A time course optimization method for wind energy tracking control of large wind turbine generators considering aging effect

Abstract

The application discloses a kind of time course optimization methods of wind energy tracking control of large wind turbine considering aging effect, comprising: S100. Quantitative evaluation of unit performance degradation, SCADA parameter selection and its data preprocessing are carried out, unit performance degradation quantitative evaluation model is constructed, and unit aging state is evaluated by field SCADA data;S300. Optimize unit control time course, judge unit degradation state by SCADA data;S200. Maximum wind energy tracking control, according to the evaluated unit aging state, draw the tip speed ratio-wind energy utilization coefficient curve under the unit aging state, approximately describe the basic performance of wind turbine, and obtain the maximum value of wind energy utilization coefficient. This optimization method is proposed for the problem that the decrease of wind energy utilization coefficient caused by natural aging of wind turbine will directly affect the power output of wind turbine, and the best operating condition parameters of wind turbine under aging state can be more accurately obtained, the operating condition is optimized, and the operation power generation efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wind power generation, and particularly provides a method for optimizing the time schedule of wind energy tracking control of large wind turbine generators considering aging effects. BACKGROUND

[0002] A wind turbine generator is a device that converts wind energy into electrical energy. The ability to capture aerodynamic energy is closely related to the aerodynamic performance of the wind turbine generator. Therefore, in the design of a wind turbine generator, the analysis of structure and aerodynamic performance is the focus of attention. Some aerodynamic performance analysis methods, such as the classic blade element momentum theory (BEM) and computational fluid dynamics (CFD), have received extensive attention. Aerodynamic design combined with control design can design a wind turbine generator power characteristic curve as the basis for the operation control strategy of the wind turbine generator. With the increase of service life, the performance degradation caused by natural aging of the wind turbine generator is inevitable. This problem is increasingly ignored as a large number of wind turbine generators enter the “golden old age”. Research on the aging of wind turbine generators and its evaluation has attracted more and more attention from researchers. In a wind farm, describing the historical evolution behavior of the aerodynamic performance of the wind turbine generator on site is a difficult problem related to the operation, maintenance and optimization of the wind turbine generator. First, the wind turbine generator is a super large mechanical structure, and on-site aerodynamic flow field testing cannot be performed. Second, there is limited on-site data, and there are no direct operation parameters to describe the aerodynamic performance.

[0003] For a wind turbine generator in service on site, it is not practical to additionally increase a hardware testing system, and it is feasible to make full use of the existing SCADA data. The SCADA data records many operation parameter data such as the nacelle wind speed, the wind wheel (generator) speed, the generator power and the like, and can indirectly reflect the aerodynamic performance of the wind turbine generator. For example, the aerodynamic performance of the wind turbine generator affects its power coefficient, and the power coefficient can be calculated based on the SCADA data, thereby establishing a correlation between the aerodynamic performance of the wind turbine generator and the SCADA parameters. However, the SCADA data cannot be directly used, and some zero values, null values and abnormal values intermingled therein need to be processed.

[0004] However, general data preprocessing cannot solve the problems of "randomness" and "bias" in wind speed. These two issues are the main difficulties currently affecting the reliable calculation of the power coefficient of wind turbines in the field. The randomness of wind speed means that the wind speed is unstable and changes uncertainly, and it also means that the wind speed contains high-frequency components. From a cybernetics perspective, for a large inertial system like a wind turbine, high-frequency wind speed will not affect its rotational speed. When performing power coefficient analysis, its influence should be suppressed as much as possible, otherwise it will lead to deviations in the power coefficient calculation. The bias of wind speed means that the wind speed measured by the nacelle anemometer is smaller than the wind speed at the leading edge of the turbine, and deviation compensation is required. Although some literature has carried out research on wind speed filtering, there is a lack of effective filtering methods that conform to the actual characteristics of wind turbines. For example, the problem of how to determine the filter cutoff frequency (the time constant of the inertial system) has not been effectively solved. Summary of the Invention

[0005] Based on this, the present invention provides a time-history optimization method for wind energy tracking control of large wind turbine units that takes into account aging effects. It objectively assesses the natural aging condition of wind turbine units and evaluates the wind energy utilization coefficient of the units based on the aging condition, thereby obtaining the optimal operating parameters of wind turbine units under aging conditions more accurately, optimizing operating conditions, and improving power generation efficiency.

[0006] To achieve the above objectives, the present invention provides a time-history optimization method for wind energy tracking control of large wind turbine units that takes into account aging effects, comprising:

[0007] S100. Quantitative evaluation of unit performance degradation: Select SCADA parameters for the unit and perform data preprocessing, construct a quantitative evaluation model for unit performance degradation, and evaluate the aging status of the unit through on-site SCADA parameters;

[0008] S200. Optimize unit control schedule: Determine the unit's deterioration status through SCADA data, update parameters based on the deterioration status results, and determine the impact of deterioration degree in the speed-power control curve;

[0009] S300. Maximum Wind Energy Tracking Control: Based on the assessed aging state of the unit, plot the tip speed ratio versus wind energy utilization coefficient curve under aging conditions to approximately describe the basic performance of the wind turbine unit, combined with the unit's wind energy utilization coefficient C. P The relationship with the tip speed ratio λ is used to determine when the tip speed ratio is at its optimal value λ. opt At that time, the maximum value of the wind energy utilization coefficient is obtained.

