A method for separating environmental influences from bridge monitoring data

Abstract

The present application relates to bridge monitoring data analysis and processing technical field, specifically to a kind of bridge monitoring data's environmental influence separation method, comprising: obtaining bridge monitoring data;Bridge monitoring data is subjected to environmental influence separation, and bridge structure damage signal is obtained;Environmental influence includes live load influence, dead load influence, noise influence and temperature influence;According to bridge structure damage signal, the safety monitoring of bridge is carried out.Obtain bridge monitoring data, then bridge monitoring data is subjected to environmental influence separation, and bridge structure damage signal is obtained;According to bridge structure damage signal, the safety monitoring of bridge is carried out again.Compared with directly setting threshold to the relevant monitoring item of bridge monitoring data in prior art to carry out safety monitoring, the present application considers the influence of environment on relevant monitoring item, and the bridge structure damage signal obtained after separating environmental influence can better reflect the actual situation of bridge structure, thereby improving the accuracy of bridge safety monitoring and reducing safety hazards.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of bridge monitoring data analysis and processing technology, and particularly relates to a bridge monitoring data environment influence separation method. BACKGROUND

[0002] A bridge structure operation health monitoring system stores a large amount of bridge monitoring data, how to mine bridge structure loss information from the large amount of bridge monitoring data, so as to obtain the health evaluation of the in-service bridge and realize the monitoring and early warning of the slowly emerging safety hazards, is the main difficulty of the existing technical personnel.

[0003] In the prior art, the related monitoring items in the bridge monitoring data are usually directly used to represent the bridge structure damage information, and then a threshold is set for the related monitoring items to perform safety monitoring. However, in the actual monitoring situation, many monitoring items of the bridge structure monitoring system are affected by the monitoring environment, so that the monitoring signals of the monitoring items (settlement, deflection, strain, etc.) reflecting the real structure change of the bridge cannot directly obtain the actual situation of the bridge structure. Therefore, the method of setting a threshold for the related monitoring items to perform bridge safety monitoring in the prior art has low accuracy. SUMMARY

[0004] In view of the deficiencies in the prior art, the present application provides a bridge monitoring data environment influence separation method, which improves the accuracy of bridge safety monitoring.

[0005] The technical scheme adopted by the present application is a bridge monitoring data environment influence separation method.

[0006] In a first implementation mode, a bridge monitoring data environment influence separation method comprises: obtaining bridge monitoring data; performing environment influence separation on the bridge monitoring data to obtain a bridge structure damage signal; the environment influence includes live load influence, dead load influence, noise influence and temperature influence; and performing safety monitoring on the bridge according to the bridge structure damage signal.

[0007] In a second implementation mode, in combination with the first implementation mode, the environment influence separation on the bridge monitoring data comprises: performing load effect separation on the bridge monitoring data to obtain load separation data; performing noise separation on the load separation data to obtain an effective local feature signal; and performing temperature effect separation on the effective local feature signal to obtain the bridge structure damage signal.

[0008] In a third implementation mode, in combination with the second implementation mode, the load effect separation on the bridge monitoring data to obtain the load separation data comprises: performing high-frequency filtering on the bridge monitoring data to obtain the load separation data.

[0009] In the fourth implementation manner, the load separation data is subjected to noise separation to obtain an effective local characteristic signal, including: obtaining a noisy signal from the load separation data; the noisy signal includes an original signal and a noise signal; adding auxiliary white noise to the original signal to obtain an auxiliary original signal; decomposing the auxiliary original signal to obtain a candidate local characteristic signal; obtaining a similarity between the candidate local characteristic signal and the auxiliary original signal; and determining the candidate local characteristic signal as the effective local characteristic signal when the similarity is greater than a preset threshold.

