A day-ahead dispatching method for a twin-tower gravity energy storage power station considering uncertainty

Abstract

The present invention discloses a method for day-ahead scheduling of a twin-tower gravity energy storage power station taking uncertainty into consideration, comprising the following steps: obtaining actual twin-tower gravity energy storage system parameters; constructing a twin-tower gravity energy storage day-ahead scheduling model according to system parameters when the day-ahead electricity price is known, solving the model to obtain an optimized scheduling scheme for the gravity energy storage system; scheduling according to the scheme; constructing a probability distribution robust day-ahead scheduling model for gravity energy storage power stations according to system parameters when the day-ahead electricity price is unknown, solving the model to obtain a probability distribution robust day-ahead scheduling scheme for gravity energy storage power stations, and scheduling according to the scheme. The present invention enables the twin-tower gravity energy storage system to follow the electricity price signal to move bricks to achieve arbitrage purposes in day-ahead scheduling, and even in the case of the worst probability distribution of electricity prices, it can maximize the system's expectation of obtaining revenue in the day-ahead electricity market, thereby enabling the twin-tower gravity energy storage to better exert its energy storage potential.

Description

Technical Field

[0001] The present invention belongs to the technical field of efficient energy storage in power systems, and in particular to the problem of day-ahead scheduling of a twin-tower gravity energy storage power station, and to a day-ahead scheduling method of a twin-tower gravity energy storage power station taking uncertainty into consideration. Background Art

[0002] The increase in the penetration rate of renewable energy has brought huge challenges to the dispatching and operation of the power system. The reasonable configuration of energy storage can hedge the uncertainty of renewable energy, but traditional energy storage has certain limitations, such as high unit cost, harsh construction conditions and low efficiency. As a new type of mechanical energy storage technology, double-tower gravity energy storage has excellent regulation performance, no restrictions on the geographical environment, and low unit power generation cost. Its physical structure meets the safety indicators of mechanical design. The moving bricks are in a symmetrical state during operation, and the crane arm at the top of the high truss is in a state of force balance and torque balance at any time. The overall architecture is highly safe; based on modular design, its rated capacity can be expanded exponentially to meet the capacity requirements of distributed to large-scale energy storage power stations; the design and installation are flexible, and the internal modules are adjustable, and the specified capacity can be deployed in any area as required.

[0003] Twin-tower gravity energy storage has so many advantages that it will definitely be valued and built on a large scale in the future. Therefore, the dispatching system of twin-tower gravity energy storage becomes particularly important. At present, there is no dispatching method for twin-tower gravity energy storage. The traditional energy storage dispatching system lacks modeling of the characteristics of twin-tower gravity energy storage. Twin-tower gravity energy storage still lacks a practical and effective dispatching system, and cannot fully call on the application potential of twin-tower gravity energy storage. Summary of the invention

[0004] In order to overcome the problems in the prior art, the purpose of the present invention is to provide a day-ahead scheduling method for a twin-tower gravity energy storage power station taking into account uncertainty. The method can guide the scheduling and operation of the twin-tower gravity energy storage system, and help it to follow the electricity price signal to move bricks to achieve arbitrage purposes in day-ahead scheduling. Even under the worst probability distribution of electricity prices, the system's expectation of obtaining revenue in the day-ahead electricity market can be maximized, thereby enabling the twin-tower gravity energy storage to better exert its energy storage potential.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A day-ahead dispatching method for a double-tower gravity energy storage power station considering uncertainty comprises the following steps:

[0007] S1. Obtaining actual parameters of the double-tower gravity energy storage system;

[0008] S2. Under the condition of known day-ahead electricity price, a day-ahead dispatching model of the double-tower gravity energy storage is constructed according to the actual parameters of the double-tower gravity energy storage system, and the optimal dispatching scheme of the gravity energy storage system is obtained by solving the model; the dispatching is performed according to the scheme;

[0009] When the day-ahead electricity price is unknown, a probabilistic blue-stick day-ahead dispatching model of the gravity energy storage power station is constructed according to the actual parameters of the double-tower gravity energy storage system. The probabilistic blue-stick day-ahead dispatching scheme of the gravity energy storage power station is obtained and dispatched according to the scheme.

