A wind farm dispatching method and system based on fatigue distribution

Abstract

This application relates to the field of wind farm technology, and in particular to a wind farm dispatching method and system based on fatigue distribution. It includes: establishing multiple wind turbine locations based on wind turbine parameters; establishing multiple regulation cycles based on historical environmental parameters; generating the expected power of each wind turbine location within the current regulation cycle based on a preset power prediction model; generating an initial dispatching strategy based on the grid dispatching instructions within the current regulation cycle and the expected power of all wind turbine locations; obtaining fatigue evaluation values ​​for each wind turbine location based on preset feedback time nodes; determining whether to modify the initial dispatching strategy based on all fatigue evaluation values; generating the initial dispatching strategy based on the operating parameters of each wind turbine; and optimizing and adjusting the initial dispatching strategy based on all fatigue evaluation values. This optimizes the uniformity of fatigue distribution in the wind farm without sacrificing power generation, extends the service life of the turbines, and reduces the operation and maintenance costs of the wind farm.

Description

Technical Field

[0001] This application relates to the field of wind farm technology, and in particular to a wind farm scheduling method and system based on fatigue distribution. Background Technology

[0002] In mountainous and hilly areas, the elevation difference of the wind turbine installation location is large, and the change in elevation will affect the change in wind speed; coupled with the effect of wake, the wind speed input to each unit is seriously uneven, resulting in an uncertain load.

[0003] Currently, most wind farm active power control strategies focus on whether the output power can meet the needs of the power grid dispatch center, thus ignoring the wind farm's own operating conditions. This increases the operating burden of the units, ultimately leading to higher failure rates, reduced service life, and poorer overall economic efficiency of the wind farm. Summary of the Invention

[0004] The purpose of this application is to provide a wind farm scheduling method and system based on fatigue distribution to solve the above-mentioned technical problems, aiming to improve the operating efficiency and operating benefits of wind turbine units, extend the service life of the units, and reduce the operation and maintenance costs of wind farms.

[0005] In some embodiments of this application, an initial scheduling strategy is generated based on the operating parameters of each wind turbine, and a fatigue evaluation value for each wind turbine is generated based on a preset feedback time node. The initial scheduling strategy is then optimized and adjusted based on all fatigue evaluation values. Without losing power generation, the uniformity of fatigue distribution in the wind farm is optimized, the service life of the turbines is extended, and the operation and maintenance costs of the wind farm are reduced.

[0006] In some embodiments of this application, by periodically monitoring the uniformity of fatigue distribution in the wind farm, the operating parameters of excessively worn wind turbine units are adjusted in a timely manner. The fatigue characteristics of the wind turbine units are improved by controlling the unit speed, pitch angle, or by shutdown maintenance, so as to avoid the reduction in operating efficiency and service life of the wind turbine units due to excessive wear.

[0007] In some embodiments of this application, a wind farm dispatching method based on fatigue distribution is provided, including:

[0008] Establish multiple wind turbine points based on wind turbine parameters;

[0009] Multiple adjustment cycles are established based on historical environmental parameters, and the expected power of each wind turbine point in the current adjustment cycle is generated based on the preset power prediction model.

[0010] An initial dispatch strategy is generated based on the grid dispatch instructions within the current adjustment cycle and the expected power of all wind turbines.

[0011] The fatigue evaluation value of each wind turbine is obtained according to the preset feedback time node, and the initial scheduling strategy is determined based on all fatigue evaluation values.

[0012] When establishing multiple wind turbine locations, this includes:

[0013] Generate a sequence of wind turbine points A, A=(a1,a2…a ... i …a n ), where a i Let n be the i-th wind turbine location, and n be the number of wind turbine locations.

[0014] In some embodiments of this application, the generation of the initial scheduling policy includes:

[0015] Establish the expected power sequence P, P=(p1,p2…p) within the current adjustment period. i …p n ), where pi is the expected power of the i-th wind turbine point in the current adjustment cycle;

[0016] The required power b' is generated according to the power grid dispatch instructions;

[0017] Generate the first-level power of each wind turbine point in the current adjustment cycle based on the expected power sequence P and the demand power b'.

[0018] Establish a first-order power sequence B, B=(b1, b2…b i …b n ),b i The first-level power of the i-th fan point during the current adjustment cycle;

[0019] Among them, b i =b'*[p i / ( p i )];

[0020] The initial scheduling strategy for the current adjustment cycle is generated based on the first-level power sequence B.

