Bridge state comprehensive evaluation method based on multi-source data fusion

Abstract

The invention provides a bridge state comprehensive evaluation method based on multi-source data fusion, which is suitable for the field of bridge health evaluation and comprises the steps of selecting multi-source data indexes, determining multi-source data evaluation scales, analyzing multi-source data, distributing weights, calculating comprehensive scores and evaluating bridge states. The method has the advantages of high accuracy, good predictability and the like, and can be widely applied to the field of bridge technical condition evaluation.

Description

Technical Field

[0001] The invention relates to a comprehensive bridge status evaluation method based on multi-source data fusion, which is applicable to the field of bridge technical status evaluation. Background Art

[0002] As an important transportation infrastructure, the safety status of bridges is directly related to the safety of people's lives and property and the normal operation of social and economic activities. As time goes by and the load increases, bridge structures will gradually experience problems such as aging and damage. Therefore, real-time monitoring and comprehensive evaluation of bridge status are particularly important. Traditional bridge inspection methods often rely on manual regular inspections, which have problems such as high cost, low efficiency, and discontinuous data.

[0003] In recent years, with the development of sensor technology, Internet of Things technology, big data technology and artificial intelligence technology, the comprehensive evaluation method of bridge status based on multi-source data fusion has gradually become a research hotspot. This method can achieve comprehensive monitoring of bridge status by integrating data from different sensors, such as stress, displacement, vibration, etc., as well as environmental factors such as temperature and humidity. The bridge status is evaluated by monitoring the bridge status. The bridge status evaluation result is an important task of bridge maintenance and a necessary basis for bridge maintenance decision-making. At present, the widely used method is the hierarchical comprehensive evaluation method of the "Highway Bridge Technical Condition Evaluation Standard" (JTG / T H21-2011). Since the bridge monitoring data is a real-time changing process, this method does not take this into account. Therefore, there is an urgent need for a comprehensive evaluation method of bridge status based on multi-source data fusion that can consider the temporal change characteristics of monitoring data. Summary of the invention

[0004] The purpose of the present invention is to solve the current urgent need for a comprehensive bridge status evaluation method based on multi-source data fusion that can take into account the temporal variation characteristics of monitoring data. Therefore, a comprehensive bridge status evaluation method based on multi-source data fusion is proposed.

[0005] The purpose of the present invention can be achieved by adopting the following technical solutions:

[0006] A comprehensive evaluation method for bridge status based on multi-source data fusion, characterized in that the steps of the comprehensive evaluation method for bridge status based on multi-source data fusion are as follows:

[0007] S101, select multi-source data indicators;

[0008] The selection of multi-source data indicators includes selecting n real-time monitoring indicators for evaluating the bridge status, which are marked as D i, i=1-n, the real-time monitoring index refers to an index for obtaining monitoring values ​​at a certain time interval, and the real-time monitoring index varies according to the type of the bridge;

[0009] S102, determining a multi-source data assessment scale;

[0010] Determining the multi-source data evaluation scale includes dividing the monitoring data evaluation results of the monitoring indicators into four scales according to the real-time monitoring indicators, in combination with relevant specifications and actual conditions, and the four scales are: very stable, stable, basically stable, and very unstable;

[0011] S103, multi-source data analysis;

[0012] The multi-source data analysis includes dividing the n monitoring indicators into periodic indicators and non-periodic indicators, wherein the periodic indicators are j in number and are marked as D 1 i (i=1~j), the number of the non-periodic indicators is k, marked as D 2 i (i=1~k), where j+k=n;

[0013] The multi-source data analysis includes the periodicity indicator D 1 i The detection data of the measuring points included in (i=1~j) are sorted to obtain the monitoring data of the same time interval, and the autoregressive integrated moving average model is used to construct the model prediction to obtain the predicted value of the next time interval of the measuring point. According to the overall change situation of the data of the measuring point, the obtained measuring point prediction values ​​are combined as a data set, and the quarter quantile (Q1) or the third quarter quantile (Q3) of the data set is used as the representative value. According to the obtained representative value and the evaluation scale, the periodicity index D is obtained. 1 i (i=1~j) evaluation scale E 1 i (i=1~j);

