Prediction method, system, equipment and medium for ultimate load of fan unit

Abstract

The present invention discloses a method and system, equipment and medium for predicting the ultimate load of a wind turbine unit. The prediction method includes: respectively collecting environmental parameters of each unit in the target site; inputting the environmental parameters into an ultimate load prediction model to obtain the ultimate load corresponding to each unit; and generating a prediction result corresponding to the target site according to the ultimate load. The present invention establishes an explicit prediction model for the ultimate load of a wind turbine, and can predict the ultimate load of the wind turbine unit points at the target wind farm site without iterative optimization calculations, thereby avoiding the large amount of time spent on detailed simulation calculations and greatly saving computing resources. At the same time, it avoids the drawbacks of the existing technology of excessively large redundant deviations in ultimate load estimation, resulting in poor accuracy and practicality, and can accurately locate the unit point with the largest ultimate load, thereby improving the prediction efficiency while ensuring that the prediction results have high accuracy, and has industry promotion value.

Description

Technical Field

[0001] The present invention relates to the field of wind power technology, and in particular to a method and system, equipment and medium for predicting the ultimate load of a wind turbine unit. Background Art

[0002] During the load suitability assessment process for a wind farm site, it is necessary to analyze the ultimate load of each wind turbine at the site to ensure that the ultimate load over its lifecycle is within the design value. Two approaches are commonly used in the existing technology. One approach is to conduct a point-by-point assessment, simulating the ultimate load of each unit based on the wind parameter information at each unit site, and assessing the ultimate safety of each unit. The other approach is to take the maximum envelope value of the environmental parameters at each point in the wind farm, equating the environmental parameters of all points to the environmental parameters of a "virtual" point. Using this environmental parameter as input, the ultimate load of the unit is calculated. This ultimate load is greater than the ultimate load of any unit at any point in the site. By assessing the ultimate load safety of the unit at the virtual point, the ultimate load safety of all units at the entire site is ensured.

[0003] However, the first approach requires high simulation performance, consumes large amounts of computing resources, and takes a long time to simulate. Therefore, it is difficult to implement when simulation capabilities are insufficient or the number of site locations is large. While the second approach is computationally efficient, the resulting extreme load results have a large safety margin for all units at the site, resulting in low accuracy and poor practicality. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the above-mentioned problems existing in the process of calculating the limit load of wind turbines at specific sites in the prior art, and to provide a method and system, equipment and medium for predicting the limit load of wind turbine units.

[0005] The present invention solves the above technical problems through the following technical solutions:

[0006] The present invention provides a method for predicting the ultimate load of a fan unit, comprising:

[0007] Collect environmental parameters of each unit in the target site separately;

[0008] Inputting the environmental parameters into a limit load prediction model to obtain the limit load corresponding to each unit;

[0009] A prediction result corresponding to the target site is generated according to the ultimate load.

[0010] Preferably, the prediction method further includes:

[0011] Based on the design environmental parameters of each unit in the target site, obtaining the corresponding reference limit load of each unit;

[0012] Based on the design environmental parameters, obtaining the ultimate load sensitivity samples and environmental parameter samples corresponding to each of the units;

[0013] Using the extreme load sensitivity sample as a function and the environmental parameter sample as an independent variable to perform model training, respectively obtaining sensitivity models corresponding to different extreme loads;

[0014] The limit load prediction model corresponding to the design environment parameter is obtained according to the reference limit load and the corresponding sensitivity model.

[0015] Preferably, after the step of obtaining the ultimate load prediction model corresponding to the design environment parameters, the method further comprises:

[0016] Based on the design environment parameters, generating environment parameters for verification;

[0017] Obtaining a load verification value corresponding to the environmental parameter used for verification;

[0018] inputting the environmental parameters for verification into the ultimate load prediction model to obtain the load prediction value;

[0019] The ultimate load prediction model is adjusted according to the load verification value and the load prediction value.

[0020] Preferably, the step of obtaining the extreme load sensitivity samples and environmental parameter samples corresponding to each of the units based on the design environmental parameters includes:

[0021] determining a variation range of the design environment parameter;

[0022] performing discrete processing on the environmental parameter within the variation interval to generate an environmental parameter sample;

[0023] Obtaining an ultimate load sample corresponding to the environmental parameter sample through simulation calculation;

[0024] The ultimate load sample is divided by the reference ultimate load to obtain the ultimate load sensitivity sample.

[0025] Preferably, the step of generating the prediction result corresponding to the target site according to the extreme load includes:

[0026] Respectively obtaining the maximum value of the limit loads of all units corresponding to each target load component as the limit load of the target load component;

[0027] The ultimate load of each target load component is compared with the design ultimate load of each unit, and the safety of the ultimate load corresponding to each unit in the target site is evaluated to generate the prediction result.

