Power unit combination optimization method and system based on fast frequency response of wind farm

Abstract

The application belongs to the technical field of novel power system operation optimization, and particularly discloses a power unit combination optimization method and system based on wind power plant fast frequency response; the method comprises the following steps: a system frequency response model considering wind power fast frequency response characteristics is established; the difference between active step disturbance and wind power response power and wind power ramp-off power are taken as the input of the system frequency response model, and the time domain expressions of the first frequency drop minimum point and the second frequency drop minimum point are obtained respectively; under the premise of meeting the system safety and stability operation constraint, a unit combination optimization model containing frequency change constraint is constructed with the minimum operation cost as the target; the nonlinear constraint is equivalently processed, the unit combination optimization model is solved, and the optimal power unit combination operation strategy is obtained; the application can fully exert the performance of wind power fast response, and can fully exert the adjustable capacity of wind power while coordinating frequency modulation with synchronous machines.

Description

Technical Field

[0001] This invention relates to the field of novel power system operation optimization technology, and in particular to a method and system for optimizing the combination of power units based on the fast frequency response of wind farms. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] In the development of new power systems, synchronous machines, primarily powered by coal, still need to play a crucial role in supporting the flexibility and power generation of the power system to cope with the intermittent and fluctuating characteristics of new energy sources. Against this backdrop, with the continuous increase in the penetration rate of new energy sources, the active power support provided by synchronous machines alone is no longer sufficient to meet the needs of safe system operation, posing new challenges to the frequency stability and reliability of the power system. The joint role of new energy sources and synchronous machines as the main power source, and their coordinated participation in active power and frequency support, has become an inevitable trend in the development of new power systems.

[0004] However, wind and solar fluctuations cause changes in the operation mode of new energy sources. Under different operation modes, new energy sources can implement various frequency regulation control strategies. The differences between wind and solar fluctuations and control strategies will jointly affect the frequency regulation energy of new energy sources, resulting in changes in their support capabilities.

[0005] Existing unit combination studies mostly consider the integrated inertia control under wind power load shedding and reserve mode, neglecting the wind power frequency regulation capability under maximum power point tracking (MPPT) operation. As a result, the efficient utilization of wind power frequency regulation energy cannot be achieved, and the economic efficiency of power system operation needs to be further improved. Summary of the Invention

[0006] To address the aforementioned issues, this invention proposes a power unit combination optimization method and system based on the rapid frequency response of wind farms. It considers both MPPT and load shedding reserve operation modes of wind power, and proposes a power system unit combination optimization strategy based on the rapid frequency response of wind farms while taking frequency security constraints into account. This fully leverages the frequency support capability of wind power, achieves coordinated frequency regulation between wind power and synchronous machines, and reduces the operating costs of the power system.

[0007] In some implementations, the following technical solutions are adopted:

[0008] A method for optimizing the combination of power units based on the fast frequency response of wind farms includes:

[0009] Based on the optimized power curves of wind farms participating in fast frequency response under maximum power point tracking and unloaded operation, a system frequency response model considering the fast frequency response characteristics of wind power is established.

[0010] By using the difference between the active step disturbance and the wind power response power and the wind power ramp-out power as inputs to the system frequency response model, the time-domain expressions for the first frequency drop minimum point and the second frequency drop minimum point are obtained, respectively.

[0011] Under the premise of satisfying the constraints of safe and stable system operation, and with the goal of minimizing operating costs, a unit combination optimization model including frequency variation constraints is constructed.

[0012] The nonlinear constraints are equivalently processed by a piecewise linearization method, and the unit combination optimization model is solved to obtain the optimal power unit combination operation strategy.

[0013] As a further approach, the difference between the active step disturbance and the wind power response is used as the input to the system frequency response model to obtain the time-domain expression for the first frequency drop minimum point. Specifically:

[0014] ;

[0015] in, R , K m These are the equivalent droop coefficient and mechanical power gain coefficient of the synchronous machine prime mover and speed governor, respectively. D This represents the equivalent damping coefficient of the power grid. ω r The frequency of the damped oscillation. ζ For the damping ratio, ω n For natural frequency, The time of the lowest frequency point; Phase angle; The disturbance power; For coefficients; This represents the response power of the wind turbine under MPPT conditions. This represents the response power of wind power in a reduced load standby state.

[0016] As a further approach, the wind power slope withdrawal power is used as the input to the system frequency response model to obtain the time-domain expression for the second frequency drop minimum point, specifically:

[0017] ;

[0018] ;

[0019] in, This is the quasi-steady-state frequency of the first frequency drop; This refers to the power output removed from wind power. The minimum frequency time under equivalent load input. The time it takes for the frequency to drop to a quasi-steady state after the first drop. This represents the deviation from the lowest point of the second frequency drop. For equivalent load input t Frequency deviation value at time 0; For coefficients, R , K m These are the equivalent droop coefficient and mechanical power gain coefficient of the synchronous machine prime mover and speed governor, respectively. D This represents the equivalent damping coefficient of the power grid. ω r The frequency of the damped oscillation. ζ For the damping ratio, ω n It is the natural frequency.