[0014] The multi-source data analysis also includes a method of dividing the monitoring indicators into periodic indicators and non-periodic indicators by using a visualization analysis method or an autocorrelation analysis method, wherein the visualization analysis method includes drawing a time series diagram of the monitoring data of the detection indicator, and observing whether the monitoring data has a repeated waveform trend, and if so, it indicates that the monitoring data is periodic data, and the monitoring indicator of the monitoring data is also a periodic indicator;

[0015] The multi-source data analysis also includes the non-periodic indicator D 2 iThe detection data of the measuring points included in (i=1-k) are sorted to obtain the monitoring data of the same time interval, and the simple exponential smoothing method is used to construct the model prediction to obtain the predicted value of the next time interval of the measuring point. According to the overall change situation of the data of the measuring point, the obtained measuring point prediction values ​​are combined as a data set, and the quarter quantile (Q1) or the third quarter quantile (Q3) of the data set is used as the representative value. According to the obtained representative value and the evaluation scale, the non-periodic index D is obtained. 2 i (i=1~k) evaluation scale E 2 i (i=1~k);

[0016] S104, weight allocation;

[0017] The weight distribution includes obtaining the periodicity index D by using the hierarchical analysis method. 1 i The weight X of (i=1~j) 1 i (i=1~j) and the non-periodic indicator D 2 i The weight X of (i=1~k) 2 i (i=1~k), the periodicity index D is obtained by expert evaluation method 1 i The weight Y of (i=1~j) 1 i (i=1~j) and the non-periodic indicator D 2 i The weight Y of (i=1~k) 2 i (i=1~k), the periodicity index D is calculated using formula (1) 1 i (i=1~j) the intermediate weight W 1 i (i=1~j) and the non-periodic index D 2 i The final weight W of (i=1~k) 2 i (i=1~k),

[0018]

[0019] S105, comprehensive score calculation;

[0020] The comprehensive score calculation includes the four scales of very stable, stable, basically stable, and very unstable, which correspond to very stable (4 points), stable (3 points), basically stable (2 points), and very unstable (1 point), respectively. 1 i (i=1~j) and E 2 i (i=1~k) The corresponding score is expressed as F 1 i (i=1~j) and F 2 i (i=1~k), according to F 1 i (i=1~j), F 2 i (i=1-k) and the final weight, calculate the comprehensive evaluation value S representing the bridge status, the calculation formula is formula (2),

[0021]

[0022] S106, Bridge Condition Assessment;

[0023] The bridge status assessment includes assessing the status of the bridge according to the calculated comprehensive assessment value S combined with the assessment index, wherein the assessment index is: when S∈(0.75S1, S1], the bridge is judged to be in a stable state; when S∈(0.6S1, 0.75S1], the bridge is judged to be in a basically stable state; when S∈(0.5S1, 0.6S1], the bridge is judged to be in an unstable state; when S∈(0, 0.5S1], the bridge is judged to be in a dangerous state; the assessment index can be appropriately adjusted according to actual conditions.

[0024] Furthermore, in the above S103, the implementation steps of the autocorrelation analysis method are: a) obtaining monitoring data of the monitoring indicator, b) calculating the autocorrelation coefficient R(k), c) drawing an autocorrelation graph, d) observing the autocorrelation graph, and using the confidence interval to determine whether the peak is significant. If the autocorrelation coefficient exceeds the confidence interval, it is considered that there is significant periodicity.

[0025] Furthermore, the calculation formula of the autocorrelation coefficient R(k) is formula (3),

[0026]

[0027] Where, X t is the observation at time point t, X is the mean of the observations, and N is the total number of observations.

