Self-sensing rotation monitoring system and intelligent risk prediction method

Abstract

The invention discloses a self-sensing swivel monitoring system and an intelligent risk prediction method. The self-sensing swivel monitoring system is composed of a strain gauge, a pressure box, a strain rosette, a three-axis tilt angle sensor and a three-axis gyroscope. The strain gauges are embedded in the spherical hinges; the pressure box is embedded in the top surface of the upper spherical hinge; the strain rosettes are embedded in the side surfaces of the main piers; the tilt angle sensor and the gyroscope are installed in the center of the top face of the main beam. The identification method comprises the following steps: determining a risk index; collecting data in real time, calculating risk values corresponding to the risk indexes, and generating a risk value time sequence; inputting the risk value time sequence into a trained risk prediction model to generate a prediction risk level signal; the system has the active sensing capability of multiple types of monitoring data, can realize dynamic real-time monitoring and risk state estimation of multi-dimensional rotation states such as the deflection state, the rotation speed, the local load and the bridge balance weight of the swivel bridge, is beneficial to mastering the safety state in the construction process of the swivel bridge, guides the construction, and improves the construction efficiency. And digitization, precision and intelligentization of the construction process are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of transportation civil engineering, and more specifically, to a self-sensing rotation monitoring system and an intelligent risk prediction method. Background Art

[0002] The construction monitoring of a swing bridge refers to the real-time risk data monitoring of relevant structures during the construction of the foundation structure of the swing bridge and the rotation process to ensure that the construction quality and structural safety of the swing bridge meet relevant regulations. By collecting and analyzing the effective real-time response data during the construction period, construction personnel can take effective response measures in a timely manner after abnormal situations occur. In recent years, with the continuous improvement of bridge construction technology, the swing construction method has emerged in bridge construction. The swing bridge construction method refers to a construction method in which the bridge structure is first constructed on a non-design axis and then in place after being rotated by traction. Due to its advantages such as less interference with the existing traffic operation below, it is widely used in the bridge engineering practice across existing lines (especially railways).

[0003] In actual construction, due to the complex swing construction process and many uncertain factors, it is difficult to achieve precise swing bridge construction monitoring and risk management. Currently, the existing construction monitoring means are relatively single, mainly focusing on the linearity of the cast-in-place main beam before rotation, the foundation settlement, and the stress state of key sections. This monitoring content is basically the same as that of conventional bridge monitoring. During the rotation process, the monitoring methods for the rotation speed and displacement are very scarce, the dynamic monitoring indicators are small, and the monitoring content is relatively single, which cannot meet the requirements of dynamically monitoring the overall safety information of the bridge during the construction of the swing bridge and comprehensively realizing the risk identification and prediction alarm in the swing project. Moreover, the overall monitoring process has a low degree of automation, large errors in the collected data, and unsatisfactory visualization.

[0004] Therefore, there is an urgent need in this field for a monitoring system that can perform real-time monitoring of the swing construction of a swing bridge with multi-dimensions, precision, and visualization. At the same time, this system should also have the ability to intelligently predict and discriminate risks, be able to dynamically self-sense and capture the specific location and possible occurrence time of risks, and establish a risk evaluation index system under the influence of different factors. Summary of the Invention

[0005] In view of this, the present invention provides a self-sensing rotation monitoring system, which has the monitoring ability of various data, can monitor the safety state during the construction process of the swing bridge in real time, guide the construction, and helps to achieve a higher degree of digitization, precision, and intelligence in the construction process.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A self-sensing rotation monitoring system, including a self-sensing sensing system, which is composed of strain gauges, pressure cells, strain rosettes, three-axis inclinometers and three-axis gyroscopes, and is used to obtain the monitoring data of multiple components in the rotating bridge;

[0008] The multiple components in the rotating bridge include: rotating ball hinge, main girder and main pier; wherein, the rotating ball hinge includes an upper ball hinge, a lower ball hinge and a plurality of ball hinge sliders distributed between the spherical surface of the upper ball hinge and the lower ball hinge.

[0009] Preferably, it further includes a data processing terminal, which is used to obtain the monitoring data and display it externally through a display screen; used to analyze the monitoring data, automatically identify risks, and generate risk information; used to generate an alarm signal according to the risk identification result.

