Mixed tower steel cable pre-tightening force real-time monitoring method and system

Abstract

The invention discloses a mixed tower steel cable pre-tightening force real-time monitoring method and system, and relates to the technical field of wind power generation. The method comprises the following steps: acquiring known dynamic characteristic data of the mixed tower steel cable under multiple working conditions and multiple structures; analyzing and screening key parameters which effectively represent the dynamic response, the system structure and the working condition load of the mixed tower steel cable system in the dynamic characteristic data by utilizing global sensitivity, and constructing a dynamic characteristic database according to the key parameters; constructing a structure-working condition-response relation neural network agent model by means of a neural network or a regression prediction method based on the dynamic characteristic database; and performing reverse calculation on real-time working condition data and motion response data sensed by the sensor by using the structure-working condition-response relation neural network agent model to obtain the pre-tightening force of the mixed tower steel cable. According to the invention, the pre-tightening force of the steel cable of the mixed tower can be accurately monitored in real time, and the pre-tightening force prediction of the mixed tower under the extreme working condition and the fault working condition is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a method and system for real-time monitoring of pre-tightening force of a hybrid tower steel cable. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Currently, in the study of the stress on the mixed tower steel cables, due to the large number of steel cables, the close arrangement of each bundle of steel cables, and the limited space, traditional wire rope tension measurement methods and equipment are unable to measure the preload of the steel cables inside the cable bundle. The development of a real-time online monitoring system for preload is of great engineering significance. First, the real-time online monitoring system for preload can promptly detect steel cable failures and take maintenance measures in advance to ensure the effective preload of the mixed tower steel cables and reduce the probability of accidents. Secondly, the real-time monitoring data of preload strongly supports the calculation of steel cable fatigue life and realizes the monitoring maintenance of faults such as insufficient steel cable preload and steel cable fatigue failure. Finally, the real-time monitoring system for steel cable preload can collect preload data under multiple working conditions for a long time, clarify the relationship between the preload of the mixed tower steel cables and the working conditions and the motion response of the mixed tower, and provide a basis for the subsequent design of the mixed tower steel cable system.

[0004] Existing technologies often calculate the pre-tension of the mixed tower steel cables by the hydraulic pressure of the tension cables. However, after the hydraulic tensioning system is tensioned, the hydraulic station no longer provides hydraulic pressure during the natural contraction stage of the steel cables, making it impossible to monitor the tension of the steel cables after contraction. In addition, since the wind load has large random fluctuations and average values ​​when the wind turbine is operating normally, it is necessary to monitor the pre-tension of the steel cables in real time. Although methods such as optical fiber and magnetic leakage can achieve real-time monitoring, sensors need to be placed on each steel cable, resulting in high monitoring costs and poor applicability. Therefore, there is currently a lack of equipment and methods for effectively measuring the pre-tension of steel cables, which plagues wind farm operations and maintenance.

[0005] In summary, it is necessary to develop an economical and reliable real-time monitoring method for the preload of hybrid tower cables to reduce the cost of real-time monitoring and prevent the risk of preload failure. Summary of the Invention

[0006] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a real-time monitoring method and system for the preload of the mixed tower steel cable, which can monitor the preload of the mixed tower steel cable in real time and accurately, and realize the preload prediction of the mixed tower under extreme working conditions and fault conditions.

[0007] In order to achieve the above object, the present invention is implemented through the following technical solutions: A first aspect of the present invention provides a method for real-time monitoring of preload force of a hybrid tower steel cable, comprising the following steps: Obtaining the known dynamic characteristic data of the hybrid tower cable under multiple working conditions and multiple structures, wherein the dynamic characteristic data includes structural data, working condition data, and corresponding motion response data; Global sensitivity analysis is used to screen key parameters in the dynamic characteristic data that effectively characterize the dynamic response of the hybrid tower cable system, the system structure, and the working load. A dynamic characteristic database is constructed based on these key parameters. Based on the dynamic characteristics database, a neural network proxy model of the structure-operating condition-response relationship is constructed with the help of neural network or regression prediction methods; The structure-operating condition-response relationship neural network proxy model is used to reversely calculate the real-time operating condition data and motion response data perceived by the sensor to obtain the preload force of the hybrid tower steel cable.

[0008] Furthermore, the specific steps for obtaining the dynamic characteristic data of the known hybrid tower steel cable under multiple working conditions and multiple structures are as follows: A dynamic model of the hybrid tower cable system is constructed. Based on the dynamic model of the hybrid tower cable system, hybrid towers with different structural sizes are simulated under different working conditions to determine the motion response of the hybrid tower under wind load.

