Method and system for wind power plant frequency modulation characteristic online modeling and parameter identification

Abstract

The invention discloses a wind power plant frequency modulation characteristic online modeling and parameter identification method and system, and belongs to the technical field of new energy grid connection. The method comprises the steps that for a wind power plant, coherent wind turbine generators in the wind power plant are clustered in groups; according to a grouping clustering result, collecting operation data of the wind power plant, and based on the operation data, obtaining grouping index data and parameter identification data; on the basis of the grouping index data, on-line modeling is carried out for the frequency modulation characteristics of the wind power plant, and a wind power plant multi-machine equivalent model is generated; and based on the established wind power plant frequency modulation control system model, identifying the wind power plant frequency modulation control parameters according to the parameter identification data. According to the method, on-line modeling and parameter identification of the frequency modulation characteristics of the wind power plant are realized on the basis of measurable information, an accurate wind power plant equivalent aggregation model capable of being refreshed on line is provided for characteristic analysis of the wind power plant participating in system frequency modulation, and frequency stability analysis of the wind power plant accessing a power grid is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy grid connection, and more specifically, to a method and system for online modeling and parameter identification of frequency regulation characteristics of a wind farm. Background Art

[0002] With the increasing penetration of renewable energy sources such as wind power in the power system, the need for wind power to participate in system frequency regulation has become more urgent. The revised Wind Power Grid Connection Guidelines (GB / T19963.1-2021) released in 2021 has put forward specific requirements for the frequency regulation function of wind farms. At present, most newly built wind farms are equipped with frequency regulation control systems with inertia and primary frequency regulation capabilities. However, after wind power participates in the frequency regulation of the power system, there is a lack of accurate simulation analysis models for the coordination with synchronous machines and other renewable energy stations in the system and the impact on the system frequency, making it difficult to accurately judge the interactive impact of grid sources under the support of wind power frequency.

[0003] Modeling and simulation are important means to conduct stability analysis, grid-source interaction impact and safety risk assessment of high-proportion renewable energy systems. Accurate model simulation helps to provide research analysis and verification means in stability analysis, parameter optimization and performance improvement of large-scale renewable energy grid connection. Summary of the invention

[0004] In view of the above problems, the present invention proposes a method for online modeling and parameter identification of wind farm frequency regulation characteristics, comprising:

[0005] For a wind farm, clustering the coherent wind turbines in the wind farm;

[0006] According to the clustering result, the operation data of the wind farm is collected, and based on the operation data, the clustering index data and the parameter identification data are obtained;

[0007] Based on the grouping index data, online modeling is performed for the frequency regulation characteristics of the wind farm to generate a multi-machine equivalent model of the wind farm;

[0008] Based on the established wind farm frequency regulation control system model, the wind farm frequency regulation control parameters are identified according to the parameter identification data.

[0009] Optionally, for a wind farm, clustering the coherent wind turbines in the wind farm includes:

[0010] Grouping based on the types of double-fed and direct-drive generator sets;

[0011] In the case of double-fed and direct-drive generator types, wind turbines are grouped according to active power. If the wind turbine power P = P max , then they are divided into one category;

[0012] If P<Pmax , then the pitch angles are divided into groups, and the pitch angles β=0 and ≠ are clustered separately;

[0013] According to the speed grouping, according to the speed Ω<Ω m_min ,Ω=Ω m_min ,Ω m_min <Ω<Ω m_max ,Ω=Ω m_max Perform clustering;

[0014] For Ω m_min <Ω<Ω m_max The wind turbines under the fuzzy clustering algorithm are grouped according to the rotation speed.

[0015] Optionally, the clustering indicator data includes at least one of the following:

[0016] Wind turbine active power data, pitch angle data and speed information data.

[0017] Optionally, the parameter identification data includes at least one of the following:

[0018] The frequency of the wind farm grid connection point or the analog signal frequency data of the on-site test, the active power data of the wind farm grid connection point, the superior active power command value data received by the site frequency regulation control system, the active power command value data issued by the site frequency regulation control system to the unit, the wind turbine wind speed data, the wind turbine speed data and the pitch angle data.

[0019] Optionally, the method further includes: preprocessing the clustering indicator data, including:

[0020] The grouping index data is smoothed to eliminate noise data in the grouping index data.

[0021] Optionally, based on the grouping index data, online modeling is performed for the frequency regulation characteristics of the wind farm to generate a multi-machine equivalent model of the wind farm, including:

[0022] A clustering algorithm based on the load-shedding standby mode and the operating status of the units is used to perform equivalent processing of wind turbine multi-machine clustering on the clustering index data. Based on the processing results of the wind turbine multi-machine clustering, online modeling of the frequency regulation characteristics of the wind farm is performed to generate a multi-machine equivalent model of the wind farm.

[0023] Optionally, identifying the wind farm frequency regulation control parameters according to the parameter identification data includes:

[0024] Power distribution mode identification, power regulation link parameter identification, limiting link parameter identification and primary frequency modulation and inertia response link parameter identification.

[0025] Optionally, power distribution mode identification includes: average distribution mode identification, distribution mode identification based on the proportion of rated power, equal margin distribution mode identification and equal proportion distribution mode identification.

[0026] Optional, power regulation link parameter identification, including:

[0027] Get the input P of the control system at the station plant_ref When the command imposes disturbance or the wind farm is monitored during operation, the dispatcher issues P plant_ref The particle swarm algorithm is used to identify the parameters of the active power regulation link according to the operating data when the instruction changes.

[0028] Optionally, after the power distribution mode identification and the power regulation link parameter identification are performed, the limiting link parameter identification is performed;

[0029] The limiting link parameter identification includes:

[0030] The parameters are identified by defining the amount of power increase.

