A wind energy conversion model construction and optimization method based on fan index analysis

Abstract

The present application relates to the technical field of wind energy conversion and system optimization, and particularly relates to a wind energy conversion model construction and optimization method based on wind turbine index analysis, comprising the following steps: step one, wind turbine data preprocessing; step two, establishing a wind energy conversion model; step three, establishing a main shaft torque and tower thrust model; step four, establishing a material fatigue damage model; step five, optimizing main shaft torque, tower thrust and fatigue damage calculation parameters with the aid of a genetic algorithm; step six, comparing reference data and analyzing error results, and analyzing the error of model results by comparing with historical data and industry standards. The accuracy of the model and the reliability in actual application are evaluated using statistical methods to ensure that the model can effectively reflect the dynamic characteristics of the wind turbine operation in actual engineering. Through systematic modeling steps, the present application can effectively improve the wind energy conversion efficiency, enhance the accuracy of structural safety analysis, and provide a scientific basis for the maintenance and optimization of wind turbines.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind energy conversion and system optimization, and particularly relates to a wind energy conversion model construction and optimization method based on wind turbine index analysis. BACKGROUND

[0002] With the increasing use of renewable energy, wind energy as an important green energy has attracted more and more research and application. However, the wind turbine is affected by various factors during operation, and how to effectively improve its conversion efficiency and optimize the system performance has become an important research topic. Traditional wind energy conversion models often only focus on a single factor, such as wind speed or power generation, and lack comprehensive analysis of various indicators. This leads to insufficient robustness and adaptability of the model in actual application, especially when facing environmental noise and communication delay, the performance of the model will be significantly affected.

[0003] Therefore, the present application proposes a wind energy conversion model construction and optimization method based on wind turbine index analysis, aiming to realize comprehensive analysis and optimization of the wind turbine operation process through multi-step data processing and model construction. SUMMARY

[0004] The purpose of the present application is to solve the problems existing in the prior art, and to propose a wind energy conversion model construction and optimization method based on wind turbine index analysis. Through innovative data analysis and modeling strategies, this method can effectively improve the energy conversion efficiency and operation stability of the wind turbine, ensure the coordination between various indicators, and thus reduce system loss and enhance the adaptability of the wind turbine in different working environments.

[0005] In order to achieve the above purpose, the present application adopts the following technical solutions:

[0006] A wind energy conversion model construction and optimization method based on wind turbine index analysis, comprising the following steps:

[0007] Step 1: Wind turbine data preprocessing, first collect the operation data of the wind turbine under different wind speeds and load conditions, and perform cleaning and normalization processing on the data to eliminate noise and outliers, thereby providing accurate basic data for subsequent model construction;

[0008] Step 2: Establish a wind energy conversion model, based on the preprocessed data, construct a wind energy conversion model. This model is based on the relationship between wind speed and energy conversion efficiency, and uses mathematical formulas to describe the process of wind energy conversion into mechanical energy, ensuring that the model can reflect the actual operating state;

[0009] Step three: Establish the main shaft torque and tower thrust model, model the main shaft torque and tower thrust of the wind turbine, analyze the mechanical characteristics of the wind turbine under different working conditions, and establish the functional relationship between the main shaft torque and the wind speed, load; this model is used to calculate the stress of the wind turbine and evaluate its structural safety;

[0010] Step four: Establish a material fatigue damage model, combine the actual operating environment, and establish a material fatigue damage model. Analyze the fatigue life of the wind turbine structure using stress-strain relationship and rainflow counting method, and evaluate the damage and failure mode of the material in long-term operation to provide basis for maintenance decision;

[0011] Step five: Use genetic algorithm to optimize main shaft torque, tower thrust and fatigue damage calculation parameters. Based on the optimization search method of natural selection and genetics, various parameters in the wind energy conversion model, such as main shaft torque, tower thrust and fatigue damage calculation, are effectively optimized.

[0012] Step six: Compare the data and analyze the error results. By comparing with historical data and industry standards, the error of the model results is analyzed.

[0013] Preferably, in step one, the operating data of the wind turbine under different wind speed and load conditions are collected and cleaned and normalized to eliminate noise and outliers. Through these processes, the accuracy and effectiveness of the data used are ensured, providing a reliable data basis for the construction of subsequent models. At the same time, by analyzing the data distribution characteristics, key parameters and influencing factors are identified for further analysis and modeling in subsequent steps.