[0028] The present invention also provides a system for predicting the ultimate load of a wind turbine unit, comprising:

[0029] Environmental parameter collection module, used to collect environmental parameters of each unit in the target site;

[0030] A limit load acquisition module, configured to input the environmental parameters into a limit load prediction model to obtain the limit load corresponding to each unit;

[0031] A prediction result generating module is used to generate a prediction result corresponding to the target site according to the ultimate load.

[0032] Preferably, the prediction system further includes:

[0033] A reference load acquisition module, configured to acquire a reference limit load corresponding to each unit in the target site based on the design environmental parameters of each unit;

[0034] A sample acquisition module, configured to acquire, based on the design environmental parameters, the ultimate load sensitivity samples and environmental parameter samples corresponding to each of the units;

[0035] A sensitivity model training module is used to perform model training using the extreme load sensitivity sample as a function and the environmental parameter sample as an independent variable to obtain sensitivity models corresponding to different extreme loads;

[0036] A prediction model generation module is used to obtain the limit load prediction model corresponding to the design environment parameter based on the benchmark limit load and the corresponding sensitivity model.

[0037] Preferably, the prediction system further includes:

[0038] A parameter generation module for verification, used to generate environmental parameters for verification based on the design environment parameters;

[0039] A verification value acquisition module, used to obtain a load verification value corresponding to the environmental parameter used for verification;

[0040] A prediction value acquisition module, configured to input the environmental parameters for verification into the ultimate load prediction model to obtain load prediction values ​​corresponding to the simulated environmental parameters;

[0041] The prediction model adjustment module is used to adjust the extreme load prediction model according to the load verification value and the load prediction value.

[0042] Preferably, the sample acquisition module includes:

[0043] a parameter interval determining unit, configured to determine a variation interval of the design environment parameter;

[0044] a discrete processing unit, configured to perform discrete processing on the environmental parameter within the variation interval to generate an environmental parameter sample;

[0045] An extreme load sample acquisition unit, configured to acquire an extreme load sample corresponding to the environmental parameter sample through simulation calculation;

[0046] The sensitivity sample acquisition unit is configured to divide the limit load sample by the reference limit load to acquire the limit load sensitivity sample.

[0047] Preferably, the prediction result generation module includes:

[0048] a limit load acquisition unit, configured to respectively acquire the maximum value of the limit loads of all units corresponding to each target load component as the limit load of the target load component;

[0049] The wind turbine unit evaluation unit is used to compare the ultimate load corresponding to each target load component with the design ultimate load of each unit, evaluate the safety of the ultimate load corresponding to each unit in the target site, and generate the prediction result.

[0050] The present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for predicting the ultimate load of a wind turbine unit when executing the computer program.

[0051] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for predicting the limit load of a wind turbine unit is implemented.

[0052] The positive and progressive effects of the present invention are as follows: the wind turbine unit limit load prediction method, system, equipment, and medium of the present invention establish an explicit wind turbine limit load prediction model, which can predict the limit load of wind turbine unit points at the target wind farm site without iterative optimization calculations, avoiding the time-consuming detailed simulation calculations and greatly saving computing resources. At the same time, it avoids the drawbacks of the existing technology of excessive redundant deviation in limit load estimation, resulting in poor accuracy and practicality, and can accurately locate the unit point with the largest limit load, thereby improving prediction efficiency while ensuring high accuracy of the prediction results, and has strong industry promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flow chart of a method for predicting the ultimate load of a wind turbine unit according to embodiment 1 of the present invention.

[0054] Figure 2 This is an example flow chart of establishing an extreme load prediction model in Example 1 of the present invention.

[0055] Figure 3 This is a module schematic diagram of a system for predicting the ultimate load of a wind turbine unit according to embodiment 2 of the present invention.

[0056] Figure 4 This is a structural block diagram of an electronic device according to embodiment 3 of the present invention. DETAILED DESCRIPTION

[0057] The present invention is further described below by way of examples, but the present invention is not limited to the scope of the examples.

[0058] Example 1

[0059] See also Figure 1 This embodiment specifically provides a method for predicting the ultimate load of a wind turbine unit, including:

[0060] S101. Collect environmental parameters of each unit in the target site respectively;

[0061] S102. Input the environmental parameters into the limit load prediction model to obtain the limit load corresponding to each unit;

[0062] S103. Generate target site ultimate load safety prediction results based on the comparison between the ultimate load and the design ultimate load.

[0063] This embodiment provides a prediction model-based method for predicting the ultimate load of a wind turbine group to achieve prediction and assessment of the ultimate load of wind turbines at a specific site. Specifically, the method includes two steps: establishing a prediction model for the ultimate load of wind turbines and evaluating the ultimate load of wind turbines at each point in the specific site.

