Iron tower structure load intelligent early warning and dynamic balancing method and system thereof

Abstract

The invention relates to the technical field of iron tower early warning, in particular to an iron tower structure load intelligent early warning and dynamic balancing method and system, and the method comprises the steps: obtaining multi-mode sensor data of an iron tower structure, the multi-mode sensor data comprises vibration data, wind load data, deformation data, stress data and temperature data; constructing a hypercomplex group representation based on the multi-modal sensor data; performing topology continuous coherence analysis according to the hypercomplex group representation; a hypercomplex Laplacian matrix is constructed on the basis of a topological continuous coherence analysis result; according to the hypercomplex Laplacian matrix, establishing a fractional order nonlinear power system; based on the fractional order nonlinear power system, calculating related dimensions of the system; and according to the relevant dimensions and the characteristic values of the hypercomplex Laplacian matrix, the health index of the iron tower structure is generated, and comprehensive and accurate evaluation and timeliness of the health state of the iron tower structure are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of iron tower early warning, and more specifically to an iron tower structure load intelligent early warning and dynamic balancing method and a system thereof. Background Art

[0002] With the rapid development of the communications industry and the rapid construction of communications networks, the safety and reliability of communications tower structures have increasingly become the focus of the industry. Traditional tower monitoring methods mainly rely on regular manual inspections and simple sensor monitoring, which have many limitations. Manual inspections have long cycles and low efficiency, making it difficult to detect potential safety hazards in a timely manner; while simple sensor monitoring can provide some basic data, it is often difficult to fully reflect the health status of the tower.

[0003] In recent years, with the development of the Internet of Things and big data technology, some new tower monitoring methods have gradually emerged. Among them, the method based on vibration analysis has been widely used. This method installs an accelerometer on the tower to collect the vibration response data of the tower, and then uses signal processing technology to extract features to evaluate the health status of the tower. Although it has improved compared with traditional methods, there are still some obvious shortcomings.

[0004] First, this method mainly focuses on a single type of data (vibration data), which makes it difficult to fully reflect the complex dynamic characteristics of the tower. As a complex engineering structure, the health status of the tower is affected by many factors, and relying solely on vibration data is far from enough. Secondly, the existing data analysis methods are mostly based on traditional signal processing techniques, such as Fourier transform, which have obvious limitations when processing nonlinear and non-stationary signals. The load conditions faced by the tower in actual operation are often complex and changeable, and simple linear analysis is difficult to accurately describe its dynamic behavior.

[0005] In addition, most existing technologies remain at the passive monitoring stage, and even if problems are found, it is difficult to make timely adjustments and balances. This lag may cause small problems to gradually evolve into serious failures, increasing the difficulty and cost of maintenance. At the same time, due to the lack of in-depth understanding of the overall dynamic characteristics of the tower structure, the existing methods also have a lot of room for improvement in the accuracy and timeliness of early warning.

[0006] In the face of these problems, a new method is urgently needed to comprehensively, accurately and timely evaluate the health status of the tower and actively adjust the dynamic balance. This method should be able to comprehensively utilize multi-source heterogeneous data, use advanced mathematical models to describe the complex dynamic characteristics of the tower, and have intelligent early warning and automatic adjustment functions. Summary of the invention

[0007] The present invention is proposed to solve the above technical problems. The present invention provides a tower structure load intelligent early warning and dynamic balance method and system, aiming to solve the problems of insufficient data utilization, single analysis method, untimely early warning, and inability to actively adjust in the prior art.

[0008] The present invention provides an intelligent early warning and dynamic balancing method for a tower structure load, comprising:

[0009] The acquisition steps include:

[0010] Acquire multimodal sensor data of the tower structure, wherein the multimodal sensor data includes vibration data, wind load data, deformation data, stress data and temperature data;

[0011] Processing steps include:

[0012] constructing a hypercomplex group representation based on the multimodal sensor data;

[0013] According to the hypercomplex number group representation, topological persistent homology analysis is performed;

[0014] Based on the topological persistence homology analysis results, construct a hypercomplex Laplacian matrix;

[0015] According to the hypercomplex Laplacian matrix, a fractional-order nonlinear dynamic system is established;

[0016] Based on the fractional-order nonlinear dynamic system, calculating the correlation dimension of the system;

[0017] Output steps include:

[0018] A health index of the tower structure is generated according to the correlation dimension and the eigenvalue of the hypercomplex Laplacian matrix.

[0019] Preferably, the constructing of the hypercomplex group representation specifically includes:

[0020] Mapping the multimodal sensor data to a hypercomplex space to obtain a hypercomplex group element;

[0021] Group operations between elements of the hypercomplex group are defined.

[0022] Preferably, the hypercomplex group element is represented as follows:

[0023] h=a+bi+cj+dk+el

[0024] Among them, h is the hypercomplex group element, a is the vibration data, b is the wind load data, c is the deformation data, d is the stress data, e is the temperature data, and i, j, k, and l are hypercomplex units.

[0025] Preferably, the performing topological continuous coherence analysis specifically includes:

[0026] constructing a simple complex based on the hypercomplex group representation;

[0027] Compute the persistent homology group of the simplicial complex.

[0028] Preferably, the constructing of the hypercomplex Laplacian matrix specifically includes:

[0029] Constructing a degree matrix and an adjacency matrix based on the topological continuous homology analysis results;

[0030] The difference between the degree matrix and the adjacency matrix is ​​calculated to obtain a hypercomplex Laplacian matrix.