[0014] Preferably, in step two, by collecting and preprocessing various operating data of the wind turbine, a wind energy conversion model is established to analyze the working state of the wind turbine, including wind speed, power generation power and main shaft speed, etc. The mathematical relationship between wind speed and power generation power is established by using regression analysis method, thereby providing a basis for subsequent improvement of wind energy conversion efficiency.

[0015] Preferably, in step three, combined with sensor data and related physical models, the torque change of the wind turbine under different working angles is calculated from the main shaft torque of the wind turbine using dynamic equations; through numerical simulation, the operating parameters of the wind turbine under certain working conditions are obtained, and its efficiency and stability are evaluated.

[0016] Preferably, in step four, a material fatigue damage model is established, considering the stress condition of each component of the wind turbine in long-term operation; the fatigue life of the component is analyzed by using material mechanics theory, and the fatigue accumulation process is simulated by software to monitor the health status of the wind turbine component in real time and effectively predict the failure risk.

[0017] Preferably, in step five, first, define the fitness function, generate new generation of parameter combinations through genetic operations such as selection, crossover and mutation, then repeatedly evaluate its fitness, continue iteration until the optimization criteria are met, to improve the overall performance and reliability of the model. Finally, apply the optimization results to the wind energy conversion model, perform performance verification and result analysis, and provide data support for practical application.

[0018] Preferably, in step six, by comparing with historical data and industry standards, analyze the error of the model results, use statistical methods to evaluate the accuracy of the model and the reliability in practical application, and ensure that the model can effectively reflect the dynamic characteristics of the wind turbine operation in actual engineering.

[0019] Compared with the prior art, the present application has the following beneficial effects:

[0020] 1、The present application uses data mining and machine learning techniques to effectively predict the nonlinear relationship between wind speed, rotational speed and power generation based on in-depth analysis of various indicator data of the wind turbine. The model can dynamically adjust and optimize the operating parameters of the wind turbine to adapt to different wind speed conditions, thereby significantly improving the energy conversion efficiency of the wind turbine.

[0021] 2、The present application accurately evaluates the stress of the wind turbine under different working conditions by establishing a mechanical analysis model of the main shaft torque and the tower thrust. This model integrates various dynamic factors in the operation of the wind turbine and uses numerical simulation techniques to provide reliable basis for the operation of the wind turbine in complex environments through accurate calculation, thereby reducing the risk of system failure.

[0022] 3、The present application also introduces a material fatigue damage evaluation model, which analyzes the life and reliability of wind turbine components under changing operating conditions based on the theory of material science. This method not only monitors the health status of the components in real time, but also provides data-based maintenance recommendations, thereby optimizing the maintenance strategy of the wind turbine and prolonging the service life.

[0023] 4、The present application continuously improves the prediction accuracy of the model through model error analysis and feedback optimization mechanism. Real-time data updating and learning mechanism is used to ensure that the model can adapt to different environmental conditions, thereby reducing the uncertainty in operation and improving the economy and safety of the wind turbine. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 is an error distribution diagram of the calculated value and the reference value of the wind turbine in the embodiment of the present application;

[0025] Figure 2 is a comparison diagram of the estimated value and the reference value of the five wind turbines in the embodiment of the present application;

[0026] Figure 3 Error analysis diagram for the sum of squares of differences between estimated values and reference values of the fan in the embodiment of the present application;

[0027] Figure 4 Stability test diagram of the model in the embodiment of the present application;

[0028] Figure 5 Flowchart of the present application. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings, so that those skilled in the art can better understand the advantages and features of the present application, and the protection scope of the present application can be more clearly defined. The described embodiments of the present application are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0030] Embodiment:

[0031] Reference Figure 5 A wind energy conversion model construction and optimization method based on fan index analysis, comprising the following steps:

[0032] Step one: fan data preprocessing. First, collect the running data of the fan under different wind speeds and load conditions. Clean and normalize the data to eliminate noise and outliers, so as to provide accurate basic data for subsequent model construction.

[0033] Step two: establish a wind energy conversion model. Based on the preprocessed data, a wind energy conversion model is constructed. The model is mainly based on the relationship between wind speed and energy conversion efficiency, using mathematical formula to describe the process of wind energy conversion into mechanical energy, to ensure that the model can reflect the actual running state.

