TOPOLOGY DESIGN OPTIMIZATION BASED ON BIOTIC BRANCH NET

The biotic branch nets are extreme high-tech product. In order to achieve acertain functional objective, they can adjust their growth direction and growth velocity byaccording to the varying growth environment. An innovative and effective methodology of topologydesign optimization based on the growth mechanism of biotic branch nets is suggested, and it isapplied to a layout design problem of a conductive cooling channel in a heat transfer system. Theeffectiveness of the method is validated by the FEM analysis.

MW-OBS:An Improved Pruning Method for Topology Design of Neural Networks

Topology design of artificial neural networks （ANNs） is an important problem for large scale applications. This paper describes a new efficient pruning method, the multi-weight optimal brain surgeon （MWOBS） method, to optimize neural network topologies. The advantages and disadvantages of the OBS and unit-OBS were analyzed to develop the method. Actually, optimized topologies are difficult to get within reasonable times for complex problems. Motivating by the mechanism of natural neurons, the MW-OBS method balances the accuracy and the time complexity to achieve better neural network performance. The method will delete multiple connections among neurons according to the second derivative of the error function, so the arithmetic converges rapidly while the accuracy of the neural network remains high. The stability and generalization ability of the method are illustrated in a Java program. The results show that the MWOBS method has the same accuracy as OBS, but time is similar to that of unit-OBS. Therefore, the MWOBS method can be used to efficiently optimize structures of neural networks for large scale applications.

A Unified Optimization-Based Framework to Adjust Consensus Convergence Rate and Optimize the Network Topology in Uncertain Multi-Agent Systems

This paper deals with the consensus problem in an uncertain multi-agent system whose agents communicate with each other through a weighted undirected(primary)graph.The considered multi-agent system is described by an uncertain state-space model in which the involved matrices belong to some matrix boxes.As the main contribution of the paper,a unified optimization-based framework is proposed for simultaneously reducing the weights of the edges of the primary communication graph(optimizing the network topology)and synthesizing a controller such that the consensus in the considered uncertain multi-agent system is ensured with an adjustable convergence rate.Considering the NP-hardness nature of the optimization problem related to the aforementioned framework,this problem is relaxed such that it can be solved by regular LMI solvers.Numerical/practical-based examples are presented to verify the usefulness of the obtained results.

Space truss construction modeling based on on-orbit assembly motion feature

More space truss construction has been planned to develop and utilize space resources.These trusses are designed in the way of large-scale,complex,modular,and on-orbit assembly.To meet the upcoming challenge of large-scale space infrastructure construction,it is necessary to study space truss automation design and robotic construction.This paper proposes an ordinal finite screw adjacency matrix model(OFSAMM),focusing on the relationship between assembly motions,to express and compute a space truss structure.In this model,a space truss is abstracted as a set of ordered assembly motions,each of which is recorded as a finite screw as the basic element of the truss and its assembly.The operation of truss transformation is also derived under this model.Therefore,the truss configuration,the assembly sequence,the truss sub-assembly,the truss components,and the on-orbit assembly task can be expressed and calculated in a unified model,which is calculated and stores the truss topology and assembly with the minimum storage cost.At the end of this paper,we introduce how to synthesize and optimize space truss design through two cases.The study will help to improve design efficiency.Furthermore,it provides a theoretical basis for the automatic construction of space truss structures,especially in the next stage.

Non-interference topology scheme in wireless sensor networks

A novel topology scheme, cell with multiple mobile sinks method （CMMSM）, is proposed in this article for the collection of information and for the environment monitoring in wireless sensor networks. The system consists of many static sensors, scattered in a large scale sensing field and multiple mobile sinks, cruising among the clusters. Conservation of energy and simplification of protocol are important design considerations in this scheme. The noninterference topology scheme largely simplifies the full-distributed communication protocol with the ability of collision avoidance and random routing. The total number of cluster heads in such a topology was analyzed, and then an approximate evaluation of the total energy consumption in one round was carded out. Simulation results show that CMMSM can save considerable energy and obtain higher throughput than low-energy adaptive clustering hierarchy （LEACH） and geographical adaptive fidelity （GAF）.

COMPUTER PROGRAM FOR DIRECTED STRUCTURE TOPOLOGY OPTIMIZATION

To compensate for the imperfection of traditional bi-directional evolutionary structural optimization, material interpolation scheme and sensitivity filter functions are introduced. A suitable filter can overcome the checkerboard and mesh-dependency. And the historical information on accurate elemental sensitivity numbers are used to keep the objective function converging steadily. Apart from rational intervals of the relevant important parameters, the concept of distinguishing between active and non-active elements design is proposed, which can be widely used for improving the function and artistry of structures directly, especially for a one whose accurate size is not given. Furthermore, user-friendly software packages are developed to enhance its accessibility for practicing engineers and architects. And to reduce the time cost for large timeconsuming complex structure optimization, parallel computing is built-in in the MATLAB codes. The program is easy to use for engineers who may not be familiar with either FEA or structure optimization. And developers can make a deep research on the algorithm by changing the MATLAB codes. Several classical examples are given to show that the improved BESO method is superior for its handy and utility computer program software.

Optimizing Fiber Topologies for WDM Optical Networks Based on Multi-Granularity Optical Switching Technology

For the quality of service （QoS） and fairness considerations, the hop counts of various lightpaths in a wavelength division multiplexing （WDM） optical network should be short and compact. The development of multi-granularity optical switching technology has made it possible to construct various fiber topologies over a fixed physical topology. This paper describes a fiber topology design （FTD） problem, which minimizes the maximum number of required fibers in the physical links for a maximum lightpath hop count in the fiber topology. After the formular description for the FTD problem, a method was given to obtain the lower bound on the maximum number of required fibers. For large or moderate scale networks, three heuristic algorithms are given to efficiently solve the FTD problem. This study gives a new way to optimize the resource configuration performance in WDM optical networks at the topology level and proves its effectiveness via both analyses and numerical experiments.

Development of lunar regolith composite and structure via laser-assisted sintering

Aiming at the exploration and resource utilization activities on the Moon,in situ resource utilization and in situ manufacturing are proposed to minimize the dependence on the ground transportation supplies.In this paper,a laser-assisted additive manufacturing process is developed to fabricate lunar regolith composites with PA12/SiO2 mixing powders.The process parameters and composite material compositions are optimized in an appropriate range through orthogonal experiments to establish the relationship of process–structure–property for lunar regolith composites.The optimal combination of composite material compositions and process parameters are mixing ratio of 50/50 in volume,laser power of 30 W,scanning speed of 3500 mm/s,and scanning hatch space of 0.2 mm.The maximum tensile strength of lunar regolith composites reaches 9.248 MPa,and the maximum depth of surface variation is 120.79μm,which indicates poor powder fusion and sintering quality.Thereafter,the mechanical properties of laser-sintered lunar regolith composites are implemented to the topology optimization design of complex structures.The effectiveness and the feasibility of this laser-assisted process are potentially developed for future lightweight design and manufacturing of the solar panel installed on the lunar rover.