Method for calculating temperature-dependent mid-span vertical displacement of girder bridge

Abstract

A method for calculating a temperature-dependent mid-span vertical displacement of a girder bridge includes: setting a joint rotation of a main girder at each support as an unknown quantity, and establishing an equation according to a bending moment equilibrium condition at the joint; then introducing a sequence to establish a quantitative relationship between each unknown quantity; substituting the relationship into the equation, to obtain an analytical formula for a rotation at each joint; establishing an analytical formula for a bending moment at each joint through a principle of superposition; and finally, establishing an analytical formula for a mid-span vertical displacement of each span girder through a principle of virtual work. This method provides an analytical formula with exact solutions for prismatic girder bridges which have equal side spans yet have any number of spans.

Description

CROSS REFERENCE TO THE RELATED APPLICATIONS

This application is based upon and claims priority to Chinese Patent Application No. 202010814388.4, filed on Aug. 13, 2020, the entire contents of which are incorporated herein by reference.

TECHNICAL FIELD

The disclosure relates to the technical field of structural analysis and monitoring of bridges, in particular to a method for calculating a temperature-dependent mid-span vertical displacement of a girder bridge.

BACKGROUND

Simply supported girder bridge and continuous girder bridge are the most common bridge forms in the transportation network, which are collectively referred to as “girder bridge” below. The structural deformation of the girder bridge is the focus of structural health monitoring during the operation period of the bridge. It is usually characterized by the expansion and contraction displacement at the girder ends and the mid-span vertical displacement of the main girder. There have been extensive studies on the girder-end expansion and contraction displacement of the girder bridge, which can be well estimated by the one-dimensional (1D) thermal expansion and contraction formula. However, there is much less research on temperature-induced mid-span vertical displacement of the main girder. Field monitoring shows that in sunny days, the top surface of the bridge is exposed to direct sunlight. This results in a significant vertical temperature difference between the top and the bottom surfaces of the girder, which affects the vertical deformation of the bridge and even changes the internal forces of the structure. In order to correctly understand the behavior of the girder bridge under temperature changes and provide a theoretical basis for the sensor layout in the bridge structural health monitoring system, it is necessary to study the dependency of the girder mid-span vertical displacement on the temperature difference between the top and bottom surfaces of the main girder.

At present, there are mainly two methods for calculating the thermal deformation of the girder bridge, namely theoretical analysis based on the principles of structural mechanics and numerical simulation based on the finite element (FE) calculation software. The theoretical analysis method usually calculates the thermal deformation by using the three-moment equation. It is inconvenient to solve the three-moment equation; also, the solution is of poor generality as a separate analysis is required for the structure of different span layouts. The numerical simulation method is time-consuming and laborious. It hardly reveals the mechanism behind the phenomenon, and demands necessary expertise, which restrains its on-site application by the field engineers. Regarding the mid-span vertical displacement of the girder bridge induced by the temperature difference between the top and bottom surfaces, there is no unified calculation formula with strong versatility, simple form and clear parameter relationship.

SUMMARY

The technical problem to be solved by the disclosure is to provide a method for calculating a temperature-dependent mid-span vertical displacement of a girder bridge.

This method is suitable for a straight bridge with a constant cross section, where main girders of each span are made of a same material, and a temperature difference between top and bottom surfaces of the main girders is equal everywhere along the bridge. If a total number of bridge spans is odd, the middlemost span is a main span and the rest are side spans; if the total number of bridge spans is even, the middlemost two spans are main spans and the rest are side spans. A length of the main span of the bridge may be different from that of the side spans, but the length of all side spans is the same. The calculation method includes the following steps:

(1) setting a joint rotation of a main girder at each support as an unknown quantity, and establishing an equation according to a bending moment equilibrium condition at the joint:

when the total number of the spans of the girder bridge is odd:

$\{\begin{array}{cc}4i_0·z_k+2i_0·z_{k+1}+M_T=0& \mathrm{when}k=1\left(1-1\right)\\ 2i_0·z_{k-1}+8i_0·z_k+2i_0·z_{k+1}=0& \mathrm{when}k=2,3,Lu-1\left(1-2\right)\\ 2i_0·z_{k-1}+\left(4i_0+2i_1\right)·z_k=0& \mathrm{when}k=u\left(1-3\right)\end{array}$

when the total number of the spans of the girder bridge is even:

$\{\begin{array}{cc}4i_0·z_k+2i_0·z_{z+1}+M_T=0& \mathrm{when}k=1\left(2-1\right)\\ 2i_0·z_{k-1}+8i_0·z_k+2i_0·z_{k+1}=0& \mathrm{when}k=2,3,Lu-1\left(2-2\right)\\ 2i_0·z_{k-1}+\left(4i_0+4i_1\right)·z_k=0& \mathrm{when}k=u\left(2-3\right)\end{array}$

where, z_k, z_k+1, and z_k−1 are unknowns of joint rotations, with clockwise rotation as positive; k is a subscript of the variable, k=1, 2, . . . , u; u is the number of the unknowns, u=┌n/2┐, which means the smallest integer not less than n/2; n is the total number of the spans of the girder bridge; because the structure and temperature change are symmetrical about the centerline of the bridge, the number of independent unknowns of joint rotations is u;

when the total number n of the spans is an odd number, a u-th span in the middlemost is designated as a main span, with a length of l₁, and the rest spans are side spans, with a length of l₀; when the total number n of the spans is an even number, a u-th span and a (u+1)-th span in the middlemost are designated as main spans, with a length of l₁, and the rest spans are side spans, with a length of l₀;

i₀ and i₁ are relative flexural stiffness of main girders of the side span and the main span respectively, defined as i₀=(E·I)/l₀ and i₁=(E·I)/l₁, where E is an elastic modulus of a main girder material, and I is an area moment of inertia of a main girder section;