[0020] As a further option, the unit combination optimization model aims to minimize the sum of the power generation cost, start-up and shutdown cost, reserve cost of thermal power units, and reserve cost of wind farms.

[0021] As a further embodiment, the frequency variation constraint includes:

[0022] Frequency single drop minimum point constraint: ;

[0023] Frequency double drop minimum point constraint: ;

[0024] in, f min The safety threshold for the lowest point of system frequency;

[0025] , , These represent the minimum frequency drop on the first attempt, the minimum frequency drop on the second attempt, and the system's rated frequency, respectively.

[0026] As a further solution, a piecewise linearization method is used to equivalently process the frequency single-drop minimum point constraint or the frequency double-drop minimum point constraint, specifically as follows:

[0027] Aggregate parameters of thermal power units;

[0028] The parameter space of the thermal power unit is divided into pieces using a piecewise linearization method. N J Each series of subspaces;

[0029] In each subspaceS j In this process, the model is optimized to find the hyperplane in each subspace that is closest to the time-domain representation of the first or second frequency drop minimum point;

[0030] For each sampling point, a suitable linear segment is selected so that the value of the linear segment at the sampling point is closest to the actual curve value, thereby obtaining a linearized expression for the minimum point constraint of the first frequency drop or the minimum point constraint of the second frequency drop.

[0031] As a further solution, the wind power load shedding rate in the wind farm output power constraint is linearized, specifically as follows:

[0032] Range of values ​​for wind power load factor D j,t Discretize into N L Points , ,…, ,…, At the same time, add N L Auxiliary binary variables serve as segmentation selection indicator variables. , ,…, ,…, ,as well as N L Auxiliary continuous variables , , ,… ; n =1, 2,…, N L ;

[0033] Introduce a set of linear constraints that satisfy:

[0034] when When =1, = σ w ,when When =0, =0;

[0035] Only one binary variable is allowed. =1, then The value is the selected first n wind power load reduction rate .

[0036] In other embodiments, the following technical solutions are adopted:

[0037] A power unit combination optimization system based on the fast frequency response of wind farms includes:

[0038] The frequency response model building module is used to establish a system frequency response model that considers the fast frequency response characteristics of wind power based on the optimized power curves of wind farms participating in fast frequency response under maximum power point tracking operation and unloaded operation.

[0039] The frequency response module is used to take the difference between the active step disturbance and the wind power response power and the wind power slope-off power as inputs to the system frequency response model, and obtain the time-domain expressions for the first frequency drop minimum point and the second frequency drop minimum point, respectively.

[0040] The unit optimization model construction module is used to construct a unit combination optimization model with frequency variation constraints, under the premise of meeting the constraints of safe and stable system operation and with the goal of minimizing operating costs.

[0041] The model solving module is used to perform equivalent processing on the nonlinear constraints using a piecewise linearization method, solve the unit combination optimization model, and obtain the optimal power unit combination operation strategy.

[0042] In other embodiments, the following technical solutions are adopted:

[0043] A terminal device includes a processor and a memory, wherein the processor is used to implement instructions; and the memory is used to store multiple instructions adapted to be loaded and executed by the processor to optimize the power unit combination based on the fast frequency response of wind farms as described above.

[0044] In other embodiments, the following technical solutions are adopted:

[0045] A computer-readable storage medium storing a plurality of instructions adapted for loading and execution by a processor of a terminal device of the above-described method for optimizing the power unit combination based on the fast frequency response of wind farms.

[0046] Compared with the prior art, the beneficial effects of the present invention are:

[0047] (1) This invention describes the optimized power curves of wind farms participating in fast frequency response under maximum power point tracking operation and load reduction operation respectively, and establishes a system frequency response model that considers the fast frequency response characteristics of wind power. It can give full play to the fast response performance of wind power and fully utilize the adjustable capability of wind power while coordinating frequency regulation with synchronous machines.

[0048] (2) The present invention establishes and derives analytical expressions for the lowest point of the first frequency drop and the lowest point of the second frequency drop after disturbance, which can realize the accurate characterization of frequency dynamics and help to rationally allocate wind power frequency regulation resources.

[0049] (3) The method of the present invention realizes the efficient use of wind power frequency regulation resources, effectively leverages the frequency support capability of wind farms, and ensures the safety and stable operation of system frequency while improving the economic efficiency of dispatch.

