Home Artificial Intelligence Model-Free Reinforcement Learning for Chemical Process Development Introduction Problem Definition

Model-Free Reinforcement Learning for Chemical Process Development Introduction Problem Definition

0
Model-Free Reinforcement Learning for Chemical Process Development
Introduction
Problem Definition

Towards Universal Chemical Process Operators

Towards Data Science
Photo by Alex Kondratiev on Unsplash

Process development, design, optimization, and control are a number of the most important duties inside chemical and process engineering. In concrete terms, the scope is finding an optimal recipe or suitable configuration of apparatus or process parameters (via laboratory experiments) in order that certain objectives (e.g., yield or throughput) are maximized while potential constraints (e.g., input concentrations, flow rates, reactor volumes, or boiling points of solvents) are respected. By automating these tasks, e.g., through laboratory robots, an awesome deal of manual labor could possibly be saved.

The recent progress in reinforcement learning (RL) made it clear that agents can master complex tasks and play a wide range of games, and even discover more efficient mathematical procedures, e.g., for matrix operations. With the supply of kinetic parameters, either from experiments or numerical simulations, agents may find optimal configurations and synthesis recipes. In contrast to convex optimization, nonetheless, the algorithm/model might be directly used for process control. Such experiments can happen either on the pc or directly within the laboratory, depending on the sample efficiency of the strategy. In the long run, this may (partially) automate process development. The scope of this text is for instance this on the instance of paracetamol using proximal policy optimization (PPO).

We now have a pc program, a so-called agent, here we call it an universal chemical process operator. This operator finds itself in an environment by which it will probably perform chemical operations, i.e., actions. Such actions include dosing component A, increasing/decreasing input/output flow, increasing/decreasing temperature, and so forth. Because the agent perform actions in certait states corresponding to concentrations of certain components, it transitions into latest states.

Paracetamol (PC) is synthesized from p-aminophenol (AP) and acetic anhydride (AA), shown in Fig. 1a. Under known kinetics, this process might be modeled and represents the environment, e.g., in a continuous stirred-tank reactor (CSTR) as shown in Fig…

LEAVE A REPLY

Please enter your comment!
Please enter your name here