M_T is a fixed-end bending moment of a single-span girder fixed at both ends, caused by a temperature difference between top and bottom surfaces of the main girder, with the bottom of the main girder being stretched as positive, M_T=(α·ΔT·E·I)/h, where α is a linear expansion coefficient of the main girder material; ΔT is the temperature difference between the top and bottom surfaces of the main girder, with a top surface temperature higher than a bottom surface temperature as positive; h is a section depth;

among the above equations, when u=1, only Eqs. (1-1) and (2-1) are valid, and the girder bridge has only the main span and no side spans, so it is assumed that l₀=l₁; when u=2, only Eqs. (1-1), (1-3), (2-1) and (2-3) are valid; when u≥3, Eqs. (1-1), (1-2), (1-3), (2-1), (2-2) and (2-3) are all valid;

(2) introducing a sequence to establish a quantitative relationship between the unknown joint rotations:

introducing a sequence {a_k}, which has a general term of:

$a_k=\frac{\lambda ·{\phi }_1^{k-1}+{\phi }_2^{k-1}}{\lambda ·{\phi }_1^k+{\phi }_2^k}$

where, as constants, φ₁=2+√{square root over (3)}, φ₂=2−√{square root over (3)}; λ is also a constant, which is equal to λ=λ₁=(√{square root over (3)}+ξ)/(√{square root over (3)}−ξ) for a girder bridge with an odd number of spans, and is equal to λ=λ₂=(√{square root over (3)}+2ξ)/(√{square root over (3)}−2ξ) for a girder bridge with an even number of spans; ξ is a side-to-main span ratio, that is, ξ=l₀/l₁; k=1, 2, . . . , u−1 represents a subscript of the sequence {a_k};

• • expressing the relationship between z_k and z_k with the help of the sequence {a_k} as follows:z_k=−a_u−k+1·z_k−1

where, the subscript k=2, 3, . . . , u;

(3) substituting the relationship between the variables in step (2) into the equation in step (1), to obtain an analytical formula for each joint rotation:

$z_k={\left(-1\right)}^k\frac{\alpha ·\Delta T·l_0}{2\sqrt{3}·h}·\frac{\lambda ·{\phi }_1^{u-k}+{\phi }_2^{u-k}}{\lambda ·{\phi }_1^{u-1}-{\phi }_2^{u-1}};$

where, k=1, 2, . . . , u; for a girder bridge with an odd number of spans, λ takes λ₁, and according to symmetry, z_2u+1−k=−z_k; for a girder bridge with an even number of spans, λ takes λ₂, and according to symmetry, z_2u+2−k=−z_k, z_u+1=0;

(4) using a result of step (3) to obtain an analytical formula for a bending moment at each joint by a principle of superposition, where

if a bending moment that makes the girder bottom stretched is assumed as a positive bending moment, when the total number of the spans of the girder bridge is an odd number, a bending moment M_k at a k-th joint is:

$M_k=\left[1+{\left(-1\right)}^k·\frac{{\lambda }_1·{\phi }_1^{u-k}-{\phi }_2^{u-k}}{\lambda ·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h},$ where, k=1, 2, . . . , u−1;

a bending moment M_u at a u-th joint is:

$M_u=\left[1+{\left(-1\right)}^u\frac{\xi }{\sqrt{3}}·\frac{{\lambda }_1+1}{\lambda ·{\phi }_1^{u-1}+{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h}$

according to symmetry, a bending moment at a (2u+1−k)-th joint is equal to the bending moment at the k-th joint, namelyM_2u+1−k=M_k, where, k=1,2, . . . ,u;

• • when the total number of the spans of the girder bridge is an even number, the bending moment M_k at the k-th joint is:

$M_k=\left[1+{\left(-1\right)}^k·\frac{{\lambda }_2·{\phi }_1^{u-k}-{\phi }_2^{u-k}}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h},$ where, k=1, 2, . . . , u;

a bending moment M_u+1 at a (u+1)-th joint is:

$M_{u+1}=\left[1+{\left(-1\right)}^{u+1}\frac{\xi }{\sqrt{3}}·\frac{{\lambda }_2+1}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h}$

according to symmetry, a bending moment at a (2u+2−k)-th joint is equal to the bending moment at the k-th joint, namelyM_2u+2−k=M_k, where, k=1,2, . . . ,u;

(5) using a result of step (4) to obtain an analytical formula for the mid-span vertical displacement of each span girder by a principle of virtual work, where

if it is assumed that the mid-span downward movement of the main girder corresponds to a positive displacement, when the total number of the spans of the girder bridge is an odd number, the mid-span vertical displacement ΔS_k of the k-th span girder is:

${\Delta }_k={\left(-1\right)}^k\frac{l_0^2·\alpha ·\Delta T}{16h}·\frac{\left({\lambda }_1·{\phi }_1^{u-k-1}+{\phi }_2^{u-k}\right)\left(1+\sqrt{3}\right)}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}},$ where, k=1, 2, . . . , u−1;

the mid-span vertical displacement ΔS_u of the u-th span girder is:

${\Delta }_u={\left(-1\right)}^u\frac{l_0^2·\alpha ·\Delta T}{8\sqrt{3}·\xi ·h}·\frac{{\lambda }_1+1}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}}$

according to symmetry, the mid-span vertical displacement ΔS_2u−k of the (2u−k)-th span girder is:ΔS_2u−k=ΔS_k, where, k=1,2, . . . ,u−1;

when the total number of the spans of the girder bridge is an even number, the mid-span vertical displacement ΔS_k of the k-th span girder is:

$\begin{array}{cc}{\Delta }_k={\left(-1\right)}^k\frac{l_0^2·\alpha ·\Delta T}{16h}·\frac{\left({\lambda }_2·{\phi }_1^{u-k-1}+{\phi }_2^{u-k}\right)\left(1+\sqrt{3}\right)}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}},& \end{array}$ where, k=1, 2, . . . , u−1;

the mid-span vertical displacement ΔS_u of the u-th span girder is:

${\Delta }_u={\left(-1\right)}^u\frac{l_0^2·\alpha ·\Delta T}{16\sqrt{3}·\xi ·h}·\frac{{\lambda }_2+1}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}}$

according to symmetry, the mid-span vertical displacement ΔS_2u+1−k of the (2u+1−k)-th span girder is:ΔS_2u+1−k=ΔS_k, where, k=1,2, . . . ,u.

