In this paper, we focus on distributed generation systems composed of power generators such as gas-turbines and communication channels. For the systems, we consider determining operation modes (on/off modes) of generators based on distributed cooperative optimization. We first show a distributed cooperative control algorithm. Then, we verify the effectiveness of the algorithm by several numerical simulations.

View full abstract

In this paper, we consider the optimal jumping motion planning problem for a legged robot composed of one leg and the body, and examine the jumping motion obtained by numerical simulations for the purpose of clarifying the effective usage of singular configuration of the robot. The integral of the squared joint torques during jump is minimized on the assumptions that the joint torques supply a fixed amount of energy to the body and that the motion of the robot's center of mass is generated only in vertical direction. By choosing the initial posture of the robot and the spline functions representing the body position as optimization parameters, we apply a numerical optimization method to find the optimal motion. For the optimized motion,the final posture of the robot is close to the singular configuration where the robot is fully extended.

View full abstract

Robot arm is desired not to vibrate on operation. Then, we examine the relationship between vibration suppression and shape of robot arm. We prepare some beams which have different shapes but equal weight, which are regarded as robot arm. And we measure the strain of the rotating arm for the examination. The robot arm rotates horizontally and the gravitation can be ignored. From experiment results, we show the tapered arm is more effective to suppress the vibration of robot arm.

View full abstract

The present paper proposes a multidisciplinary design optimization method for smart micro-composite structures to maximize vibration suppression performance, and an innovative experimental technique with laser excitation is applied to validate numerical results. The smart structures are composed of graphite/epoxy (CFRP) composites and PZT actuators. The performance of smart structures for vibration suppression strongly depends on vibration mode shapes and PZT actuator placements. It is possible to specify vibration mode shapes for the laminated composite by varying the stacking angles of reinforcing fibers, and both fiber orientation angles and PZT placements are optimized simultaneously by using a simple Genetic Algorithm (GA) method. Then, the calculated results are validated by an experiment using laser excitation which excites the micro-sized structures with high reproducibility since the laser is precisely irradiated to the structure at same position with same power repeatedly.

View full abstract

In the actual design, the problem by which a continuation value and a discrete value are also included in a design variable exists. It examines that Differential Evolution algorithm having local search and global search can expect any models to obtain the optimal solution by high probability to such a problem. By this Differential Evolution algorithm, the example of the modal control which used the point of "node" of vibration which does not move by geometry change using FEM on the structural vibration problem of the machine design is introduced.

View full abstract

In structural design, nonsmooth and noisy responses are frequently encountered, e.g. for dynamic problems. It is well known, that for optimization of systems with nonsmooth and noisy responses, conventional derivative-based methods might not converge. On the contrary, the directional Direct Search Method (DSM), which is a derivative-free method, is a promising candidate providing global convergence even in nonsmooth optimization. DSM combines so-called poll steps, ensuring convergence, and search steps, improving efficiency. However, some derivatives may be smooth and easily available, i.e. inexpensive to calculate. Clearly, derivative-free methods ignore this valuable information, rendering them inefficient. To remedy this, we attempt a practical compromise between derivative-based and derivative-free methods by enhancing the DSM. The main ingredients of the proposed augmented DSM are: (i) a tangent cone which is identified efficiently by using derivatives of some, but not all, response functions, (ii) randomly generated poll directions in a tangent cone, (iii) Search steps facilitating a large flexibility, particularly for Kriging model with subset selection. The efficiency and robustness of the method is studied on a seismic design problem of a 10-story building model. The studied problem confirms the efficiency and robustness of the proposed method.

View full abstract

[in Japanese]

View full abstract

This paper proposes a topology optimization method incorporating level set boundary expressions, based on the concept of the phase field method and under the steady-state Navier-Stokes equation, which we apply to a minimum flow friction problem. Using a reaction-diffusion equation to update the level set function, we construct a new topology optimization method that is able to create new holes and maintain clear boundaries in the target structures of an internal flow channel for an incompressible viscous fluid problem.