[0031] As a preference, it also includes:

[0032] Compute the eigenvalues ​​and eigenvectors of the hypercomplex Laplacian matrix.

[0033] Preferably, the establishing of the fractional-order nonlinear dynamic system specifically includes:

[0034] Use Caputo fractional derivative operators to construct fractional differential equations;

[0035] Substitute the eigenvalues ​​and eigenvectors of the hypercomplex Laplacian matrix into the fractional differential equation.

[0036] Preferably, the relevant dimensions of the computing system specifically include:

[0037] Calculating relevant integrals based on the solution of the fractional-order nonlinear dynamic system;

[0038] According to the correlation integral, the correlation dimension is obtained by limit calculation.

[0039] Preferably, the specific formula for generating the health index of the iron tower structure is:

[0040]

[0041] Among them, Ψ is the health index, w i and v j is the weight coefficient, λ i is the eigenvalue of the hypercomplex Laplacian matrix, D2 is the correlation dimension, and b is the bias term.

[0042] The tower structure load intelligent early warning and dynamic balance system for implementing the method comprises:

[0043] A data acquisition module, used to obtain multimodal sensor data of the tower structure;

[0044] Data processing module, used to perform hypercomplex group representation, topological persistent homology analysis, hypercomplex Laplacian matrix construction, fractional-order nonlinear dynamical system establishment and correlation dimension calculation;

[0045] Health assessment module, used to generate the health index of the tower structure;

[0046] An early warning module, used for intelligent early warning based on the health index;

[0047] The dynamic balance module is used to perform dynamic balance adjustment of the tower structure according to the health index and the early warning result.

[0048] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0049] The core of the present invention is to build a complete technology chain from data acquisition to intelligent early warning and dynamic balance. By innovatively introducing advanced mathematical tools such as hypercomplex group representation, topological persistent homology analysis, hypercomplex Laplacian matrix and fractional-order nonlinear dynamic system, the present invention realizes a comprehensive and accurate assessment of the health status of the tower structure. This method can not only effectively integrate multimodal data, but also deeply explore the topological features and nonlinear dynamic characteristics in the data, thereby greatly improving the accuracy and timeliness of early warning.

[0050] More importantly, the present invention introduces a dynamic balance adjustment mechanism based on monitoring. This innovative design enables the system to automatically adjust when an abnormality is detected, effectively preventing small problems from evolving into major failures. This proactive prevention approach represents an important development direction in the field of tower structural health monitoring.

[0051] There is a close logical connection and synergy between the various steps of the method of the present invention. For example, the hypercomplex group representation provides a solid foundation for the subsequent topological analysis, and the results of the topological analysis provide key input for constructing the hypercomplex Laplacian matrix. This organic combination of steps ensures the coherence and effectiveness of the entire analysis process, and ultimately achieves an accurate assessment of the health status of the tower.

[0052] In practical applications, the present invention has shown significant superiority. Compared with traditional methods, the present invention has significantly improved key indicators such as warning accuracy, false alarm rate, and warning lead time. In particular, in terms of dynamic balance adjustment effect, the performance of the present invention far exceeds that of the existing technology. These improvements not only improve the safety and reliability of communication towers, but also greatly reduce operation and maintenance costs, providing strong technical support for the stable operation of communication systems.

[0053] In general, the present invention has effectively solved a series of technical problems in the field of tower structure load monitoring and early warning by innovatively combining multiple advanced technologies. It not only promotes the development of relevant theories and methods, but also provides reliable technical guarantees for safe production in the communications industry, and has important theoretical significance and practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a system logic block diagram of the present invention.

[0055] Figure 2 This is a time series variation diagram of the health index of the present invention.

[0056] Figure 3 It is a visualization diagram of the load and structural response of the present invention.

[0057] Figure 4 This is the health warning and adjustment decision diagram of the present invention. DETAILED DESCRIPTION

[0058] Please refer to Figure 1-4 The present invention discloses an intelligent early warning and dynamic balance method and system for tower structure load. The method realizes comprehensive evaluation and early warning of the health status of the tower structure by innovatively combining multimodal data fusion, topological analysis and nonlinear dynamic system theory.

[0059] First, the method of the present invention includes an acquisition step, a processing step and an output step. In the acquisition step, the system acquires multimodal sensor data of the tower structure, which includes vibration data, wind load data, deformation data, stress data and temperature data. The collection of such multimodal data provides a comprehensive information basis for subsequent analysis.

[0060] The multimodal sensor data of the tower structure is the basis of this invention, which mainly includes the following categories:

[0061] Vibration data is obtained through accelerometers or displacement sensors installed at key locations of the tower to monitor the vibration characteristics of the tower.

[0062] Wind load data is obtained through wind speed sensors or wind pressure sensors to evaluate the impact of wind on the tower.

[0063] The deformation data is obtained through strain gauges or laser rangefinders and is used to monitor the deformation of the tower.

[0064] Stress data is obtained through strain gauges or pressure sensors and is used to evaluate the stress distribution of the tower.

[0065] Temperature data is obtained through temperature sensors to monitor the impact of ambient temperature on the performance of tower materials.

[0066] These sensors are distributed at key locations of the tower to collect dynamic response data of the structure in real time. The data acquisition module is responsible for transmitting this data to the data processing module to ensure the integrity and accuracy of the data.