The above girder bridge is a straight bridge, having a constant cross section and made of the same material. The temperature difference between the top and bottom surfaces of the main girders is equal everywhere along the bridge. If the total number of bridge spans is odd, the middlemost span is a main span and the rest are side spans. If the total number of bridge spans is even, the middlemost two spans are main spans and the rest are side spans. The length of all side spans of the bridge is the same.

When the length of each span of the girder bridge is identical:

the mid-span vertical displacement ΔS_k of the k-th span girder in a girder bridge with odd-numbered spans has a unified formula:

${\Delta }_k={\left(-1\right)}^k\frac{l_0^2·\alpha ·\Delta T}{16h}·\frac{\left({\phi }_1^{u-k}+{\phi }_2^{u-k}\right)\left(1+\sqrt{3}\right)}{{\phi }_1^u-{\phi }_2^{u-1}},$ where, k=1, 2, . . . , u;

according to symmetry, the mid-span vertical displacement ΔS_2u−k of the (2u−k)-th span girder is:ΔS_2u−k=ΔS_k, where, k=1,2, . . . ,u−1;

the mid-span vertical displacement ΔS_k of the k-th span girder in a girder bridge with even-numbered spans has a unified formula:

${\Delta }_k={\left(-1\right)}^k\frac{l_0^2·\alpha ·\Delta T}{16h}·\frac{\left({\phi }_1^{u-k}-{\phi }_2^{u-k+1}\right)\left(1+\sqrt{3}\right)}{{\phi }_1^u+{\phi }_2^u},$ where, k=1, 2, . . . , u;

according to symmetry, the mid-span vertical displacement ΔS_2u+1−k of the (2u+1−k)-th span girder is:ΔS_2u−1−k=ΔS_k, where, k=1,2, . . . ,u.

When the length of each span of the girder bridge is identical:

a change magnitude of the mid-span vertical displacement of the outermost span girder caused by the temperature difference between the top and bottom surfaces of the main girder is larger than that of the other spans; when the total number n of the spans increases, the mid-span vertical displacement of the outermost span girder approaches a limit value ΔS_lim^max:

${\Delta }_{l\mathrm{im}}^{\max }=\frac{\left(\sqrt{3}-1\right)l_0^2·\alpha ·\Delta T}{16h};$

the change magnitude of the mid-span vertical displacement of the middlemost span girder caused by the temperature difference between the top and bottom surfaces of the main girder is smaller than that of the other spans; when the total number n of the spans increases, the mid-span vertical displacement of the middlemost span girder approaches 0.

The above technical solution of the disclosure has the following beneficial effects.

The above solution provides a unified calculation formula for the mid-span vertical displacement of girder bridges with any number of spans induced by the temperature difference between the top and bottom surfaces of the main girder. The formula provides a simple way to calculate the exact thermal deformation, even with an electronic calculator. Also, the formula has general applicability and is convenient to perform parametric analysis. This method avoids cumbersome structural mechanics analysis, as well as finite element (FE) numerical simulation. It is very suitable for on-site calculations and verification of testing results, and thus has significant practical application value. More importantly, based on the calculation formula, this method reveals the universal law of the change of the internal force and structural deformation in the girder bridge with temperature, and gives the limit value of thermal deformation as the spans of the girder bridge increase, which has clear theoretical significance. In addition, the results obtained by this method can be used to guide the layout of measuring points in the structural health monitoring system of the girder bridge, and provide prior knowledge for the establishment of thermal deformation benchmark models.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is an analytical model for a girder bridge with odd-numbered spans according to an example of the disclosure.

FIG. 2 is an analytical model for a girder bridge with even-numbered spans according to an example of the disclosure.

DETAILED DESCRIPTION OF THE EMBODIMENTS

In order to make the to-be-solved technical problems, technical solutions and advantages of the disclosure clearer, the disclosure is described in detail below with reference to the accompanying drawings and specific examples.

The disclosure provides a method for calculating a temperature-dependent mid-span vertical displacement of a girder bridge.

The method includes the following steps:

(1) Set a joint rotation of a main girder at each support as an unknown quantity, and establish an equation according to a bending moment equilibrium condition at the joint:

when the total number of the spans of the girder bridge is odd:

$\{\begin{array}{cc}4i_0·z_k+2i_0·z_{k+1}+M_T=0\mathrm{when}k=1& \left(1-1\right)\\ 2i_0·z_{k-1}+8i_0·z_k+2i_0·z_{k+1}=0\mathrm{when}k=2,3,\mathrm{Lu}-1& \left(1-2\right)\\ 2i_0·z_{k-1}+\left(4i_0+2i_1\right)·z_k=0\mathrm{when}k=u& \left(1-3\right)\end{array}$

when the total number of the spans of the girder bridge is even:

$\{\begin{array}{cc}4i_0·z_k+2i_0·z_{k+1}+M_T=0\mathrm{when}k=1& \left(1-1\right)\\ 2i_0·z_{k-1}+8i_0·z_k+2i_0·z_{k+1}=0\mathrm{when}k=2,3,\mathrm{Lu}-1& \left(1-2\right)\\ 2i_0·z_{k-1}+\left(4i_0+4i_1\right)·z_k=0\mathrm{when}k=u& \left(1-3\right)\end{array}$

where, z_k, z_k+1, and z_k−1 are unknowns of joint rotations, with clockwise rotation as positive; k is a subscript of the variable, k=1, 2, . . . , u; u is the number of the unknowns, u=┌n/2┐, which is the smallest integer not less than n/2; n is the total number of the spans of the girder bridge.

When the total number n of the spans is an odd number, a u-th span in the middlemost is designated as a main span, with a length of l₁, and the rest spans are side spans, with a length of l₀; when the total number of the spans n is an even number, a u-th span and a (u+1)-th span in the middlemost are designated as main spans, with a length of l₁, and the rest spans are side spans, with a length of l₀.

i₀ and i₁ are relative flexural stiffness of main girders of the side span and the main span respectively, defined as i₀=(E·I)/l₀ and i₁=(E·I)/l₁, where E is an elastic modulus of a main girder material, and I is an area moment of inertia of a main girder section.

