A Sewage Treatment Control Method Based on Reinforcement Learning

Abstract

The invention provides a sewage treatment control method based on reinforcement learning, comprising the following steps: the invention trains the reinforcement learning model in a real scene in three stages, which can effectively avoid the collection of data required by the existing training reinforcement learning model , The process of establishing a virtual environment, so as to effectively reduce the cost required for enterprises to apply reinforcement learning for automatic control, and facilitate users to complete the process of switching from traditional control to reinforcement learning control.

Description

technical field [0001] The invention relates to a sewage treatment control method based on reinforcement learning. Background technique [0002] At present, the application of reinforcement learning in industrial control is gradually increasing, but it is generally limited to a specific model. The main reason is that the training of reinforcement learning requires environmental support, and in many sewage treatment control, it is necessary to completely simulate the environment of the real scene. The amount of calculation is far greater than the amount of calculation required for reinforcement learning model training itself, resulting in more losses than gains, and in terms of current enterprise technology development, the accumulation of original data is also very problematic. Contents of the invention [0003] In order to solve the above technical problems, the present invention provides a sewage treatment control method based on reinforcement learning. The sewage treatm...

AI Technical Summary

Problems solved by technology

[0002] At present, the application of reinforcement learning in industrial control is gradually increasing, but it is generally limited to a specific model. The main reason is that the training of reinforcement learning requires environmental support, and in many sewage treatment control, it is necessary to completely simulate the environment of the real scene. The amount of calculation is far greater than the amount of calculation required for reinforcement learning model training itself, resulting in more losses than gains, and in terms of current enterprise technology development, the accumulation of original data is also very problematic

Images