[0021] In some embodiments of this application, when determining whether to modify the initial scheduling strategy based on all fatigue evaluation values, the following are included:

[0022] Set multiple feedback time points within the current adjustment cycle;

[0023] Obtain the fatigue evaluation values ​​of each wind turbine point at the current feedback time point;

[0024] Establish a fatigue evaluation value sequence C, C=(c1,c2…c i …c n ), where c i This represents the fatigue evaluation value of the i-th fan at the current feedback time point;

[0025] Generate a corrected evaluation value g for the current feedback time node based on the fatigue evaluation value sequence C;

[0026] Judge whether to generate a correction instruction according to the corrected evaluation value g.

[0027] In some embodiments of the present application, when generating the corrected evaluation value g for the current feedback time node, it includes:

[0028] g = e1 * Q1 * (c i - c') 2 + e2 * Q2 * (c i - ∆c) 2 + e3 * Q3 * U;

[0029] U = Y(i) * (c i - c');

[0030] Where, e1 is a preset first weight coefficient; e2 is a preset second weight coefficient; e3 is a preset third weight coefficient; Q1 is a preset first fixed coefficient; Q2 is a preset second fixed coefficient; Q3 is a preset third fixed coefficient; c' is the expected fatigue evaluation value of the current feedback time node; ∆c is the average value of all data in the fatigue evaluation value sequence C; Y(i) is a selection coefficient. If (c i - c') > 0, Y(i) = 1 / (c i - c'); if (c i - c') < 0, Y(i) = 0.

[0031] In some embodiments of the present application, when judging whether to generate a correction instruction according to the corrected evaluation value g, it includes:

[0032] Preset a first corrected evaluation value threshold G1 and a second corrected evaluation value threshold G2;

[0033] If g < G1, no correction instruction is generated;

[0034] If G1 ≤ g < G2, generate a first-level correction instruction, and generate a power compensation coefficient for each fan point according to the first-level correction instruction;

[0035] Correct the initial scheduling strategy according to all power compensation coefficients;

[0036] If g ≥ G2, generate a second-level correction instruction and correct the initial scheduling strategy according to the second-level correction instruction.

[0037] In some embodiments of the present application, when correcting the initial scheduling strategy according to the second-level correction instruction, it includes:

[0038] Generate the fatigue evaluation value threshold C1 for the current feedback time point;

[0039] If c i If C1 is selected, the i-th fan point is removed, and the i-th fan point is set as the fan point to be inspected.

[0040] A maintenance plan is generated based on all the wind turbines to be inspected.

[0041] Based on the elimination results, establish a sequence A1 of primary wind turbine points at the current feedback time node;

[0042] A1=(a 11 ,a 12 …a 1i …a 1n1 ), where is the i-th wind turbine point that has not been removed at the current feedback time point; n1 is the number of wind turbine points that have not been removed at the current feedback time point, and n1≤n;

[0043] The allocation coefficients for each primary wind turbine are generated based on the expected power and fatigue evaluation values ​​of each primary wind turbine.

[0044] The first-level scheduling strategy for the current feedback time node is generated based on the allocation coefficient and the required power b'.

[0045] In some embodiments of this application, a wind farm dispatching system based on fatigue distribution is provided, comprising:

[0046] The central control unit is used to establish multiple wind turbine points based on the wind turbine parameters;

[0047] The monitoring unit includes multiple monitoring sub-modules, which are set at the wind turbine points to collect the operating parameters of each wind turbine point.

[0048] The central control unit includes:

[0049] The first processing module is used to establish multiple adjustment cycles based on historical environmental parameters. The first processing module is also used to generate the expected power of each wind turbine point in the current adjustment cycle based on a preset power prediction model.

[0050] The second processing module is used to generate an initial scheduling strategy based on the grid dispatch instructions and the expected power of all wind turbines in the current adjustment cycle.

[0051] The third processing module is used to obtain the fatigue evaluation value of each wind turbine point according to the preset feedback time node;

[0052] The correction module is used to determine whether to correct the initial scheduling strategy based on all fatigue evaluation values.

