Controlling a chemical reactor for the production of polymer compounds
Techniques regarding the synthesis of one or more polymers of a target polymer class are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise a recommendation component that can generate a recommended chemical reactor control setting for inverse synthesis of a polymer based on a target polymer characteristic and reactor training data.
1. A system, comprising:
a memory;
a processor, operably coupled to the memory, wherein the processor:
operates a chemical reactor in accordance with a recommended chemical reactor control setting to generate a polymer;
determines a characteristic of the polymer generated by the chemical reactor; and
determines whether the characteristic is within a permissible range defined bya target polymer characteristic; and
generates a recommended chemical reactor control setting for inverse synthesis of the polymer based on the target polymer characteristic and reactor training data.
2. The system of claim 1 , wherein the processor also:
generates an adjacency matrix based on the reactor training data, wherein the reactor training data comprises a plurality of reactor control settings, and generates the adjacency matrix by combining the plurality of reactor control settings into a plurality of control setting tuples.
3. The system of claim 2 , wherein the plurality of reactor control settings regard reactor parameters previously used to generate a synthesized polymer, and wherein the reactor training data further comprises a characteristic of the synthesized polymer.
4. The system of claim 2 wherein the processor also:
performs a link prediction in network based on the adjacency matrix with at least one partition.
5. The system of claim 4 , wherein the link prediction in network comprises a process selected from a group consisting of a matrix factorization, a tensor factorization, and a deep learning model for a network completion task applied to the adjacency matrix.
6. The system of claim 1 , wherein the processor also:
updates the reactor training data with the characteristic of the polymer and the recommended chemical reactor control setting.
7. The system of claim 1 , wherein the processor also:
computes latent embedding of an initial adjacency matrix to generate a reconstructed adjacency matrix, wherein the initial adjacency matrix comprises the reactor training data, and
performs an entry-wise comparison of the initial adjacency matrix and the reconstructed adjacency matrix.
8. A computer-implemented method, comprising:
operating, by a system, operatively coupled to a processor, a chemical reactor in accordance with a recommended chemical reactor control setting to generate the polymer,
determining, by the system, a characteristic of the polymer generated by the chemical reactor;
determining, by the system, whether the characteristic is within a permissible range defined by a target polymer characteristic; and
generating, by a system operatively coupled to a processor, the recommended chemical reactor control setting for inverse synthesis of the polymer based on the target polymer characteristic and reactor training data.
9. The computer-implemented method of claim 8 , further comprising:
training, by the system, a machine learning algorithm based on the reactor training data via an adjacency matrix, wherein the reactor training data regards a synthesis of a previously generated polymer.
10. The computer-implemented method of claim 9 , wherein the training comprises:
computing a latent embedding of the adjacency matrix to generate a reconstruction adjacency matrix.
11. The computer-implemented method of claim 9 , further comprising:
performing, by the system, a link prediction in network analysis based on the adjacency matrix with at least one partition, wherein the link prediction in network analysis is selected from a group consisting of: a matrix factorization, a tensor factorization, and a deep learning model for a network completion task applied to the adjacency matrix.
12. A computer program product for controlling a chemical reactor to produce a polymer, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:
operate, by a system operatively coupled to the processor, a chemical reactor in accordance with the recommended chemical reactor control setting to generate the polymer; and
generate, by a system operatively coupled to the processor, a recommended chemical reactor control setting for inverse synthesis of the polymer based on a target polymer characteristic and reactor training data.
13. The computer program product of claim 12 , wherein the program instructions further cause the processor to:
collect, by the system, the reactor training data via a cloud computing environment.
14. The computer program product of claim 12 , wherein the program instructions further cause the processor to:
train, by the system, a machine learning algorithm based on the reactor training data via an adjacency matrix, wherein the reactor training data regards a synthesis of a previously generated polymer.
15. The computer program product of claim 14 , wherein the program instructions further cause the processor to:
perform, by the system, a link prediction in network analysis based on the adjacency matrix with at least one partition, wherein the link prediction in network analysis is selected from a group consisting of: a matrix factorization, a tensor factorization, and a deep learning model for a network completion task applied to the adjacency matrix.