[0010] Further, the data preprocessing for SCADA parameters includes:

[0011] S110. Filling in missing data and handling outliers: Find and fill in missing data to facilitate subsequent processing; identify and handle null values, zero values, and outliers that are obviously inconsistent with reality;

[0012] S120. High-frequency processing of temperature data: High-frequency processing of temperature data with long sampling periods to achieve the same sampling frequency as parameters such as wind speed, rotational speed, and power, and to place it in the same data package as the data of parameters such as wind speed, rotational speed, and power for convenient subsequent analysis;

[0013] S130. Supplementary data on air density in wind farms: using the formula ρ=f(T) C The instantaneous air density of the wind farm is obtained by numerical calculation of temperature, humidity, and atmospheric pressure (RH), where ρ is the air density and T is the atmospheric pressure. C The value is Celsius, obtained through the SCADA system; h is the altitude, which can be obtained from wind farm technical data; RH is the relative humidity, obtained from a meteorological website.

[0014] S140. Wind speed data deviation compensation and filtering: Compensate for wind speed deviation and eliminate high-frequency components mixed in the random changes of wind speed through filtering.

[0015] Furthermore, a quantitative evaluation model for unit performance degradation is constructed as follows:

[0016]

[0017] This paper presents a time-history optimization method for wind energy tracking control of large wind turbine units that takes into account aging effects.

[0018] 1. The power coefficient is used as the basic indicator for the field analysis of the aerodynamic performance of wind turbine blades. The flow field characteristics of the field wind speed data are fully considered. Data compensation and data filtering are used to overcome the adverse effects of the two unfavorable factors of "randomness" and "bias" in SCADA wind speed data.

[0019] 2. Treating the relationship between wind speed and rotor speed as an inertial system, a data-driven time constant determination method is proposed. This method determines the system time constant from the data itself, overcoming the inconvenience of requiring precise unit parameters in the model-driven time constant determination method and eliminating the need to consider the influence of blade flexibility.

[0020] 3. An algorithm for dividing the calculation operating condition region was designed to conform to the actual characteristics of the field data. The speed-power curve was selected as the judgment basis, making full use of the stability and gradual change advantages of these two parameters relative to wind speed. The data binning method and the "conservative" data comparison method were used to solve the problems of "single-value" of scattered data and determination of interval critical values ​​that need to be solved when dividing the calculation operating condition region.

[0021] 4. The effectiveness of the proposed "four-index" multi-angle observation algorithm was demonstrated through computational case analysis, showing its reliability and effectiveness. The algorithm revealed the evolution characteristics of the aerodynamic performance of large wind turbines in the field. The analysis results are consistent with theoretical analysis and provide new ideas and basis for quantitative evaluation. This provides important knowledge support for the operation, maintenance, and performance improvement of large wind turbines after they have reached their "old age." Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of an embodiment of the time-history optimization method for wind energy tracking control of large wind turbine units that takes into account aging effects.

[0024] Figure 2 This is a logic block diagram of the SCADA parameter selection and processing provided;

[0025] Figure 3 This is a flowchart of the data preprocessing for the provided SCADA parameters;

[0026] Figure 4 This is the provided time-scheduling optimization control block diagram for wind turbine units;

[0027] Figure 5 It is the performance curve of the wind turbine unit;

[0028] Figure 6 It is the wind turbine speed-power curve;

[0029] Figure 7 High-frequency processing of temperature data;

[0030] Figure 8 It is a three-dimensional surface representing the average wind energy utilization coefficient from 2017 to 2020;

[0031] Figure 9 Three-dimensional scatter plot of wind energy utilization coefficient based on wind speed and power;

[0032] Figure 10 This refers to the wind energy utilization coefficient and degradation rate from 2017 to 2020.

[0033] Figure 11 It is the performance curve of the wind turbine unit;

[0034] Figure 12 It is a time-history optimization performance analysis under stable wind speed conditions;

[0035] Figure 13 It is the unit power increase rate after process optimization;

[0036] Figure 14 This refers to the unit's power characteristics under varying wind speed conditions. Detailed Implementation

[0037] This application provides a time-history optimization method for wind energy tracking control of large wind turbines that takes into account aging effects. It studies the historical evolution of the aerodynamic performance of large wind turbines in the field from several dimensions: description of blade aerodynamic performance, preprocessing of field wind speed data, division of calculation operating conditions, and analysis of calculation cases. The power coefficient is used as the basic indicator for the field analysis of wind turbine blade aerodynamic performance. The flow field characteristics of field wind speed data are fully considered, and data compensation and filtering are used to overcome the adverse effects of the "randomness" and "bias" of SCADA wind speed data. The system time constant is determined from the data itself, proposing a data-driven time constant determination method. An algorithm for dividing the calculation operating conditions according to the actual characteristics of field data is designed, and the effectiveness of the "four-index" multi-angle observation algorithm is proposed.

[0038] (I) Maximum wind energy tracking control

[0039] Based on wind speed characteristics, wind turbines can be divided into two operating conditions: above rated wind speed and below rated wind speed. Below rated wind speed, wind turbines operate without power limitations, while above rated wind speed, power capture and output need to be limited by adjusting the blade pitch angle. Below rated wind speed, the main task of the wind turbine control system is to achieve the maximum wind energy capture efficiency, primarily through speed regulation.

