A GNSS and acceleration fusion method for deformation monitoring of arch bridges in complex environments

Abstract

The application discloses a GNSS and acceleration fusion method for arch bridge deformation monitoring in a complex environment and relates to the field of bridge engineering. The method comprises the following steps: acquiring acceleration and GNSS original data for arch bridge deformation monitoring; performing arch bridge vertical deformation calculation to obtain arch bridge deformation data; acquiring arch bridge deformation frequency based on the obtained arch bridge deformation data, establishing an arch bridge finite element model to determine the critical frequency of arch bridge dynamic deformation; establishing a GNSS and acceleration fusion method based on the critical frequency of arch bridge dynamic deformation, calculating the arch bridge deformation frequency, and judging whether the arch bridge deformation frequency is greater than the critical frequency; when the arch bridge deformation frequency is greater than the critical frequency, a G-type fusion method is selected; and when the arch bridge deformation frequency is less than the critical frequency, a D-type fusion method is selected. The GNSS original data collected is subjected to orthogonal double-star system calculation, the influence of GNSS signal shielding is reduced, and one-step solution of arch bridge deformation in a complex environment is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bridge engineering, and more particularly to a GNSS and acceleration fusion method for arch bridge deformation monitoring in complex environments. BACKGROUND

[0002] Arch bridge structures are mostly located in dangerous mountainous areas, and the bridge site environment is complex. Traditional deformation monitoring methods cannot meet the health monitoring needs. GNSS (Global Navigation Satellite System) and acceleration fusion are usually used to detect structural deformation, but in complex deformation monitoring environments, GNSS signals are easily blocked by vegetation and other obstructions, which interferes with the accuracy of GNSS calculation. Existing fusion algorithms mainly target large buildings and flexible structures such as cable-supported bridges. For arch bridges with large stiffness and small deformation amplitude, the monitoring needs cannot be met, and there are limitations in real-time measurement of arch bridge dynamic accuracy in the vertical direction.

[0003] Currently, the GNSS deformation monitoring method first calculates the coordinate deformation of the monitoring station, and then calculates the deformation value of the bridge structure. The GNSS and acceleration data fusion method mainly uses Kalman filtering, multi-rate Kalman filtering and particle filtering to fuse the two types of data, with an accuracy of centimeters, which still cannot meet the needs of arch bridge deformation monitoring. In the prior art, the following problems exist:

[0004] 1. The existing GNSS deformation calculation method limits the accuracy of arch bridge dynamic deformation calculation. Traditional calculation methods introduce coordinate conversion, phase distance deformation monitoring model linearization errors and other factors;

[0005] 2. The existing fusion method does not consider the deformation characteristics of the arch bridge, and the fusion method is difficult to calculate the true deformation of the arch bridge.

[0006] Therefore, to solve the problems in the prior art, a GNSS and acceleration fusion method for arch bridge deformation monitoring in complex environments is proposed, which is a problem that needs to be solved by those skilled in the art. SUMMARY

[0007] Therefore, the present application provides a GNSS and acceleration fusion method for arch bridge deformation monitoring in complex environments to solve the technical problems existing in the prior art.

[0008] To achieve the above purpose, the present application provides the following technical solutions:

[0009] A GNSS and acceleration fusion method for arch bridge deformation monitoring in complex environments, comprising the following steps:

[0010] S1: Obtain acceleration and GNSS raw data for arch bridge deformation monitoring;

[0011] S2: adopt the orthogonal double star system to solve the arch bridge vertical deformation of the GNSS raw data, and obtain the arch bridge deformation data based on the GNSS solution;

[0012] S3: obtain the arch bridge deformation frequency based on the obtained arch bridge deformation data, establish the arch bridge finite element model to determine the critical frequency of the arch bridge dynamic deformation;

[0013] S4: based on the critical frequency of the arch bridge dynamic deformation, establish the GNSS and acceleration fusion method, calculate the actual arch bridge deformation frequency, and judge whether the actual arch bridge deformation frequency is greater than the critical frequency, when the actual arch bridge deformation frequency is greater than the critical frequency, select the G type fusion method, when the actual arch bridge deformation frequency is less than the critical frequency, select the D type fusion method.