[0010] A further improvement of the present invention is that the actual parameters of the double-tower gravity energy storage system include the number of bricks per layer of tower a / tower b X a / X b ; Bottom area / volume / mass per unit brick S unit / V unit / m unit ;Minimum / maximum volume of tower a b tower minimum / maximum volume Initial volume of tower a The maximum volume of bricks raised / lowered per unit time Actual charging / discharging efficiency η of a double-tower gravity energy storage system ch / η dis ; Maximum charge / discharge power Maximum height of tower a Operation and maintenance cost per unit maximum charge and discharge power ρ om .

[0011] A further improvement of the present invention is that the objective function of the day-ahead scheduling model of the double-tower gravity energy storage is as follows:

[0012]

[0013] Where:

[0014] —Charging power of the gravity energy storage system at time t;

[0015] —The discharge power of the gravity energy storage system at time t;

[0016] —The maximum charging power of the gravity energy storage system;

[0017] —The maximum discharge power of the gravity energy storage system;

[0018] —Day-ahead market electricity price at time t;

[0019] ρ om —Operation and maintenance cost per unit maximum charge and discharge power.

[0020] A further improvement of the present invention is that the constraints of the day-ahead scheduling model of the double-tower gravity energy storage include brick transfer volume constraints, upper and lower limit constraints on the volumes of towers a and b, initial and final state constraints, brick transfer rate constraints, charge and discharge power upper limit constraints, and charge and discharge power constraints.

[0021] The present invention is further improved in that: 1) the brick transfer volume constraint is as follows:

[0022]

[0023] Where:

[0024] —Time of day collection;

[0025] —The volume of tower a at time t;

[0026] —The volume of tower b at time t;

[0027] —the volume of bricks transferred upward at time t;

[0028] —The volume of bricks transferred downward at time t

[0029] 2) The upper and lower limits of the volume of towers a and b are as follows:

[0030]

[0031] Where:

[0032] —Minimum / maximum volume of tower a;

[0033] —Minimum / maximum volume of tower b;

[0034] 3) The initial and final state constraints are as follows:

[0035]

[0036] Where:

[0037] — volume of tower a at the end of the day;

[0038] —The volume of tower a at the beginning of the day;

[0039] 4) The brick transfer rate constraints are as follows:

[0040]

[0041] Where:

[0042] —The maximum volume of bricks transferred upward per unit time;

[0043] —The maximum volume of bricks transferred downward per unit time;

[0044] X a —The number of bricks on each floor of a tower;

[0045] X b —Number of bricks per floor of tower b;

[0046] S unit —the base area of ​​a unit brick;

[0047] —a maximum height of the tower;

[0048] 5) Charge and discharge power upper limit constraint:

[0049]

[0050] —discharge power;

[0051] —discharge power;

[0052] —Charging power limit;

[0053] —Discharge power upper limit;

[0054] —Charge state variable of the gravity energy storage system at time t;

[0055] —Discharge state variables of the gravity energy storage system at time t;

[0056] 6) The charging and discharging power constraints are as follows:

[0057]

[0058]

[0059] Where:

[0060] g—acceleration due to gravity;

[0061] S unit —the base area of ​​a unit brick;

[0062] V unit —The volume of a unit brick;

[0063] munit —The mass of a unit brick;

[0064] η ch —The charging efficiency of the system;

[0065] η dis —The discharge efficiency of the system.

[0066] A further improvement of the present invention is that, according to the actual parameters of the double-tower gravity energy storage system, before constructing the probability distribution robust day-ahead dispatching model of the gravity energy storage power station, a probability distribution uncertainty set of the day-ahead electricity price is constructed.

[0067]

[0068] — Random variables are based on empirical distributions of historical data sets;

[0069] Φ—the set of all possible probability distributions under the support set Ω;

[0070] ε—Radius of the Wasserstein sphere.

[0071] A further improvement of the present invention is that the radius of the Wasserstein sphere is calculated by the following formula:

[0072]

[0073] r — diameter of the random variable support set Ω;

[0074] N—the number of samples in the historical data set;

[0075] α—Probability distribution uncertainty set Contains the confidence level of the true probability distribution.

