An Optimal Bidding Method and System for Wind-Power Storage Integration in Energy and Frequency Regulation Markets

Abstract

The present invention discloses an optimal bidding method and system for wind power and energy storage to participate in the energy-frequency regulation market. The bidding method includes: obtaining the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market; substituting the predicted output of the wind farm and the energy storage capacity for each trading period, and the clearing prices of the energy market and the frequency regulation market into a pre-constructed power market bidding model for calculation to obtain the capacities of wind power and / or energy storage participating in each scenario; the scenarios include wind power participating in the energy market, energy storage participating in the energy market, wind power and energy storage jointly participating in the frequency regulation market, and energy storage participating in the frequency regulation market; the power market bidding model is constructed with the goal of maximizing the revenue when wind power and / or energy storage participate in each scenario. The present invention effectively improves the frequency regulation performance of wind power and to a certain extent reduces the loss cost of energy storage.

Description

Technical Field

[0001] The present invention relates to the field of power markets, and particularly to an optimal bidding method and system for a wind energy storage combined participation in the energy-frequency regulation market. Background Art

[0002] With the continuous increase in the penetration rate of renewable energy, the demand for frequency regulation services in the power system is also increasing day by day. In a power system with a relatively high penetration rate of renewable energy, renewable energy, such as wind power, will inevitably undertake part of the frequency regulation services of conventional generating units. When wind power participates in frequency regulation market transactions, the uncertainty of wind power output will reduce the quality of its frequency regulation services. Energy storage has great potential in the frequency regulation market due to its better frequency regulation accuracy and faster response speed. Therefore, energy storage can well make up for the defects of wind power frequency regulation. Therefore, it has certain research significance to provide power generation and frequency regulation services for the power market by a virtual power plant composed of wind power and energy storage. At present, after wind power and energy storage are combined to participate in frequency regulation services, there are not only situations where the frequency regulation capacity does not match the declared capacity, but also the energy storage battery faces the risk of accelerated aging during the frequency regulation process of frequent charge and discharge, resulting in a shortened life of the energy storage battery. The inventor proposes to determine the capacities of different combinations of wind power and energy storage participating in the energy market and the frequency regulation market with the goal of maximizing the benefits of wind power and / or energy storage in different scenarios. Summary of the Invention

[0003] In order to solve the above-mentioned deficiencies in the prior art, the present invention provides an optimal bidding method for a wind energy storage combined participation in the energy-frequency regulation market, including:

[0004] Obtaining the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market;

[0005] Substituting the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market into a pre-constructed power market bidding model for calculation to obtain the capacities of wind power and / or energy storage participating in each scenario;

[0006] wherein the scenarios include wind power participating in the energy market, energy storage participating in the energy market, wind energy storage combined participating in the frequency regulation market, and energy storage participating in the frequency regulation market;

[0007] The power market bidding model is constructed with the goal of maximizing the benefits of wind power and / or energy storage participating in each scenario.

[0008] Preferably, the construction of the power market bidding model includes:

[0009] Based on the day-ahead clearing revenue of wind power participating in the energy market, the day-ahead clearing revenue of energy storage participating in the energy market, the day-ahead clearing revenue of wind-storage combined participating in the frequency regulation market, the day-ahead revenue of energy storage participating in the frequency regulation market alone, and the historical daily loss cost of energy storage, an optimal expected revenue objective function is constructed with the goal of maximizing the expected revenue;

[0010] Based on the optimal expected revenue objective function and the constraint conditions constructed for the optimal expected revenue objective function, the relationships between the energy clearing price, the frequency regulation mileage clearing price, the frequency regulation capacity clearing price and the day-ahead bidding capacity are obtained;

[0011] Based on the net revenue of wind power and energy storage participating in the energy market, the frequency regulation capacity revenue of wind-storage combined, the frequency regulation mileage revenue of wind-storage combined, the day-ahead revenue of energy storage participating in the frequency regulation market alone, the intraday revenue of energy storage participating in the frequency regulation market alone, the penalty cost, the loss cost of energy storage, and the day-ahead bidding capacity, a total revenue optimal objective function is constructed with the goal of maximizing the total revenue;

[0012] Based on the total revenue optimal objective function and the constraint conditions constructed for the total revenue optimal objective function, the relationships between the energy clearing price, the frequency regulation mileage clearing price, the frequency regulation capacity clearing price and the capacities of wind power participating in the energy market, energy storage participating in the energy market, wind-storage combined participating in the frequency regulation market, energy storage participating in wind-storage combined frequency regulation, and energy storage participating in the frequency regulation market alone are obtained.

[0013] Preferably, the optimal expected revenue objective function is as shown in the following formula:

[0014]

[0015] In the formula: is the day-ahead clearing price of the power market at time t; is the day-ahead clearing capacity of energy storage participating in the energy market at time t; is the day-ahead clearing capacity of wind power participating in the energy market at time t; Δt is the time scale of day-ahead clearing; is the day-ahead clearing capacity of wind-storage combined participating in frequency regulation; ρ is the frequency regulation performance index; is the day-ahead capacity clearing price of the frequency regulation market at time t; η is the mileage revenue factor; is the day-ahead mileage clearing price of the frequency regulation market at time t; is the day-ahead clearing capacity of wind-storage combined participating in frequency regulation; H invest is the investment cost of energy storage; N 100％ is the maximum number of cycles in a complete discharge cycle; is the historical average daily loss cost in a complete discharge cycle; Hour is a set containing 24 integer time points.

[0016] Preferably, the total revenue optimization objective function is as follows:

[0017]

[0018] In the formula: is the net revenue of wind power and energy storage participating in the energy market; is the capacity revenue of wind - storage combined participating in the frequency regulation market; is the mileage revenue of wind - storage combined participating in the frequency regulation market; is the capacity revenue of energy storage alone participating in the frequency regulation market; is the mileage revenue of energy storage alone participating in the frequency regulation market; is the penalty cost at time t; is the loss cost of energy storage.

[0019] Preferably, the penalty cost is as follows:

[0020]

[0021] In the formula: is the under - generation penalty coefficient of the frequency regulation market; is the day - ahead clearing capacity of energy storage alone participating in frequency regulation at time t; is the real - time clearing capacity of energy storage alone participating in frequency regulation at time τ.

[0022] Preferably, the loss cost of energy storage is as follows:

[0023]

[0024] In the formula: is the capacity cost per unit time of energy storage; is the energy storage cost at time τ;

[0025] Among them, the capacity cost per unit time of energy storage is as follows:

[0026]

[0027] In the formula: h capacity is the unit capacity cost of energy storage; C battery is the rated output power of energy storage; r is the discount rate; T float is the floating - charge life of energy storage;

[0028] The energy storage cost at time τ is as follows:

[0029]

[0030] Where: H invest is the investment cost of energy storage; N 100％ is the maximum number of cycles in a complete discharge cycle; is the number of cycles in the time period [k, k+Δk] at a cycle depth of a complete discharge cycle.

[0031] Preferably, the number of cycles at a cycle depth of a complete discharge cycle is calculated according to the following formula:

[0032]

[0033] Where: is the daily cycle number at a discharge depth d; d is the cycle depth; k p is a constant;

[0034] Among them, the cycle depth d is calculated according to the following formula:

[0035]

[0036] Where: d k is the cycle depth of energy storage in two adjacent control moments; is the charge and discharge state of the energy storage battery at the (k + 1)th moment; is the charge and discharge state of the energy storage battery at the kth moment; ξ is the charging efficiency of the battery; is the charge and discharge power of the battery at the kth moment, equal to the power of the energy storage participating in the energy market at the kth moment and the power of the energy storage participating in the frequency regulation market at the kth moment The sum, where is determined by the capacity of the energy storage participating in wind-storage combined frequency regulation at the kth moment and the real-time cleared capacity of the energy storage at the τth moment decide.

[0037] Preferably, the capacity revenue of the energy storage participating in the frequency regulation market alone is calculated according to the following formula:

[0038]

[0039] Where: ρ is the value of the frequency regulation performance index; is the day-ahead capacity clearing price of the frequency regulation market at the tth moment; is the intra-day cleared capacity of the energy storage participating in frequency regulation alone at the τth moment; is the day-ahead cleared capacity of the energy storage participating in frequency regulation alone at the tth moment; Δt is the time scale of day-ahead clearing; Δτ is the real-time clearing scale.

