Hasil Pencarian  ::  Simpan CSV :: Kembali

"This work presents an investigation of the influence of different modeling approaches on the quality of fuel economy simulations of hybrid electric powertrains. The main focus is on the challenge to accurately include transient effects and reduce the computation time of complex models. Methods for the composition of entire powertrain models are analyzed as well as the modeling of the individual components internal combustion engine and battery. The results shall help with the selection of suitable models for specific simulation tasks and provide a deeper understanding of the dynamic processes within simulations of hybrid electric vehicles."

"Permodelan menggunakan program komputer untuk melihat reaksi suatu struktur terhadap beban dinamik yang diberikan pada pondasinya telah banyak digunakan untuk memudahkan proses perhitungan dalam sebuah penelitian. Dengan memberikan input yang akan dianalisis dan batasan-batasan permodelan yang sesuai dengan teori, sistem permodelan dengan komputer dapat memberikan jaminan terhadap valid data output yang dihasilkan. Sifat tanah lunak yang memberi efek redaman yang semakin besar terhadap gelombang seiring bertambahnya volume tanah yang dilaluinya, dianggap sulit diabaikan. Untuk itu, batasan terhadap mesh yang terlibat sebagai representasi tanah di sekitar struktur yang dimodelkan, dirasa perlu diverifikasi. (Mesh adalah elemen-elemen kecil yang menghubungkan nodal-nodal sebagai diskritisasi suatu sistem yang lebih kontinyu sebagai dasar analisis elemen hingga yang digunakan komputer). Berdasarkan efek di atas, lebar mesh terlibat yang relatif sempit diduga akan memberikan output yang berbeda apabila permodelan dilakukan dengan lebar mesh yang lebih luas. Untuk melihat pengaruh batasan ini, dilakukan pendekatan dengan permodelan uji seismic down-hole dengan program PLAXIS. Dengan menggunakan kecepatan gelombang geser sebagai indikator, Akan dilakukan perbandingan hasil output yang diperoleh terhadap pemakaian mesh yang berbeda-beda. Dengan pendekatan ini diharapkan mampu memberikan gambaran mengenai pengaruh lebar mesh yang terlibat.

---

Modeling using computer programs to see the response of a structure when it is subjected to dynamic loads have been widely used to facilitate the calculation process at any interest. By providing the input required and setting up the boundaries in accordance with theoretical manner, the computer system can provide an assured valid output of data generated. However, in some cases, it is often hard to determine the maximum effective boundaries for the mesh involved (Mesh is the small elements that connected by the nodes as a result of discretization process of a more continuous system as the basic principle for finite element analysis used by the computer system). With concomitant increase in the volume of soil along its path, the effect of the damping will cause the amplitude of the travelling wave to be gradually decreased. Thus, it is difficult to ignore the boundaries by setting a point of boundaries at random distance from the structure. Boundaries of the mesh involved as the representation of the soil around the structure, perceived need to be verified. Based on the above effects, the relatively narrow mesh involved will allegedly give a different output compared with when it is modeled with a wider mesh geometry. To see the effect of this mesh boundaries, the approach by modeling the down-hole seismic test with program PLAXIS is conducted. By doing so, clearer explanation about the mesh boundary is expected."

"Komponen otomotif, terutama casing dan mounting umumnya memiliki bentuk yang sangat rumit. Perancangan dari awal akan memakan banyak waktu, namun dapat dihindari dengan proses rekayasa balik. Komponen yang direkayasa balik pada penelitian ini adalah casing transmisi Nissan March. Metode yang digunakan adalah pemindaian tiga dimensi dengan menggunakan Faro Arm dan Geomagic Studio. Permukaan komponen dipindai untuk mendapatkan titik - titik koordinat x, y, z persatuan jarak tertentu. Data titik ini dibentuk menjadi model polygon dengan proses triangulasi. Model polygon disempurnakan dan dijadikan dasar untuk pemodelan permukaan CAD dengan metode parametrik dan NURBS. Model CAD kemudian digunakan untuk pembuatan cetakan, estimasi kebutuhan material, dan berbagai simulasi.

---

Automotive components, especially casing and mounting, have very complex geometry. Normal design procedure will take too much time, but it can be avoided by doing reverse engineering process. Component that is reverse engineered in this research is Nissan March transmission’s case. The method is 3D scanning using Faro Arm and Geomagic Studio. Component’s surfaces are scanned to get x, y, z coordinate points at every certain distance. This point clouds are used to make a polygon model with triangulation process. Polygon model is refined and used as a base for CAD surface modeling with parametric and NURBS methods. Then the CAD model is used for dies modeling, raw materials needs estimation, and many kinds of simulations."

