A basic classifier system, ZCS, is presented which keeps much of Holland’s original framework but simplifies it to increase understandability and performance. ZCS’s relation to Q-learning is brought out, and their performances compared in environments of two difficulty levels. Extensions to ZCS are proposed for temporary memory, better action selection, more efficient use of the genetic algorit...

The application of MATLAB/SIMULINK is developed for simulation and learning feedback controller design of dc–dc switching converters. With the help of its intuitive graphical user interface and some basic circuit theories, the corresponding SIMULINK models of the switching converter circuits can be easily constructed. Students can then use these models to learn and evaluate the closed-loop beha...

This paper studies the switching of trading strategies and its effect on the market volatility in a continuous double auction market. We describe the behavior when some uninformed agents, who we call switchers, decide whether or not to pay for information before they trade. By paying for the information they behave as informed traders. First we verify that our model is able to reproduce some of...

Empirical studies reveal a surprisingly wide variety of pricing strategies among retailers, even among Internet sellers of undifferentiated homogeneous goods, such as books and music CDs. Several empirical findings remain puzzling; for example, within the same market, some small retailers decide to discount deeply, whereas others forgo the price-sensitive switchers and price high. The authors p...

XCS and later ZCS are considered as adaptive intelligent controllers for road traffic networks. A simple traffic network simulator is employed and results are presented showing that within this environment both classifier system models are capable of learning reasonable policies. It is concluded that for reasons discussed, ZCS may be the more suitable system for this class of problem. Initial t...

We investigate Learning Classifier Systems for online environments that consist of real-valued states and which require every action made by the agent to count towards its performance. Two Learning Classifier System architectures are considered, ZCS and XCS. We use an interval representation with these Learning Classifier Systems for the rule conditions together with roulette wheel action selec...

Models of propulsion motors, generators, gas turbines, and power converters are used to determine weights and volumes, evaluate designs, and predict performance of power system components for all-electric navy ships. The finite element analysis method using 3D CAD models is used to validate designs and predict performance under prescribed constraints that simulate realistic operational conditio...

A new zero voltage switching (ZVS) dc / dc converter operating on constant frequency and having wide linearity is proposed. ZVS operations are achieved not only for the primary switches but also for the secondary rectifier diodes to reduce the switching stresses and losses. The proposed converter also overcomes the other shortcomings of the conventional resonant dc / dc converters such as high ...

This paper considers the transient stability enhancement of a power system containing large amounts of solar and wind generation in Japan. Following the Fukushima Daiichi nuclear disaster there has been an increasing awareness on the importance of a distributed architecture, based mainly on renewable generation, for the Japanese power system. Also, the targets of CO2 emissions can now be approa...

CERN is working on a new particle accelerators that will require a very large number of power converters. In that view, the reliability of the whole powering will be a major issue. The use of a redundancy and modularity may help increasing the overall machine availability. However, the reliability of the redundancy system must be high enough to add a significant improvement when compared to sim...