[0040] Since the blade pitch angle of wind turbines generally remains constant below the rated wind speed, it can be used Figure 5 The "tip speed ratio - wind energy utilization coefficient" curve shown approximates the basic performance of the wind turbine. As can be seen from the figure, the wind energy utilization coefficient C of the wind turbine... P It is closely related to the tip speed ratio λ. When the tip speed ratio is at a certain optimal value λ opt At this time, the wind energy utilization coefficient reaches its maximum. The wind energy utilization coefficient reflects the ability of a wind turbine to capture energy from natural wind. From existing wind turbine operating theories, it is known that the mechanical power captured by a wind turbine from the wind can be written as...

[0041]

[0042] In the formula, ρ is the air density (kg / m³). 3 S is the swept area of ​​the wind turbine (m²) 2 ), where v is the wind speed at the leading edge of the wind turbine (m / s) and β is the pitch angle (°).

[0043] Tip speed ratio is used to represent the state of the wind turbine at different wind speeds. It is the ratio of the circumferential speed at the blade tip to the wind speed, expressed as:

[0044]

[0045] In the formula, ω t R is the angular frequency of the wind turbine (rad / s) and R is the radius of the wind turbine (m).

[0046] Will Figure 5 λ opt and C Pmax Substituting into formula (1), we can obtain the optimal wind energy capture expression below the rated wind speed as follows:

[0047]

[0048] Then, substituting formula (2) into formula (3), we have

[0049]

[0050] Rewrite the above equation as follows:

[0051]

[0052] During the maximum wind energy tracking phase, the rotational speed ω r With power P opt The above relationship should be satisfied, that is, when the tip speed ratio and wind energy utilization coefficient are kept constant, the power and rotational speed are cubic. Figure 6 This is a speed-power curve plotted based on actual SCADA data from a 2MW wind turbine. The graph shows that the wind turbine's operation can be divided into four regions: transition zone 1, maximum wind energy tracking zone, transition zone 2, and constant power zone. It can be seen that significant speed changes primarily occur in the maximum wind energy tracking zone, where the wind turbine is controlled by the given speed-power curve.

[0053] For newly installed wind turbines, achieving optimal energy capture performance hinges on ensuring the set speed-power curve aligns with the turbine's actual characteristics. However, for existing wind turbines, performance degradation is inevitable with extended service life, leading to discrepancies between the initially set speed-power curve and the turbine's actual performance. Ignoring these degradation-related deviations will inevitably impact energy capture efficiency. Therefore, the authors propose a "time-based optimization" approach, adjusting the speed-power curve in real-time based on the wind turbine's dynamic performance degradation. However, the aging state of wind turbines during service is difficult to describe effectively, and there is no prior knowledge to draw upon. Assessing the turbine's aging state relies solely on on-site SCADA data. Therefore, determining the turbine's aging state based on SCADA data is a prerequisite for time-based performance optimization.

[0054] (II) Quantitative Evaluation of Unit Performance Deterioration

[0055] SCADA Parameter Selection and Data Preprocessing: SCADA systems record many parameters, but not all of them need to be used as the basis for performance degradation assessment. The parameter selection and processing flow is as follows: Figure 4 As shown, the entire process includes relevant parameter analysis, parameter data classification, supplementary parameter analysis, and data preprocessing. This application uses the wind energy utilization coefficient (FE) as an evaluation indicator for performance degradation caused by turbine aging. Therefore, in the "relevant parameter analysis," only SCADA parameters related to its calculation need to be considered, including hub speed, instantaneous wind speed, active power, ambient temperature, and pitch angle. On the other hand, although these parameters are related to the calculation of the FE, they can be divided into two categories: one directly used in the FE calculation, and the other used as observation values ​​for data extraction during the FE calculation. For example, the pitch angle parameter is not directly used in the FE calculation, but its value should remain unchanged within the data range used for calculation; therefore, it is a status monitoring parameter during calculation. This process is called "parameter data classification."

[0056] Data analysis revealed that air density is generally not directly recorded in SCADA data. Ignoring the dynamic changes in air density would inevitably affect the accuracy of wind energy utilization factor calculations. Therefore, the "supplementary parameter analysis" process primarily involves adding air density data to the SCADA data table, which in turn requires supplementary calculations using data such as temperature, humidity, and air pressure. Furthermore, in some SCADA systems, temperature data is stored at the minute level, which is inconvenient for batch data analysis and calculation. Therefore, it is also necessary to convert minute-level data to second-level data for higher frequency processing.

[0057] During the acquisition and transmission of SCADA data from wind turbines, data loss, null values, zero values, and outliers are unavoidable due to internal and external signal interference (noise). Therefore, "data preprocessing" is an indispensable step in SCADA data analysis. Different data preprocessing strategies are employed depending on the specific data processing and analysis objectives. The following outlines four main steps for data preprocessing:

[0058] Step 1: Missing data filling and outlier handling.

[0059] SCADA data analysis revealed a phenomenon called "data missing" – discontinuous data storage time. To facilitate subsequent processing, missing data needs to be located and supplemented. Each set of SCADA data contains not only status parameter information but also a sampling time series, and data is stored in packets with a certain time period. In the process of identifying and searching for missing data, the start time of the data packet is first recorded. Then, the time is accumulated and compared item by item according to the sampling time interval (usually one second). If the time item in the packet does not match the accumulated time item, it indicates that a missing item has occurred. The supplementation algorithm is as follows:

[0060] X=(time,parameter1,parameter2,···)

[0061] i = start_time:1:end_time

[0062] If X.time≠i,

[0063] insert X.time = i;

[0064] X.parameters = 0.