[0064] The process for assessing the ultimate load of wind turbines at a specific site is as follows: For each wind turbine location, the environmental parameters of each location are first obtained. Using these environmental parameters as input, the ultimate load prediction model is used to determine the ultimate load under these environmental parameters. This ultimate load assessment is then performed point by point for the wind farm. For each load component, the maximum ultimate load of all wind turbines for that load component is taken as the ultimate load for that load component at the site. By comparing the site ultimate load with the unit's design ultimate load, the site's ultimate load safety is determined. Because the ultimate load prediction model is explicit, this method can save significant computational resources and time compared to detailed calculations.

[0065] The above two links realize the rapid and accurate assessment of the ultimate load of each point in the wind farm.

[0066] As a preferred embodiment, the ultimate load prediction model in step S102 is obtained by the following steps:

[0067] S1. Based on the design environmental parameters of each unit at the target site, including air density, turbulence intensity, vertical wind shear index, inflow angle, and 50-year maximum turbulent wind speed, obtain the corresponding benchmark ultimate load for each unit;

[0068] S2. Based on the design environmental parameters, obtain the corresponding extreme load sensitivity samples and environmental parameter samples for each unit;

[0069] S3. Use the extreme load sensitivity samples as functions and the environmental parameter samples as independent variables to perform model training, and obtain sensitivity models corresponding to different extreme loads;

[0070] S4. Obtain an ultimate load prediction model corresponding to the design environment parameters based on the benchmark ultimate load and the corresponding sensitivity model.

[0071] The process of developing a limit load prediction model includes four steps: generating a baseline limit load, generating a limit load sensitivity model, generating a limit load prediction model, and verifying the accuracy of the limit load prediction model. During the baseline limit load generation process, step S1 first selects the unit requiring load assessment. Based on the unit's designed environmental conditions, detailed limit load calculations are performed using wind turbine load simulation software to obtain the baseline limit load.

[0072] Step S2 obtains the extreme load sensitivity samples and environmental parameter samples corresponding to each unit based on the design environmental parameters. Preferably, step S2 includes:

[0073] Determine the variation range of design environment parameters;

[0074] Discretize the environmental parameters within the variation range to generate environmental parameter samples;

[0075] Obtain the ultimate load sample corresponding to the environmental parameter sample through simulation calculation;

[0076] Divide the ultimate load sample by the benchmark ultimate load to obtain the ultimate load sensitivity sample.

[0077] Specifically, in the process of generating the extreme load sensitivity model, the unit design environmental conditions are taken as the center, the variation range of the environmental parameters is set, the environmental parameters are discretized within the range to form environmental parameter samples, the extreme load is calculated using wind turbine load simulation software, the extreme load samples under each set of environmental parameters are obtained, the extreme load samples are divided by the benchmark extreme load, the extreme load sensitivity samples are obtained, and the extreme load sensitivity prediction model is set.

[0078] In step S3, the model is trained using the environmental parameter samples as independent variables and the extreme load sensitivity samples as functions to obtain sensitivity models for different extreme loads. In step S4, the extreme load sensitivity model is multiplied by the reference extreme load to obtain an extreme load prediction model.

[0079] As a preferred embodiment, after the ultimate load prediction model is generated, it is verified and adjusted. Specifically, after step S4, the following steps are further included:

[0080] S5. Based on the design environment parameters, generate environmental parameters for model verification;

[0081] S6. Obtaining the load verification value corresponding to the environmental parameter used for model verification;

[0082] S7. Input the environmental parameters used for model verification into the ultimate load prediction model to obtain the corresponding load prediction value;

[0083] S8. Adjust the ultimate load prediction model based on the load verification value and the load prediction value.

[0084] Specifically, within the environmental parameter definition domain of the extreme load prediction model, several sets of environmental parameters are randomly generated. These parameters include air density, turbulence intensity, vertical wind shear index, inflow angle, and the 50-year maximum turbulent wind speed. These parameters are used as input for a detailed load calculation to obtain verified extreme load values. Simultaneously, these sets of environmental parameters are substituted into the extreme load prediction model to obtain predicted extreme load values. By comparing the two, the extreme load prediction model is verified and adjusted. The verified extreme load prediction model can be used to predict the extreme loads of wind turbines.

[0085] As a preferred embodiment, step S103 includes:

[0086] Obtain the maximum value of the limit loads of all units corresponding to each target load component as the limit load of the target load channel;

[0087] The ultimate load corresponding to each target load component is compared with the design ultimate load of each unit to evaluate the safety of the ultimate load corresponding to each unit in the target site and generate prediction results.

