A BIM-based online performance evaluation and prediction method for cable-stayed bridges

Abstract

The present invention proposes an online performance evaluation and prediction method for a cable-stayed bridge based on BIM, which includes creating a BIM model and creating an online evaluation and prediction system based on the BIM model. The present invention integrates real-time monitoring data collected by an existing bridge health management system and regular manual inspection data, and adopts a hierarchical analysis method to achieve comprehensive and accurate monitoring and evaluation of large-span cable-stayed bridges. At the same time, based on big data and time series methods, a prediction model of cable force is created to extract its corresponding trend changes, which can not only obtain the true degradation trend of the structure, but also evaluate the sequence indicators.

Description

Technical Field

[0001] The invention relates to the field of cable-stayed bridge monitoring, and in particular to a BIM-based cable-stayed bridge performance online evaluation and prediction method. Background Art

[0002] At present, the maintenance and management of long-span bridges is usually achieved through the following two methods:

[0003] (1) Bridge health monitoring system: By installing sensors at the response position of the bridge, data can also be collected with the help of technologies such as cameras and drones. However, the bridge health monitoring system is still not perfect. Due to the limitations of the scale of the health monitoring system and monitoring technology, the bridge health monitoring system cannot provide complete bridge health information;

[0004] (2) Manual regular inspections. The information obtained by conventional manual visual inspections or portable one-time measurements has a long manual inspection cycle, many blind spots, and high subjectivity, which often have great limitations in actual use. In addition, the safety and durability assessment of long-span bridges still relies on experienced engineering and technical personnel. However, the investigation methods and judgment criteria used in this "traditional empirical method" are directly related to the personal knowledge level and experience of the assessor. The expert evaluation results under different background conditions will be highly subjective and may be misjudged. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a BIM-based online performance evaluation and prediction method for a cable-stayed bridge.

[0006] The main contents of the present invention include:

[0007] A BIM-based online performance evaluation and prediction method for a cable-stayed bridge comprises the following steps:

[0008] S1. Create a corresponding BIM model according to the object to be monitored, and arrange the same sensor model at the corresponding position of the BIM model according to the monitoring plan of the monitored object;

[0009] S2. Create an online evaluation and prediction system based on the BIM model, which includes a data acquisition module, a data processing module and an early warning display module; the data acquisition module is connected to the existing real-time monitoring system to transmit the data corresponding to each sensor in the existing real-time monitoring system to the corresponding sensor model in the BIM model; the data acquisition module can also receive manual inspection data; the data processing module receives the data transmitted by the data acquisition module, and creates a bridge performance evaluation index after data preprocessing and information fusion; the early warning display module receives the bridge performance evaluation index obtained by the data processing module, and evaluates the service status of the bridge based on the variable weight analysis fusion method according to the created evaluation criteria, and establishes a mathematical model of the trend change of the cable-stayed bridge status based on big data and time series methods, and displays the estimation and prediction results.

[0010] Preferably, the data acquisition module includes a real-time monitoring unit and a manual inspection unit, wherein the real-time monitoring unit is used to receive sensor data transmitted by an existing real-time monitoring system; and the manual inspection unit receives input of scheduled inspection data.

[0011] Preferably, the data processing module receives the data from the real-time monitoring unit, and after self-calibration and fusion, performs mutual calibration and fusion with the data transmitted by the manual inspection unit, and generates bridge performance evaluation indicators based on the created evaluation model.

[0012] Preferably, based on big data and time series methods, a mathematical model of the state trend change of a cable-stayed bridge is established, and the estimation and prediction results are displayed, including the following steps:

[0013] S21. updating the BIM model according to the evaluation results of the service status of the bridge;

[0014] S22. According to the prediction scheme, select the acceleration sensor model at the corresponding position on the BIM model, and obtain the monitoring data of the corresponding acceleration sensor on the existing real-time monitoring system to obtain the cable force of the corresponding cable;

[0015] S23. Use the Holt-Winters triple exponential smoothing method to create a cable force prediction model, enter the cable force data obtained in S22 into the cable force prediction model, and perform seasonal statistical tests using the Krusakal-Wallis test to evaluate the trend of future performance evolution of the bridge.

