Line galloping three-dimensional state analysis method and system based on Gaussian process

Abstract

The invention discloses a line galloping three-dimensional state analysis method and system based on a Gaussian process, and relates to the technical field of information technology, and the method comprises the steps: obtaining line galloping position vector time sequence data; constructing a Gaussian process model, and fitting line galloping position vector time sequence data to obtain a model hyper-parameter at a next moment; constructing a state transition matrix; a Markov Gaussian process model is constructed; predicting a line galloping position vector and a covariance matrix; obtaining a weighted and predicted vector and a covariance matrix of each Gaussian process model; receiving an actual line galloping position vector; calculating a mahalanobis distance; calculating an abnormal value detection threshold value; comparing the mahalanobis distance with an abnormal value detection threshold value; and updating the state estimation. According to the method, a plurality of Gaussian process models are used, each model has a different covariance function, and the dynamic change of the power transmission line under the influence of complex environmental factors can be captured and simulated, so that the prediction precision and robustness are improved.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a method and system for analyzing three-dimensional state of line galloping based on Gaussian process. Background Art

[0002] Conductor line gallop monitoring is a critical safety and maintenance task in power systems, which involves real-time tracking and analysis of the dynamic position changes of transmission lines under the action of natural wind. Traditional line gallop monitoring technologies usually rely on physical sensors installed directly on transmission lines, which can measure the position change data of the conductors. However, these methods usually face the following challenges:

[0003] Limitations in data processing capabilities: As the transmission line environment is complex and affected by multiple factors, such as wind speed, wind direction, and temperature changes, traditional methods may be inefficient in processing large-scale real-time data.

[0004] Insufficient prediction capabilities: Most existing technologies lack the ability to predict the future state of wire movement, which limits their application in risk prevention and decision support.

[0005] Abnormality handling capability: Under extreme weather conditions, data outliers and noise may cause false positives or negative positives in the monitoring system, affecting the overall reliability of the system. Summary of the invention

[0006] In view of the above-mentioned problems, the present invention is proposed.

[0007] Therefore, the technical problem solved by the present invention is to solve the technical problems of low data processing efficiency, insufficient prediction ability and weak abnormality processing ability in transmission line dancing monitoring.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions: a three-dimensional state analysis method of line dancing based on Gaussian process, which comprises the following steps:

[0009] Obtain the time series data of the line dancing position vector; construct a Gaussian process model, use the regression framework of the Gaussian process model to fit the time series data of the line dancing position vector, and obtain the model hyperparameters corresponding to the next moment; construct the state transfer matrix of the next moment according to the model hyperparameters of each Gaussian process model; construct a Markov Gaussian process model according to the state transfer matrix of the next moment; use the state transfer matrix and the line dancing position vector of the previous moment to predict the line dancing position vector and covariance matrix corresponding to each Gaussian process model at the next moment; perform weighted summation on the line dancing position vector and covariance matrix corresponding to each Gaussian process model at the next moment to obtain the weighted sum prediction vector and weighted sum covariance matrix; receive the actual line dancing position vector; calculate the Mahalanobis distance between the weighted sum prediction value and the actual line dancing position vector; calculate the outlier detection threshold according to the dimension of the line dancing position vector and the set significance level; compare the size of the Mahalanobis distance and the outlier detection threshold; update the state estimate of each Markov Gaussian process through the line dancing position vector to be detected.

[0010] As a preferred solution of the Gaussian process-based line dancing three-dimensional state analysis method of the present invention, wherein: the obtaining of the model hyperparameters corresponding to the next moment includes selecting the covariance function k(t i ,t j ).

[0011] The covariance K between each pair of vectors in the line dancing position vector time series data is calculated by the covariance function ij .

[0012] According to K ij Construct the covariance matrix K.

[0013] Constructing the covariance matrix taking noise into account It is expressed as,

[0014]

[0015] Among them, t i and t j Respectively represent the time points numbered i and j in the time series, represents the variance of process noise and I represents the identity matrix.

[0016] Among them, t i and t j Respectively represent the time points numbered i and j in the time series, represents the variance of process noise and I represents the identity matrix.