[0076] A further improvement of the present invention is that the support set is calculated by the following formula:

[0077]

[0078] B—correlation constant matrix;

[0079] b—association constant vector;

[0080] —Day-ahead electricity price vector,

[0081] —The uncertainty value of the day-ahead electricity price at time t.

[0082] A further improvement of the present invention is that, within the probability distribution uncertainty set of the day-ahead electricity price, under the worst probability distribution of the day-ahead electricity price, the objective function of the probability distribution robust day-ahead dispatching model of the gravity energy storage power station is:

[0083]

[0084] — mathematical expectation operator;

[0085] <>—scalar product (dot product) operator;

[0086] p—the day-ahead output power vector of the gravity energy storage power station.

[0087] A further improvement of the present invention is that the constraints of the probabilistic blue day-ahead scheduling model of the gravity energy storage power station include: brick transfer volume constraints, upper and lower limit constraints on the volumes of towers a and b, initial and final state constraints, brick transfer rate constraints, charge and discharge power upper limit constraints, and charge and discharge power constraints.

[0088] Compared with the prior art, the present invention has at least the following beneficial effects: In the present invention, for the day-ahead scheduling problem of the twin-tower gravity energy storage power station, compared with other energy storage scheduling systems, the model is modeled according to the actual parameters of the twin-tower gravity energy storage system. Due to the use of the twin-tower gravity energy storage characteristics, the twin-tower gravity energy storage day-ahead scheduling model and the gravity energy storage power station probability distribution robust day-ahead scheduling model are more in line with reality. Compared with traditional energy storage day-ahead scheduling methods, the present invention has stronger adaptability to the twin-tower gravity energy storage system and the day-ahead electricity price uncertainty. The present invention enables the twin-tower gravity energy storage system to follow the electricity price signal to move bricks in the day-ahead scheduling to achieve arbitrage purposes, and even under the worst probability distribution of electricity prices, it can maximize the system's expectation of obtaining revenue in the day-ahead electricity market, thereby enabling the twin-tower gravity energy storage to better exert its energy storage potential.

[0089] Furthermore, the objective function of the twin-tower gravity energy storage day-ahead dispatching model is to maximize the participation of the gravity energy storage system in the day-ahead electricity market and realize profits, so that the dispatching results obtained can more effectively help the twin-tower gravity energy storage system to make profits.

[0090] Furthermore, the constraints of the gravity energy storage system take into account the brick transfer volume constraints, brick transfer rate constraints, charging and discharging power constraints, etc., making the entire model closer to the actual operation mode, and the obtained scheduling plan is more realistic and instructive for the gravity energy storage system.

[0091] Furthermore, by constructing an uncertainty set of probability distribution of day-ahead electricity prices, the uncertainty of electricity prices can also be taken into consideration in the gravity energy storage scheduling system.

[0092] Furthermore, the objective function of the probability distribution robust day-ahead dispatching model of the gravity energy storage power station is to maximize the profit of the gravity energy storage system under the probability distribution of the worst day-ahead electricity price, so that the gravity energy storage system can be reasonably dispatched even under the worst electricity price condition. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 It is a flow chart of the day-ahead dispatching method of a double-tower gravity energy storage power station considering uncertainty according to the present invention;

[0094] Figure 2 To clear the electricity price on the day before;

[0095] Figure 3 The historical data of electricity price for the day before clearing;

[0096] Figure 4 It is the day-ahead dispatch result of deterministic gravity energy storage;

[0097] Figure 5 It is the day-ahead dispatch result of deterministic gravity energy storage;

[0098] Figure 6 This is the day-ahead dispatch result of gravity energy storage considering uncertainty. DETAILED DESCRIPTION

[0099] The present invention will be further described below in conjunction with the accompanying drawings. However, the content of the present invention is not limited to this.