[0009] Furthermore, the multi-structure includes the height of the concrete tower, the number of concrete sections, the radius of the concrete tower, the wall thickness of the concrete tower, the radius of the steel cable, the length of the steel cable, the arrangement of the steel cable, and the variable parameters of the pre-tensioning force. The multi-working conditions include the normal working condition of the steel cable pre-tensioning force, the working condition of insufficient pre-tensioning force, the working condition of excessive pre-tensioning force, the unbalanced pre-tensioning force and the working condition of the steel cable breakage. The structural data include the height of the concrete tower, the number of concrete sections, the radius of the concrete tower, the wall thickness of the concrete tower and the arrangement position, quantity, diameter, axial stiffness, damping, anchor point position and pre-tensioning force of the steel cable. The working condition data include wind speed, wind direction, wind turbine power, wind rotor speed, blade airfoil, blade chord length and blade pitch angle. The motion response parameters include the dynamic inclination angle, angular velocity and angular acceleration of the tower top.

[0010] Furthermore, the specific steps to determine the motion response of the concrete tower under wind load are as follows: Based on the Markov chain analysis of local wind speed data, the wind speed data and time data are combined, i.e. one-to-one correspondence is established to establish a speed-time state; Traverse the local wind speed condition data, integrate all states from the current moment to the next second with a collection range of 1m / s, and count the frequency of different states in the next second; Calculate the state transition probability from the current second to the next second; Calculate the speed-time state transition probability matrix from the current second to the next second at different times; The acceptance-rejection sampling method is used to perform Monte Carlo simulation on the wind speed time state transition probability matrix to obtain the equivalent wind speed curve. Multiple equivalent wind speed curves are generated through multiple calculations to ensure that their mean and standard deviation are the same as the local wind speed working condition data, and multi-condition simulation wind speed curves are obtained.

[0011] Furthermore, global sensitivity analysis is used to screen key parameters in the dynamic characteristic data that effectively characterize the dynamic response of the hybrid tower cable system and the system structure and working load. The specific steps for constructing a dynamic characteristic database based on the key parameters are as follows: Evaluate the contribution of structural data and operating condition data to the dynamic response of the hybrid tower cable system, including the contribution of each structural data and operating condition data to the dynamic response of the hybrid tower cable system, as well as the contribution of the interaction between each structural data and operating condition data to the dynamic response of the hybrid tower cable system, and screen out representative structures, operating conditions and corresponding motion response data; The response experimental data are fused based on the variance weight fusion method, and the optimal weight of data fusion is obtained through the variance of data signals. A dynamic characteristics database of the hybrid tower cable system is established based on the fused response data.

[0012] Furthermore, based on the dynamic characteristics database of the mixed tower steel cable system, with structural data and working condition data as input, high-dimensional features are extracted through neural networks, motion response data are output, the structure-working condition-response relationship is clarified, and with the help of neural networks and regression prediction methods, a neural network proxy model of the structure-working condition-response relationship is obtained.

[0013] Furthermore, the reverse inference capability of the structure-working condition-response relationship neural network agent model is developed. The motion response data and working condition data are used as input, high-dimensional features are extracted through the neural network, and structural data is output to realize the monitoring of the cable preload.

[0014] A second aspect of the present invention provides a real-time monitoring system for pre-tensioning force of a hybrid tower steel cable, comprising: A data acquisition module is configured to acquire known dynamic characteristic data of a hybrid tower cable under multiple working conditions and multiple structures, wherein the dynamic characteristic data includes structural data, working condition data, and corresponding motion response data; A data screening module is configured to use global sensitivity analysis to screen key parameters in the dynamic characteristic data that effectively characterize the dynamic response of the hybrid tower cable system and the system structure and working load, and to construct a dynamic characteristic database based on the key parameters; A model building module is configured to build a neural network proxy model of structure-operating condition-response relationship based on a dynamic characteristics database using a neural network or regression prediction method; The preload monitoring module is configured to use the structure-operating condition-response relationship neural network agent model to reversely infer the real-time operating condition data and motion response data sensed by the sensor to obtain the preload of the hybrid tower steel cable.

[0015] The third aspect of the present invention provides a computer-readable storage medium storing a computer program, which is suitable for being loaded by a processor and executing the steps of the real-time monitoring method for pre-tensioning force of a hybrid tower steel cable as described in the first aspect of the present invention.

[0016] A fourth aspect of the present invention provides a computer device, comprising: a processor adapted to execute a computer program; A computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the method for real-time monitoring of the pre-tensioning force of the hybrid tower steel cable as described in the first aspect of the present invention is implemented.