[0014] The above method, optionally, the specific content of S2 is:

[0015] S201: select observation zenith satellite s ⊥ And horizontal direction satellite s, establish the arch bridge deformation solution system based on GNSS;

[0016] S202: s ⊥ Difference between the two satellites eliminates the error of the monitoring station receiver, and the error of the satellite receiver is eliminated;

[0017] S203: repair and process the carrier phase data collected by GNSS on the arch bridge, and perform quadratic polynomial fitting on the data without abnormality in a short time domain window:

[0018] S(t)=a×(t-t0) 2 +b×(t-t0)+c

[0019] Wherein, S(t) is the carrier phase observation value at t time, t0 is the start point of the short time domain, a, b and c are fitting parameters;

[0020] S204: establish the coefficient matrix X about a, b and c:

[0021] X=(A T A) -1 (A T L)

[0022] Wherein, L is the carrier phase observation matrix without abnormality, A is the time matrix, A T Is the transpose matrix of A;

[0023] S205: substitute the first m ephemeris observation values, and establish the equation about error matrix V:

[0024] V=L-A×X;

[0025] S206: Calculate the average fitting error σ:

[0026]

[0027] Among them, V i is the error matrix at the i-th moment;

[0028] S207: Use the coefficient matrix to solve the carrier phase observation value at the next moment

[0029] S208: Repair abnormal data:

[0030] y' i =y i-1 +V i-1 ;

[0031] Among them, y' i is the observed value after repair, V i-1 is the fitting error of the previous moment when no abnormal data occurred, y i-1 is the carrier phase observation value at the i-1th moment;

[0032] S209: Establish the double-difference observation equation of the carrier phase:

[0033]

[0034] Among them, p and q are the ID numbers of the reference satellite and the measurement satellite respectively, k and m are the ID numbers of the reference station and the mobile station respectively, and λ is the wavelength. is the double-difference carrier phase observation, is the double difference geometric distance, S is the double difference ambiguity distance, is the double difference phase residual value, N is the random noise error;

[0035] S2010: Using polynomial functions to calculate the differential geometric distance of orthogonal binary star systems in a short time domain window:

[0036]

[0037] Among them, a n 、a n-1 , a0 is the polynomial coefficient, is the double difference geometric distance;

[0038] S2011: Solve the phase residual value R(t):

[0039]

[0040] S2012: Obtain the deformation data of the arch bridge at each moment based on GNSS solution through the relationship between phase and coordinates.

[0041] The method, and optionally, a specific content of S3 is: establishing a finite element model for monitoring the arch bridge, and extracting a fundamental frequency of the arch bridge as a critical frequency of dynamic deformation of the arch bridge.

[0042] The method, and optionally, a specific content of the G-type fusion method in S4 is:

[0043] S41: calculating filtering parameters according to sample data;

[0044] S42: performing low-frequency filtering on arch bridge deformation data based on GNSS solution, and performing adjacent data interpolation to obtain low-frequency data of GNSS deformation monitoring;

[0045] S43: reconstructing high-frequency data of acceleration deformation monitoring by using an acceleration-based moving window time-domain integral reconstruction dynamic displacement method;

[0046] S44: adding the low-frequency data and the high-frequency data to obtain arch bridge displacement.

[0047] The method, and optionally, a specific content of S43 is:

[0048] S431: reading acceleration data and performing high-pass filtering to remove low-frequency noise;

[0049] S432: determining a window length and a sliding window step;

[0050] S433: performing twice integral on acceleration data in each window, the first integral obtaining velocity, and the second integral obtaining displacement;

[0051] S434: performing polynomial fitting on displacement data, and subtracting a fitting result;

[0052] S435: splicing reconstructed velocity and displacement in each window to form a complete reconstruction result.

[0053] The method, and optionally, in S432, the longest vibration period of the arch bridge is set as the window length.

