Analisis Segemental PC-I Girder Dengan Finite Element Method Menggunakan Software Midas

Dinda Mutiara Az Zahra, Prof. Ir. Bambang Suhendro, M.Sc., Ph.D., IPU.

2025 | Skripsi | TEKNIK SIPIL

Jembatan segmental merupakan jembatan yang dibangun dengan metode konstruksi precast beton dalam bentuk segmen-segmen yang dibuat di pabrik lalu dirakit secara bertahap di lapangan. Penelitian dilakukan pada Jembatan Jalan Tol Solo-Yogyakarta-NYIA Kulon Progo STA 6+388 pada bentang 45,8 m. Penelitian ini memiliki tujuan utama untuk menganalisis tegangan dan lendutan pada setiap construction stage yang terdiri dari transfer, konstruksi, dan layan. Pada tahap transfer terdiri dari 7 tahap penarikan tendon.

Metode analisis dilakukan secara manual/Excel sesuai Manual Bina Marga No. 02 / M / BM / 2021 dan menggunakan Finite Element Method dengan dukungan perangkat lunak Midas Civil. Pada analisis Elemen Hingga, girder dimodelkan sebagai line element. Terdapat dua jenis pemodelan, yaitu model girder tunggal dan model jembatan utuh. Model girder tunggal difokuskan untuk meninjau kondisi transfer, sedangkan pemodelan jembatan difokuskan untuk meninjau tahap konstruksi dan layan. Pemodelan girder tunggal dilakukan dua metode, yaitu stressing tidak bertahap dan stressing bertahap. Saat pengaplikasian beban truk, pada pemodelan jembatan dilakukan dua metode, yaitu menggunakan jarak antar girder (Sg) dan lebar efektif (beff).

Hasil analisis menunjukkan bahwa perbedaan tegangan dan lendutan antara perhitungan manual dan pemodelan Midas Civil relatif kecil pada seluruh construction stage. Pada tahap transfer, perbedaan tegangan < 1>lendutan 9,514 mm (9,923%); pada tahap konstruksi, perbedaan tegangan < 1>lendutan 5,344 mm (21,441%). Pada masa layan, perbedaan tegangan mencapai ±3 MPa pada serat atas girder, ±0,5 MPa pada serat bawah girder, ±1 MPa pada serat atas dan serat bawah pelat lantai. Sedangkan perbedaan lendutan beban layan maksimal mencapai 9,663 mm pada kombinasi Layan II. Pemodelan jembatan utuh menghasilkan tegangan dan lendutan lebih kecil dibandingkan girder tunggal maupun perhitungan manual karena distribusi beban lebih merata. Metode manual menghasilkan nilai tegangan dan lendutan yang lebih besar serta bersifat konservatif, namun memiliki keunggulan berupa transparansi perhitungan sehingga berguna untuk preliminary design dan verifikasi hasil pemodelan. Sebaliknya, Midas Civil mampu memperhitungkan efek time-dependent seperti creep dan shrinkage, memberikan hasil cepat, menampilkan visualisasi gaya dalam, tegangan, dan deformasi, serta memodelkan kondisi lapangan lebih realistis. Metode ini lebih tepat digunakan untuk final analysis yang menuntut akurasi. Nilai immediate loss of prestress terbesar diperoleh dari pemodelan stressing bertahap (81,251 MPa), diikuti perhitungan manual (78,620 MPa) dan tidak bertahap (78,570), dengan selisih 0,198 % antara stressing bertahap dan tidak bertahap. Kehilangan tegangan pada stressing bertahap lebih tinggi karena akumulasi setiap tahap penegangan, sehingga tegangan efektif akhir lebih kecil. Nilai camber terbesar diperoleh dari perhitungan manual (163,865 mm), diikuti stressing tidak bertahap (153,190 mm) dan stressing bertahap (150,724 mm). Meski perbedaannya kecil, pemodelan stressing bertahap lebih disarankan untuk mengurangi tegangan awal yang berlebihan serta meminimalkan risiko retak beton. Selain itu, penggunaan lebar efektif tidak berpengaruh signifikan terhadap tegangan, tetapi analisis lendutan dapat mereduksi ±10% dibandingkan perhitungan manual.

Segmental bridges are constructed using a precast concrete method in which individual segments, manufactured in a controlled plant environment, are subsequently assembled sequentially on site. This study focuses on the Solo–Yogyakarta–NYIA Kulon Progo Toll Road Bridge at STA 6+388, featuring a main span of 45.8 m. The primary objective is to analyze stress and deflection behavior throughout the construction stages, namely transfer, construction, and service. The transfer stage comprises seven successive tendon stressing steps.

The analysis was conducted both manually, based on Manual Bina Marga No. 02/M/BM/2021, and numerically using the Finite Element Method implemented in MIDAS Civil software. In the finite element analysis, the girder was modeled as a line element. Two modeling approaches were adopted: a single-girder model and a full-bridge model. The single-girder model was employed to assess transfer conditions through both non-staged and staged stressing schemes, whereas the full-bridge model was used to evaluate construction and service phases. For live load analysis, two approaches were considered: load distribution based on girder spacing (Sg) and on effective flange width (beff).

The results indicate that differences in stress and deflection between manual calculations and MIDAS Civil simulations are relatively small across all construction stages. During the transfer stage, stress deviations were below 1 MPa and deflection differences reached 9.514 mm (9.923%); in the construction stage, stress differences remained below 1 MPa with deflection variations of 5.344 mm (21.441%). At the service stage, stress discrepancies reached approximately ±3 MPa at the top flange, ±0.5 MPa at the bottom flange, and ±1 MPa at both the top and bottom fibers of the deck slab, while the maximum deflection difference under Service Load Combination II was 9.663 mm. The full-bridge model produced smaller stress and deflection values compared to the single-girder and manual analyses due to more uniform load distribution. Manual analysis yielded slightly higher and more conservative results, offering transparency beneficial for preliminary design and verification. In contrast, MIDAS Civil efficiently accounted for time-dependent effects such as creep and shrinkage, provided rapid computation, and enabled realistic visualization of internal forces, stresses, and deformations—making it more suitable for final analyses requiring higher accuracy. The largest immediate prestress loss was obtained from the staged stressing model (81.251 MPa), followed by manual calculation (78.620 MPa) and non-staged stressing (78.570 MPa), with only a 0.198% difference between staged and non-staged methods. Staged stressing produced slightly greater losses due to cumulative effects, resulting in lower effective prestress. The greatest camber occurred in the manual analysis (163.865 mm), followed by non-staged (153.190 mm) and staged stressing (150.724 mm). Although the differences were minor, staged stressing is recommended to mitigate excessive initial stresses and reduce cracking risk in concrete. Furthermore, the use of effective width had negligible influence on stress analysis but improved deflection accuracy by approximately 10% compared to manual calculations.

Kata Kunci : segmental PC-I girder, post-tensioned, stressing bertahap, construction stage, loss of prestress, metode elemen hingga