[0010] In the fifth implementation manner, the similarity between the candidate local characteristic signal and the auxiliary original signal is obtained by the following formula in combination with the fourth implementation manner:

[0011]

[0012] In the above formula, sim(imf,x) is the similarity between the candidate local characteristic signal and the auxiliary original signal, is a feature vector of the candidate local characteristic signal, is a feature vector of the auxiliary original signal.

[0013] In the sixth implementation manner, the effective local characteristic signal is subjected to temperature effect separation to obtain a bridge structure damage signal in combination with the second implementation manner, including: obtaining an observation signal from the effective local characteristic signal; and obtaining the bridge structure damage signal from the observation signal.

[0014] In the seventh implementation manner, the bridge structure damage signal is obtained from the observation signal in combination with the sixth implementation manner, including: obtaining a mutual observation degree; obtaining a separation matrix from the mutual observation degree; and obtaining the bridge structure damage signal from the separation matrix and the observation signal.

[0015] In the eighth implementation manner, the separation matrix is obtained from the mutual observation degree in combination with the seventh implementation manner, including: obtaining a nonlinear function of the mutual observation degree from the mutual observation degree; and updating an initial separation matrix to obtain a latest separation matrix according to the mutual observation degree and the nonlinear function.

[0016] In the ninth implementation manner, the nonlinear function of the mutual observation degree is obtained from the mutual observation degree in combination with the eighth implementation manner, including: obtaining the nonlinear function of the mutual observation degree by the following formula:

[0017]

[0018] In the above formula, μ(k+1) is the nonlinear function of the mutual observation degree, D(k) is the mutual observation degree, α is a first control parameter, γ is a second control parameter, β is a third control parameter, and exp represents an exponential function with a natural constant e as the base.

[0019] In combination with the eighth implementation manner, in the tenth implementation manner, the initial separation matrix is updated according to the mutual observation degree and the nonlinear function to obtain the latest separation matrix, including: the separation matrix is updated through the following formula:

[0020] W(k+1) = W(k) + μ(k+1) [I - f(Y(k))Y T (k) + Y(k)f T (Y(k)) - Y(k)Y T (k)] W(k);

[0021] In the above formula, W(k) is the separation matrix corresponding to the data length value k, W(k+1) is the separation matrix corresponding to the data length value K+1, μ(k+1) is the nonlinear function of the mutual observation degree, I is the unit matrix, Y(k) is the bridge structure damage signal, and f(Y(k) represents solving the bridge structure damage signal function.

[0022] From the above technical solution, the beneficial technical effects of the present application are as follows:

[0023] 1. The bridge monitoring data is obtained, and then the environmental influence separation is performed on the bridge monitoring data to obtain the bridge structure damage signal; and then the safety monitoring is performed on the bridge according to the bridge structure damage signal. Compared with the safety monitoring by setting the threshold value for the related monitoring items of the bridge monitoring data in the prior art, the present application considers the influence of the environment on the related monitoring items, and the bridge structure damage signal obtained after separating the environmental influence can better reflect the actual situation of the bridge structure, thereby improving the accuracy of the bridge safety monitoring and reducing the safety hidden danger.

[0024] 2. The environmental influence includes the live load influence, the dead load influence, the noise influence and the temperature influence, and by separating the load influence, the noise influence and the temperature influence in the bridge monitoring data, the bridge structure damage signal obtained can better reflect the real change situation of the bridge structure, thereby further improving the accuracy of the bridge safety monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0025] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the specific embodiments or the prior art description will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual proportion.

[0026] Figure 1 A schematic diagram of a bridge monitoring data environmental influence separation method provided by an embodiment of the present application;

[0027] Fig. 2(a) is a diagram of the relationship between the measured temperature and strain of a bridge according to an embodiment of the present application;

[0028] Fig. 2(b) is a diagram of the relationship between the measured temperature and deflection of a bridge according to an embodiment of the present application;

[0029] Figure 3 Fig. 3 is a diagram of the settlement deformation results of a settlement monitoring point according to an embodiment of the present application. DETAILED DESCRIPTION

[0030] The embodiments of the technical solutions of the present application will be described in detail below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application, and therefore only serve as examples, and cannot limit the protection scope of the present application.

