Method for determining the structural integrity of an infrastructural element

Abstract

The invention relates to a method for determining the structural integrity of an infrastructural element, comprising the steps of: measuring deformations, such as displacements or rotations, during a predetermined time period with deformation measurement means arranged at or near a main structural body of the infrastructural element, in particular supports of the main structural body, characterized by determining the load configuration of the main structural body over the course of the predetermined time period, such as the load configuration concerning the loading perpendicular to a longitudinal direction of the main structural body, calculating the bending stiffness (EI) of the main structural body over the course of the predetermined time period, from the load configuration and deformations measured by the deformation measurement means, and comparing the bending stiffness (EI) at the end of the predetermined time period to the bending stiffness (EI) at the start of the predetermined time period to establish a difference in bending stiffness (EI) over the course of the predetermined time period.

Claims

1. A method for determining the structural integrity of a bridge, comprising the steps of: measuring deformations using sensors positioned at supports of a bridge deck of the bridge; determining a load configuration of the bridge deck, the load configuration including loading perpendicular to a longitudinal direction of the bridge deck, wherein the load configuration is derived from reaction forces measured by the sensors positioned at the supports; calculating bending stiffness (EI) of the bridge deck at each of a beginning and an end of a predetermined time period, from the load configuration and deformations measured by the sensors; and comparing the calculated bending stiffness (EI) at the end of the predetermined time period to the calculated bending stiffness (EI) at the beginning of the predetermined time period to determine a change in the bending stiffness (EI) over the predetermined time period; wherein the change in the bending stiffness (EI) of the bridge deck over the predetermined time period is calculated based on a moving load passing over the bridge deck, comprising the steps of: determining a type of the moving load, wherein the moving load is a vehicle having axles; determining a magnitude of axle loads of the moving load, wherein the axle loads have a static component, a dynamic component and a noise component, and wherein the dynamic component and the noise component of the axle loads are excluded from the calculation of the change in the bending stiffness (EI); determining a position of the moving load; determining a speed of the moving load, wherein the speed of the moving load over the bridge deck is constant; calculating influence lines; establishing kinematic relations between the load configuration of the bridge deck caused by the moving load and the reaction forces measured by the sensors; and calculating the change in the bending stiffness (EI) of the bridge based on the load configuration resulting from the moving load and the deformations measured by the sensors positioned at the bridge supports.

2. The method according to claim 1, wherein the bridge supports comprise bridge bearings and the sensors are arranged at the bridge bearings, and wherein the sensors establish the reaction forces at the bridge bearings from an elastic deformation and a spring constant of the bridge bearings.

3. The method according to claim 1, further comprising the step of inspecting the bridge when the change in the bending stiffness (EI) exceeds a predetermined value.

4. The method according to claim 1, wherein the deformations include at least one of displacements and rotations.

5. The method according to claim 1, wherein the change in bending stiffness calculated is utilized as a global indicator of an amount of occurred damage.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The present invention will be explained hereafter with reference to the drawings. Therein:

(2) FIG. 1 shows the development of the equivalent bending stiffness over the years;

(3) FIG. 2 shows a simply supported beam model;

(4) FIG. 3 shows influence lines support reactions;

(5) FIG. 4 shows summed reaction forces;

(6) FIG. 5 shows influence line rotations at support;

(7) FIG. 6 shows an orthotropic plate on spring supports; and

(8) FIG. 7 shows support rotations.

DETAILED DESCRIPTION

(9) In FIG. 1 the development of an “equivalent bending stiffness” of a bridge deck can be seen in a period from 1995 (commissioning) up until 2040. A rapid decrease in bending stiffness can be observed from 1995 to 1997 which is caused by initial changes to the structure (e.g. expected cracking of the concrete with small crack width). After this a gradual decrease of the bending stiffness from year 1997 2019 can be seen. Further a sudden drop in 2019-2020 and a further gradual decrease from 2020 2040 are shown.