[0050] Other features and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0051] Figure 1 This is a flowchart of the power unit combination optimization method based on the fast frequency response of wind farms in an embodiment of the present invention;

[0052] Figure 2 This is the power response curve of wind power under MPPT state in an embodiment of the present invention;

[0053] Figure 3 This is a wind power-speed curve in an embodiment of the present invention;

[0054] Figure 4 This is the power response curve of wind power under unloaded conditions in an embodiment of the present invention;

[0055] Figure 5 This is a schematic diagram of the system frequency response model considering wind power in an embodiment of the present invention;

[0056] Figure 6 This is a schematic diagram of the equivalent disturbance of the second frequency drop in an embodiment of the present invention;

[0057] Figure 7 This is an improved IEEE RTS-24 wiring diagram in an embodiment of the present invention;

[0058] Figure 8 The above are the daytime wind speed and load forecast curves in this embodiment of the invention;

[0059] Figure 9 This is a stacked diagram of the power output of thermal power units obtained by the method in an embodiment of the present invention;

[0060] Figure 10 This is a power output curve obtained by the method in an embodiment of the present invention;

[0061] Figure 11 The results of wind turbine grouping optimization for each time period obtained by the method in the embodiments of the present invention;

[0062] Figure 12 This is a schematic diagram comparing the linearized fitting curves of the first and second minimum points of the frequency in an embodiment of the present invention.

[0063] Figure 13 This is a schematic diagram comparing the lowest frequency points of different strategies in an embodiment of the present invention;

[0064] Figure 14 This is a schematic diagram comparing different RoCoF strategies in embodiments of the present invention. Detailed Implementation

[0065] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0066] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0067] Example 1

[0068] In one or more embodiments, a power unit combination optimization method based on the fast frequency response of wind farms is disclosed, combining... Figure 1 Specifically, it includes:

[0069] S101: Based on the optimized power curves of wind farms participating in fast frequency response under maximum power point tracking and unloaded operation, establish a system frequency response model that considers the fast frequency response characteristics of wind power.

[0070] In MPPT operation mode, wind power can provide short-term power support to the system by releasing rotor kinetic energy; simultaneously, to ensure rotational speed safety, the wind power speed needs to return to its original operating state after the frequency returns to quasi-steady state. Therefore, this embodiment proposes a control method combining fast frequency response and ramp-out, such as... Figure 2 and Figure 3 As shown: Under normal operating conditions, the wind power operates in MPPT state, i.e., point A; after a disturbance occurs and exceeds the wind power frequency regulation dead zone, the wind power increases its power rapidly to point B in a step manner, and then continues to operate at this power to point C, during which the speed continuously decreases; when the system frequency returns to quasi-steady state, the wind power gradually decreases in a ramp manner, and after reaching point D, the power remains unchanged, and the speed gradually recovers; when the wind power equals the MPPT, it switches back to MPPT operation at point E, and finally recovers to point F.

[0071] Among them, wind power support duration t on and ramp exit duration t offThe inertial response of the synchronous machine needs to be matched with the frequency modulation time. t on It should be ensured that the time from the start of the synchronous machine's speed decrease to the recovery process is greater than that required. Based on typical synchronous machine parameters, this can be taken as... t on =10s; t off It should be approximately equal to the time it takes for the synchronous machine to reach its first frequency modulation response, which can be expressed as:

[0072] (1)

[0073] in, ω r is the damped oscillation frequency.

[0074] Wind power response power Δ P f and exit power Δ P The wind speed can be constructed using time-domain simulation, depending on the active power support capacity of the wind turbine. v w -Response power Δ P f -Exit power Δ P The dispatcher can look up the table based on the wind speed forecast results to determine the wind power parameters.

[0075] Under unloaded operation, the wind power output is less than the maximum MPPT power, such as Figure 3 and Figure 4 As shown at point A1, when a disturbance occurs in the system and exceeds the wind power frequency regulation dead zone, the wind power output increases sharply in a step manner. It reaches point A2, and then continues to operate at that power until it eventually stabilizes at point A in MPPT.

[0076] This represents the response power of the wind turbine under MPPT conditions. This represents the response power of wind power in a reduced load standby state.

[0077] In order to fully utilize the active power support capacity of wind turbines, it is necessary to ensure that the wind turbine load reduction power is fully released under the expected maximum disturbance. Therefore, the wind power response power... It can be represented as:

[0078] (2)

[0079] In the formula: Δ P L For the disturbance power, Δ P Lmax For the maximum expected disturbance power, Δ P d Reduce the load on wind power.

[0080] S102: Using the difference between the active step disturbance and the wind power response power and the wind power ramp-out power as inputs to the system frequency response model, the time-domain expressions for the first frequency drop minimum point and the second frequency drop minimum point are obtained respectively.

