2 results about "Batch process control" patented technology

The embodiment of the invention discloses a quality data evaluation method for a multi-variety small-batch manufacturing process, and the method comprises the steps: 1, employing a tolerance coefficient method as a data conversion algorithm, and obtaining a tolerance utilization rate of a single monitoring characteristic; step 2, performing process capability evaluation on each monitoring characteristic, combining standardized data of multiple monitoring characteristics for single monitoring characteristics capable of accumulating the total sample number N and single monitoring characteristics incapable of accumulating the total sample number N so as to perform control process capability evaluation, and performing process control in an auxiliary single-value monitoring mode; and step 3, evaluating the control process capability in the step 2, and formulating a quality state evaluation criterion of multi-variety small-batch process control. According to the technical scheme provided by the embodiment of the invention, the problems of low accuracy of quality evaluation, difficulty in establishing an effective control model and control limit and the like of a quality control mode of multi-variety and small-batch products are solved.

The application discloses a kind of reinforcement learning batch process control methods based on improved AC algorithm, it is related to the field of deep reinforcement learning and batch process control field.The method will be based on reinforcement learning method The batch process control is modeled as an optimal control problem on the basis of Markov decision process;Control action constraint is introduced in the reward function of reinforcement learning controller, the number of effective reward samples is increased to improve the learning rate of reinforcement learning controller, and the control cycle is shortened.Priority sampling method is introduced in the Actor-Critic algorithm of deep reinforcement learning, and a soft actor-critic algorithm with priority sampling is proposed to improve the sampling efficiency in the experience replay pool.The present application does not depend on prior knowledge and process model, and can realize model-free control of batch process.