[0017] As a preferred solution of the Gaussian process-based line galloping three-dimensional state analysis method described in the present invention, wherein: the covariance function k(t i ,t j ) includes the following forms,

[0018]

[0019] Among them, σ 2 , l and σ n is the model hyperparameter, σ 2 represents the signal amplitude corresponding to each component of the road dancing position vector, l represents k1(t i ,t j ), Γ(ν) represents the gamma function, ν represents the smoothing parameter, ν = 1.5, K ν represents the modified Bessel function, ρ represents k2(t i ,t j ) length scale, p represents k3(t i ,t j ) period, d represents k3(t i ,t j ) length scale.

[0020] As a preferred solution of the Gaussian process-based line galloping three-dimensional state analysis method described in the present invention, the state transfer matrix includes: when the covariance function is k1(t i ,t j ), the state transfer matrix at the next moment is,

[0021]

[0022] When the covariance function is k2(t i ,t j ), the state transfer matrix at the next moment is:

[0023]

[0024] When the covariance function is k3(t i ,t j ), the state transfer matrix at the next moment is:

[0025]

[0026] As a preferred solution of the Gaussian process-based line galloping three-dimensional state analysis method described in the present invention, wherein: the Markov Gaussian process model includes, when the covariance function is k1(t i ,t j ), the Markov Gaussian process model is,

[0027] x(t+1)=A1x(t+1)

[0028] When the covariance function is k2(t i ,t j ), the Markov Gaussian process model is,

[0029] x(t+1)=A2x(t+1)

[0030] When the covariance function is k3(t i ,t j ), the Markov Gaussian process model is,

[0031] x(t+1)=A3x(t+1).

[0032] As a preferred solution of the Gaussian process-based line galloping three-dimensional state analysis method described in the present invention, the line galloping position vector and covariance matrix corresponding to each Gaussian process model at the next moment are expressed as:

[0033]

[0034] in,

[0035]

[0036] Where A represents the state transfer matrix, A includes A1, A2 and A3, P k represents the posterior covariance matrix of the state variables at the kth iteration, Q represents the processing noise covariance matrix, H represents the observation matrix, R represents the observation noise covariance matrix, and Z k represents the observation vector of the kth iteration, represents the predicted value of the prior covariance matrix of the kth iteration, I represents the identity matrix, represents the predicted value of the line dancing position vector at the kth iteration, Represents the output value of the line dancing position vector at the kth iteration.

[0037] As a preferred solution of the Gaussian process-based line galloping three-dimensional state analysis method described in the present invention, the weighted representation of each Gaussian process model is:

[0038]

[0039] Where k represents the sequence number of the Gaussian process model, n represents the number of time points in the line dancing position vector time series data, x(t) represents the actual observation value of the line dancing position vector at time point t, Represents the predicted value of the line dancing position vector at time point t calculated by the kth Gaussian process model.

[0040] Another object of the present invention is to provide a three-dimensional state analysis system for line galloping based on Gaussian process, which can efficiently process large-scale real-time data and accurately predict the future state of the conductor, thereby solving the problems of low efficiency of existing technologies in processing data under the influence of complex environmental factors, insufficient prediction ability, and false alarms or omissions caused by outliers and noise.

[0041] To solve the above technical problems, the present invention provides the following technical solutions: a three-dimensional state analysis system for line dancing based on Gaussian process, comprising a data acquisition module, a Gaussian process model construction module, a model hyperparameter acquisition module and a state estimation update module.

[0042] The data acquisition module is responsible for acquiring the line dancing position vector time series data.

[0043] The Gaussian process model building module is responsible for building a Gaussian process model and using the model to fit the line dancing position vector time series data.

[0044] The model hyperparameter acquisition module selects a covariance function for each line dancing position vector dimension and calculates the model hyperparameters.

[0045] The state estimation updating module updates the state estimation of each Markov Gaussian process by the line galloping position vector to be detected.

[0046] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned method for analyzing the three-dimensional state of line galloping based on Gaussian process are implemented.

[0047] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a three-dimensional state analysis method for line galloping based on a Gaussian process as described above.

[0048] Beneficial effects of the present invention: By using multiple Gaussian process models, each model has a different covariance function. The present invention can more accurately capture and simulate the dynamic changes of transmission lines under the influence of complex environmental factors, thereby improving the accuracy and robustness of the prediction.

[0049] The present invention uses the Markov Gaussian process to make an arbitrarily accurate approximation to the Gaussian process with a stationary kernel function, and then uses the Kalman filter to make predictions and inferences, with a time complexity of O(N), which is relatively faster than the time complexity of O(N) for predictions and inferences only through the Gaussian process. 3). This method is more computationally efficient, achieves continuous real-time prediction of line galloping status, optimizes data processing capabilities, and ensures efficiency in large-scale real-time data processing scenarios.

