Techniques for automated component classification

1 . A method for generating intelligent drawings, comprising:

obtaining a current set of design drawings for a physical plant comprising a plurality of components;

generating rasterized drawings from the current set of design drawings;

applying an object detection model to the rasterized drawings to extract a set of shapes corresponding to components in the current set of design drawings;

applying a trained library to the extracted shapes to obtain a classification of components in the current set of design drawings; and

propagating the classification to instances of the components in the current set of design drawings by applying the classification to vector drawings corresponding to the current set of design drawings to obtain vector standardized drawings.

2 . The method of claim 1 , wherein the trained library is trained by:

providing a previous set of design drawings prior to the current set of design drawings;

generating prior rasterized drawings from the previous set of design drawings;

applying the object detection model to the rasterized drawings to extract a prior set of shapes corresponding to components in the previous set of design drawings;

providing a user interface comprising one or more of the prior set of shapes and a user input component to classify the one of more of the prior set of shapes; and

receiving, via the user interface, a classification for the one or more prior set of shapes, wherein the trained library is trained based on the received classification for the one or more prior sets of shapes.

3 . The method of claim 1 , wherein applying a trained library to the extracted shapes to obtain a classification of components in the current set of design drawings comprises:

obtaining an initial classification for the extracted shapes; and

presenting a prompt, in a user interface, to a user to confirm the initial classification for the extracted shapes, wherein the classification is propagated in accordance with a received confirmation of the initial classification.

4 . The method of claim 3 , further comprising:

receiving, in response to the prompt, a correction to the initial classification to obtain the classification; and retraining the trained library based on the correction.

5 . The method of claim 1 , wherein the current set of design drawings comprises inconsistent representations of components, and wherein the vector standardized drawings comprise standardized representations of the components.

6 . The method of claim 1 , wherein the current set of design drawings are obtained from a plurality of sources.

7 . The method of claim 1 , further comprising:

obtaining a 3D model of a system corresponding to the current set of design drawings;

applying a mapping process to identify the plurality of components from a 2D model against the 3D model; and

enhancing the 3D model to include the classified components.

8 . A non-transitory computer readable medium comprising computer readable code for generating intelligent drawings, the computer readable code executable by one or more processors to:

obtain a current set of design drawings for a physical plant comprising a plurality of components;

obtain rasterized drawings from the current set of design drawings;

apply a trained model to the rasterized drawings, wherein the trained model identifies and classifies components in the current set of design drawings based on a trained library;

update a data structure to include classified components based on the application of the trained model; and

apply the data structure to vector drawings corresponding to the current set of design drawings to obtain vector standardized drawings.

9 . The non-transitory computer readable medium of claim 8 , wherein the trained library is trained by:

providing a previous set of drawings prior to the current set of drawings;

generating prior rasterized drawings from the previous set of drawings;

applying an object detection model to the rasterized drawings to extract a prior set of shapes corresponding to components in the drawings;

providing a user interface comprising one or more of the prior set of shapes and a user input component to classify the one of more of the prior set of shapes; and

receiving, via the user interface, a classification for the one or more prior set of shapes, wherein the trained library is trained based on the received classification for the one or more prior sets of shapes.

10 . The non-transitory computer readable medium of claim 8 , wherein the computer readable code to update the data structure to include the classified components comprises computer readable code to:

obtain an initial classification for the components; and

present a prompt, in a user interface, to a user to confirm the initial classification for the components, wherein the classification is propagated in accordance with a received confirmation of the initial classification.

11 . The non-transitory computer readable medium of claim 10 , further comprising computer readable code to:

receive, in response to the prompt, a correction to the initial classification to obtain the classification; and

retrain the trained library based on the correction.

12 . The non-transitory computer readable medium of claim 8 , wherein propagating the classification comprises applying the classification to vector drawings corresponding to the current set of design drawings to obtain vector standardized drawings.

13 . The non-transitory computer readable medium of claim 12 , wherein the current set of design drawings comprises inconsistent representations of components, and wherein the vector standardized drawings comprise standardized representations of the components.

14 . The non-transitory computer readable medium of claim 12 , wherein the current set of design drawings are obtained from a plurality of sources.

15 . The non-transitory computer readable medium of claim 8 , further comprising computer readable code to:

obtain a 3D model of a system corresponding to the current set of design drawings;

apply a mapping process to identify the plurality of components from a 2D model against the 3D model; and

enhance the 3D model to include the classified components.

16 . A system for generating intelligent drawings, comprising:

one or more processors; and one or more computer readable media comprising computer readable code executable by the one or more processors to:

obtain a current set of design drawings for a physical plant comprising a plurality of components;

obtain rasterized drawings from the current set of design drawings;

apply a trained model to the rasterized drawings, wherein the trained model identifies and classifies components in the current set of design drawings based on a trained library;

update a data structure to include classified components based on the application of the trained model; and

apply the data structure to vector drawings corresponding to the current set of design drawings to obtain vector standardized drawings.

17 . The system of claim 16 , wherein the trained library is trained by:

providing a previous set of design drawings prior to the current set of design drawings;

generating prior rasterized drawings from the previous set of design drawings;

applying an object detection model to the rasterized drawings to extract a prior set of shapes corresponding to components in the previous set of design drawings;

providing a user interface comprising one or more of the prior set of shapes and a user input component to classify the one of more of the prior set of shapes; and

receiving, via the user interface, a classification for the one or more prior set of shapes, wherein the trained library is trained based on the received classification for the one or more prior sets of shapes.

18 . The system of claim 16 , wherein the computer readable code to update the data structure to include the classified components comprises computer readable code to:

obtain an initial classification for the components; and

present a prompt, in a user interface, to a user to confirm the initial classification for the components, wherein the classification is propagated in accordance with a received confirmation of the initial classification.

19 . The system of claim 18 , further comprising computer readable code to:

receive, in response to the prompt, a correction to the initial classification to obtain the classification; and

retrain the trained library based on the correction.