[0040] Preferably, the mileage revenue of the energy storage participating in the frequency regulation market alone Calculate according to the following formula:

[0041]

[0042] Where: ρ is the value of the frequency modulation performance index; η is the mileage revenue factor; is the day-ahead mileage clearing price of the frequency modulation market at time t; is the real-time clearing capacity of the energy storage participating in frequency modulation alone at time τ; is the day-ahead clearing capacity of the energy storage participating in frequency modulation alone at time t.

[0043] Preferably, the capacity revenue of the wind-energy storage combined participation in the frequency modulation market Calculate according to the following formula:

[0044]

[0045] Where: ρ is the value of the frequency modulation performance index; is the day-ahead capacity clearing price of the frequency modulation market at time t; is the real-time clearing capacity of the wind-energy storage combined participation in the frequency modulation market at time τ; is the day-ahead clearing capacity of the wind-energy storage combined participation in the frequency modulation at time t.

[0046] Preferably, the mileage revenue of the wind-energy storage combined participation in the frequency modulation market Calculate according to the following formula:

[0047]

[0048] Where: η is the mileage revenue factor; is the day-ahead mileage clearing price of the frequency modulation market at time t.

[0049] Preferably, after obtaining the optimal objective function of the total revenue and the constraint conditions constructed for the optimal objective function of the total revenue, it further includes:

[0050] Using a robust optimization model to update the uncertainty variables in the optimal objective function of the total revenue and the constraint conditions constructed for the optimal objective function of the total revenue;

[0051] The uncertainty variables include the energy market clearing price, the frequency modulation capacity clearing price, the frequency modulation mileage clearing price, and the real-time output of wind power.

[0052] Preferably, the updated optimal objective function of the total revenue is as shown in the following formula:

[0053]

[0054] Where: is the predicted value of the day-ahead clearing price of the power market at time t; is the profit of the battery at time t; is the profit of wind power at time t; is the penalty cost at time t; is the predicted value of the clearing price in the energy market; is the predicted value of the clearing price of the frequency regulation capacity; is the capacity profit of the wind - storage combination participating in the frequency regulation market; is the capacity profit of the energy storage participating in the frequency regulation market alone; is the penalty cost at time t; is the predicted value of the day - ahead capacity clearing price in the frequency regulation market at time t; is the predicted value of the day - ahead mileage clearing price in the frequency regulation market at time t; is the mileage profit of the wind - storage combination participating in the frequency regulation market; is the mileage profit of the energy storage participating in the frequency regulation market alone; is the predicted value of the number of cycles in the time period [k, k + Δk] under a complete discharge cycle depth; is the energy storage cost at time τ; is the capacity cost per unit time of the energy storage; z1 is the first - type optimization parameter corresponding to the energy market clearing price; Λ1 is the second - type optimization parameter corresponding to the energy market clearing price; is the third - type optimization parameter corresponding to the energy market clearing price; z2 is the first - type optimization parameter corresponding to the frequency regulation capacity clearing price; Λ2 is the second - type optimization parameter corresponding to the frequency regulation capacity clearing price; is the third - type optimization parameter corresponding to the frequency regulation capacity clearing price; z3 is the first - type optimization parameter corresponding to the frequency regulation mileage clearing price; Λ3 is the second - type optimization parameter corresponding to the frequency regulation mileage clearing price; is the third - type optimization parameter corresponding to the frequency regulation mileage clearing price; Hour is a set containing 24 whole - point moments.

[0055] Preferably, the updated constraint conditions are as shown in the following formula:

[0056]

[0057] In the formula: λ1 is the first dual multiplier; λ2 is the second dual multiplier; λ3 is the third dual multiplier; λ4 is the fourth dual multiplier, Λ wind is the conservatism parameter of the real - time output of wind power; y1 is the fourth - type optimization parameter corresponding to the energy market clearing price; y2 is the fourth - type optimization parameter corresponding to the frequency regulation capacity clearing price; y3 is the fourth - type optimization parameter corresponding to the frequency regulation mileage clearing price; is the real - time clearing capacity of the wind - storage combination participating in the frequency regulation market at time τ; is the real - time clearing capacity of wind power participating in the energy market at time τ; is the real - time clearing capacity of energy storage participating in the energy market at time τ; is the real - time clearing capacity of wind power participating in upward frequency regulation at time τ; is the output of the wind farm at time τ.

[0058] Based on the same inventive concept, the present invention also provides an optimal bidding system for wind - storage joint participation in the energy - frequency regulation market, including:

[0059] An acquisition module, configured to acquire the predicted output of the wind farm and the energy storage capacity in each trading period, as well as the clearing prices of the energy market and the frequency regulation market;

[0060] A result module, configured to input the predicted output of the wind farm and the energy storage capacity in each trading period, as well as the clearing prices of the energy market and the frequency regulation market into a pre - constructed power market bidding model for calculation, and obtain the capacities of wind power and / or energy storage participating in each scenario;

[0061] wherein, the scenarios include wind power participating in the energy market, energy storage participating in the energy market, wind - storage joint participation in the frequency regulation market, and energy storage participating in the frequency regulation market;

[0062] The power market bidding model is constructed with the goal of maximizing the revenue when wind power and / or energy storage participate in each scenario.

[0063] Preferably, the system further includes a module for constructing a power market bidding model, specifically used for:

[0064] Based on the day - ahead clearing revenue of wind power participating in the energy market, the day - ahead clearing revenue of energy storage participating in the energy market, the day - ahead clearing revenue of wind - storage joint participation in the frequency regulation market, the day - ahead revenue of energy storage participating in the frequency regulation market alone, and the historical daily loss cost of energy storage, construct an expected revenue optimal objective function with the goal of maximizing the expected revenue;

[0065] Based on the expected revenue optimal objective function and the constraint conditions constructed for the expected revenue optimal objective function, obtain the relationships between the energy clearing price, the frequency regulation mileage clearing price, the frequency regulation capacity clearing price and the day - ahead bidding capacity;

[0066] Based on the net revenue of wind power and energy storage participating in the energy market, the frequency regulation capacity revenue of wind - storage joint participation, the frequency regulation mileage revenue of wind - storage joint participation, the day - ahead revenue of energy storage participating in the frequency regulation market alone, the intraday revenue of energy storage participating in the frequency regulation market alone, the penalty cost, the loss cost of energy storage, and the day - ahead bidding capacity, construct a total revenue optimal objective function with the goal of maximizing the total revenue;

[0067] Based on the total revenue optimal objective function and the constraint conditions constructed for the total revenue optimal objective function, the relationships between the energy clearing price, the frequency regulation mileage clearing price, and the frequency regulation capacity clearing price and the capacities of wind power participating in the energy market, energy storage participating in the energy market, wind-storage combined participating in the frequency regulation market, energy storage participating in wind-storage combined frequency regulation, and energy storage participating in the frequency regulation market alone are obtained.

[0068] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0069] The technical solution provided by the present invention obtains the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market; substitutes the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market into a pre-constructed electricity market bidding model for calculation to obtain the capacities of wind power and / or energy storage participating in each scenario; constructs the electricity market bidding model with the goal of maximizing the revenue when wind power and / or energy storage participate in each scenario, considering scenarios where wind power participates in the energy market, energy storage participates in the energy market, wind-storage combined participates in the frequency regulation market, and energy storage participates in the frequency regulation market; flexibly determines the capacities of wind power and energy storage participating in the frequency regulation market and the energy market under the condition of maximum economic revenue, effectively improving the frequency regulation performance of wind power and at the same time reducing the loss cost of energy storage.

[0070] The preferred solution provided by the present invention fully considers the advantages and disadvantages of two resources, namely wind power and energy storage, introduces penalty costs and energy storage loss costs to make the influence of the characteristics of the two resources on revenue specific, and effectively improves the frequency regulation performance of wind power and reduces the loss cost of energy storage without affecting the lifespan of the energy storage battery according to the electricity market bidding model. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 is a flowchart of an optimal bidding method for wind-storage combined participation in the energy-frequency regulation market provided by the present invention;

[0072] Figure 2 is a schematic diagram of the model for wind-storage combined participation in the electricity market in an embodiment of the present invention;

[0073] Figure 3 is a flowchart of an optimal bidding strategy for wind-storage combined participation in the energy-frequency regulation market based on penalty and energy storage loss costs in an embodiment of the present invention;

[0074] Figure 4 is a graph of the historical output of a certain wind farm on a certain day and the historical trading data of the electricity market in an embodiment of the present invention;

[0075] Figure 5 is the real-time capacity of wind-storage participating in the electricity market alone in an embodiment of the present invention (Λ wind= 5%, Λ1 = Λ2 = Λ3 = 5%) curve graph;

[0076] Figure 6 is the real-time capacity of the wind-storage combination participating in the power market in the embodiment of the present invention (Λ wind = 5%, Λ1 = Λ2 = Λ3 = 5%) curve graph;

[0077] Figure 7 is the real-time capacity of the wind-storage combination participating in the power market in the embodiment of the present invention (Λ wind = 10%, Λ1 = Λ2 = Λ3 = 5%) curve graph. Detailed implementation manners

[0078] To better understand the present invention, the content of the present invention will be further described below in conjunction with the specification drawings and examples.

