STUDY OF CABLE-STAYED BRIDGES FOR PIPELINES

Abstract

Cable-stayed bridges for pipelines (CSP-Bridges) are generally narrow and need additional stiffness in the horizontal plane. These bridges serve for the transportation of oil, natural gas, and water and may be called as energy bridges. These are specialised structure, carrying light loads and simultaneously withstand high wind pressures. As in any cable-stayed bridge, it has inclined stays emanating from one or more points in the pylons and holding the deck of the bridge at intermediate locations between the main supports, thus imparting a high degree of vertical stiffness to the bridge. Information available for cable-stayed bridges for pipelines is very limited. There are also no suitable guidelines for the designer to select the geometry of the bridge, dimensions and the sectional properties of the elements of the bridge. The work contained in this thesis is a step towards generating some of the behavioural informations on the CSP-bridges.Keeping in view the size of the pipelines and access required for repair and maintenance gangs, the width of the bridge has been adopted between 2.5 m and 7.5 m, so that the plan aspect ratio (= span/width) of the bridge in the range 25 to 35. The study in this thesis covers two types of bridges as follows; Type-A: Single span with towers at the ends, as used in the hills and, Type-B: Three span with towers in-between, as commonly used in the plains. In Type-B Bridges two cases have been investigated : i Without any horizontal "offsets" at the towers for supporting the stiffening cable and with an "offset" projecting horizontally at right angles on both sides of the bridge axis at the tower locations and supporting the stiffening cables. (iii) Abridge span range of50m to 400m total span (main span 50m to 200m for Type-A and 55m to 220m for Type-B) has been considered, which is close to the reported economical span range of 90m to 270m for cable-stayed bridges [124,150]. As narrow bridges have high degree of susceptibility to wind/earthquake oscillations, the effectiveness of horizontal stiffening arrangements needed to stabilize the bridge against lateral loads is investigated systematically. The stiffening arrangements studied are (1) cables and (2) cable-trusses of different configurations in the horizontal plane. The most effective location for the cable connection with the deck was first investigated and the same has been used for the following four cases studied. Case-I : Unstiffened deck without pipeline Case-II : Unstiffened deck with pipeline Case-Ill: Stiffened deck without pipeline. Case-IV : Stiffened deck with pipeline. The criteria adopted for the design of the unstiffened bridge deck for the lateral stiffness is that its deflection under the lateral load at basic wind speed of 44 m/s should be about 1/180 of the span. Static analysis for various loading cases has been carried out by (1) considering the structure of be linearly elastic, and (2) including the geometrical non-linearity due to cable sag. The stiffness matrix method has been used for both static and dynamic analysis. Dynamic analysis study includes the dynamic behaviour of CSP-Bridges under wind as well as (IV) earthquake loads. To carry out the dynamic analysis the mode shapes and natural frequencies of vibration were first determined by a three dimensional free vibration analysis which uses the Inverse Iteration technique coupled with Strum sequence property of the characteristic polynomials of the eigen value problem. In all, 15 modes of vibration have been considered in the analysis and the dynamic response is computed for a specified base motion. However, the dominant modes are only 2 to 5. The maximum seismic displacement responses (SRSS-values) for the six bridges have been evaluated using the average response spectra specified in the IS: code(IS:1893-1984) for 2% damping (Zone -V, I = 1.5, (3 = 1.0 for hills and 1.2 for plains). The wind loading on the CSP-bridges is considered in two parts; the static (mean) wind loads due to the steady component of wind and the fluctuating wind loads due to the horizontal gustiness of the wind. The response to the fluctuating load is determined using statistical concepts of stationary time series. The vertical vibration (cross-wind) caused by wind is insignificant due to the use of highly perforated decks in pipeline bridges coupled with the large vertical stiffness available in the bridge system. The dynamic analysis for wind loads has been carried out using Daveport's approach. However, some modifications had to be made for the cable-stayed bridges.Both the bridge configurations (Type A and B) have been analyzed using the same procedure. The non-linearity in vertical cables are found less then 1% and all six bridges are found within the specified limit of deflection to span ratio i.e. 1/225. The maximum deflection response under earthquake loading for bridges #Ai, #A2, M3, #Bj, #B2 and #63 are found as 26.6 mm 69.1 mm, 25.3mm, 35.2mm, 73.7mm 201.0mm respectively. The along wind response of bridges Type-A are showing deflection more than Type-B bridges under 49m/s design wind speed. The along wind response of all six bridges are effectively controlled by stiffening cable system 'f (Fig. 5.3).