[0031] Optionally, the identification parameter objects of the limiting link parameter identification include: active power input limiting parameters and active power control link integrator limiting parameters.

[0032] Optional, primary frequency modulation and inertia response link parameter identification, including:

[0033] Dead zone parameter identification, primary frequency modulation droop coefficient identification and moment of inertia constant identification.

[0034] On the other hand, the present invention also proposes a system for online modeling and parameter identification of wind farm frequency regulation characteristics, comprising:

[0035] A grouping unit, for grouping and clustering the coherent wind turbines in the wind farm;

[0036] A data collection unit, used to collect the operation data of the wind farm according to the clustering result, and obtain the clustering index data and parameter identification data based on the operation data;

[0037] An online modeling unit, used to perform online modeling on the frequency regulation characteristics of the wind farm based on the grouping index data, and generate a multi-machine equivalent model of the wind farm;

[0038] The identification unit is used to identify the wind farm frequency regulation control parameters based on the established wind farm frequency regulation control system model and according to the parameter identification data.

[0039] Optionally, for a wind farm, clustering the coherent wind turbines in the wind farm includes:

[0040] Grouping based on the types of double-fed and direct-drive generator sets;

[0041] In the case of double-fed and direct-drive generator types, wind turbines are grouped according to active power. If the wind turbine power P = P max , then they are divided into one category;

[0042] If P<P max , then the pitch angles are divided into groups, and the pitch angles β=0 and ≠ are clustered separately;

[0043] According to the speed grouping, according to the speed Ω<Ω m_min ,Ω=Ω m_min ,Ω m_min <Ω<Ω m_max ,Ω=Ω m_max Perform clustering;

[0044] For Ω m_min <Ω<Ω m_max The wind turbines under the fuzzy clustering algorithm are grouped according to the rotation speed.

[0045] Optionally, the clustering indicator data includes at least one of the following:

[0046] Wind turbine active power data, pitch angle data and speed information data.

[0047] Optionally, the parameter identification data includes at least one of the following:

[0048] The frequency of the wind farm grid connection point or the analog signal frequency data of the on-site test, the active power data of the wind farm grid connection point, the superior active power command value data received by the site frequency regulation control system, the active power command value data issued by the site frequency regulation control system to the unit, the wind turbine wind speed data, the wind turbine speed data and the pitch angle data.

[0049] Optionally, the data collection unit is further used to pre-process the clustering index data, including:

[0050] The grouping index data is smoothed to eliminate noise data in the grouping index data.

[0051] Optionally, based on the grouping index data, online modeling is performed for the frequency regulation characteristics of the wind farm to generate a multi-machine equivalent model of the wind farm, including:

[0052] A clustering algorithm based on the load-shedding standby mode and the operating status of the units is used to perform equivalent processing of wind turbine multi-machine clustering on the clustering index data. Based on the processing results of the wind turbine multi-machine clustering, online modeling of the frequency regulation characteristics of the wind farm is performed to generate a multi-machine equivalent model of the wind farm.

[0053] Optionally, identifying the wind farm frequency regulation control parameters according to the parameter identification data includes:

[0054] Power distribution mode identification, power regulation link parameter identification, limiting link parameter identification and primary frequency modulation and inertia response link parameter identification.

[0055] Optionally, power distribution mode identification includes: average distribution mode identification, distribution mode identification based on the proportion of rated power, equal margin distribution mode identification and equal proportion distribution mode identification.

[0056] Optional, power regulation link parameter identification, including:

[0057] Get the input P of the control system at the station plant_ref When the command imposes disturbance or the wind farm is monitored during operation, the dispatcher issues P plant_ref The particle swarm algorithm is used to identify the parameters of the active power regulation link according to the operating data when the instruction changes.

[0058] Optionally, after the power distribution mode identification and the power regulation link parameter identification are performed, the limiting link parameter identification is performed;

[0059] The limiting link parameter identification includes:

[0060] The parameters are identified by defining the amount of power increase.

[0061] Optionally, the identification parameter objects of the limiting link parameter identification include: active power input limiting parameters and active power control link integrator limiting parameters.

[0062] Optional, primary frequency modulation and inertia response link parameter identification, including:

[0063] Dead zone parameter identification, primary frequency modulation droop coefficient identification and moment of inertia constant identification.

[0064] In yet another aspect, the present invention further provides a computing device, comprising: one or more processors;

[0065] a processor for executing one or more programs;

[0066] When the one or more programs are executed by the one or more processors, the above-described method is implemented.

[0067] In yet another aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed, the method described above is implemented.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] The present invention provides a method for online modeling and parameter identification of wind farm frequency regulation characteristics, including: for a wind farm, clustering the coherent wind turbines in the wind farm; collecting the operation data of the wind farm according to the clustering results, and obtaining the clustering index data and parameter identification data based on the operation data; based on the clustering index data, online modeling is performed on the frequency regulation characteristics of the wind farm to generate a multi-machine equivalent model of the wind farm; based on the established wind farm frequency regulation control system model, the wind farm frequency regulation control parameters are identified according to the parameter identification data. The present invention is based on the measurable information of the wind farm, realizes the online modeling and parameter identification of the wind farm frequency regulation characteristics, provides an accurate and online refreshable wind farm equivalent aggregation model for the characteristic analysis of the wind farm participating in the system frequency regulation, and facilitates the frequency stability analysis of the wind farm connected to the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 is a flow chart of the method of the present invention;

[0071] Figure 2 A flow chart of online modeling and parameter identification of wind farm frequency regulation characteristics of row data according to an embodiment of the present invention;