[0017] One or more of the above technical solutions have the following beneficial effects: The present invention discloses a real-time monitoring method and system for the pre-tensioning force of a mixed-tower steel cable. The dynamic response characteristics of the mixed-tower steel cable system are closely related to the structural characteristics of the system and the system operating load, and the pre-tensioning force of the steel cable affects the stiffness characteristics of the system. Therefore, when the pre-tensioning force of the steel cable is insufficient, excessive or uneven, the dynamic response characteristics of the mixed-tower steel cable system are different under the same operating conditions. Structural characteristics such as the pre-tensioning force of the steel cable can also be monitored based on the system operating load and dynamic response characteristics. In view of the fact that the wind turbine cabin is currently equipped with a wind measurement system to capture the system operating information, and the installed inclination, acceleration and other sensors capture the system macro-dynamic response characteristics, the present invention intends to realize the calculation of the pre-tensioning force of the steel cable based on the operating data of the existing wind turbine and the macro-dynamic response characteristics of the wind turbine. Based on the neural network proxy model, the structure-operating condition-response relationship of the mixed-tower steel cable system is constructed. Based on the reverse calculation capability of the model, the operating condition and response information obtained by the sensor are input into the structure-operating condition-response relationship neural network proxy model to realize the monitoring of the pre-tensioning force of the steel cable.

[0018] This invention achieves real-time or near-real-time monitoring. Once the neural network model is trained, the monitoring process is computationally expedited, which is crucial for the safe monitoring of hybrid tower cable systems. Existing methods for directly and accurately measuring cable preload (such as hydraulic jack and frequency methods) are often difficult, time-consuming, and costly. This method provides a new approach for real-time, continuous monitoring based on readily available indirect signals.

[0019] This invention also reduces reliance on high-precision, high-cost specialized preload force measurement equipment, instead utilizing more economical and easily installed sensors (such as angle and acceleration sensors) to acquire signals. Indirect measurement sensors are generally simpler to install than direct measurement equipment, with less structural disruption and lower maintenance costs. This allows for automated, continuous preload force monitoring, significantly reducing the frequency of costly and high-risk manual testing.

[0020] This invention can promptly detect preload anomalies, providing early warning of potential tower failures and avoiding safety incidents. Based on accurate preload status information, it enables monitoring maintenance and precise tensioning adjustments when necessary, avoiding over- or under-maintenance, thereby optimizing maintenance costs and extending the service life of the structure.

[0021] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0023] Figure 1 This is a flow chart of a method for real-time monitoring of pre-tensioning force of a hybrid tower steel cable in a first embodiment of the present invention; Figure 2 Schematic diagram of the neural network proxy model for the structure-operating condition-response relationship of the hybrid tower cable system in Example 1 of the present invention; Figure 3 Schematic diagram of the cable preload monitoring process in the first embodiment of the present invention. DETAILED DESCRIPTION

[0024] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0025] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof; The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0026] Example 1: In view of the problem that the number of mixed tower cables is large and the layout space is tight, and there is a risk of fatigue failure, it is difficult for traditional equipment to effectively monitor their preload. The first embodiment of the present invention provides a real-time monitoring method for the preload of mixed tower cables, which realizes the preload monitoring of the mixed tower during the tensioning process and under extreme and fault conditions. Figure 1 As shown, first, dynamic simulations of multiple working conditions and structures were conducted based on a parametric model of the hybrid tower cable system. Next, sensitivity analysis was conducted, integrating response experimental data to establish a dynamic characteristics database. A neural network proxy model of the structure-working condition-response relationship was built and trained. The sensor-perceived working condition and motion response data were fed into the proxy model, and the proxy model's reverse inference capabilities were leveraged to monitor cable preload.

[0027] The specific steps include: Step 1: Obtain the known dynamic characteristic data of the hybrid tower cable under multiple working conditions and multiple structures, wherein the dynamic characteristic data includes structural data, working condition data and corresponding motion response data.

[0028] This example constructs a dynamic model of a hybrid tower cable system. Based on this model, hybrid towers of different structural sizes are simulated under different working conditions to determine the motion response of the hybrid tower under wind load. The specific steps are as follows: Step 1.1: Construct a dynamic model of the hybrid tower cable system and perform dynamic characteristic analysis to determine the motion response of the hybrid tower under wind load.

[0029] In one specific implementation, a dynamics analysis is performed based on a dynamics model of the hybrid tower cable system to clarify the kinematic response of the hybrid tower under wind loads. During operation, the hybrid tower is primarily subject to wind loads, which in this embodiment are considered to be the effects of tower wind pressure loads and blade aerodynamic loads.

[0030] The calculation formula for tower wind pressure load is as follows: , .

[0031] Where, is the tower wind pressure load, is the basic wind pressure, is the air density, is the reference wind speed, is the wind vibration coefficient, is the wind load shape coefficient, is the wind pressure height variation coefficient.

[0032] The aerodynamic load of the fan is as follows: .