[0028] Beneficial effects:

[0029] The present invention proposes a comprehensive bridge status assessment method based on multi-source data fusion, which has the following beneficial effects: by dividing the monitoring data into periodic and non-periodic indicators and adopting different data processing and prediction methods, the bridge status can be predicted and evaluated more accurately; the monitoring data is predicted by using the autoregressive integral moving average model and the simple exponential smoothing method, which can predict the future development of the bridge status and provide a basis for preventive maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 : A flow chart of a comprehensive bridge status assessment method based on multi-source data fusion of the present invention. DETAILED DESCRIPTION

[0031] The specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings; it should be understood that the specific embodiments given here are only used to illustrate and explain the present invention and cannot be used to limit the present invention.

[0032] The following is a specific embodiment of a comprehensive bridge status assessment method based on multi-source data fusion.

[0033] S101, select multi-source data indicators;

[0034] The selection of multi-source data indicators includes selecting n real-time monitoring indicators for evaluating the bridge status, which are marked as D i , i=1-n, the real-time monitoring index refers to an index for obtaining monitoring values ​​at a certain time interval, and the real-time monitoring index varies according to the type of the bridge;

[0035] S102, determining a multi-source data assessment scale;

[0036] Determining the multi-source data evaluation scale includes dividing the monitoring data evaluation results of the monitoring indicators into four scales according to the real-time monitoring indicators, in combination with relevant specifications and actual conditions, and the four scales are: very stable, stable, basically stable, and very unstable;

[0037] S103, multi-source data analysis;

[0038] The multi-source data analysis includes dividing the n monitoring indicators into periodic indicators and non-periodic indicators, wherein the periodic indicators are j in number and are marked as D 1 i (i=1~j), the number of the non-periodic indicators is k, marked as D 2 i (i=1~k), where j+k=n;

[0039] The multi-source data analysis includes the periodicity indicator D 1i The detection data of the measuring points included in (i=1~j) are sorted to obtain the monitoring data of the same time interval, and the autoregressive integrated moving average model is used to construct the model prediction to obtain the predicted value of the next time interval of the measuring point. According to the overall change situation of the data of the measuring point, the obtained measuring point prediction values ​​are combined as a data set, and the quarter quantile (Q1) or the third quarter quantile (Q3) of the data set is used as the representative value. According to the obtained representative value and the evaluation scale, the periodicity index D is obtained. 1 i (i=1~j) evaluation scale E 1 i (i=1~j);

[0040] The multi-source data analysis also includes a method for dividing the monitoring indicators into periodic indicators and non-periodic indicators by using a visualization analysis method or an autocorrelation analysis method. The visualization analysis method includes drawing a time series diagram of the monitoring data of the detection indicator, observing whether the monitoring data has a repeated waveform trend, and if so, it indicates that the monitoring data is periodic data, and the monitoring indicator of the monitoring data is also a periodic indicator.

[0041] The multi-source data analysis also includes the non-periodic indicator D 2 i The detection data of the measuring points included in (i=1-k) are sorted to obtain the monitoring data of the same time interval, and the simple exponential smoothing method is used to construct the model prediction to obtain the predicted value of the next time interval of the measuring point. According to the overall change situation of the data of the measuring point, the obtained measuring point prediction values ​​are combined as a data set, and the quarter quantile (Q1) or the third quarter quantile (Q3) of the data set is used as the representative value. According to the obtained representative value and the evaluation scale, the non-periodic index D is obtained. 2 i (i=1~k) evaluation scale E 2 i (i=1~k);

[0042] S104, weight allocation;

[0043] The weight distribution includes obtaining the periodicity index D by using the hierarchical analysis method. 1 i The weight X of (i=1~j) 1 i (i=1~j) and the non-periodic indicator D 2 i The weight X of (i=1~k) 2 i(i=1~k), the periodicity index D is obtained by expert evaluation method 1 i The weight Y of (i=1~j) 1 i (i=1~j) and the non-periodic indicator D 2 i The weight Y of (i=1~k) 2 i (i=1~k), the periodicity index D is calculated using formula (1) 1 i (i=1~j) the intermediate weight W 1 i (i=1~j) and the non-periodic index D 2 i The final weight W of (i=1~k) 2 i (i=1~k),