[0088] Specifically, the process for assessing the ultimate load of wind turbines at a specific site, point by point, is as follows: For each wind turbine location, environmental parameters are first obtained. Using these environmental parameters as input, the ultimate load under the influence of these environmental parameters is calculated using an ultimate load prediction model. This ultimate load prediction is then performed point by point for the wind farm. For each load component, the maximum value of the ultimate load of that load component across all locations is taken as the ultimate load of that load component at the site. By comparing the site's ultimate load with the unit's design ultimate load, the site's ultimate load safety is determined. Because the ultimate load prediction model is explicit, this method can save significant computing resources and time compared to detailed calculations. Thus, through these two steps, rapid and accurate assessment of the ultimate load at each wind farm location can be achieved.

[0089] In a specific prediction example, Figure 2 The diagram shows the process for establishing a wind turbine limit load prediction model. The model establishment process, in sequential order, includes: generating a baseline limit load for the unit 10; determining the possible range of environmental parameter variations 11; discretely combining environmental parameters to form input parameter samples 12; simulating using load simulation software to generate limit load samples corresponding to the input parameter samples 13; establishing a mapping relationship between environmental parameters and limit load sensitivity coefficients to form a load sensitivity coefficient regression model 14; considering each limit load component independently, and continuously multiplying the load sensitivity regression model for each environmental parameter by the baseline limit load for that load component to form a limit load prediction model 15; and verifying the accuracy of the limit load prediction model 16.

[0090] The generation of the unit's benchmark limit load can be achieved through detailed load simulation according to industry specifications. It is necessary to obtain the limit load under each operating condition, each load channel, and each wind speed condition to form a limit load matrix. Environmental parameters include the following variables: air density, turbulence intensity, vertical wind shear index, inflow angle, and wind speed of the maximum turbulent wind once in 50 years. The possible range of environmental parameter changes is determined according to the range of site environmental parameters in the target market of the unit. The discrete processing of environmental parameters can be carried out in the following way:

[0091] Air density is discretized within the possible range of site variations, with a step size of 0.01 kg / m³. Under a given air density, the turbulence intensity at each wind speed is discretized to the possible range of site turbulence intensity, with 10 equally spaced points. The wind speed range is from the cut-in speed to the cut-out speed, with a step size of 1 m / s. The vertical wind shear index, inflow angle, and the 50-year maximum turbulent wind speed are also discretized to the possible range of site variations, with 10 equally spaced points. The environmental parameter combination utilizes a controlled variable method. When the vertical wind shear index, inflow angle, and the 50-year maximum turbulent wind speed are varied, the air density and turbulence intensity remain unchanged at their design values. When the air density and turbulence intensity are varied, the vertical wind shear index, inflow angle, and the 50-year maximum turbulent wind speed remain unchanged at their design values. However, for each air density condition, discrete values ​​for turbulence intensity are required for each wind speed.

[0092] Through simulation, we obtained samples of wind turbine extreme loads. For each combination of the aforementioned environmental parameters, we generated a matrix of extreme load samples for each operating condition, each load component, and each wind speed. For the extreme load sensitivity parameter prediction model, we divided the extreme load samples by the baseline extreme load matrix to obtain the percentage of the sample load to the baseline load. Separate analyses were conducted for each operating condition, each load component, and each wind speed. Regression analysis was performed using air density, turbulence intensity, inflow angle, wind shear index, and the 50-year maximum turbulent wind speed as independent variables, and the load percentage as the dependent variable. A polynomial regression model was selected.

[0093] Regression analysis is divided into four categories. The first category is the regression function of inflow angle and ultimate load, which is set as a univariate polynomial and includes a constant term. The second category is the regression function of wind shear index and ultimate load, which is set as a univariate polynomial and includes a constant term. The third category is the regression function of wind speed of the maximum turbulent wind once in 50 years and ultimate load, which is set as a univariate polynomial and includes a constant term. The fourth category is the regression function of air density, turbulence intensity and ultimate load, which is set as a two-variable polynomial and includes a constant term. The order of the above polynomials is adjusted according to the regression error, and the maximum order does not exceed three.

[0094] The ultimate load prediction model analyzes each operating condition, each load component, and each wind speed separately. The benchmark ultimate load and the four types of ultimate load sensitivity models mentioned above are continuously multiplied to obtain the ultimate load prediction model. The ultimate load for a given load component is defined as the maximum ultimate load of the load channel under all operating conditions and all wind speeds. The ultimate load prediction model and parameters are described as follows:

[0095] Ultimate load prediction model: L' = L*f1*f2*f3*f4, L" = max{L' i，j};

[0096] L': The predicted value of the ultimate load under given load components, given working conditions and given wind speed.

[0097] L: The ultimate load reference value under given load components, given working conditions, and given wind speed.

[0098] f1: Regression function of the ultimate load and inflow angle sensitivity coefficient, independent variable: inflow angle.

[0099] f2: Regression function of ultimate load and wind shear index sensitivity coefficient, independent variable: wind shear index.