M_T is a fixed-end bending moment of a sing-span girder fixed at both ends, caused by a temperature difference between top and bottom surfaces of the main girder, with the bottom of the main girder being stretched as positive, M_T=(α·ΔT·E·I)/h, where α is a linear expansion coefficient of the main girder material; ΔT is the temperature difference between the top and bottom surfaces of the main girder, with a top surface temperature higher than a bottom surface temperature as positive; h is a section height.

(2) Introduce a sequence to establish a quantitative relationship between the unknown joint rotations:

A sequence {a_k} is introduced, which has a general term of:

$a_{\kappa }=\frac{\lambda ·{\phi }_1^{k-1}+{\phi }_2^{k-1}}{\lambda ·{\phi }_1^k+{\phi }_2^k}$

where, as constants, φ₁=2+√{square root over (3)}, φ₂=2−√{square root over (3)}; λ is also a constant, which is equal to λ=λ₁=(√{square root over (3)}+ξ)/(√{square root over (3)}−ξ) for a girder bridge with an odd number of spans, and is equal to λ=λ₂=(√{square root over (3)}+2ξ)/(√{square root over (3)}−2ξ) for a girder bridge with an even number of spans; ξ is a side-to-main span ratio, that is, ξ=l₀/l₁; k=1, 2, . . . , u−1 represents a subscript of the sequence {a_k}.

With the help of the sequence {a_k}, the relationship between z_k and z_k−1 is expressed as follows:z_k=a_u−k+1·z_k−1 where, the subscript k=2, 3, . . . , u.

(3) Substitute the relationship between the variables in step (2) into the equation in step (1), to obtain an analytical formula for each joint rotation:

$z_k={\left(-1\right)}^k\frac{\alpha ·\Delta T·l_0}{2\sqrt{3}·h}·\frac{\lambda ·{\phi }_1^{u-k}+{\phi }_2^{u-k}}{\lambda ·{\phi }_1^{u-1}-{\phi }_2^{u-1}}$

where, k=1, 2, . . . , u; for a girder bridge with an odd number of spans, λ takes λ₁, and according to symmetry, z_2u+1−k=−z_k; for a girder bridge with an even number of spans, λ takes λ₂, and according to symmetry, z_2u+2−k=−z_k, z_u+1=0.

(4) Use a result of step (3) to obtain an analytical formula for the bending moment of each joint by a superposition principle.

If a bending moment that makes the girder bottom stretched is assumed as a positive bending moment, when the total number of the spans of the girder bridge is an odd number, a bending moment M_k at a k-th joint is:

$M_k=\left[1+{\left(-1\right)}^k·\frac{{\lambda }_1·{\phi }_1^{u-k}-{\phi }_2^{u-k}}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h},$ where, k=1, 2, . . . , u−1;

a bending moment M_u at a u-th joint is:

$M_u=\left[1+{\left(-1\right)}^u\frac{\xi }{\sqrt{3}}·\frac{{\lambda }_1+1}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h}$

according to symmetry, a bending moment at a (2u+1−k)-th joint is equal to the bending moment at the k-th joint, namelyM_2u+1−k=M_k where, k=1,2, . . . ,u.

When the total number of the spans of the girder bridge is an even number, the bending moment M_k at the k-th joint is:

$M_k=\left[1+{\left(-1\right)}^k·\frac{{\lambda }_2·{\phi }_1^{u-k}-{\phi }_2^{u-k}}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h},$ where, k=1, 2, . . . , u;

a bending moment M_u+1 at a (u+1)-th joint is:

$M_{u+1}=\left[1+{\left(-1\right)}^{u+1}\frac{\xi }{\sqrt{3}}·\frac{{\lambda }_2+1}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h}$

according to symmetry, the bending moment at a (2u+2−k)-th joint is equal to the bending moment at the k-th joint, namelyM_2u+2−k=M_k, where, k=1,2, . . . ,u.

(5) Use a result of step (4) to obtain an analytical formula for the mid-span vertical displacement of each span girder by a principle of virtual work.

If it is assumed that the mid-span downward movement of the main girder corresponds to a positive displacement, when the total number of the spans of the girder bridge is an odd number, the mid-span vertical displacement ΔS_k of the k-th span girder is:

${\Delta }_k={\left(-1\right)}^k\frac{l_0^2·\alpha ·\Delta T}{16h}·\frac{\left({\lambda }_1·{\phi }_1^{u-k-1}+{\phi }_2^{u-k}\right)\left(1+\sqrt{3}\right)}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}},$ where, k=1, 2, . . . , u−1;

the mid-span vertical displacement ΔS_u of the u-th span girder is:

${\Delta }_u={\left(-1\right)}^u\frac{l_0^2·\alpha ·\Delta T}{8\sqrt{3}·\xi ·h}·\frac{{\lambda }_1+1}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}}$

according to symmetry, the mid-span vertical displacement ΔS_2u−k of the (2u−k)-th span girder is:ΔS_2u−k=ΔS_k, where, k=1,2, . . . ,u−1.

When the total number of the spans of the girder bridge is an even number, the mid-span vertical displacement ΔS_k of the k-th span girder is:

${\Delta }_k={\left(-1\right)}^k\frac{l_0^2·\alpha ·\Delta T}{16h}·\frac{\left({\lambda }_2·{\phi }_1^{u-k-1}+{\phi }_2^{u-k}\right)\left(1+\sqrt{3}\right)}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}},$ where, k=1, 2, . . . , u−1;

the mid-span vertical displacement ΔS_u of the u-th span girder is:

${\Delta }_u={\left(-1\right)}^u\frac{l_0^2·\alpha ·\Delta T}{16\sqrt{3}·\xi ·h}·\frac{{\lambda }_2+1}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}}$

according to symmetry, the mid-span vertical displacement ΔS_2u+1+k of the (2u+1−k)-th span girder is:ΔS_2u+1−k=ΔS_k, where, k=1,2, . . . ,u.