[0079] Embodiment 1: An auction method for a wind-storage combination to participate in the energy market and the frequency regulation market proposed in the embodiment of the present invention includes a revenue model for the wind-storage participating in the power market under four scenarios. The advantages and disadvantages of the two resources are fully considered, and penalty costs and loss costs are introduced to make the influence of the characteristics of the two resources on the revenue specific. Corresponding objective functions and constraint conditions are set according to the auction model. The real-time capacity and net revenue are optimized. Finally, the effectiveness and correctness of the model are verified. As a combined auction scheme, this scheme effectively improves the frequency regulation performance of wind power without affecting the life of the energy storage battery, and to a certain extent reduces the loss cost of the energy storage.

[0080] As Figure 1 shown, an optimal auction method for a wind-storage combination to participate in the energy-frequency regulation market provided by the embodiment of the present invention includes:

[0081] S1 Obtain the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market;

[0082] S2 Substitute the predicted output of the wind farm and the energy storage capacity for each trading period, and the clearing prices of the energy market and the frequency regulation market into a pre-constructed power market auction model for calculation to obtain the capacities of the wind power and / or the energy storage participating in each scenario;

[0083] wherein, the scenarios include the wind power participating in the energy market, the energy storage participating in the energy market, the wind-storage combination participating in the frequency regulation market, and the energy storage participating in the frequency regulation market;

[0084] The power market auction model is constructed with the goal of maximizing the revenue when the wind power and / or the energy storage participate in each scenario.

[0085] As Figure 2 shown, a model of the wind-storage combination participating in the power market is shown.

[0086] As Figure 3 shown, in the embodiment of the present invention, the optimal bidding strategy process for wind energy storage combined participation in the energy-frequency regulation market based on penalty and energy storage loss cost includes:

[0087] (1) Analyze the energy-frequency regulation service market mechanism, including the bidding mechanism and market operation mechanism of the energy market and the frequency regulation service market. According to the proposed penalty mechanism, apply it to cost analysis and revenue analysis at different time scales of day-ahead and real-time, and adopt the sequential bidding mode of the energy market first and then the frequency regulation market;

[0088] (2) Propose a revenue model for wind energy storage combined participation in the power market considering energy storage loss cost and deviation penalty cost, including the revenue model of wind power and energy storage participating in the energy market, the revenue model of wind energy storage combined participation in the frequency regulation market, the revenue model of energy storage participating in the frequency regulation market alone, the cost model of energy storage participating in the power market, and the model of wind energy storage combined participation in the power market;

[0089] (3) Establish a bidding model based on optimal economic revenue, establish an optimization model of expected revenue, and obtain the real-time clearing capacity and total revenue through total revenue optimization. Establish a robust optimization model, fully consider the uncertainty of wind power output and market price fluctuations, and adjust the fluctuation degree of uncertain parameters through the conservatism parameter, and can effectively obtain the bidding strategies under different conservatism degrees.

[0090] (4) Set parameters for relevant equipment such as energy storage, verify the bidding model for wind energy storage combined participation in the power market, compare it with wind energy storage participating in the power market alone, analyze the differences in the real-time capacity of the two resources participating in the energy market and the frequency regulation market under the two methods, and obtain the net revenue of the two resources under the two scenarios.

[0091] The inventor considered that if the actual power generation of wind power is less than the declared capacity, it will bring a certain increase in the power generation cost of wind power, that is, the penalty cost; if the real-time power generation of wind power is over-generated, the electricity price of the over-generated part will be lower than the clearing price within the time period [t, t+Δt]. The same situation also exists in the frequency regulation service market where the frequency regulation capacity does not match the declared capacity. Considering that the proportion of new energy in the power system is increasing year by year, its uncertainty also brings a greater impact on the power grid, and the actual situation of energy storage batteries is fully considered in the revenue analysis, and the deviation penalty cost and battery loss cost are introduced into the bidding model to make the revenue accurate. The specific steps are as follows:

[0092] Step 1: Establish a revenue model for wind power and energy storage participating in the energy market

[0093] Both wind power and energy storage participate in the energy market transactions separately, so the revenue models of the two in the energy market are the same. The revenue of the resource at time t Including the day-ahead clearing revenue and the real-time clearing revenue:

[0094] In Equation (1): is the day-ahead clearing price of the electricity market; Δt is the time scale of day-ahead clearing; is the real-time clearing capacity of the resource at time τ. The first term is the day-ahead clearing revenue, and the second term is the real-time clearing revenue.

[0095] In the energy market, a deviation penalty cost is introduced to incentivize suppliers to provide the generated electricity that meets the declared capacity. The penalty cost at time t is the sum of the penalty costs within each Δτ period. That is:

[0096]

[0097]

[0098] Finally, the net revenue of participating in the energy market at time t is obtained by subtracting the penalty cost from the revenue where Δτ is the time scale of real-time clearing; is the over-generation penalty coefficient of the frequency regulation market; is the under-generation penalty coefficient of the frequency regulation market.

[0099]

[0100] When constructing the revenue model of wind power and energy storage participating in the energy market, considering the penalty cost can improve the frequency regulation performance of wind power and accurately represent the revenue of participating in the electricity market.

[0101] Step 2: Establish the revenue model of wind-storage joint participation in the frequency regulation market

[0102] Energy storage and wind power are complementary in terms of frequency regulation capacity and frequency regulation accuracy. Since the time period of AGC frequency regulation is much shorter than the time interval of real-time clearing, the control period of AGC is set as Δk (the period is 5 s), and k is the time node of AGC frequency regulation. When wind power is insufficient to meet the AGC frequency regulation command, energy storage will fill this part of the capacity gap. In this model, the key parameter is the capacity of the two resources of wind and storage participating in frequency regulation. Assuming that energy storage can meet the frequency regulation capacity shortage of wind power, the calculation is as follows:

[0103]

[0104] In Equation (5), is the capacity of the battery participating in wind-storage joint frequency regulation at time k when the control command is issued; is the control instruction at time k. If the AGC issues an upward frequency regulation instruction, then while it is -1 for a downward frequency regulation instruction; is the day-ahead cleared capacity participating in frequency regulation at time t; and are respectively the real-time upward and downward cleared capacities of wind power participating in frequency regulation at time τ.

[0105] The revenue of the virtual power plant participating in frequency regulation is finally obtained through the following formula:

[0106]

[0107]

[0108] where, and are respectively the day-ahead capacity clearing price and day-ahead mileage clearing price of the downward frequency regulation market at time t; and are respectively the real-time cleared capacity and day-ahead cleared capacity of the virtual power plant participating in frequency regulation; η is the mileage revenue factor. Equation (6) is for solving the revenue of the frequency regulation capacity, while equation (7) is for solving the revenue of the frequency regulation mileage. Among them, the frequency regulation performance index ρ has different values for different models. When wind power participates in frequency regulation alone, it belongs to A signal resources, ρ≈1, while when energy storage participates in the frequency regulation market alone, the frequency regulation performance index is ρ≈3. In the invention, the wind-storage combined frequency regulation has a slower response speed in terms of frequency regulation capacity but higher accuracy. Therefore, the performance index of the wind-storage combination takes ρ≈1 - 3.

[0109] Step 3: Establish the revenue model of energy storage participating in the frequency regulation market alone

[0110]

[0111]

[0112] In the formula: and are respectively the day-ahead and intra-day cleared capacities of energy storage participating in frequency regulation alone. The frequency regulation performance index is much higher than that of A signal resources, about 2 - 4 times that of A signal (ρ = 3 is taken in this invention). If the real-time cleared capacity of energy storage is less than its day-ahead cleared capacity, a penalty cost needs to be deducted, otherwise it is not required.