[0054] The method, and optionally, a specific content of the D-type fusion method in S4 is:

[0055] S45: calculating filtering parameters according to sample data;

[0056] S46: performing high-pass filtering on acceleration data;

[0057] S47: performing GNSS and acceleration fusion displacement detection by using a forward multi-rate Kalman filter to obtain arch bridge displacement.

[0058] The method, and optionally, a specific content of S47 is:

[0059] S471: initialize state vector and covariance matrix;

[0060] S472: perform state and covariance prediction update for each time step k, with the following formulas:

[0061]

[0062] where, is the state prediction at time k, i.e. the prior estimate, F k is the state transition matrix, is the state estimate at time k-1, i.e. the posterior estimate, B k is the control input matrix, u k is the external control applied to the system at time k, P k|k-1 is the prediction covariance matrix at time k, i.e. the prior covariance, P k-1|k-1 is the posterior covariance matrix at time k-1, Q k is the process noise covariance matrix, F k T is the transpose of the state transition matrix;

[0063] S473: update state vector and covariance matrix from GNSS and acceleration sensor data;

[0064] When GNSS data arrives, update the formulas:

[0065] K k = P k|k-1 H T (HP k|k-1 H T + R GNSS ) -1

[0066]

[0067] P k|k = (I - K K H)P k|k-1

[0068] where K k is the Kalman gain matrix, R GNSS is the GNSS measurement noise covariance matrix, H is the observation matrix, H T is the transpose of the observation matrix, P k|k-1 is the prediction covariance matrix at time k, z GNSS,k is the GNSS measurement at time k, is the state prediction at time k, i.e. the prior estimate, I is the identity matrix, is the updated state vector at time k, P k|k is the updated covariance matrix at time k;

[0069] When the acceleration data arrives, the update formula is:

[0070] K k = P k|k-1 H T (HP k|k-1 H T +R ACC ) -1

[0071]

[0072] P k|k = (I-K K H)P k|k-1

[0073] wherein K k is the Kalman gain matrix, P k|k-1 is the predicted covariance matrix at time k, H is the observation matrix, H T is the transpose matrix of the observation matrix, is the prediction of the state at time k, i.e., the prior estimate, R ACC is the acceleration measurement noise covariance matrix, z ACC,k is the acceleration measurement value at time k, I is the unit matrix, is the updated state vector at time k, P k|k is the updated covariance matrix at time k;

[0074] S474: repeat S472 and S473 until all data processing is completed;

[0075] S475: obtain the displacement of the arch bridge.

[0076] The method described above, optionally, the initialization state vector and covariance matrix in S471 include the initial state estimate and the initial covariance matrix P0.

[0077] The method described above, optionally, further comprises: S5: arranging the deformation data obtained at different frequencies in time to obtain the arch bridge deformation information in the overall time domain.

[0078] According to the technical solution described above, compared with the prior art, the application provides a GNSS and acceleration fusion method for arch bridge deformation monitoring in a complex environment, which has the following beneficial effects:

[0079] 1) The collected GNSS raw data is solved by an orthogonal double star system to reduce the influence of GNSS signal shielding and realize one-step solution of arch bridge deformation in a complex environment.

[0080] 2) According to the different real-time deformation frequencies of the arch bridge, a G-type and D-type fusion method is established, the deformation characteristics of the arch bridge are considered, and the purpose of solving the accurate deformation of the arch bridge is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0081] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of the provided drawings.

[0082] Figure 1 A flowchart of a GNSS and acceleration fusion method for arch bridge deformation monitoring in a complex environment provided by the present application;

[0083] Figure 2 A flowchart of a GNSS and acceleration fusion method for arch bridge deformation monitoring in a complex environment provided by the present application;

[0084] Figure 3 A schematic diagram of an orthogonal double star system provided by the present application;

[0085] Figure 4 A flowchart of GNSS orthogonal double star system solution deformation provided by the present application;

[0086] Figure 5 A flowchart of a mobile window-based time-domain integral reconstruction dynamic displacement method provided by the present application;

[0087] Figure 6 A Kalman filter flowchart provided by the present application;