[0033] Where T is the aerodynamic load of the fan, is the air density, is the wind speed at the rotor hub height, is the axial induction factor, is the swept area of ​​the wind wheel.

[0034] During the impeller startup process, the impeller speed gradually increases, and the aerodynamic load also changes from zero to the maximum. In addition, during the actual operation process, the wind speed also changes randomly, and the wind pressure load on the mixed tower is also changing every moment. In order to make the constructed wind speed equation more accurately reflect the actual wind load on the mixed tower, this embodiment uses local actual wind speed data based on the Markov chain to construct a transient wind speed condition. In the Markov chain, the current state is only related to the state of the previous moment, and has nothing to do with the past state such as the previous moment. Specifically, when the current state is given, the probability distribution of the future state depends only on the current state and is not affected by the past state. This property is called the Markov property. In this embodiment, the wind load condition during the operation of the mixed tower is regarded as a random process with Markov properties, and the wind speed that constitutes the wind load condition is used as the basic unit of construction. The specific steps are as follows: Step 1.1.1: Analyze the local wind speed condition data based on the Markov chain, combine the wind speed condition data with the time data, that is, one-to-one correspondence, and establish a speed-time state.

[0035] Specifically, the current speed-time state is expressed as Indicates that The possible speed-time states in the next second can be represented by the set Q: .

[0036] Where, for The speed-time state of the next second, where represents the speed-time state index in the set Q, represents the total number of speed-time states in the set Q.

[0037] Step 1.1.2: Traverse the local wind speed condition data, integrate all states of the next second from the current moment with a collection range of 1 m / s, and count the frequency of different states in the next second.

[0038] Specifically, the frequency set of different states appearing in the next second is represented by B: .

[0039] Where, represent The frequency of .

[0040] Step 1.1.3: Calculate the state transition probability from the current second to the next second.

[0041] Specifically, the calculation is from arrive The state transition probability is: .

[0042] Step 1.1.4: Calculate the speed-time state transition probability matrix from the current second to the next second at different times.

[0043] Specifically, the speed-time state transition probability matrix at different moments is described by the following formula: .

[0044] Where P is the speed-time state transition probability matrix. When the state transitions from the current moment b to the next moment j, it can be described by the following formula: .

[0045] Where S is the state space composed of local wind speed condition data, Indicates the previous state of the wind speed condition, Indicates the latter state of the wind speed condition.

[0046] Step 1.1.5: Use the acceptance-rejection sampling method to perform Monte Carlo simulation on the wind speed time state transition probability matrix to obtain the equivalent wind speed curve.

[0047] Because the probabilities in the wind speed-time state transition probability matrix are not uniformly distributed, using uniform distribution sampling for Monte Carlo simulations can result in significant errors, affecting the accuracy and precision of the model. Therefore, the Monte Carlo simulations are performed using the acceptance-rejection sampling method. Programmatically, the acceptance-rejection sampling method within the Monte Carlo method is used to construct operating condition results, which are then stored in an alternative chain. Finally, an error evaluation function is used to calculate the error between each operating condition curve and the actual operating condition, and the optimal operating condition with the smallest error is output.

[0048] The specific steps are as follows: 1. Sampling from distribution G, obtain a sample Y.

[0049] 2. Sample from a uniform distribution in [0,1] to obtain a sample U.

[0050] 3. Judge, if , then accept this Y as the recorded sampling value, otherwise reject this sampling value and discard it to resample.

[0051] Among them, G is uniform distribution; c is a constant value, is the probability density function of the target distribution, is the probability density function of the proposed distribution. For any x, ,In order to improve the efficiency of sampling, c should take a smaller value when the above conditions are met.

[0052] Step 1.1.6: Generate multiple equivalent wind speed curves through multiple calculations, ensuring that their mean and standard deviation are consistent with the local wind speed condition data, and obtain multi-condition simulation wind speed curves.

[0053] Specifically, the time is set to 100 s, and the wind speed results of 100 s are sampled according to the above steps to achieve the equivalence of local wind speed operating data within one year.

[0054] Step 1.2: Based on the dynamic model of the hybrid tower cable system, simulate hybrid towers with different structural sizes under different working conditions.

[0055] In a specific embodiment, the multiple structures include the height of the concrete tower, the number of concrete sections, the radius of the concrete tower, the wall thickness of the concrete tower, the radius of the steel cable, the length of the steel cable, the arrangement of the steel cable, and the variable parameters of the pre-tensioning force. The multiple working conditions include the normal working condition of the steel cable pre-tensioning force, the working condition of insufficient pre-tensioning force, the working condition of excessive pre-tensioning force, the unbalanced pre-tensioning force and the working condition of the steel cable breakage. The structural data include the height of the concrete tower, the number of concrete sections, the radius of the concrete tower and the wall thickness of the concrete tower, and the arrangement position, quantity, diameter, axial stiffness, damping, anchor point position and pre-tensioning force of the steel cable. The working condition data include wind speed, wind direction, wind turbine power, wind rotor speed, blade airfoil, blade chord length, blade pitch angle, and the motion response parameters include the dynamic inclination angle, angular velocity and angular acceleration of the tower top.