In a specific application, in step (1), the joint rotation of the main girder at each support is set as an unknown quantity to establish a bending moment equilibrium equation at the joint, specifically:

In an analysis model for a girder bridge with odd-numbered spans in FIG. 1, the total number of spans is n=2·u−1, where an integer u≥1, u=┌n/2┐ (i.e., the smallest integer not less than n/2). The u-th span is designated as the main span, with a length of l₁, and the rest spans are side spans, with a length of l₀, i₀ and i₁ are the relative flexural stiffness of the main girders of the side span and the main span respectively, i₀=(E·I)/l₀, i₁=(E·I)/l₁. E is an elastic modulus of a main girder material, and I is an area moment of inertia of a main girder section. h is a section depth of the main girder, ΔT is a temperature difference between the top and bottom surfaces of the main girder, and α is a linear expansion coefficient of the main girder material. If a side-to-main span ratio ξ=l₀/l₁ is introduced, i₁=i₀ξ.

The bending moment equilibrium equation of the main girder at each joint can be written according to the slope-deflection equation in structural mechanics. At joint 1, the bending moment equilibrium equation is as follows:4i₀·z₁+2i₀·z₂+M_T=0  (1)

In the equation, M_T is a fixed-end bending moment of a single-span girder fixed at both ends, caused by the temperature difference between the top and the bottom surfaces. If the bending moment which makes the girder bottom stretched is considered positive, by referring to the ready-made tables in textbooks, the fixed-end bending moment is

$\begin{array}{cc}{M}_T=\frac{\alpha ·\Delta T}{h}\mathrm{EI}& \left(2\right)\end{array}$

At joints k=2, 3, . . . , u−1, the bending moment equilibrium equation is as follows:2i₀·z_k−1+8i₀·z_k+2i₀·z_k+1=0  (3)

At joint u, the bending moment equilibrium equation is as follows:2i₀·z_u−1+(4i₀+4i₁)·z_u+2i₁·z_u+1=0  (4)

Because the structure and temperature change are symmetrical about the centerline of the girder, z_k=−z_2u+1−k, where k=1, 2, . . . , u, so Eq. (4) can be written as:2i₀·z_u−1+(4i₀+2i₁)·z_u=0  (5)

Meanwhile, the bending moment equilibrium equations at joints k=u+1, u+2, . . . , 2u are equivalent to Eqs. (1), (3) and (5). Therefore, the number of independent unknown joint rotations is u.

Similarly, in an analysis model for a girder bridge with even-numbered spans in FIG. 2, the total number of spans is n=2·u, where the integer u≥1. The u-th span and the (u+1)-th span in the middlemost are designated as main spans, with a length of l₁, and the rest spans are side spans, with a length of l₀. According to the slope-deflection equation in structural mechanics, the bending moment equilibrium equation of the main girder at joint 1 is the same as Eq. (1), the bending moment equilibrium equation of the main girder at joints k=2, 3, . . . , u−1 is the same as Eq. (3), and the bending moment equilibrium equation of the main girder at joint u is:2i₀·z_u−1+(4i₀+4i₁)·z_u+2i₁·z_u+1=0  (6)

Because the structure and temperature change are symmetrical about the centerline of the girder, z_k=−z_2u+2−k, where k=1, 2, . . . , u, and z_n+1=0. Thus, Eq. (6) can be written as:2i₀·z_u−1+(4i₀+4i₁)·z_u=0  (7)

Meanwhile, the bending moment equilibrium equation at joint k=u+1 is reduced to an identity, and the bending moment equilibrium equation at joints k=u+2, u+3, . . . , 2u+1 are equivalent to those written for the joints k=1, 2, . . . , u. Therefore, the number of independent unknown joint rotations is still u.

Among the above derivation, when u=1, only Eq. (1) is valid, and the girder bridge has only the main span and no side spans, so it is assumed that l₀=l₁. When u=2, only Eqs. (1) and (5) or Eqs. (1) and (7) are valid. When u≥3, the above equations are all valid.

In step (2), a sequence is introduced to establish a quantitative relationship between the unknown joint rotations, specifically:

For the girder bridge with odd-numbered spans, it is derived from Eq. (5) that

$\begin{array}{cc}{z}_u=-\frac{2i_0}{4i_0+2i_1}{z}_{u-1}=-\frac{1}{2+\xi }{z}_{u-1}& \left(8\right)\end{array}$

By introducing a₁=1/(2+ξ) and substituting z_u=−a₁·z_u−1 into Eq. (3), the relationship between z_k and z_k−1 (k=u−1, u−2, . . . , 2) can be rewritten as:z_k=−a_u−k+1·z_k−1  (9)

In the equation, the sequence a_k (k=2, 3, . . . , u−1) satisfies:

$\begin{array}{cc}{a}_k=\frac{1}{4-a_{k-1}}& \left(10\right)\end{array}$

Subtracting a constant φ from both sides of Eq. (10) yields

$\begin{array}{cc}{a}_k-\phi =\frac{1}{4-a_{k-1}}-\phi & \left(11\right)\end{array}$

Through an identical transformation, the right hand side of Eq. (11) is equivalent to

$\begin{array}{cc}\frac{1}{4-a_{k-1}}-\phi =\frac{1-\phi \left(4-a_{k-1}\right)}{4-a_{k-1}}=\frac{\phi a_{k-1}+1-4\phi }{4-a_{k-1}}=\frac{\phi \left(a_{k-1}-\phi \right)+1-4\phi +{\phi }^2}{4-a_{k-1}}& \left(12\right)\end{array}$

Let 1−4φ+φ²=0 in the numerator, and solutions to this quadratic equation are φ₁=2+√{square root over (3)} and φ₂=2−√{square root over (3)}. According to Eq. (11), the following two equations are obtained:

$\begin{array}{cc}{a}_k-{\phi }_1=\frac{{\phi }_1\left(a_{k-1}-{\phi }_1\right)}{4-a_{k-1}}& \left(13\right)\end{array}$ $\begin{array}{cc}{a}_k-{\phi }_2=\frac{{\phi }_2\left(a_{k-1}-{\phi }_2\right)}{4-a_{k-1}}& \left(14\right)\end{array}$

Dividing Eq. (13) by Eq. (14) leads to:

