Predictive Fuzzy Logic Control Research Articles

Piston motion control for a dual free piston linear generator: predictive-fuzzy logic control approach

The free-piston engine has a lot of potential to alter the conventional combustion engine due to mechanical structural simplicity, high engine efficiency, reducing hazardous gas emission and fuel flexible combustion. However, accuracy of piston motion control is a major challenge in all types of free-piston engine as well a dual free piston engine generator (DFPG) because of not having a crankshaft mechanism. This article describes a simulation investigation of a predictive-fuzzy logic control method conducted on the DFPG prototype in University of Ulsan. First, the simulation model of the DFPG consisted of a free-piston engine and a linear alternator is established and running in Matlab/Simulink environment. Then simulation result is validated against experimental data to prove validity of the simulation model. Second, a predictive-fuzzy logic (PFL) control method is proposed to apply to the DFPG and its performance is carried out using the full-cycle simulation model. The PFL strategy is demonstrated to improve the DFPG performance and more effective than the PID one in terms of settling time, TDC error, and cycle-to-cycle variation. According to results, the PFL controller reduces settling time 30 % in comparison to PID controller. The PFL controller has also 0.52 mm better TDC position error than the PID controller is 0.76 mm. Coefficient of variation values is found to be 1.0 %, and 1.9 % for the PFL controller and PID controller, respectively. This means that the PFL controller is a powerful method to dampen cycle to cycle variations in the DFPG.

WLAN Cell Handoff Latency Abatement Using an FPGA Fuzzy Logic Algorithm Implementation

Following the path toward 4 G set by its wireless siblings LTE and WiMax, IEEE 802.11 technology, universally known as WiFi, is evolving to become a high data rate QoS-enabled mobile platform. The IEEE 802.11n standard yields data rates up to 450 Mbp s and the 802.11e standard ensures proficient QoS for real-time applications. Still in need of better performance, multicell environments that provide extended coverage allow the mobile station nomadic passage beyond a single cell by means of cell dissociation-association process known as handoff. This process poses a challenge for real-time applications like voice over IP (150 ms maximum delay) and video (200–400 ms) sessions, to give the user a seamless cell-crossing without data loss or session breakage. It presented an approach of a predictive fuzzy Logic controller to reduce the channel scanning process to a tenth of the standard time, and its efficient FPGA implementation to speed up the processing time. The algorithm of the fuzzy controller was implemented in C language. Experimental results are provided.

Introduction of Predictive Fuzzy Inflow Control for Urban Expressway

The dynamic change of traffic condition is always monitored in traffic control centre. Therefore, it seems to be profitable to consider the estimation of traffic change in near future. The predictive fuzzy controller has been proposed to realize the decision mechanism to consider the future influence. Therefore, the predictive fuzzy logic controller might be regarded as an adaptive way to the dynamic change of traffic conditions. It is confirmed as well that the predictive fuzzy logic controller provide dynamically the efficient traffic flow on urban expressway as an advanced traffic control measure. Therefore, the combination of intelligent devices as ETC and the advanced traffic control technique show the direction of future effective traffic management on urban expressway.

A predictive fuzzy logic controller with an adaptive loop for the manufacture of resin adhesives

Control of exothermic reaction in manufacturing of resin adhesives has been performed, for long, through human reasoning and expertise. Human involvement has been a source of control errors, which affect the quality of resin product. An earlier paper on this topic has incorporated fuzzy logic control (FLC) in the control loop of the manufacturing process to support and accommodate human expertise and to avoid human errors in maintaining a sustained product quality. A one-step predictor has also been included in the FLC loop to compensate for the inherent time delay in the process system. The current research substantiates the FLC methodology and proposes further modifications in the fuzzy logic controller in order to overcome some shortfalls in the earlier system such as reduced speed of control and varying steady state errors in temperature responses. The main reason for these shortfalls is some unmeasurable parameters of the system that vary slowly and unpredictably. An approach of adaptive control is found appropriate in compensating such parameter variations and is applied as an outer loop to the earlier FLC system. Several experiments performed on this system confirm that the new proposal is effective in overcoming the shortfalls of the earlier FLC system.

A PREDICTIVE FUZZY LOGIC CONTROLLER FOR RESIN MANUFACTURING

Exothermic reactions in polymerization processes require precise control of the reaction temperature in order to avoid the danger of temperature run-away and the consequent damage to expensive materials. Mathematical modeling of the reaction process, in order to construct a model-based control, is highly complex due to the unpredictable and nonlinear exothermic reaction temperatures. In such cases, Fuzzy Logic Control (FLC) is very useful since it does not require a mathematical model for generating a control law. This paper details a design methodology for a sensor-based FLC control system in producing phenol formaldehyde resin adhesives. A control structure called predictive FLC is developed and employed. Several experiments were carried out to ascertain the accuracy of predictive FLC in controlling the reaction temperature. Two of the temperature responses are included in the paper.