Dynamic Graph Representation for Chemical Process Network Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing process engineering models for chemical plants are static, cumbersome to design, and struggle to adapt to different optimization problems or capture interdependencies between multiple plants, leading to less robust monitoring and control results.
Innovation Solution
A computer-implemented method generates a dynamic, problem-specific representation of a process network by combining classical balance-equation based models with data-driven models, using a graph structure that collapses unobservable quantities, allowing for flexible adaptation to changing conditions and improved accuracy in monitoring and controlling interconnected chemical plants.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If classical balance-equation based models are used, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the process network into modular graph components (vertices representing units, edges representing streams) that can be independently modeled and recombined. This allows complex systems to be broken down into manageable units while maintaining overall modeling accuracy through systematic assembly of balance equations for each segment.
Solution Approach 2:
The patent creates a universal graph-based modeling framework that can represent diverse chemical engineering units (reactors, separators, heat exchangers) using common vertex and edge elements. This multi-functional approach maintains precision across different unit types while reducing overall model complexity through standardized representations.
2Device complexity
If static models are used, then device complexity is reduced, but adaptability worsens
Solution Approach 1:
The patent implements dynamic model generation where the graph structure and balance equations are automatically assembled based on the specific problem requirements. The system can dynamically add, remove, or modify units and streams in the graph representation, allowing the same framework to adapt to different optimization problems without manual reconfiguration.
Solution Approach 2:
The patent allows problem-specific parameters and constraints to be modified within the unified graph framework. Users can change physical properties, operating conditions, or objective function parameters while the underlying graph structure remains intact, enabling flexible adaptation to different scenarios without reconstructing the entire model.
3Reliability
If detailed models capturing all interdependencies are used, then reliability is improved, but computation time increases
Solution Approach 1:
The patent extracts and represents only the essential interdependencies between process units in the graph structure, focusing on mass and energy flow relationships that are critical for monitoring and control. Less critical detailed interactions can be omitted or simplified, maintaining reliability for key functions while reducing overall computation time.
Solution Approach 2:
The patent implements a hierarchical modeling approach where full detailed models are used only when necessary, while simpler aggregate models handle routine monitoring. The system can selectively apply detailed balance equations to specific critical units or time periods, achieving sufficient reliability without the computational burden of exhaustive detailed modeling everywhere.
Data Source
AI summary
A computer implemented method for generating a problem specific representation of a process network to enable controlling or monitoring a process network with at least two interconnected chemical plants, the method comprising the steps of providing a first digital representation of the process network comprising a digital process representation of each plant, its connections to other plants and sensor elements placed in the process network, generating based on the first digital representation a graph structure including vertices representing unit operations, edges linking unit operations representing at least physico-chemical quantities, wherein the edges include edge meta data representing at least physico-chemical quantities, and a measurable tag, generating based on the graph structure a collapsed graph structure including, vertices representing virtual unit operations, edges linking virtual unit operations representing at least physico-chemical quantities, wherein the edges include edge meta data representing observable physico chemical quantities, and their relation to vertices, deriving a set of balance equations from the collapsed graph structure, providing the set of balance equations, and physico- chemical quantities for monitoring and/or controlling operation of a process network is proposed.