[0081] The power system frequency response model considering wind power frequency regulation is as follows: Figure 5 As shown, the wind power response power Δ P w It can be used as input to the traditional SFR model, effectively changing the magnitude of the disturbance. Two frequency drops occurred during the entire frequency regulation period. The first frequency drop was caused by a power step disturbance, and the second frequency drop was caused by wind power withdrawing from frequency support. The two frequency drops are analyzed below:

[0082] (1) First frequency drop:

[0083] The first frequency dip uses the difference between the active step disturbance and the wind power response as the input to the SFR model, and its time-domain expression Δ f 1( t This can be represented as:

[0084] (3)

[0085] (4)

[0086] (5)

[0087] (6)

[0088] (7)

[0089] (8)

[0090] (9)

[0091] Where, Δ f n1 This indicates the lowest point of frequency deviation during the first frequency drop; P step The equivalent perturbation magnitude; D This represents the equivalent damping coefficient of the power grid. H The equivalent inertial time constant of the power grid; R , K m , T R , F HThese are the equivalent parameters of the synchronous machine prime mover and governor, namely the droop coefficient, mechanical power gain coefficient, prime mover time constant, and reheat power percentage, respectively. ω r The frequency of the damped oscillation. ζ For the damping ratio, ω n For natural frequency, φ , φ 1. φ 2 are both phase angles (introduced to express the frequency response in the time domain form of damped oscillation, and have no specific meaning). a For coefficients; , , , These represent the incremental power of wind power, the disturbance power, the response power of the wind turbine under MPPT state, and the response power of wind power under load shedding and standby state, respectively.

[0092] Taking the derivative of equation (3) yields the time of the first minimum frequency point. t n1 for:

[0093] (10)

[0094] Therefore, the first lowest frequency point Δ f n1 It can be represented as:

[0095] (11)

[0096] (2) Second frequency drop:

[0097] The second frequency drop uses the wind power slope exit power as input to the SFR model, such as... Figure 6 As shown, its time-domain expression Δ f 2( t This can be represented as:

[0098] (12)

[0099] (13)

[0100] Differentiating equation (12) yields Figure 6 The minimum frequency time under the equivalent load input shown t 0 is:

[0101] (14)

[0102] (15)

[0103] in,n 0 indicates that t The smallest integer greater than zero.

[0104] These are intermediate variables in the derivation process and have no specific physical meaning. For wind power to be withdrawn, , These are all phase angles generated during the derivation process and have no specific physical meaning.

[0105] Therefore, the time of the second lowest frequency point can be obtained. t n2 for:

[0106] (16)

[0107] The second frequency drop to the lowest point Δ f n2 It can be represented as:

[0108] (17)

[0109] in, This is the quasi-steady-state frequency of the first frequency drop.

[0110] express Figure 6 The equivalent load input shown t Frequency deviation value at time 0.

[0111] S103: Under the premise of satisfying the constraints of safe and stable operation of the system, construct a unit combination optimization model with the goal of minimizing operating costs, including frequency variation constraints.

[0112] In this embodiment, the objective of unit combination is to minimize operating costs while meeting the constraints of safe and stable system operation. This mainly includes the power generation cost, start-up and shutdown cost, and reserve cost of traditional thermal power units, as well as the reserve cost of wind farms. The objective function can be expressed as:

[0113] (18)

[0114] In the formula: For thermal power units i The coal consumption cost coefficient, P i,t For thermal power units i exist t Power generation at any given moment and thermal power units i exist t The costs of starting up and shutting down at any time. and thermal power units i Adjusting reserve costs upwards and downwards, and thermal power units i exist t Adjusting reserve capacity up or down over a given time period and Wind turbine j Adjusting reserve costs upwards and downwards, and Wind turbine j exist t Adjusting the reserve capacity up or down over a given time period.

[0115] The constraints of the above objective function specifically include: system operation constraints, thermal power plant operation constraints, and wind farm operation constraints.

[0116] System operational constraints specifically include:

[0117] (1) System power balance constraints:

[0118] (19)

[0119] In the formula: G, W, L, D These are the number of thermal power units, the number of wind power units, the number of power lines, and the number of loads. P i,t For thermal power units i During the period t contribution ，P j,t For wind turbines j During the period t contribution ，P l,t For the line l During the period t The size of the current ，P d,t For load d During the period t The power requirements.

[0120] (2) Power flow constraints of the line:

[0121] (20)

[0122] (twenty one)

[0123] In the formula: θ m,t and θ n,t They are time t The node voltage phase angle, θ ref,t The reference phase angle for time t,P min l and P max l The lines are respectively l The constraints of current trends.

[0124] (3) Frequency change rate constraint:

[0125] (twenty two)

[0126] In the formula: f 0 represents the system's rated frequency, RoCoF lim This is the safe threshold for the system's frequency change rate.

[0127] (4) Frequency drop minimum point constraint:

[0128] (twenty three)

[0129] (5) Frequency double drop minimum point constraint:

[0130] (twenty four)

[0131] in, f min The safety threshold for the lowest point of system frequency; , These represent the minimum value of the first frequency drop and the minimum value of the second frequency drop, respectively.

[0132] The specific constraints on thermal power plant operation include:

[0133] (1) Upper and lower limits of thermal power output power constraints:

[0134] (25)

[0135] In the formula: and thermal power units i Minimum and maximum output limits, For thermal power units i During the period t The start / stop status is indicated by 1 for power on and 0 for power off. , , They represent thermal power units i During the period t The reduction of reserve, the output size, and the increase of reserve.

[0136] (2) Upper and lower limits of thermal power reserve capacity:

[0137] (26)

[0138] (27)

[0139] In the formula: and thermal power units i Adjusting the standby capacity limit.