[0050] The present invention enhances the system's ability to process abnormal data under extreme weather conditions and reduces the risks of false positives and false negatives by introducing Mahalanobis distance calculation and outlier detection mechanisms. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0052] Figure 1 An overall flow chart of a method for analyzing three-dimensional state of line galloping based on Gaussian process provided in the first embodiment of the present invention.

[0053] Figure 2 An overall framework diagram of a line galloping three-dimensional state analysis system based on Gaussian process provided in the second embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0055] Example 1, reference Figure 1 , which is an embodiment of the present invention, provides a three-dimensional state analysis method of line galloping based on Gaussian process, characterized in that:

[0056] S1. Obtain line dancing position vector time series data.

[0057] S2. Construct more than one Gaussian process model, each with a different covariance function.

[0058] S3. Use the regression framework of the Gaussian process model to fit the line dancing position vector time series data to obtain the model hyperparameters corresponding to the next moment.

[0059] Obtaining the model hyperparameters corresponding to the next moment includes selecting the covariance function k(t i ,tj ).

[0060] The covariance K between each pair of vectors in the line dancing position vector time series data is calculated by the covariance function ij .

[0061] According to K ij Construct the covariance matrix K.

[0062] Constructing the covariance matrix taking noise into account It is expressed as,

[0063]

[0064] Among them, t i and t j Respectively represent the time points numbered i and j in the time series, represents the variance of process noise and I represents the identity matrix.

[0065] Among them, t i and t j Respectively represent the time points numbered i and j in the time series, represents the variance of process noise and I represents the identity matrix.

[0066] Covariance function k(t i ,t j ) includes the following forms,

[0067]

[0068] Among them, σ 2 , l and σ n is the model hyperparameter, σ 2 represents the signal amplitude corresponding to each component of the road dancing position vector, l represents k1(t i ,t j ), Γ(ν) represents the gamma function, ν represents the smoothing parameter, ν = 1.5, K ν represents the modified Bessel function, ρ represents k2(t i ,t j ) length scale, p represents k3(t i ,t j ) period, d represents k3(t i ,t j ) length scale.

[0069] S4, constructing a state transfer matrix at the next moment according to the model hyperparameters of each Gaussian process model;

[0070] The state transfer matrix includes, when the covariance function is k1(t i ,tj ), the state transfer matrix at the next moment is,

[0071]

[0072] When the covariance function is k2(t i ,t j ), the state transfer matrix at the next moment is:

[0073]

[0074] When the covariance function is k3(t i ,t j ), the state transfer matrix at the next moment is:

[0075]

[0076] S5. Construct a Markov Gaussian process model according to the state transfer matrix at the next moment.

[0077] The Markov Gaussian process model includes, when the covariance function is k1(t i ,t j ), the Markov Gaussian process model is,

[0078] x(t+1)=A1x(t+1)

[0079] When the covariance function is k2(t i ,t j ), the Markov Gaussian process model is,

[0080] x(t+1)=A2x(t+1)

[0081] When the covariance function is k3(t i ,t j ), the Markov Gaussian process model is,

[0082] x(t+1)=A3x(t+1).

[0083] S6. Use the state transfer matrix and the line dancing position vector at the previous moment to predict the line dancing position vector and covariance matrix corresponding to each Gaussian process model at the next moment.

[0084] The line dancing position vector and covariance matrix corresponding to each Gaussian process model at the next moment are expressed as,

[0085]

[0086] in,

[0087]

[0088] Where A represents the state transfer matrix, A includes A1, A2 and A3, P k represents the posterior covariance matrix of the state variables at the kth iteration, Q represents the processing noise covariance matrix, H represents the observation matrix, R represents the observation noise covariance matrix, and Z k represents the observation vector of the kth iteration, represents the predicted value of the prior covariance matrix of the kth iteration, I represents the identity matrix, represents the predicted value of the line dancing position vector at the kth iteration, Represents the output value of the line dancing position vector at the kth iteration.

[0089] S7. Perform weighted summation on the line dancing position vector and covariance matrix corresponding to each Gaussian process model at the next moment to obtain a weighted sum prediction vector and a weighted sum covariance matrix.