(10) The gradual decrease in bending stiffness is expected to be caused by gradual deterioration of the concrete and reinforcement/pre-stressing steel while the sudden drop is expected to be caused by an unexpected event like an earthquake, extremely heavy transport or damaging of a pre-stressing tendon. The bending stiffness in this graph has been calculated at certain moments in time when a load is present on the structure.

(11) The Simply Supported Beam Model

(12) As stated before, the simply supported beam model will now be explained in more detail with reference to FIG. 2.

(13) A simply supported beam with length L will be considered. A number N of point loads each representing the axle of a truck with magnitude P.sub.i will move over this beam in the x-direction all with the same speed v. The spacing distance of each axle with respect to the first one is called d.sub.i. By definition d.sub.1=0. At t=0 the location of P.sub.1 will be by definition at x=0.

(14) From the given model the support reactions at support A and support B will be determined as a function of time. To do this first the influence lines for the support reactions are determined. These influence lines are given by the magnitude of the support reaction at the concerning support resulting from a unit load located at location x:

(15) $\begin{array}{cc}\mathrm{For}0>x>L& I_{R,A}=I_{R,B}=0\\ \mathrm{For}0<x<L& I_{R,A}=1-\frac{x}{L}\\ & I_{R,B}=\frac{x}{L}\end{array}$
The influence lines for the reaction force at both supports are shown in FIG. 3.
From these influence lines which are a function of x the influence lines are determined as a function of time. This is done by substitution of x.sub.n:

(16) $\begin{array}{cc}& x_n=\left(v.Math.t\right)-d_n\\ \mathrm{For}0>\left(v.Math.t\right)-d_n>L& I_{R,A}=I_{R,B}=0\\ \mathrm{For}0<\left(v.Math.t\right)-d_n<L& I_{R,A}=1-\frac{\left(v.Math.t\right)-d_n}{L}\\ & I_{R,B}=\frac{\left(v.Math.t\right)-d_n}{L}\end{array}$

(17) Multiplying the results with P.sub.n gives the support reactions as a function of t for the passing of a single axle. The summation of N of these equation results in the total support reaction at A and B as a function of time.

(18) 0$R_A\left(t\right)=\underset{n=1}{\overset{N}{.Math.}}{P}_n.Math.I_{R,A}$$R_B\left(t\right)=\underset{n=1}{\overset{N}{.Math.}}{P}_n.Math.I_{R,B}$

(19) These two reaction forces summed up results in the total reaction force of the bridge deck:

(20) $R_{A+B}\left(t\right)=\underset{\underset{0<\left(v.Math.t\right)-d_n<L}{n=1}}{\overset{N}{.Math.}}{P}_n$

(21) Therein, the total vertical reaction force of the bridge should be equal to the sum of the axle loads of the axles that are present on the bridge deck.

(22) Axle Load Magnitude

(23) As can be seen in FIGS. 2 and 3 the magnitude of the axle load is equal to the change in support reaction A upon entering the bridge or the change in support reaction B upon leaving the bridge (ΔR.sub.i,A and ΔR.sub.i,B). This change is also equal to the change in the total reaction force (R.sub.i,A+B). By measuring these changes for each individual axle or axle group it is possible to determine the magnitude of the axle loads or axle group loads.

(24) Axle Load Location

(25) X-Location

(26) By looking for a two corresponding changes in support reaction at support A and support B (ΔR.sub.i,A and ΔR.sub.i,B), or two corresponding changes in the sum of support reactions (R.sub.i,A+B) it is possible to identify the time moment at which an axle enters t.sub.i,e and leaves t.sub.i,l the bridge. From these two time moments the axle speed can be determined:

(27) $v_i=\frac{L}{t_{i,l}-t_{i,e}}.$
As we now have a time moment of entering the bridge and a speed the x-location can be calculated at any time moment:
x.sub.i=v.sub.i.Math.(t−t.sub.i,e)

(28) It should be noted that this method to determine the x-location of an axle will only work under the assumption that there are no axles overtaking each other while on the bridge.