[0031] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in the present application should be understood as the usual meanings understood by the skilled person in the field of the present application.

[0032] In combination Figure 1 As shown in the drawings, the present embodiment provides a bridge monitoring data environmental influence separation method, comprising:

[0033] Step S01, obtaining bridge monitoring data;

[0034] Step S02, separating the environmental influence of the bridge monitoring data to obtain a bridge structure damage signal; the environmental influence includes live load influence, dead load influence, noise influence and temperature influence;

[0035] Step S03, performing safety monitoring on the bridge according to the bridge structure damage signal.

[0036] In some embodiments, a long-term monitoring is performed on a bridge in operation to obtain bridge monitoring data S. S = s T +s P +s L +s R , wherein s T is the temperature influence, s P is the load influence, s L is the long-term change value of the bridge structure, s R is the noise influence. The load influence includes live load influence and dead load influence; the long-term change value of the bridge structure is mainly caused by changes such as prestress loss, structural damage and concrete shrinkage and creep, and the long-term change value of the bridge structure includes the bridge structure damage signal; the noise influence is the test error of the bridge monitoring system and other random influences.

[0037] Optionally, the bridge monitoring data is subjected to environmental influence separation, comprising: load effect separation of the bridge monitoring data to obtain load separation data; noise separation of the load separation data to obtain effective local feature signals; temperature effect separation of the effective local feature signals to obtain bridge structure damage signals.

[0038] Optionally, the load effect separation of the bridge monitoring data to obtain load separation data comprises: high-frequency filtering of the bridge monitoring data to obtain the load separation data.

[0039] In some embodiments, the load effect is high in frequency, and a filter is used to filter out high-frequency signals in the bridge monitoring data to achieve load effect separation and obtain load separation data S1, S1=s T +s L +s R .

[0040] Optionally, the noise separation of the load separation data to obtain effective local feature signals comprises: obtaining a noisy signal from the load separation data; the noisy signal comprising an original signal and a noise signal; adding auxiliary white noise to the original signal to obtain an auxiliary original signal; decomposing the auxiliary original signal to obtain an alternative local feature signal; obtaining a similarity between the alternative local feature signal and the auxiliary original signal; and determining the alternative local feature signal as an effective local feature signal when the similarity is greater than a preset threshold.

[0041] In some embodiments, the noisy signal is x(t)=f(t)+n(t); wherein x(t) is the noisy signal, f(t) is the original signal, and n(t) is the noise signal. The number of original signals is 2n.

[0042] In some embodiments, n sets of positive and negative paired auxiliary white noise are added to the original signal to obtain an auxiliary original signal as follows:

[0043]

[0044] In the above formula, x is the original signal, N is the auxiliary white noise, x1 and x2 are the auxiliary original signals x' after adding positive and negative paired noise. x'=(x1,x2,…,x n ), x1 is the first auxiliary original signal, x n is the nth auxiliary original signal.

[0045] Optionally, each auxiliary original signal is decomposed to obtain each set of alternative local feature signals corresponding to each auxiliary original signal, the average value of each set of alternative local feature signals is determined, and the average value is determined as the corresponding alternative local feature signal.

[0046] In some embodiments, the EMD (Empirical Mode Decomposition) decomposition is performed on all the auxiliary original signals to obtain a plurality of groups of candidate local characteristic signals corresponding to the auxiliary original signals. The gth candidate local characteristic signal of the fth auxiliary original signal is IMF fg , where f<2n. The average value of the groups of candidate local characteristic signals is determined, and the average value is determined as the corresponding candidate local characteristic signal X. The final candidate local characteristic signal IMF=(X1,X2,…,X n , where X1 is the first local characteristic signal, X n is the nth local characteristic signal.

