Role of automatic differentiation in concurrent subspace optimization method

2000 ◽  
Author(s):  
Jongsoo Lee ◽  
Changkue Park
Keyword(s):  
Automatic Differentiation ◽  
Optimization Method ◽  
Subspace Optimization ◽  
Concurrent Subspace Optimization

scholarly journals Reviews and syntheses: Flying the satellite into your model: on the role of observation operators in constraining models of the Earth system and the carbon cycle

2017 ◽  
Vol 14 (9) ◽  
pp. 2343-2357 ◽  
Author(s):  
Thomas Kaminski ◽  
Pierre-Philippe Mathieu
Keyword(s):  
Data Assimilation ◽  
Carbon Cycle ◽  
Automatic Differentiation ◽  
Satellite Observations ◽  
Spectral Radiance ◽  
Special Focus ◽  
Earth System ◽  
The Earth ◽  
Retrieval Scheme

Abstract. The vehicles that fly the satellite into a model of the Earth system are observation operators. They provide the link between the quantities simulated by the model and the quantities observed from space, either directly (spectral radiance) or indirectly estimated through a retrieval scheme (biogeophysical variables). By doing so, observation operators enable modellers to properly compare, evaluate, and constrain their models with the model analogue of the satellite observations. This paper provides the formalism and a few examples of how observation operators can be used in combination with data assimilation techniques to better ingest satellite products in a manner consistent with the dynamics of the Earth system expressed by models. It describes commonalities and potential synergies between assimilation and classical retrievals. This paper explains how the combination of observation operators and their derivatives (linearizations) form powerful research tools. It introduces a technique called automatic differentiation that greatly simplifies both the development and the maintenance of code for the evaluation of derivatives. Throughout this paper, a special focus lies on applications to the carbon cycle.

Concurrent Subspace Optimization Using Design Variable Sharing in a Distributed Computing Environment

1995 ◽  
Author(s):  
Brett A. Wujek ◽  
John E. Renaud ◽  
Stephen M. Batill ◽  
Jay B. Brockman
Keyword(s):  
Distributed Computing ◽  
Design Variable ◽  
High Performance ◽  
Low Cost ◽  
Multidisciplinary Design ◽  
Computing Environment ◽  
Subspace Optimization ◽  
System A ◽  
Implementation Studies ◽  
Concurrent Subspace Optimization

Abstract This paper reviews recent implementation advances and modifications in the continued development of a Concurrent Subspace Optimization (CSSO) algorithm for Multidisciplinary Design Optimization (MDO). The CSSO-MDO algorithm implemented in this research incorporates a Coordination Procedure of System Approximation (CP-SA) for design updates. Implementation studies detail the use of a new discipline based decomposition strategy which provides for design variable sharing across discipline design regimes (i.e., subspaces). The algorithm is implemented in a distributed computing environment, providing for concurrent discipline design. Implementation studies introduce a new multidisciplinary design test problem, the optimal design of a high performance, low cost structural system. A graphical user interface is developed which provides for menu driven execution and results display; this new programming environment highlights the modularity of the algorithm. Significant time savings are observed when using distributed computing for concurrent design across disciplines. The use of design variable sharing across disciplines does not introduce any difficulties in implementation as the design update in the CSSO-MDO algorithm is generated in the coordination procedure of system approximation (CP-SA).

scholarly journals SISTEM PENDUKUNG KEPUTUSAN PRIORITAS CALON PENERIMA BANTUAN PROGRAM INDONESIA PINTAR (PIP) PADA SISWA TINGKAT SEKOLAH DASAR MENGGUNAKAN METODE MOORA

2021 ◽  
pp. 1-8
Author(s):  
Laurensia Agustin Manik ◽  
Yani Maulita ◽  
Indah Ambarita
Keyword(s):  
Decision Making ◽  
Decision Support ◽  
Optimization Method ◽  
Ratio Analysis ◽  
Multi Objective Optimization ◽  
A Value ◽  
Parental Work ◽  
Moora Method ◽  
To Receive

Regarding assistance from the recipients of the Smart Indonesia Program (PIP) because there are so many prospective recipients of PIP assistance who volunteered to get the assistance, the process of delivering aid is sometimes not on target. Invalid data causes errors in the distribution of PIPs given to recipients who are entitled to receive them. Special Decision Support System (SPK) was built to support the solution of a problem or for an opportunity, one of the PIP aid recipient's approval in the Binjai City Education Office so that it is right on target for students who will get PIP assistance. The role of decision support systems is needed to improve the efficiency of decision making in determining students who are eligible for PIP assistance. The method used in this system uses the Multi-Objective Optimization method on the basis of Ratio Analysis (MOORA) and by using 7 (seven) criteria, namely family conditions, report card scores, parental work, number of dependents, number of dependents of children who attend school, home conditions and parents' income as well as 10 (ten) alternatives that are student data. Based on the results of calculations using the MOORA method, students on behalf of MM (A9) are the best alternative to be given PIP assistance with a value of 0.2843

Sign in / Sign up

Export Citation Format

Share Document