[0056] After the simulation, the motion response parameters such as the dynamic inclination angle, angular velocity and angular acceleration of the hybrid tower top are recorded as the simulation results.

[0057] Step 2: Use global sensitivity analysis to screen the key parameters in the dynamic characteristics data that effectively characterize the dynamic response of the hybrid tower cable system and the system structure and working load, and build a dynamic characteristics database based on the key parameters.

[0058] Step 2.1: Evaluate the contribution rate of structural data and operating condition data to the dynamic response of the hybrid tower steel cable system, including the contribution rate of each structural data and operating condition data to the dynamic response of the hybrid tower steel cable system, as well as the contribution rate of the interaction between each structural data and operating condition data to the dynamic response of the hybrid tower steel cable system, and screen out representative structures, operating conditions and corresponding motion response data.

[0059] This example screens key variables that significantly influence the dynamic response of the hybrid tower cable system (such as the dynamic inclination angle of the tower top, angular velocity, angular acceleration, etc.) from numerous input parameters (hybrid tower geometry, cable size and arrangement, external loads, cable preload, etc.), and evaluates the contribution of each parameter individually and through interaction to the output response.

[0060] Specifically, one of the many input parameters is first selected as a single variable and the kinetic response is observed. The more pronounced the result, the greater the contribution. Then, these input parameters are combined as combined variables to observe the degree of influence of each combined variable on the kinetic response. Again, the more pronounced the result, the greater the contribution. The representative parameters with the largest contribution are selected and used as input parameters for the subsequent proxy model.

[0061] Based on the hybrid tower cable system model with different geometric sizes, the dynamic inclination angle, angular velocity, angular acceleration and other motion response parameters of the hybrid tower top are recorded with a step length of 5 seconds.

[0062] Step 2.2: Fuse the response experimental data using a variance-weighted fusion method, and determine the optimal weight for data fusion based on the data signal variance. This step increases the amount of sample data and the accuracy of the trained monitoring model, ensuring that the fused data retains the obvious fault characteristics of the simulation data and the true environmental impact characteristics of the experimental data to the greatest extent possible. The response experimental data is obtained through response experiments, which are well-known techniques in the field and will not be further described here.

[0063] The variance weight fusion method is as follows: is the experimental data in state i; is the variance corresponding to the experimental data; is the simulation data in state i; is the variance corresponding to the simulated pressure data.

[0064] The formula for weighted fusion of experimental and simulation data is: .

[0065] Where: is the fusion data of state i; is the assignment weight obtained in state i.

[0066] To alleviate the best condition in i , so that the fused data Contains the most effective features of the two data, so that the fused data Variance Minimize to find the optimal weight in reverse , since the two sets of signals are independent of each other, we can calculate the variance on both sides of the above equation and simplify it to get: .

[0067] right Find the derivative and let =0, and the solution is: .

[0068] Substituting the above formula into the weighted fusion formula of experimental and simulation data can obtain the optimal estimate. At this time, the fusion signal The variance of is minimized, so that the fused data contains the most effective features of the two data, thus achieving data fusion.

[0069] Step 2.3: Based on the fused response data, a dynamic characteristics database of the hybrid tower cable system is established to provide data support for establishing the structure-operating condition-response relationship of the hybrid tower cable system.

[0070] Step 3: Based on the dynamic characteristics database, a neural network proxy model of the structure-operating condition-response relationship is constructed with the help of neural network or regression prediction method.

[0071] In a specific implementation, this embodiment is based on the dynamic characteristics database of the mixed tower steel cable system, takes structural data and operating condition data as input, extracts high-dimensional features through a neural network, outputs motion response data, clarifies the structure-operating condition-response relationship, and uses neural networks and regression prediction methods to obtain an agent model based on the structure-operating condition-response relationship.

[0072] Specifically, there is a highly nonlinear relationship between the input variables and output variables of the mixed tower structure-working condition-response relationship neural network proxy model. Thanks to the nonlinear activation function in the neural network unit, the neural network algorithm performs well in processing the nonlinear relationship between variables. In this embodiment, a neural network algorithm is selected to establish a mapping relationship between the input variables and output variables of the mixed tower structure-working condition-response relationship. With structural data such as steel cable preload and time-varying external working condition load data as input, high-dimensional features are extracted through the neural network, and the dynamic characteristic data of the mixed tower is output. The model realizes accurate monitoring of system response under complex working conditions at a speed a thousand times faster than traditional simulation, provides real-time warning for the safety status of steel cables, and significantly improves the monitoring efficiency of mixed tower structures.