[0053] The fourth processing module is used to generate the sequence of wind turbine points A, A=(a1,a2…a…).i …a n ), where a i Let i be the i-th wind turbine point, and n be the number of wind turbine points;

[0054] The second processing module is also used for:

[0055] Establish the expected power sequence P, P=(p1,p2…p) within the current adjustment period. i …p n ), where pi is the expected power of the i-th wind turbine point in the current adjustment cycle;

[0056] The required power b' is generated according to the power grid dispatch instructions;

[0057] Generate the first-level power of each wind turbine point in the current adjustment cycle based on the expected power sequence P and the demand power b'.

[0058] Establish a first-order power sequence B, B=(b1, b2…b i …b n ),b i The first-level power of the i-th fan point during the current adjustment cycle;

[0059] Among them, b i =b'*[p i / ( p i )];

[0060] The initial scheduling strategy for the current adjustment cycle is generated based on the first-level power sequence B.

[0061] In some embodiments of this application, the third processing module is further configured to:

[0062] Set multiple feedback time points within the current adjustment cycle;

[0063] Obtain the fatigue evaluation values ​​of each wind turbine point at the current feedback time point;

[0064] Establish a fatigue evaluation value sequence C, C=(c1,c2…c i …c n ), where c i This represents the fatigue evaluation value of the i-th fan at the current feedback time point.

[0065] In some embodiments of this application, the correction module is further configured to:

[0066] Generate the corrected evaluation value g for the current feedback time point based on the fatigue evaluation value sequence C;

[0067] Whether to generate a correction instruction is determined based on the correction evaluation value g.

[0068] g = e1 * Q1 * (c i - c') 2 + e2 * Q2 * (c i - ∆c) 2 + e3 * Q3 * U;

[0069] U = Y(i) * (c i - c');

[0070] Where, e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; e3 is the preset third weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; Q3 is the preset third fixed coefficient; c' is the expected fatigue evaluation value at the current feedback time node; ∆c is the average value of all data in the fatigue evaluation value sequence C; Y(i) is the selection coefficient. If (c i - c') > 0, Y(i) = 1 / (c i - c'); if (c i - c') < 0, Y(i) = 0;

[0071] Judge whether to generate a correction instruction according to the corrected evaluation value g.

[0072] In some embodiments of the present application, when judging whether to generate a correction instruction according to the corrected evaluation value g, it includes:

[0073] Preset the first corrected evaluation value threshold G1 and the second corrected evaluation value threshold G2;

[0074] If g < G1, no correction instruction is generated;

[0075] If G1 ≤ g < G2, generate a first-level correction instruction, and generate a power compensation coefficient for each wind turbine point according to the first-level correction instruction;

[0076] Correct the initial scheduling strategy according to all power compensation coefficients;

[0077] If g ≥ G2, generate a second-level correction instruction, and correct the initial scheduling strategy according to the second-level correction instruction.

[0078] Compared with the prior art, the beneficial effects of a wind farm scheduling method and system based on fatigue distribution in the embodiments of the present application are as follows:

[0079] Generate an initial scheduling strategy according to the operating parameters of each wind turbine, generate the fatigue evaluation value of each wind turbine according to the preset feedback time node, and optimize and adjust the initial scheduling strategy according to all fatigue evaluation values. Without sacrificing power generation, the uniformity of the fatigue distribution of the wind farm is optimized, the service life of the unit is extended, and the operation and maintenance cost of the wind farm is reduced.

[0080] By periodically monitoring the uniformity of fatigue distribution in wind farms, the operating parameters of excessively worn wind turbines can be adjusted in a timely manner. By controlling the turbine speed, pitch angle, or performing maintenance shutdowns, the fatigue characteristics of wind turbines can be improved, thus avoiding a decrease in operating efficiency and a reduction in service life due to excessive wear. Attached Figure Description

[0081] Figure 1 This is a flowchart illustrating a wind farm scheduling method based on fatigue distribution in a preferred embodiment of this application. Detailed Implementation

[0082] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0083] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0084] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0085] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0086] like Figure 1 As shown in the preferred embodiment of this application, a wind farm dispatching method based on fatigue distribution includes:

[0087] S101: Establish multiple wind turbine points based on wind turbine parameters;

[0088] S102: Establish multiple adjustment cycles based on historical environmental parameters, and generate the expected power of each wind turbine point in the current adjustment cycle based on the preset power prediction model;

[0089] S103: Generate an initial dispatch strategy based on the grid dispatch instructions and the expected power of all wind turbines in the current regulation cycle;

[0090] S104: Obtain the fatigue evaluation value of each wind turbine according to the preset feedback time node, and determine whether to correct the initial scheduling strategy based on all fatigue evaluation values.