[0067] The data processing module is the core part of the present invention and is responsible for executing the complex data analysis and modeling process. The specific steps are as follows:

[0068] Hypercomplex group representation: Multimodal sensor data is mapped to the hypercomplex space to form hypercomplex group elements. This step unifies different types of sensor data into a mathematical framework for subsequent analysis.

[0069] Topological persistence homology analysis: Based on the hypercomplex group representation, a simple complex is constructed and its persistence homology group is calculated. This step can capture the topological features in the data and identify potential anomalies in the tower structure.

[0070] Hypercomplex Laplacian matrix construction: Based on the results of topological persistence homology analysis, the degree matrix and adjacency matrix are constructed, and the difference between the two is calculated to obtain the hypercomplex Laplacian matrix. This step converts topological information into algebraic structure, preparing for further dynamical system analysis.

[0071] Fractional-order nonlinear dynamic system establishment: The fractional-order differential equation is constructed using the Caputo fractional-order derivative operator, and the eigenvalues ​​and eigenvectors of the hypercomplex Laplacian matrix are substituted into the equation. This step can more accurately describe the complex dynamic behavior of the tower structure, especially the memory effect and nonlinear behavior.

[0072] Health index calculation: Based on the solution of the fractional-order nonlinear dynamic system, the correlation dimension of the system is calculated, and combined with the eigenvalue of the hypercomplex Laplacian matrix, the health index of the tower structure is generated. This step provides a comprehensive assessment of the health status of the tower.

[0073] Through these steps, the present invention realizes the whole process from multi-modal data acquisition to complex dynamic system modeling, and provides strong technical support for intelligent load warning and dynamic balance of the tower structure.

[0074] In the processing step, the method first constructs a hypercomplex group representation based on the acquired multimodal sensor data. This is an innovative data fusion method that can unify different types of sensor data into a mathematical framework. Specifically, the method maps the data of each sensor to a hypercomplex space to form a hypercomplex group element.

[0075] In the multimodal data fusion of the tower structure, traditional data processing methods are usually difficult to effectively integrate different types of sensor data (such as vibration, wind load, deformation, stress and temperature). In order to overcome this problem, the present invention introduces the hypercomplex group representation. Hypercomplex is a mathematical structure that extends the concept of complex numbers, can process data in multiple dimensions at the same time, and maintains the good properties of complex number operations.

[0076] Preferably, in one embodiment of the present invention, the representation form of the hypercomplex group element is:

[0077] h=a+bi+bj+dk+el,

[0078] Among them, h is the hypercomplex group element, a is the vibration data, b is the wind load data, c is the deformation data, d is the stress data, e is the temperature data, i, j, k, l are hypercomplex units that meet specific multiplication rules (such as i 2 =j 2 =k 2 = l 2 =-1). This representation allows for the processing of multiple types of data within a unified mathematical structure, laying the foundation for subsequent analysis.

[0079] Introducing hypercomplex groups to represent and fuse multimodal sensor data:

[0080] H={h i |h i =a+bi+cj+dk+el,i=1,2,…,n},

[0081] Where H is the hypercomplex group, h i is a group element, representing the data of the i-th sensor. a is vibration data, b is wind load data, c is deformation data, d is stress data, and e is temperature data. i, j, k, and l are hypercomplex units.

[0082] Define the group operation:

[0083]

[0084] Next, the method performs topological persistence homology analysis based on the constructed hypercomplex group representation. The purpose of this step is to extract valuable information from the topological structure of the data. Specifically, a simple complex is first constructed based on the hypercomplex group representation, and then the persistence homology group of the simple complex is calculated. This analysis method can capture the topological features in the data and help identify potential anomalies in the tower structure.

[0085] By mapping multimodal sensor data to hypercomplex space, the present invention can process different types of data in a unified mathematical framework. This representation method not only simplifies the data fusion process, but also retains the original characteristics of each sensor data to avoid information loss. For example, the combination of vibration data and wind load data can more accurately reflect the dynamic response of the tower under wind force, while the combination of deformation data and stress data is helpful to evaluate the structural integrity of the tower.

[0086] Assuming that the tower is located in an area with strong winds, changes in wind speed recorded by the wind load sensor will affect the vibration characteristics of the tower. Through the hypercomplex group representation, wind load data and vibration data can be included in the analysis at the same time, thereby more comprehensively evaluating the stability of the tower under wind. In addition, changes in temperature data may affect the mechanical properties of the material, so processing it together with other data can better predict the behavior of the tower under different environmental conditions.

[0087] The health status of the tower structure depends not only on the data of a single sensor, but also on its overall topological structure. Traditional data analysis methods often ignore the topological relationship between data, resulting in insufficient sensitivity in identifying structural anomalies. To this end, the present invention introduces topological persistence homology analysis, a mathematical tool based on topology that can capture global and local features in the data.

[0088] Based on the topological analysis, this method further constructs the hypercomplex Laplacian matrix. This step converts the topological information into an algebraic structure to prepare for the subsequent dynamical system analysis. Specifically, based on the results of the topological persistence homology analysis, the degree matrix and the adjacency matrix are constructed, and then the difference between the two matrices is calculated to obtain the hypercomplex Laplacian matrix.

[0089] Persistent homology groups can reveal the topological characteristics of data at different scales and help identify potential anomalies in the tower structure.