"Pemodelan dinamis merupakan penyederhanaan fenomena yang terjadi di alam dengan melibatkan berbagai faktor empiris. Dalam sejarah dunia, pandemi merupakan salah satu fenomena yang menimbulkan kerugian. Merebaknya pandemi COVID-19 di awal tahun 2020 dengan cepat dengan tingkat kematian yang relatif tinggi, menjadi topik baru dan hangat dalam dunia penelitian, khususnya dalam pencarian model prediksi. Berbagai model penyebaran pun dikembangkan. Gap penelitian yang dapat diidentifikasi, antara lain pertimbangan faktor spasial dalam pengembangan model prediksi penyebaran COVID-19 dan; penggunaan teknik machine learning/deep learning. Tujuan studi ini adalah mendapatkan model prediksi penyebaran penyakit menggunakan teknik machine learning/deep learning yang melibatkan faktor spasial untuk diterapkan pada kasus COVID-19. Model matematis Susceptibles–Infected–Recovered–Deaths (SIRD) dipilih sebagai dasar pengembangan model. Metode yang digunakan dalam membangun model yang melibatkan faktor spasial adalah Cellular Automata (CA). Adapun metode Long Short Term Memory (LSTM) dengan strategi Multiple Parallel Input Multi-Step Output pada arsitektur Stacked-LSTM digunakan untuk menyelesaikan model. Kontribusi studi adalah menghasilkan rumusan matematis untuk faktor spasial pada model penyebaran penyakit, serta modifikasi terhadap konsep CA yang menghasilkan model ”Gabungan Non-uniform Continuous Cellular Automata (N-CCA) dan Stacked-LSTM”. Keakuratan model yang diusulkan, diberikan oleh nilai RMSPE dan MAPE, masing-masing sekitar 8% dan 7%, untuk prediksi COVID-19 di USA selama 21 hari ke depan, dan 4% dan 3% untuk tujuh hari prediksi di Indonesia.

---

Dynamic modeling is a simplification of naturally occurring phenomena. Pandemics are one of the phenomena that cause losses. The outbreak of the COVID-19 pandemic in early 2020 has become a new and hot topic of research, especially in modeling disease spread. This outbreak spread rapidly and had a relatively high mortality rate. Various deployment models begin to develop. Gaps in the research results can also be identified, among others considering spatial factors on the predictive models using machine learning for the spread of COVID-19. The purpose of the studies is to obtain the formula and the predictive of disease spreads spatial model using machine learning and apply it to COVID-19 cases. A model is built using Cellular Automata (CA) configuration based on the Susceptibles–Infected–Recovered–Deaths (SIRD) model and its completion with the Long-short-term memory (LSTM) method. Multiple-parallel input multi-step output strategies are chosen on the Stacked-LSTM architecture to complete a model. The contributions of this study are the mathematical formulation for the spatial factors of the disease spread model and, as well as modifications to the CA concept. The final model was named ”A Combined Non-uniform Continuous Cellular Automata (N-CCA) Modification and Stacked-LSTM Model.” The accuracy of the proposed model, given by RMSPE and MAPE values, which is around 8% and 7%, respectively, for predictions of COVID-19 in the USA for the next 21-days, and 4% and 3% for seven days predictions in Indonesia."

"This volume of the Lecture Notes in Mobility series contains papers written by speakers at the 22nd International Forum on Advanced Microsystems for Automotive Applications (AMAA 2018) "Smart Systems for Clean, Safe and Shared Road Vehicles" that was held in Berlin, Germany in September 2018. The authors report about recent breakthroughs in electric and electronic components and systems, driver assistance, vehicle automation and electrification as well as data, clouds and machine learning. Furthermore, innovation aspects and impacts of connected and automated driving are covered. The target audience primarily comprises research experts and practitioners in industry and academia, but the book may also be beneficial for graduate students alike. "

"This volume includes selected and reviewed papers from the 4th International Congress of Automotive and Transport Engineering, held in Cluj, Romania, in September 2018. Authors are experts from research, industry and universities coming from 14 countries worldwide. The papers are covering the latest developments in automotive vehicles and environment, advanced transport systems and road traffic, heavy and special vehicles, new materials, manufacturing technologies and logistics, accident research and analysis and innovative solutions for automotive vehicles.The conference is organized by SIAR (Society of Automotive Engineers from Romania) in cooperation with FISITA. "