$\begin{array}{cc}\frac{a_k-{\phi }_1}{a_k-{\phi }_2}=\frac{{\phi }_1}{{\phi }_2}·\frac{a_{k-1}-{\phi }_1}{a_{k-1}-{\phi }_2}& \left(15\right)\end{array}$

Therefore, the sequence {(a_k−φ₁)/(a_k−φ₂)} is a geometric sequence with φ₁/φ₂ as a common ratio and (a₁−φ₁)/(a₁−φ₂) as the first term, thus:

$\begin{array}{cc}\frac{a_k-{\phi }_1}{a_k-{\phi }_2}={\left(\frac{{\phi }_1}{{\phi }_2}\right)}^{k-1}·\frac{a_1-{\phi }_1}{a_1-{\phi }_2}& \left(16\right)\end{array}$

As a₁=1/(2+ξ), the expression of a_k (k=1, 2, . . . , u−1) is obtained from Eq. (16):

$\begin{array}{cc}{a}_k=\frac{\lambda ·{\phi }_1^{k-1}+{\phi }_2^{k-1}}{\lambda ·{\phi }_1^k+{\phi }_2^k}& \left(17\right)\end{array}$

where the constants λ=λ₁=(√{square root over (3)}+ξ)/(√{square root over (3)}−ξ), φ₁=2+√{square root over (3)}, and φ₂=2−√{square root over (3)}.

For the girder bridge with even-numbered spans, it is derived from Eq. (7) that

$\begin{array}{cc}{z}_u=-\frac{2i_0}{4i_0+4i_1}{z}_{u-1}=-\frac{1}{2+2\xi }{z}_{u-1}& \left(18\right)\end{array}$

By introducing a₁=1/(2+2ξ) and following the same derivation process as that of the girder bridge with odd-numbered spans, Eq. (9) still applies to the relationship between z_k and z_k−1. With the replacement of λ=λ₁ with λ=λ₂=(√{square root over (3)}+2ξ)/(√{square root over (3)}−2ξ), the expression of a_k (k=1, 2, . . . , u−1), namely Eq. (17) is also valid.

In step (3), the relationship between the variables in step (2) is substituted into the equations in step (1), to obtain an analytical formula of each joint rotation. Specifically:

The general term of a_k in Eq. (17) produces:

$\begin{array}{cc}{a}_{u-1}=\frac{\lambda ·{\phi }_1^{u-2}+{\phi }_2^{u-2}}{\lambda ·{\phi }_1^{u-1}+{\phi }_2^{u-1}}& \left(19\right)\end{array}$

By substituting z₂=−a_u−1·z₁ into Eq. (1), z₁ is solved as:

$\begin{array}{cc}{z}_1=-\frac{M_T}{2\sqrt{3}·i_0}·\frac{\lambda ·{\phi }_1^{u-1}+{\phi }_2^{u-1}}{\lambda ·{\phi }_1^{u-1}-{\phi }_2^{u-1}}& \left(20\right)\end{array}$

The rest unknowns are further solved according to Eq. (9):

$\begin{array}{cc}{z}_k={\left(-1\right)}^k\frac{M_T}{2\sqrt{3}·i_0}·\frac{\lambda ·{\phi }_1^{u-k}+{\phi }_2^{u-k}}{\lambda ·{\phi }_1^{u-1}-{\phi }_2^{u-1}}& \left(21\right)\end{array}$

Eq. (21) is applicable for k=1, 2, . . . , u. For the girder bridge with odd-numbered spans, λ takes λ₁, and according to symmetry, z_2u+1−k=−z_k. For the girder bridge with even-numbered spans, λ takes λ₂ and according to symmetry, z_2u+2−k=−z_k and z_u+1=0.

In step (4), a result of step (3) is used to obtain an analytical formula for the bending moment at each joint by a superposition principle. Specifically:

The bending moment that makes the girder bottom stretched is assumed as a positive bending moment. For the girder bridge with odd-numbered spans, the bending moment M_k at the k-th joint (k=1, 2, . . . , u−1) isM_k=4i₀·z_k+2i₀·z_k+1+M_T  (22)

Substituting Eqs. (2) and (21) into Eq. (22) and simplifying lead to

$\begin{array}{cc}{M}_k=\left[1+{\left(-1\right)}^k·\frac{{\lambda }_1·{\phi }_1^{u-k}-{\phi }_2^{u-k}}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h}& \left(23\right)\end{array}$

Because of symmetry (z_u+1=−z_u), the bending moment at the u-th joint is:

$\begin{array}{cc}{M}_u=4i_1·z_u+2i_1·z_{u+1}+M_T=2i_1·z_u+M_T=\left[1+{\left(-1\right)}^u\frac{\xi }{\sqrt{3}}·\frac{{\lambda }_1+1}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h}& \left(24\right)\end{array}$

Because of symmetry, the bending moment at the (2u+1−k)-th joint (k=1, 2, . . . , u) is:M_2u+1−k=M_k  (25)

For the girder bridge with even-numbered spans, the bending moment M_k at the k-th joint (k=1, 2, . . . , u) can be calculated by Eq. (22). However, different from the case of the odd-numbered-span bridge, the expression of z_k uses λ₂ instead of λ₁. Thus, for the girder bridge with even-numbered spans, the bending moment M_k at the k-th joint (k=1, 2, . . . , u) is:

$\begin{array}{cc}{M}_k=\left[1+{\left(-1\right)}^k·\frac{{\lambda }_2·{\phi }_1^{u-k}-{\phi }_2^{u-k}}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h}& \left(26\right)\end{array}$

Because z_u+1=0, the bending moment M_u+1 at the (u+1)-th joint is:

$\begin{array}{cc}{M}_u=-2i_1·z_u-4i_1·z_{u+1}+M_T=-2i_1·z_u+M_T=\left[1+{\left(-1\right)}^u\frac{\xi }{\sqrt{3}}·\frac{{\lambda }_1+1}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}}\right]·\frac{\alpha \mathrm{EI}·\Delta T}{h}& \left(27\right)\end{array}$