View full abstract

In this paper we propose a partial differential equation (PDE) to calculate distance from finite element nodes to a threshold value of a variable. The PDE is applied to reduce the grey transition area in topology optimization based on a density approach. The density approach essentially generates grey transition areas, which appear around interfaces between different phases. As the grey areas may affect numerical results and lead to undesirable optimization results, they have to be reduced. For reduction of the grey areas, finite element nodes are repositioned by using the distance from the interfaces calculated by the PDE. In order to demonstrate the effectiveness of the proposed method, a minimum compliance problem is solved.

View full abstract

Recently, the growing demand for hybrid vehicle(HV) and electric vehicle(EV) is increasing the competition for development of power semiconductor devices, which is indispensable for HV and EV. Therefore, topology optimization for power semiconductor devices design has been studied; however, there is no research satisfying both of low calculation cost and high calculation accuracy. In this study, we used the adjoint method and the appropriate physical models to solve this problem. We demonstrated the improvement in the trade-off between on-resistance and breakdown voltage of a p-n junction diode by topology optimization.

View full abstract

This paper presents Density based topology optimization targeting dielectric resonator based metamaterials. In this paper, we aim to obtain left-handed metamaterials by density approach based topology optimization. Left-handed metamaterials are artificial materials that show negative permittivity and negative permeability. Dielectric resonator based metamaterials are obtained left-handed metamaterials by topology change. To achieve this goal, we design the dielectric resonator structure of metamaterials using density approach based topology optimization. We aim to obtain minimizing the effective permeability to obtain Left-handed metamaterials. We provide a numerical example to demonstrate the usefulness of the presented method.

View full abstract

In order to improve a design of structure, it is important to know the actual load condition of failed structure. However, in the case of crane-hook, failed samples do not clearly indicate the actual loading condition that caused their failure. In order to obtain the critical information for the design improvement, we utilize a load estimation process based on the failed crane-hook samples. An optimal design problem that takes account of the information of the estimated critical load condition is formulated as a minimization problem with a multi-objective function; the stiffness and the structural weight are also adopted as the evaluation items that make up the objective function. The multi-objective particle swarm optimization (MOPSO) is adopted as the optimization algorithm. As a result of the optimization, we obtain the Pareto optima. Compared with the actual design, the obtained solutions have good performance in the adopted evaluation items.

View full abstract

This paper presents sizes and shapes optimization of edge ribs of the low-yield steel shear panel damper (SPD) for improving the deformation ability. A minimization problem of maximum accumulated equivalent plastic strain, an index of the deformation ability of SPD, is formulated . The response surface methodology combined with the design of experiment technique is applied as the optimization methodology. In this study, finite element analysis with isotropic/kinematic hardening model was used to simulate the cyclic elasto-plastic behavior instead of experimental approach, and the numerical solutions were confirmed by comparing to previous experimental results. Based on the numerical analysis, the shape parameters effects were investigated and second order polynomials were fitted to obtain the regression equations of the maximum accumulated equivalent plastic strain. The final optimal size and shape were determined by using the regression equations.

View full abstract

RTLS is a system for detecting the position information of the tag attached to the person or object in real time. In factories and distribution centers, inventory management and safety support using the RTLS is expected, there is a need for fast and accurate positioning algorithms. In this study, we apply the Levenberg-Marquardt algorithm to 3D location estimation method of UWB real-time location systems. We verify its effectiveness by experiment using actual equipment.

View full abstract

In manufacturing, process planning considers how the order in which a product is manufactured affects elements such as machining time, power consumption and quality of the final product. However, it is common that a large number of tools and processes are required for the completion of a product, therefore the number of possible processing sequences can be enormous. It is impossible to quickly select the most suitable processing sequence in such cases. Therefore, we solve the problem of deciding the processing sequence in manufacturing using solutions obtained by fc-Shortest Paths algorithm. We develop a decision support system that can evaluate the efficiency of each processing sequence according to the criteria desired.

View full abstract

In our study, a design system is developed for generating gear trains under design constraints on a specified area. The design specifications are input position, output position, input torque, output torque, rotation direction, and so on. Spur gear trains are generated by using random method combined with local optimization method. The objective function evaluates weight, efficiency and life of generated gear train. For several examples, optimal gear trains are generated successfully. Multi-objective design optimizations in spur gear train generation problems are examined by the gear train generation system. The advantage of the optimization approach to generating effective gear trains is reported through several examples.