[0091] When establishing multiple wind turbine locations, this includes:

[0092] Generate a sequence of wind turbine points A, A=(a1,a2…a ... i …a n ), where a i Let n be the i-th wind turbine location, and n be the number of wind turbine locations.

[0093] Specifically, a single wind turbine represents a single wind turbine unit. The duration of a single adjustment cycle is set based on historical demand fluctuation parameters of the power grid and historical environmental fluctuation parameters. The more severe the fluctuations, the shorter the corresponding adjustment cycle duration, thereby ensuring the accuracy of the power prediction model's prediction values ​​within a single adjustment cycle and improving the scheduling efficiency for all wind turbine units in the wind farm.

[0094] Specifically, generating the initial scheduling policy includes:

[0095] Establish the expected power sequence P, P=(p1,p2…p) within the current adjustment period. i …p n ), where pi is the expected power of the i-th wind turbine point in the current adjustment cycle;

[0096] The required power b' is generated according to the power grid dispatch instructions;

[0097] Generate the first-level power of each wind turbine point in the current adjustment cycle based on the expected power sequence P and the demand power b'.

[0098] Establish a first-order power sequence B, B=(b1, b2…b i …b n ),b i The first-level power of the i-th fan point during the current adjustment cycle;

[0099] Among them, b i =b'*[p i / ( p i )];

[0100] The initial scheduling strategy for the current adjustment cycle is generated based on the first-level power sequence B.

[0101] Specifically, the expected power refers to the maximum active power that the corresponding wind turbine can provide within the current adjustment cycle, and the corresponding initial scheduling strategy is generated based on the averaging principle.

[0102] It is understood that in the above embodiments, by generating an initial scheduling strategy based on the operating parameters of each wind turbine, generating fatigue evaluation values ​​for each wind turbine based on preset feedback time nodes, and optimizing and adjusting the initial scheduling strategy based on all fatigue evaluation values, the uniformity of fatigue distribution in the wind farm is optimized without losing power generation, thereby extending the service life of the turbines and reducing the operation and maintenance costs of the wind farm.

[0103] In a preferred embodiment of this application, when determining whether to modify the initial scheduling strategy based on all fatigue evaluation values, the following steps are included:

[0104] Set multiple feedback time points within the current adjustment cycle;

[0105] Obtain the fatigue evaluation values ​​of each wind turbine point at the current feedback time point;

[0106] Establish a fatigue evaluation value sequence C, C=(c1,c2…c i …c n ), where c i This represents the fatigue evaluation value of the i-th fan at the current feedback time point;

[0107] Generate the corrected evaluation value g for the current feedback time point based on the fatigue evaluation value sequence C;

[0108] Whether to generate a correction instruction is determined based on the correction evaluation value g.

[0109] Specifically, the fatigue load of a wind turbine is concentrated in the transmission system load caused by shaft torsion and the tower structure load caused by tower deflection. The fatigue load of a wind turbine can be judged by parameters such as shaft torque and tower thrust, thereby generating a corresponding fatigue evaluation value. The larger the fatigue evaluation value, the greater the fatigue degree of the current wind turbine.

[0110] Specifically, by establishing multiple feedback time nodes, the fatigue status of each wind turbine is periodically monitored, thereby optimizing and correcting the initial scheduling strategy and improving the uniformity of fatigue distribution within the wind farm.

[0111] Specifically, when generating the corrected evaluation value g for the current feedback time point, it includes:

[0112] g=e1*Q1* (c i -c') 2+e2*Q2* (c i -∆c) 2 +e3*Q3*U;

[0113] U= Y(i)*(c i -c');

[0114] where e1 is a preset first weight coefficient; e2 is a preset second weight coefficient; e3 is a preset third weight coefficient; Q1 is a preset first fixed coefficient; Q2 is a preset second fixed coefficient; Q3 is a preset third fixed coefficient; c' is the expected fatigue evaluation value at the current feedback time node; ∆c is the average value of all data in the fatigue evaluation value sequence C; Y(i) is a selection coefficient. If (c i -c')>0, Y(i)=1 / (c i -c'); if (c i -c')<0, Y(i)=0.