[0090] Through topological continuous coherence analysis, the present invention can capture the complex topological relationships in the tower structure and identify abnormal patterns that are difficult to detect with traditional methods. For example, if a small crack or deformation occurs in a certain part of the tower, these changes may appear as new topological features (such as new holes or connections) in the topological structure. Through continuous coherence analysis, the system can detect these changes in time, issue early warning signals in advance, and avoid potential safety hazards.

[0091] Suppose a support pole of a tower is slightly bent. This change may not immediately affect the stability of the entire structure, but it may cause bigger problems over time. Through topological continuous homology analysis, the system can detect changes in the topological characteristics of the support pole at an early stage, so that maintenance measures can be taken in time to prevent the problem from getting worse.

[0092] After constructing the hypercomplex Laplacian matrix, this method calculates the eigenvalues ​​and eigenvectors of the matrix. These eigenvalues ​​and eigenvectors contain important dynamic characteristics information of the tower structure and provide key input for subsequent dynamic system analysis.

[0093] Based on the above processing results, this method establishes a fractional-order nonlinear dynamic system. This is another innovation of the present invention. By introducing fractional-order differential equations, the complex dynamic behavior of the tower structure can be described more accurately. Specifically, the Caputo fractional-order derivative operator is used to construct the fractional-order differential equation, and the eigenvalues ​​and eigenvectors of the hypercomplex Laplacian matrix are substituted into the equation.

[0094] This fractional-order nonlinear dynamic system can be expressed as:

[0095]

[0096] Among them, D α is the Caputo fractional derivative operator, α∈(0,1) is the fractional order, x(t) is the system state vector, and is a nonlinear function, λ is the eigenvalue, and v is the eigenvector. By solving this fractional-order nonlinear dynamic system, the dynamic response of the tower structure under different load conditions can be obtained. This provides an important basis for evaluating the health status of the structure.

[0097] Finally, in the output step, the method generates a health index of the tower structure based on the above analysis results. This health index takes into account multiple factors, including the dynamic characteristics and topological characteristics of the system, and can fully reflect the health status of the tower structure.

[0098] In practical applications, the threshold of the health index can be set based on experience. For example, 0.8 can be used as the warning threshold. When the health index is lower than 0.8, the system will issue a warning signal. The selection of this threshold needs to consider the specific characteristics and safety requirements of the tower structure, and can be optimized through long-term monitoring and data analysis.

[0099] Through this method, the present invention realizes intelligent early warning and dynamic balance of the tower structure load. Compared with the traditional method, the present invention has the following advantages: first, the effective fusion of multimodal data is realized through the hypercomplex group representation; second, the topological analysis method is introduced to capture the structural characteristics of the data; third, the fractional-order nonlinear dynamic system is used to more accurately describe the dynamic behavior of the tower; finally, a comprehensive health status assessment is provided by comprehensively considering the health index of multiple factors.

[0100] This method can not only timely discover the potential risks of the tower structure, but also provide a basis for subsequent maintenance and adjustment, thereby improving the safety and reliability of the communication tower. In practical applications, this method can be combined with an intelligent control system to achieve automatic adjustment and balance of the tower structure, further improving the stability and efficiency of the system. The method of the present invention further includes topological persistence coherence analysis in the processing steps. This innovative analysis method provides a new perspective for the health status assessment of the tower structure. Specifically, the topological persistence coherence analysis is performed based on the aforementioned constructed hypercomplex group representation.

[0101] In a preferred embodiment of the present invention, topological persistence homology analysis specifically includes two main steps: first, constructing a simple complex based on the hypercomplex group representation; second, calculating the persistence homology group of the simple complex. This analysis method can effectively capture the topological features in the data, thereby identifying possible abnormal patterns in the tower structure.

[0102] The construction of simple complexes is the basis of topological analysis. In this method, the simple complex K can be expressed as:

[0103] K={σ0,σ1,…,σ m},in

[0104] Where H is the aforementioned hypercomplex group, σ i is a simplex in K. This construction method maps the relationship between the elements of the hypercomplex group into the topological space, laying the foundation for subsequent analysis.

[0105] Next, the method computes the persistent homology group of the simplicial complex. The calculation formula is:

[0106]

[0107] in, is the k-dimensional continuous homology group, is a k-dimensional cyclic group, This calculation method can reveal the topological characteristics of data at different scales, which is of great significance for identifying potential problems in tower structures.

[0108] In the dynamic analysis of tower structures, the Laplacian matrix is ​​an important tool for describing the connection relationship between nodes and the energy distribution of the system. However, the traditional Laplacian matrix can only process scalar data and cannot be directly applied to multimodal data. To this end, the present invention proposes a hypercomplex Laplacian matrix that can process multiple types of data at the same time and retains the excellent properties of the Laplacian matrix.

[0109] Preferably, the method of the present invention further constructs a hypercomplex Laplacian matrix based on topological analysis. This step converts topological information into an algebraic structure to prepare for subsequent dynamical system analysis. Specifically, constructing a hypercomplex Laplacian matrix includes two key steps: first, constructing a degree matrix and an adjacency matrix based on the results of topological continuous homology analysis; second, calculating the difference between the two matrices to obtain a hypercomplex Laplacian matrix.

[0110] In one embodiment of the present invention, the hypercomplex Laplacian matrix The construction formula is:

[0111]

[0112] in, is the degree matrix, is the adjacency matrix. The diagonal elements of are defined as:

[0113]

[0114] This definition takes into account the strength of the connections between nodes and can better reflect the overall characteristics of the tower structure.

