Gas flow control is one of the critical factors for affecting process yield and throughput as well as device performance in a semiconductor device fabrication process. Semiconductor deposition and etching process uniformity are strongly dependent on process gas delivery condition. Process gas delivery is controlled by a gas flow splitter with multi-lines gas distribution across the wafer. Following recent requirement for further improving process uniformity across the wafer, the number of flow split control lines is increasing, and the flow splitting device is required to stabilize the gas flow rate as fast as possible in each gas line. This paper proposes a novel control method of gas flow split device combined with pressure control to avoid flow interference between multi-lines. Additionally, the controller parameters are optimized based on mathematical control system models, which is different from most of the conventional flow split control methods. Finally, the effectiveness of the proposed method is validated through simulations and experiments.

We investigate an identification of the dynamical behavior of wave segments propagating in excitable media from viewpoints of root locus, stability region, and bode plot. The investigation reveals that the behavior around an operating point can be precisely described by a third-order transfer function in wide frequency range. The precise function is used to design an optimal servo system. This system allows us to maintain the stabilization of wave segments propagating through obstacles that induce high frequency perturbations.

Flat network topology for digital transformation (DX) in industrial control systems (ICS) brings many merits, but also draws threats of cyberattack. Considering cybersecurity, countermeasures against DoS (Denial of Service) attack is mandatory for ICS, especially to maintain its availability. To solve this issue, we propose a flood type DoS mitigation method “SFAT”(Synchronized Filtering based on Arrival Time), which focuses on the periodicity of control communications at the edge of ICS. This SFAT passes network frames only at the timing synchronized to periodic and legitimate communication frames, and then mitigates the amount of DoS frames. This timing is detected by a synchronous detection process that is implemented on hardware. We prototyped this SFAT on a firewall on an FPGA, and then confirmed that it can mitigate the amount of DoS frames to 1/10.

In this paper, we propose an evaluation method of synthetic population. We calculate the difference between actual statistics and synthetic statistics summarized from synthetic populations in our previous method. If we can use actual individual data, we compare its and synthetic population. However, we can not use it. This paper proposes an evaluation method that calculates the difference between the synthetic population and the virtual population generated from actual statistics.

In this study, I proposed a novel data-driven control method for updating controllers and models. The proposed method can obtain a plant model and a feedback controller by using initial control signals and initial controllers. I verified the validity of the proposed methods via experiment verifications.