[0073] Specifically, relying on the established database of dynamic characteristics of hybrid tower cable systems, 200,000 sets of data were randomly generated for training the neural network, and 20,000 sets of data were generated for verification. Both the training set and the verification set data were fused by simulation and experimental data.

[0074] The structure-condition-response neural network proxy model consists of a linear weighting function and a nonlinear activation function. The linear weighting function linearly transforms inputs from different sources, while the nonlinear activation function performs nonlinear processing on the transformed values. The choice of activation function significantly impacts network performance. In this example, the ReLU function, commonly used in regression tasks, is used.

[0075] Neural network units are connected to each other according to certain rules to form a deep neural network. Figure 2 As shown in the figure, the forward relational neural network has two hidden layers, where hidden layer 1 contains 20 operation nodes and hidden layer 2 contains 20 operation nodes. V , blade pitch angle β , mixing tower radius r , Number of steel cables a , preload F The working conditions and structural data are input into the input layer as input variables, and the dynamic inclination angle of the tower top is θ , angular velocity ω and angular acceleration α The isomotion response parameters are output by the output layer as output variables, and the nonlinear relationship between the input variables and the output variables can be gradually approximated by the deep neural network.

[0076] Normalized mean square error is an indicator used to quantify the accuracy of model monitoring. The calculation method is: .

[0077] Where, and are the monitoring value and the true response value of the forward relational neural network, m is the total number of iterations, and o is the current step number.

[0078] In order to avoid overfitting of the model, the learning rate is set to The Adam optimizer was selected as the optimizer, and the number of training rounds was set to 1000. The mean square error after 1000 rounds was used to determine whether the forward relationship neural network training process had converged.

[0079] like Figure 2 As shown in the figure, similar to the establishment process of the forward relationship neural network, the reverse monitoring neural network also uses a series double hidden layer structure to filter and process the input information, where the hidden layer 1 contains 20 operation nodes and the hidden layer 2 contains 40 operation nodes. V , blade pitch angle β , dynamic inclination angle of the top of the mixing tower θ , angular velocity ω and angular acceleration αThe working conditions and motion response parameters are input into the input layer as input variables, and the preload F , mixing tower radius r , Number of steel cables a The equal structure data is output by the output layer as the output variable. Reverse monitoring also randomly generates 220,000 initial data sets for learning, of which 200,000 sets of data are used as training sets and the other 20,000 sets of data are used as validation sets.

[0080] Step 4: Use the structure-operating condition-response relationship neural network proxy model to reversely calculate the real-time operating condition data and motion response data perceived by the sensor to obtain the preload force of the hybrid tower steel cable.

[0081] In a specific implementation, this embodiment develops the reverse inference capability of the structure-working condition-response relationship neural network agent model, takes motion response data and working condition data as input, extracts high-dimensional features through the neural network, outputs structural data, and realizes the monitoring of steel cable preload.

[0082] Specifically, cabin sensors sense operating condition information, while inclination and acceleration sensors sense response information, which are input into a trained proxy model. The structure-operating condition-response relationship neural network proxy model uses a neural network algorithm to establish a nonlinear mapping relationship between the input and output variables of the hybrid tower structure-operating condition-response relationship. This structure-operating condition-response relationship neural network proxy model comprises a forward relationship neural network and a reverse monitoring neural network. The forward relationship neural network first establishes a forward relationship between structure, operating condition, and response, thereby determining the expected response of the hybrid tower under certain structures and operating conditions. The reverse monitoring neural network then performs inverse inversion based on this forward relationship between structure, operating condition, and response. Specifically, the forward relationship neural network takes structural data such as cable preload and time-varying external operating loads as input, extracts high-dimensional features through the neural network, and outputs the hybrid tower's motion response parameters. The reverse monitoring neural network takes the hybrid tower's motion response parameters and time-varying external operating loads as input, extracts high-dimensional features through the neural network, and outputs structural data such as cable preload. Based on the real-time working condition data and motion response data perceived by the sensor, the structural data such as preload force are reversely calculated to achieve real-time monitoring of the preload force of the steel cable. Figure 3As shown in the figure, after comprehensively considering the wind turbine's operating status and external environmental conditions, operating information such as wind speed, wind direction, turbine power, rotational speed, rotor speed, blade airfoil, blade chord length, and blade pitch angle are monitored and recorded. Response information such as the dynamic tilt angle, angular velocity, and angular acceleration at the tower top, as sensed by tilt and acceleration sensors, is then constructed to represent the structure-operating-condition-response relationship of the hybrid-tower cable system. The proxy model's reverse inference capabilities are then developed to enable real-time monitoring of structural parameters such as cable preload. Multiple sensor fusion acquisition ensures information integrity and data quality. Since the blade airfoil and chord length of a given wind turbine model are fixed, while other parameters change dynamically, nacelle sensors are used to sense operating information such as turbine power, rotational speed, rotor speed, and blade pitch angle. Inclination sensors sense the dynamic tilt angle at the tower top, and acceleration sensors sense motion response information such as angular velocity and angular acceleration at the tower top. These operating and response data are then transmitted via optical fiber to a proxy model of the structure-operating-condition-response relationship in a computer.