[0047] Optionally, the similarity between the candidate local characteristic signal and the auxiliary original signal is obtained by the following formula:

[0048]

[0049] In the above formula, sim(IMF,x′) is the similarity between the candidate local characteristic signal and the auxiliary original signal, is the feature vector of the candidate local characteristic signal, is the feature vector of the auxiliary original signal.

[0050] Optionally, the mean value, entropy value and hyper-entropy of the candidate local characteristic signal IMF are obtained, and the feature vector of the candidate local characteristic signal is obtained according to the mean value, entropy value and hyper-entropy of the candidate local characteristic signal , where is the mean value of the candidate local characteristic signal, En X is the entropy value of the candidate local characteristic signal, He X is the hyper-entropy of the candidate local characteristic signal.

[0051] Optionally, the mean value, entropy value and hyper-entropy of the auxiliary original signal x′ are obtained, and the feature vector of the auxiliary original signal is obtained according to the mean value, entropy value and hyper-entropy of the auxiliary original signal , where is the mean value of the auxiliary original signal, En x′ is the entropy value of the auxiliary original signal, He x′ is the hyper-entropy of the auxiliary original signal.

[0052] Optionally, the temperature effect separation is performed on the effective local characteristic signal to obtain a bridge structure damage signal, including: obtaining an observation signal according to the effective local characteristic signal; and obtaining the bridge structure damage signal according to the observation signal.

[0053] In some embodiments, the effective local feature signals are combined into an observation signal X(k). X(k) is an N-dimensional observation signal, X(k) = (x1(k), …, xN(k))T, k = 1, 2, …, N, where N and K are positive integers. N (k)) T , k = 1, 2, …, N, where N and K are positive integers.

[0054] Optionally, the bridge structure damage signal is obtained according to the observation signal, comprising: obtaining a mutual observation degree; obtaining a separation matrix according to the mutual observation degree; and obtaining the bridge structure damage signal according to the separation matrix and the observation signal.

[0055] Optionally, the separation matrix is obtained according to the mutual observation degree, comprising: obtaining a nonlinear function of the mutual observation degree; and updating the initial separation matrix according to the mutual observation degree and the nonlinear function to obtain the latest separation matrix.

[0056] Optionally, the bridge structure damage signal is calculated according to the initial separation matrix and the observation signal, the mutual observation degree is obtained according to the bridge structure damage signal, the nonlinear function of the mutual observation degree is obtained according to the mutual observation degree, the initial separation matrix is updated according to the mutual observation degree and the nonlinear function to obtain the latest separation matrix; the latest separation matrix is taken as the initial separation matrix and the next observation signal to recalculate the next bridge damage signal, and the above method is iterated until the Nth bridge structure damage signal is obtained.

[0057] In some embodiments, the first separation matrix is a unit matrix, the first bridge structure damage signal y1(k) is obtained according to the unit matrix and the first observation signal, the first mutual observation degree is obtained according to the first bridge structure damage signal, the first nonlinear function of the first mutual observation degree is obtained according to the first mutual observation degree, the first separation matrix is updated according to the first mutual observation degree and the first nonlinear function to obtain the second separation matrix, and the above method is iterated until the Nth bridge structure damage signal yN(k) is obtained. N (k) is obtained according to each bridge structure damage signal. Y(k) = (y1(k), …, yN(k))T. N (k)) T , where y1(k) is the first bridge structure damage signal, yN(k) is the Nth bridge structure damage signal. N

[0058] Optionally, the bridge structure damage signal is obtained according to the separation matrix and the observation signal, comprising: the bridge structure damage signal is obtained by the following formula:

[0059] Y(k) = WX(k);

[0060] In the above formula, Y(k) is the bridge structure damage signal, Y(k) = (y1(k), …, yN(k))T.​N (k)) T , W is a separation matrix, and X(k) is an observation signal.

