Set to 49.8Hz, f max2 Set to 50.3Hz, f dead1 Set to 49.95Hz, f dead2 With the frequency set to 50.05Hz, the droop coefficient K1 for upward frequency modulation is 13.33MW / Hz, and the droop coefficient K2 for downward frequency modulation is 8MW / Hz.

[0144] Those skilled in the art will readily understand that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements 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 method for determining a multi-service coordinated operation strategy of a flow battery energy storage power station, characterized in that, The method includes the following steps: S1 clusters the power curves for energy storage to mitigate fluctuations, obtaining the power and capacity requirements for energy storage to mitigate fluctuations. S2 calculates the secondary frequency regulation standby power based on the total historical frequency regulation signal value and obtains revenue from energy storage participating in secondary frequency regulation ancillary services, specifically: S21 uses the positive frequency modulation signal as the frequency modulation for energy storage discharge and the negative frequency modulation signal as the frequency modulation for energy storage charging. It then statistically analyzes a specific year, quarter, and weather characteristic (the...). r +1)T-( r +2) T The power values ​​and durations of the up and down frequency modulation signals within a period are calculated using the following formula. T The total value of up and down frequency modulation signals within the period , In the formula, This represents the total value of the up-modulated signal within period T. This represents the total value of the down-modulated signal within period T. For the first q Frequency modulation mileage of one frequency modulation signal For the first q The duration of the frequency modulation signal d The frequency modulation signal for a certain day. s The total number of days in a certain year, a certain quarter, and a certain weather characteristic; S22 is based on the total value of the up and down frequency modulation signals obtained in step S21. The upper and lower reserve power of the secondary frequency modulation is calculated using the following formula. In the formula, The energy storage is used for frequency regulation reserve power during cycle T; This refers to the reserve power for downward frequency regulation of energy storage within cycle T. This represents the total number of up and down frequency modulation signals. This represents the total number of down-modulated signals. S23 Based on the total value of the up and down frequency modulation signals obtained in step S21 The revenue from energy storage participating in secondary frequency regulation ancillary services is calculated using the following formula. , In the formula, For the overall performance indicators of frequency modulation, To clear out prices for paid frequency regulation of energy storage; S3 calculates the energy storage peak-shaving capacity demand and the revenue from energy storage participating in peak-shaving ancillary services. The energy storage peak-shaving capacity demand is calculated using the following formula. In the formula, The capacity required for energy storage peak shaving To provide the capacity required for energy storage and valley filling, This refers to the peak-shaving charging power demand for energy storage within cycle T. This represents the energy storage discharge power during cycle T. The duration of peak shaving for energy storage within cycle T. The duration of energy storage valley filling within cycle T. This refers to the energy storage discharge power; S4 uses the energy storage overload capacity to set the primary frequency regulation droop coefficient of the energy storage; S5 sets the energy storage operating power strategy. Mode 1 is when the energy storage operates in peak shaving and frequency regulation conditions. If the revenue of the energy storage participating in secondary frequency regulation ancillary services is lower than or equal to the revenue of the energy storage peak shaving ancillary services, then no frequency regulation power is reserved. If the revenue of the energy storage participating in secondary frequency regulation ancillary services is higher than the revenue of the energy storage peak shaving ancillary services, then the peak shaving power is reduced for frequency regulation reserve according to the secondary frequency regulation upper and lower reserve power obtained in step S2. Mode 2 is when the energy storage operates in frequency regulation and fluctuation smoothing conditions. The energy storage prioritizes the reserve power for fluctuation smoothing, and the frequency regulation reserve power is adjusted according to the energy storage fluctuation smoothing power demand obtained in step S1. S6 sets the energy storage state of charge range strategy, wherein the fluctuation smoothing SOC working range is set according to the energy storage fluctuation smoothing capacity demand obtained in step S1, the secondary frequency regulation upper and lower reserve power of mode 1 and mode 2 secondary frequency regulation SOC working range is set according to the secondary frequency regulation upper and lower reserve power obtained in step S2, and the peak shaving SOC working range is set according to the energy storage peak shaving capacity demand obtained in step S3, thereby realizing the coordinated operation of multiple services of the power type flow battery energy storage power station.