[0115] It is worth noting that in the construction of the hypercomplex Laplacian matrix, special attention should be paid to the definition of the matrix elements. For example, the adjacency matrix The elements of can be defined as:

[0116]

[0117] Among them, h i and h j is an element of the hypercomplex group, This definition takes into account the topological relationship between data and can better reflect the overall characteristics of the tower structure.

[0118] By constructing a hypercomplex Laplacian matrix, the present invention can transform topological information into an algebraic structure, providing a basis for further dynamic system analysis. The eigenvalues ​​and eigenvectors of the Laplacian matrix contain important dynamic information of the tower structure, which can help understand the stability and response characteristics of the system. For example, the distribution of eigenvalues ​​can reflect the stiffness and flexibility of the tower structure, while the eigenvectors can reveal the vibration mode of the system.

[0119] If a tower resonates under strong winds, some parts of the system may vibrate excessively. By analyzing the eigenvalues ​​and eigenvectors of the hypercomplex Laplacian matrix, the system can identify which parts are vibrating most violently and adjust the tension of the guy wires or other supporting structures accordingly to restore the balance of the system. In addition, the change in eigenvalues ​​can also serve as an important indicator of the health of the tower structure, helping to assess the long-term stability of the structure.

[0120] After constructing the hypercomplex Laplacian matrix, this method further calculates the eigenvalues ​​and eigenvectors of the matrix. This step is crucial to understanding the dynamic characteristics of the tower structure. The calculation of eigenvalues ​​and eigenvectors satisfies the following equations:

[0121]

[0122] Among them, λ is the eigenvalue and v is the corresponding eigenvector. These eigenvalues ​​and eigenvectors contain important dynamic information of the tower structure and provide key input for subsequent health status assessment.

[0123] The dynamic behavior of the tower structure is usually highly nonlinear, especially under complex environmental conditions (such as strong winds, earthquakes, etc.). Traditional integer-order differential equations are difficult to accurately describe these nonlinear behaviors, especially when the system has memory effects. To this end, the present invention introduces a fractional-order nonlinear dynamic system, which can more accurately characterize the complex dynamic behavior of the tower structure.

[0124] In a preferred embodiment of the present invention, based on the above analysis results, the method further establishes a fractional-order nonlinear dynamic system. This is another innovation of the present invention. By introducing fractional-order differential equations, the complex dynamic behavior of the tower structure can be described more accurately. Specifically, establishing a fractional-order nonlinear dynamic system includes two main steps: first, constructing a fractional-order differential equation using the Caputo fractional-order derivative operator; second, substituting the eigenvalues ​​and eigenvectors of the hypercomplex Laplacian matrix into the equation.

[0125] The general form of the fractional-order nonlinear dynamic system can be expressed as:

[0126]

[0127] Among them, D α is the Caputo fractional derivative operator, α∈(0,1) is the fractional order, x(t) is the system state vector, and is a nonlinear function, λ is the eigenvalue and v is the eigenvector.

[0128] In practical applications, the selection of fractional order α has an important influence on the system behavior. Generally, the closer the α value is to 1, the weaker the memory effect of the system; conversely, the closer the α value is to 0, the stronger the memory effect of the system. For tower structures, a suitable α value can be selected based on actual monitoring data and experience. For example, α can be initially set to 0.8, and then adjusted through experiments to obtain the best effect.

[0129] Compared with the traditional integer-order model, this fractional-order nonlinear dynamic system model can better describe the dynamic response of the tower structure in a complex environment, especially the memory effect and long-range correlation in the structure. This provides a more reliable theoretical basis for evaluating the health status of the tower structure.

[0130] Fractional-order derivative operators can capture the memory effect and long-range correlation of the system, which is very important for describing the dynamic behavior of the tower structure in a complex environment. For example, the vibration of the tower under the action of wind depends not only on the current wind speed, but also on the changes in wind speed over a period of time in the past. By introducing fractional-order derivatives, the system can more accurately simulate this memory effect, thereby improving the accuracy of the early warning. In addition, fractional-order nonlinear dynamic systems can also better describe the nonlinear behavior of the system, especially under extreme conditions (such as strong winds, earthquakes, etc.), the response of the system may be highly nonlinear.

[0131] Assuming that the tower is located in an area where wind speed changes frequently, changes in wind speed will cause the vibration characteristics of the tower to change significantly. By establishing a fractional-order nonlinear dynamic system, the system can more accurately predict the dynamic response of the tower under different wind speed conditions, especially when the wind speed suddenly increases or decreases, the vibration mode of the system may change significantly. By analyzing the fractional-order derivatives of the system, the system can identify potential risks in advance and take corresponding measures (such as adjusting the tension of the guy wire or activating the stress release mechanism) to ensure the safety of the tower.

[0132] Through the above steps, the method of the present invention realizes the whole process from multimodal data acquisition to complex dynamic system modeling. This innovative method combines data fusion, topological analysis and nonlinear dynamic system theory, and provides strong technical support for intelligent load warning and dynamic balance of tower structure. In a preferred embodiment of the present invention, based on the aforementioned established fractional-order nonlinear dynamic system, the method further calculates the correlation dimension of the system. This step aims to evaluate the complexity and stability of the tower structure from the perspective of chaos theory. Specifically, calculating the correlation dimension of the system includes two main steps: first, based on the solution of the fractional-order nonlinear dynamic system, calculating the correlation integral; second, according to the correlation integral, the correlation dimension is obtained by limit calculation.