[0061] Optionally, the cross observation degree is obtained by the following formula:

[0062]

[0063]

[0064]

[0065] In the above formula, D(k) is the cross observation degree, sc ij (k) is a second-order correlation coefficient, hc ji (k) is a high-order correlation coefficient, k represents a data length value, cov represents a cross-correlation coefficient, cov[y i (k), y j (k)] is a cross-correlation coefficient between the i-th bridge structure damage signal y i (k) and the j-th bridge structure damage signal y j (k), var represents a variance between signals, var[y i (k)] and var[y j (k)] are variances between the i-th bridge structure damage signal y i and the j-th bridge structure damage signal y j , and φ() is a nonlinear function.

[0066] Optionally, the nonlinear function of the cross observation degree is obtained according to the cross observation degree, and the nonlinear function of the cross observation degree comprises: the nonlinear function of the cross observation degree is obtained by the following formula:

[0067]

[0068] In the above formula, μ(k+1) is the nonlinear function of the cross observation degree, D(k) is the cross observation degree, α is a first control parameter, γ is a second control parameter, β is a third control parameter, and exp represents an exponential function with a natural constant e as a base. In some embodiments, α=50 and β=0.005. The nonlinear function is controlled by the first control parameter and the second control parameter, and the value range of the nonlinear function is controlled by the third control parameter. By controlling the size of the nonlinear function of the cross observation degree, the adaptive step of iterative solving is controlled, so that the separation matrix is better converged, and the separated bridge structure damage signal is more accurate.

[0069] Optionally, the initial separation matrix is updated according to the mutual observation degree and a nonlinear function of the mutual observation degree to obtain the latest separation matrix, including: updating the initial separation matrix by the following formula:

[0070] W(k+1) = W(k) + μ(k+1) [I - f(Y(k))Y T (k) + Y(k)f T (Y(k)) - Y(k)Y T (k)]W(k)

[0071] In the above formula, W(k) is the initial separation matrix corresponding to the data length value k, W(k+1) is the latest separation matrix corresponding to the data length value K+1, μ(k+1) is a nonlinear function of the mutual observation degree, I is a unit matrix, Y(k) is a bridge structure damage signal, f(Y(k) represents solving a bridge structure damage signal function, (·) T represents transposition.

[0072] Optionally, the bridge is safety monitored according to the bridge structure damage signal, including: statistically analyzing the bridge structure damage signal to obtain a bridge health state evaluation; and performing a safety alarm in a case where the bridge health state evaluation is lower than a preset safety value.

[0073] Optionally, the safety alarm includes: issuing a preset alarm signal for prompting a bridge safety hidden danger.

[0074] In combination with FIG. 2, in some embodiments, FIG. 2(a) is measured data of a strain measuring point of a certain bridge in Chongqing from January 14, 2022 to March 31, 2022, and the dashed line at the bottom of the figure represents strain, and the solid line at the top represents temperature. FIG. 2(b) is measured data of a settlement measuring point of a certain bridge in Chongqing from September 28, 2021 to March 31, 2022, and the dashed line at the top of the figure represents temperature, and the solid line at the bottom represents deflection. As can be seen from FIG. 2(a) and FIG. 2(b), in the whole monitoring period, the strain value and the deflection value of the measuring point fluctuate with temperature fluctuations, and there is a strong correlation between the two. At the same time, in addition to the influence of temperature, the bridge structure monitoring system is also affected by traffic load and many other random factors.