[0115] Specifically, all parameters in the model are normalized by presetting the first fixed coefficient, the second fixed coefficient, and the third fixed coefficient, so that each parameter is within the same value range.

[0116] Specifically, the expected fatigue evaluation value at the current feedback time node is generated according to historical parameters, and the expected fatigue evaluation value refers to the fatigue loss of the wind turbine when it operates in an ideal state and accumulates to the current feedback time node as the operation time increases.

[0117] Specifically, the larger the correction evaluation value is, the worse the uniformity of the fatigue distribution in the current wind farm is, and the more wind turbines with fatigue greater than expected there are. It is necessary to promptly correct and optimize the current initial scheduling strategy.

[0118] It can be understood that in the above embodiments, by periodically monitoring the uniformity of the fatigue distribution in the wind farm, the operation parameters of the overly worn wind turbines are adjusted in a timely manner. By controlling the turbine speed, pitch angle or shutdown maintenance, etc., the fatigue characteristics of the wind turbines are improved, and the decline in the operation efficiency and the reduction in the operation life of the wind turbines caused by excessive wear are avoided.

[0119] In a preferred embodiment of the present application, when judging whether to generate a correction instruction according to the correction evaluation value g, it includes:

[0120] Presetting a first correction evaluation value threshold G1 and a second correction evaluation value threshold G2;

[0121] If g < G1, no correction instruction is generated;

[0122] If \(G1\leq g < G2\), generate a first-level correction instruction, and generate a power compensation coefficient for each wind turbine point according to the first-level correction instruction;

[0123] Correct the initial scheduling strategy according to all power compensation coefficients;

[0124] If \(g\geq G2\), generate a second-level correction instruction, and correct the initial scheduling strategy according to the second-level correction instruction.

[0125] Specifically, the first correction evaluation value threshold and the second correction evaluation value threshold can be set according to historical parameters.

[0126] Specifically, when generating the first-level correction instruction, it indicates that the uniformity of the fatigue distribution in the current wind farm is poor. Proportionally allocate the superior scheduling instruction according to the output capacity and fatigue degree of each wind turbine unit, and through optimization processing, generate the power compensation coefficient of each wind turbine unit, so as to achieve the scheduling optimization of each wind turbine unit, optimize the uniformity of the fatigue distribution of the wind farm without losing power generation, extend the service life of the unit, and reduce the operation and maintenance cost of the wind farm.

[0127] Specifically, when correcting the initial scheduling strategy according to the second-level correction instruction, it includes:

[0128] Generate the fatigue evaluation value threshold \(C1\) of the current feedback time node;

[0129] If \(c\) i \(>C1\), then eliminate the \(i\)-th wind turbine point and set the \(i\)-th wind turbine point as the to-be-maintained wind turbine point;

[0130] Generate a maintenance plan according to all to-be-maintained wind turbine points;

[0131] Establish a first-level wind turbine point sequence \(A1\) of the current feedback time node according to the elimination result;

[0132] \(A1=(a\) 11 ,a\) 12 …a\) 1i …a\) 1n1 ), where \(a_i\) is the \(i\)-th uneliminated wind turbine point of the current feedback time node; \(n1\) is the number of uneliminated wind turbine points of the current feedback time node, and \(n1\leq n\);

[0133] Generate the distribution coefficient of each first-level wind turbine point according to the expected power and fatigue evaluation value of each first-level wind turbine point;

[0134] Generate the first-level scheduling strategy of the current feedback time node according to the distribution coefficient and the demand power \(b'\).

[0135] Specifically, the fatigue evaluation value threshold can be set according to historical parameters. When \(c\) iWhen C1 is greater than 1, it indicates that the i-th wind turbine has abnormal losses and needs to be repaired in time to improve the service life of the wind turbine.

[0136] Specifically, by shutting down some wind turbines with abnormal losses, their losses are reduced and their service life is extended. At the same time, the remaining primary wind turbines are allocated proportionally to the upper-level dispatch instructions based on their output capacity and fatigue level, and a primary dispatch strategy is generated after optimization.