[0072] Figure 3 The speed and pitch angle characteristic diagram of the wind turbine generator set in different standby modes according to the embodiment of the present invention;

[0073] Figure 4 A flow chart of clustering wind turbine components according to an embodiment of the present invention;

[0074] Figure 5 This is a wind farm frequency regulation control block diagram according to an embodiment of the present invention;

[0075] Figure 6 A topological diagram of a wind farm according to an embodiment of the present invention;

[0076] Figure 7 A comparison diagram of the primary frequency modulation output power of a wind turbine generator system according to an embodiment of the present invention;

[0077] Figure 8 A comparison diagram between a frequency modulation simulation and an actual frequency modulation when the frequency decreases in an embodiment of the present invention;

[0078] Fig. 9 A comparison diagram between a frequency modulation simulation and an actual frequency modulation when the frequency rises in an embodiment of the present invention;

[0079] Fig.10 It is a structural diagram of the system of the present invention. DETAILED DESCRIPTION

[0080] Now, exemplary embodiments of the present invention are described with reference to the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to disclose the present invention in detail and completely and to fully convey the scope of the present invention to those skilled in the art. The terms used in the exemplary embodiments shown in the accompanying drawings are not intended to limit the present invention. In the accompanying drawings, the same units / elements are marked with the same reference numerals.

[0081] Unless otherwise specified, the terms (including technical terms) used herein have the commonly understood meanings to those skilled in the art. In addition, it is understood that the terms defined in commonly used dictionaries should be understood to have the same meanings as those in the context of the relevant fields, and should not be understood as idealized or overly formal meanings.

[0082] Embodiment 1:

[0083] The present invention proposes a method for online modeling and parameter identification of wind farm frequency regulation characteristics, such as Figure 1 As shown, including:

[0084] Step 1: for a wind farm, clustering the coherent wind turbines in the wind farm;

[0085] Step 2: According to the clustering result, the operation data of the wind farm is collected, and based on the operation data, the clustering index data and parameter identification data are obtained;

[0086] Step 3: Based on the grouping index data, online modeling is performed for the frequency regulation characteristics of the wind farm to generate a multi-machine equivalent model of the wind farm;

[0087] Step 4: Based on the established wind farm frequency regulation control system model and according to the parameter identification data, the wind farm frequency regulation control parameters are identified.

[0088] Wherein, for a wind farm, clustering the coherent wind turbines in the wind farm includes:

[0089] Grouping based on the types of double-fed and direct-drive generator sets;

[0090] In the case of double-fed and direct-drive generator types, wind turbines are grouped according to active power. If the wind turbine power P = P max , then they are divided into one category;

[0091] If P<P max , then the pitch angles are divided into groups, and the pitch angles β=0 and ≠ are clustered separately;

[0092] According to the speed grouping, according to the speed Ω<Ω m_min ,Ω=Ω m_min ,Ωm_min <Ω<Ω m_max ,Ω=Ω m_max Perform clustering;

[0093] For Ω m_min <Ω<Ω m_max The wind turbines under the fuzzy clustering algorithm are grouped according to the rotation speed.

[0094] The clustering indicator data includes at least one of the following:

[0095] Wind turbine active power data, pitch angle data and speed information data.

[0096] The parameter identification data includes at least one of the following:

[0097] The frequency of the wind farm grid connection point or the analog signal frequency data of the on-site test, the active power data of the wind farm grid connection point, the superior active power command value data received by the site frequency regulation control system, the active power command value data issued by the site frequency regulation control system to the unit, the wind turbine wind speed data, the wind turbine speed data and the pitch angle data.

[0098] The method further includes: preprocessing the clustering indicator data, including:

[0099] The grouping index data is smoothed to eliminate noise data in the grouping index data.

[0100] Based on the grouping index data, online modeling is performed for the frequency regulation characteristics of the wind farm to generate a multi-machine equivalent model of the wind farm, including:

[0101] A clustering algorithm based on the load-shedding standby mode and the operating status of the units is used to perform equivalent processing of wind turbine multi-machine clustering on the clustering index data. Based on the processing results of the wind turbine multi-machine clustering, online modeling of the frequency regulation characteristics of the wind farm is performed to generate a multi-machine equivalent model of the wind farm.

[0102] According to the parameter identification data, the wind farm frequency regulation control parameters are identified, including:

[0103] Power distribution mode identification, power regulation link parameter identification, limiting link parameter identification and primary frequency modulation and inertia response link parameter identification.

[0104] The power distribution mode identification includes: average distribution mode identification, distribution mode identification according to the proportion of rated power, equal margin distribution mode identification and equal proportion distribution mode identification.

[0105] Among them, the power regulation link parameter identification includes:

[0106] Get the input P of the control system at the station plant_ref When the command imposes disturbance or the wind farm is monitored during operation, the dispatcher issues P plant_ref The particle swarm algorithm is used to identify the parameters of the active power regulation link according to the operating data when the instruction changes.

[0107] After the power distribution mode and power regulation link parameter identification are performed, the limiting link parameter identification is performed;

[0108] The limiting link parameter identification includes:

[0109] The parameters are identified by defining the amount of power increase.

[0110] Among them, the identification parameter objects of the limiting link parameter identification include: active power input limiting parameters and active power control link integrator limiting parameters.

[0111] Among them, the parameter identification of primary frequency modulation and inertia response link includes:

[0112] Dead zone parameter identification, primary frequency modulation droop coefficient identification and moment of inertia constant identification.