[0100] f3: Regression function of ultimate load, air density and turbulence intensity sensitivity coefficient, independent variables: air density and turbulence intensity.

[0101] f4: Regression function of the sensitivity coefficient between the ultimate load and the wind speed of the maximum turbulent wind once in 50 years. Independent variable: the wind speed of the maximum turbulent wind once in 50 years.

[0102] L”: Ultimate load for a given load component.

[0103] i: Working condition number.

[0104] j: wind speed number.

[0105] The extreme load prediction model is applicable to the three moments, three forces, and the load components of the resultant moment in the orthogonal coordinate system of the blade root, rotating hub, fixed hub, yaw bearing, tower top section, and tower bottom section. To verify the accuracy of the extreme load prediction model, several sets of environmental parameter combinations are randomly generated within the environmental parameter definition domain corresponding to the extreme load prediction model. The extreme loads are then calculated in detail using load simulation software to obtain verification values. The environmental parameters are then substituted into the extreme load prediction model to obtain predicted values. The verification values ​​are then compared with the predicted values ​​to verify the accuracy of the extreme load prediction model. If the accuracy is poor, the order of the sensitivity prediction model is adjusted and the regression analysis is repeated.

[0106] Regarding the process of predicting the site's ultimate load using the ultimate load prediction model: First, obtain the point environmental parameters and substitute them into the ultimate load prediction model to predict the ultimate values ​​of each load component. Repeat the "obtaining point environmental parameters" and "substituting them into the ultimate load prediction model to predict the ultimate values ​​of each load component" steps for each point to obtain the ultimate loads for all load channels at all points. For each load component, take the maximum ultimate load for that load component at all points to obtain the ultimate load envelope value for that load component at the site. Compare each site's ultimate load envelope value with the unit's design ultimate load to determine ultimate safety.

[0107] The wind turbine unit limit load prediction method of this embodiment establishes an explicit wind turbine limit load prediction model, enabling prediction of the limit loads at the target wind farm site without requiring iterative optimization calculations. This method avoids the time-consuming and labor-intensive detailed simulations and significantly conserves computing resources. Furthermore, it avoids the drawbacks of prior art techniques, where excessive redundancy and deviation in limit load estimations lead to poor accuracy and practicality. It accurately locates the unit locations with the highest limit loads, improving prediction efficiency while ensuring high accuracy in prediction results, thus demonstrating strong industry promotion value.

[0108] Example 2

[0109] See also Figure 3 This embodiment specifically provides a system for predicting the ultimate load of a wind turbine unit, including:

[0110] Environmental parameter collection module 51, used to collect environmental parameters of each unit in the target site;

[0111] The limit load acquisition module 52 is used to input the environmental parameters into the limit load prediction model to obtain the limit load corresponding to each unit;

[0112] The prediction result generating module 53 is used to generate the ultimate load safety prediction result corresponding to the target site according to the ultimate load.

[0113] This embodiment provides a prediction system for the ultimate load of a wind turbine group based on a prediction model to achieve prediction and assessment of the ultimate load of wind turbines at a specific site. Specifically, it includes two parts: establishing a prediction model for the ultimate load of wind turbines and assessing the ultimate load of wind turbines point by point at a specific site.

[0114] The process for assessing the ultimate load of wind turbines at a specific site, point by point, is as follows: For each wind turbine location, the environmental parameters of each location are first obtained. Using these environmental parameters as input, a limit load prediction model is used to determine the ultimate load under the influence of these environmental parameters. This ultimate load assessment is then performed point by point on the wind farm. For each load component, the maximum value of the ultimate load of that load component across all locations is taken as the ultimate load of that load component at the site. By comparing the site ultimate load with the unit's design ultimate load, the site's ultimate load safety is determined. Because the ultimate load prediction model is explicit, this method can save significant computational resources and time compared to detailed calculations, enabling rapid and accurate assessment of the ultimate load at each wind farm location.

[0115] As a preferred embodiment, the prediction system is further used to generate an ultimate load prediction model, including:

[0116] A reference load acquisition module 61 is used to obtain the reference limit load corresponding to each unit based on the design environmental parameters of each unit in the target site;

[0117] The sample acquisition module 62 is used to obtain the extreme load sensitivity samples and environmental parameter samples corresponding to each unit based on the design environmental parameters;

[0118] Sensitivity model training module 63, used to perform model training using extreme load sensitivity samples as functions and environmental parameter samples as independent variables, to obtain sensitivity models corresponding to different extreme loads;

[0119] The prediction model generation module 64 is used to obtain the ultimate load prediction model corresponding to the design environment parameters based on the benchmark ultimate load and the corresponding sensitivity model.