[0083] The response and working condition data sensed by the sensor are input into the input layer as input variables, and structural data such as the preload force of the mixed tower cables are output. The monitoring results are compared with the results under working conditions such as normal, insufficient, excessive, uneven, and broken preload force to determine the current health status of the mixed tower. The monitoring results are displayed on the existing SCADA system user interface, thereby realizing accurate monitoring of the cable preload force.

[0084] Example 2: A second embodiment of the present invention provides a real-time monitoring system for pre-tensioning force of a hybrid tower steel cable, comprising: A data acquisition module is configured to acquire known dynamic characteristic data of a hybrid tower cable under multiple working conditions and multiple structures, wherein the dynamic characteristic data includes structural data, working condition data, and corresponding motion response data; A data screening module is configured to use global sensitivity analysis to screen key parameters in the dynamic characteristic data that effectively characterize the dynamic response of the hybrid tower cable system and the system structure and working load, and to construct a dynamic characteristic database based on the key parameters; a model building module configured to build a neural network proxy model of structure-operating condition-response relationship based on dynamic characteristics; The preload monitoring module is configured to use the structure-operating condition-response relationship neural network agent model to reversely infer the real-time operating condition data and motion response data sensed by the sensor to obtain the preload of the hybrid tower steel cable.

[0085] Example 3: Embodiment 3 of the present invention provides a computer-readable storage medium storing a computer program, which is suitable for being loaded by a processor and executing the steps of the real-time monitoring method for pre-tensioning force of a mixed tower steel cable as described in embodiment 1 of the present invention.

[0086] Example 4: A fourth embodiment of the present invention provides a computer device, comprising: a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium. When the computer program is executed by the processor, the steps in the method for real-time monitoring of pre-tensioning force of a hybrid tower steel cable as described in the first embodiment of the present invention are implemented.

[0087] The steps involved in the above embodiments 2, 3 and 4 correspond to those in the method embodiment 1. For the specific implementation methods, please refer to the relevant description part of the embodiment 1.

[0088] Those skilled in the art will appreciate that the units and algorithmic steps of each example described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technical personnel may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application. In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted via a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data processing device such as a server or data center that integrates one or more available media. Available media can include magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)). The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any technical object of a person skilled in the art that can be easily conceived of within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A real-time monitoring method for pre-tensioning force of a mixed tower steel cable, characterized in that: The following steps are involved: Obtaining the known dynamic characteristic data of the hybrid tower cable under multiple working conditions and multiple structures, wherein the dynamic characteristic data includes structural data, working condition data, and corresponding motion response data; Global sensitivity analysis is used to screen key parameters in the dynamic characteristic data that effectively characterize the dynamic response of the hybrid tower cable system, the system structure, and the working load. A dynamic characteristic database is constructed based on these key parameters. Based on the dynamic characteristics database, a neural network proxy model of the structure-operating condition-response relationship is constructed with the help of neural network or regression prediction methods; The structure-operating condition-response relationship neural network agent model is used to reversely predict the preload of the hybrid tower steel cable based on the real-time operating condition data and motion response data sensed by the sensors.

2. The method for real-time monitoring of preload force of a hybrid tower steel cable according to claim 1, characterized in that: The specific steps to obtain the known dynamic characteristics database of hybrid tower cables under multiple working conditions and multiple structures are as follows: A dynamic model of the hybrid tower steel cable system is constructed. Based on the dynamic model of the hybrid tower steel cable system, hybrid towers with different structural sizes are simulated under different working conditions to determine the motion response of the hybrid tower under wind load. In addition, the measured data of the existing hybrid tower steel cable system is collected, and the simulation data and measured data are integrated to create a dynamic characteristics database of the hybrid tower steel cable system under multiple working conditions and multiple structures.