[0113]

[0114] Similar to the wind-storage combined participation in frequency regulation, in addition to the penalty cost, there is also the loss cost of energy storage, which will be described in the cost model of energy storage participating in the power market.

[0115] If the benefits of energy storage participating in the frequency regulation market alone are not considered, the benefits of energy storage will be reduced to a certain extent, which is inconsistent with the actual situation. Therefore, considering the scenario of energy storage participating in the frequency regulation market alone, on the one hand, resource participation scenarios can be flexibly combined to maximize the value of energy storage, and on the other hand, the benefits of energy storage participating in frequency regulation can be obtained.

[0116] Step 4: Build a cost model for energy storage to participate in the electricity market

[0117] The capacity cost per unit time of energy storage refers to the depreciation cost of allocating the capacity investment cost of energy storage to each real-time clearing cycle. The following calculation formula can be obtained through the workload algorithm.

[0118]

[0119] In formula (11), h capacity is the unit capacity cost of energy storage; C battery is the rated output power of the energy storage; r is the discount rate; T float The floating charge life of energy storage refers to the life of the energy storage battery under normal operation. It usually depends on the material and corrosion cost of the battery and can be considered as a constant.

[0120] definition The mileage cost of energy storage depends on the attenuation of battery life caused by the energy storage cycle depth d. The daily cycle number at different cycle depths is converted to the cycle number at 100% cycle depth.

[0121]

[0122] In formula (12): k p Usually provided by the battery manufacturer, it is a constant ranging from 0.8 to 2.1; is the number of daily cycles at discharge depth d. The loss cost of the battery is related to the charge and discharge state of the battery, because the cycle depth d depends on the difference between the charge and discharge states at two adjacent control moments.

[0123]

[0124] In the formula is the charge and discharge state of the energy storage battery at time k. The difference between the charge and discharge states of the two control cycles can be determined by the capacity of the energy storage participating in the upward and downward frequency modulation at time k, as shown in formula (14):

[0125]

[0126] Where ξ is the charging efficiency of the battery; is the charging and discharging power of the battery at time k, Greater than zero indicates that the battery is currently in the charging state, and vice versa for the discharging state. The charging and discharging power of the battery is affected by the energy market and the frequency regulation market. Therefore, the charging and discharging power is equivalent to the power of the energy storage participating in the energy market at time k and the power of the energy storage participating in the frequency regulation market at time k The sum of, and is determined by the capacity of the energy storage participating in the combined wind and energy storage frequency regulation at time k and the real-time cleared capacity of the energy storage at time τ As shown in Equation (15):

[0127]

[0128]

[0129] According to the above formula, the cycle depth d between two adjacent control times can be obtained k . For the Δk period, the battery only cycles once at the cycle depth d k And the state of the battery only has one state (charging or discharging), so its cycle times are half of the actual. From Equation (12), the cycle times with a cycle depth of 100% d in the time period [k, k + Δk] are Finally, the energy storage cost at time τ is obtained from formula (17)

[0130]

[0131] In Equation (17): H invest is the investment cost of the energy storage; N 100％ is the maximum cycle times at 100% d, where 100% d is a complete discharge cycle

[0132] Finally, adding the capacity cost and the mileage cost of the energy storage gives the loss cost of the energy storage

[0133]

[0134] When constructing the energy storage cost model, the inventor found that the loss cost of the battery can be divided into two aspects: capacity cost and mileage cost. The capacity cost depends on the construction cost of the energy storage, while the mileage cost depends on the operating loss life of the battery. The larger the capacity of the energy storage, the greater its loss cost will be. On the other hand, the frequency modulation mileage determines the cyclic discharge depth of the energy storage battery. The greater the charge-discharge depth, the faster the battery life is lost, and the higher the loss cost. If only the operating loss cost of the battery is considered, without considering the inherent loss of the energy storage, that is, the cost of the upfront investment allocated to each scheduling cycle, ignoring the capacity cost will lead to an overly optimistic bidding income for the energy storage under the existing conditions, with a certain deviation from the actual income. Therefore, when constructing the energy storage cost model, both aspects are considered to make the battery loss cost more accurate.

[0135] In the constructed electricity market bidding model, wind energy and energy storage first participate in the energy market transaction, that is, first obtain the real-time clearing capacity of the energy market, and then participate in the frequency modulation market. The participation of energy storage in the frequency modulation market can be divided into two parts: participating alone and participating jointly with wind power. Among them, the capacity of the energy storage participating in wind-storage joint frequency modulation can be obtained from Equation (5), while the capacity of the energy storage participating in the energy market and participating in frequency modulation alone is obtained from the bidding model based on optimal economic operation. Similarly, wind power first participates in the energy market transaction, and the remaining capacity participates in the frequency modulation market jointly with the energy storage.

[0136] The bidding models for wind power and energy storage to participate in the energy market and the frequency modulation market are essentially an optimization model of economic benefits. In this bidding model, there are two types of components that need to be optimized. One is the expected income, and the other is the total income. The expected income determines the day-ahead bidding capacity, and the day-ahead bidding capacity is the input variable for the optimization of the total income. The specific steps are as follows:

[0137] Step 5: Establish an optimization model for the expected income

[0138] The objective function for the optimization of the expected income is:

[0139]

[0140] In Equation (19): The first term is the day-ahead clearing income of wind power and energy storage participating in the energy market respectively, the second term is the day-ahead clearing income of wind-storage joint participation in the frequency modulation market, and the third term is the day-ahead income of the energy storage participating in the frequency modulation market alone; the finally deducted cost is the historical daily loss cost of the energy storage battery. It is the historical average daily loss cost at 100% d. Hour is a set containing 24 integral time points.

[0141] The constraint conditions must satisfy the capacity constraints, including the energy storage capacity constraint and the wind power capacity constraint:

[0142]

[0143]

[0144]

[0145] Where: C wind is the rated output power of the wind power.

[0146] Step 6: Establish an optimization model for the total revenue

[0147] The total revenue considers the sum of the net revenues of wind power and energy storage, which is equivalent to the sum of the revenues of the two resources participating in the electricity market minus their penalty costs and the loss costs of the energy storage.

[0148]

[0149] The constraint functions of the total revenue optimization model also need to satisfy the capacity constraints. In addition to the constraints on the day-ahead clearing capacity in equations (20), (21), and (22), the real-time clearing capacity and the frequency regulation capacities in the upward and downward directions also need to be restricted. Equations (24) and (25) are the restrictions on the sum of the capacities of wind power participating in the frequency regulation market and the energy market at time τ.

[0150]

[0151]

[0152] The ramping ability of wind power is limited, and its maximum upward and downward ramping abilities are limited to less than β% of the rated power of the wind power:

[0153]

[0154] The sum of the capacity of the energy storage participating in the energy market and the capacities in the upward and downward directions must be kept within its rated power, where k ∈ [τ, τ + Δτ]:

[0155]

[0156]

[0157] The energy storage battery needs to ensure that a certain remaining capacity (γ% of the frequency regulation capacity cleared during the day) is available for upward and downward frequency regulation. This constraint condition ensures that when the capacity of wind power participating in frequency regulation is too low due to external factors, the energy storage battery can make up for the part where the combined wind and energy storage real-time frequency regulation capacity is lower than the day-ahead clearing capacity.

[0158]

[0159] Where, C battery,stored is the rated capacity of the energy storage, k ∈ [τ, τ + Δτ].

[0160] The objective function of the electricity market bidding model is divided into two parts. The first part is used to solve the day-ahead bidding capacity, while the second part is used to solve the real-time bidding capacity. The result obtained by optimizing the day-ahead bidding and real-time bidding together is more accurate and more in line with practical applications. Because in the actual bidding process of the electricity market, the day-ahead bidding volume needs to be submitted one day in advance, and the exact value of the real-time bidding capacity cannot be known at the time of submission. In addition, if the day-ahead bidding volume and the actual bidding capacity are optimized together, in order to avoid deviations during the optimization process, the software will make the two types of bidding volumes very close through corresponding algorithms, which cannot reflect the advantages of the algorithm. Therefore, the method of optimizing the day-ahead and real-time bidding capacities separately is used to solve the impact of the above problems on the algorithm results.