[0133] The calculation formula of the correlation integral C(∈) is:

[0134]

[0135] Where N is the number of sampling points of the system state vector, Θ is the Heaviside step function, ∈ is the distance threshold, and x i and x j is the value of the system state vector at different times.

[0136] In practical applications, the choice of distance threshold ∈ has an important influence on the calculation results. Preferably, multiple ∈ values ​​can be selected for calculation, for example, ∈ = 0.01, 0.1, 1.0, etc., and then the trend of the correlation integral changing with ∈ can be observed to ensure the stability and reliability of the results.

[0137] Based on the correlation integral, the calculation formula of the correlation dimension D2 is:

[0138]

[0139] This limit calculation process is usually approximated by linear fitting in actual operation. The correlation dimension D2 reflects the complexity of the system's dynamic behavior and is of great significance for evaluating the health status of the tower structure. For example, if the calculated correlation dimension suddenly increases, it may mean that the dynamic behavior of the tower structure has become more complex, which may be a precursor to structural abnormalities.

[0140] The health status of the tower structure is a complex multi-factor problem, involving multiple aspects such as the dynamic characteristics, topological characteristics and complexity of the system. In order to comprehensively evaluate the health status of the tower, the present invention proposes a health index Ψ, which comprehensively considers multiple factors and can comprehensively reflect the health status of the tower.

[0141] In one embodiment of the present invention, based on all the above analysis results, the method finally generates a health index of the tower structure. This health index comprehensively considers multiple factors, including the dynamic characteristics, topological characteristics and complexity of the system, and can fully reflect the health status of the tower structure. The calculation formula of the health index is:

[0142]

[0143] Among them, Ψ is the health index, w i and v j is the weight coefficient, λ i is the eigenvalue of the hypercomplex Laplacian matrix, D2 is the correlation dimension, b is the bias term, k is the number of selected eigenvalues, and l is the selected fractal dimension order.

[0144] In practical applications, the weight coefficient w i and v j The choice of is crucial to the calculation of the health index. Preferably, these weight coefficients can be optimized based on historical data and expert knowledge by machine learning methods, such as support vector machines or neural networks. For example, w can be initially set to i =0.1(i=1,2,...,k),v j =0.2(j=1,2,...,l), and then continuously adjust through the training process to obtain the best effect.

[0145] In addition, the choice of bias term b will also affect the distribution of the health index. Usually b can be initialized to 0 and then fine-tuned according to the actual application scenario. For example, if you want the health index to be closer to 1 in a normal state, you can increase the value of b appropriately.

[0146] The health index Ψ provides an intuitive quantitative indicator that can fully reflect the health status of the tower structure. i and v j ,The system can flexibly adjust the contribution of each factor according to actual needs. For example, if the wind force is strong in a certain area, the system can increase the weight of the wind load data to pay more attention to the impact of the wind force on the tower. In addition, the changing trend of the health index can help identify potential problems of the tower structure, and take maintenance measures in advance to avoid accidents.

[0147] Assuming that the health index of the tower gradually decreases over a period of time, this may mean that there are potential problems in some parts of the tower (such as fatigue or corrosion of the support rods). By analyzing the changing trend of the health index, the system can issue early warning signals in time to remind maintenance personnel to check and repair. In addition, the health index can also be used as an input to the automatic control system to help the system automatically adjust the tension of the guy wires or other supporting structures according to the health status of the tower to maintain the stability and safety of the tower.

[0148] The present invention also provides an iron tower structure load intelligent early warning and dynamic balance system for implementing the above method. The system includes multiple functional modules, each module has its specific function and role.

[0149] First, the data acquisition module 1 is used to obtain multimodal sensor data of the tower structure. This module can include multiple types of sensors, such as vibration sensors, strain sensors, wind speed sensors, etc. These sensors are distributed at key locations of the tower to collect dynamic response data of the structure in real time.

[0150] The data processing module 2 is the core part of the system, responsible for executing complex data analysis and modeling processes. Together, they realize the extraction and analysis from raw data to advanced features.

[0151] The health assessment module 3 is responsible for generating the health index of the tower structure. This module receives the output of the data processing module 2, comprehensively considers multiple factors, and calculates an index reflecting the overall health status of the tower structure.

[0152] The early warning module 4 performs intelligent early warning based on the health index generated by the health assessment module 3. This module can set multiple early warning thresholds corresponding to different levels of alarms. For example, the health index 0.8 can be set as the yellow early warning threshold, 0.6 as the orange early warning threshold, and 0.4 as the red early warning threshold. When the health index is lower than the corresponding threshold, the system will automatically issue an alarm of the corresponding level.

[0153] Finally, the dynamic balance module 5 performs dynamic balance adjustment of the tower structure according to the health index and the warning result. This module can include multiple adjustment mechanisms, such as tension adjustment, support structure adjustment, etc. For example, when the tower is detected to be tilted, the system can automatically adjust the tension of the guy wire to restore the balance; when it is found that the stress in some parts is too large, the corresponding stress release mechanism can be activated.

[0154] Through the collaborative work of these modules, the system can achieve real-time monitoring, intelligent early warning and dynamic balance of the tower structure load, greatly improving the safety and reliability of the communication tower. At the same time, the modular design of the system also makes it have good scalability and adaptability, and can be flexibly adjusted and upgraded according to actual needs.

[0155] To verify the superiority of the present invention, three different methods were selected for comparative experiments. The experiment selected a communication tower as the research object, and these towers were monitored and analyzed for a period of one year. The following is a detailed description of the embodiments and comparative examples and the experimental results.

