18-04-2023 дата публикации
Номер: CN115982588A
Принадлежит:
The embodiment of the invention relates to the technical field of reinforcement learning, and provides a cement kiln condition judgment method and device based on reinforcement learning, equipment and a medium, and the method comprises the steps: obtaining a training sample set, obtaining current state data in the training sample set, inputting the current state data into a cement kiln condition judgment model, and obtaining a predicted cement kiln condition, determining a reward and punishment feedback value and a first loss function value according to the predicted cement kiln condition and the correct cement kiln condition, forming an interaction pair by the current state data, the predicted cement kiln condition, the reward and punishment feedback value and the next state data, storing the interaction pair in an experience pool, and determining a second loss function value according to a plurality of randomly selected interaction pairs, and updating the model according to the two loss ...
Подробнее