Alkane Dehydrogenation Control Using Yield and Coking Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for optimizing alkane dehydrogenation processes, such as propane dehydrogenation, face challenges in accurately predicting product yield and coking rate, leading to inefficiencies and production waste, as they do not account for both yield and coking rate optimization simultaneously.
Innovation Solution
A method and system that utilize historical data filtering, machine learning processing, and mass/heat balance adjustments to develop predictive models for product yield and coking rate, allowing for real-time optimization of production parameters to maximize yield while controlling coking rates below predetermined levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional prediction methods are used for product yield, then prediction can be made, but prediction accuracy is insufficient and cannot predict coking rate
Solution Approach 1:
The patent combines product yield prediction and coking rate prediction into a single integrated machine learning model. The model simultaneously processes multiple input parameters (temperature, pressure, flow rates, catalyst properties) to predict both output variables (product yield and coking rate) together, rather than using separate conventional prediction methods. This merging approach enables accurate prediction of both parameters while capturing their interrelationships.
Solution Approach 2:
The machine learning model serves multiple functions: it predicts product yield, predicts coking rate, identifies optimal operating conditions, and provides early warnings for coking issues. This multi-functional model replaces multiple specialized conventional models, achieving comprehensive prediction capabilities with a single unified system that improves overall accuracy and information retention.
2Productivity
If production parameters are adjusted without accurate prediction, then process optimization is attempted, but process stability deteriorates and production waste increases
Solution Approach 1:
The system continuously monitors actual process parameters and compares them with predicted values from the machine learning model. The prediction results feed back into the control system to adjust operating conditions, creating a closed-loop feedback mechanism. This ensures that parameter adjustments are based on accurate predictions, maintaining process stability while improving efficiency and reducing waste.
Solution Approach 2:
The machine learning model predicts future process outcomes and coking rates before actual changes occur. This preliminary prediction capability allows operators to anticipate problems and make proactive adjustments to prevent instability and waste, rather than reacting to issues after they arise. The system performs virtual experiments through prediction to identify optimal adjustment paths.
3Loss of information
If multiple production parameters are monitored simultaneously, then comprehensive data is collected, but system complexity increases
Solution Approach 1:
The patent replaces complex manual data processing and analysis systems with a machine learning-based computational system. The machine learning model automatically processes multiple input parameters (temperature, pressure, flow rates, catalyst properties) through algorithms, eliminating the need for complex manual monitoring and analysis procedures. This substitution reduces operational complexity while maintaining comprehensive data utilization.
Solution Approach 2:
The system transforms multiple raw process parameters into optimized feature representations that the machine learning model can process efficiently. By changing the form and organization of input data into meaningful features, the system reduces processing complexity while preserving all necessary information for accurate prediction of both product yield and coking rate.
Data Source
AI summary
The present invention relates to a method for optimizing an alkane dehydrogenation operation. The method comprises the steps of obtaining historical data of a plurality of production parameter data, filtering abnormal data of the said historical data, developing a product yield prediction model and a coking rate prediction model, and processing the product yield prediction model and the coking rate prediction model, and determining optimum production parameter data based on the predicted product yield and the predicted coking rate. The present invention further relates to a system for optimizing an alkane dehydrogenation operation which comprises a production data detecting and storage unit and a processor configured to obtain historical data of the production parameter data, filter abnormal data of the historical data of the production parameter data, develop a product yield prediction model and a coking rate prediction model, and provide a process efficiency analysis model with a machine learning processing unit for processing the product yield prediction model and the coking rate prediction model to predict a product yield and a coking rate, and determine optimum production parameter data based on the predicted product yield and the predicted coking rate.