[0161] Step 8: After constructing the electricity market bidding model, establish a robust optimization model, fully consider the uncertainty of wind power output and market price fluctuations, and adjust the fluctuation degree of uncertain parameters in the electricity market bidding model through the conservatism parameter, so as to effectively obtain the bidding strategies under different conservatism degrees.

[0162] In the total revenue optimization model, the existence of uncertain variables will lead to fluctuations in the values of decision variables and the optimal revenue. To reduce the impact of the fluctuations of uncertain variables on the optimization results, the objective functions of both are set as a max-min problem as shown in Equation (30). The improved robust optimization algorithm is used to solve the expected revenue model, the double-layer max-min model is transformed into a single-layer linear robust optimization model to quantify the impact of uncertain variables, and the use of the robust optimization algorithm is proposed.

[0163] For all linear programming functions containing uncertain variables, they can be transformed into the robust equivalent form as shown in (30):

[0164]

[0165] s.t.l≤x≤u (31)

[0166] where c is a parameter matrix independent of uncertain variables; a is a parameter matrix of uncertain variables; a i is the column vector of the i-th column corresponding to a certain uncertain variable, is the predicted value; x is the decision variable matrix.

[0167] Introduce an integer variable Λ to adjust the balance between the optimality of the solution of the robust optimization model result and the system robustness, then Equation (30) changes to the following equation:

[0168]

[0169] where: J is the set of uncertain variables; S is the set of J, representing the set of the expected values of uncertain variables.

[0170] Using the duality principle, the above model is converted into an equivalent mixed-integer linear optimization model (NP-hard) as follows:

[0171]

[0172] In Equation (33): z, p i , y i are optimization variables; the value range of Λ is Λ ∈ [0, |J|]. When Λ = 0, the model is equivalent to the linear programming model without considering uncertainty factors composed of Equations (30)-(31); when Λ = |J|, the robustness degree of the model is the highest, which means the solution obtained is the most conservative and the optimality of the solution is relatively the worst. Reasonably setting the value of Λ can effectively adjust the balance between the optimality of the solution and the system robustness.

[0173] In the total revenue optimization model, the set of decision variables in the objective function of the total revenue is Similar to the predicted revenue model, the input variables also include the real-time price of the market and the real-time output of wind power. In the real-time market, a higher precision requirement for the input variables is needed, so their uncertainty needs to be considered. Define as the set of uncertain variables, including the clearing price of the energy market the clearing price of the frequency regulation capacity the clearing price of the frequency regulation mileage and the real-time output of wind power Assume that the value range of the above uncertain variables fluctuates symmetrically above and below the predicted value, then the constraints shown in Equation (34) are satisfied:

[0174]

[0175] where: is the predicted value of the uncertain variable, and Δc is the fluctuation amount of the predicted value.

[0176] According to the robust optimization model, the total revenue optimization model is updated, and the objective function and constraint conditions are updated as shown in Equations (35)-(36).

[0177]

[0178] where: is the predicted value of the day-ahead clearing price of the power market at time t; is the revenue of the battery at time t; is the revenue of wind power at time t; is the penalty cost at time t; is the predicted value of the clearing price of the energy market; is the predicted value of the clearing price of the frequency regulation capacity; is the capacity revenue for the combined wind and energy storage participating in the frequency regulation market; is the capacity revenue for the energy storage participating in the frequency regulation market alone; is the penalty cost at time t; is the predicted value of the day-ahead capacity clearing price in the frequency regulation market at time t; is the predicted value of the day-ahead mileage clearing price in the frequency regulation market at time t; is the mileage revenue for the combined wind and energy storage participating in the frequency regulation market; is the mileage revenue for the energy storage participating in the frequency regulation market alone; is the predicted value of the number of cycles in the time period [k, k+Δk] under a full discharge cycle depth; is the energy storage cost at time τ; is the capacity cost per unit time of the energy storage; z1 is the first type of optimization parameter corresponding to the clearing price in the energy market; Λ1 is the second type of optimization parameter corresponding to the clearing price in the energy market; is the third type of optimization parameter corresponding to the clearing price in the energy market; z2 is the first type of optimization parameter corresponding to the clearing price of the frequency regulation capacity; Λ2 is the second type of optimization parameter corresponding to the clearing price of the frequency regulation capacity; is the third type of optimization parameter corresponding to the clearing price of the frequency regulation capacity; z3 is the first type of optimization parameter corresponding to the clearing price of the frequency regulation mileage; Λ3 is the second type of optimization parameter corresponding to the clearing price of the frequency regulation mileage; is the third type of optimization parameter corresponding to the clearing price of the frequency regulation mileage; Hour is the set containing 24 integer hours.

[0179] The constraint conditions of the objective function (35) are:

[0180]

[0181] In the formula: λ1 is the first dual multiplier; λ2 is the second dual multiplier; λ3 is the third dual multiplier; λ4 is the fourth dual multiplier, Λ wind is the conservatism parameter of the real-time wind power output; y1 is the fourth type of optimization parameter corresponding to the clearing price in the energy market; y2 is the fourth type of optimization parameter corresponding to the clearing price of the frequency regulation capacity; y3 is the fourth type of optimization parameter corresponding to the clearing price of the frequency regulation mileage; is the real-time clearing capacity of the combined wind and energy storage participating in the frequency regulation market at time τ; is the real-time clearing capacity of the wind power participating in the energy market at time τ; is the real-time clearing capacity of the energy storage participating in the energy market at time τ; is the real-time clearing capacity of the wind power participating in upward frequency regulation at time τ; is the output of the wind farm at time τ. Formulas (20)-(22), (25)-(29) do not need to be updated.

[0182] Due to the output of wind power, the prices in each market are uncertain. The robust optimization model can well describe this kind of uncertain random change, and the conservatism parameter can adjust the fluctuation range of each uncertain variable according to the situation of the trading day.

[0183] Step 9: Set parameters for relevant equipment such as energy storage, verify the bidding model for wind and energy storage participating in the electricity market jointly, compare it with the situation where wind and energy storage participate in the electricity market separately, analyze the differences in the real-time capacity of the two resources participating in the energy market and the frequency regulation market under the two methods, and obtain the net benefits of the two resources under the two scenarios.

[0184] First, construct the bidding scenario and set parameters for relevant equipment such as energy storage. In the present invention, two power stations participate in the bidding for the energy market and the frequency regulation market. The energy storage power station is composed of 30MW sodium-sulfur batteries, k p = 1.3, and the specific parameters are shown in Table 1;

[0185] Table 1 Basic parameter settings of the energy storage power station composed of lithium batteries

[0186]

[0187] As Figure 4 shown, the wind power station uses the actual power generation data of a 200MW wind power station in a certain regional power grid on a certain day, and the electricity market data uses the actual trading data of PJM on a certain day in 2017. To exclude the interference of other factors, the dates of the wind power generation data and the trading data are taken on the same day.

[0188] The experiment is mainly carried out in two scenarios; Scenario 1: Wind and energy storage participate in the electricity market transaction separately; Scenario 2: Wind and energy storage participate in the electricity market transaction under multiple operation modes. The optimization goal of Scenario 1 is to obtain the optimum of the sum of the economic benefits of energy storage and wind power; while Scenario 2 is optimized through the joint bidding model proposed in this embodiment. Since the uncertainty of wind power output and market price is considered in the optimization process, the optimization results of the wind and energy storage joint bidding model under different conservatism parameters are obtained in the simulation. Figure 5 is the real-time declared capacity for wind and energy storage participating in the electricity market separately, divided into 96 time periods; Figure 6 is the real-time cleared capacity of each resource for wind and energy storage participating in the electricity market jointly; Figure 7 is the real-time declared capacity of each resource for wind and energy storage participating in the electricity market jointly under different wind power conservatism parameters (10%).

[0189] According to the real-time capacity in the two cases, through the net benefit calculation method in Steps 1-8, the net benefit within the trading day is finally obtained. Table 2 compares the net benefit values of wind power and energy storage when wind and energy storage participate in the electricity market jointly and when wind and energy storage participate in the electricity market independently.