[0137] Specifically, if the total expected power of the remaining primary wind turbines is lower than the required power b', each primary wind turbine will generate electricity according to the expected power, and the operating parameters of each eliminated wind turbine will be set according to the power difference between the total expected power and the required power b'.

[0138] In another preferred embodiment of the wind farm dispatching method based on fatigue distribution based on any of the above preferred embodiments, this preferred embodiment provides a wind farm dispatching method based on fatigue distribution, including:

[0139] The central control unit is used to establish multiple wind turbine points based on the wind turbine parameters;

[0140] The monitoring unit includes multiple monitoring sub-modules, which are set at the wind turbine points to collect the operating parameters of each wind turbine point.

[0141] The central control unit includes:

[0142] The first processing module is used to establish multiple adjustment cycles based on historical environmental parameters. The first processing module is also used to generate the expected power of each wind turbine point in the current adjustment cycle based on the preset power prediction model.

[0143] The second processing module is used to generate an initial scheduling strategy based on the grid dispatch instructions and the expected power of all wind turbines in the current adjustment cycle.

[0144] The third processing module is used to obtain the fatigue evaluation value of each wind turbine point according to the preset feedback time node;

[0145] The correction module is used to determine whether to correct the initial scheduling strategy based on all fatigue evaluation values.

[0146] The fourth processing module is used to generate the sequence of wind turbine points A, A=(a1,a2…a…). i …a n ), where a i Let i be the i-th wind turbine point, and n be the number of wind turbine points;

[0147] The second processing module is also used for:

[0148] Establish the expected power sequence P, P=(p1,p2…p) within the current adjustment period. i …p n ), where pi is the expected power of the i-th wind turbine point in the current adjustment cycle;

[0149] The required power b' is generated according to the power grid dispatch instructions;

[0150] Generate the first-level power of each wind turbine point in the current adjustment cycle based on the expected power sequence P and the demand power b'.

[0151] Establish a first-order power sequence B, B=(b1, b2…b i …b n ),b i The first-level power of the i-th fan point during the current adjustment cycle;

[0152] Among them, b i =b'*[p i / ( p i )];

[0153] The initial scheduling strategy for the current adjustment cycle is generated based on the first-level power sequence B.

[0154] Specifically, the third processing module is also used for:

[0155] Set multiple feedback time points within the current adjustment cycle;

[0156] Obtain the fatigue evaluation values ​​of each wind turbine point at the current feedback time point;

[0157] Establish a fatigue evaluation value sequence C, C=(c1,c2…c i …c n ), where c i This represents the fatigue evaluation value of the i-th fan at the current feedback time point.

[0158] In a preferred embodiment of this application, the correction module is further configured to:

[0159] Generate the corrected evaluation value g for the current feedback time point based on the fatigue evaluation value sequence C;

[0160] Whether to generate a correction instruction is determined based on the correction evaluation value g.

[0161] g=e1*Q1* (c i -c') 2 ]+e2*Q2* (c i -∆c) 2 ]+e3*Q3*U;

[0162] U = Y(i)*(c i - c');

[0163] Among them, e1 is a preset first weight coefficient; e2 is a preset second weight coefficient; e3 is a preset third weight coefficient; Q1 is a preset first fixed coefficient; Q2 is a preset second fixed coefficient; Q3 is a preset third fixed coefficient; c' is the expected fatigue evaluation value at the current feedback time node; ∆c is the average value of all data in the fatigue evaluation value sequence C; Y(i) is a selection coefficient. If (c i - c') > 0, Y(i) = 1 / (c i - c'); if (c i - c') < 0, Y(i) = 0;

[0164] Judge whether to generate a correction instruction according to the corrected evaluation value g.

[0165] Specifically, when judging whether to generate a correction instruction according to the corrected evaluation value g, it includes:

[0166] Preset a first corrected evaluation value threshold G1 and a second corrected evaluation value threshold G2;

[0167] If g < G1, no correction instruction is generated;

[0168] If G1 ≤ g < G2, generate a first-level correction instruction, and generate a power compensation coefficient for each wind turbine point according to the first-level correction instruction;

[0169] Correct the initial scheduling strategy according to all power compensation coefficients;

[0170] If g ≥ G2, generate a second-level correction instruction, and correct the initial scheduling strategy according to the second-level correction instruction.