[0120] The wind turbine limit load prediction model development process includes four steps: generating a baseline limit load, generating a limit load sensitivity model, generating a limit load prediction model, and verifying the accuracy of the limit load prediction model. During the baseline limit load generation process, the baseline load acquisition module 61 first selects the unit requiring load assessment. Based on the unit's designed environmental conditions, it uses wind turbine load simulation software to perform detailed limit load calculations to obtain the baseline limit load.

[0121] The sample acquisition module 62 acquires the extreme load sensitivity samples and environmental parameter samples corresponding to each unit based on the design environmental parameters. Preferably, the sample acquisition module 62 includes:

[0122] The parameter interval determination unit 621 is used to determine the variation interval of the design environment parameter;

[0123] A discrete processing unit 622, configured to perform discrete processing on the environmental parameters within a variation interval to generate environmental parameter samples;

[0124] The extreme load sample acquisition unit 623 is used to acquire the extreme load sample corresponding to the environmental parameter sample through simulation calculation;

[0125] The sensitivity sample acquisition unit 624 is configured to divide the limit load sample by the reference limit load to acquire the limit load sensitivity sample.

[0126] Specifically, in the process of generating the extreme load sensitivity model, the unit design environmental conditions are taken as the center, the variation range of the environmental parameters is set, the environmental parameters are discretized within the range to form environmental parameter samples, the extreme load is calculated using wind turbine load simulation software, the extreme load samples under each set of environmental parameters are obtained, the extreme load samples are divided by the benchmark extreme load, the extreme load sensitivity samples are obtained, and the extreme load sensitivity prediction model is set.

[0127] The sensitivity model training module 63 uses environmental parameter samples as independent variables and extreme load sensitivity samples as functions to train the model and obtain sensitivity models for different extreme loads. The prediction model generation module 64 can obtain an extreme load prediction model by multiplying the extreme load sensitivity model by the reference extreme load.

[0128] As a preferred embodiment, the prediction system for the ultimate load of the wind turbine unit further includes:

[0129] A model verification parameter generation module 65 is used to generate environment parameters for model verification based on the design environment parameters;

[0130] A load verification value acquisition module 66 is used to obtain load verification values ​​corresponding to environmental parameters used for model verification;

[0131] A load prediction value acquisition module 67 is used to input the environmental parameters used for model verification into the extreme load prediction model to obtain corresponding load prediction values;

[0132] The prediction model adjustment module 68 is used to adjust the extreme load prediction model according to the load verification value and the load prediction value.

[0133] Specifically, within the environmental parameter definition domain of the extreme load prediction model, several sets of environmental parameters are randomly generated and used as input for detailed load calculations to obtain verified load values. Simultaneously, these sets of environmental parameters are substituted into the extreme load prediction model to obtain predicted load values. By comparing the two, the extreme load prediction model is verified and adjusted. The verified extreme load prediction model can be used to predict extreme loads of wind turbines.

[0134] As a preferred embodiment, the prediction result generation module 53 includes:

[0135] The limit load acquisition unit 531 is used to respectively acquire the maximum value of the limit loads of all units corresponding to each target load component as the limit load of the target load component;

[0136] The site safety assessment unit 532 is used to compare the ultimate load corresponding to each target load component with the design ultimate load of each unit, evaluate the safety of the ultimate load corresponding to each unit in the target site, and generate a prediction result.

[0137] Specifically, the process for assessing the ultimate load of wind turbines at a specific site, point by point, is as follows: For each wind turbine location, the environmental parameters of each location are first obtained. Using these environmental parameters as input, a mathematical model for the ultimate load is used to determine the ultimate load under the influence of these environmental parameters. This ultimate load assessment is then performed point by point for the wind farm. For each load component, the maximum value of the ultimate load of that load component across all locations is taken as the ultimate load of that load component at the site. By comparing the site ultimate load with the unit's design ultimate load, the site's ultimate load safety is determined. Because the ultimate load prediction model is explicit, this method can save significant computing resources and time compared to detailed calculations. Thus, through these two steps, a rapid and accurate assessment of the ultimate load at each point in the wind farm can be achieved.

[0138] The wind turbine unit limit load prediction system of this embodiment establishes an explicit wind turbine limit load prediction model, enabling prediction of the limit loads at the target wind farm site without requiring iterative optimization calculations. This avoids the time-consuming and labor-intensive detailed simulations and significantly conserves computing resources. Furthermore, it avoids the drawbacks of prior art techniques, where excessive redundancy and deviation in limit load estimations lead to poor accuracy and practicality. It accurately locates the unit locations with the highest limit loads, improving prediction efficiency while ensuring high accuracy in prediction results, thus possessing strong industry-wide value.