Because of symmetry, the bending moment at the (2u+2−k)-th joint (k=1, 2, . . . , u) is:M_2u+2−k=M_k  (28)

In step (5), a result of step (4) is used to obtain an analytical formula for the mid-span vertical displacement of each span girder by a principle of virtual work. Specifically:

The mid-span downward movement of the main girder is assumed to be a positive displacement. When the total number of the spans of the girder bridge is an odd number, according to the principle of virtual work combined with a method of diagram multiplication, the mid-span vertical displacement of the k-th span girder (k=1, 2, . . . , u−1) is:

$\begin{array}{cc}{\Delta }_k=\frac{1}{\mathrm{EI}}·\frac{l_0}{4}·\frac{l_0}{2}·\frac{M_k^L+M_k^R}{2}-\frac{\alpha ·\Delta T}{h}\frac{l_0}{2}·\frac{l_0}{4}={\left(-1\right)}^k\frac{l_0^2·\alpha ·\Delta T}{16h}·\frac{\left({\lambda }_1·{\phi }_1^{u-k-1}+{\phi }_2^{u-k}\right)\left(1+\sqrt{3}\right)}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}}& \left(29\right)\end{array}$

Similarly, the mid-span vertical displacement of the u-th span girder is:

$\begin{array}{cc}{\Delta }_u=\frac{1}{\mathrm{EI}}·\frac{l_1}{4}·\frac{l_1}{2}·\frac{M_u^L+M_u^R}{2}-\frac{\alpha ·\Delta T}{h}\frac{l_1}{2}·\frac{l_1}{4}={\left(-1\right)}^u\frac{l_0^2·\alpha ·\Delta T}{8\sqrt{3}·\xi ·h}·\frac{{\lambda }_1+1}{{\lambda }_1·{\phi }_1^{u-1}-{\phi }_2^{u-1}}& \left(30\right)\end{array}$

According to symmetry, the mid-span vertical displacement of the (2u−k)-th span girder (k=1, 2, . . . , u−1) is:ΔS_2u−k=ΔS_k  (31)

When the total number of the spans of the girder bridge is an even number, the mid-span vertical displacement of the k-th span girder (k=1, 2, . . . , u−1) is:

$\begin{array}{cc}{\Delta }_k=\frac{1}{\mathrm{EI}}·\frac{l_0}{4}·\frac{l_0}{2}·\frac{M_k^L+M_k^R}{2}-\frac{\alpha ·\Delta T}{h}\frac{l_0}{2}·\frac{l_0}{4}={\left(-1\right)}^k\frac{l_0^2·\alpha ·\Delta T}{16h}·\frac{\left({\lambda }_2·{\phi }_1^{u-k-1}+{\phi }_2^{u-k}\right)\left(1+\sqrt{3}\right)}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}}& \left(32\right)\end{array}$

Similarly, the mid-span vertical displacement of the u-th span girder is:

$\begin{array}{cc}{\Delta }_u=\frac{1}{\mathrm{EI}}·\frac{l_1}{4}·\frac{l_1}{2}·\frac{M_u^L+M_u^R}{2}-\frac{\alpha ·\Delta T}{h}\frac{l_1}{2}·\frac{l_1}{4}={\left(-1\right)}^u\frac{l_0^2·\alpha ·\Delta T}{16\sqrt{3}·\xi ·h}·\frac{{\lambda }_2+1}{{\lambda }_2·{\phi }_1^{u-1}-{\phi }_2^{u-1}}& \left(33\right)\end{array}$

According to symmetry, the mid-span vertical displacement of the (2u+1−k)-th span girder (k=1, 2, . . . , u) is:ΔS_2u+1−k=ΔS_k  (34)

When the length of each span of the girder bridge is identical, ξ=1. For the girder bridge with odd-numbered spans, λ₁=(√{square root over (3)}+1)/(√{square root over (3)}−1), and Eqs. (29) and (30) for calculating the mid-span vertical displacement of each span girder can be unified as:

$\begin{array}{cc}{\Delta }_k={\left(-1\right)}^k\frac{l_0^2·\alpha ·\Delta T}{16h}·\frac{\left({\phi }_1^{u-k}+{\phi }_2^{u-k}\right)\left(1+\sqrt{3}\right)}{{\phi }_1^u-{\phi }_2^{u-1}}& \left(35\right)\end{array}$

where k=1, 2, . . . , u. According to symmetry, the mid-span vertical displacement of the (2u−k)-th span girder (k=1, 2, . . . , u−1) is calculated by Eq. (31).

When the length of each span of the girder bridge is equal, for the girder bridge with even-numbered spans, λ₂=(√{square root over (3)}+2)/(√{square root over (3)}−2), and Eqs. (32) and (33) for calculating the mid-span vertical displacement of each span girder can be unified as:

$\begin{array}{cc}{\Delta }_k={\left(-1\right)}^k\frac{l_0^2·\alpha ·\Delta T}{16h}·\frac{\left({\phi }_1^{u-k}+{\phi }_2^{u-k+1}\right)\left(1+\sqrt{3}\right)}{{\phi }_1^u-{\phi }_2^u}& \left(35\right)\end{array}$

where k=1, 2, . . . , u. According to symmetry, the mid-span vertical displacement of the 2u+1−k-th span girder (k=1, 2, . . . , u) is calculated by Eq. (34).

According to Eq. (35), for the odd-numbered-span girder bridge with a given u, the mid-span vertical displacement magnitude |ΔS_k| of each span girder is proportional to φ₁^u−k+φ₂^u−k. As φ₂=1/φ₁, two functions ƒ₁(x)=x+1/x and g₁(x)=φ₁^u−x are introduced to define a composite function, which leads to ƒ₁(g₁(k))=φ₁^u−k+φ₂^u−k. For k=1, 2, . . . , u, g₁(k) decreases with the increase of k and g₁(k)≥1. On the other hand, the derivative ƒ′₁(x)=1−1/x² implies that ƒ₁(x) is a monotonically increasing function in the region of x≥1. Therefore, ƒ₁(g₁(k)) decreases monotonically with the increase of k, that is, the outermost span girder (k=1) has the largest magnitude of the mid-span vertical displacement while the middlemost span girder (k=u) has the smallest magnitude.