[0034] Step three: establish the main shaft torque and tower thrust model. Model the main shaft torque and tower thrust of the fan. By analyzing the mechanical properties of the fan under different working conditions, a function relationship between the main shaft torque and wind speed, load is established. This model can be used to calculate the stress of the fan and evaluate its structural safety.

[0035] Step four: establish a material fatigue damage model. Combined with the actual operating environment, a material fatigue damage model is established. The stress-strain relationship and rainflow counting method are used to analyze the fatigue life of the fan structure, to evaluate the possible damage and failure modes of the material in long-term operation, and to provide basis for maintenance decision.

[0036] Step five: Optimization of main shaft torque, tower thrust, and fatigue damage calculation parameters using genetic algorithms. Based on the principles of natural selection and genetics, this optimization search method can effectively optimize various parameters in the wind energy conversion model, such as main shaft torque, tower thrust, and fatigue damage calculation.

[0037] Step six: Analysis of model results' errors by comparing with historical data and industry standards. Statistical methods are used to evaluate the accuracy and reliability of the model in practical applications, ensuring that the model effectively reflects the dynamic characteristics of the wind turbine in actual engineering.

[0038] In the process of wind power generation, wind speed and wind turbine output power have a nonlinear relationship. When the wind speed is low, the wind turbine efficiency is high, and it can efficiently convert wind energy into electrical energy. When the wind speed is too high, the wind turbine power reaches saturation, and the excess wind energy will increase the main shaft torque and tower thrust, affecting the fatigue damage of the wind turbine mechanical components.

[0039] The working process of the wind turbine is influenced by various factors:

[0040] 1. Wind energy capture efficiency: power coefficient C p , which represents the efficiency of the wind turbine in converting wind energy into mechanical power, depends on factors such as wind speed and rotor speed.

[0041] 2. Rated wind speed: the rated wind speed of the wind turbine is 11.2 m / s, and when the wind speed exceeds this value, the output power remains constant (5 MW).

[0042] 3. Aerodynamic effect: as wind speed increases, tower thrust and main shaft torque also increase, especially when the wind speed exceeds the rated value.

[0043] To address this, a mathematical model will be established to estimate the main shaft torque and tower thrust of the wind turbine.

[0044] (1) Establishing a wind energy conversion model

[0045] Based on the Betz-Equation model and combined with fluid dynamics principles, the wind energy captured by the wind turbine is calculated:

[0046]

[0047] where P w is the wind energy power, ρ is the air density (1.225 kg / m 3 ), A = πr 2 is the swept area of the wind turbine, r is the wind turbine radius, and V w is the wind speed.

[0048] The relationship between the mechanical power P ref output by the wind turbine and the captured wind energy is determined by the power coefficient C pη represents the efficiency of the wind turbine in capturing wind energy:

[0049]

[0050] Power coefficient C p is a nonlinear coefficient that is affected by factors such as wind speed and rotor speed. At low wind speeds, C p increases with increasing wind speed; when the wind speed exceeds the rated wind speed, the power of the wind turbine reaches saturation, C p decreases accordingly.

[0051] Main shaft torque: Through energy conservation, the output power of wind energy and the main shaft torque and angular velocity have the following relationship:

[0052] P ref = T s · ω (3.)

[0053] where T s is the main shaft torque and ω is the angular velocity of the rotor. Thus, the main shaft torque can be calculated as:

[0054]

[0055] Tower thrust: The tower thrust F t is affected by wind speed and aerodynamic force, and is proportional to the square of the wind speed:

[0056]

[0057] where C t is the thrust coefficient, representing the force on the wind turbine at different wind speeds.

[0058] (2) Establishing the main shaft torque and tower thrust model

[0059] According to the wind speed, the model is divided into three regions: low wind speed, rated wind speed, and high wind speed.

[0060] (a) Low wind speed range (V w < 11.2 m / s)

[0061] Main shaft torque calculation:

[0062]

[0063] Tower thrust calculation:

[0064]

[0065] In this range, the wind speed output power is low, and the mechanical components are under less stress, with fatigue damage mainly coming from high cycle times.

[0066] (b) Rated wind speed range (Vw ≈11.2m / s)

[0067] Main shaft torque calculation:

[0068]

[0069] Tower thrust calculation:

[0070]

[0071] At rated wind speed, the wind speed output power reaches the maximum, and the fatigue damage accumulation is more uniform.