[0139] Example 3

[0140] See also Figure 4 As shown, this embodiment provides an electronic device 30, including a processor 31, a memory 32, and a computer program stored in the memory 32 and executable on the processor 31. When the processor 31 executes the program, the method for predicting the ultimate load of the wind turbine unit in Example 1 is implemented. Figure 4 The electronic device 30 shown is only an example and should not limit the functionality and scope of use of the embodiments of the present invention.

[0141] The electronic device 30 may be a general-purpose computing device, such as a server device. Components of the electronic device 30 may include, but are not limited to, the at least one processor 31, the at least one memory 32, and a bus 33 connecting various system components (including the memory 32 and the processor 31).

[0142] The bus 33 includes a data bus, an address bus, and a control bus.

[0143] The memory 32 may include a volatile memory, such as a random access memory (RAM) 321 and / or a cache memory 322 , and may further include a read-only memory (ROM) 323 .

[0144] The memory 32 may also include a program / utility 325 having a set (at least one) of program modules 324, such program modules 324 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0145] The processor 31 executes various functional applications and data processing by running the computer programs stored in the memory 32 , such as the method for predicting the ultimate load of the wind turbine unit in Embodiment 1 of the present invention.

[0146] The electronic device 30 can also communicate with one or more external devices 34 (e.g., a keyboard, pointing device, etc.). This communication can occur via an input / output (I / O) interface 35. Furthermore, the model-generating device 30 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 36. The network adapter 36 communicates with other modules of the model-generating device 30 via a bus 33. Other hardware and / or software modules can be used in conjunction with the model-generating device 30, including but not limited to microcode, device drivers, redundant processors, external disk drive arrays, RAID (RAID) systems, tape drives, and data backup storage systems.

[0147] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the present invention, the features and functions of two or more units / modules described above may be embodied in a single unit / module. Conversely, the features and functions of a single unit / module described above may be further divided and embodied by multiple units / modules.

[0148] Example 4

[0149] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method for predicting the ultimate load of the wind turbine unit in embodiment 1 is implemented.

[0150] The readable storage medium may include, but is not limited to, a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0151] In a possible implementation manner, the present invention can also be implemented in the form of a program product, which includes program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the method for predicting the ultimate load of the wind turbine unit in Example 1.

[0152] The program code for executing the present invention may be written in any combination of one or more programming languages, and may be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on the remote device.

[0153] Although specific embodiments of the present invention have been described above, those skilled in the art will appreciate that these are merely illustrative and that the scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, and such changes and modifications are intended to fall within the scope of the present invention.

Claims

1. A method for predicting the ultimate load of a wind turbine unit, characterized in that: include: Collect environmental parameters of each unit in the target site separately; Inputting the environmental parameters into a limit load prediction model to obtain the limit load corresponding to each unit; generating a prediction result corresponding to the target site according to the ultimate load; The prediction method further includes pre-generating the ultimate load prediction model, which includes: Obtaining a reference limit load corresponding to each unit based on design environmental parameters of each unit at the target site, wherein the design environmental parameters include: air density, turbulence intensity, vertical wind shear index, inflow angle, and 50-year maximum turbulent wind speed; Based on the designed environmental parameters, obtaining environmental parameter samples corresponding to each of the units; Obtain the ultimate load samples of the wind turbine through simulation, and for each combination of the environmental parameter samples, form an ultimate load sample matrix under each operating condition, each load component, and each wind speed condition; divide the ultimate load sample by the benchmark ultimate load to obtain the percentage of the ultimate load sample to the benchmark ultimate load, and use the percentage as the ultimate load sensitivity sample; analyze each operating condition, each load component, and each wind speed separately, using air density, turbulence intensity, inflow angle, wind shear index, and the wind speed of the maximum turbulent wind once in 50 years as independent variables, and the ultimate load sensitivity sample as the dependent variable, perform regression analysis to obtain a regression model, and use the regression model as the sensitivity model; The regression analysis includes four types of regression functions: the first type is the regression function of inflow angle and ultimate load; the second type is the regression function of wind shear index and ultimate load; the third type is the regression function of wind speed of the maximum turbulent wind once in 50 years and ultimate load; the fourth type is the regression function of air density, turbulence intensity and ultimate load; According to the benchmark limit load and the corresponding sensitivity model, the limit load prediction model corresponding to the design environment parameter is obtained; the limit load prediction model is obtained by analyzing each working condition, each load component, and each wind speed separately, and continuously multiplying the benchmark limit load and each sensitivity model.

2. The method for predicting the ultimate load of a wind turbine unit according to claim 1, wherein: After the step of obtaining the ultimate load prediction model corresponding to the design environment parameters, the following steps are included: Based on the design environment parameters, generating environment parameters for verification; Obtaining the load value corresponding to the environmental parameter used for verification; Inputting the environmental parameters for verification into the ultimate load prediction model to obtain a load prediction value; The extreme load prediction model is adjusted according to the load value and the load prediction value.