[0075] In some embodiments, from October 14, 2021 to March 14, 2022, a settlement deformation measuring point of a certain bridge in Chongqing is monitored to obtain measured values of the settlement deformation measuring point of the certain bridge from October 14, 2021 to March 14, 2022 (as shown by the dashed line at the bottom of FIG. 3(a), Figure 3 Figure 3 ​The horizontal axis represents time, and the vertical axis represents deflection. The curve fluctuation range of the measured values ​​is close to 15 mm. Using the environmental impact separation method for bridge monitoring data provided in this embodiment, load effect separation, noise separation, and temperature effect separation are performed on the bridge monitoring data to obtain the values ​​after separating the environmental impact (e.g., ...). Figure 3 The solid line at the top center shows a fluctuation range of only 5mm. Compared to the measured value, the fluctuation range is much lower, and the monitoring curve is relatively stable. Further analysis of the value after separating environmental influences, i.e., the bridge structural damage signal, reveals that the dead load settlement deformation at this monitoring point was approximately 0.72mm during the 152-day monitoring period. Considering calculation errors and other factors, the bridge's health status can be assessed as unchanged in its structural operational state. On-site investigation confirmed that the actual condition of the bridge matched the analysis results, verifying that the bridge structural damage signal obtained through the environmental influence separation method of bridge monitoring data is more accurate and better characterizes the actual condition of the bridge structure, thus validating the effectiveness and accuracy of the method.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for separating the environmental impact of bridge monitoring data, characterized in that, include: Obtain bridge monitoring data; The bridge monitoring data is subjected to environmental impact separation to obtain bridge structural damage signals; the environmental impacts include live load impact, dead load impact, noise impact, and temperature impact. Safety monitoring of the bridge is carried out based on the bridge structural damage signals. The environmental impact separation of the bridge monitoring data includes: The bridge monitoring data is subjected to load effect separation to obtain load separation data; The load separation data is subjected to noise separation to obtain effective local feature signals; Temperature effect separation is performed on the effective local feature signals to obtain bridge structural damage signals, including: The observation signal is obtained based on the effective local feature signal; Based on the observed signals, bridge structural damage signals are obtained, including: Obtain mutual observability; The separation matrix is ​​obtained based on the mutual observability. Bridge structural damage signals are obtained based on the separation matrix and the observed signals; The step of obtaining the separation matrix based on the mutual observability includes: The nonlinear function of the mutual observability is obtained based on the mutual observability. The initial separation matrix is ​​updated based on the mutual observability and the nonlinear function to obtain the latest separation matrix; Wherein, the nonlinear function for obtaining the mutual observability based on the mutual observability includes: The nonlinear function of mutual observability is obtained using the following formula: ； In the above formula, For the nonlinear function of mutual observability, For mutual observation degree, The first control parameter, This is the second control parameter. The third control parameter is exp, which represents an exponential function with the natural constant e as its base.

2. The method according to claim 1, characterized in that, The bridge monitoring data is subjected to load effect separation to obtain load separation data, including: The bridge monitoring data is subjected to high-frequency filtering to obtain load separation data.

3. The method according to claim 1, characterized in that, The load separation data is subjected to noise separation to obtain effective local feature signals, including: The noisy signal is obtained based on the load separation data; the noisy signal includes the original signal and the noise signal. Add auxiliary white noise to the original signal to obtain an auxiliary original signal; The auxiliary original signal is decomposed to obtain candidate local feature signals; Obtain the similarity between the candidate local feature signals and the auxiliary original signal; If the similarity is greater than a preset threshold, the candidate local feature signal is determined to be a valid local feature signal.

4. The method according to claim 3, characterized in that, The similarity between the candidate local feature signal and the auxiliary original signal is obtained using the following formula: In the above formula, The similarity between the candidate local feature signals and the auxiliary original signal is used as a reference. The feature vector of the candidate local feature signal, This is the feature vector that assists the original signal.

5. The method according to claim 1, characterized in that, The initial separation matrix is ​​updated based on the mutual observability and the nonlinear function to obtain the latest separation matrix, including: The initial separation matrix is ​​updated using the following formula: ； In the above formula, This is the separation matrix corresponding to the data length value k. This is the separation matrix corresponding to the data length value k+1. For the nonlinear function of mutual observability, It is the identity matrix. This is a signal of bridge structural damage. This represents the function for solving the bridge structure damage signal. (·) T This indicates transpose.

Images