[0100] See also Figure 1 The present invention provides a method for day-ahead scheduling of a double-tower gravity energy storage power station taking uncertainty into consideration, comprising the following steps:

[0101] S1. Obtain relevant parameters and power market data of the actual double-tower gravity energy storage system;

[0102] S2. According to the power market data, determine whether the day-ahead electricity price is known. If the day-ahead electricity price is known, use the relevant parameters of the twin-tower gravity energy storage system and consider the physical characteristics of the twin-tower gravity energy storage to build a day-ahead dispatch model based on the twin-tower gravity energy storage. Use the linearization method to reconstruct the charging and discharging power constraints to improve the efficiency of model solution.

[0103] S3. Consider the uncertainty of the day-ahead electricity price, that is, when the day-ahead electricity price is unknown, maximize the profit of the gravity energy storage system under the worst probability distribution of the day-ahead electricity price, build a probability distribution robust day-ahead dispatching model for the gravity energy storage power station, and reconstruct the model so that the model can be solved.

[0104] When applying the model proposed in the present invention, it is necessary to obtain the relevant parameters of the double-tower gravity energy storage system and the electricity price data related to the electricity market. The calculation model input data that needs to be obtained include:

[0105] The parameters of the double-tower gravity energy storage system are the basic technical data of the gravity energy storage system, including: the number of bricks per layer of tower a / tower b X a / X b; Bottom area / volume / mass per unit brick S unit / V unit / m unit ;Minimum / maximum volume of tower a b tower minimum / maximum volume Initial volume of tower a The maximum volume of bricks raised / lowered per unit time The charge / discharge efficiency η of the system ch / η dis ; Maximum charge / discharge power Maximum height of tower a Operation and maintenance cost per unit maximum charge and discharge power ρ om .

[0106] The power market data refers to the power market-related electricity price data, including: the day-ahead market electricity price at time t The uncertainty value of the day-ahead electricity price at time t

[0107] After obtaining the above information, please refer to Figure 1 The day-ahead dispatching method of a double-tower gravity energy storage power station considering uncertainty includes the following steps:

[0108] Step 1: Construct a dual-tower gravity energy storage day-ahead dispatch model, including:

[0109] Step 1.1: Construct the objective function for maximizing the profit of the gravity energy storage system by participating in the day-ahead electricity market:

[0110]

[0111] Where:

[0112] —Charging power of the gravity energy storage system at time t;

[0113] —The discharge power of the gravity energy storage system at time t;

[0114] —The maximum charging power of the gravity energy storage system;

[0115] —The maximum discharge power of the gravity energy storage system;

[0116] —Day-ahead market electricity price at time t;

[0117] ρ om —Operation and maintenance cost per unit maximum charge and discharge power.

[0118] Step 1.2: Construct the constraints of the gravity energy storage system:

[0119] 1) Brick transfer volume constraints

[0120]

[0121] Where:

[0122] —Time of day collection;

[0123] —The volume of tower a at time t;

[0124] —The volume of tower b at time t;

[0125] —The volume of bricks transferred upward (from tower b to tower a) at time t;

[0126] —The volume of bricks transferred downward (from tower a to tower b) at time t.

[0127] 2) Upper and lower limit constraints of ab tower volume

[0128]

[0129] Where:

[0130] —Minimum / maximum volume of tower a;

[0131] —Minimum / maximum volume of tower b.

[0132] It can be seen that the sum of the volumes of tower a and tower b is always equal to the maximum volume of tower a at any time.

[0133] 3) Initial and final state constraints

[0134] For the operation of the gravity energy storage system, the state at the end of the day should be maintained at the initial level to facilitate the scheduling of the next stage:

[0135]

[0136] Where:

[0137] — volume of tower a at the end of the day;

[0138] —The volume of column a at the beginning of the day (i.e. t=0).

[0139] 4) Brick transfer rate constraints

[0140]

[0141] Where:

[0142] —The maximum volume of bricks transferred upward per unit time;

[0143] —The maximum volume of bricks transferred downward per unit time;

[0144] X a —The number of bricks on each floor of a tower;

[0145] X b —Number of bricks per floor of tower b;

[0146] S unit —the base area of ​​a unit brick;

[0147] —a The maximum height of the tower.