[0072] (c) High wind speed interval (V w > 11.2m / s)

[0073] Main shaft torque calculation:

[0074]

[0075] Tower thrust calculation:

[0076]

[0077] At high wind speed, the wind turbine power is reduced, the main shaft torque is reduced, but the tower thrust is increased, and the fatigue damage accumulation is accelerated.

[0078] In summary, the main shaft torque and tower thrust model can be represented by a piecewise function:

[0079] Main shaft torque calculation:

[0080]

[0081] Tower thrust calculation:

[0082]

[0083] (3) Establish material fatigue damage model

[0084] Using Palmgren-Miner linear cumulative damage theory:

[0085]

[0086] Where D is the cumulative fatigue damage, n i is the number of cycles at stress level F i , N Fi is the maximum number of cycles of the material at stress level F i , usually obtained by S-N curve.

[0087] (4) Optimize parameters such as spindle torque, tower thrust, and fatigue damage calculation using genetic algorithms.

[0088] To achieve the goals of minimizing spindle torque, reducing tower thrust, and minimizing fatigue damage, a fitness function of the following form can be designed:

[0089] F = w1T s +w2F t +w3D(15.)

[0090] Among them, w1, w2, and w3 are weights, reflecting the importance of different objectives.

[0091] Next, the model is solved, and the results are compared with the reference data and the error results are analyzed.

[0092] (1) The stress / torque results of the 5 wind turbines are shown in Table 1 and Table 2.

[0093] Table 1. Main shaft torque values ​​of 5 wind turbines at all times.

[0094]

[0095]

[0096] Table 2. Tower thrust values ​​of the five wind turbines at all times.

[0097]

[0098] (2) Comparison with reference data

[0099] The following section will present the comparison between the calculation results and the actual data from three aspects: error distribution, comparison between the estimated values ​​and reference values ​​of the five wind turbines, and the sum of squares of the differences between the estimated values ​​and reference values.

[0100] (a) Error distribution of the wind turbine as shown in Figure 1 As shown

[0101] Figure 1 The error distribution of spindle torque and tower thrust is shown, displaying the differences between estimated and reference values. Spindle torque error distribution. Figure 1 In the graph above, the X-axis represents error (Nm), and the Y-axis represents frequency. The error distribution is mainly concentrated between -2 and 0 Nm, exhibiting a left-leaning distribution. The peak region is between -1 and -0.5 Nm, with a high frequency, indicating that the estimated spindle torque differs most from the reference value within this range. From the deviation analysis, the left-leaning error distribution suggests that the estimated values ​​are generally lower than the reference values. This may be due to certain parameters in the model (such as the power coefficient C). P Or thrust coefficient C Tinaccurate), leading to systematically lower estimated values. The error range is large, from -2 to 1 Nm, indicating that the estimation model has large errors in some cases. It can be considered to optimize the model by adjusting the power coefficient C P and the thrust coefficient C T so that the estimated values are closer to the reference values. Further analyze the accuracy of wind speed and other input parameters to ensure the accuracy of input data.

[0102] Tower thrust error distribution Figure 1 From the error distribution pattern, the error is mainly concentrated between 6 and 10 N, showing a relatively symmetric distribution. The peak area is between 8 and 9 N, with a high frequency, indicating that in this interval, the difference between the estimated value of the tower thrust and the reference value is the largest. From the bias analysis, the error distribution is relatively symmetric, but overall biased to the right, indicating that the estimated value is generally higher than the reference value. This may be due to the thrust coefficient C t being set too high, leading to a systematic bias in the estimated value. The error range is large, from 6 to 12 N, indicating that the estimation model has large errors in some cases.

[0103] (b) Comparison of the estimated values of the first 5 wind turbines with the reference values as shown in Figure 2

[0104] Figure 2 shows the comparison of the estimated values of the main shaft torque and the tower thrust of the 5 wind turbines with the reference values. The following is a detailed analysis of each subgraph:

[0105] Figure 2 The left chart inshows the main shaft torque of wind turbines 1-5, with the X-axis being the time step (from 0 to 100) and the Y-axis being the main shaft torque (Nm), with the solid line being the estimated value and the dashed line being the reference value. From the overall trend, the estimated values and reference values of the main shaft torque of all wind turbines show a fluctuating trend over time. From the error analysis, the estimated value of wind turbine 1 has a significant difference from the reference value at most time steps, especially in the first 50 time steps, where the estimated value is lower. The estimated value of wind turbine 2 is relatively close to the reference value at some time steps, but there is still an error overall, especially in the high torque area. The difference between the estimated value and the reference value of wind turbine 3 is large in the first 50 time steps, and improves in the second half, but there is still an error. The difference between the estimated value and the reference value of wind turbine 4 is relatively small, especially after time step 50, showing good estimation effect. The estimated value of wind turbine 5 is relatively close to the reference value at most time steps, but there is still an error in some peak areas.

