Publications Scientifiques

Permanent URI for this communityhttps://dspace.univ-boumerdes.dz/handle/123456789/10

Browse

Has files: NoSubject: search.filters.subject.Multi-objective optimization

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Optimization of WAG Process Using Dynamic Proxy, Genetic Algorithm and Ant Colony Optimization
    (Springer, 2018) Nait Amar, Menad; Zeraibi, Noureddine; Kheireddine, Redouane
    The optimization of water alternating gas injection (WAG) process is a complex problem, which requires a significant number of numerical simulations that are time-consuming. Therefore, developing a fast and accurate replacing method becomes a necessity. Proxy models that are light mathematical models have a high ability to identify very complex and non-straightforward problems such as the answers of numerical simulators in brief deadlines. Different static proxy models have been used to date, where a predefined model is employed to approximate the outputs of numerical simulators such as field oil production total (FOPT) or net present value, at a given time and not as functions of time. This study demonstrates the application of time-dependent multi Artificial Neural Networks as a dynamic proxy to the optimization of a WAG process in a synthetic field. Latin hypercube design is used to select the database employed in the training phase. By coupling the established proxy with genetic algorithm (GA) and ant colony optimization (ACO), the optimum WAG parameters, namely gas and water injection rates, gas and water injection half-cycle, WAG ratio and slug size, which maximize FOPT subject to some time-depending constraints, are investigated. The problem is formulated as a nonlinear optimization problem with bound and nonlinear constraints. The results show that the established proxy is found to be robust and an efficient alternative for mimicking the numerical simulator performances in the optimization of the WAG. Both GA and ACO are strongly shown to be highly effective in the combinatorial optimization of the WAG process.
  • No Thumbnail Available
    Item
    Multiobjective optimization of an agile machining type linear delta structure
    (2018) Mansouri, Khaled; Belaidi, Idir; Atia, Abdelmalek
    The innovative architectures design of agile machines dedicated to the machining at high speed requires the implementation of analytical and numerical models for the optimization of the kinematic, static and dynamic behavior of the machine, taking into account the elastic deformations and their compensation at level of machine control. In the context of multi-objective optimization, the first part is to identify the parameters and variables inherent to each constituent element of a DELTA robot type machine, the purpose is to optimize the essential elements of its structure. This requires a formulation of the multi-objective problem by expressing the objective functions, the constraints and the corresponding search spaces, as well as the resolution of the problem by the use of high-performance mathematical methods and tools (genetic algorithms, etc.).