In this paper, a revised GMDH (Group Method of Data Handling) of multiinput-singleoutput type algorithm is developed by using principal component-regression analysis. In previous GMDH algorithms, the optimal partial polynomials, which are generated by using the second degree polynomial of the two variables in each selection layer, are accumulated in multilayerd structure to construct the complete polynomial. But, in a high order selection layer, many combinations of two variables generate multicolinearity in partial polynomials and this presents a severe problem in estimation accuracy of model parameters. The nonlinear models which contain multicolinearity in the partial polynomials lack in stability and can not be used in prediction problems. The revised GMDH in this paper generates some optimal partial polynomials which are identified by using principal component-regression analysis in each selection layer, and the complete description of the system is constructed by combining these optimal partial polynomials in multilayerd structure. The principal components which construct the partial polynomials are perendicular each other and so the partial polynomials generate no multicolinearity. The revised GMDH algorithm is applied to a simple illustrative example and compared with the result obtained by the previous GMDH algorithm.
The authors have previously proposed a design method of time-optimal feedback control system by using switching hypersurface. In the present paper, the proposed method is applied to construct the control system of the training ship Ohshimamaru 330 GT, and then the practical applicability is shown in the sea trial. The dynamics of a ship is presented by a complicated mathematical model with some nonlinear elements and dead time elements. To make the design of the dual mode controller tractable, the model of ship dynamics is simplified from the practical point of view. And dead time problem is evaded by making use of predictive control. In the simulated course changing control using the nonlinear mathematical model as a controlled object, it is confirmed that the time-optimal control is approximately realized, It is demonstrated that the proposed method is efective showing the resemblance of the results in the sea trial, with those in the simulation.
Pneumatic drive systems have higher power/weight ratio. So, it is easy to make a compact and lightweight robot manipulator. On the other hand, it is very difficult to control this manipulator because of interference force between each axis and large friction which is caused by mechanical seal. Therefore, we applied model based control to a pneumatic drive manipulator, and we proposed the disturbance observer which used the concept of model based control. The effectiveness of the proposed control scheme is shown through experiments. Further more, we designed a compact robot manipulator using the link built-in rotary cylinder. As a result, we can state that we can use a pneumatic drive manipulator, which is very small and very light, as a industrial robot.
As well known, computational fluid dynamics consists of two main parts, i. e. (1) generation of computational grid, and (2) numerical scheme to slove the conservation law. In this report, new methods for adaptive grid generation through the optimal regulator of digital control are presented. The performance index of linear optimal regulator problem has a strong relation to the characteristics of adaptive grid generation. Paying attention to this point, one and two dimensional Poisson equations are transformed into the state-space form. The numerical examples of NACA-0012 airfoil flow field show strong clustering of grid lines at shock waves through these two kinds of adaptation.
When a hovercraft is navigated, it is frequently heaved by some complex causes. However, most methods of decreasing the heave are only in the structural design of a vehicle. In this paper we considered the application of H∞-control theory to designing the controllers which decreased the heave of hovercraft. To begin with the study, we made a simple model of a hovercraft which was restricted in two degrees of mechanical freedom (vertical motion and rotation) and made some configuration control experiments on it. As the result, we could show the effectiveness of our method.