[0065] In the expression above, start_time and end_time are the start and end times of packet sampling, respectively.

[0066] Furthermore, during the on-site data sensing, acquisition, and transmission process, various interferences may cause null values, zero values, and values ​​that are clearly inconsistent with reality. These are all referred to as "outliers" and require judgment and processing. Zero and null values ​​can be handled using the following algorithm:

[0067] ifX.parameters is null

[0068] X.parameters = 0;

[0069] ifX i .parameter=0andX i-1.parameter>0&X i+1 .parameter>0

[0070] X i .parameter = (X i-1 .parameter+X i+1 .parameter) / 2.

[0071] Step 2: High-frequency processing of temperature data.

[0072] The sampling period for different wind turbine parameters varies in the stored SCADA data. SCADA data for parameters such as wind speed, RPM, and power are sampled at 1 second, while temperature data is sampled at 60 seconds. This is, of course, related to the SCADA system's own settings. For ease of subsequent analysis, the temperature data needs to be processed to a higher frequency and then placed in the same data package as other parameters. Considering that temperature is a slowly changing variable, the high-frequency processing can be simplified. Specifically, minute-level temperature data can be evenly distributed into second-level data, assuming that the temperature data remains constant within one minute. Figure 7 As shown.

[0073] Step 3: Supplement air density data for wind farms.

[0074] The wind energy utilization coefficient of a wind turbine is an important indicator for assessing the aging status of the unit. According to equation (1), the wind energy utilization coefficient is related to factors such as wind speed, power, and air density. The SCADA system generally does not record the real-time air density of the wind farm, but it can be obtained by calculating the values ​​of temperature, humidity, and atmospheric pressure.

[0075] ρ=f(T C (6)

[0076] In the formula, ρ is the air density, and T C is the temperature in Celsius, obtained through the SCADA system; h is the altitude, which can be obtained from wind farm technical data; RH is the relative humidity, which can be found on a meteorological website. The specific calculation expression for equation (6) can be expressed as follows:

[0077]

[0078] In the formula, T K P is the Kelvin temperature; V R is the water vapor pressure (Pa); d R is the specific gas constant of dry air (J / (kg·K)); v ρ is the specific gas constant of water vapor (J / (kg·K)); Po is the sea level pressure; g is the acceleration due to gravity, taken as g = 9.8 m / s². 2M is the molar mass of air (kg / mol); R is the universal gas constant (N·m / [mol·K]).

[0079] Step 4: Wind speed data deviation compensation and filtering.

[0080] Among various SCADA parameters, wind speed requires special attention due to its random uncertainty. Furthermore, wind speed data is mostly measured by anemometers installed in the nacelle (behind the rotor). Due to their spatial location, the wind speed measured by the anemometer is not necessarily the actual incoming wind speed. As airflow passes through the rotor, some of the air's kinetic energy is converted into mechanical energy, thus causing a discrepancy between the wind speed measured by the anemometer and the actual wind speed.

[0081] For random variations in wind speed (wind speed containing high-frequency components), filtering can be used, for example, by employing a moving average filtering algorithm:

[0082]

[0083] In the formula, y i x is the output of the filter. i is the input to the filter, and n is the window length of the MAF filter.

[0084] For the problem of wind speed deviation compensation, the compensation expression is used:

[0085]

[0086] In the formula, v1 is the compensated wind speed, P is the generator output power, v0 is the wind speed measured by the sensor, and S is the swept area of ​​the wind turbine rotor.

[0087] 2.2 Quantitative Evaluation Model for Unit Performance Deterioration

[0088] As wind turbines age, natural aging inevitably leads to performance degradation. Wind turbines are equipment that converts wind energy into electrical energy, and the wind energy utilization factor (TEF) is the most important indicator for measuring turbine aging. Therefore, this application designs a quantitative evaluation model for wind turbine performance degradation based on the TEF, as follows. This model describes the entire process from SCADA data to the degree of turbine degradation (aging).

[0089] In the model, x represents a set of SCADA system parameters, where each set of data represents a single SCADA sampling result. iThis represents a specific operating parameter of the wind turbine. X represents several sets of parameter data stored in a SCADA data packet. These data sets may be continuous in time, or they may be discontinuous during data acquisition. D1 is a new data packet composed of several consecutive data packets. This is mainly because SCADA system data packets are stored in packets, and the data in a single packet, regardless of size, is finite (e.g., 10 minutes or 1 day). Multiple data packets need to be integrated into a new data packet with a sufficiently large data volume for analysis, such as selecting data from 10 consecutive days (N). day =10). D2 is the data space after relevant parameter selection and data preprocessing. System parameters unrelated to aging have been removed from this data space, and the data has been processed, including supplementing unrecorded relevant parameters, supplementing missing data, high-frequency data processing, and data bias compensation and filtering. D3 selects several subspaces within the D2 data space that meet the calculation conditions, using wind speed as the selection criterion. Using wind speed as the data selection criterion is mainly to ensure that the selected data are all within the maximum wind energy tracking zone, because theoretically, the wind energy utilization coefficient is constant in this area, which is beneficial for obtaining more reliable wind energy utilization coefficient calculation results. In addition, the data screening also needs to set the data capacity of the subspaces (e.g., 3000 sets of data per subspace). Too large a data volume makes it difficult to achieve the same data volume for different years, while too small a data volume easily leads to large deviations in the calculation results. Data analysis found that the number of subspaces obtained by the above method is relatively large. To simplify the calculation, several subspaces without data overlap are randomly selected from the D3 data subspace to obtain D4; then, for each subspace in D4, a series of C values ​​are obtained according to the wind energy utilization coefficient calculation method. Pi It should be noted that although the wind energy utilization coefficient is theoretically constant in the maximum wind energy tracking zone, the calculated C value is affected by factors such as the time-varying nature of wind speed and the inertia of the wind turbine. P The value is dynamic; therefore, for each subspace, the average wind energy utilization coefficient can be used. This represents the wind energy utilization coefficient for this interval; using this method, r subspaces of D4 can be obtained. Values, select the maximum value among them. As the final maximum wind energy tracking area of ​​that year, C P Value. Following the above method, the wind energy utilization coefficient for different years is calculated and compared with the results of the selected base year to obtain the unit deterioration (aging) degree Γ that evolves from year to year.

