An approach to improving attenuation cutoffs is proposed to linear phase filters based on the Lagrangean B-filters approximating the ideal low-pass filter. The impulse response of the mth order Lagragean B-filter holds a non-negative even function symmetry, and all its derivatives of order up to (m-1) are continuos. Hence, the B-filters have linear phase properties in nature. The mth order B-filters are connected in parallel in view of more convenient implementations of circuits than conventional cases, and then each peak component of sidelobe spectra standing close by the main lobes of their frequency domains is cancelled out. This can result in sharper attenuation cutoffs than the mth order B-filter. These mth order linear phase filters realized by this method preserve the bandwith and the asymptotic decay rate of the sidelobe envelope in the mth order B-filter. This approach is illustrated by a few numerical examples.
This paper gives a Dempster-Shafer (DS) theoretic approach for realizing decision support in safety-control for large-complex systems. We show that rules of combination in the conventional framework of the DS theory are not always adequate for representing various knowledge on characterisitics of systems which must be controlled. This paper extends the current framework of the DS theory into a general one so that it can reflect operators' system-specific knowledge. Usefulness of our extension is proven by taking a safety-control problem as an example.
At present, print boad checking devices (called jig) are widely used for electrical checking of multi-layer print boards. To design these devices, it is required to extract points, which correspond to ends of electrical circuits, from the print films. The point extraction work has been done by human beings and it makes their eyes extremely tired. This paper describes an image processing system for newly developed for automatic point extraction, after feeding print films by one-dimensional line sensor with high resolution. The features of the system to extract the points are : (1) Local image processing is carried out on the large amount of data for processing efficiency. (2) Global processing that signal connection is checked between print boards is incorporated for processing accuracy. (3) The system is designed to be universal so that multi-layer print boards can be dealt with.
In this paper H∞ control has been applied to an interactive steam pressure and steam flow control system of a Heat Recovery Steam Generator (HRSG) of a thermal power plant. The controller structure is taken account for practical applications on a commercial plant. H∞ control for the interactive system is verified the experiments on a commercial sized test facility. In addition to the nominal performance, robustness of H∞control has been also verified. To demonstrate a practical applicability, the experiments for signal saturation and noise reduction have been accomplished.
A two-degree-of-freedom model following servosystem is proposed. First, a model following control system with no integral compensator is designed for a nominal plant, a given reference model, and a specified performance index. The obtained control system is augmented with an integral compensator. Then, additional feedforward and feedback are introduced at the output of the integrator to cancel the integral action for reference signals under the assumptions that there is no modeling error of the plant and no disturbance input to the plant. When the modeling error or the disturbance input exists, the integral effect appears so that the steady state error in the control output is eliminated.