Batch Process Control With Cascaded MPC for Refinery Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing industrial process control systems struggle to seamlessly integrate and optimize continuous and batch process operations in refineries, leading to inefficiencies and disruptions.
Innovation Solution
An IoT platform for industrial process control and optimization that uses real-time models and control to harmonize plantwide optimization with batch operations, employing a cascaded MPC architecture and virtual content trackers to minimize disruptions and optimize performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If batch operations are integrated with continuous process sections in a refinery, then the batch section can utilize blending components from the continuous section to produce refined products, but the batch operations may disrupt the continuous process section
Solution Approach 1:
The control system is segmented into multiple independent controllers - a first controller for the continuous process section and a second controller for the batch process section. This segmentation allows each controller to independently manage its respective process without mutual disruption, while still enabling integration through coordinated operation. The continuous process maintains its stability through dedicated control while the batch section utilizes blending components from the continuous section.
2Device complexity
If a single control system manages both continuous and batch operations, then integration is simplified, but the complexity of controlling both operation types increases
Solution Approach 1:
The control system is divided into separate first and second controllers that independently manage continuous and batch operations respectively. This segmentation reduces the complexity within each controller while maintaining overall system integration. Each controller is optimized for its specific operation type, making operation simpler and more intuitive for operators while avoiding the overwhelming complexity of a monolithic control system.
Solution Approach 2:
A coordination mechanism acts as an intermediary between the first and second controllers, enabling seamless integration of continuous and batch operations. This intermediary layer manages the interaction between the two operation types, handling the exchange of blending components and coordinating schedules without requiring direct complex interactions between the controllers, thus simplifying operation while maintaining integration.
3Adaptability or versatility
If batch operations run on unpredictable schedules, then flexibility in production is improved, but planning and optimization become more difficult
Solution Approach 1:
The second controller receives feedback regarding the schedule of the batch process and uses this information to coordinate with the first controller. This feedback mechanism allows the system to adapt to unpredictable batch schedules while maintaining optimal planning and optimization. The continuous process controller can adjust its operation based on real-time feedback about batch timing and requirements, ensuring both flexibility and informed decision-making.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
Various embodiments described herein relate to management of industrial process optimization related to batch operations. In this regard, an optimization request to optimize an industrial process that produces an industrial process product is received. Thereafter, a first time period and an optimization time period are determined to produce the industrial process product and to update industrial process product characteristics. In addition, blending component characteristics indicative of availability of one or more blending components are updated for the batch process. Furthermore, in response to completing the production of the industrial process product, product spent characteristics associated are determined to update demand data for the one or more feed products utilized for blending the one or more blending products based on the product spent characteristics. Finally, a control signal is transmitted to control the flow of the feed component.