[0088] Figure 7 A forward multi-rate Kalman filter schematic diagram provided by the present application. DETAILED DESCRIPTION

[0089] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0090] Reference Figure 1As shown, the application discloses a GNSS and acceleration fusion method for arch bridge deformation monitoring in complex environment, and comprises the following steps:

[0091] S1: acquiring acceleration and GNSS original data for arch bridge deformation monitoring;

[0092] S2: using a quadruple satellite system to solve arch bridge vertical deformation of the GNSS original data, and obtaining arch bridge deformation data based on GNSS solution;

[0093] S3: obtaining arch bridge deformation frequency based on the obtained arch bridge deformation data, establishing an arch bridge finite element model to determine the critical frequency of arch bridge dynamic deformation;

[0094] S4: establishing a GNSS and acceleration fusion method based on the critical frequency of arch bridge dynamic deformation, calculating the actual arch bridge deformation frequency, and judging whether the actual arch bridge deformation frequency is greater than the critical frequency; when the actual arch bridge deformation frequency is greater than the critical frequency, a G-type fusion method is selected; and when the actual arch bridge deformation frequency is less than the critical frequency, a D-type fusion method is selected.

[0095] Further, referring to Figure 4 As shown, the specific content of S2 is:

[0096] S201: selecting observation zenith satellites s ⊥ and horizontal direction satellites s, and establishing a GNSS-based arch bridge deformation solution system;

[0097] S202: s ⊥ and s two satellites are subtracted to eliminate the error of the monitoring station receiver end, and the satellite end error is eliminated by subtracting the monitoring station receiver end;

[0098] S203: repairing and processing the carrier phase data collected by the GNSS on the arch bridge, and performing quadratic polynomial fitting on the data without abnormality in a short time domain window:

[0099] S(t) = a x (t-t0) 2 + b x (t-t0) + c

[0100] Wherein, S(t) is the carrier phase observation value at t time, t0 is the start point of the short time domain, a, b and c are fitting parameters;

[0101] S204: establishing a coefficient matrix X about a, b and c:

[0102] X = (A T A) -1 (A T L)

[0103] Wherein, L is the carrier phase observation matrix without abnormality, A is a time matrix, and AT is the transpose matrix of A;

[0104] S205: Substitute the first m epoch observation values to establish the equation for solving error matrix V:

[0105] V = L - A x X;

[0106] S206: Solve the average error σ of fitting:

[0107]

[0108] wherein, V i is the error matrix of the i th moment;

[0109] S207: Use the coefficient matrix to solve the carrier phase observation value of the next moment

[0110] Specifically, when it is considered that the observation data is abnormal.

[0111] S208: Repair abnormal data:

[0112] y' i = y i-1 + V i-1 ;

[0113] wherein, y' i is the repaired observation value, V i-1 is the fitting error of the previous moment without abnormal data, y i-1 is the carrier phase observation value of the i-1 th moment;

[0114] S209: Establish the double difference observation equation of carrier phase:

[0115]

[0116] wherein, p and q are the ID numbers of the reference satellite and the measured satellite respectively, k and m are the ID numbers of the reference station and the mobile station respectively, λ is the wavelength, is the double difference carrier phase observation, is the double difference geometric distance, S is the double difference ambiguity distance, is the double difference phase residual value, N is the random noise error;

[0117] Specifically, as shown in Figure 3 is the schematic diagram of the orthogonal double star system.

[0118] S2010: In the short time domain window, the differential geometric distance of the orthogonal double star system is calculated by using the polynomial function:

[0119]

[0120] wherein a n , a n-1 , a0is a polynomial coefficient, is a double-difference geometric distance;

[0121] S2011: solving the phase residual value R(t):

[0122]

[0123] S2012: obtaining the arch bridge deformation data based on GNSS solution at each moment by the relationship between phase and coordinates.

[0124] Further, the specific content of S3 is to establish a finite element model for monitoring the arch bridge, and extract the fundamental frequency of the arch bridge as the critical frequency of the dynamic deformation of the arch bridge.