[0016] Preferably, S23 includes the following sub-steps:

[0017] S231. Use the Krusakal-Wallis test to identify the seasonal effects of the index. The test statistic is:

[0018]

[0019] Among them, R i Indicates that in the i-th season y i The rank sum of n i Indicates that in the i-th season y i ; n represents the number of all y, represents the average rank of the i-th season; ——overall average rank;

[0020] S232. If H>χ calculated in S231 α 2 (L-1), the cable is considered to have seasonal influence, and the cable force prediction model is created according to the Holt-Winters triple exponential smoothing method, and the time when the predicted structure degrades to the threshold is calculated as:

[0021]

[0022] in, Assume that the current time of a certain measuring point is t1, and the predicted value of the cable force at time t is At (t1+t m ) reaches the limit value;

[0023] S233. If H calculated in S231 is not greater than χ α 2 (L-1), it is considered that the cable has no seasonal influence, and the cable force variation trend is obtained by the method of fitting, that is,

[0024]

[0025] The corresponding predicted structure degenerates to the threshold time:

[0026]

[0027] S234. The time t at which the predicted structure of a single cable degenerates to a threshold value m , calculate the cable force evaluation value of the monitoring point.

[0028] Preferably, the method for calculating the cable force evaluation value of the monitoring point in S234 is as follows:

[0029] When t m >T0,x j =100, when t m =0,x j = 0, then the cable force evaluation value of the measuring point is

[0030]

[0031] Preferably, creating a cable force prediction model using the Holt-Winters triple exponential smoothing method includes:

[0032] Build the additive mode model:

[0033]

[0034] In the formula, a t is the horizontal part of the sequence, b t is the trend part of the series, s t is the seasonal part of the sequence, L is the length of a season; α, β, γ are smoothing coefficients, ranging from 0 to 1. In the equation system, the first equation is the level equation, the second is the trend equation, and the third equation is the seasonal equation, which means that the seasonal change value in period t is the weighted average of the difference between the observed value and the level value in period t and the seasonal change value in period tL;

[0035] The additive mode model is used for sequence prediction, and the predicted value of l steps forward in period t is:

[0036]

[0037] Among them, a t 、b t are the level and trend values ​​in period t, s t+l-L is the seasonal variation value of t+lL.

[0038] Preferably, the initial value in formula (1) is:

[0039] Among them, the initial value of the first formula is

[0040] The initial value of the second formula

[0041] The initial value of the third formula: s1 = y1-a0, s2 = y2-a0, ...s L =y L -a0.

[0042] Preferably, the method for creating the evaluation criteria includes: generating a cable-stayed bridge judgment matrix based on an expert questionnaire method, performing fuzzy C-means clustering through an FCM algorithm, classifying the weights given by each expert, and constructing an indicator to distinguish the logic of the expert judgment matrix through the maximum eigenvalue root of the judgment matrix.

[0043] Preferably, the method for evaluating the service status of a bridge based on the variable weight analysis fusion method includes: using the hierarchical analysis method to perform status evaluation, wherein the evaluation index of the cable force is the constant load eigenvalue extracted by eliminating the temperature effect and the live load effect, and the cable forces at the symmetrical positions of the structure are selected for mutual evaluation.

[0044] Compared with the prior art, the BIM-based online performance evaluation and prediction method for cable-stayed bridges proposed in the present invention has the following beneficial effects: by integrating the real-time monitoring data collected by the existing bridge health management system and the regular manual inspection data, the hierarchical analysis method is used to achieve comprehensive and accurate monitoring and evaluation of large-span cable-stayed bridges. At the same time, based on big data and time series methods, by creating a prediction model of cable force and extracting its corresponding trend changes, not only the real degradation trend of the structure can be obtained, but also the sequence indicators can be evaluated. DETAILED DESCRIPTION

[0045] The technical solution protected by the present invention will be described in detail below.