View full abstract

In the city traffic flow, it is often observed that a new additional road causes more terrible traffic jam although the new road was constructed for improving the traffic jam. The aim of this study is to determine the road that causes the traffic jam from the traffic data of the traffic simulator. The problem is formulated as the optimization problem in which the traffic amount is maximized with respect to the configuration of the road network. The present algorithm is applied to the simple numerical example. The simulation result shows that the present algorithm can determine the road like that causes the traffic jam.

View full abstract

The aim of the present study was to formulate the ergonomic design problem with considering the response to individual differences. Ergonomic design strategies are classified into three categories: a) design for average participant (Strategy A), b) design for weak participant (Strategy B), and c) design for divided clusters (Strategy C). The regret was defined as the difference from the objective function value of optimum solution for each participant. Then, the strategy A, B, and C were formulated as the minimization of the average regret, maximum regret, and average regret for each divided cluster. The proposed method was applied to the design problem of cylindrical diameter for power grip. The cylindrical diameter is the only design variable, and the average pressure and the coefficient of variance were objective functions. The result showed the average regret of the strategy C was significantly lower than the other strategies.

View full abstract

This paper discusses the plastic limit analysis of a truss with some deficient structural components for evaluating redundancy of the truss. Given the upper bound for the number of deficient members, we consider uncertainty in the locations of deficient members. In other words, the set of deficient members is not specified in advance. Then we attempt to find the worst scenario of deficiency, in which the limit load factor attains the minimum value. It is shown that this optimization problem can be formulated as a mixed integer linear programming problem, which can be solved with an algorithm with guaranteed global convergence.

View full abstract

The "Shitamachi Bobsleigh" project, which is established mainly by the small factories in the Ota distinct of Tokyo, is making domestically-build high performance bobsleigh to win a medal at 2014 Winter Olympic Games. The optimization aerodynamic shape design which is combined with computational fluid dynamics, played a part in the development of this bobsleigh. In this article, we present the design approach and the optimization results for creating an aerodynamically high performance bobsleigh model.

View full abstract

The present paper concerns the development and validation of a CFD-based multiobjective stochastic optimization method. The numerical optimization method is applied for the initial design of a waterjet propelled high-speed ship. The scheme investigated in the present study is based on Robust Design Optimization theory with Evolutionary Algorithm. In addition, the concepts of Variable Fidelity/Physics and hierarchical optimization approach are adopted, so as to increase the efficiency of the whole optimization problem by using two different levels of accuracy. That is, both Unsteady RANS and Potential Flow based CFD methods together with Asynchronous Evaluator model for interface with optimization module are used in the present study. A practical geometry modeling method to yield new designs is used as successfully demonstrated in the previous work. In the following, an overview of the present method is given, and results are presented and discussed for Delft Catamaran test cases. The results appear very promising, which support overall validity and effectiveness of the optimization framework developed in the present study.

View full abstract

Most researches of ship shape optimization aim to optimize one factor that influences performance of ship. However Multidisciplinary factors influence ship performance in real sailing. So we have to consider about many factors to optimize Ship Shape. In this report, we take Solar Boat as ship model. Solar Boat gains the power output by solar energy with solar panel. In recent years, solar energy is been getting some attention as new power resources but its volume is influenced by weather. We propose a Multidisciplinary Optimization of Trimaran Solar Boat Shape by the method of Multi-objective Particle Swarm Optimization with considering uncertainty of Solar Boat's power output. By this method we could optimize the Trimaran Solar Boat Shape including diversity and robustness.

View full abstract

We have developed the new optimization method called Stochastic Process Optimization Technique (SPOT). In this paper, integrated optimization of SSTO for two different engines was performed. As a result, similar flight path with different airframe configuration was obtained. In addition, different flight path with similar airframe configuration was obtained as temporary optimum solutions in optimization process. These transitions of optimum solution indicate the influence of SSTO's engine performance on its airframe configuration and flight path through the integrated optimization. The necessity for the integrated optimization for SSTO was shown. Simultaneously, the integrated optimization ability of SPOT was shown.