[0090]

[0091] In the formula, v is the wind speed parameter in the D2 data space, which is the wind speed after filtering and compensation; P iIt is the active power data recorded in the D4 subspace, ρ i This is the air density data (supplementary value) in the D4 subspace, v i These are wind speed data in the D4 subspace; s is the number of parameters retained after data selection, and k is the sample data length; C pi It is a dynamic wind energy utilization coefficient obtained based on the D4 subspace. C represents the average wind energy utilization coefficient of the sub-interval; p C represents the maximum wind energy utilization coefficient among r randomly selected sub-intervals; pb C is the wind energy utilization factor for the baseline (historical) year. pc To assess the annual wind energy utilization coefficient.

[0092] (III) Optimization of Unit Control Schedule

[0093] In a variable-speed permanent magnet wind turbine, the torque equation of the permanent magnet generator is:

[0094]

[0095] In the formula, T e The electromagnetic torque of the generator; p is the number of pole pairs; ψ f For permanent magnet flux linkage; L d L q These are the quadrature and direct-axis inductances of a permanent magnet generator, i d i q Let ψ be the AC and DC axis currents. If the rotor flux linkage ψ f If the d-axis current component is constant and always zero, then the electromagnetic torque is proportional to the q-axis current. Let the d-axis command current i... d =0, the mechanical dynamic model of the wind turbine can be written as

[0096]

[0097] In the formula, J t k is the moment of inertia. g K represents the gearbox speed ratio; t T is the coefficient of friction; r For aerodynamic torque. Equation (12) can be rewritten as

[0098]

[0099] From equation (13), it can be seen that changing the generator quadrature-axis current i q This allows adjustment of the wind turbine's speed. By selecting a first-order dynamic speed tracking error, the expression for the generator's quadrature-axis current is obtained as follows:

[0100]

[0101] In the formula, a0 is a constant; ε is the dynamic speed tracking error, and its expression is:

[0102] ε=ω tref -ω t (15)

[0103] ω t It refers to the wind turbine speed, which can be measured in real time using a high-precision speed sensor. The speed command is ω. tref It can be obtained by setting the "speed-power" curve. Equation (5) gives the relationship between speed and power in the maximum wind energy tracking zone.

[0104] The method provided in this application uses the power coefficient as the basic indicator for the field analysis of the aerodynamic performance of wind turbine blades, fully considers the flow field characteristics of the field wind speed data, and adopts data compensation and data filtering to overcome the adverse effects of the two unfavorable factors of "randomness" and "bias" in SCADA wind speed data.

[0105] Assuming the aerodynamic degradation of a wind turbine after a certain period of service is Γ0, what is the maximum wind energy utilization coefficient C corresponding to the optimal tip speed ratio? Pmax Become

[0106]

[0107] The speed command that takes into account aging effects can be written as:

[0108]

[0109] Considering the aerodynamic torque T in equation (14) r Since real-time values ​​are difficult to measure, the calculation of the quadrature-axis current only considers obtaining it from the dynamic speed tracking error ε, i.e.

[0110]

[0111] In the formula, K p K i These are the proportional coefficient and integral coefficient, respectively; ε(k) is the dynamic speed tracking error at the k-th sampling time.

[0112] Based on the above analysis, the time-history optimization control mode of the wind turbine is plotted as follows: Figure 4 As shown. First, the deterioration status of the unit is determined by SCADA data. Then, the parameters are updated according to the deterioration status results, that is, the influence of the degree of deterioration is considered in the "speed-power" control curve according to formula (17).

[0113] Treating the relationship between wind speed and rotor speed as an inertial system, a data-driven time constant determination method is proposed. This method determines the system time constant from the data itself, overcoming the inconvenience of model-driven time constant determination methods requiring precise unit parameters and eliminating the need to consider the influence of blade flexibility.

[0114] Simultaneously, an algorithm for dividing the calculation operating condition region was designed to conform to the actual characteristics of the field data. The speed-power curve was selected as the judgment criterion, fully utilizing the stability and gradual variation advantages of these two parameters relative to wind speed. The data binning method and the "conservative" data comparison method were used to solve the problems of "single-valued" scattered data and the determination of interval critical values ​​that needed to be addressed when dividing the calculation operating condition region.