[0106] Figure 2The right-hand chart shows the tower thrust for turbines 1-5, with the Y-axis representing tower thrust (N), otherwise identical to the left-hand chart. Overall, the estimated tower thrust for all turbines shows a fluctuating trend over time, compared to the reference value. Error analysis shows that the estimated value for turbine 1 is significantly higher than the reference value for most time steps, especially in the first 30 time steps. The estimated value for turbine 2 is significantly higher than the reference value for most time steps, especially in the first 50 time steps. The estimated value for turbine 3 is significantly higher than the reference value for most time steps, indicating a large systematic error. The estimated value for turbine 4 is significantly higher than the reference value for the first 50 time steps, improving somewhat in the latter half but still showing errors. The estimated value for turbine 5 is significantly higher than the reference value for most time steps, especially in the first 50 time steps.

[0107] In summary, although the estimated values ​​are close to the reference values ​​at certain time steps and for certain wind turbines, significant errors still exist overall. Further optimization of the model and calibration parameters can improve the accuracy and stability of the estimation.

[0108] (c) The sum of squares of the differences between the estimated value and the reference value, as shown in the figure. Figure 3 As shown

[0109] Figure 3 This shows the trend of the sum of squared errors of spindle torque and tower thrust over 100 time steps. For the sum of squared errors of spindle torque as shown... Figure 3 The above figure shows that in the initial stage (time steps 0 to 10), from 6 × 10 13 It rose rapidly and then stabilized at 7.5 × 10⁻⁶ in about 10 time steps. 13 This means the model rapidly reduced its error by 25% in the initial stage. In the intermediate stage (10 to 50 time steps), the sum of squared errors was 7 × 10⁻⁶. 13 Up to 8×10 13 Fluctuations ranged between 1×10 13 This represents 13.3% of the stable values, indicating that the model performs well in capturing dynamic changes in wind speed and power. In the later stages (50 to 100 time steps), the sum of squared errors remained at 7.5 × 10⁻⁶. 13 Near the same timeframe, the fluctuation range did not increase significantly, indicating that the model's estimate of the spindle torque remained stable throughout the entire time period.

[0110] For the sum of squares of tower thrust error, such as Figure 3 The diagram below shows the initial stage (time steps 0 to 10), starting from 1×10 13 It rose rapidly and stabilized at 7×10 at approximately 10 time steps. 13 This means the model rapidly reduced its error by 30% in the initial stage. In the intermediate stage (10 to 50 time steps), the sum of squared errors fluctuated by 6.8 × 10⁻⁶.13 to 7.2 x 10 13 12 , which is 5.7% of the stable value, indicating that the model has certain stability in dealing with the relationship between wind speed and power. In the later stage (50 to 100 time steps), the sum of squared errors continues to fluctuate around 7 x 10 13 , showing certain periodicity and stability.

[0111] Table 3 Sum of squared differences between estimated and reference values

[0112]

[0113] The sum of squared differences between estimated and reference values at all times is calculated in Table 3, which shows that the sum of squared differences of main shaft torque is relatively high, but the model shows certain stability in dealing with wind speed and power data. The stability of the mean of the sum of squared errors indicates that the model performs consistently at different time steps without significant error fluctuations, providing a good foundation for further optimization. The mean of the sum of squared errors of tower thrust is relatively high, but the fluctuation range is small, showing the stability of the model in dealing with the relationship between wind speed and power. Although the error is large, the model maintains a relatively stable error level throughout the entire period, indicating that the model has certain reliability in dealing with wind distribution and tower stress.