3. The method for real-time monitoring of pre-tensioning force of a hybrid tower steel cable according to claim 2, characterized in that: Multiple structures include concrete tower height, number of concrete sections, concrete tower radius, concrete tower wall thickness, steel cable radius, steel cable length, steel cable arrangement method, and preload variable parameters. Multiple working conditions include normal steel cable preload working conditions, insufficient preload working conditions, excessive preload working conditions, unbalanced preload and steel cable break working conditions. Structural data include concrete tower height, number of concrete sections, concrete tower radius and concrete tower wall thickness, as well as steel cable arrangement position, quantity, diameter, axial stiffness, damping, anchor point position and preload. Working condition data include wind speed, wind direction, wind turbine power, rotor speed, blade airfoil, blade chord length, blade pitch angle, and motion response parameters include inclination angle, angular velocity, and angular acceleration.

4. The method for real-time monitoring of pre-tensioning force of a hybrid tower steel cable according to claim 2, characterized in that: The specific steps for determining the wind load effects of multiple working conditions on a concrete tower are as follows: Based on the Markov chain analysis of local wind speed data, the wind speed data and time data are combined, i.e. one-to-one correspondence is established to establish a speed-time state; Traverse the local wind speed condition data, integrate all states from the current moment to the next second with a collection range of 1m / s, and count the frequency of different states in the next second; Calculate the state transition probability from the current second to the next second; Calculate the speed-time state transition probability matrix from the current second to the next second at different times; The acceptance-rejection sampling method is used to perform Monte Carlo simulation on the wind speed time state transition probability matrix to obtain the equivalent wind speed curve. Multiple equivalent wind speed curves are generated through multiple calculations to ensure that their mean and standard deviation are the same as the local wind speed working condition data, and multi-condition simulation wind speed curves are obtained.

5. The method for real-time monitoring of pre-tensioning force of a hybrid tower steel cable according to claim 1, characterized in that: Global sensitivity analysis is used to screen key parameters in the dynamic characteristics data that effectively characterize the dynamic response of the hybrid tower cable system, the system structure, and the working load. The specific steps for constructing a dynamic characteristics database based on the key parameters are as follows: Evaluate the contribution of structural data and operating condition data to the dynamic response of the hybrid tower cable system, including the contribution of each structural data and operating condition data to the dynamic response of the hybrid tower cable system, as well as the contribution of the interaction between each structural data and operating condition data to the dynamic response of the hybrid tower cable system, and screen out representative structures, operating conditions and corresponding motion response data; The response experimental data are fused based on the variance weight fusion method, and the optimal weight of data fusion is obtained through the variance of data signals. A dynamic characteristics database of the hybrid tower cable system is established based on the fused response data.

6. The method for real-time monitoring of pre-tensioning force of a hybrid tower steel cable according to claim 5, characterized in that: Based on the dynamic characteristics database of the hybrid tower steel cable system, the structural data and operating condition data of the hybrid tower steel cable system are used as input, high-dimensional features are extracted through a neural network, and motion response data are output. The structure-operating condition-response relationship of the hybrid tower steel cable system is clarified. With the help of neural network and regression prediction methods, a neural network proxy model of the structure-operating condition-response relationship is obtained.

7. The method for real-time monitoring of pre-tensioning force of a hybrid tower steel cable according to claim 1, characterized in that: Develop the reverse inference capability of the structure-operating condition-response relationship neural network agent model, use the motion response data and operating condition data of the mixed tower cable system as input, extract high-dimensional features through the neural network, output structural characteristic data, predict the cable preload, and realize the monitoring of the cable preload.

8. A real-time monitoring system for pre-tensioning force of a mixed tower steel cable, characterized in that: include: A data acquisition module is configured to acquire known dynamic characteristic data of a hybrid tower cable under multiple working conditions and multiple structures, wherein the dynamic characteristic data includes structural data, working condition data, and corresponding motion response data; A data screening module is configured to use global sensitivity analysis to screen key parameters in the dynamic characteristic data that effectively characterize the dynamic response of the hybrid tower cable system and the system structure and working load, and to construct a dynamic characteristic database based on the key parameters; A model building module is configured to build a neural network proxy model of structure-operating condition-response relationship based on a dynamic characteristics database using a neural network or regression prediction method; The preload monitoring module is configured to use the structure-operating condition-response relationship neural network agent model to reversely infer the real-time operating condition data and motion response data sensed by the sensor to obtain the preload of the hybrid tower steel cable.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing the real-time monitoring method for pre-tensioning force of a hybrid tower steel cable according to any one of claims 1 to 7.

10. A computer device, characterized in that: include: a processor adapted to execute a computer program; A computer-readable storage medium having a computer program stored therein, wherein when the computer program is executed by the processor, the method for real-time monitoring of pre-tensioning force of a hybrid tower steel cable according to any one of claims 1 to 7 is implemented.

Images