[0115] (IV) Calculation Case Analysis

[0116] Analysis of turbine deterioration based on SCADA data: Taking a 2MW doubly fed wind turbine in a mountainous wind farm in southern China as an example, this type of wind turbine has a variable speed and variable pitch power regulation method, a design life of 20 years, a rated power of 2100kW, a cut-in wind speed of 3m / s, and a cut-out wind speed of 20m / s. SCADA data for 10 consecutive days from 2017 to March 2020 from a wind turbine were extracted. Parameter selection (supplement) and data preprocessing were performed according to the method described in section 2.1. Wind speed range [5.0 m / s, 9.5 m / s] was used as the condition (this wind speed range is known to be within the maximum wind energy tracking zone based on the wind turbine's design parameters and operating data). Continuous data was selected with a sampling interval of 1 second and a data length of 3000 points. Among the selected 10-day data from 2017 to March 2020, 18156, 51791, 10940, and 45671 data sets met the selection criteria, respectively. From these sets, 8 sets (per year) of data with non-overlapping data and a fixed blade angle were randomly selected to ensure that the selected wind turbine operating data were under maximum wind energy tracking zone conditions. The selected data set numbers were r = 1, 2, 3, ..., 8. Using groups as units, the wind energy utilization coefficient calculation method according to formula (10) is used to obtain a series of C values ​​at the sampling time. Pi Value and average wind energy utilization coefficient for each group Take 8 groups each year The maximum value among the mean values ​​is taken as the wind energy utilization coefficient C in the maximum wind energy tracking area for that year. P .

[0117] Figure 8 , Figure 9 As shown, the average wind energy utilization coefficient is calculated from 8 sets (32 sets in total) of data randomly selected each year from 2017 to 2020 according to formula (10). The three-dimensional surface plot shows the average wind energy utilization coefficient calculated from 32 data sets. The range is [0.332, 0.442], with the maximum value in 2017 (wind utilization coefficient of 0.442) and the minimum value in 2017 (wind energy utilization coefficient of 0.332). There are 8 years in 2020. The value fluctuations were smaller compared to the other three years, with the largest fluctuations in 2018. According to formula (10), eight values ​​were selected for each year. The highest average value among the values ​​is taken as the wind energy utilization coefficient C in the maximum wind energy tracking area for that year. P The figures for 2017-2020 were 0.442, 0.420, 0.397, and 0.392, respectively.

[0118] To further analyze the degradation characteristics exhibited by the unit as its service time increases, Figure 9 The paper presents a three-dimensional scatter plot relationship between wind speed (after compensation) and active power and wind energy utilization coefficient for the maximum mean wind energy utilization coefficient data set from 2017 to 2020. It can be seen that the blade angle values ​​in the four data sets from 2017 to 2020 are 0.02°, 0.03°, 0.02°, and -0.01°, respectively. The blade angle values ​​in 2017 and 2019 are the same, while the blade angle values ​​in the data sets in 2017 and 2020 are different, but all are fixed values. This indicates that the pitch system of the wind turbine does not have pitch in this operating condition area, that is, the operating conditions of the selected data set of the maximum mean wind energy utilization coefficient are all within the maximum wind energy tracking zone (MPPT). The four-year dynamic Cp values ​​(3000 values) represented by the three-dimensional scatter plot show that in 2019, the dynamic Cp values ​​were relatively concentrated, mainly within the range of [0.281, 0.530]. In 2020, the dynamic Cp values ​​were more dispersed, ranging from [0.251, 0.593], which is lower than the dynamic Cp values ​​of the other three sets of data. p The value is wider, with a variation of 0.342, which is related to the changes in the aerodynamic performance of the unit. Obviously, the larger the range of variation, the more significant the decline in aerodynamic performance. At the same time, among the four sets of data, the wind speed range in 2018 was [5.02m / s, 7.75m / s], which was the most concentrated and had the smallest variation (fluctuation), indicating that the wind speed was the most stable compared to the other three sets of data. The wind speed range in 2019 was [5.01m / s, 8.85m / s], which had the largest variation (fluctuation). The smallest change in active power was in 2017, with a range of [286kW, 1000kW] and a change of 714kW. The largest change was in 2019, with a range of [313kW, 1262kW] and a change of 949kW.

[0119] Figure 10 The paper presents the wind energy utilization coefficient and historical degradation characteristic curves of the maximum wind energy tracking area over four years (2018-2020) compared to 2017 (base year). The C value for 2017-2020 is also provided. p The values ​​were 0.442, 0.420, 0.397, and 0.392, respectively. The curves show that the C value of this unit increases with its service life. pThe value shows an overall downward trend. According to the definition of degradation degree in equation (10), the degradation degree (Γ) for 2017-2020 can be calculated to be 0, 5.1%, 10.2%, and 10.9%, respectively. The degradation degree value increases year by year with the increase of service life, which also indicates that the aerodynamic performance of the unit decreases year by year. Figure 8 The given unit deterioration curves show an overall upward trend in deterioration, which is consistent with the above-mentioned aging assessment method.