[0125] Further, the specific content of the G-type fusion method in S4 is:

[0126] S41: calculating filtering parameters according to sample data;

[0127] S42: performing low-frequency filtering on the arch bridge deformation data based on GNSS solution, and performing adjacent data interpolation to obtain low-frequency data of GNSS deformation monitoring;

[0128] S43: reconstructing acceleration deformation monitoring high-frequency data by acceleration based on mobile windowing time domain integral reconstruction dynamic displacement method;

[0129] S44: adding the low-frequency data and the high-frequency data to obtain the displacement of the arch bridge.

[0130] Further, referring to Figure 5 , the specific content of S43 is:

[0131] S431: reading acceleration data and performing high-pass filtering to remove low-frequency noise;

[0132] S432: determining window length and sliding window step;

[0133] S433: performing twice integration on the acceleration data in each window, the first integration obtaining velocity, and the second integration obtaining displacement;

[0134] S434: performing polynomial fitting on the displacement data and subtracting the fitting result;

[0135] Specifically, the displacement data is polynomial fitted, and the fitting result is subtracted to eliminate the influence of baseline drift.

[0136] S435: concatenating the reconstructed velocity and displacement in each window to form a complete reconstruction result.

[0137] Furthermore, in S432, the longest vibration period of the arch bridge is set as the window length.

[0138] Furthermore, the specific content of the D-type fusion method in S4 is:

[0139] S45: Calculate filtering parameters according to sample data;

[0140] S46: performing high-pass filtering on the acceleration data;

[0141] S47: Use forward multi-rate Kalman filtering to perform GNSS and acceleration fusion displacement detection to obtain the arch bridge displacement.

[0142] For details, see Figure 7 Figure 2 shows the schematic diagram of the forward multi-rate Kalman filter. The sampling frequencies of acceleration and GNSS are different, resulting in time intervals when no GNSS displacement measurements are available. When there are no GNSS displacement measurements, the Kalman filter only performs the prediction process and does not perform measurement updates. When both GNSS displacement and acceleration measurements are present, the full Kalman filter process is completed at that moment.

[0143] For further information, see Figure 6 As shown, the specific content of S47 is:

[0144] S471: Initialize the state vector and covariance matrix;

[0145] Specifically, a state vector with added acceleration is adopted to capture more complex dynamic characteristics.

[0146] S472: Update the state and covariance prediction for each time step k. The formula is as follows:

[0147]

[0148] P k|k-1 =F k P k-1|k-1 F k T +Q k

[0149] in, is the prediction of the state at time k, that is, the prior estimate, F k is the state transition matrix, is the state estimate at time k-1, that is, the posterior estimate, B k is the control input matrix, u k is the external control quantity applied to the system at time k, P k|k-1 is the predicted covariance matrix at time k, that is, the prior covariance, P k-1|k-1is the posterior covariance matrix at time k-1, Q k is the process noise covariance matrix, F k T is the transposed matrix of the state transfer matrix;

[0150] S473: Update the state vector and covariance matrix according to the data of the GNSS and acceleration sensors;

[0151] When GNSS data arrives, update the formula:

[0152] K k =P k|k-1 H T (HP k|k-1 H T +R GNSS ) -1

[0153]

[0154] P k|k =(IK K H)P k|k-1

[0155] Among them, K k is the Kalman gain matrix, R GNSS is the GNSS measurement noise covariance matrix, H is the observation matrix, and H T is the transposed matrix of the observation matrix, P k|k-1 is the prediction covariance matrix at time k, z GNSS,k is the GNSS measurement value at time k, is the prediction of the state at time k, i.e. the prior estimate, I is the unit matrix, is the updated state vector at time k, P k|k Updated covariance matrix at time k;

[0156] When acceleration data arrives, update the formula:

[0157] K k =P k|k-1 H T (HP k|k-1 H T +R ACC ) -1

[0158]

[0159] P k|k =(IK K H)P k|k-1

[0160] Among them, Kk is the Kalman gain matrix, P k|k-1 is the prediction covariance matrix at time k, H is the observation matrix, H T is the transpose matrix of the observation matrix, is the prediction of the state at time k, i.e., the prior estimate, R ACC is the acceleration measurement noise covariance matrix, z ACC,k is the acceleration measurement value at time k, I is the identity matrix, is the updated state vector at time k, P k|k is the updated covariance matrix at time k;

[0161] Specifically, the GNSS data usually has a low sampling rate, and the acceleration data has a high sampling rate, and multi-rate processing is adopted.