[0010] Preferably, the upper ball hinge and the lower ball hinge are of steel-concrete structure, and the strain gauges include 4 first strain gauges buried in the steel-concrete bonding surface of the lower ball hinge, 8 second strain gauges buried in the steel-concrete bonding surface of the upper ball hinge, and 35 third strain gauges buried on the surface of the lower ball hinge;

[0011] The first strain gauge, the second strain gauge and the third strain gauge are respectively used to obtain the stress data at the steel-concrete bonding surface of the lower ball hinge, the steel-concrete bonding surface of the upper ball hinge and the surface of the lower ball hinge.

[0012] Preferably, the strain gauges further include 15 fourth strain gauges buried on the side surface of the ball hinge slider, which are used to obtain the axial stress at the ball hinge slider.

[0013] Preferably, the number of the pressure cells is 4, and multiple pressure cells are buried on the top surface of the upper ball hinge to obtain the compression data of the upper ball hinge.

[0014] Preferably, the number of the strain rosettes is 4, and multiple strain rosettes are buried on the side surface of the main pier to obtain the torque on the pier column when the main girder rotates.

[0015] Preferably, the three-axis inclinometer is installed at the center position of the top surface of the main girder to obtain the skew and rotation angle data of the main girder.

[0016] Preferably, the three-axis gyroscope is installed at the center position of the top surface of the main girder, and the three-axis gyroscope is used to measure the angular acceleration and angular velocity of the bridge rotation.

[0017] A risk intelligent prediction method includes the following steps:

[0018] Determine risk indicators;

[0019] Obtain real-time acquisition data, calculate the risk values corresponding to each of the risk indicators according to the risk indicators, and generate a risk value time series;

[0020] Input the risk value time series into a trained risk prediction model to generate a predicted risk level signal.

[0021] Preferably, the steps further include: dynamically monitoring the collected data, and the dynamic monitoring includes: data screening, data verification, and data storage.

[0022] Preferably, the risk indicators include primary indicators and secondary indicators corresponding to the primary indicators;

[0023] The primary indicators include: stress risk, inclination risk, and overspeed risk,

[0024] The secondary indicators corresponding to the stress risk include: axial compressive stress of the outermost spherical hinge slider, circumferential shear stress of the outermost spherical hinge slider, torsional shear stress of the pier column, and rotational traction force;

[0025] The secondary indicators corresponding to the inclination risk include: rotational inclination angle of the main girder and predicted inclination angle of the spherical hinge slider;

[0026] The secondary indicators corresponding to the overspeed risk include: rotational angular velocity and rotational angular acceleration.

[0027] Advantages of the present invention:

[0028] As can be seen from the above technical solutions, compared with the prior art, the present invention discloses a self-sensing rotation monitoring system and a risk intelligent prediction method, which have the ability of real-time digital characterization of multiple structural components of a rotating bridge, can monitor the safety status during the construction process of a rotating bridge in real time, guide the construction, and contribute to achieving higher precision in the digitization, precision, and intelligence of the construction process. Description of the Drawings

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0030] Figure 1 The drawings are schematic diagrams of the layout of the self-sensing sensing system in the self-sensing rotation monitoring system provided by the present invention;

[0031] Figure 1 (a) The drawings are schematic diagrams of the measuring points of the self-sensing sensing system on the main girder, the measuring points of the main pier column, and the structure of the rotating bridge provided by the present invention;

[0032] Figure 1 (b) Schematic diagram of measuring points of the self-sensing sensing system of the present invention in the slewing spherical hinge;

[0033] Figure 2 Schematic diagram of the distribution of measuring points of the self-sensing sensing system provided by the present invention in the slewing spherical hinge;

[0034] Figure 2 (a) Schematic diagram of the distribution of measuring points of the first strain gauge of the present invention at the bonded surface of the lower spherical hinge of steel and concrete;

[0035] Figure 2 (b) Schematic diagram of the distribution of measuring points of the second strain gauge at the bonded surface of the upper spherical hinge of steel and concrete in the present invention;

[0036] Figure 2 (c) Schematic diagram of the distribution of measuring points of the third strain gauge on the surface of the lower spherical hinge in the present invention;

[0037] Figure 2 (d) Schematic diagram of the distribution of the fourth strain gauge and the spherical hinge slider in the present invention. The strain gauge is on the side surface of the slider, and the installation position is at the perpendicular line of the connecting line between the center of the spherical hinge and the center of the slider;

[0038] Figure 2 (e) Schematic diagram of the distribution of the pressure cell on the top surface of the upper spherical hinge in the spherical hinge sensor system of the present invention;

[0039] Figure 3 The accompanying drawings are schematic diagrams of a risk intelligent prediction method provided by the present invention;

[0040] Figure 4 The accompanying drawings are schematic diagrams of the key index system for risk intelligent identification;

[0041] Figure 5 The accompanying drawings are schematic diagrams of the risk prediction results of axial stress and skew angle in the present invention;

[0042] Figure 5 (a) The accompanying drawings are schematic diagrams of the prediction results of the GM(1,1) model for the axial compressive stress of the spherical hinge slider;

[0043] Figure 5 (b) The accompanying drawings are schematic diagrams of the prediction results of the GM(1,N) model for the skew angle;

[0044] Among them, 1 - strain gauge, 2 - pressure cell, 3 - strain rosette, 4 - main beam sensor; 5 - slewing spherical hinge, 51 - upper spherical hinge, 52 - lower spherical hinge, 53 - spherical hinge slider; 6 - main beam, 7 - main pier; 8 - lower bearing platform, 9 - upper bearing platform. Specific implementation manners

[0045] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0046] Embodiment 1

[0047] As Figure 1 and Figure 2 , an embodiment of the present invention discloses a self-sensing swivel monitoring system, including a self-sensing sensing system, which is composed of a strain gauge 1, a pressure cell 2, a strain rosette 3 and a main girder sensor 4. Among them, the main girder sensor 1 includes a three-axis inclinometer and a three-axis gyroscope, and is used to obtain the monitoring data of multiple components in the swivel bridge.

[0048] Multiple components in the swivel bridge include: a swivel spherical hinge 5, a main girder 6 and a main pier 7; the specific structure of the swivel bridge is as Figure 1 (a) shown, and the specific structure of the swivel spherical hinge 5 is as Figure 1 (b) shown, which includes an upper spherical hinge 51, a lower spherical hinge 52 and a plurality of spherical hinge sliders 53 distributed between the spherical surface of the upper spherical hinge 51 and the lower spherical hinge 52; the upper spherical hinge 51 and the lower spherical hinge 52 are of reinforced concrete structure or concrete structure. The lower spherical hinge 52 in the swivel spherical hinge 5 is buried inside the lower bearing platform 8, the upper spherical hinge is connected to the main pier 7 through the upper bearing platform 9, and the upper part of the main pier 7 is connected to the main girder 6.

[0049] Among them, the monitoring data includes the skew state data of the main girder of the swivel bridge, the local compression data in the swivel spherical hinge and the torque data received by the main pier during rotation. Specifically, the skew state data of the main girder includes the rotation angle and angular acceleration of the swivel bridge, the rotation speed and rotational acceleration.

[0050] In order to externally display the collected data, a data processing terminal is further included, which is used to obtain the monitoring data and externally display it through a display screen; in addition, in order to achieve intelligent and automated applications, the data processing terminal is also used to analyze the monitoring data, automatically identify risks, and generate risk information; and is used to generate an alarm signal according to the risk identification result. The strain gauge can adopt a high-precision strain gauge, such as a sensitivity coefficient of (2.00±1)%, and an ultimate strain of 20000μm / m.

[0051] As Figure 2, To further implement the above solution, the upper spherical hinge 51 and the lower spherical hinge 52 are of steel-concrete structure. The strain gauge 1 includes 4 first strain gauges embedded in the steel-concrete bonding surface of the lower spherical hinge 52, 8 second strain gauges embedded in the steel-concrete bonding surface of the upper spherical hinge 51, and 35 third strain gauges embedded on the surface of the lower spherical hinge 52. The first strain gauge, the second strain gauge, and the third strain gauge are respectively used to obtain the stress data at the steel-concrete bonding surface of the lower spherical hinge 52, the steel-concrete bonding surface of the upper spherical hinge 51, and the surface of the lower spherical hinge 52. Among them, the first strain gauge, the second strain gauge, and the third strain gauge are evenly installed on the corresponding installation surfaces. When arranging the strain gauge 1, the edge positions of the installation surfaces can be preferentially considered. Specifically, multiple first strain gauges are at least evenly distributed at the edge of the steel-concrete bonding surface of the lower spherical hinge 52; multiple first strain gauges are at least evenly distributed at the edge of the steel-concrete bonding surface of the upper spherical hinge 51.

[0052] Further, the strain gauge 1 further includes a fourth strain gauge embedded on the side surface of the spherical hinge slider 53. At least one fourth strain gauge is arranged on the side surface of each spherical hinge slider 53, which is used to obtain the axial stress at the spherical hinge slider 53. Taking the surface of the lower spherical hinge as a reference, the arrangement of the fourth strain gauge and the spherical hinge slider 53 is as Figure 2 (d) shown.

[0053] To further implement the above solution, the number of pressure cells 2 is 4. The 4 pressure cells 2 are evenly embedded in the top surface of the upper spherical hinge 51, which is used to obtain the compression data of the upper spherical hinge 51. The pressure cell 2 is a vibrating wire pressure sensor. Since there are circumferential rib plates on the top surface of the upper spherical hinge 51, it is necessary to pre-cast it into a plane before embedding the pressure sensor, and then perform the secondary pouring of the main pier 7. The four pressure sensors are evenly embedded in a cross distribution form, which can not only analyze the compression state of the rotating spherical hinge 5 but also verify whether the main pier 7 is eccentric.

[0054] The number of strain rosettes 3 is 4. The 4 strain rosettes are embedded on the side surface of the main pier 7, which is used to obtain the torque generated by the action on the pier column when the main beam rotates. The three-axis inclination sensor and the three-axis gyroscope are both installed at the center position of the top surface of the main beam 6. The three-axis sensor is used to obtain the deflection and rotation angle data of the main beam 6; the three-axis gyroscope is used to measure the angular acceleration and angular velocity of the rotation of the main beam 6.

[0055] Further, the three-axis gyroscope is internally provided with a temperature detection element, and through the built-in temperature compensator, the temperature drift problems of the gyroscope itself and the inclination sensor are solved.

[0056] Embodiment 2

[0057] As Figures 3 - 5 , based on the same inventive concept as Embodiment 1, the embodiment of the present invention discloses a risk intelligent judgment method. This method can collect data from key measuring points during the rotation process by using the structure provided in Embodiment 1 and perform risk intelligent judgment, including the following steps:

[0058] Determine risk indicators;

[0059] Obtain real-time collected data, calculate the risk values corresponding to each risk indicator according to the risk indicators, and generate a risk value time series;

[0060] Input the risk value time series into the trained risk prediction model to generate a predicted risk level signal. Among them, the risk indicators include one or more of stress risk, inclination risk, and overspeed risk; the overspeed risk situations include angular velocity overspeed or acceleration overspeed; the stress risk situations include the axial compressive stress of the outermost spherical hinge slider, the axial shear stress of the outermost spherical hinge slider, the torsional shear stress of the pier column, and the rotational traction force; the inclination risk includes the detected inclination angle, the inclination angle inversely calculated from the main tower verticality, the predicted inclination angle of the 4# spherical hinge slider, and the predicted inclination angle of the 5# spherical hinge slider. Among them, the predicted inclination angle can predict the deflection angle of the rotating bridge through the axial stress of the 4# and 5# spherical hinge sliders.

[0061] In one embodiment, the selection of key measurement points includes analyzing the correlation between the data of each measurement point and the rotation state based on the grey entropy theory, and selecting the measurement points with high correlation coefficients as key measurement points.

[0062] In this implementation, determine the risk evaluation indicators according to the key measurement points and establish a dynamic risk evaluation index system based on the movement process. Then, based on the GM model prediction theory and the comprehensive fuzzy algorithm, establish a dynamic monitoring and risk warning system for the rotation process.

[0063] Specifically, taking the data of the first 10 s of the indoor rotation model as an example, conduct a risk identification analysis of the rotation process, and the risk indicator data is shown in Table 1. σ Y

[0064] Table 1 Risk Indicator Data of Each Risk Indicator in the First 10 s of the Indoor Rotation Model

[0065]

[0066] Among them, σ Y : Axial compressive stress of the outermost ring spherical hinge slider. σ J : Circumferential shear stress of the outermost ring spherical hinge slider. σ N : Torsional shear stress of the pier column. F: Rotational traction force. θ Q : Inclination angle monitored by the inclination sensor. θ G : Inclination angle inversely calculated from the main tower verticality. σ 4# : Predicted inclination angle of the 4# spherical hinge slider. σ 5# : Predicted inclination angle of the 5# spherical hinge slider. ω: Rotation angular velocity. ω a : Rotation angular acceleration.

[0067] Establish a GM model:

[0068] The GM(1,1) prediction model is established for the axial compressive stress and circumferential shear stress of the outermost ball joint slider, the torsional shear stress, rotational traction, angular velocity and angular acceleration of the pier column, and the GM(1,N) prediction model is established for the deflection angle. The respective GM prediction models are solved, and the above-mentioned grey models are tested in this paper. The detection indicators include relative residual, variance ratio test and small error probability test.

[0069] The variance ratios of the gray model of the axial compressive stress and deflection angle of the outermost ball joint slider are 0.1525 and 0.2376, respectively, both less than 0.35, which indicates that the gray model of the axial compressive stress and deflection angle of the outermost ball joint slider has a high prediction accuracy. By collecting the axial compressive stress and deflection angle of the outermost ball joint slider at the 11th second of the indoor rotation model, it is found that the predicted values ​​of the gray model of the axial compressive stress and deflection angle of the outermost ball joint slider are very close to the calculated data, with a relative error range of 3.9% to 12.3%, indicating that the gray prediction model has a high accuracy in predicting the state of the rotation process.

[0070] According to the above method, by establishing grey prediction models for different indicators and conducting accuracy tests, it is found that the grey models of the axial compressive stress and circumferential shear stress of the outermost ball joint slider, the torsional shear stress of the pier column, the rotation traction force and the deflection angle have high accuracy and good prediction effect. The grey prediction model of angular velocity and angular acceleration is not accurate, and the initial data needs to be transformed by square root, logarithmic or translation. Finally, the predicted value of each indicator at the next time point can be obtained, as shown in Table 2.

[0071] Table 2 Data of risk indicators of indoor rotation model at the 11th second

[0072]

[0073]

[0074] It can be seen from Table 2 that the predicted values ​​of the grey model of each risk indicator are very close to the calculated data, and the grey predicted values ​​are slightly larger than the measured values ​​as a whole, with a relative error range of -4.41% to 20.69%, which shows that the grey prediction model is suitable and reliable for predicting the state of the rotation process.

[0075] The calculation steps of the GM(1,N) model are as follows:

[0076] 1) Set is the system characteristic data sequence, that is, the inclination angle measured by the triaxial inclinometer is the reference series. The engineering monitoring inclination angle and the inclination angle estimated by the axial stress of the 4# and 5# ball joint sliders are the related factor sequences:

[0077]

[0078]

[0079]

[0080] 2) Calculate the 1-AGO sequence. The 1-AGO sequence is an accumulative sequence. The first element of the 1-AGO sequence to be obtained is the first element of the known sequence, and the second element of the 1-AGO sequence is the sum of the first two elements of the known sequence, the third element is the sum of the first three elements of the known sequence, and so on.

[0081] Let The first-order accumulative (1-AGO) generated sequence of

[0082]

[0083]

[0084] 3) The adjacent mean generated sequence of the sequence is

[0085]

[0086] where X 1 (k) is the first-order accumulative generated value corresponding to the original sequence X0(k). Therefore, let be the reference column, be the comparison column, and the GM(1,N) model established accordingly is:

[0087]

[0088] In the formula, k = 1, 2, 3,..., n, the parameter a is the system development coefficient, is called the driving term, b i is the driving coefficient, is the system characteristic data sequence (i.e., the reference column), is 's adjacent mean generated sequence, is 's first-order accumulative (1-AGO) generated sequence, i = 1, 2,..., n.

[0089] 4) The least squares estimation of the parameter column of the GM(1,N) model differential equation satisfies:

[0090] a) When n < N + 1, [B T B] ≠ 0

[0091] b) When n = N + 1, [B] ≠ 0

[0092] c) When n > N + 1, [B T B] ≠ 0

[0093] Among them,

[0094]

[0095]

[0096] It is called

[0097]

[0098] the whiting equation (shadow equation) of the GM(1,N) model, and its solution is:

[0099]

[0100] By performing cumulative subtraction reduction on the formula and taking as the (k + 1)-th predicted value can be obtained as:

[0101]

[0102] According to the fuzzy algorithm above, the estimated risk level at the 11th second of the indoor rotation model can be obtained, which is 78.5 points. The risk level of the rotation is safe, and the risk has increased slightly compared to the previous second.

[0103] In this specification, each embodiment is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method part.

[0104] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A self-sensing rotation monitoring system, characterized in that It includes a self-sensing sensing system, which is composed of strain gauges, pressure cells, strain rosettes, three-axis inclinometers and three-axis gyroscopes, and is used to obtain the monitoring data of multiple components in the rotating bridge in real time; The multiple components in the rotating bridge include: rotating ball joints, main girders and main piers; among them, the rotating ball joint includes an upper ball joint, a lower ball joint and a plurality of ball joint sliders distributed between the upper ball joint and the lower ball joint.

2. The self-sensing swivel monitoring system according to claim 1, characterized in that, The upper ball joint and the lower ball joint are of steel-concrete structure. The strain gauges include 4 first strain gauges embedded in the steel-concrete bonding surface of the lower ball joint, 8 second strain gauges embedded in the steel-concrete bonding surface of the upper ball joint, and 35 third strain gauges embedded on the surface of the lower ball joint; The first strain gauge, the second strain gauge and the third strain gauge are respectively used to obtain the stress data at the steel-concrete bonding surface of the lower ball joint, the steel-concrete bonding surface of the upper ball joint and the surface of the lower ball joint.

3. The self-sensing swivel monitoring system according to claim 2, characterized in that, The strain gauges also include fourth strain gauges embedded on the side surface of the ball joint slider, and one fourth strain gauge is arranged on the side surface of each ball joint slider, which is used to obtain the axial stress at the ball joint slider.

4. The self-sensing swivel monitoring system according to claim 1, characterized in that, The number of the pressure cells is 4, and the 4 pressure cells are orthogonally embedded in the top surface of the upper ball joint along the longitudinal bridge direction and the transverse bridge direction, and are used to obtain the compression data of the upper ball joint.

5. The self-sensing swivel monitoring system according to claim 1, characterized in that, The number of the strain rosettes is 4, and the 4 strain rosettes are embedded on the side surface of the main pier, and are used to obtain the torque on the pier column when the main girder rotates.

6. The self-sensing swivel monitoring system according to claim 1, characterized in that, The three-axis inclinometer is installed at the central position of the top surface of the main girder, and is used to obtain the skew and rotation angle data of the main girder.

7. The self-sensing swivel monitoring system according to claim 1, characterized in that The three-axis gyroscope is installed at the central position of the top surface of the main girder, and the three-axis gyroscope is used to measure the angular acceleration and angular velocity of the rotation of the main girder.

8. A risk intelligent prediction method, characterized in that, It includes the following steps: Determine the risk indicators; Obtain the real-time collected data, calculate the risk values corresponding to each risk indicator according to the risk indicators, and generate a risk value time series; Input the risk value time series into the trained risk prediction model to generate a predicted risk level signal.

9. A risk intelligent prediction method according to claim 8, characterized in that, The steps also include: dynamically monitoring the collected data, and the dynamic monitoring includes: data screening, data verification and data storage.

10. A risk intelligent prediction method according to claim 8, characterized in that, The risk indicators include primary indicators and secondary indicators corresponding to the primary indicators; The primary indicators include: stress risk, inclination risk and overspeed risk, The secondary indicators corresponding to the stress risk include: axial compressive stress of the outermost ball joint slider, circumferential shear stress of the outermost ball joint slider, torsional shear stress of the pier column and rotational traction force; The secondary indicators corresponding to the inclination risk include: rotation inclination of the main girder and predicted inclination of the ball joint slider; The secondary indicators corresponding to the overspeed risk include: rotation angular velocity and rotation angular acceleration.

Images