[0171] According to the first concept of the present application, generate an initial scheduling strategy according to the operating parameters of each wind turbine generator, generate the fatigue evaluation value of each wind turbine generator according to the preset feedback time node, optimize and adjust the initial scheduling strategy according to all fatigue evaluation values, and optimize the uniformity of the fatigue distribution of the wind farm without losing power generation, extend the service life of the unit, and reduce the operation and maintenance costs of the wind farm.

[0172] According to the second concept of the present application, by periodically monitoring the uniformity of the fatigue distribution of the wind farm, timely adjust the operating parameters of the wind turbine generators with excessive wear, and improve the fatigue characteristics of the wind turbine generators by controlling the unit speed, pitch angle or shutdown maintenance, etc., to avoid the decline of the operating efficiency and the reduction of the operating life of the wind turbine generators caused by excessive wear.

[0173] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application.

Claims

1. A wind farm dispatching method based on fatigue distribution, characterized in that, including: establishing multiple wind turbine points according to the parameters of the wind turbine; establishing multiple adjustment cycles according to historical environmental parameters, and generating the expected power of each wind turbine point within the current adjustment cycle according to a preset power prediction model; generating an initial scheduling strategy according to the grid dispatching instruction within the current adjustment cycle and the expected power of all wind turbine points; obtaining the fatigue evaluation value of each wind turbine point according to the preset feedback time node, and judging whether to correct the initial scheduling strategy according to all fatigue evaluation values; wherein, when establishing multiple wind turbine points, it includes: Generate a sequence of wind turbine points A, A=(a1,a2…a ... i …a n ), where a i Let i be the i-th wind turbine point, and n be the number of wind turbine points; when generating an initial scheduling strategy, it includes: Establish the expected power sequence P, P=(p1,p2…p) within the current adjustment period. i …p n ), where pi is the expected power of the i-th wind turbine point in the current adjustment cycle; generating a demand power b' according to the grid dispatching instruction; generating the primary power of each wind turbine point within the current adjustment cycle according to the expected power sequence P and the demand power b'; Establish a first-order power sequence B, B=(b1, b2…b i …b n ),b i The first-level power of the i-th fan point during the current adjustment cycle; Among them, b i =b'*[p i / ( p i )]; generating an initial scheduling strategy for the current adjustment cycle according to the primary power sequence B; when judging whether to correct the initial scheduling strategy according to all fatigue evaluation values, it includes: setting multiple feedback time nodes within the current adjustment cycle; obtaining the fatigue evaluation value of each wind turbine point at the current feedback time node; Establish a fatigue evaluation value sequence C, C=(c1,c2…c i …c n ), where c i This represents the fatigue evaluation value of the i-th fan at the current feedback time point; generating a correction evaluation value g for the current feedback time node according to the fatigue evaluation value sequence C; judging whether to generate a correction instruction according to the correction evaluation value g; when generating the correction evaluation value g for the current feedback time node, it includes: g=e1*Q1* (c i -c') 2 ]+e2*Q2* (c i -∆c) 2 ]+e3*Q3*U; U= Y(i)*(c i -c'); Where e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; e3 is the preset third weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; Q3 is the preset third fixed coefficient; c' is the expected fatigue evaluation value at the current feedback time node; ∆c is the average value of all data in the fatigue evaluation value sequence C; Y(i) is the selection coefficient, if (c i -c')>0,Y(i)=1 / (c i -c'); if (c i -c')<0,Y(i)=0.

2. The wind farm dispatching method based on fatigue distribution as described in claim 1, characterized in that, when judging whether to generate a correction instruction according to the correction evaluation value g, it includes: presetting a first correction evaluation value threshold G1 and a second correction evaluation value threshold G2; if g < G1, no correction instruction is generated; if G1 ≤ g < G2, generating a primary correction instruction, and generating a power compensation coefficient for each wind turbine point according to the primary correction instruction; correcting the initial scheduling strategy according to all power compensation coefficients; if g ≥ G2, generating a secondary correction instruction, and correcting the initial scheduling strategy according to the secondary correction instruction.