[0113] The present invention is described below in conjunction with specific implementation cases:

[0114] Implementation steps of the case Figure 2 As shown, including:

[0115] (1) Selection of model clustering indicators;

[0116] For multiple wind turbines in a wind farm, clustering the same-speed wind turbines according to simulation requirements and reducing the complexity of the online model is the first step in equivalent modeling. For the frequency modulation simulation requirements of the equivalent model, the processing method of the grouping index is based on the wind turbines in the wind farm working in two frequency modulation standby modes, one is speed standby and the other is pitch angle standby. First, the two modes are grouped. As shown in the attached Figure 3 The curves of speed and pitch angle changing with wind speed in speed reserve and pitch angle reserve modes are shown. In speed reserve mode, the speed deviates from the optimal speed, the active power is reduced, and power reserve is obtained. In operating areas 1-3, the pitch angle β is kept at 0, while in area 4, the pitch angle is required to be adjusted to avoid excessive power. In pitch angle reserve mode, a smaller power C is obtained by adjusting the pitch angle. P curve.

[0117] By the attached Figure 3 It can be seen that the wind turbine operates in four areas. Wind speed V v ≤V v_in When not working, when V v >V v_inWhen the unit is put into operation, it runs in the low speed area 1, in which Ω = Ω m_min ; As the wind speed increases further, Ω≥Ω m_min The fan then enters region 2, which is the MPPT mode, and the speed Ω is maintained at the optimal tip speed ratio λ with the wind speed V. opt ; As the wind speed increases further, the rotation speed no longer changes, and Ω = Ω m_max , the wind turbine enters area 3. The speeds of areas 3 and 4 are unchanged, except that the wind turbine in area 3 does not enter the full power state, that is, the power P <P max ; and wind speed V ≥ V V_nom After that, the speed and power of the wind turbine will no longer change.

[0118] When load shedding standby is taken into account, there will be certain differences between variable speed standby control and variable pitch angle standby control. The units in area 1 generally do not have load shedding standby. The pitch angle of the units with variable speed standby is generally maintained at 0°, while the pitch angle of the variable pitch standby wind turbine may not be 0°. As the speed increases further, both types of units in working area 2 will reserve reserves on the basis of MPPT. The difference is that the variable speed standby wind turbine will control the speed by controlling the electromagnetic torque, and the pitch angle will not change; the load shedding standby unit will reduce the load by changing the pitch angle control loop. At this time, the pitch angle will increase with the increase of wind speed, and the speed of the variable pitch load shedding unit will also decrease accordingly. Although the two methods control the speed differently, the logic of load shedding standby is the same. The intuitive way to distinguish them is the difference in the pitch angles of the units. The main difference between the two types of units in areas 3 and 4 is still reflected in the pitch angle. Generally, the variable speed control mode will not adjust the pitch angle in area 3, and control the speed through electromagnetic torque until the torque and speed no longer change in area 4, and then control the pitch angle to reduce wind energy absorption. The variable pitch control unit may control the speed through electromagnetic torque in area 3, and the pitch angle will no longer change. In area 4, it can only reduce wind energy absorption by reducing the pitch angle.

[0119] Considering the speed and pitch angle reserve, a wind farm frequency regulation characteristic clustering method using fuzzy clustering (FCM) is proposed. In addition to the conventional clustering based on wind turbine type, the active power P of the wind turbine is selected. m , speed Ω m , pitch angle β as the grouping index, which can be obtained through wind farm SCADA and other monitoring systems. The grouping process is based on the attached Figure 4 As shown, the specific process is as follows.

[0120] 1) Grouping by type of double-fed and direct-drive generator sets;

[0121] 2) Group according to the active power of wind turbines. If P = P max Divide into one category; if P<P max , further grouping;

[0122] 3) Clustering based on pitch angle, with pitch angles β = 0 and ≠ clustered separately;

[0123] 4) Group by speed, according to speed Ω<Ω m_min ,Ω=Ω m_min (Working Area 1), Ω m_min <Ω<Ω m_max (Working Area 2), Ω=Ω m_max (Workspace 3) Clustering;

[0124] 5) Work area 2 is further subdivided according to the rotation speed using fuzzy clustering algorithm.

[0125] (2) Collect operation data and obtain clustering index data and parameter identification data;

[0126] 1) Grouping index data: Through the wind farm panoramic monitoring system or SCADA system, the active power, pitch angle and speed information of wind turbines in the wind farm are collected and obtained as grouping index data;

[0127] 2) Parameter identification data: Taking the primary frequency regulation and inertia response field test of the wind farm and the frequency fluctuation of the wind farm grid connection point as the frequency disturbance event, the wind farm grid connection point frequency and test frequency, the wind farm grid connection point active power, the superior active power command value received by the site frequency regulation control system, the active power command value issued by the site frequency regulation control system to the unit, the wind speed of the wind turbine unit, the speed of the wind turbine unit and the pitch angle are collected.

[0128] 3) Clustering index data processing: The Savitzky-Golay method, also known as the five-point cubic filtering method, is used to smooth the running data, eliminate noise, and avoid frequent changes in clustering indicators. Let x represent the clustering index data, and the data processing process is as follows:

[0129] First, select the current time and the previous 2 points and the next 2 points, a total of 5 points x[-2], x[-1], x[0], x[1], x[2], and construct a quadratic parabola based on these 5 points:

[0130] f(i)=a0+a1i+a2i 2 (1)

[0131] In the formula, i=-2,-1,0,1,2.

[0132] Find the optimal a0, a1, and a2 to minimize the least squares fit. The function of the least squares fit is:

[0133]

[0134] Let the least squares E be the smallest and its derivative equal to 0, that is:

[0135]

[0136] The three partial derivatives of the unknown variables a0, a1 and a2 can give us three equations, which can be combined to obtain a0, a1 and a2.

[0137] (3) Equivalent modeling of multiple machines in wind farms;

[0138] Based on the clustering index obtained from the wind farm operation data, the clustering algorithm proposed in step (1) that takes into account the load reduction standby mode and the unit operation status is used to perform multi-machine clustering of wind turbines.

[0139] Establish a wind farm frequency regulation control system model to form a wind farm frequency regulation equivalent model of station control + wind turbines. The structure diagram of the wind farm frequency regulation control system is shown in the attached figure. Figure 5 shown.

[0140] (4) Parameter identification of wind farm frequency regulation control system;

[0141] according to Figure 3 Structure, wind farm frequency regulation control parameter identification is divided into power distribution mode identification, power regulation link parameter identification, primary frequency regulation and inertia response link parameter identification.

[0142] 1) Power allocation mode identification;

[0143] In the control of wind farms, power allocation is the process of allocating the total power command value of the wind farm to each wind turbine in a certain proportion. Different power allocation methods directly affect the dynamic characteristics of the wind farm. Therefore, in the process of identifying the field control, the power allocation method used should be identified first. There are four main power allocation methods in actual wind farms: equal margin, equal proportion, average allocation and capacity.

[0144] First, there are four main power allocation modes:

[0145] ①Equal distribution;

[0146]

[0147] The active reference value is evenly distributed to each operating unit in the wind farm. The reference power of each unit is the total reference power of the site divided by the total number of allocable units. ord is the total power value command value of wind farm frequency regulation, p ordi is the frequency regulation active power command value of the i-th wind turbine.

[0148] ② Allocate in proportion to rated power;

[0149] Since the models and rated powers of the units in the station are different, the output capacity of each unit varies. At this time, the power output is generally allocated according to the rated power ratio of each unit. At this time, the output power reference value of each unit is:

[0150]

[0151] Among them, p ni is the rated power of the i-th unit.

[0152] ③Equal margin allocation;

[0153]

[0154] Among them, p i is the current active power of the i-th wind turbine, l pi is the adjustable margin of active power of the i-th wind turbine, l pi It can be calculated from the wind speed, rotation speed and pitch angle of the wind turbine.

[0155] ④Equal proportion distribution

[0156] According to the measured wind speed value of each unit, the units with larger wind power values ​​are allocated more. At this time, the reference value of the output power of each unit is:

[0157]

[0158] Where: v i is the wind power value of the i-th unit.

[0159] Secondly, according to the calculation principle of the power allocation mode, the power allocation mode is identified. If the frequency modulation power instructions of all wind turbines are the same, it is an average allocation; otherwise, if the frequency modulation power instruction ratio of each wind turbine is the same as the rated power ratio of the wind turbine, it is allocated according to the rated power ratio; otherwise, if the frequency modulation power instruction ratio is the same as the cubic ratio of the wind speed of each wind turbine, it is an equal proportion allocation; otherwise, it is an equal margin allocation.

[0160] 2) Parameter identification of active power regulation link;

[0161] The wind farm active power PI regulator parameter identification of the present invention is to input P plant_ref When a disturbance is applied or a dispatching instruction is monitored during wind farm operation, P plant_ref The operating data when the command changes, the particle swarm algorithm is used to identify the parameters of the active power regulation link, and the signal transfer function and parameters are shown in formula (8).

[0162]

[0163] P plant As input quantity, P ord As the output, the current power command response is obtained. plant_ref When a disturbance is applied, there is also a power command response. Subtracting the previous power command response from the current power command response can obtain the following transfer function response characteristics:

[0164]

[0165] The first-order delay coefficient of the active power measurement link can be obtained through particle swarm algorithm identification.

[0166] 3) Identification of limiting link parameters;

[0167] The identification of the limiting link needs to be carried out after the identification of the above parameters is completed. The main identification object is the active power input limiting parameter P ermax , P ermin and active power control link integrator limit parameter P poImax , P poImin Two types of parameters are identified. Among them, the two types of parameters are identified by defining the power increase variation δ.

[0168]

[0169] Input a set of active power command step signals P with equal amplitude and increasing step length plant_ref , and record each group of output signals. By calculating the power rise change rate δ in the first n seconds, it can be found that as the input step signal increases, the change rate δ at the same time should increase. Due to the existence of the limit parameter, there must be an input threshold. After reaching the threshold, the change rate δ has relatively no change in the first n seconds, which means that the input signal has reached the active power input limit parameter P at this time. ermax , P ermin This threshold is equal to P ref The step signal amplitude. After identification, P ermax , P ermin .

[0170] For the integrator limiting parameter P poImax , P poImin Identification, when the integrator reaches the limit value P ermax , P ermin After that, the control signal cannot achieve zero-difference regulation, and the feedback signal P plant Unable to track input command signal P plant_ref In combination with this feature, a set of active power command step signals P with equal amplitude and increasing step length are input. plant_ref , observe the steady-state feedback signal P plant , when P plantWhen the command signal cannot be tracked, the steady-state active power command P is calculated. plant_ref Subtract the wind farm feedback signal P plant The difference ΔP is

[0171] P poImax poImin =(1-K P )(P plant_ref -P plant ) (11)

[0172] Finally, we identify P poImax ,P poImin .

[0173] 4) Identification of primary frequency modulation and inertia response links;

[0174] according to Figure 4 It can be seen that the parameters of the frequency modulation and inertia response link include the primary frequency modulation dead zone parameter F dbd1 / dbd2 , primary frequency modulation droop coefficient D, inertia response dead zone parameter F dbd3 / dbd4 , moment of inertia constant T j .

[0175] ① Dead zone parameter identification;

[0176] The dead zone parameters of primary frequency regulation and inertia response are usually determined by the wind farm after communication with the dispatcher. The parameters are fixed, so the dead zone parameters do not need to be identified and can be obtained directly.

[0177] ② Identification of primary frequency modulation droop coefficient;

[0178] When passing the primary frequency regulation test of the wind farm, the ref The injected frequency step up or down disturbance, the grid frequency changes when the wind farm is running and the wind farm active power command P plant_ref The disturbance condition data for frequency modulation parameter identification is obtained. ord The signal of (the sum of the power commands of all wind turbines) can be used to identify the primary frequency modulation droop coefficient D. plant_ref and frequency signal Freq ref As input, P ord As the output, the transfer function can be expressed as formula (12). The stable frequency change brings about the change in active power. Further ignoring the response process of the power regulation link, the steady-state transfer function of the primary frequency modulation is rewritten as formula (13), and the droop coefficient D can be calculated.

[0179]

[0180] ③Identification of moment of inertia constant;

[0181] By F req Apply a frequency-varying disturbance signal and observe Pord Signal. The control system transfer function is shown in equation (14), and the improved least squares method can be used to obtain T J .

[0182]

[0183] (5) Model accuracy verification and online update;

[0184] Taking the grid frequency change as the event drive, the same frequency change is simulated in the equivalent model to calculate the active power P measured at the wind farm grid connection point. meas Simulate active power P with equivalent model sim The absolute value of the average error is shown in formula (15).

[0185]

[0186] Where i is the event sequence of each sampled value, and N is the total number of sampled values.

[0187] When the average model error exceeds 5% of the installed capacity of the wind farm, the model clustering and parameter identification are re-performed according to the operating data recorded by the wind farm frequency disturbance event to achieve online updating of the equivalent model.

[0188] According to the frequency regulation standby mode and control characteristics of wind turbines, the present invention uses the active power, rotation speed and pitch angle of wind turbines as grouping indicators, and adopts fuzzy clustering to group and cluster the frequency regulation characteristic model of wind farms. Based on the frequency regulation characteristic test and grid-connected operation data of wind farms, grouping indicators and parameter identification data are obtained, and based on the power distribution mode analysis and the station frequency regulation controller structure, the power distribution mode, power regulation, primary frequency regulation and inertia response link parameter identification are performed to realize the online modeling of the frequency regulation characteristics of wind farms. Finally, the model accuracy error calculation method is used as the online model update judgment index to realize the automatic update of the model. The method uses the measurable information of the wind farm as the basis to realize the online modeling and parameter identification of the frequency regulation characteristics of the wind farm, and provides an accurate and online refreshable wind farm equivalent aggregation model for the characteristic analysis of the wind farm participating in the system frequency regulation, which is convenient for the frequency stability analysis of the wind farm connected to the power grid.

[0189] Taking a wind farm with 20 units and a capacity of 2.5MW as the research object, the implementation process and results of the present invention are introduced. The topological structure of the wind farm is as follows Figure 6As shown in Table 1, there are two types of wind turbines. The operating status of wind turbines before frequency disturbance is shown in Table 1. First, they are divided into Class A and Class B units according to the type of wind turbines. Secondly, based on the clustering index and fuzzy clustering algorithm, Class A and Class B units are clustered to obtain the multi-machine clustering results of wind turbines, as shown in Table 2. Class A units are divided into three groups: A1, A2 and A3, and Class B units are divided into three groups: B1, B2 and B3.

[0190] Table 1

[0191]

[0192] Table 2

[0193] Cluster Group A Cluster B group Cluster A1 1，4，7 Cluster B1 11，14 Cluster A2 2，5，8 Cluster B2 12，15，17，18 Cluster A3 3，6，9，10 Cluster B3 13，16，19，20

[0194] Figure 7 The figure shows the simulation comparison of the response characteristics of the mechanical power and electromagnetic power simulation of the aggregated equivalent machine and the aggregated multiple wind turbines under frequency disturbance. It can be seen that the output characteristics of the mechanical and electromagnetic power equivalents before and after frequency modulation are consistent.

[0195] The data recorded during the frequency disturbance event during the operation of the wind farm are collected to identify the model parameters. First, the wind farm power distribution is identified using an average distribution method. Then, the wind farm frequency control parameter identification results are obtained through the parameter identification algorithm, as shown in Table 3. The comparison between the model simulation and the actual operation characteristics under the frequency upward and downward disturbances is shown in Table 3. Figure 8 and Fig. 9 shown.

[0196] Table 3

[0197]

[0198] Embodiment 2:

[0199] The present invention also proposes a system 200 for online modeling and parameter identification of wind farm frequency regulation characteristics, such as Fig.10 As shown, including:

[0200] A grouping unit 201 is used for grouping and clustering the coherent wind turbines in the wind farm;

[0201] A data collection unit 202 is used to collect the operation data of the wind farm according to the clustering result, and obtain clustering index data and parameter identification data based on the operation data;

[0202] An online modeling unit 203 is used to perform online modeling on the frequency regulation characteristics of the wind farm based on the grouping index data to generate a multi-machine equivalent model of the wind farm;

[0203] The identification unit 204 is used to identify the wind farm frequency regulation control parameters based on the established wind farm frequency regulation control system model and according to the parameter identification data.

[0204] Wherein, for a wind farm, clustering the coherent wind turbines in the wind farm includes:

[0205] Grouping based on the types of double-fed and direct-drive generator sets;

[0206] In the case of double-fed and direct-drive generator types, wind turbines are grouped according to active power. If the wind turbine power P = P max , then they are divided into one category;

[0207] If P<P max , then the pitch angles are divided into groups, and the pitch angles β=0 and ≠ are clustered separately;

[0208] According to the speed grouping, according to the speed Ω<Ω m_min ,Ω=Ω m_min ,Ω m_min <Ω<Ω m_max ,Ω=Ω m_max Perform clustering;

[0209] For Ω m_min <Ω<Ω m_max The wind turbines under the fuzzy clustering algorithm are grouped according to the rotation speed.

[0210] The clustering indicator data includes at least one of the following:

[0211] Wind turbine active power data, pitch angle data and speed information data.

[0212] The parameter identification data includes at least one of the following:

[0213] The frequency of the wind farm grid connection point or the analog signal frequency data of the on-site test, the active power data of the wind farm grid connection point, the superior active power command value data received by the site frequency regulation control system, the active power command value data issued by the site frequency regulation control system to the unit, the wind turbine wind speed data, the wind turbine speed data and the pitch angle data.

[0214] The data collection unit 202 is also used to pre-process the grouping index data, including:

[0215] The grouping index data is smoothed to eliminate noise data in the grouping index data.

[0216] Based on the grouping index data, online modeling is performed for the frequency regulation characteristics of the wind farm to generate a multi-machine equivalent model of the wind farm, including:

[0217] A clustering algorithm based on the load-shedding standby mode and the operating status of the units is used to perform equivalent processing of wind turbine multi-machine clustering on the clustering index data. Based on the processing results of the wind turbine multi-machine clustering, online modeling of the frequency regulation characteristics of the wind farm is performed to generate a multi-machine equivalent model of the wind farm.

[0218] According to the parameter identification data, the wind farm frequency regulation control parameters are identified, including:

[0219] Power distribution mode identification, power regulation link parameter identification, limiting link parameter identification and primary frequency modulation and inertia response link parameter identification.

[0220] The power distribution mode identification includes: average distribution mode identification, distribution mode identification according to the proportion of rated power, equal margin distribution mode identification and equal proportion distribution mode identification.

[0221] Among them, the power regulation link parameter identification includes:

[0222] Get the input P of the control system at the station plant_ref When the command imposes disturbance or the wind farm is monitored during operation, the dispatcher issues P plant_ref The particle swarm algorithm is used to identify the parameters of the active power regulation link according to the operating data when the instruction changes.

[0223] After the power distribution mode and power regulation link parameter identification are performed, the limiting link parameter identification is performed;

[0224] The limiting link parameter identification includes:

[0225] The parameters are identified by defining the amount of power increase.

[0226] Among them, the identification parameter objects of the limiting link parameter identification include: active power input limiting parameters and active power control link integrator limiting parameters.

[0227] Among them, the parameter identification of primary frequency modulation and inertia response link includes:

[0228] Dead zone parameter identification, primary frequency modulation droop coefficient identification and moment of inertia constant identification.

[0229] The present invention is based on the measurable information of the wind farm, realizes the online modeling and parameter identification of the frequency regulation characteristics of the wind farm, provides an accurate and online refreshable wind farm equivalent aggregation model for the characteristic analysis of the wind farm participating in the system frequency regulation, and facilitates the frequency stability analysis of the wind farm connected to the power grid.

[0230] Embodiment 3:

[0231] Based on the same inventive concept, the present invention also provides a computer device, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding functions, so as to implement the steps of the method in the above embodiment.

[0232] Embodiment 4:

[0233] Based on the same inventive concept, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both a built-in storage medium in a computer device and an extended storage medium supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the steps of the method in the above embodiment.

[0234] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0235] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0236] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0237] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0238] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0239] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for online modeling and parameter identification of wind farm frequency regulation characteristics, characterized in that: include: For a wind farm, clustering the coherent wind turbines in the wind farm; According to the clustering result, the operation data of the wind farm is collected, and based on the operation data, the clustering index data and the parameter identification data are obtained; Based on the grouping index data, online modeling is performed for the frequency regulation characteristics of the wind farm to generate a multi-machine equivalent model of the wind farm; Based on the established wind farm frequency regulation control system model, the wind farm frequency regulation control parameters are identified according to the parameter identification data.

2. The method according to claim 1, characterized in that The method of clustering the coherent wind turbines in the wind farm comprises: Grouping based on the types of double-fed and direct-drive generator sets; In the case of double-fed and direct-drive generator types, wind turbines are grouped according to active power. If the wind turbine power P = P max , then they are divided into one category; If P<P max , then the pitch angles are divided into groups, and the pitch angles β=0 and ≠ are clustered separately; According to the speed grouping, according to the speed Ω<Ω m_min 、Ω=Ω m_min ,Ω m_min <Ω<Ω m_max 、Ω=Ω m_max Perform clustering; For Ω m_min <Ω<Ω m_max The wind turbines under the fuzzy clustering algorithm are grouped according to the rotation speed.

3. The method according to claim 1, characterized in that The grouping indicator data includes at least one of the following: Wind turbine active power data, pitch angle data and speed information data.

4. The method according to claim 1, characterized in that The parameter identification data includes at least one of the following: The frequency of the wind farm grid connection point or the analog signal frequency data of the on-site test, the active power data of the wind farm grid connection point, the superior active power command value data received by the site frequency regulation control system, the active power command value data issued by the site frequency regulation control system to the unit, the wind turbine wind speed data, the wind turbine speed data and the pitch angle data.

5. The method according to claim 1, characterized in that: The method further includes: preprocessing the grouping index data, including: The grouping index data is smoothed to eliminate noise data in the grouping index data.

6. The method according to claim 1, characterized in that Based on the grouping index data, online modeling is performed for the frequency regulation characteristics of the wind farm to generate a multi-machine equivalent model of the wind farm, including: A clustering algorithm based on the load-shedding standby mode and the operating status of the units is used to perform equivalent processing of wind turbine multi-machine clustering on the clustering index data. Based on the processing results of the wind turbine multi-machine clustering, online modeling of the frequency regulation characteristics of the wind farm is performed to generate a multi-machine equivalent model of the wind farm.

7. The method according to claim 1, characterized in that The step of identifying the wind farm frequency modulation control parameters according to the parameter identification data includes: Power distribution mode identification, power regulation link parameter identification, limiting link parameter identification and primary frequency modulation and inertia response link parameter identification.

8. The method according to claim 7, characterized in that The power distribution mode identification includes: average distribution mode identification, rated power proportional distribution mode identification, equal margin distribution mode identification and equal proportion distribution mode identification.

9. The method according to claim 7, characterized in that: The power regulation link parameter identification includes: Get the input P of the control system at the station plant_ref When the command imposes disturbance or the wind farm is monitored during operation, the dispatcher issues P plant_ref The particle swarm algorithm is used to identify the parameters of the active power regulation link according to the operating data when the instruction changes.

10. The method according to claim 7, characterized in that After the power distribution mode and power regulation link parameter identification are performed, the limiting link parameter identification is performed; The limiting link parameter identification includes: The parameters are identified by defining the amount of power increase.

11. The method according to claim 10, characterized in that The identification parameter objects of the limiting link parameter identification include: active power input limiting parameters and active power control link integrator limiting parameters.

12. The method according to claim 7, characterized in that The primary frequency modulation and inertia response link parameter identification includes: Dead zone parameter identification, primary frequency modulation droop coefficient identification and moment of inertia constant identification.

13. A system for online modeling and parameter identification of wind farm frequency regulation characteristics, characterized in that: include: A grouping unit, for grouping and clustering the coherent wind turbines in the wind farm; A data collection unit, used to collect the operation data of the wind farm according to the clustering result, and obtain the clustering index data and parameter identification data based on the operation data; An online modeling unit, used to perform online modeling on the frequency regulation characteristics of the wind farm based on the grouping index data, and generate a multi-machine equivalent model of the wind farm; The identification unit is used to identify the wind farm frequency regulation control parameters based on the established wind farm frequency regulation control system model and according to the parameter identification data.

14. The system according to claim 13, characterized in that The method of clustering the coherent wind turbines in the wind farm comprises: Grouping based on the types of double-fed and direct-drive generator sets; In the case of double-fed and direct-drive generator types, wind turbines are grouped according to active power. If the wind turbine power P = P max , then they are divided into one category; If P<P max , then the pitch angles are divided into groups, and the pitch angles β=0 and ≠ are clustered separately; According to the speed grouping, according to the speed Ω<Ω m_min 、Ω=Ω m_min ,Ω m_min <Ω<Ω m_max 、Ω=Ω m_max Perform clustering; For Ω m_min <Ω<Ω m_max The wind turbines under the fuzzy clustering algorithm are grouped according to the rotation speed.

15. The system according to claim 13, characterized in that The grouping indicator data includes at least one of the following: Wind turbine active power data, pitch angle data and speed information data.

16. The system according to claim 13, characterized in that The parameter identification data includes at least one of the following: The frequency of the wind farm grid connection point or the analog signal frequency data of the on-site test, the active power data of the wind farm grid connection point, the superior active power command value data received by the site frequency regulation control system, the active power command value data issued by the site frequency regulation control system to the unit, the wind turbine wind speed data, the wind turbine speed data and the pitch angle data.

17. The system according to claim 13, characterized in that The data collection unit is also used to pre-process the clustering index data, including: The grouping index data is smoothed to eliminate noise data in the grouping index data.

18. The system according to claim 13, characterized in that Based on the grouping index data, online modeling is performed for the frequency regulation characteristics of the wind farm to generate a multi-machine equivalent model of the wind farm, including: A clustering algorithm based on the load-shedding standby mode and the operating status of the units is used to perform equivalent processing of wind turbine multi-machine clustering on the clustering index data. Based on the processing results of the wind turbine multi-machine clustering, online modeling of the frequency regulation characteristics of the wind farm is performed to generate a multi-machine equivalent model of the wind farm.

19. The system according to claim 13, characterized in that The identification unit identifies the wind farm frequency modulation control parameters according to the parameter identification data, including: Power distribution mode identification, power regulation link parameter identification, limiting link parameter identification and primary frequency modulation and inertia response link parameter identification.

20. The system according to claim 19, characterized in that The power distribution mode identification includes: average distribution mode identification, rated power proportional distribution mode identification, equal margin distribution mode identification and equal proportion distribution mode identification.

21. The system according to claim 19, characterized in that The power regulation link parameter identification includes: Get the input P of the control system at the station plant_ref When the command imposes disturbance or the wind farm is monitored during operation, the dispatcher issues P plant_ref The particle swarm algorithm is used to identify the parameters of the active power regulation link according to the operating data when the instruction changes.

22. The system according to claim 19, characterized in that The identification unit performs the identification of the parameters of the limiting link after the identification of the power distribution mode and the power regulation link parameters; The limiting link parameter identification includes: The parameters are identified by defining the amount of power increase.

23. The system according to claim 22, characterized in that The identification parameter objects of the limiting link parameter identification include: active power input limiting parameters and active power control link integrator limiting parameters.

24. The system according to claim 19, characterized in that The primary frequency modulation and inertia response link parameter identification includes: Dead zone parameter identification, primary frequency modulation droop coefficient identification and moment of inertia constant identification.

25. A computer device, characterized in that: include: one or more processors; a processor for executing one or more programs; When the one or more programs are executed by the one or more processors, the method according to any one of claims 1 to 12 is implemented.

26. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed, the method according to any one of claims 1 to 12 is implemented.

Images