[0044]

[0045] S105, comprehensive score calculation;

[0046] The comprehensive score calculation includes the four scales of very stable, stable, basically stable, and very unstable, which correspond to very stable (4 points), stable (3 points), basically stable (2 points), and very unstable (1 point), respectively. 1 i (i=1~j) and E 2 i (i=1~k) The corresponding score is expressed as F 1 i (i=1~j) and F 2 i (i=1~k), according to F 1 i (i=1~j), F 2 i (i=1-k) and the final weight, calculate the comprehensive evaluation value S representing the bridge status, the calculation formula is formula (2),

[0047]

[0048] S106, Bridge Condition Assessment;

[0049] The bridge status assessment includes assessing the status of the bridge according to the calculated comprehensive assessment value S combined with the assessment index, wherein the assessment index is: when S∈(0.75S1, S1], the bridge is judged to be in a stable state; when S∈(0.6S1, 0.75S1], the bridge is judged to be in a basically stable state; when S∈(0.5S1, 0.6S1], the bridge is judged to be in an unstable state; when S∈(0, 0.5S1], the bridge is judged to be in a dangerous state; the assessment index can be appropriately adjusted according to actual conditions.

[0050] Furthermore, in the above S103, the steps of implementing the autocorrelation analysis method are: a) obtaining monitoring data of the monitoring indicator, b) calculating the autocorrelation coefficient R(k), c) drawing an autocorrelation graph, d) observing the autocorrelation graph, and using the confidence interval to determine whether the peak is significant, if the autocorrelation coefficient exceeds the confidence interval, it is considered that there is significant periodicity;

[0051] Furthermore, the calculation formula of the autocorrelation coefficient R(k) is formula (3),

[0052]

[0053] Where, X t is the observation at time point t, X is the mean of the observations, and N is the total number of observations.

[0054] In the above embodiment, the present invention discloses a comprehensive bridge status assessment method based on multi-source data fusion, including selecting multi-source data indicators, determining multi-source data assessment scales, multi-source data analysis, weight allocation, comprehensive score calculation and bridge status assessment; the present invention has the advantages of high accuracy and good predictability, and can be widely used in the field of bridge technical condition assessment.