3. The wind farm dispatching method based on fatigue distribution as described in claim 2, characterized in that, when correcting the initial scheduling strategy according to the secondary correction instruction, it includes: generating a fatigue evaluation value threshold C1 for the current feedback time node; If c i If C1 is selected, the i-th fan point is removed, and the i-th fan point is set as the fan point to be inspected. generating a maintenance plan according to all wind turbine points to be maintained; establishing a primary wind turbine point sequence A1 for the current feedback time node according to the elimination result; A1=(a 11 ,a 12 …a 1i …a 1n1 ), where is the i-th wind turbine point that has not been removed at the current feedback time point; n1 is the number of wind turbine points that have not been removed at the current feedback time point, and n1≤n; generating an allocation coefficient for each primary wind turbine point according to the expected power and fatigue evaluation value of each primary wind turbine point; generating a primary scheduling strategy for the current feedback time node according to the allocation coefficient and the demand power b'.

4. A wind farm dispatching system based on fatigue distribution, employing the wind farm dispatching method based on fatigue distribution as described in any one of claims 1-3, characterized in that, including: a central control unit for establishing multiple wind turbine points according to the parameters of the wind turbine; a monitoring unit including multiple monitoring sub-modules, the monitoring sub-modules are arranged at the wind turbine points for collecting the operation parameters of each wind turbine point; the central control unit includes: a first processing module for establishing multiple adjustment cycles according to historical environmental parameters, and the first processing module is also used for generating the expected power of each wind turbine point within the current adjustment cycle according to a preset power prediction model; a second processing module for generating an initial scheduling strategy according to the grid dispatching instruction within the current adjustment cycle and the expected power of all wind turbine points; a third processing module for obtaining the fatigue evaluation value of each wind turbine point according to the preset feedback time node; a correction module for judging whether to correct the initial scheduling strategy according to all fatigue evaluation values; The fourth processing module is used to generate the sequence of wind turbine points A, A=(a1,a2…a…). i …a n ), where a i Let i be the i-th wind turbine point, and n be the number of wind turbine points; the second processing module is also used for: Establish the expected power sequence P, P=(p1,p2…p) within the current adjustment period. i …p n ), where pi is the expected power of the i-th wind turbine point in the current adjustment cycle; generating a demand power b' according to the grid dispatching instruction; Generate the primary power of each fan point within the current adjustment period according to the expected power sequence P and the required power b'. Establish a first-order power sequence B, B=(b1, b2…b i …b n ),b i The first-level power of the i-th fan point during the current adjustment cycle; Among them, b i =b'*[p i / ( p i )]; Generate the initial scheduling strategy for the current adjustment period according to the primary power sequence B.

5. The wind farm dispatching system based on fatigue distribution as described in claim 4, characterized in that, The third processing module is further configured to:[[]] Set multiple feedback time nodes within the current adjustment period; Obtain the fatigue evaluation values of each fan point at the current feedback time node; Establish a fatigue evaluation value sequence C, C=(c1,c2…c i …c n ), where c i This represents the fatigue evaluation value of the i-th fan at the current feedback time point.

6. The wind farm dispatching system based on fatigue distribution as described in claim 5, characterized in that, The correction module is further configured to:[[]] Generate the correction evaluation value g at the current feedback time node according to the fatigue evaluation value sequence C; Judge whether to generate a correction instruction according to the correction evaluation value g; g=e1*Q1* (c i -c') 2 ]+e2*Q2* (c i -∆c) 2 ]+e3*Q3*U; U= Y(i)*(c i -c'); Where e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; e3 is the preset third weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; Q3 is the preset third fixed coefficient; c' is the expected fatigue evaluation value at the current feedback time node; ∆c is the average value of all data in the fatigue evaluation value sequence C; Y(i) is the selection coefficient, if (c i -c')>0,Y(i)=1 / (c i -c'); if (c i -c') < 0, Y(i) = 0; Judge whether to generate a correction instruction according to the correction evaluation value g.

7. The wind farm dispatching system based on fatigue distribution as described in claim 6, characterized in that, When judging whether to generate a correction instruction according to the correction evaluation value g, it includes:[[]] Preset the first correction evaluation value threshold G1 and the second correction evaluation value threshold G2; If g < G1, no correction instruction is generated; If G1 ≤ g < G2, generate a primary correction instruction, and generate the power compensation coefficient of each fan point according to the primary correction instruction; Correct the initial scheduling strategy according to all the power compensation coefficients; If g ≥ G2, generate a secondary correction instruction, and correct the initial scheduling strategy according to the secondary correction instruction.

Images