[0114] Next, the stability of the model is tested:

[0115] Figure 4 The stability of the sum of squared errors of main shaft torque and tower thrust is tested. By analyzing the sum of squared errors of main shaft torque of 10 experiments, the following results can be observed:

[0116] (1) For the range and fluctuation of the sum of squared errors of main shaft torque: the range is 7.62 x 10 15 to 7.626 x 10 15 , the maximum value is 7.625 x 10 15 (experiment number 8), and the minimum value is 7.62 x 10 15 (experiment number 2). The fluctuation range is about 0.006 x 10 15 , showing that the overall fluctuation is small, indicating that the system maintains good stability in most cases.(2) For the trend analysis of the sum of squared errors of main shaft torque: the data shows slight fluctuations between experiment numbers 1 to 10, but the overall trend is relatively stable. Although the peak values of experiment numbers 3 and 8 are significant, the remaining data points are concentrated, showing the stability of the system in most cases.(3) For the statistical characteristics of the sum of squared errors of shaft torque: the mean is about 7.623 x 10 15 ​The standard deviation is smaller, and the data is concentrated around the mean, indicating that the system performs stably in terms of spindle torque. (4) Analysis of outliers for the sum of squares of spindle torque error: The values of experiment numbers 3 and 8 are much higher than the others, which may be outliers, but the rest of the data points show that the system performs stably in most cases. (5) Stability evaluation for the sum of squares of spindle torque error: Although there are a few peaks, the overall data fluctuation is small, showing the stability of the system in most cases. The system can maintain a low error under most experimental conditions, indicating that the model has a certain robustness in handling spindle torque. (6) Analysis of the influence of experimental conditions on the sum of squares of spindle torque error: The peaks of experiment numbers 3 and 8 may be related to specific experimental condition changes, and further analysis of these conditions can help optimize the model. (7) Analysis of the robustness of the model: Although there are a few outliers, the model performs well in most cases, indicating a certain robustness.

[0117] By analyzing the tower thrust error sum of squares of 10 experiments, the following results can be observed: (1) The range and fluctuation of the tower thrust error sum of squares: The range is 7.006×10 15 to 7.009×10 15 , the maximum value is 7.009×10 15 (experiment number 4), and the minimum value is 7.006×10 15 (experiment number 2). The fluctuation range of the tower thrust error sum of squares is about 0.003×10 15 , showing extremely small fluctuations, indicating that the system has very high stability in terms of tower thrust. (2) Trend analysis of the tower thrust error sum of squares: The data remains basically stable between experiment numbers 1 to 10, with minimal fluctuations. The values of experiment numbers 4 and 7 are slightly higher, but the overall trend is stable, indicating that the system performs well in most cases. (3) Statistical characteristic analysis of the tower thrust error sum of squares: The mean of the tower thrust error sum of squares is about 7.0075×10 15 The standard deviation is smaller, and the data is more concentrated, indicating that the system performs very stably in terms of tower thrust. (4) Analysis of outliers for the tower thrust error sum of squares: The values of experiment numbers 4 and 7 are slightly higher than the others, but the overall fluctuation range is extremely small, showing the stability of the system. (5) Qualitative analysis of the tower thrust error sum of squares: The data fluctuation is extremely small, showing that the system has very high stability in terms of tower thrust. The system can maintain a low error under all experimental conditions, indicating that the model is very robust in handling tower thrust. (6) Analysis of the influence of experimental conditions on the tower thrust error sum of squares: The peaks of experiment numbers 4 and 7 also need to be further analyzed to determine whether there are specific experimental conditions that cause the error to increase. (7) Robustness analysis of the model for the tower thrust error sum of squares: The model performs stably under all experimental conditions, showing extremely high robustness.

[0118] From the above, the following conclusions are obtained: (1) Stability: the variation range of the error sum of squares of the main shaft torque and the tower thrust is small in 10 experiments, and the mean and standard deviation are at a low level, indicating that the model has good stability under different experimental conditions. (2) Robustness: the overall trend of the error sum of squares shows that the model has strong robustness to random disturbances and can maintain stable performance under different wind speed conditions.

[0119] In addition, the basic symbols in the embodiments of the present application are explained as follows:

[0120]

[0121] In summary, through the innovative modeling and optimization method, the present application can effectively improve the wind energy conversion efficiency and enhance the operation stability of the system, and provide strong support for the sustainable development of the wind energy industry.

[0122] The description and practice disclosed in the present application are easy to think and understand for ordinary skilled in the art, and some improvements and refinements can be made without departing from the principles of the present application. Therefore, the modifications or improvements made without departing from the spirit of the present application should be considered as the protection scope of the present application.