[0140] (3) Cost constraints of thermal power plant start-up and shutdown:

[0141] (28)

[0142] (29)

[0143] In the formula: H i , J i The units i The cost of a single start / stop.

[0144] (4) Start-up and shutdown time constraints for thermal power plants:

[0145] (30)

[0146] (31)

[0147] In the formula: T off and T on These are the minimum shutdown and startup times, respectively.

[0148] (5) Thermal power plant ramping constraint, that is, the output slope of two adjacent time periods is within a certain range:

[0149] (32)

[0150] In the formula: R d , R u These are the downhill and uphill ramp speed limits for the unit, respectively.

[0151] The specific operational constraints of wind farms are as follows:

[0152] (1) Wind farm output power constraints:

[0153] The power of a wind turbine is a piecewise nonlinear function of wind speed, which can be expressed as:

[0154] (33)

[0155] In the formula: p j,t For wind turbines j existt Maximum power at a given wind speed. v w For wind speed, v in , v out , v R These are the cut-in wind speed, cut-out wind speed, and rated wind speed, respectively. p R This is the rated power.

[0156] The total output power of a wind farm can be expressed as:

[0157] (34)

[0158] In the formula: For wind farms in t Total power at time , N MPPT , N DE These represent the number of fans operating under MPPT (Multi-Purpose Test) and those operating under reduced load conditions, respectively. d j,t For wind turbines j exist t The rate of load reduction at any given moment u j,t For wind turbines j During the period t The grid connection status is indicated by 1 for grid-connected operation and 0 for off-grid operation.

[0159] (2) Constraints on the number of grid-connected wind power plants:

[0160] (35)

[0161] (36)

[0162] (37)

[0163] (38)

[0164] (39)

[0165] In the formula: This is the grid connection coefficient for wind farms. , These are the upper and lower limits of the grid connection coefficient for wind farms, set at 0.4-1. This represents the number of wind turbine units.

[0166] (3) Constraints on wind farm reserve adjustments:

[0167] (40)

[0168] (41)

[0169] In the formula: and These are the upward and downward reserve auxiliary variables for wind turbine units, respectively. d max This represents the maximum load reduction rate of the wind turbine.

[0170] (4) Wind farm response power and power withdrawal constraints:

[0171] To prevent frequency reversal, the wind farm response power should be less than the disturbance magnitude, which can be expressed as:

[0172] (42)

[0173] (43)

[0174] (44)

[0175] (45)

[0176] In the formula: Δ P f,j,t and Δ P j,t Wind turbines under MPPT status j exist t The response power and exit power at any given time are both functions of wind speed and can be obtained by looking up a table, Δ. P fd,j,t Wind turbine under reduced load conditions j exist t Response power at any given time. for t The power of the disturbance at any given moment. This represents the maximum disturbance power. , These are lookup functions for the response power and exit power of wind power under MPPT conditions, respectively.

[0177] S104: The piecewise linearization method is used to perform equivalent processing on the nonlinear constraints, and the above unit combination optimization model is solved to obtain the optimal power unit combination operation strategy.

[0178] In this embodiment, since the constraints of the two lowest frequency drop points are highly nonlinear and do not meet the constraints of convex optimization, the solver cannot solve the problem and needs to be linearized.

[0179] First, the expression for the lowest frequency point (11) is: H , D , K , R , F Functions of variables, research found K , R They always appear in pairs, therefore, by introducing new variables... F g , R g The parameters of thermal power units are aggregated, defined as follows:

[0180] (46)

[0181] (47)

[0182] (48)

[0183] (49)

[0184] In the formula: H i , D i , K i , R i , F i The first i The inertial time constant, damping coefficient, mechanical power gain coefficient, droop coefficient, and reheat power percentage of the thermal power unit.

[0185] The expression for the lowest point of the frequency drop (11) can be rewritten as:

[0186] (50)

[0187] It can be observed that the lowest point of the frequency drop is F g , R g , H , D A function with four variables can be represented by Δ. f n1 ( F g , R g , H , D )express. For thermal power units i The start / stop status is indicated by 1 for power on and 0 for power off. For thermal power units i Maximum output.

[0188] Then, the piecewise linearization (PWL) method was used to select ( F g , R g , H , D )∈ S The parameter space will S Divided into N J Subspaces of each series S j ( j =1, 2, ..., N J In each subspace S j In, function It has local convexity, so the optimization model shown in equation (51) can be used to find the closest one in each subspace. The hyperplane.

[0189] (51)

[0190] In the formula: , , , , It is the optimization variable for the lowest point of the frequency drop. η Indicates the sampling point. v This indicates the number of PWL segments in each subspace. A suitable PWL segment is selected for each sampling point so that the PWL segment value at the sampling point is closest to the actual curve value.