(29) Y-Location

(30) Since the simply supported beam model is a one-dimensional model there is no y-location to be determined. The y-location will be determined for the two-dimensional models in the following chapters.

(31) The Orthotropic Plate on Spring Supports

(32) The orthotropic plate on spring supports is the model that is closest to a real bridge deck. The change in reaction forces due to a change in stiffness of the bearings has a significant impact on the applicability of the method of obtaining the axle load and locations. In this model an orthotropic slab will be modelled with the following properties:

(33) $I_x=1.0.Math.{10}^{10}{\mathrm{mm}}^4/m$$I_y=1.0.Math.{10}^9{\mathrm{mm}}^4/m$$\frac{I_y}{I_x}=\frac{1}{10}$

(34) The element properties are shown in the box below:

(35) $\mathrm{Modulus}\mathrm{of}\mathrm{elasticity}\text{:}$$E_{\mathrm{element}}=E_{\mathrm{deck}}=32.800\mathrm{MPa}\left(\mathrm{uncracked}C30/37\right)$$\mathrm{Shear}\mathrm{modulus}\text{:}{G}_{\mathrm{xy},\mathrm{element}}=\frac{E_{\mathrm{deck}}}{2\left(1+v\right)}\sqrt{\frac{I_{y,\mathrm{deck}}}{1_{x,\mathrm{deck}}}}=4321.8\mathrm{MPa}$$\mathrm{Element}\mathrm{thickness},x\text{:}{d}_{x,\mathrm{element}}=\sqrt[3]{12I_{x,\mathrm{deck}}}=493.2\mathrm{mm}$$\mathrm{Element}\mathrm{thickness},y\text{:}{d}_{y,\mathrm{element}}=\sqrt[3]{12I_{y,\mathrm{deck}}}=228.9\mathrm{mm}$

(36) Further the plate is supported on fourteen spring supports in total. The stiffness of these spring supports is varied between 2500, 1000 and 500 kN/mm. These stiffnesses have been chosen for the following reasons: 2500 kN/mm, because this is an approximation for the lower boundary stiffness for a reinforced rubber bearing with dimensions 350×280×45, 1000 and 500 kN/mm, because it is expected that a lower stiffness will cause a larger change in support reactions compared to the infinitely stiff case.

(37) All assumptions that hold for the isotropic bridge deck also hold for the orthotropic bridge deck on spring supports. The properties of this bridge deck are the following: Length: 20 m Width: 15 m Model type: Two heights Thickness, x: As described Thickness, y: As described E-modulus: 32.800 MPa (uncracked concrete C30/37) k: Varied between 2500, 1000 and 500 kN/mm

(38) The resulting support reactions from the axle passing over the lanes are then calculated for each of the supports.

(39) It can be seen that there is hardly any difference between an orthotropic plate with infinitely stiff supports and the orthotropic plate with spring supports. Again the reaction forces at x=0 and x=L are summed up. This sum of the reaction forces at x=0 and x=L are shown in FIG. 4.

(40) Axle Load Magnitude

(41) It can be concluded that the axle load magnitude can be found in exactly the same manner as for the simply supported beam.

(42) Axle Load Location

(43) X-Location

(44) Since the bridge deck is still simply supported the summed reaction forces at x=0 and x=L remain the same. This means that nothing changes for the determination of the x-location.

(45) Y-Location

(46) The y-location is determined in the same way as for an infinitely stiff plate, an isotropic plate and an orthotropic plate. It can be seen in table 1 that it is still possible to determine this location very accurately. Again since nothing has changed about the moment equilibrium this is not surprising:

(47) TABLE-US-00001 TABLE 1 y-locations Real y-location y-location k Lane y-location x = 0 x = L [kN/mm] nr. [mm] [mm] [mm] 2500 1 1875 1859 1859 −0.9% −0.9% 2 5625 5653 5641 +0.5% +0.3 1000 1 1875 1854 1859 −1.1% −0.9% 2 5625 5639 5628 +0.2% +0.1%  500 1 1875 1887 1893 +0.6% +0.1% 2 5625 5712 5662 +1.5% +0.7%
Determining the Bending Stiffness (EI)