[0090] The weighted representation of each Gaussian process model is,

[0091]

[0092] Where k represents the sequence number of the Gaussian process model, n represents the number of time points in the line dancing position vector time series data, x(t) represents the actual observation value of the line dancing position vector at time point t, Represents the predicted value of the line dancing position vector at time point t calculated by the kth Gaussian process model.

[0093] S8. Receive the actual line dancing position vector.

[0094] S9. Calculate the Mahalanobis distance between the weighted and predicted value and the actual line dancing position vector.

[0095] S10. Calculate an outlier detection threshold according to the dimension of the road dancing position vector and the set significance level.

[0096] S11. Compare the Mahalanobis distance with the outlier detection threshold. If the Mahalanobis distance is greater than the outlier detection threshold, determine that the actual line dancing position vector is an outlier, exclude the actual line dancing position vector from the line dancing position vector time series data, and jump to step S08. Otherwise, add the actual line dancing position vector to the line dancing position vector time series data in chronological order.

[0097] S12. Update the state estimation of each Markov Gaussian process by the line galloping position vector to be detected, where the state estimation includes the line galloping position vector and the covariance matrix.

[0098] S13. Jump to step S06.

[0099] Example 2, reference Figure 2 , which is an embodiment of the present invention, provides a system for a line dancing three-dimensional state analysis method based on Gaussian process, characterized by: including a data acquisition module, a Gaussian process model construction module, a model hyperparameter acquisition module and a state estimation update module.

[0100] The data acquisition module is responsible for obtaining the time series data of the line dancing position vector.

[0101] The Gaussian process model building module is responsible for building a Gaussian process model and using the model to fit the line dancing position vector time series data.

[0102] The model hyperparameter acquisition module selects the covariance function for each line dancing position vector dimension and calculates the model hyperparameters.

[0103] The state estimation update module updates the state estimation of each Markov Gaussian process by the line galloping position vector to be detected.

[0104] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0105] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0106] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0107] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0108] Embodiment 3, in this embodiment, in order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0109] In the operation and maintenance of transmission lines, line galloping is an issue that requires close attention. Line galloping not only affects the transmission efficiency, but in severe cases may also cause line damage and even cause power grid accidents. Although the traditional line galloping monitoring method can capture the movement state of the line to a certain extent, it has many limitations in practical applications. These limitations are mainly reflected in low data processing efficiency, insufficient prediction ability, and weak anomaly detection ability. In order to overcome these limitations, this study proposes a three-dimensional state analysis method for line galloping based on Gaussian process. The following examples are intended to demonstrate the advantages of the method of the present invention in practical applications by comparison with traditional methods, and to verify its effectiveness in data processing, prediction accuracy and anomaly detection. Through these examples, we will explain in detail the specific implementation steps of the method of the present invention and the beneficial effects achieved. As shown in Table 1.

[0110] Table 1 Experimental effect comparison chart

[0111] Experimental indicators Traditional methods Method of the present invention Data processing time 120 30 Prediction accuracy (%) 70 95 Anomaly detection accuracy (%) 60 90

[0112] The method of the present invention can quickly fit large-scale real-time data by constructing a Gaussian process model, thereby significantly reducing data processing time.

[0113] Prediction accuracy: The method of the present invention utilizes the Markov Gaussian process model to improve the prediction accuracy of the future state of the conductor.

[0114] Anomaly detection accuracy: By calculating the Mahalanobis distance and dynamically updating the outlier detection threshold, the method of the present invention significantly improves the accuracy of anomaly detection.

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A three-dimensional state analysis method of line galloping based on Gaussian process, characterized in that: include: Get the line dancing position vector time series data; Construct a Gaussian process model and use the regression framework of the Gaussian process model to fit the line dancing position vector time series data to obtain the model hyperparameters corresponding to the next moment; Construct the state transfer matrix at the next moment according to the model hyperparameters of each Gaussian process model; Construct a Markov Gaussian process model based on the state transition matrix at the next moment; Use the state transfer matrix and the line dancing position vector at the previous moment to predict the line dancing position vector and covariance matrix corresponding to each Gaussian process model at the next moment; The line dancing position vector and covariance matrix corresponding to each Gaussian process model at the next moment are weighted and summed to obtain a weighted sum prediction vector and a weighted sum covariance matrix; Receive actual line galloping position vector; Calculate the Mahalanobis distance between the weighted sum prediction value and the actual line galloping position vector; Calculate the outlier detection threshold based on the dimension of the line galloping position vector and the set significance level; Compare the Mahalanobis distance with the outlier detection threshold; The state estimate of each Markov Gaussian process is updated by the line galloping position vector to be detected.