[0190] Table 2 Net benefits of wind and energy storage participating in the electricity market under different scenarios

[0191]

[0192] Comparison Figure 5 and Figure 6 of the data, the following conclusions can be obtained:

[0193] ① From Figure 5 and Figure 6 , it can be seen that the sum of the real-time wind power capacities participating in the energy market and the frequency regulation market changes with the Figure 4 wind power output curve of the day. During the time periods with relatively high frequency regulation prices (the sum of the mileage price and the capacity price), the proportion of the frequency regulation capacity of wind power / wind-storage combination is also relatively high; in order to reduce the large penalty cost caused by under-generation, the bidding model will slightly increase the capacity of wind power participating in frequency regulation alone to avoid this situation, while the accuracy of wind-storage combined frequency regulation is relatively high, and there is no need to take measures of over-generation to reduce the penalty cost. During the time periods from 1:00 to 2:00, from 6:00 to 8:00, and from 19:00 to 21:00, Figure 5 the capacity of wind power participating in frequency regulation alone in Figure 6 is slightly larger than the capacity of wind-storage combination participating in frequency regulation in Figure 5 , about 10%. When the frequency regulation price is relatively low, the bidding model prefers the energy market, and the declared capacity of wind power participating in the frequency regulation market will be further reduced to reduce the excessive capacity invested in the frequency regulation market to suppress its output fluctuation, thereby reducing the penalty cost. During the time periods from 3:00 to 6:00 and from 15:00 to 17:00, Figure 6 the capacity of wind power participating in frequency regulation alone in

[0194] is much smaller than the capacity of wind-storage combination participating in frequency regulation in Figure 5 and Figure 6 ② Due to the constraint of formula (29), the energy storage must reserve a part of its capacity to prevent large-scale wind power; when the wind and energy storage participate in the power market alone, there is no need to reserve backup capacity, so the capacity of the energy storage participating in the energy market and the frequency regulation market in each time period has a certain degree of increase compared with the case of wind-storage combined participation in the power market. In addition, by comparing the capacity of the energy storage participating in the power market between

[0195] Setting different conservativeness also affects the declared capacity of wind-storage participating in the energy-frequency regulation market. The higher the conservativeness, the worse the economy of the optimization result. Comparing Figure 6 and Figure 7It can be obtained that: due to the increase in the wind power conservatism parameter, the robust optimization model will reduce the declared capacity of wind power resources in each market to reduce the penalty cost caused by the fluctuation of wind power output. During the entire period, the participation of wind power in the energy market and the declared capacity decrease with the increase in the wind power conservatism parameter, about 4%; the sum of the declared capacities of energy storage participating in the power market decreases to make up for the impact of the increase in the wind power conservatism parameter on the declared capacity of wind energy storage participating in the frequency regulation market, decreasing by about 5%; the declared capacity of wind energy storage participating in the power market remains almost unchanged.

[0196] The net income results of wind energy storage jointly participating in the power market are shown in Table 2. Analyzing Table 2, the following conclusions can be obtained:

[0197] ① Under the bidding model proposed in the present invention, the total income growth mainly comes from the frequency regulation market, about 12.8%. Since the resources composed of energy storage and wind power have high accuracy, the bidding model prefers the frequency regulation market more during the program optimization process, so the proportion of the energy market decreases significantly.

[0198] ② In the case of wind energy storage combination, the net income of participating in frequency regulation increases by 12.3%, while the income of participating in the energy market decreases by 6.7%, indicating that the bidding model of wind energy storage jointly participating in the power market effectively improves the frequency regulation performance of wind power and the income of participating in the power market, and to a certain extent reduces the loss cost of energy storage.

[0199] A virtual power plant composed of wind power and energy storage provides power generation and frequency regulation services for the power market. A real-time bidding scheme including wind power and energy storage is proposed under different time scales of the day-ahead market and the real-time market. By using the complementary characteristics of the two, the income of wind power and energy storage participating in the power market is maximized, and an income model under different scenarios is established. In addition, in the income analysis, the actual situation is fully considered, and the deviation penalty cost and battery loss cost are introduced into the bidding model to make the income accurate. Finally, based on the economic optimal objective function, a bidding model of wind energy storage combination is established, and the net income of wind energy storage participating in the power market under different scenarios is obtained.

[0200] In the embodiment of the present invention, combined with the complementary characteristics of energy storage and wind power in terms of frequency regulation capacity and frequency regulation accuracy, the income models including penalty costs in four scenarios are analyzed, namely, wind power / energy storage participating in energy market transactions, wind energy storage jointly participating in frequency regulation market transactions, and energy storage alone participating in frequency regulation market transactions. Then, the loss cost caused by energy storage participating in the power market is analyzed; then, based on the economic optimal objective function, the expected income and total income bidding models of wind energy storage combination are established respectively. The expected income determines the day-ahead bidding capacity, and the day-ahead bidding capacity is the input variable for the total income optimization. Through the total income optimization, the real-time clearing capacity and total income are obtained. Give full play to the complementary advantages of wind power and battery energy storage in terms of frequency regulation accuracy and durability, as a reference for energy merchants when bidding.

[0201] To account for the deviation of the bidding capacity caused by uncertain variables, a robust optimization model with a degree of conservatism is also established, which can set different degrees of conservatism according to the trading day conditions to simulate the actual wind power, actual market price, and obtain a more reasonable bidding capacity.

[0202] Embodiment 2: Based on the same inventive concept, the embodiment of the present invention further provides an optimal bidding system for a wind energy storage system to participate in the energy-frequency regulation market, including:

[0203] An acquisition module, configured to acquire the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market;

[0204] A result module, configured to input the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market into a pre-constructed power market bidding model for calculation, and obtain the capacity of the wind power and / or the energy storage to participate in each scenario;

[0205] Wherein, the scenarios include the wind power participating in the energy market, the energy storage participating in the energy market, the wind energy storage system jointly participating in the frequency regulation market, and the energy storage participating in the frequency regulation market;

[0206] The power market bidding model is constructed with the goal of maximizing the revenue when the wind power and / or the energy storage participate in each scenario.

[0207] In the embodiment, the system further includes a module for constructing a power market bidding model, specifically for:

[0208] Based on the day-ahead clearing revenue of the wind power participating in the energy market, the day-ahead clearing revenue of the energy storage participating in the energy market, the day-ahead clearing revenue of the wind energy storage system jointly participating in the frequency regulation market, the day-ahead revenue of the energy storage participating in the frequency regulation market alone, and the historical daily loss cost of the energy storage, construct an expected revenue optimal objective function with the goal of maximizing the expected revenue;

[0209] Based on the expected revenue optimal objective function and the constraint conditions constructed for the expected revenue optimal objective function, obtain the relationship between the energy clearing price, the frequency regulation mileage clearing price, the frequency regulation capacity clearing price and the day-ahead bidding capacity;

[0210] Based on the net revenue of the wind power and the energy storage participating in the energy market, the frequency regulation capacity revenue of the wind energy storage system jointly participating in the frequency regulation market, the frequency regulation mileage revenue of the wind energy storage system jointly participating in the frequency regulation market, the day-ahead revenue of the energy storage participating in the frequency regulation market alone, the intraday revenue of the energy storage participating in the frequency regulation market alone, the penalty cost, and the loss cost of the energy storage, as well as the day-ahead bidding capacity, construct a total revenue optimal objective function with the goal of maximizing the total revenue;

[0211] Based on the total revenue optimal objective function and the constraint conditions constructed for the total revenue optimal objective function, the relationships between the energy clearing price, the frequency regulation mileage clearing price, the frequency regulation capacity clearing price and the capacities of wind power participating in the energy market, energy storage participating in the energy market, wind-storage combined participating in the frequency regulation market, energy storage participating in wind-storage combined frequency regulation and energy storage participating in the frequency regulation market alone are obtained.

[0212] In the embodiment, the expected revenue optimal objective function is shown as the following formula:

[0213]

[0214] In the formula: is the day-ahead clearing price of the power market at time t; is the day-ahead clearing capacity of energy storage participating in the energy market at time t; is the day-ahead clearing capacity of wind power participating in the energy market at time t; Δt is the time scale of day-ahead clearing; is the day-ahead clearing capacity of wind-storage combined participating in frequency regulation; ρ is the frequency regulation performance index; is the day-ahead capacity clearing price of the frequency regulation market at time t; η is the mileage revenue factor; is the day-ahead mileage clearing price of the frequency regulation market at time t; is the day-ahead clearing capacity of wind-storage combined participating in frequency regulation; H invest is the investment cost of energy storage; N 100％ is the maximum number of cycles in a complete discharge cycle; is the historical average daily loss cost in a complete discharge cycle; Hour is a set containing 24 integral time points.

