Search results

The purpose of this paper is to develop a new rectilinear branch pipe‐routing algorithm for automatic generation of rectilinear branch pipe routes in constrained spaces of aero‐engines.

Rectilinear branch pipe routing that connects multiple terminals in a constrained space with obstacles can be formulated as a rectilinear Steiner minimum tree with obstacles (RSMTO) problem while meeting certain engineering rules, which has been proved to be an NP‐hard and discrete problem. This paper presents a discrete particle swarm optimization (PSO) algorithm for rectilinear branch pipe routing (DPSO‐RBPRA) problems, which adopts an attraction operator and an energy function to plan the shortest collision‐free connecting networks in a discrete graph space. Moreover, this paper integrates several existing techniques to evaluate particles for the RSMTO problem in discrete Manhattan spaces. Further, the DPSO‐RBPRA is extended to surface cases to adapt to requirements of routing pipes on the surfaces of aero‐engines.

Pipe routing numeral computations show that, DPSO‐RBPRA finds satisfactory connecting networks while considering several engineering rules, which demonstrates the effectiveness of the proposed method.

This paper applies the Steiner tree theory and develops a DPSO algorithm to plan the aero‐engine rectilinear branch pipe‐routing layouts.

The paper aims to present a modified particle swarm optimization (MPSO) approach for automatic generation of near‐optimal pipe routes in constrained aero‐engine 3D rotational space.

Pipe assembly for aero‐engine is formulated as searching for the optimal pipe paths meeting certain objectives in a constrained 3D rotational space. The routing space is first modelled by grid discretization in the cylindrical coordinate system, and then is simplified into several 2D planes by mapping development. The objective function is formulated to minimize the pipe lengths and the number of pipe turns, to place pipes next to the inner jacket as close as possible, and also to make pipe trajectories closely follow around obstacle contours while avoiding collisions. Then, an MPSO approach, which adopts a discrete operator and a fixed‐length encoding mechanism, is developed to seek optimal solutions to the objective function. The convergence of MPSO is theoretically proved. Finally, numerical computations of pipe‐routing examples are conducted by using Matrix Laboratory and Unigraphics NX 4.0 system, which demonstrates effectiveness and efficiency of the proposed method.

Results show that MPSO can quickly find the optimal pipe routes meeting certain engineering constraints, and also manifests better computation convergences.

The application of the MPSO approach in pipe routing for aero‐engines is demonstrated. MPSO is a general modified particle swarm optimization version that it is not restricted to the pipe‐routing problems, and the routing approach can also be applied in similar path‐planning problems such as robot path‐planning and very large‐scale integration design.

The paper develops a new formulation for aero‐engine pipe‐routing problems, and presents an MPSO approach to find the optimal pipe paths.

Detecting objects in images and videos is a difficult task that has challenged the field of computer vision. Most of the algorithms for object detection are sensitive to background clutter and occlusion, and cannot localize the edge of the object. An object's shape is typically the most discriminative cue for its recognition by humans. The purpose of this paper is to introduce a model‐based object detection method which uses only shape‐fragment features.

The object shape model is learned from a small set of training images and all object models are composed of shape fragments. The model of the object is in multi‐scales.

The major contributions of this paper are the application of learned shape fragments‐based model for object detection in complex environment and a novel two‐stage object detection framework.

The results presented in this paper are competitive with other state‐of‐the‐art object detection methods.

During the design of a new product, the generation of assembly sequences plans (ASPs) has become one of the most important problems taken into account by researchers. In fact, a good mounting order allows the time decrease of the assembly process which leads to the reduction of production costs. In this context, researchers developed several methods to generate and optimize ASP based on various criteria. Although this paper aims to improve the quality of ASP it is necessary to increase the number of criteria which must be taken into account when generating ASPs.

In this paper, an ASP generation approach, which is based on three main algorithms, is proposed. The first one generates a set of assembly sequences based on stability criteria. The obtained results are treated by the second algorithm which is based on assembly tools (ATs) workspace criterion. An illustrative example is used to explain the different steps of this proposed approach. Moreover, a comparative study is done to highlight its advantages.

The proposed algorithm verifies, for each assembly sequence, the minimal required workspace of used AT and eliminates the ASPs non-respecting this criterion. Finally, the remaining assembly sequences are treated by the third algorithm to reduce the AT change during the mounting operation.

The proposed approach introduces the concept of AT workspace to simulate and select ASPs that respect this criterion. The dynamic interference process allows the eventual collision detection between tool and component and avoids it. The proposed approach reduces the AT change during the mounting operations.

Ant colony algorithm is widely used in recent years as a heuristic algorithm. It provides a new way to solve complicated combinatorial optimization problems. Having been enlightened by the behavior of ant colony's searching for food, positive feedback construction and distributed computing combined with certain heuristics are adopted in the algorithm, which makes it easier to find better solution. This paper introduces a series of ant colony algorithm and its improved algorithm of the basic principle, and discusses the ant colony algorithm application situation. Finally, several problems existing in the research and the development prospect of ACO are reviewed.