Claims

1. A wind energy conversion model construction and optimization method based on wind turbine index analysis, characterized in that, Comprising the following steps: Step one: fan data preprocessing, first collect fan running data under different wind speed and load conditions, clean and normalize the data to eliminate noise and outliers, and provide accurate basic data for subsequent model construction; Step two: establish a wind energy conversion model, based on the preprocessed data, construct a wind energy conversion model, which is based on the relationship between wind speed and energy conversion efficiency, uses mathematical formulas to describe the process of converting wind energy into mechanical energy, and ensures that the model can reflect the actual operating state; Step three: establish a main shaft torque and tower thrust model, model the main shaft torque and tower thrust of the fan, analyze the mechanical properties of the fan under different working conditions, and establish a functional relationship between the main shaft torque and wind speed, load; this model is used to calculate the stress of the fan and evaluate its structural safety; According to the wind speed, the model is divided into three regions: low wind speed, rated wind speed, and high wind speed; The main shaft torque and tower thrust model is represented by a piecewise function: Main shaft torque calculation: ， Tower thrust calculation: ， in, Let the air density be denoted as . , The sweeping area of ​​the wind turbine. Where is the radius of the wind turbine. For wind speed, power coefficient A nonlinear coefficient, affected by wind speed and rotor speed, at low wind speeds... As wind speed increases; when the wind speed exceeds the rated wind speed, the fan power reaches saturation. It then decreased; It is the spindle torque. It is the angular velocity of the wind turbine; For tower thrust, It is the thrust coefficient, which indicates the stress on the wind turbine at different wind speeds; Step 4: Establish a material fatigue damage model. In combination with the actual operating environment, establish a material fatigue damage model, use stress-strain relationship and rainflow counting method to analyze the fatigue life of the wind turbine structure, evaluate the possible damage and failure modes of the material during long-term operation, and provide a basis for maintenance decisions. Using Palmgren-Miner linear cumulative damage theory: wherein, is the cumulative fatigue damage, is the stress level, is the number of cycles at stress level is the maximum number of cycles at stress level obtained by S-N curve; Step five: optimization of main shaft torque, tower thrust and fatigue damage calculation parameters by means of genetic algorithm, optimization search method based on natural selection and genetic principles, effectively optimizing main shaft torque, tower thrust and fatigue damage calculation in wind energy conversion model, involving fitness function : wherein, is the weight, reflecting the importance of different objectives; Step six: comparison and error result analysis of reference data, analyzing the error of model results by comparing with historical data and industry standards.

2. The method of claim 1, wherein, In step one, collect fan running data under different wind speed and load conditions, and clean and normalize these data to eliminate noise and outliers; Through these processes, ensure the accuracy and effectiveness of the data used, and provide a reliable data basis for subsequent model construction; by analyzing the data distribution characteristics, identify key parameters and influencing factors to facilitate more in-depth analysis and modeling in subsequent steps.

3. The method of claim 2, wherein, In step two, by collecting and preprocessing various running data of the fan, a wind energy conversion model is established to analyze the working state of the fan, including wind speed, power generation power and main shaft speed; using regression analysis method, the mathematical relationship between wind speed and power generation power is established, which provides the basis for the subsequent improvement of wind energy conversion efficiency.

4. The method of claim 3, wherein, In step three, combined with sensor data and physical model, starting from the main shaft torque of the fan, the torque change of the fan under different working angles is calculated by using dynamics equation; through numerical simulation, the running parameters of the fan under certain working conditions are obtained, and its efficiency and stability are evaluated.

5. The method of claim 4, wherein, In step four, a material fatigue damage model is established to consider the stress of each component of the fan in long-term operation; the fatigue life of the component is analyzed by using material mechanics theory, and the fatigue accumulation process is simulated by software to monitor the health status of the fan components in real time and effectively predict the failure risk during operation.

6. The method of claim 5, wherein, In step five, first, define the fitness function, generate new generation parameter combinations through selection, crossover and mutation genetic operations, then repeatedly evaluate their fitness, continue iteration until the optimization criteria are met to improve the overall performance and reliability of the model; finally, apply the optimization results to the wind energy conversion model for performance verification and result analysis to provide data support for actual application.

7. The method of claim 6, wherein the method further comprises: In step six, by comparing with historical data and industry standards, analyze the error of the model results, use statistical methods to evaluate the accuracy and reliability of the model in actual application, and ensure that the model can effectively reflect the dynamic characteristics of the fan operation in actual engineering.

Images