[0215] In the embodiment, the total revenue optimal objective function is shown as the following formula:

[0216]

[0217] In the formula: is the net revenue of wind power and energy storage participating in the energy market; is the capacity revenue of wind-storage combined participating in the frequency regulation market; is the mileage revenue of wind-storage combined participating in the frequency regulation market; is the capacity revenue of energy storage participating in the frequency regulation market alone; is the mileage revenue of energy storage participating in the frequency regulation market alone; is the penalty cost at time t; is the loss cost of energy storage.

[0218] In the embodiment, the penalty cost is shown as the following formula:

[0219]

[0220] In the formula: is the under-generation penalty coefficient of the frequency modulation market; is the day-ahead clearing capacity of the energy storage participating in frequency modulation alone at time t; is the real-time clearing capacity of the energy storage participating in frequency modulation alone at time τ.

[0221] In the embodiment, the loss cost of the energy storage is shown in the following formula:

[0222]

[0223] In the formula: is the capacity cost per unit time of the energy storage; is the energy storage cost at time τ;

[0224] Among them, the capacity cost per unit time of the energy storage is shown in the following formula:

[0225]

[0226] In the formula: h capacity is the unit capacity cost of the energy storage; C battery is the rated output power of the energy storage; r is the discount rate; T float is the floating charge life of the energy storage;

[0227] The energy storage cost at time τ is shown in the following formula:

[0228]

[0229] In the formula: H invest is the investment cost of the energy storage; N 100％ is the maximum number of cycles in a complete discharge cycle; is the number of cycles in the time period [k, k + Δk] at a cycle depth of a complete discharge cycle.

[0230] In the embodiment, the number of cycles in a complete discharge cycle is calculated according to the following formula:

[0231]

[0232] In the formula: is the daily number of cycles at a discharge depth of d; d is the cycle depth; k p is a constant;

[0233] Among them, the cycle depth d is calculated according to the following formula:

[0234]

[0235] Where: d k is the cycle depth of energy storage within two adjacent control instants; is the charge and discharge state of the energy storage battery at time k; ξ is the charging efficiency of the battery; is the charging and discharging power of the battery at time k, which is equal to the power of the energy storage participating in the energy market at time k and the power of the energy storage participating in the frequency regulation market at time k The sum, where is determined by the capacity of the energy storage participating in the wind - energy storage combined frequency regulation at time k and the real - time clearing capacity of the energy storage at time τ decides.

[0236] In the embodiment, the capacity revenue of the energy storage participating in the frequency regulation market alone is calculated according to the following formula:

[0237]

[0238] Where: ρ is the value of the frequency regulation performance index; is the day - ahead capacity clearing price of the frequency regulation market at time t; is the intra - day clearing capacity of the energy storage participating in the frequency regulation alone at time τ; is the day - ahead clearing capacity of the energy storage participating in the frequency regulation alone at time t; Δt is the time scale of day - ahead clearing; Δτ is the time scale of real - time clearing.

[0239] In the embodiment, the mileage revenue of the energy storage participating in the frequency regulation market alone is calculated according to the following formula:

[0240]

[0241] Where: ρ is the value of the frequency regulation performance index; η is the mileage revenue factor; is the day - ahead mileage clearing price of the frequency regulation market at time t; is the real - time clearing capacity of the energy storage participating in the frequency regulation alone at time τ; is the day - ahead clearing capacity of the energy storage participating in the frequency regulation alone at time t.

[0242] In the embodiment, the capacity revenue of the wind - energy storage combined participation in the frequency regulation market is calculated according to the following formula:

[0243]

[0244] Where: ρ is the value of the frequency regulation performance index; is the day - ahead capacity clearing price of the frequency regulation market at time t; is the real-time clearing capacity of the wind-storage combination participating in the frequency regulation market at time τ; is the day-ahead clearing capacity of the wind-storage combination participating in the frequency regulation at time t.

[0245] In the embodiment, the mileage revenue of the wind-storage combination participating in the frequency regulation market is calculated according to the following formula:

[0246]

[0247] In the formula: η is the mileage revenue factor; is the day-ahead mileage clearing price of the frequency regulation market at time t.

[0248] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0249] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0250] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0251] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Therefore, the instructions executed on the computer or other programmable device provide for implementing the functions in the processFigure 1 One process or multiple processes and / or boxes Figure 1 Steps of the functions specified in one box or multiple boxes

[0252] The above are only embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention are included within the scope of the claims of the present invention pending approval of the application.

Claims

1. An optimal bidding method for wind - storage combined participation in the energy - frequency regulation market, characterized in that, Including: Obtain the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market; Input the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market into a pre-constructed power market bidding model for calculation to obtain the capacities of wind power and / or energy storage participating in each scenario; Wherein, the scenarios include wind power participating in the energy market, energy storage participating in the energy market, wind-storage combined participating in the frequency regulation market, and energy storage participating in the frequency regulation market; The power market bidding model is constructed with the goal of maximizing the revenue when wind power and / or energy storage participate in each scenario; The construction of the power market bidding model includes: Based on the day-ahead clearing revenue of wind power participating in the energy market, the day-ahead clearing revenue of energy storage participating in the energy market, the day-ahead clearing revenue of wind-storage combined participating in the frequency regulation market, the day-ahead revenue of energy storage participating in the frequency regulation market alone, and the historical daily loss cost of the energy storage, construct an expected revenue optimal objective function with the goal of maximizing the expected revenue; Based on the expected revenue optimal objective function and the constraint conditions constructed for the expected revenue optimal objective function, obtain the relationships between the energy clearing price, the frequency regulation mileage clearing price, the frequency regulation capacity clearing price and the day-ahead bidding capacity; Based on the net revenue of wind power and energy storage participating in the energy market, the frequency regulation capacity revenue of wind-storage combined, the frequency regulation mileage revenue of wind-storage combined, the day-ahead revenue of energy storage participating in the frequency regulation market alone, the intraday revenue of energy storage participating in the frequency regulation market alone, the penalty cost, and the loss cost of the energy storage, as well as the day-ahead bidding capacity, construct a total revenue optimal objective function with the goal of maximizing the total revenue; Based on the total revenue optimal objective function and the constraint conditions constructed for the total revenue optimal objective function, obtain the relationships between the energy clearing price, the frequency regulation mileage clearing price, the frequency regulation capacity clearing price and the capacities of wind power participating in the energy market, energy storage participating in the energy market, wind-storage combined participating in the frequency regulation market, energy storage participating in wind-storage combined frequency regulation, and energy storage participating in the frequency regulation market alone.

2. The method according to claim 1, characterized in that, The expected revenue optimal objective function is shown as follows: Wherein: is the day-ahead clearing price of the electricity market at time t; is the day-ahead clearing capacity of the energy storage participating in the energy market at time t; is the day-ahead clearing capacity of the wind power participating in the energy market at time t; Δt is the time scale of the day-ahead clearing; is the day-ahead clearing capacity of the combined wind and energy storage participating in frequency regulation; ρ is the frequency regulation performance index; is the day-ahead capacity clearing price of the frequency regulation market at time t; η is the mileage benefit factor; is the day-ahead mileage clearing price of the frequency modulation market at time t; H is the day-ahead clearing capacity of wind and storage units participating in frequency regulation; invest is the investment cost of energy storage; N 100％ It is the maximum number of cycles under a complete discharge cycle; is the historical average daily loss cost under a complete discharge cycle; Hour is a set of 24 hourly moments.

3. The method according to claim 1, characterized in that, The total revenue optimal objective function is shown as follows: Wherein: is the net revenue of wind power and energy storage participating in the energy market; is the capacity revenue of wind-storage combined participating in the frequency regulation market; is the mileage revenue of wind-storage combined participating in the frequency regulation market; is the capacity revenue of energy storage participating in the frequency regulation market alone; is the mileage revenue of energy storage participating in the frequency regulation market alone; is the penalty cost at time t; is the loss cost of energy storage.

4. The method according to claim 3, characterized in that, The penalty cost is as shown in the following formula: Wherein: is the under-generation penalty coefficient of the frequency regulation market; is the day-ahead clearing capacity of the energy storage participating in frequency regulation alone at time t; is the real-time clearing capacity of the energy storage participating in frequency regulation alone at time τ.