[0148] 5) Charge and discharge power upper limit constraint:

[0149]

[0150] —discharge power;

[0151] —discharge power;

[0152] —Charging power limit;

[0153] —Discharge power upper limit;

[0154] —Charge state variable of the gravity energy storage system at time t (1 / 0 for charging, 0 / 1 for discharging).

[0155] —Discharge state variable of the gravity energy storage system at time t (1 / 0 represents charging, 0 / 1 represents discharging).

[0156] 6) Charge and discharge power constraints:

[0157]

[0158]

[0159] Where:

[0160] g—acceleration due to gravity;

[0161] S unit —the base area of ​​a unit brick;

[0162] V unit —The volume of a unit brick;

[0163] m unit —The mass of a unit brick;

[0164] η ch —The charging efficiency of the system.

[0165] η dis —The discharge efficiency of the system.

[0166] Step 1.3: Linearize equations (7) and (8) to transform the entire problem into a mixed integer linear optimization problem that is easy to solve (both equations (7) and (8) are in the form of p = g(h, q), and p = g(h, q) is used to refer to both equations (7) and (8) below):

[0167] 1) Divide h and q equally within their definition intervals:

[0168]

[0169] Where:

[0170] —The segmented set of variables h, namely

[0171] —The piecewise set of variables q, i.e.

[0172] 2) Determine the interval of h and q:

[0173]

[0174] Where:

[0175] —Variable representing the interval in which h is located;

[0176] h i —Intermediate variable used to calculate the interval in which h lies.

[0177]

[0178] Where:

[0179] —Variable representing the interval in which q lies;

[0180] q i,j —Intermediate variable used to calculate the interval in which q lies.

[0181] 3) Calculate the approximate power p:

[0182]

[0183] Where:

[0184] g i,j —Representative value of function g in each interval;

[0185] —The representative value of h in interval i.

[0186] Step 1.4: Model solution:

[0187] Formula (1)-(6) and Formula (9)-(12) constitute a complete dual-tower gravity energy storage day-ahead dispatching model. After solving, the optimal dispatching scheme of the gravity energy storage system can be obtained. When the future day-ahead electricity price can be obtained, the dual-tower gravity energy storage system can be operated according to the dispatching scheme to ensure the maximum benefit.

[0188] In the present invention, the linearization of the charge and discharge power constraints transforms the day-ahead scheduling model of the twin-tower gravity energy storage into a mixed integer linear programming model, which can significantly improve the solution speed of the model.

[0189] Step 2: Construct a probability distribution blue day-ahead dispatch model for gravity energy storage power stations:

[0190] Step 2.1: Construct the probability distribution uncertainty set of the day-ahead electricity price

[0191] 1) Wasserstein distance calculation method:

[0192]

[0193] Where:

[0194] —any two random variables;

[0195] and The set of all possible joint distributions;

[0196] and The marginal distribution of

[0197] Ω—Random variable The support set of

[0198] ||·||—norm function, generally 1 norm is selected.

[0199] 2) Day-ahead electricity price (historical data) vector The support set Ω of :

[0200] The meaning is that the day-ahead electricity price fluctuates within a certain probability range, expressed as:

[0201]

[0202] B—correlation constant matrix;

[0203] b—Association constant vector.

[0204] —Day-ahead electricity price vector,

[0205] —The uncertainty value of the day-ahead electricity price at time t is taken from historical data.

[0206] 3) Calculation of Wasserstein sphere radius:

[0207] The general Wasserstein sphere radius is related to the number of samples and can be calculated using the following empirical formula:

[0208]

[0209] r — diameter of the random variable support set Ω;

[0210] N—the number of samples in the historical data set;

[0211] α—Probability distribution uncertainty set The confidence level that the true probability distribution is contained is generally taken as 95%.

[0212] 4) Uncertain set of probability distribution of day-ahead electricity prices calculate:

[0213]

[0214] — Day-ahead electricity prices are based on empirical distributions of historical data sets;

[0215] —Probability distribution of possible day-ahead electricity prices.

[0216] Φ—the set of all possible day-ahead electricity price probability distributions under the support set Ω.