[0046] A BIM-based online performance evaluation and prediction method for a cable-stayed bridge comprises the following steps:

[0047] S1. Create a BIM model corresponding to the object to be monitored, including bridge towers, cables, main beams, foundations, railings, expansion joints and other components, and deploy the same sensor model at the corresponding position of the BIM model according to the monitoring plan of the monitored object to simulate a real bridge;

[0048] S2. Create an online evaluation and prediction system based on the BIM model, which includes a data acquisition module, a data processing module and an early warning display module; the data acquisition module is connected to the existing real-time monitoring system to transmit the data corresponding to each sensor in the existing real-time monitoring system to the corresponding sensor model in the BIM model; the data acquisition module can also receive manual inspection data; specifically, the data acquisition module includes a real-time monitoring unit and a manual inspection unit, the real-time monitoring unit is used to receive the sensor data transmitted by the existing real-time monitoring system; the manual inspection unit receives the input of scheduled inspection data.

[0049] The data processing module receives the data transmitted by the data acquisition module, and after data preprocessing and information fusion, creates a bridge performance evaluation index based on the hierarchical analysis method. Specifically, the data processing module receives the data of the real-time monitoring unit, and after self-calibration and fusion, it performs mutual calibration and fusion with the data transmitted by the manual inspection unit, and generates a bridge performance evaluation index according to the created evaluation model; the above-mentioned hierarchical analysis method refers to dividing the evaluation indicators into qualitative indicators, quantitative indicators and sequence indicators, such as coating degradation area, rust area, bolt defect rate, etc., which are indicators that can be evaluated by quantification, while cable force, displacement and stress are called sequence indicators.

[0050] The early warning display module receives the bridge performance evaluation index obtained by the data processing module, evaluates the service status of the bridge based on the variable weight analysis fusion method according to the created evaluation criteria, and establishes a mathematical model of the state trend change of the cable-stayed bridge based on big data and time series methods, and displays the estimation and prediction results. The method for evaluating the service status of the bridge based on the variable weight analysis fusion method includes: using the hierarchical analysis method for status evaluation, wherein the evaluation index of the cable force is the constant load characteristic value extracted by eliminating the temperature effect and the live load effect, and the cable force at the symmetrical position of the structure is selected for mutual verification evaluation.

[0051] Preferably, based on big data and time series methods, a mathematical model of the state trend change of a cable-stayed bridge is established, and the estimation and prediction results are displayed, including the following steps:

[0052] S21. updating the BIM model according to the evaluation results of the service status of the bridge;

[0053] S22. According to the prediction scheme, select the acceleration sensor model at the corresponding position on the BIM model, and obtain the monitoring data of the corresponding acceleration sensor on the existing real-time monitoring system to obtain the cable force of the corresponding cable;

[0054] S23. Use the Holt-Winters triple exponential smoothing method to create a cable force prediction model, enter the cable force data obtained in S22 into the cable force prediction model, and perform seasonal statistical tests using the Krusakal-Wallis test to evaluate the trend of future performance evolution of the bridge.

[0055] In order to establish the relationship between the cable force and time through the Holt-Winters model, it is also necessary to determine whether the cable force sequence has seasonality from a quantitative perspective, and then evaluate the cable force according to the influence of seasonality. In this embodiment, the Krusakal-Wallis test is used to perform a seasonal statistical test. Specifically, S23 includes the following sub-steps:

[0056] S231. Use the Krusakal-Wallis test to identify the seasonal effects of the index. The test statistic is:

[0057]

[0058] Among them, R i Indicates that in the i-th season y i The rank sum of n i Indicates that in the i-th season y i ; n represents the number of all y, represents the average rank of the i-th season; ——overall average rank;

[0059] S232. If H>χ calculated in S231 α 2 (L-1), the cable is considered to have seasonal influence, and the cable force prediction model is created according to the Holt-Winters triple exponential smoothing method, and the time when the predicted structure degrades to the threshold is calculated as:

[0060]

[0061] in, Assume that the current time of a certain measuring point is t1, and the predicted value of the cable force at time t is At (t1+t m ) reaches the limit value;

[0062] S233. If H calculated in S231 is not greater than χ α 2 (L-1), it is considered that the cable has no seasonal influence, and the cable force variation trend is obtained by the method of fitting, that is,

[0063]

[0064] The corresponding predicted structure degenerates to the threshold time:

[0065]

[0066] S234. The time t at which the predicted structure of a single cable degenerates to a threshold value m , calculate the cable force evaluation value of the monitoring point.

[0067] Preferably, the method for calculating the cable force evaluation value of the monitoring point in S234 is as follows:

[0068] When t m >T0,x j =100, when t m =0,x j = 0, then the cable force evaluation value of the measuring point is

[0069]

[0070] Preferably, creating a cable force prediction model using the Holt-Winters triple exponential smoothing method includes:

[0071] Build the additive mode model:

[0072]

[0073] In the formula, a t is the horizontal part of the sequence, b t is the trend part of the series, s tis the seasonal part of the sequence, L is the length of a season; α, β, γ are smoothing coefficients, ranging from 0 to 1. In the equation system, the first equation is the level equation, the second is the trend equation, and the third equation is the seasonal equation, which means that the seasonal change value in period t is the weighted average of the difference between the observed value and the level value in period t and the seasonal change value in period tL;

[0074] The additive mode model is used for sequence prediction, and the predicted value of l steps forward in period t is:

[0075]

[0076] Among them, a t 、b t are the level value and trend value in period t, s t+l-L is the seasonal variation value of t+lL.

[0077] Preferably, the initial value in formula (1) is:

[0078] Among them, the initial value of the first formula is

[0079] The initial value of the second formula

[0080] The initial value of the third formula: s1 = y1-a0, s2 = y2-a0, ...s L =y L -a0.

[0081] Among them, the method for creating the evaluation criteria includes: generating a cable-stayed bridge judgment matrix based on the expert questionnaire method, performing fuzzy C-means clustering through the FCM algorithm, classifying the weights given by each expert, and constructing an indicator to distinguish the logic of the expert judgment matrix through the maximum eigenvalue root of the judgment matrix.

[0082] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A BIM-based online performance evaluation and prediction method for a cable-stayed bridge, characterized in that: The steps include: S1. Create a corresponding BIM model according to the object to be monitored, and arrange the same sensor model at the corresponding position of the BIM model according to the monitoring plan of the monitored object; S2. Create an online evaluation and prediction system based on the BIM model, the online evaluation and prediction system includes a data acquisition module, a data processing module and an early warning display module; the data acquisition module is connected to the existing real-time monitoring system, and transmits the data corresponding to each sensor in the existing real-time monitoring system to the corresponding sensor model in the BIM model; the data acquisition module can also receive manual inspection data; The data processing module receives the data transmitted by the data acquisition module, and creates bridge performance evaluation indicators based on the hierarchical analysis method after data preprocessing and information fusion; The early warning display module receives the bridge performance evaluation index obtained by the data processing module, evaluates the service status of the bridge based on the variable weight analysis fusion method according to the created evaluation criteria, and establishes a mathematical model of the trend change of the cable-stayed bridge status based on big data and time series methods, and displays the estimation and prediction results; Based on big data and time series methods, the mathematical model of the state trend change of cable-stayed bridges is established, and the estimation and prediction results are displayed, which includes the following steps: S21. updating the BIM model according to the evaluation results of the service status of the bridge; S22. According to the prediction scheme, select the acceleration sensor model at the corresponding position on the BIM model, and obtain the monitoring data of the corresponding acceleration sensor on the existing real-time monitoring system to obtain the cable force of the corresponding cable; S23. Use the Holt-Winters triple exponential smoothing method to create a cable force prediction model, enter the cable force data obtained in S22 into the cable force prediction model, and perform seasonal statistical tests through the Krusakal-Wallis test to evaluate the trend of future performance evolution of the bridge; S23 includes the following sub-steps: S231. Use the Krusakal-Wallis test to identify the seasonal effects of the index. The test statistic is: ； in, Indicated in In the seasons rank and; Indicated in In the seasons The number of Indicates all The number of ; Indicates The average rank of the seasons; ——overall average rank; S232. If the calculation in S231 , it is considered that the cable has seasonal influence, and the cable force prediction model is created according to the Holt-Winters triple exponential smoothing method, and the time when the predicted structure degrades to the threshold is calculated as: ； in, , assuming that the current time of a certain measuring point is , The predicted value of the cable force at time ,exist The measured value of the cable force reaches the limit; S233. If the calculation in S231 , it is considered that the cable has no seasonal influence, and the cable force variation trend is obtained by the method of fitting, that is, ； The corresponding predicted structure degenerates to the threshold time: ； S234. The time when the predicted structure of a single cable degenerates to a threshold value , calculate the cable force evaluation value of the monitoring point, and calculate the corresponding scoring weight of the cable force.