2. A Gaussian process-based line galloping three-dimensional state analysis method as claimed in claim 1, characterized in that: The obtaining of the model hyperparameters corresponding to the next moment includes selecting the covariance function k(t i , t j ); The covariance K between each pair of vectors in the line dancing position vector time series data is calculated by the covariance function ij ; According to K ij Construct the covariance matrix K; Constructing the covariance matrix taking noise into account It is expressed as, Among them, t i and t j Respectively represent the time points numbered i and j in the time series, represents the variance of process noise, and I represents the identity matrix; Among them, t i and t j Respectively represent the time points numbered i and j in the time series, represents the variance of process noise and I represents the identity matrix.

3. A Gaussian process-based line galloping three-dimensional state analysis method as claimed in claim 2, characterized in that: The covariance function k(t i , t j ) includes the following forms, Among them, σ 2 , l and σ n is the model hyperparameter, σ 2 represents the signal amplitude corresponding to each component of the road dancing position vector, l represents k1(t i , t j ), Γ(v) represents the gamma function, v represents the smoothing parameter, v = 1.5, K v represents the modified Bessel function, represents k2(t i , t j ) length scale, p represents k3(t i , t j ) period, d represents k3(t i , t j ) length scale.

4. A Gaussian process-based line galloping three-dimensional state analysis method as claimed in claim 3, characterized in that: The state transfer matrix includes: when the covariance function is k1(t i , t j ), the state transfer matrix at the next moment is, When the covariance function is k2(t i , t j ), the state transfer matrix at the next moment is: When the covariance function is k3(t i , t j ), the state transfer matrix at the next moment is:

5. A Gaussian process-based line galloping three-dimensional state analysis method as claimed in claim 4, characterized in that: The Markov Gaussian process model includes: when the covariance function is k1(t i , t j ), the Markov Gaussian process model is: x(t+1)=A1x(t+1) When the covariance function is k2(t i , t j ), the Markov Gaussian process model is: x(t+1)=A2x(t+1) When the covariance function is k3(t i , t j ), the Markov Gaussian process model is: x(t+1)=A3x(t+1).

6. A Gaussian process-based line galloping three-dimensional state analysis method as claimed in claim 5, characterized in that: The line dancing position vector and covariance matrix corresponding to each Gaussian process model at the next moment are expressed as: in, Where A represents the state transfer matrix, A includes A1, A2 and A3, P k represents the posterior covariance matrix of the state variables at the kth iteration, Q represents the processing noise covariance matrix, H represents the observation matrix, R represents the observation noise covariance matrix, and Z k represents the observation vector of the kth iteration, represents the predicted value of the prior covariance matrix of the kth iteration, I represents the identity matrix, represents the predicted value of the line dancing position vector at the kth iteration, Represents the output value of the line dancing position vector at the kth iteration.

7. A Gaussian process-based line galloping three-dimensional state analysis method as claimed in claim 6, characterized in that: The weighted representation of each Gaussian process model is: Where k represents the sequence number of the Gaussian process model, n represents the number of time points in the line dancing position vector time series data, x(t) represents the actual observation value of the line dancing position vector at time point t, Represents the predicted value of the line dancing position vector at time point t calculated by the kth Gaussian process model.

8. A system using a Gaussian process-based line galloping three-dimensional state analysis method as claimed in any one of claims 1 to 7, characterized in that: It includes data acquisition module, Gaussian process model building module, model hyperparameter acquisition module and state estimation update module; The data acquisition module is responsible for acquiring the line dancing position vector time series data; The Gaussian process model building module is responsible for building a Gaussian process model and using the model to fit the line dancing position vector time series data; The model hyperparameter acquisition module selects a covariance function for each line dancing position vector dimension and calculates the model hyperparameters; The state estimation updating module updates the state estimation of each Markov Gaussian process by the line galloping position vector to be detected.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a line galloping three-dimensional state analysis method based on Gaussian process described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a line galloping three-dimensional state analysis method based on Gaussian process described in any one of claims 1 to 7 are implemented.

Images