[0217] Step 2.2: Construct the objective function to maximize the profit of the gravity energy storage system under the worst probability distribution of the day-ahead electricity price:

[0218]

[0219] Where:

[0220] — mathematical expectation operator;

[0221] <>—scalar product (dot product) operator;

[0222] p—the day-ahead output power vector of the gravity energy storage power station,

[0223] Step 2.3: Reconstruction of the probability distribution blue day-ahead dispatch model of the gravity energy storage power station:

[0224] Combined with the constraints in steps 1.2 and 1.3, a probability distribution robust day-ahead dispatch model for gravity energy storage power stations is constructed:

[0225]

[0226] The gravitational energy storage probability distribution blue bar optimization problem (18) based on Wasserstein distance cannot be solved directly, but it can be reconstructed into a solvable form using the duality theory:

[0227]

[0228] θ—auxiliary variable introduced by dual transformation;

[0229] — sample data set;

[0230] λ i —The i-th auxiliary variable of the sample data;

[0231] —The i-th sample data of the day-ahead electricity price vector ρ;

[0232] μ i —The i-th auxiliary variable vector of sample data;

[0233] ||·|| * —Infinity norm, taking the maximum absolute value of the elements.

[0234] In the present invention, the probability distribution robust day-ahead scheduling model of the gravity energy storage power station is reconstructed into a solvable mixed integer linear programming model by using duality theory, which can significantly improve the solution speed of the model.

[0235] Step 2.4: Model solution:

[0236] The reconstructed probabilistic blue-robust day-ahead dispatch model of the gravity energy storage power station is a mixed integer linear programming model, which can be solved by calling commercial solvers such as CPLEX. After solving, the probabilistic blue-robust day-ahead dispatch plan of the gravity energy storage power station can be obtained. In the case of uncertainty in the future day-ahead electricity price, the operation according to the probabilistic blue-robust day-ahead dispatch plan of the gravity energy storage power station can guarantee the benefits of the dual-tower gravity energy storage system even in the worst case.

[0237] To further demonstrate the advantages of the present invention, based on the gravity energy storage demonstration project designed by the energy storage company, the day-ahead charging and discharging behavior of the gravity energy storage power station is optimized and dispatched. Figure 2 and Figure 3To input the parameters of the day-ahead dispatch of the twin-tower gravity energy storage power station:

[0238] Table 1 Basic parameters of gravity energy storage power station

[0239]

[0240] After the day-ahead optimization scheduling, the gravity energy storage outputs the optimal scheduling plan (i.e. the 24-hour output plan for the next day), such as Figure 4 As shown, Figure 4 The middle curve is the day-ahead electricity price, and the bar graph is the charge and discharge power. The power generation of the gravity energy storage system can be regarded as the release of energy, and the gravitational potential energy of the bricks is converted into electrical energy, while the volume of the central tower decreases. In contrast, charging is regarded as the consumption of electrical energy, and the volume of the central tower increases. It can be seen that the electricity price is low at 4-8 o'clock, and the gravity energy storage is quickly charged, that is, electricity is purchased from the power grid; the electricity price is high at 15-19 o'clock, and the gravity energy storage is discharged on a large scale, that is, electricity is sold to the power grid. The deterministic day-ahead scheduling deterministic model makes a profit of $363.90. The gravity energy storage power station on the power generation side participates in the day-ahead electricity market, captures electricity price signals, and uses the peak-to-valley difference in electricity prices to arbitrage from the power grid to achieve the profit purpose of the energy storage power station.

[0241] In order to analyze the changes in the physical properties of the system (tower height, volume, etc.) during the charging and discharging of gravity energy storage, Figure 5 The central tower height and the remaining capacity of gravity energy storage are depicted. The tower height and the remaining capacity increase and decrease synchronously. The remaining capacity of gravity energy storage is the gravitational potential energy stored in the central tower. Figure 5 It is proved that the selection of tower height to represent the remaining capacity is consistent with the characteristics of gravity energy storage and objective facts. In addition, height is the most intuitive dimension of the central tower in reality. The volume of the tower is the tower height (variable) multiplied by the bottom area (constant). The volume and height can be replaced equivalently; modeling with height (volume) as the independent variable is closer to the physical structure of gravity energy storage. Gravity energy storage needs to leave enough margin for the next day's day scheduling. The volume of the central tower remains consistent from beginning to end, so its remaining capacity and the height of the central tower are also restored to the initial level at 24:00.