2. The BIM-based online performance evaluation and prediction method for a cable-stayed bridge according to claim 1 is characterized in that: The data acquisition module includes a real-time monitoring unit and a manual inspection unit. The real-time monitoring unit is used to receive sensor data transmitted by an existing real-time monitoring system; the manual inspection unit receives input of scheduled inspection data.

3. The BIM-based online performance evaluation and prediction method for a cable-stayed bridge according to claim 2 is characterized in that: The data processing module receives the data from the real-time monitoring unit, performs self-calibration and fusion, and then performs mutual calibration and fusion with the data transmitted by the manual inspection unit, and generates bridge performance evaluation indicators based on the created evaluation model.

4. The BIM-based online performance evaluation and prediction method for a cable-stayed bridge according to claim 1 is characterized in that: S234 The method for calculating the cable force evaluation value of the monitoring point is as follows: when , , when , , then the cable force evaluation value of the measuring point is 5. The BIM-based online performance evaluation and prediction method for a cable-stayed bridge according to claim 1 is characterized in that: The cable force prediction model created using the Holt-Winters triple exponential smoothing method includes: Build the additive mode model: （1）； In the formula, is the horizontal part of the sequence, is the trend part of the series, is the seasonal part of the sequence, is the length of a season cycle; is the smoothing coefficient, which is between 0 and 1. In the equation system, the first equation is the horizontal equation, the second is the trend equation, and the third equation is the seasonal equation, which means The seasonal variation value of period t is the sum of the difference between the observed value and the level value of period The weighted average of the seasonal variation values ​​over the period; The additive mode model is used for sequence prediction, and the predicted value of l steps forward in period t is: ； in, , are the level value and trend value in period t, is the seasonal variation value of t+lL.

6. The BIM-based online performance evaluation and prediction method for a cable-stayed bridge according to claim 5 is characterized in that: The initial value in the formula (1) is: Among them, the initial value of the first formula is ; The initial value of the second formula ; The initial value of the third formula is: .

7. The BIM-based online performance evaluation and prediction method for a cable-stayed bridge according to claim 1 is characterized in that: The method for creating the evaluation criteria includes: generating a cable-stayed bridge judgment matrix based on an expert questionnaire method, performing fuzzy C-means clustering through an FCM algorithm, classifying the weights given by each expert, and constructing an indicator for distinguishing the logic of the expert judgment matrix through the maximum eigenvalue root of the judgment matrix.

8. The BIM-based online performance evaluation and prediction method for a cable-stayed bridge according to claim 1 is characterized in that: The method for evaluating the service status of a bridge based on the variable weight analysis fusion method includes: using the hierarchical analysis method for status evaluation, wherein the evaluation index of the cable force is the constant load eigenvalue extracted by eliminating the temperature effect and the live load effect, and the cable forces at the symmetrical positions of the structure are selected for mutual evaluation.

Images