(48) A value for the bending stiffness of the bridge deck can be determined for a vehicle passing over this bridge deck. This bending stiffness provides information about the structural condition of the bridge deck. This will be done by looking at the actual (measured) deformations of the bridge deck and comparing these too the expected deformations of this bridge deck for different models of statically determined bridge decks. The two simple models as discussed before will be considered. Below an explanation will be given on how to translate an input in terms of: The magnitude of the axle loads Defined as the static component of the axle force exerted on the bridge deck in kN The x-location of the axles as a function of time Defined as a time of entering and leaving the bridge Speed is assumed to be constant The y-location of the axle Defined as a number assigned to the lane on which the axle is present y-location is assumed to be constant The measured rotation at the support Determined by measuring two deformations on opposite sides of the support

(49) To an output in terms of: The equivalent bending stiffness of the bridge deck Defined as a value of EI.sub.eq in Nmm.sup.2

(50) The models that will be considered in this chapter are the following: Simply supported one dimensional beam Orthotropic plate on spring supports

(51) These models are considered because the simply supported one dimensional beam is the most simple possible model for a bridge, and the orthotropic plate on springs supports is the most advanced model.

(52) The Simply Supported Beam

(53) For the same model for the simply supported beam as described earlier, the equivalent bending stiffness will be calculated as a function of the loading determined in the same section, and the measured rotations at the support. To do this first the influence line for rotations at a single support is determined. This influence line is found using a “vergeet-me-nietje” as shown below:

(54) The “vergeet-me-nietje” for θ.sub.1 has been used where a is substituted by x, and b is substituted by (L−x). Further the equation for θ.sub.1 has been divided by P.

(55) $\begin{array}{cc}\mathrm{For}0>x>L& I_{\phi ,\mathrm{support}A}=0\\ \mathrm{For}0<x<L& I_{\phi ,\mathrm{supportA}}=\frac{-x\left(L-x\right)\left(2L-x\right)}{6\mathrm{EIL}}=\frac{-x^3+3{\mathrm{Lx}}^2-2L^{2}x}{6\mathrm{EIL}}\end{array}$

(56) The resulting influence line for the rotations at the left support (support A) is shown in FIG. 5 for a beam with a length of 1000 mm and a EI of 1.000.000 Nmm.sup.2. Multiplying the influence line with the magnitude of the load n, P.sub.n, and changing x to x.sub.n gives us the rotation at the support due to load n. Summing up all these rotations for load n=1 to n=N gives us the total rotation.

(57) $\begin{array}{cc}\mathrm{For}0>x>L& {\phi }_{A,n}=0\\ \mathrm{For}0<x<L& {\phi }_{A,n}=\frac{P_n.Math.\left(-x_n^3+L^2{x}_n\right)}{6\mathrm{EIL}}\\ & \begin{array}{c}{\phi }_A=\underset{n=1}{\overset{N}{.Math.}}\left({\phi }_{A,n}\right)=\underset{n=1}{\overset{N}{.Math.}}\left(\frac{P_n.Math.\left(-x_n^3+L^2{x}_n\right)}{6\mathrm{EIL}}\right)=\\ \frac{1}{\mathrm{EI}}\underset{n=1}{\overset{N}{.Math.}}\left(\frac{P_n.Math.\left(-x_n^3+L^2{x}_n\right)}{6L}\right)\end{array}\end{array}$

(58) Now the bending stiffness can be determined as:

(59) $\mathrm{EI}=\frac{1}{\phi A}\underset{n=1}{\overset{N}{.Math.}}\left(\frac{P_n.Math.\left(-x_n^3+L^2{x}_n\right)}{6L}\right)$

(60) Now the output variables calculated before together with the measured support rotations can be used to determine the bending stiffness of the simply supported beam as a function of time.