[0055] The above description is a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A comprehensive bridge condition assessment method based on multi-source data fusion, characterized in that: The steps of the comprehensive bridge status assessment method based on multi-source data fusion are as follows: 1) Select multi-source data indicators; 2) Determine the evaluation scale of multi-source data; 3) Multi-source data analysis; 4) Weight allocation; 5) Calculation of comprehensive score; 6) Bridge condition assessment; The selection of multi-source data indicators includes selecting n real-time monitoring indicators for evaluating the bridge status, which are marked as D i , i=1-n, the real-time monitoring index refers to an index for obtaining monitoring values ​​at a certain time interval, and the real-time monitoring index varies according to the type of the bridge; Determining the multi-source data evaluation scale includes dividing the monitoring data evaluation results of the monitoring indicators into four scales according to the real-time monitoring indicators, in combination with relevant specifications and actual conditions, and the four scales are: very stable, stable, basically stable, and very unstable; The multi-source data analysis includes dividing the n monitoring indicators into periodic indicators and non-periodic indicators, wherein the periodic indicators are j in number and are marked as D 1 i (i=1~j), the number of the non-periodic indicators is k, marked as D 2 i (i=1~k), where j+k=n; The multi-source data analysis includes the periodicity indicator D 1 i The detection data of the measuring points included in (i=1~j) are sorted to obtain the monitoring data of the same time interval, and the autoregressive integrated moving average model is used to construct the model prediction to obtain the predicted value of the next time interval of the measuring point. According to the overall change situation of the data of the measuring point, the obtained measuring point prediction values ​​are combined as a data set, and the quarter quantile (Q1) or the third quarter quantile (Q3) of the data set is used as the representative value. According to the obtained representative value and the evaluation scale, the periodicity index D is obtained. 1 i (i=1~j) evaluation scale E 1 i (i=1~j); The multi-source data analysis also includes a method of dividing the monitoring indicators into periodic indicators and non-periodic indicators by using a visualization analysis method or an autocorrelation analysis method, wherein the visualization analysis method includes drawing a time series diagram of the monitoring data of the detection indicator, and observing whether the monitoring data has a repeated waveform trend, and if so, it indicates that the monitoring data is periodic data, and the monitoring indicator of the monitoring data is also a periodic indicator; The multi-source data analysis also includes the analysis of the non-periodic indicator D 2 i The detection data of the measuring points included in (i=1-k) are sorted to obtain the monitoring data of the same time interval, and the simple exponential smoothing method is used to construct the model prediction to obtain the predicted value of the next time interval of the measuring point. According to the overall change situation of the data of the measuring point, the obtained measuring point prediction values ​​are combined as a data set, and the quarter quantile (Q1) or the third quarter quantile (Q3) of the data set is used as the representative value. According to the obtained representative value and the evaluation scale, the non-periodic index D is obtained. 2 i (i=1~k) evaluation scale E 2 i (i=1~k); The weight distribution includes obtaining the periodicity index D by using the hierarchical analysis method. 1 i The weight X of (i=1~j) 1 i (i=1~j) and the non-periodic index D 2 i The weight X of (i=1~k) 2 i (i=1~k), the periodicity index D is obtained by expert evaluation method 1 i The weight Y of (i=1~j) 1 i (i=1~j) and the non-periodic index D 2 i The weight Y of (i=1~k) 2 i (i=1~k), the periodicity index D is calculated using formula (1) 1 i (i=1~j) the middle weight W 1 i (i=1~j) and the non-periodic index D 2 i The final weight W of (i=1~k) 2 i (i=1~k), The comprehensive score calculation includes the four scales of very stable, stable, basically stable, and very unstable, which correspond to very stable (4 points), stable (3 points), basically stable (2 points), and very unstable (1 point), respectively. 1 i (i=1~j) and E 2 i (i=1~k) The corresponding score is expressed as F 1 i (i=1~j) and F 2 i (i=1~k), according to F 1 i (i=1~j), F 2 i (i=1-k) and the final weight, calculate the comprehensive evaluation value S representing the bridge status, the calculation formula is formula (2), The bridge status assessment includes assessing the status of the bridge according to the calculated comprehensive assessment value S combined with the assessment index, wherein the assessment index is: when S∈(0.75S1, S1], the bridge is judged to be in a stable state; when S∈(0.6S1, 0.75S1], the bridge is judged to be in a basically stable state; when S∈(0.5S1, 0.6S1], the bridge is judged to be in an unstable state; when S∈(0, 0.5S1], the bridge is judged to be in a dangerous state; the assessment index can be appropriately adjusted according to actual conditions.

2. The method for comprehensive evaluation of bridge status based on multi-source data fusion according to claim 1 is characterized in that: In the above 3), the steps for implementing the autocorrelation analysis method are: a) obtaining monitoring data of the monitoring indicator, b) calculating the autocorrelation coefficient R(k), c) drawing an autocorrelation graph, d) observing the autocorrelation graph, and using the confidence interval to determine whether the peak is significant. If the autocorrelation coefficient exceeds the confidence interval, it is considered that there is significant periodicity.

3. The method for comprehensive evaluation of bridge status based on multi-source data fusion according to claim 2 is characterized in that: The calculation formula of the autocorrelation coefficient R(k) is formula (3), Where Xt is the observation at time point t, X is the mean of the observations, and N is the total number of observations.

Images