[0162] S474: repeat S472 and S473 until all data processing is completed;

[0163] S475: obtain the arch bridge displacement.

[0164] Further, the initialized state vector and covariance matrix in S471 include an initial state estimate and an initial covariance matrix P0.

[0165] Referring to Figure 2 , it is a whole flowchart of the GNSS and acceleration fusion method for arch bridge deformation monitoring in a complex environment provided by the application.

[0166] Further, it further includes: S5: arranging the deformation data obtained at different frequencies in time to obtain arch bridge deformation information in the whole time domain.

[0167] Each of the embodiments in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same or similar parts between each embodiment can be referred to each other.

[0168] The above description of the disclosed embodiments enables a person skilled in the art to implement or use the application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the application. Therefore, the application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1.A GNSS and acceleration fusion method for deformation monitoring of an arch bridge in a complex environment, characterized in that, The method comprises the following steps: S1: obtaining acceleration and GNSS original data of arch bridge deformation monitoring; S2: using a quadruple system to calculate the vertical deformation of the arch bridge based on the GNSS original data, and obtaining arch bridge deformation data based on GNSS calculation; S3: obtaining arch bridge deformation frequency based on the obtained arch bridge deformation data, establishing an arch bridge finite element model, and determining the critical frequency of arch bridge dynamic deformation; S4: establishing a GNSS and acceleration fusion method based on the critical frequency of arch bridge dynamic deformation, calculating the actual arch bridge deformation frequency, and judging whether the actual arch bridge deformation frequency is greater than the critical frequency; when the actual arch bridge deformation frequency is greater than the critical frequency, a G-type fusion method is selected; and when the actual arch bridge deformation frequency is less than the critical frequency, a D-type fusion method is selected; The specific content of the D-type fusion method in S4 is: S45: calculating filtering parameters according to sample data; S46: performing high-pass filtering on acceleration data; S47: performing GNSS and acceleration fusion displacement detection by using forward multi-rate Kalman filtering to obtain arch bridge displacement; The specific content of S47 is: S471: initializing a state vector and a covariance matrix; S472: Perform state and covariance prediction update for each time step k S474: Perform state and covariance correction update for each time step in, For the moment k The prediction of the state, i.e. the prior estimate, F k is the state transition matrix, For the moment k -1 state estimate, i.e., posterior estimate, B k is the control input matrix, u k For the moment k The amount of external control applied to the system, P k|k-1 For the moment k The predicted covariance matrix of , that is, the prior covariance, P k-1|k-1 For the moment k -1 posterior covariance matrix, Q k is the process noise covariance matrix, F k T is the transposed matrix of the state transfer matrix; S473: updating the state vector and the covariance matrix according to GNSS and acceleration sensor data; When GNSS data arrives, the update formula is: in, K k is the Kalman gain matrix, R GNSS is the GNSS measurement noise covariance matrix, H is the observation matrix, H T is the transposed matrix of the observation matrix, P k|k-1 For the moment k The prediction covariance matrix of For the moment k GNSS measurements, For the moment k The prediction of the state, i.e. the prior estimate, is the identity matrix, For the moment k The updated state vector, At the moment k The updated covariance matrix of ; When acceleration data arrives, the update formula is: in, K k is the Kalman gain matrix, P k|k-1 For the moment k The prediction covariance matrix of H is the observation matrix, H T is the transposed matrix of the observation matrix, For the moment k The prediction of the state, i.e. the prior estimate, is the acceleration measurement noise covariance matrix, For the moment k The acceleration measurement value, is the identity matrix, For the moment k The updated state vector, For the moment k The updated covariance matrix of ; S474: repeating S472 and S473 until all data processing is completed; S475: obtaining arch bridge displacement; The specific content of the G-type fusion method in S4 is: S41: calculating filtering