[0120] 4.2 Time-based optimization performance analysis

[0121] To achieve the desired effect of time-optimized control of the wind turbine, a numerical analysis method is employed. First, a parametric control aerodynamic characteristic analysis model for the wind turbine is established as follows:

[0122]

[0123] In the formula, c i (i = 1, 2, 3, 4, 5, 6) and k j (j = 1, 2, 3) are set coefficients, determined based on the specific characteristics of the wind turbine. The above calculation model is obtained by modifying the model based on the characteristics of the wind turbine. Given c1 = 0.312, c2 = 116, c3 = 0.4, c4 = 5, c5 = 15, c6 = 0.00912, k0 = 3, k1 = 0.08, k2 = 2.2, k3 = 0.035, the wind turbine performance curve is obtained as follows: Figure 11 As shown in the figure, when the blade pitch angle is 0°, the optimal tip speed ratio λ is... opt The value is 8.09, corresponding to C. Pmax It is 0.421.

[0124] exist Figure 4 In the control model, after obtaining the wind turbine speed command from equation (17), the generator quadrature-axis current control command is given according to equation (18). From the perspective of response time, the time required for electrical and control systems is much shorter than that for mechanical systems. The wind turbine speed tracking effect mainly depends on the inertial characteristics of the wind turbine speed. To accelerate the analysis, the control system, electrical system, and wind turbine inertial system are merged and simplified into a first-order system for speed tracking during simulation.

[0125]

[0126] In the formula, τ is the time constant of the inertial system, which can be estimated by formula (21).

[0127]

[0128] In the formula, T rated J is the rated rotor torque; r ω is the moment of inertia of the wind turbine.rated ω0 is the rated speed of the wind turbine, and ω0 is the initial speed of the wind turbine.

[0129] The wind turbine rotor radius is set to 46.7m, the cut-in wind speed is 3m / s, and the cut-out wind speed is 25m / s; the rotor inertial time constant is set to 10s. Figure 12 The time-history optimization performance analysis under stable wind speed conditions is given, where Figure 12 (a) is a scatter plot of the power variation of wind turbines under different wind speeds and different degrees of degradation. Figure 12 Figures (b)-(f) show the power changes before and after time-history optimization under different degradation conditions at wind speeds of 6 m / s, 7 m / s, 8 m / s, 9 m / s, and 10 m / s, respectively. In the figures, according to the definition of degradation in equation (10), and considering the possible situations that may occur during the long-term service of wind turbines, various cases with degradation degrees of 5%, 10%, 15%, 20%, 25%, and 30% are considered. Figure 12 (a) does not consider the impact of time-history optimization. As can be seen from the power scatter plot in the graph, the power output gradually decreases with increasing unit degradation. The higher the wind speed, the greater the power reduction. For example, at a wind speed of 6 m / s and a degradation rate of 5%, the power decreases from 381.2 kW to 361.9 kW; at a degradation rate of 30%, the power decreases to 260.4 kW. At a wind speed of 10 m / s and a degradation rate of 5%, the power decreases from 1765.5 kW to 1676.5 kW; at a degradation rate of 30%, the power decreases to 1206.3 kW.

[0130] from Figure 12 As shown in (b)-(f), time-history optimization after the aerodynamic performance of the wind turbine deteriorates can improve the power output to a certain extent. Relatively speaking, the higher the wind speed, the greater the power increase. For example, at a degradation degree of 5% and a wind speed of 6 m / s, the power increases from 361.9 kW to 362.1 kW; at a wind speed of 10 m / s, the power increases from 1676.5 kW to 1677.2 kW.

[0131] Define power improvement rate

[0132]

[0133] In the formula, P D P represents the power output of a wind turbine when its performance deteriorates. O The power output of the unit after time-scheduling optimization.

[0134] Although under the same degradation conditions, higher wind speeds result in greater power gains through time-history optimization, the improvement rate remains essentially unchanged.

[0135] Figure 13The power improvement rates obtained by time-history optimization of wind turbines under different degradation levels and wind speeds are presented. The improvement percentages vary depending on the degradation level: at 5% degradation, time-history optimization increases power output by approximately 0.05%; at 10% degradation, approximately 0.19%; at 15% degradation, approximately 0.48%; at 20% degradation, approximately 0.92%; at 25% degradation, approximately 1.55%; and at 30% degradation, approximately 2.46%. Because this is within the maximum wind energy tracking (MAXST) region, the power output improvement percentage is primarily related to the degree of degradation and less affected by wind speed. From 6 m / s to 10 m / s, as long as the degradation level remains constant, the power improvement percentage after time-history optimization remains essentially unchanged.

[0136] The above analysis is based on constant wind speed. However, the actual wind speed varies over time. Therefore, a segment of real-time wind speed at a certain wind farm is used as the input for the simulation analysis.

[0137] The wind speed in the figure varies between 4.2 m / s and 9.2 m / s, which is within the maximum wind energy tracking range of the wind turbine. Figure 14 The paper presents the power curves of wind turbines at degradation levels Γ of 5%, 15%, and 30%, including power variation curves before and after time-history optimization. Clearly, under the same wind speed input conditions, the higher the degradation level, the lower the output power. When the turbine has no performance degradation (i.e., Γ is 0), the power variation range is 301.3kW-883.5kW. At Γ of 5%, the power variation range before time-history optimization is 287kW-837.4kW. Comparatively, the power reduction is 4.7% {(287kW-301.3kW) / 301.3kW*100%) at the minimum power and 5.2% at the maximum power. The power reduction fluctuates around 5%, mainly due to the introduction of the first-order system considering the inertia of the wind turbine. After time-history optimization, the power variation range is 286.5kW-839.4kW. When the degradation degree Γ is 15%, the power variation range is 257.6kW-742.7kW; after time-history optimization, the power variation range is 256.1kW-751kW. When Γ is 30%, the power variation range is 211kW-594.2kW; after time-history optimization, the power variation range is 210.9kW-618.5kW. Comparing the variation ranges before optimization, it can be seen that the power is not greater at all times after time-history optimization than before. The main factor is the influence of the large inertia of the wind turbine; the power variation characteristics are related not only to the current wind speed but also to the wind speed variation characteristics over a period of time.