2. The method for determining the multi-service coordinated operation strategy of a flow battery energy storage power station as described in claim 1, characterized in that, Step S1 is as follows: S11 selects a certain quarter of a certain year and a certain weather characteristic (the... r +1) T -( r +2) T Within the period w In the energy storage power fluctuation curve m A typical feature of the curve, generating m In the Veuvian space w The vector set formed by the bar vectors X ; S12 is the vector set in step S11. X Random selection K Each sample point is used as K The initial center points of each cluster are used to calculate the vector sets. X Calculate the Euclidean distance from each sample to each initial center point, and assign them to the cluster with the smallest Euclidean distance; S13 calculates the sum of distances between a sample in each cluster and other samples in the cluster, and takes the point with the smallest sum of distances as the update center point of that cluster. S14 Substitute the updated center point into step S12 and repeat steps S12 and S13 until the center point of the cluster no longer changes, thereby completing one cluster analysis; S15 pairs K Assigning value 1~ w Repeat steps S11 to S14 to complete the process. w For each cluster analysis, the average silhouette coefficient is calculated using the following formula, and the cluster with the largest average silhouette coefficient is selected. K The value is used as the final cluster analysis result. In the formula, The average profile coefficient. For the sample The average Euclidean distance from samples within the same cluster. For the sample The minimum Euclidean distance to other samples in other clusters; S16 Based on the final cluster analysis results and their corresponding probability of occurrence The following formula is used to calculate the energy storage's ability to mitigate fluctuating power and capacity demands. In the formula, To mitigate fluctuating capacity demand for energy storage, To mitigate fluctuating charging power demand, To mitigate fluctuating discharge power demand, t for T A specific point in time within a period.

3. The method for determining the multi-service coordinated operation strategy of a flow battery energy storage power station as described in claim 2, characterized in that, In step S11, the typical characteristics of the curve include maximum charging power, maximum discharging power, average charging power, and average discharging power.

4. The method for determining the multi-service coordinated operation strategy of a flow battery energy storage power station as described in claim 1, characterized in that, In step S3, the revenue from energy storage participating in peak-shaving ancillary services is calculated using the following formula. In the formula, For revenue from energy storage participating in peak-shaving ancillary services, To clear out prices for paid peak-shaving of energy storage. This refers to the peak-shaving charging power demand for energy storage within cycle T. This represents the energy storage discharge power during cycle T. The duration of peak shaving for energy storage within cycle T. The duration of energy storage valley filling within cycle T. , This refers to the energy storage discharge power.

5. The method for determining the multi-service coordinated operation strategy of a flow battery energy storage power station as described in claim 1, characterized in that, In step S4, the primary frequency regulation droop coefficient of energy storage is calculated using the following formula. K , In the formula, This is the up-modulation droop factor. This refers to the droop factor of the down-modulation frequency; For energy storage charge and discharge overload multiple, Overload factor for energy storage charging; The grid frequency corresponding to the maximum value of the up-modulated active power. The grid frequency corresponding to the maximum value of down-regulation active power; To adjust the frequency dead zone, To reduce the frequency dead zone, To store the maximum discharge power, This represents the maximum charging power for energy storage.

6. The method for determining the multi-service coordinated operation strategy of a flow battery energy storage power station as described in claim 1, characterized in that, In step S5, the formula for calculating the peak-shaving power after mode 1 reduction is as follows: In the formula, This is the reduced peak-shaving power. To store the maximum discharge power, This is the maximum charging power for energy storage. The energy storage is used for frequency regulation reserve power during cycle T; This refers to the backup power for energy storage during cycle T, which is adjusted downwards.

7. The method for determining the multi-service coordinated operation strategy of a flow battery energy storage power station as described in claim 1, characterized in that, In step S5, the calculation formula for the modified frequency modulation reserve power in mode 2 is as follows: In the formula, To correct the up-modulation reserve power, To correct the down-modulation reserve power, To mitigate fluctuating charging power demand, To mitigate fluctuating discharge power demand, To store the maximum discharge power, This represents the maximum charging power for energy storage.

8. The method for determining the multi-service coordinated operation strategy of a flow battery energy storage power station as described in any one of claims 1 to 7, characterized in that, In step S6, the SOC working range for smoothing fluctuations is: SOC1∈[5%+ ,95%- ] In the formula, SOC1 is the SOC working range for smoothing fluctuations. To minimize the fluctuating capacity demand for energy storage, To mitigate fluctuations in energy storage capacity demand, This refers to the rated capacity of the energy storage. The operating range of the Mode 1 secondary frequency modulation SOC is: SOC2∈[5%+ ,95%- ] In the formula, SOC2 is the operating range of the secondary frequency modulation SOC. The energy storage is used for frequency regulation reserve power during cycle T; The energy storage power is used for downward frequency regulation during cycle T; The operating range of the Mode 2 secondary frequency modulation SOC is: SOC3∈[5%+ ,95%- ] In the formula, SOC3 is the operating range of the secondary frequency modulation SOC. To mitigate fluctuating capacity demand for energy storage; The peak-shaving SOC operating range is: SOC4∈[5%+ ,95%- ] In the formula, SOC4 is the peak-shaving SOC operating range. This represents the total value of the up-modulated signal within period T. This represents the total value of the down-modulated signal within period T. The capacity required for energy storage peak shaving The capacity required for energy storage and valley filling.

Images