View full abstract

In industrial design optimization, the objectives and constraints are generally given as the implicit form of the design variables. In addition, the objectives and constraints are not explicitly known but can be evaluated through computationally intensive numerical simulation. Under this situation, the response surface methodology (RSM) is one of helpful approaches for design optimization. In recent, sequential approximate optimization (SAG) has gained its popularity. In this paper, a sequential approximate multi-objective optimization (SAMOO) with radial basis function (RBF) network is proposed. In the SAMOO, it is important to identify highly accurate pareto-frontier with a small number of function evaluations. In order to achieve this objective, a new function called the pareto-fitness function with the RBF network is developed. The pareto-fitness function can generate local minima around a set of pareto-optimal solutions, and the optimal solution of this function is taken as a new sampling point. Also, the density function to find the unexplored region is also employed. It is possible to add the new sampling points around a set of pareto-optimal solutions effectively by using both functions. Through numerical examples, some advantages employing both functions are well discussed.

View full abstract

The aim of the function identification problem is to find the unknown function representation for the given data set. Grammatical Evolution is one of the evolutionary computations which can find the function representation by using the chromosome in Genetic Algorithm and the translation rule described in Backus Naur Form (BNF). This paper describes the application of an advanced Grammatical Evolution (GE) to function identification problem. The advanced Grammatical Evolution uses two-dimensional chromosome, instead of the one-dimensional chromosome employed in the original GE. The linear and nonlinear functions are taken as the numerical examples. The results show that Grammatical Evolution with one-dimensional chromosome can find the linear function faster than the Genetic Programming and that the advanced Grammatical Evolution with two-dimensional chromosome is more effective than that with one-dimensional chromosome for finding the nonlinear function.

View full abstract

Manufacturers over the last decade have tackled the important issues of speeding up and improving quality of product development in an extremely changeable and globalizing world market. Analysis leads design is very useful approach to solve the issues by using supercomputer. However, it is difficult to execute numerical optimization of large scale models due to the increase of computation time. To tackle this problem, shape optimization technology by combing multiple response surface models corresponding to divided domains has been developed. The effectiveness of the proposed technology was demonstrated by applying it to a flow passage design problem in steam turbine. It is shown that the error of our method is less than 6.5%

View full abstract

[in Japanese]

View full abstract

The optimization function of the method of search-space smoothing (SSS) in combination with the Metropolis algorithm (MA) is studied on the random Euclidean traveling salesman problem. The search dynamics of this combined approach (MASSS) are analyzed in both the smoothed and unsmoothed terrains and compared to those of the original SSS method. The results show that the MA can be successfully utilized as a local search algorithm in the SSS approach and the optimization characteristics of these two constituent methods are improved in a mutually beneficial manner in the MASSS run: The relaxation dynamics generated by employing the MA effectively work even in a smoothed terrain and more advantage is taken of the guiding function proposed in the idea of SSS. Also, this mechanism operates in an adaptive manner in the de-smoothing process and the MASSS method maintains its performance over a wider temperature range than does the MA.

View full abstract

The optimization function of the method of search-space smoothing in combination with the Metropolis algorithm (MASSS) is studied on the random Euclidean traveling salesman problem. Three numerical experiments conducted for the method of simulated annealing (SA) are performed to compare functional characteristics between these two methods. First, the optimization characteristics of MASSS are analyzed by using adaptive de-smoothing schedule, and the results show the existence of the effective value of the smoothing factor. Second, interbasin transition dynamics are investigated to evaluate their intensification function. The effect of the relaxation dynamics is maximized at some intermediate value of the smoothing factor, and the value is very close to that found in the first experiment. Last, rate-cycling experiment is performed to test directly the role of the relaxation dynamics in the de-smoothing process. In the optimization process of MASSS, effective relaxation dynamics and the resulting performance is sensitive to the search around the effective value of the smoothing factor. These functional characteristics are quite similar to those of SA.

View full abstract

[in Japanese]

View full abstract