[0191] Therefore, the linearized expression for the frequency-first drop minimum point constraint can be written as:

[0192] (52)

[0193] Similarly, the frequency quadratic drop minimum point constraint is also linearized using the PWL method described above, and its linearization expression is:

[0194] (53)

[0195] In the formula: , , , , , , , It is the optimization variable for the lowest point of the second frequency drop.

[0196] However, approximating the original function using piecewise linearization methods leads to a "min-max" problem. To simplify the max function, new binary and continuous variables, along with new linear inequality constraints, are introduced, as follows:

[0197] (54)

[0198] A new variable was introduced. t =max{ p , q}, and add the following new constraints:

[0199] (55)

[0200] In the formula: δ It is a binary variable. M It is a sufficiently large positive real number.

[0201] This method simplifies the max, that is, it uses the linear constraint expression of equation (55) to express the constraint of equation (54).

[0202] In this embodiment, the wind power load reduction rate is defined in the wind turbine output power equation constraint (34). d j,t For continuous variables, u j,t Since the variables are binary, the multiplication of the two variables results in a non-linear constraint, which also requires linearization.

[0203] Range of wind power load factor D j,t Divided into N L The points are obtained through discretization. , ,…, ,…, ),join in N L Auxiliary binary variables , ,…, ,…, ( n =1, 2,…, N L ) as a segmented selection indicator variable, and N L Auxiliary continuous variables ( , , ,… Introduce a set of linear constraints:

[0204] (56)

[0205] (57)

[0206] Among them, constraint (56) means that when z n When =1, = σ w ,when z n When =0, =0; constraint (57) means that there can only be one binary variable. z n =1, then d j,t The value is the selected first n wind power load reduction rate Through the above processing, continuous variables are transformed into integer variables, thus transforming the original problem into a mixed integer linear programming problem. Similarly, this applies to equations (40) and (41). and Perform similar linearization processing on each.

[0207] Finally, the mixed-integer linear programming problem was solved using the Yalmip / Gurobi solver in MATLAB, yielding the unit combination and wind farm operation optimization results. These results include thermal power unit combination results and wind farm operation results. The thermal power unit combination results include start-up and shutdown status, output, and reserve capacity for each time period, while the wind farm operation results include grid connection coefficient, operating status for each time period, and output.

[0208] In this embodiment, an improved IEEE RTS-24 system is used as a case study to verify the effectiveness of the proposed strategy in large-scale power systems. Figure 7 As shown, the system includes 12 units, one of which is connected to busbar 11. This wind farm has 250 wind turbines, each with a rated capacity of 4.87 MW, a cut-in wind speed of 3 m / s, and a cut-out wind speed of 15 m / s. Figure 8 This section presents the day-ahead wind speed and load forecast curves. The unit standby cost for wind power is $5 / MW, which is one-third of the standby cost for conventional units. References are provided for relevant parameters regarding frequency support for thermal power units, including the base frequency. f 0 = 50 Hz, initial frequency change rate limit RoCoF lim = -0.5 Hz / s, minimum frequency limit f min= 49.5 Hz, with the disturbance set at 8% of the total load for each time period.

[0209] To verify the superiority of the proposed strategy, it is compared with other strategies:

[0210] Strategy 1: Unit combination model that does not consider frequency security constraints;

[0211] Strategy 2: A unit combination model considering frequency security constraints, in which wind power does not participate in system frequency support;

[0212] Strategy 3: All wind power systems use integrated inertia control under load reduction and standby mode to support system frequency.

[0213] The simulation results of the proposed strategy are as follows: Figure 9-11 As shown, Figure 9 As shown in the unit combination results of this embodiment, it can be seen that units 1 and 9 are in operation all day due to their lower operating costs. During the 1-6 hour period, the output of thermal power unit 9 increases significantly. This is because the upward reserve capacity and cost of thermal power unit 9 are both 0, and the reserve deficit required by the system is provided by the wind turbine units. Figure 10 This is a power output curve diagram of the strategy proposed in this embodiment. Figure 11 The diagram shows the optimization results of the wind power grouping strategy proposed in this embodiment. Compared with strategy 3, it reduces the wind turbine load reduction reserve capacity and more accurately measures the overall frequency support capability of the wind turbines.

[0214] The actual values ​​of the first and second minimum frequencies of the strategy proposed in this paper are compared with the linearized fitted values, for example... Figure 12 As shown, the mean absolute errors are 1.75% and 1.83%, respectively, and the maximum errors are 7.01% and 3.06%, respectively, demonstrating the effectiveness of the linearization method. Furthermore, it can be seen that the fitted values ​​obtained by the linearization method are all less than or equal to the actual values, ensuring the conservatism of the minimum frequency point constraint.

[0215] The comparison results of the total system cost are shown in Table 1:

[0216] Table 1 Comparison of Total System Costs

[0217]

[0218] As shown in Table 1, Strategy 1 does not consider system frequency security constraints, thus having the lowest cost, but frequency security has exceeded limits; Strategy 2 considers system frequency security constraints but does not consider frequency support and reserves provided by wind farms, resulting in relatively higher operating costs; In Strategy 3, wind farms all use off-load reserve to provide frequency support, reducing the operating time of thermal power units with high generation costs, thereby lowering the system operating cost compared to Strategy 2; Compared with previous studies, the strategy proposed in this embodiment uses two methods to provide frequency support for wind farms simultaneously, which can effectively reduce the system's reserve cost and total cost. Compared with Strategy 2 and Strategy 3, the total system operating cost of the strategy proposed in this embodiment is reduced by $6812.31 and $3547.13 respectively, greatly improving the economic efficiency of power system operation.

[0219] The minimum frequency and rate of change for different strategies are as follows: Figure 13 , Figure 14 As shown, Strategy 1, due to the lack of consideration for frequency safety constraints, has significantly exceeded the limits for both the minimum frequency point and the rate of frequency change; the minimum frequency point and rate of frequency change indicators of Strategy 2, Strategy 3, and the strategy proposed in this paper are all within the safe range. Furthermore, the rate of frequency change indicator of the strategy proposed in this embodiment is significantly better than that of Strategy 2 and Strategy 3, demonstrating that wind turbines have the ability to mitigate the rate of frequency drop under both MPPT operation and load shedding operation, which helps the power system cope with more severe active power disturbances.

[0220] Example 2

[0221] In one or more embodiments, a power unit combination optimization system based on the fast frequency response of wind farms is disclosed, comprising:

[0222] The frequency response model building module is used to establish a system frequency response model that considers the fast frequency response characteristics of wind power based on the optimized power curves of wind farms participating in fast frequency response under maximum power point tracking operation and unloaded operation.

[0223] The frequency response module is used to take the difference between the active step disturbance and the wind power response power and the wind power slope-off power as inputs to the system frequency response model, and obtain the time-domain expressions for the first frequency drop minimum point and the second frequency drop minimum point, respectively.

[0224] The unit optimization model construction module is used to construct a unit combination optimization model with frequency variation constraints, under the premise of meeting the constraints of safe and stable system operation and with the goal of minimizing operating costs.

[0225] The model solving module is used to perform equivalent processing on the nonlinear constraints using a piecewise linearization method, solve the unit combination optimization model, and obtain the optimal power unit combination operation strategy.

[0226] It should be noted that the specific implementation methods of the above modules are exactly the same as those in Example 1, and will not be described in detail again.

[0227] Example 3

[0228] In one or more embodiments, a terminal device is disclosed, comprising a processor and a memory, wherein the processor is used to implement instructions; and the memory is used to store multiple instructions adapted to be loaded by the processor and executed by the processor to optimize the power unit combination based on the fast frequency response of wind farms as described in Embodiment 1.

[0229] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0230] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0231] In the implementation process, each step of the above method can be completed by the integrated logic circuits in the processor hardware or by software instructions.

[0232] Example 4

[0233] In one or more embodiments, a computer-readable storage medium is disclosed, wherein a plurality of instructions are stored, the instructions being adapted to be loaded by a processor of a terminal device and executed by the power unit combination optimization method based on the fast frequency response of a wind farm as described in Embodiment 1.

[0234] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for optimizing the combination of power units based on the fast frequency response of wind farms, characterized in that, include: Based on the optimized power curves of wind farms participating in fast frequency response under maximum power point tracking and unloaded operation, a system frequency response model considering the fast frequency response characteristics of wind power is established. By using the difference between the active step disturbance and the wind power response power and the wind power ramp-out power as inputs to the system frequency response model, the time-domain expressions for the first frequency drop minimum point and the second frequency drop minimum point are obtained, respectively. Under the premise of satisfying the constraints of safe and stable system operation, and with the goal of minimizing operating costs, a unit combination optimization model including nonlinear frequency variation constraints is constructed. The piecewise linearization method is used to perform equivalent processing on the nonlinear frequency variation constraint, and the unit combination optimization model is solved to obtain the optimal power unit combination operation strategy. The nonlinear frequency variation constraint includes: Frequency single drop minimum point constraint: ; Frequency double drop minimum point constraint: ; in, f min The safety threshold for the lowest point of system frequency; , These are the deviations from the lowest point of the first frequency drop and the deviations from the lowest point of the second frequency drop, respectively. , , These represent the minimum value of the first frequency drop, the minimum value of the second frequency drop, and the system rated frequency, respectively. The piecewise linearization method is used to perform equivalent processing on the minimum point constraints of the first frequency drop and the second frequency drop, specifically as follows: Aggregate parameters of thermal power units; The parameter space of the thermal power unit is divided into pieces using a piecewise linearization method. N J Each series of subspaces; In each subspace S j In this process, the model is optimized to find the hyperplane in each subspace that is closest to the time-domain representation of the first or second frequency drop minimum point; For each sampling point, a suitable linear segment is selected so that the value of the linear segment at the sampling point is closest to the actual curve value, thereby obtaining the linearized expressions for the minimum point constraint of the first frequency drop and the minimum point constraint of the second frequency drop.