(61) The Orthotropic Plate on Spring Supports

(62) The same method will be applied for the orthotropic plate on spring supports, with the only difference that for each individual lane a separate influence line has to be determined. These influence lines are preferably determined using Scia Engineer instead of hand calculations. The rotations are determined at the central support, indicated in FIG. 6. Rotations are considered in the x-direction. From the Scia engineer model the influence lines for the rotations are derived. This is done by placing an axle load of 1 kN along the defined lanes and determining the rotation at the indicated support. This is shown in FIG. 7.

(63) The influence lines for the rotations at the support are found by performing a fourth order polynomial fit for the obtained data points. A fourth order polynomial has been chosen because this gave a near-perfect fit for the data points. A higher order polynomial would of course give a better fit but this would make the equations harder to handle. As discussed before, this results in the following equations:
I.sub.φ,Lane 1 & 4=9.0071.Math.10.sup.−11.Math.x.sup.4−2.8203.Math.10.sup.−9.Math.x.sup.3−1.9723.Math.10.sup.−8.Math.x.sup.2+8.0688.Math.10.sup.−7.Math.x−2.7327.Math.10.sup.−7
I.sub.φ,Lane 2 & 3=−5.7965.Math.10.sup.−11.Math.x.sup.4+4.8628.Math.10.sup.−9.Math.x.sup.3−1.5914.Math.10.sup.−7.Math.x.sup.2+1.7249.Math.10.sup.−6.Math.x−6.8949.Math.10.sup.−7

(64) Now to find a value for the bending stiffness (EI) the same procedure as for the simply supported beam model is followed. However this time there is no EI in the equation. To solve this problem the following ratio is added:

(65) $\frac{{\mathrm{EI}}_{\mathrm{theoretical}}}{{\mathrm{EI}}_{\mathrm{eq}}}.$
In this ratio EI is the theoretical bending stiffness of the bridge deck used in the Scia model, while EI.sub.eq is the bending stiffness we want to obtain using the method:

(66) $I_{\phi ,\mathrm{Lane}1\&4}=\hspace{1em}\hspace{1em}\frac{{\mathrm{EI}}_{\mathrm{theoretical}}}{{\mathrm{EI}}_{\mathrm{eq}}}\left(9.0071.Math.{10}^{-11}.Math.x^4-2.8203.Math.{10}^{-9}.Math.x^3-1.9723.Math.{10}^{-8}.Math.x^2+8.0688.Math.{10}^{-7}.Math.x-2.7327.Math.{10}^{-7}\right)$$I_{\phi ,\mathrm{Lane}2\&3}=\frac{{\mathrm{EI}}_{\mathrm{theoretical}}}{{\mathrm{EI}}_{\mathrm{eq}}}\left(-5.7965.Math.{10}^{-11}.Math.x^4+4.8628.Math.{10}^{-9}.Math.x^3-1.5914.Math.{10}^{-7}.Math.x^2+1.7249.Math.{10}^{-6}.Math.x-6.8949.Math.{10}^{-7}\right)$

(67) Again the output values calculated before together with the measured support rotation at the indicated support can be used to extract values for the EI.sub.eq. These values should be looked at individually per lane.

(68) In practice it would be difficult to find the exact influence lines from a theoretical model due to the model parameters being largely unknown. It would be easier to obtain these lines by means of calibration. This would be done by having a vehicle of known, large weight with a known axle configuration and speed pass over the bridge deck and measuring the support rotations. Now the resulting rotations should be a superposition of a multiple influence lines for the considered lane.

(69) It should be clear that the description above is intended to illustrate the operation of embodiments of the invention, and not to reduce the scope of protection of the invention. Starting from the above description, many embodiments will be conceivable to the skilled person within the inventive concept and scope of protection of the present invention. Although the detailed description describes the method with reference to a bridge and a bridge deck, the method is also applicable to other infrastructural elements, such as a quay wall, a dike, a water supply duct, a sewer system duct, an electricity line, a road, a lock or a foundation.