[0138] The effectiveness of the proposed "four-index" multi-angle observation algorithm is demonstrated. Computational case studies show that the proposed method is reliable and effective, revealing the evolution characteristics of the aerodynamic performance of large wind turbines in the field. The analysis results are consistent with theoretical analysis and provide new ideas and basis for quantitative evaluation. This provides important knowledge support for the operation, maintenance, and performance improvement of large wind turbines after they have reached their "old age."

[0139] The above description is merely an exemplary embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A time-history optimization method for wind energy tracking control of large wind turbine units considering aging effects, characterized in that, include: S100. Quantitative Evaluation of Unit Performance Degradation: Select SCADA parameters for the unit and perform data preprocessing; construct a quantitative evaluation model for unit performance degradation; and assess the aging status of the unit through on-site SCADA parameters. S200. Optimize unit control schedule: The deterioration status of the unit is determined by SCADA data, and the parameters are updated according to the deterioration status results. The impact of the deterioration degree is determined in the speed-power control curve. S300 Maximum Wind Energy Tracking Control: Based on the assessed aging status of the wind turbine, a tip speed ratio-wind energy utilization coefficient curve is plotted under aging conditions to approximately describe the basic performance of the wind turbine. This curve is then combined with the wind energy utilization coefficient C. P The relationship with the tip speed ratio λ is used to determine when the tip speed ratio is at its optimal value λ. opt At that time, the maximum value of the wind energy utilization coefficient is obtained; Selecting SCADA parameters includes: In the analysis of relevant parameters, SCADA parameters relevant to the calculation were selected as assessment indicators for performance degradation caused by unit aging. The selected SCADA parameters included hub speed, instantaneous wind speed, active power, ambient temperature, and blade pitch angle. Data preprocessing for SCADA parameters included: S110. Filling in missing data and handling outliers: Find and fill in missing data to facilitate subsequent processing; identify and handle null values, zero values, and outliers that are obviously inconsistent with reality; S120. High-frequency processing of temperature data: High-frequency processing of temperature data with long sampling periods is performed to achieve the same sampling frequency as wind speed, rotational speed, and power parameters, and the data is placed in the same data package as wind speed, rotational speed, and power to facilitate subsequent analysis; S130. Supplementary data on air density in wind farms: using the formula ρ=f(T) C The instantaneous air density of the wind farm is obtained by numerical calculation of temperature, humidity, and atmospheric pressure (RH), where ρ is the air density and T is the atmospheric pressure. C Temperature is measured in Celsius and recorded via a SCADA system; h represents altitude; RH represents relative humidity. S140. Wind Speed ​​Data Deviation Compensation and Filtering: Wind speed deviation is compensated by filtering to eliminate high-frequency components mixed in with the random changes in wind speed; the following formula is used for wind speed deviation compensation: In the formula, v1 is the compensated wind speed, P is the generator output power, v0 is the wind speed measured by the sensor, and S is the swept area of ​​the wind turbine rotor.

2. The method for optimizing the time history of wind energy tracking control for large wind turbine units considering aging effects according to claim 1, characterized in that, SCADA parameters are classified, including: The selected parameter data are divided into two categories: one category is directly used for wind energy utilization coefficient calculation; the other category is used as the observation value for data extraction when calculating wind energy utilization coefficient.

3. The method for optimizing the time history of wind energy tracking control for large wind turbine units considering aging effects according to claim 1, characterized in that, In the process of high-frequency temperature data processing, since temperature is a slow variable, minute-level temperature data is evenly distributed into second-level data, and the temperature data remains unchanged within one minute.

4. The method for optimizing the time history of wind energy tracking control for large wind turbine units considering aging effects according to claim 1, characterized in that, To address the random variations in wind speed, high-frequency components mixed in with the wind speed are filtered out using the following formula: In the formula, y i x is the output of the filter. (i-j) is the input to the filter, and n is the window length of the MAF filter.

5. The method for optimizing the time history of wind energy tracking control for large wind turbine units considering aging effects according to claim 1, characterized in that, During the maximum wind energy tracking phase, the tip speed ratio and wind energy utilization coefficient are kept constant, and the rotational speed ω... r With power P opt Satisfying Relationship: In the formula, C Pmax λ is the maximum wind energy utilization coefficient obtained from the wind turbine performance curve. opt R is the corresponding tip reduction ratio obtained from the wind turbine performance curve, where R is the rotor radius.

6. The method for optimizing the time history of wind energy tracking control for large wind turbine units considering aging effects according to claim 5, characterized in that, Based on the speed-power curve plotted using SCADA parameters, the wind turbine operation process is divided into four regions: transition zone 1, maximum wind energy tracking zone, transition zone 2, and constant power zone. It is determined that large speed changes occur in the maximum wind energy tracking zone, and the wind turbine in the maximum wind energy tracking zone is controlled by the given speed-power curve.

Images