Similarly, for the even-numbered-span girder bridge with a given u, according to Eq. (36), the mid-span vertical displacement magnitude |ΔS_k| of each span girder is proportional to φ₁^u−k−φ₂^u−k+1. A new function ƒ₂(x)=x−φ₂/x is introduced and composed with g₁(x)=φ₁^u−x, which leads to ƒ₂(g₁(k))=φ₁^u−k−φ₂^u−k+1. For k=1, 2, . . . , u, g₁(k) decreases with the increase of k and g₁(k)≥1. Meanwhile, the derivative ƒ′₂(x)=1+φ₂/x²>0, so ƒ₂ (g₁(k)) decreases monotonically with the increase of k. Therefore, for the girder bridge with even-numbered spans, the mid-span vertical displacement magnitude of the outermost span girder (k=1) is the largest, while those of the middlemost span girders (k=u and k=u+1) are the smallest.

In summary, for the girder bridges with either odd- or even-numbered spans, the outermost and the middlemost span girders respectively have the largest and smallest magnitudes of the mid-span vertical displacement induced by the temperature difference between the top and bottom surfaces of the girder.

For the main girder in the span of k=1, when u approaches infinity, Eqs. (35) and (36) both converge to a constant ΔS_lim^max:

$\begin{array}{cc}{\Delta }_{\lim }^{\max }=-\frac{\left(\sqrt{3}-1\right)l_0^2·\alpha ·\Delta T}{16h}& \left(37\right)\end{array}$

Therefore, the mid-span vertical displacement of the outermost span girder in a girder bridge has a limit value as the number of bridge spans increases, namely ΔS_lim^max:

Similarly, for the main girder in the span of k=u, when u approaches infinity, Eqs. (35) and (36) both converge to 0. Therefore, the mid-span vertical displacement of the middlemost span girder in a girder bridge also has a limit value as the number of bridge spans increases, namely 0.

The disclosure is described in detail below with reference to the specific examples.

Example 1

A 6-span continuous girder bridge has equal spans of 110 m. The prismatic girder is made of steel, and has a depth of 4.5 m. The temperature difference between the top and bottom surfaces of the main girder due to direct sunlight is 20° C. The mid-span vertical displacement of the 1^st and 3^rd spans needs to be determined.

As the number of spans is n=6, the abovementioned girder bridge has an even number of spans, and the number of independent unknown joint rotations is u=3. The main and side spans have identical length of l₀=l₁=110 m; the linear expansion coefficient of the main girder is α=1.2×10⁻⁵° C.⁻¹; the girder depth is h=4.5 m; and the temperature difference between the top and bottom surfaces of the main girder is ΔT=20° C. Substituting the above parameters into Eq. (36) leads to ΔS₁=−0.0295 m for k=1, and ΔS₃=−0.0016 m for k=3. It can be seen that the mid-span vertical displacement magnitude of the 1st span girder is greater than that of the 3^rd span girder, and the negative sign before the displacement value means that the mid-span deck moves upward.

Example 2

A girder bridge has n spans, each span being 110 m in length. The prismatic girder is made of steel, and has a depth of 4.5 m. The temperature difference between the top and bottom surfaces of the main girder is 20° C. For n=1, 2, . . . , 10, the mid-span vertical displacement of the outermost span girder needs to be calculated.

It can be seen from the above that l₀=l₁=110 m, α=1.2×10⁻⁵° C.⁻¹, h=4.5 m, ΔT=20° C., and k=1. According to the parity of the span number n, these parameters are substituted into Eqs. (35) or (36) with u=┌n/2┐ (i.e., the smallest integer not less than n/2). As a result, the mid-span vertical displacement of the outermost span girder corresponding to different total span numbers is obtained, that is, −0.0807 m, −0.0202 m, −0.0323 m, −0.0288 m, −0.0297 m, −0.0295 m, −0.0295 m, −0.0295 m, −0.0295 m, and −0.0295 m, which are listed in the order of increasing n. From n=6, the mid-span vertical displacement of the outermost span girder is close to −0.0295 m, which is the limit value calculated by Eq. (37). This example clearly indicates that when the continuous girder bridge has a large number of spans, the girder mid-span vertical displacement caused by the temperature difference between the top and the bottom surfaces of the main girder is small, and is thus difficult to be accurately measured by the commonly used equipment.

The above described are preferred implementations of the disclosure. It should be noted that a person of ordinary skill in the art may further make several improvements and modifications without departing from the principle of the disclosure, but such improvements and modifications should also be deemed as falling within the protection scope of the disclosure.

Claims

1. A method for calculating a temperature-dependent mid-span vertical displacement of a girder bridge, comprising the following steps: (1) setting a joint rotation of a main girder at each support as an unknown quantity, and establishing an equation according to a bending moment equilibrium condition at a joint:when a total number of a plurality of spans of the girder bridge is an odd number:

2. The method according to claim 1, wherein the girder bridge is a straight bridge, and the girder bridge has a constant cross section and made of an identical material;the temperature difference between the top surface and the bottom surface of the main girder is equal everywhere along the girder bridge;if the total number of the plurality of spans is the odd number, a middlemost span is the main span and the plurality of spans excluding the middlemost span are the side spans;if the total number of the plurality of spans is the even number, two middlemost spans are the main span and the plurality of spans excluding the two middlemost spans are the side spans; andthe side spans of the girder bridge each have an identical length.

3. The method according to claim 1, wherein when the plurality of spans of the girder bridge each have an identical length,the temperature-dependent mid-span vertical displacement ΔSk of the main girder of the k-th span in the girder bridge with odd-numbered spans has a unified formula:

4. The method according to claim 1, wherein when the plurality of spans of the girder bridge each have an identical length:the temperature-dependent mid-span vertical displacement of the main girder of an outermost side span has a largest change magnitude caused by the temperature difference between the top surface and the bottom surface of the main girder;when the total number n of the plurality of spans increases, the temperature-dependent mid-span vertical displacement of the main girder of the outermost side span approaches a limit value ΔSlimmax: