A new intelligent temperature control system for stove was introduced. It could accomplish A/D conversion, data processing, output displaying and alarming while the temperature exceeds the given threshold. The tempera...A new intelligent temperature control system for stove was introduced. It could accomplish A/D conversion, data processing, output displaying and alarming while the temperature exceeds the given threshold. The temperature adjusting devices were bi direction SCR and heating resistor. The hardware and software were discussed in the paper.展开更多
Reinforcement learning is an excellent approach which is used in artificial intelligence,automatic control, etc. However, ordinary reinforcement learning algorithm, such as Q-learning with lookup table cannot cope wit...Reinforcement learning is an excellent approach which is used in artificial intelligence,automatic control, etc. However, ordinary reinforcement learning algorithm, such as Q-learning with lookup table cannot cope with extremely complex and dynamic environment due to the huge state space. To reduce the state space, modular neural network Q-learning algorithm is proposed, which combines Q-learning algorithm with neural network and module method. Forward feedback neural network, Elman neural network and radius-basis neural network are separately employed to construct such algorithm. It is revealed that Elman neural network Q-learning algorithm has the best performance under the condition that the same neural network training method, i.e. gradient descent error back-propagation algorithm is applied.展开更多
文摘A new intelligent temperature control system for stove was introduced. It could accomplish A/D conversion, data processing, output displaying and alarming while the temperature exceeds the given threshold. The temperature adjusting devices were bi direction SCR and heating resistor. The hardware and software were discussed in the paper.
文摘Reinforcement learning is an excellent approach which is used in artificial intelligence,automatic control, etc. However, ordinary reinforcement learning algorithm, such as Q-learning with lookup table cannot cope with extremely complex and dynamic environment due to the huge state space. To reduce the state space, modular neural network Q-learning algorithm is proposed, which combines Q-learning algorithm with neural network and module method. Forward feedback neural network, Elman neural network and radius-basis neural network are separately employed to construct such algorithm. It is revealed that Elman neural network Q-learning algorithm has the best performance under the condition that the same neural network training method, i.e. gradient descent error back-propagation algorithm is applied.