2. The method for optimizing the combination of power units based on the fast frequency response of wind farms as described in claim 1, characterized in that, Using the difference between the active step disturbance and the wind power response as input to the system frequency response model, the time-domain expression for the first frequency drop minimum point is obtained. Specifically: ； in, The deviation of the lowest point of the frequency drop. R This is the equivalent droop coefficient of the speed governor. K m This is the mechanical power gain coefficient of the synchronous machine; D This represents the equivalent damping coefficient of the power grid. ω r The frequency of the damped oscillation. ζ For the damping ratio, ω n For natural frequency, The time of the lowest frequency point; The phase angle; The disturbance power; For coefficients; This represents the response power of the wind turbine under MPPT conditions. This represents the response power of wind power in a reduced load standby state.

3. The method for optimizing the combination of power units based on the fast frequency response of wind farms as described in claim 1, characterized in that, Using the wind power slope-out power as input to the system frequency response model, the time-domain expression for the second frequency drop minimum point is obtained, specifically: ； ； in, This is the quasi-steady-state frequency of the first frequency drop; This refers to the power output removed from wind power. The minimum frequency time under equivalent load input. The time it takes for the frequency to drop to a quasi-steady state after the first drop. This represents the deviation from the lowest point of the second frequency drop. For equivalent load input t Frequency deviation value at time 0; For coefficients, R This is the equivalent droop coefficient of the speed governor. K m This is the mechanical power gain coefficient of the synchronous machine; D This represents the equivalent damping coefficient of the power grid. ω r The frequency of the damped oscillation. ζ For the damping ratio, ω n It is the natural frequency.

4. The method for optimizing the combination of power units based on the fast frequency response of wind farms as described in claim 1, characterized in that, The unit combination optimization model aims to minimize the sum of the power generation cost, start-up and shutdown cost, standby cost of thermal power units, and standby cost of wind farms.

5. The method for optimizing the combination of power units based on the fast frequency response of wind farms as described in claim 1, characterized in that, The wind power unloading rate in the wind farm output power constraint is linearized as follows: Range of values ​​for wind power load factor D j,t Discretize into N L Points , ,…, ,…, At the same time, add N L Auxiliary binary variables serve as segmentation selection indicator variables. , ,…, ,…, ,as well as N L Auxiliary continuous variables , ,… , … ; n =1, 2,…, N L ; Introduce a set of linear constraints that satisfy: when When =1, = σ w ,when When =0, =0; Only one binary variable is allowed. =1, then The value is the selected first n wind power load reduction rate .

6. A power unit combination optimization system based on the fast frequency response of wind farms, characterized in that, include: The frequency response model building module is used to establish a system frequency response model that considers the fast frequency response characteristics of wind power based on the optimized power curves of wind farms participating in fast frequency response under maximum power point tracking operation and unloaded operation. The frequency response module is used to take the difference between the active step disturbance and the wind power response power and the wind power slope-off power as inputs to the system frequency response model, and obtain the time-domain expressions for the first frequency drop minimum point and the second frequency drop minimum point, respectively. The unit optimization model construction module is used to construct a unit combination optimization model with nonlinear frequency variation constraints, under the premise of meeting the constraints of safe and stable system operation and with the goal of minimizing operating costs. The model solving module is used to perform equivalent processing on the nonlinear frequency variation constraints using a piecewise linearization method, solve the unit combination optimization model, and obtain the optimal power unit combination operation strategy. The nonlinear frequency variation constraint includes: Frequency single drop minimum point constraint: ; Frequency double drop minimum point constraint: ; in, f min The safety threshold for the lowest point of system frequency; , These are the deviations from the lowest point of the first frequency drop and the deviations from the lowest point of the second frequency drop, respectively. , , These represent the minimum value of the first frequency drop, the minimum value of the second frequency drop, and the system rated frequency, respectively. The piecewise linearization method is used to perform equivalent processing on the minimum point constraints of the first frequency drop and the second frequency drop, specifically as follows: Aggregate parameters of thermal power units; The parameter space of the thermal power unit is divided into pieces using a piecewise linearization method. N J Each series of subspaces; In each subspace S j In this process, the model is optimized to find the hyperplane in each subspace that is closest to the time-domain representation of the first or second frequency drop minimum point; For each sampling point, a suitable linear segment is selected so that the value of the linear segment at the sampling point is closest to the actual curve value, thereby obtaining the linearized expressions for the minimum point constraint of the first frequency drop and the minimum point constraint of the second frequency drop.

7. A terminal device comprising a processor and a memory, the processor for implementing instructions; the memory for storing multiple instructions, characterized in that, The instructions are adapted to be loaded by a processor and executed by the power unit combination optimization method based on the fast frequency response of wind farms as described in any one of claims 1-5.

8. A computer-readable storage medium storing a plurality of instructions, characterized in that, The instructions are adapted to be loaded and executed by the processor of the terminal device, and are based on the power unit combination optimization method according to any one of claims 1-5.

Images