[0242] Figure 6 The day-ahead dispatch results of gravity energy storage considering uncertainty are shown. The fluctuation range of electricity prices is shown in the blue area, and the peak electricity price is reached at 16:00-18:00. Gravity energy storage is charged more during the low electricity price period of 4:00-8:00 and 21:00-24:00, and discharged during the peak electricity price period of 15:00-20:00, thus achieving the purpose of arbitrage from the power grid. Figure 4 The output curves of gravity energy storage are similar, the difference lies in the charging and discharging power values. The scheduling operation considering uncertainty makes a profit of $343.70, and the deterministic day-ahead scheduling model makes a profit of $363.90. Obviously, the scheduling result considering uncertainty is more conservative.

[0243] The present invention uses piecewise linearization to reconstruct the model into a mixed integer linear programming problem, and then establishes a dual-tower gravity energy storage day-ahead scheduling model. In order to cope with the uncertainty of electricity prices, the data-driven theory is used to use historical electricity price data to establish a probability distribution uncertainty set based on the Wasserstein distance. A day-ahead scheduling model for a gravity energy storage power station that takes uncertainty into account is established to minimize the expected operating cost under the worst probability distribution of electricity prices. Using the duality theory, the model is converted into a mixed integer linear programming model for easy solution, and the scheduling operation plan is obtained through the solution.

[0244] The present invention is directed to a scheduling method for twin-tower gravity energy storage. On the one hand, in practical applications, twin-tower gravity energy storage has many advantages such as low unit power generation cost, high overall architecture safety, flexible design and installation, and has broad application prospects; on the other hand, the traditional energy storage scheduling system lacks modeling of the characteristics of twin-tower gravity energy storage, and cannot fully call on the application potential of twin-tower gravity energy storage. The present invention aims to maximize the benefits of the gravity energy storage system participating in the day-ahead power market. On the basis of considering the energy storage characteristics of the gravity energy storage system, a day-ahead scheduling method for a twin-tower gravity energy storage power station considering uncertainty is constructed. The model fully considers the physical characteristics of each part of the gravity energy storage system and the uncertainty of the day-ahead power market clearing price, and is more suitable for analyzing the energy storage characteristics of the twin-tower gravity energy storage power station. The scheduling scheme calculated by the model can better help the twin-tower gravity energy storage power station complete the scheduling decision according to the day-ahead power market information. The day-ahead scheduling method for a twin-tower gravity energy storage power station considering uncertainty proposed by the present invention has better applicability to twin-tower gravity energy storage, and can also provide guidance for the construction of twin-tower gravity energy storage power stations.