5. The method according to claim 3, characterized in that, The loss cost of the energy storage As shown in the following formula: Where: is the capacity cost of the energy storage per unit time; is the energy storage cost at time τ; Among them, the capacity cost of the energy storage per unit time is shown in the following formula: Where: h capacity is the unit capacity cost of energy storage; C battery is the rated output power of energy storage; r is the discount rate; T float is the floating charge life of energy storage; The energy storage cost at time τ As shown in the following formula: Where: H invest is the investment cost of energy storage; N 100％ is the maximum number of cycles in a complete discharge cycle; is the number of cycles in the time period [k, k + Δk] at a cycle depth of a complete discharge cycle.

6. The method according to claim 5, characterized in that, The number of cycles at a cycle depth of one complete discharge cycle Calculated according to the following formula: Wherein: is the daily cycle number at the depth of discharge d; d is the cycle depth; k p is a constant; Wherein, the loop depth d is calculated as follows: Where: d k is the cycle depth of energy storage within two adjacent control instants; is the charge and discharge state of the energy storage battery at time k; ξ is the charging efficiency of the battery; is the charge and discharge power of the battery at time k, which is equal to the power of the energy storage participating in the energy market at time k and the power of the energy storage participating in the frequency regulation market at time k The sum, where is determined by the capacity of the energy storage participating in wind-storage combined frequency regulation at time k and the real-time cleared capacity of the energy storage at time τ decides.

7. The method according to claim 3, characterized in that The capacity revenue of the energy storage participating in the frequency regulation market alone Is calculated according to the following formula: Where: ρ is the value of the frequency modulation performance index; is the day-ahead capacity clearing price of the frequency modulation market at time t; is the intraday clearing capacity of the energy storage participating in frequency modulation alone at time τ; is the day-ahead clearing capacity of the energy storage participating in frequency modulation alone at time t; Δt is the time scale of day-ahead clearing; Δτ is the time scale of real-time clearing.

8. The method according to claim 3, characterized in that The mileage revenue of the energy storage participating in the frequency regulation market alone Is calculated according to the following formula: Where: ρ is the value of the frequency modulation performance index; η is the mileage revenue factor; is the day-ahead mileage clearing price of the frequency modulation market at time t; is the real-time clearing capacity of the energy storage participating in frequency modulation alone at time τ; is the day-ahead clearing capacity of the energy storage participating in frequency modulation alone at time t.

9. The method according to claim 3, characterized in that The capacity revenue of the combined wind and energy storage participating in the frequency regulation market It is calculated according to the following formula: Where: ρ is the value of the frequency modulation performance index; is the day-ahead capacity clearing price of the frequency modulation market at time t; is the real-time clearing capacity of the wind-storage combination participating in the frequency modulation market at time τ; is the day-ahead clearing capacity of the wind-storage combination participating in frequency modulation at time t.

10. The method according to claim 9, characterized in that The mileage revenue of the combined wind and energy storage participating in the frequency regulation market It is calculated according to the following formula: Where: η is the mileage revenue factor; is the day-ahead mileage clearing price in the frequency regulation market at time t.

11. The method according to claim 1, characterized in that After obtaining the total revenue optimal objective function and the constraint conditions constructed for the total revenue optimal objective function, it further includes: Use a robust optimization model to update the uncertainty variables in the total revenue optimal objective function and the constraint conditions constructed for the total revenue optimal objective function; The uncertainty variables include the energy market clearing price, the frequency regulation capacity clearing price, the frequency regulation mileage clearing price, and the real-time output of wind power.

12. The method according to claim 11, characterized in that The updated total revenue optimal objective function is shown as follows: Wherein: is the predicted value of the day-ahead clearing price of the power market at time t; is the revenue of the battery at time t; is the revenue of the wind power at time t; is the penalty cost at time t; is the predicted value of the clearing price of the energy market; is the predicted value of the clearing price of the frequency regulation capacity; is the capacity revenue of the wind-storage combined participation in the frequency regulation market; is the capacity revenue of the energy storage participating in the frequency regulation market alone; is the penalty cost at time t; is the predicted value of the day-ahead capacity clearing price of the frequency regulation market at time t; is the predicted value of the day-ahead mileage clearing price of the frequency regulation market at time t; is the mileage revenue of the wind-storage combined participation in the frequency regulation market; is the mileage revenue of the energy storage participating in the frequency regulation market alone; is the predicted value of the number of cycles in the time period [k, k+Δk] under a complete discharge cycle depth; is τ the energy storage cost at time; is the capacity cost per unit time of the energy storage; z1 is the first type of optimization parameter corresponding to the clearing price of the energy market; Λ1 is the second type of optimization parameter corresponding to the clearing price of the energy market; is the third type of optimization parameter corresponding to the clearing price of the energy market; z2 is the first type of optimization parameter corresponding to the clearing price of the frequency regulation capacity; Λ2 is the second type of optimization parameter corresponding to the clearing price of the frequency regulation capacity; is the third type of optimization parameter corresponding to the clearing price of the frequency regulation capacity; z3 is the first type of optimization parameter corresponding to the clearing price of the frequency regulation mileage; Λ3 is the second type of optimization parameter corresponding to the clearing price of the frequency regulation mileage; is the third type of optimization parameter corresponding to the clearing price of the frequency regulation mileage; Hour is a set containing 24 whole-point times.

13. The method according to claim 12, characterized in that The updated constraint conditions are shown as follows: Where: λ1 is the first dual multiplier; λ2 is the second dual multiplier; λ3 is the third dual multiplier; λ4 is the fourth dual multiplier, Λ wind is the conservatism parameter of the real-time wind power output; y1 is the fourth type of optimization parameter corresponding to the clearing price of the energy market; y2 is the fourth type of optimization parameter corresponding to the clearing price of the frequency regulation capacity; y3 is the fourth type of optimization parameter corresponding to the clearing price of the frequency regulation mileage; is the real-time clearing capacity of the wind-storage joint participation in the frequency regulation market at time τ; is the real-time clearing capacity of the wind power participating in the energy market at time τ; is the real-time clearing capacity of the energy storage participating in the energy market at time τ; is the real-time clearing capacity of the wind power participating in upward frequency regulation at time τ; is the output of the wind farm at time τ.

14. An optimal bidding system for wind and energy storage to jointly participate in the energy-frequency regulation market, characterized in that Including: An acquisition module for obtaining the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market; A result module, configured to input the predicted output of the wind farm and the energy storage capacity for each trading period, as well as the clearing prices of the energy market and the frequency regulation market, into a pre-constructed electricity market bidding model for calculation, so as to obtain the capacities of the wind power and / or energy storage participating in each scenario; Wherein, the scenarios include wind power participating in the energy market, energy storage participating in the energy market, wind-storage combination participating in the frequency regulation market, and energy storage participating in the frequency regulation market; The electricity market bidding model is constructed with the goal of maximizing the revenue when the wind power and / or energy storage participate in each scenario; The system further includes a module for constructing an electricity market bidding model, specifically configured to: Based on the day-ahead clearing revenue of wind power participating in the energy market, the day-ahead clearing revenue of energy storage participating in the energy market, the day-ahead clearing revenue of wind-storage combination participating in the frequency regulation market, the day-ahead revenue of energy storage participating in the frequency regulation market alone, and the historical daily loss cost of the energy storage, construct an expected revenue optimal objective function with the goal of maximizing the expected revenue; Based on the expected revenue optimal objective function and the constraint conditions constructed for the expected revenue optimal objective function, obtain the relationships between the energy clearing price, the frequency regulation mileage clearing price, the frequency regulation capacity clearing price and the day-ahead bidding capacity; Based on the net revenue of wind power and energy storage participating in the energy market, the revenue of wind-storage combination participating in the frequency regulation capacity, the revenue of wind-storage combination participating in the frequency regulation mileage, the day-ahead revenue of energy storage participating in the frequency regulation market alone, the intra-day revenue of energy storage participating in the frequency regulation market alone, the penalty cost, the loss cost of the energy storage, and the day-ahead bidding capacity, construct a total revenue optimal objective function with the goal of maximizing the total revenue; Based on the total revenue optimal objective function and the constraint conditions constructed for the total revenue optimal objective function, obtain the relationships between the energy clearing price, the frequency regulation mileage clearing price, the frequency regulation capacity clearing price and the capacities of wind power participating in the energy market, energy storage participating in the energy market, wind-storage combination participating in the frequency regulation market, energy storage participating in the wind-storage combination frequency regulation, and energy storage participating in the frequency regulation market alone.

Images