parameters according to sample data; S42: performing low-frequency filtering on the arch bridge deformation data based on GNSS calculation, and performing adjacent data interpolation to obtain low-frequency data of GNSS deformation monitoring; S43: reconstructing dynamic displacement by using a moving window-based time domain integration method to obtain high-frequency data of acceleration deformation monitoring; S44: adding the low-frequency data and the high-frequency data to obtain arch bridge displacement; The specific content of S43 is: S431: reading acceleration data and performing high-pass filtering to remove low-frequency noise; S432: determining window length and sliding window step; S433: performing twice integration on acceleration data in each window, the first integration obtaining velocity, and the second integration obtaining displacement; S434: performing polynomial fitting on displacement data, and subtracting the fitting result; S435: splicing the reconstructed velocity and displacement in each window to form a complete reconstruction result. 2.The GNSS and acceleration fusion method for deformation monitoring of an arch bridge in a complex environment according to claim 1, characterized in that, The specific content of S2 is: S201: Selecting an observation zenith satellite s ⊥ and a horizontal direction satellite s, and establishing a GNSS-based arch bridge deformation calculation system; S202: s ⊥ Difference between the two satellites to eliminate the monitoring station receiver end error, monitoring station receiver end difference elimination satellite end error; S203: repairing and processing carrier phase data collected by GNSS on the arch bridge, and performing twice polynomial fitting on data without abnormality in a short time domain window: wherein, is the carrier phase observation at time t, is the start of the short time domain, a , b and c is the fitting parameter; S204: Establishing a coefficient matrix regarding a , b and c X :​ wherein, L is the carrier phase observation matrix for which no anomaly has occurred, A is the time matrix, A T is the transpose matrix of A . S205: Substitution before m of the epoch observation, establish the equation V Solve the equation: ； S206: Solve the average error of the fitting : wherein is the i error matrix at the kth time instant; S207: Solving the carrier phase observation value of the next time by using the coefficient matrix ; S208: repairing abnormal data: ； wherein, is the observed value after repair, is the fitting error of the previous time when no abnormal data occurs, is the carrier phase observation value at the i -1 time. S209: establishing a double-difference observation equation of carrier phase: ； wherein, p , q are ID numbers of reference satellites and measurement satellites, respectively, k , m are ID numbers of reference stations and rover stations, respectively, λ is a wavelength, is a double-difference carrier phase observation, is a double-difference geometric distance, S is a double-difference ambiguity distance, is a double-difference phase residual value, N is a random noise error; S2010: calculating a quadruple system differential geometric distance under a short time domain window by using a polynomial function: wherein , , are polynomial coefficients, is a double-difference geometric range. S2011: Solving phase residual value R(t) : ； S2012: obtaining arch bridge deformation data based on GNSS calculation at each time through the relationship between phase and coordinates. 3.The GNSS and acceleration fusion method for deformation monitoring of an arch bridge in a complex environment according to claim 1, characterized in that, The specific content of S3 is: establishing a finite element model of the monitored arch bridge, and extracting the fundamental frequency of the arch bridge as the critical frequency of arch bridge dynamic deformation. 4.The GNSS and acceleration fusion method for deformation monitoring of an arch bridge in a complex environment according to claim 1, characterized in that, The longest vibration period of the arch bridge is set as the window length in S432. 5.The GNSS and acceleration fusion method for deformation monitoring of an arch bridge in a complex environment according to claim 1, characterized in that, The initial state vector and covariance matrix in S471 include the initial state estimate and initial covariance matrix P 0. 6.The GNSS and acceleration fusion method for deformation monitoring of arch bridges in complex environments according to claim 1, characterized in that, Also included are: S5: arranging the deformation data obtained at different frequencies in time to obtain arch bridge deformation information in the overall time domain.

Images