[0156] Embodiment 1: The method of the present invention

[0157] The method of the present invention adopts innovative technologies such as multimodal data fusion, hypercomplex group representation, topological persistent homology analysis, hypercomplex Laplacian matrix and fractional-order nonlinear dynamical system. The specific implementation process strictly follows the steps described above, including data acquisition, hypercomplex group construction, topological analysis, dynamical system modeling and health index calculation.

[0158] Comparative Example 1: Traditional strain monitoring method

[0159] This method mainly relies on installing strain gauges at key parts of the tower to evaluate the stress condition of the tower by measuring the strain. This is a commonly used method in current engineering practice, with the advantages of simple operation and intuitiveness, but it is difficult to fully reflect the dynamic characteristics of the tower.

[0160] Comparative Example 2: Vibration Characteristics Analysis Method

[0161] This method collects the vibration response data of the tower by installing acceleration sensors, and then uses signal processing techniques such as fast Fourier transform (FFT) to extract vibration characteristics and evaluate the health status of the tower. This method is an improvement over traditional strain monitoring, but it is still difficult to process multi-source heterogeneous data.

[0162] During the experiment, we focused on the following indicators: warning accuracy, false alarm rate, average warning lead time, dynamic balance adjustment effect and calculation efficiency. The detection standards and methods of these indicators are as follows:

[0163] 1. Warning accuracy: The warning information issued by the system is verified through manual inspections, and the proportion of correct warnings is calculated.

[0164] 2. False alarm rate: The ratio of the number of false alarms issued by the system to the total number of alarms in a year.

[0165] 3. Average warning lead time: For confirmed faults, calculate the average lead time between the system warning time and the actual fault occurrence time.

[0166] 4. Dynamic balance adjustment effect: evaluated by measuring the change in the tower’s tilt angle before and after adjustment.

[0167] 5. Computational efficiency: record the average computational time for each health assessment.

[0168] The experimental results are shown in the following table:

[0169] index Example 1 Comparative Example 1 Comparative Example 2 Early warning accuracy 95.8％ 78.3％ 85.6％ False Positive Rate 2.1％ 8.7％ 5.4％ Average warning lead time 72 hours 24 hours 48 hours Dynamic balance adjustment effect 92.5％ not applicable 76.8％ Computational efficiency (per evaluation) 1.5 seconds 0.5 seconds 0.8 seconds

[0170] It can be seen from the experimental results that the method of the present invention has obvious advantages in all indicators. The early warning accuracy rate reached 95.8%, which is much higher than the traditional method and vibration analysis method. This is mainly due to the multimodal data fusion and complex system analysis technology adopted by the present invention, which can comprehensively evaluate the health status of the tower from multiple angles.

[0171] The false alarm rate is only 2.1%, which is significantly lower than the other two methods. This shows that the method of the present invention has strong anti-interference ability and can effectively distinguish normal fluctuations from real anomalies. This is particularly important in practical applications, because too high a false alarm rate will lead to fatigue of operation and maintenance personnel and waste of resources.

[0172] In terms of early warning time, the method of the present invention can issue an early warning 72 hours in advance on average, which provides sufficient time for operation and maintenance personnel to carry out preventive maintenance and emergency response. In contrast, traditional strain monitoring methods can often only detect problems when they are about to occur, and although the vibration feature analysis method has been improved, the early warning time is still insufficient.

[0173] The dynamic balance adjustment effect is a highlight of the invention. Through dynamic adjustment guided by intelligent algorithms, 92.5% of abnormal tilt situations have been effectively improved. This not only improves the safety of the tower, but also greatly reduces the need for manual intervention. Traditional strain monitoring methods cannot provide this function, and although the vibration characteristic analysis method can be adjusted simply, the effect is limited.

[0174] In terms of computational efficiency, the method of the present invention is slightly inferior to the other two methods, and each evaluation takes an average of 1.5 seconds. This is because the method involves a more complex mathematical model and calculation process. However, considering that the state of the tower usually does not change too quickly, this calculation time is completely acceptable in practical applications and will not affect the real-time performance of the system.

[0175] In summary, the present invention has shown obvious advantages in key indicators such as early warning accuracy, timeliness and active adjustment capabilities. These advantages are derived from the unique technical route of the present invention, especially the innovations such as multimodal data fusion, topological analysis and nonlinear dynamic system modeling. This method can not only more comprehensively and accurately evaluate the health status of the tower, but also actively perform dynamic balance adjustments, greatly improving the safety and reliability of communication towers.

[0176] Although the computational complexity is slightly increased, this trade-off is completely worth it considering the significant performance improvement it brings. These experimental results fully demonstrate the advancement and practicality of the method of the present invention in the field of intelligent early warning and dynamic balance of tower structure loads, providing a strong guarantee for the safe operation of the communication system.

[0177] Figure 2 The change of the health index of the tower structure over time is shown, reflecting the health status of the structure under different load conditions. By introducing hypercomplex group representation and fractional-order nonlinear dynamic system, the present invention can monitor the health changes of the tower structure in real time and accurately. The sharp fluctuation of the health index at certain time points indicates that the structure is subjected to external disturbances (such as wind load, vibration, etc.), which promptly reflects the potential structural damage or fatigue risk. The method of the present invention can not only capture the time series changes of the health index, but also dynamically adjust the health assessment model according to different environmental loads and structural responses, thereby improving the accuracy of the prediction. Through real-time health index monitoring, the present invention can effectively warn of potential damage or instability of the tower structure, and provide a scientific basis for maintenance decisions. Compared with traditional methods, the present invention provides a higher-precision health assessment, can capture more details of the dynamic changes of the structure, and has better adaptability and flexibility.