Claims

1. A day-ahead dispatching method for a twin-tower gravity energy storage power station considering uncertainty, characterized in that: The following steps are involved: S1. Obtaining actual parameters of the double-tower gravity energy storage system; S2. Under the condition of known day-ahead electricity price, a day-ahead dispatching model of the double-tower gravity energy storage is constructed according to the actual parameters of the double-tower gravity energy storage system, and the optimal dispatching scheme of the gravity energy storage system is obtained by solving the model; the dispatching is performed according to the scheme; When the day-ahead electricity price is unknown, a probability distribution blue-robust day-ahead dispatching model of the gravity energy storage power station is constructed according to the actual parameters of the double-tower gravity energy storage system. The probability distribution blue-robust day-ahead dispatching scheme of the gravity energy storage power station is obtained and dispatched according to the scheme. According to the actual parameters of the double-tower gravity energy storage system, before constructing the probability distribution blue day-ahead dispatch model of the gravity energy storage power station, the probability distribution uncertainty set of the day-ahead electricity price is constructed. — Day-ahead electricity prices are based on empirical distributions of historical data sets; —Probability distribution of possible day-ahead electricity prices; ε—Wasserstein sphere radius; Φ—the set of all possible day-ahead electricity price probability distributions under the support set Ω; The objective function of the day-ahead dispatch model of the double-tower gravity energy storage is as follows: Where: —Charging power of the gravity energy storage system at time t; —The discharge power of the gravity energy storage system at time t; —The maximum charging power of the gravity energy storage system; —The maximum discharge power of the gravity energy storage system; —Day-ahead market electricity price at time t; ρ om —Operation and maintenance cost per unit maximum charge and discharge power; The constraints of the dual-tower gravity energy storage day-ahead dispatch model include brick transfer volume constraints, upper and lower limit constraints of tower a and tower b volumes, initial and final state constraints, brick transfer rate constraints, upper limit constraints of charge and discharge power, and charge and discharge power constraints; 1) The brick transfer volume constraints are as follows: Where: —Time of day collection; —The volume of tower a at time t; —The volume of tower b at time t; —the volume of bricks transferred upward at time t; —The volume of bricks transferred downward at time t 2) The upper and lower limits of the volume of towers a and b are as follows: Where: —Minimum / maximum volume of tower a; —Minimum / maximum volume of tower b; 3) The initial and final state constraints are as follows: Where: — volume of tower a at the end of the day; —The volume of tower a at the beginning of the day; 4) The brick transfer rate constraints are as follows: Where: —The maximum volume of bricks transferred upward per unit time; —The maximum volume of bricks transferred downward per unit time; X a —a The number of bricks per floor of the tower; X b —Number of bricks per floor of tower b; S unit —the base area of ​​a unit brick; —a maximum height of the tower; 5) Charge and discharge power upper limit constraint: - Charging power; —discharge power; —Charging power upper limit; —Discharge power upper limit; —Charge state variable of the gravity energy storage system at time t; —Discharge state variables of the gravity energy storage system at time t; 6) The charging and discharging power constraints are as follows: Where: g—acceleration due to gravity; S unit —the base area of ​​a unit brick; V unit —The volume of a unit brick; m unit —The mass of a unit brick; η ch —The charging efficiency of the system; η dis —The discharge efficiency of the system.

2. A method for day-ahead scheduling of a twin-tower gravity energy storage power station considering uncertainty according to claim 1, characterized in that: The radius of the Wasserstein sphere is calculated by the following formula: r — diameter of the random variable support set Ω; N—the number of samples in the historical data set; α—Probability distribution uncertainty set Contains the confidence level of the true probability distribution.

3. A method for day-ahead scheduling of a twin-tower gravity energy storage power station considering uncertainty according to claim 1, characterized in that: The actual parameters of the double-tower gravity energy storage system include the number of bricks per floor of tower a / tower b X a / X b ; Bottom area / volume / mass per unit brick S unit / V unit / m unit ;Minimum / maximum volume of tower a b tower minimum / maximum volume Initial volume of tower a The maximum volume of bricks raised / lowered per unit time Actual charging / discharging efficiency η of a double-tower gravity energy storage system ch / η dis ; Maximum charge / discharge power Maximum height of tower a Maximum charge and discharge power per unit Row maintenance cost ρ om .

4. A method for day-ahead scheduling of a twin-tower gravity energy storage power station considering uncertainty according to claim 1, characterized in that: The support set is calculated by: B—correlation constant matrix; b—association constant vector; —Day-ahead electricity price vector, —The uncertainty value of the day-ahead electricity price at time t.

5. A method for day-ahead scheduling of a twin-tower gravity energy storage power station considering uncertainty according to claim 1, characterized in that: In the uncertain set of probability distribution of day-ahead electricity prices, under the worst probability distribution of day-ahead electricity prices, the objective function of the probability distribution robust day-ahead dispatch model of gravity energy storage power stations is: — mathematical expectation operator; < >—scalar product (dot product) operator; p—the day-ahead output power vector of the gravity energy storage power station.

6. A method for day-ahead scheduling of a twin-tower gravity energy storage power station considering uncertainty according to claim 1, characterized in that: The constraints of the probabilistic blue-bar day-ahead dispatch model of the gravity energy storage power station include: brick transfer volume constraint, upper and lower limit constraints of tower a and tower b volume, initial and final state constraints, brick transfer rate constraint, upper limit constraint of charge and discharge power, and charge and discharge power constraint.

Images