3. The method for predicting the ultimate load of a wind turbine unit according to claim 1, wherein: The step of obtaining the limit load sensitivity samples and environmental parameter samples corresponding to each of the units based on the design environmental parameters includes: determining a variation range of the design environment parameter; performing discrete processing on the environmental parameter within the variation interval to generate an environmental parameter sample; Obtaining an ultimate load sample corresponding to the environmental parameter sample through simulation calculation; The ultimate load sample is divided by the reference ultimate load to obtain the ultimate load sensitivity sample.

4. The method for predicting the ultimate load of a wind turbine unit according to claim 1, wherein: The step of generating a prediction result corresponding to the target site according to the extreme load includes: Respectively obtaining the maximum value of the limit loads of all units corresponding to each target load component as the limit load of the target load component; The ultimate load of each target load component is compared with the design ultimate load of each unit, and the safety of the ultimate load corresponding to each unit in the target site is evaluated to generate the prediction result.

5. A prediction system for the ultimate load of a wind turbine unit, characterized in that: include: Environmental parameter collection module, used to collect environmental parameters of each unit in the target site; A limit load acquisition module, configured to input the environmental parameters into a limit load prediction model to obtain the limit load corresponding to each unit; A prediction result generating module, configured to generate a prediction result corresponding to the target site according to the ultimate load; A reference load acquisition module, configured to acquire a reference limit load corresponding to each unit in the target site based on the design environmental parameters of each unit; The design environmental parameters include: air density, turbulence intensity, vertical wind shear index, inflow angle, and wind speed of the maximum turbulent wind with a return period of 50 years; A sample acquisition module, configured to acquire environmental parameter samples corresponding to each of the units based on the designed environmental parameters; a sensitivity model training module for obtaining the ultimate load samples of the wind turbine through simulation, and for each combination of the environmental parameter samples, forming an ultimate load sample matrix under each operating condition, each load component, and each wind speed condition; dividing the ultimate load sample by the benchmark ultimate load to obtain a percentage of the ultimate load sample to the benchmark ultimate load, and using the percentage as an ultimate load sensitivity sample; performing a separate analysis for each operating condition, each load component, and each wind speed, using air density, turbulence intensity, inflow angle, wind shear index, and the wind speed of the maximum turbulent wind once in 50 years as independent variables, and the ultimate load sensitivity sample as a dependent variable, performing a regression analysis to obtain a regression model, and using the regression model as a sensitivity model; The regression analysis includes four types of regression functions: the first type is the regression function of inflow angle and ultimate load; the second type is the regression function of wind shear index and ultimate load; the third type is the regression function of wind speed of the maximum turbulent wind once in 50 years and ultimate load; the fourth type is the regression function of air density, turbulence intensity and ultimate load; A prediction model generation module is used to obtain the limit load prediction model corresponding to the design environment parameter based on the benchmark limit load and the corresponding sensitivity model; the prediction model generation module is also used to analyze each working condition, each load component, and each wind speed separately, and continuously multiply the benchmark limit load and each sensitivity model to obtain the limit load prediction model.

6. The system for predicting the ultimate load of a wind turbine unit according to claim 5, wherein: The prediction system further includes: A parameter generation module for verification, used to generate environmental parameters for verification based on the design environment parameters; A verification value acquisition module, used to obtain a load verification value corresponding to the environmental parameter used for verification; A prediction value acquisition module, configured to input the environmental parameters for verification into the extreme load prediction model to obtain corresponding load prediction values; The prediction model adjustment module is used to adjust the extreme load prediction model according to the load verification value and the load prediction value.

7. The system for predicting the ultimate load of a wind turbine unit according to claim 5, wherein: The sample acquisition module includes: a parameter interval determining unit, configured to determine a variation interval of the design environment parameter; a discrete processing unit, configured to perform discrete processing on the environmental parameter within the variation interval to generate an environmental parameter sample; An extreme load sample acquisition unit, configured to acquire an extreme load sample corresponding to the environmental parameter sample through simulation calculation; The sensitivity sample acquisition unit is configured to divide the limit load sample by the reference limit load to acquire the limit load sensitivity sample.

8. The system for predicting the ultimate load of a wind turbine unit according to claim 5, wherein: The prediction result generation module includes: a limit load acquisition unit, configured to respectively acquire the maximum value of the limit loads of all units corresponding to each target load component as the limit load of the target load component; The wind turbine unit evaluation unit is used to compare the ultimate load corresponding to each target load component with the design ultimate load of each unit, evaluate the safety of the ultimate load corresponding to each unit in the target site, and generate the prediction result.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for predicting the ultimate load of the wind turbine unit according to any one of claims 1 to 4 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting the ultimate load of a wind turbine assembly according to any one of claims 1 to 4 is implemented.

Images