[0178] Figure 3 The response of the tower structure under different load conditions is shown, including vibration, deformation, stress and other responses. The present invention can more comprehensively capture the impact of loads on the tower structure by integrating multimodal sensor data (such as wind loads, vibration data, etc.) and combining hypercomplex group representation with topological persistent homology analysis. The load and structural response data in the visualization diagram show the time lag effect of different loads on the structural response, indicating that the response of the structure is not linear, which is reflected by the fractional-order nonlinear dynamic system of the present invention. The precise modeling and visual display of loads and structural responses can help engineers better understand the response characteristics of the tower structure under different environmental factors. Compared with traditional methods, the present invention can provide more accurate load response analysis, allowing engineers to more effectively evaluate the reliability and safety of the structure under different load conditions.

[0179] Figure 4The intelligent early warning and dynamic balance adjustment decisions based on the health index are demonstrated. According to the health index and the early warning results, the system can automatically adjust the dynamic balance of the structure to ensure that the tower can maintain stability when facing abnormal loads or damage. The health early warning mechanism of the present invention combines the real-time data of the tower structure (such as wind speed, temperature, etc.) with the health index to predict possible risks in advance and propose adjustment measures, rather than waiting until the damage occurs before repairing. By identifying potential damage or instability at an early stage and automatically adjusting the structure according to the health index, the risk of accidents can be greatly reduced. The present invention avoids the structure being in an unstable state for a long time by predicting and performing dynamic balance adjustments, thereby extending the service life of the tower and reducing the need for large-scale repairs. Compared with traditional manual monitoring and adjustment methods, the present invention provides a more efficient and automated early warning and decision-making system, which improves the response speed and accuracy to emergencies.

[0180] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, replacement, and improvement made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Intelligent early warning and dynamic balancing method for tower structure load, characterized in that: include: The acquisition steps include: Acquire multimodal sensor data of the tower structure, wherein the multimodal sensor data includes vibration data, wind load data, deformation data, stress data and temperature data; Processing steps include: constructing a hypercomplex group representation based on the multimodal sensor data; According to the hypercomplex number group representation, topological persistent homology analysis is performed; Based on the topological persistence homology analysis results, construct a hypercomplex Laplacian matrix; According to the hypercomplex Laplacian matrix, a fractional-order nonlinear dynamic system is established; Based on the fractional-order nonlinear dynamic system, calculating the correlation dimension of the system; Output steps include: A health index of the tower structure is generated according to the correlation dimension and the eigenvalue of the hypercomplex Laplacian matrix.

2. The method according to claim 1, characterized in that The constructing of the hypercomplex group representation specifically includes: Mapping the multimodal sensor data to a hypercomplex space to obtain a hypercomplex group element; Group operations between elements of the hypercomplex group are defined.

3. The method according to claim 2, characterized in that: The representation of the hypercomplex group element is: h=a+bi+cj+dk+el Among them, h is the hypercomplex group element, a is the vibration data, b is the wind load data, c is the deformation data, d is the stress data, e is the temperature data, and i, j, k, and l are hypercomplex units.

4. The method according to claim 1, characterized in that: The topological persistence coherence analysis specifically includes: constructing a simple complex based on the hypercomplex group representation; Compute the persistent homology group of the simplicial complex.

5. The method according to claim 1, characterized in that The constructing of the hypercomplex Laplacian matrix specifically includes: Constructing a degree matrix and an adjacency matrix based on the topological continuous homology analysis results; The difference between the degree matrix and the adjacency matrix is ​​calculated to obtain a hypercomplex Laplacian matrix.

6. The method according to claim 5, characterized in that Also includes: Compute the eigenvalues ​​and eigenvectors of the hypercomplex Laplacian matrix.

7. The method according to claim 1, characterized in that The establishment of the fractional-order nonlinear dynamic system specifically includes: Use Caputo fractional derivative operators to construct fractional differential equations; Substitute the eigenvalues ​​and eigenvectors of the hypercomplex Laplacian matrix into the fractional differential equation.

8. The method according to claim 1, characterized in that The relevant dimensions of the computing system specifically include: Calculating relevant integrals based on the solution of the fractional-order nonlinear dynamic system; According to the correlation integral, the correlation dimension is obtained by limit calculation.

9. The method according to claim 1, characterized in that: The specific formula for generating the health index of the iron tower structure is: Among them, Ψ is the health index, w i and v j is the weight coefficient, λ i is the eigenvalue of the hypercomplex Laplacian matrix, D2 is the correlation dimension, and b is the bias term.

10. An intelligent early warning and dynamic balance system for tower structure loads for implementing the method described in any one of claims 1 to 9, characterized in that: include: A data acquisition module, used to obtain multimodal sensor data of the tower structure; Data processing module, used to perform hypercomplex group representation, topological persistent homology analysis, hypercomplex Laplacian matrix construction, fractional-order nonlinear dynamical system establishment and correlation dimension calculation; Health assessment module, used to generate the health index of the tower structure; An early warning module, used for intelligent early warning based on the health index; The dynamic balance module is used to perform dynamic balance adjustment of the tower structure according to the health index and the early warning result.

Images