20240036532. METHOD AND SYSTEM FOR MODELLING INDUSTRIAL PROCESSES simplified abstract (Siemens Energy AS)
Contents
METHOD AND SYSTEM FOR MODELLING INDUSTRIAL PROCESSES
Organization Name
Inventor(s)
Esmaeil Jahanshahi of Trondheim (NO)
Arnstein Rekstad of Trondheim (NO)
Vegard Kleppe Torkelsen of Oslo (NO)
METHOD AND SYSTEM FOR MODELLING INDUSTRIAL PROCESSES - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240036532 titled 'METHOD AND SYSTEM FOR MODELLING INDUSTRIAL PROCESSES
Simplified Explanation
The patent application describes a method and system for modeling industrial processes, specifically closed loop feedback processes. The system includes a sensor to measure an input for the industrial process, and a processor that receives the measurement of the input from the sensor. The processor implements a hybrid neural network model to output a derivative of the output at the input time. This neural network model incorporates at least one neural network block and a first-principle block that incorporates a dynamic model with an ordinary differential equation defining the rate of change over time of the output as a function of the input. The derivative is then inputted to an ordinary differential equation solver to predict the output at a subsequent time, and the prediction is outputted using the measured input.
- The patent application describes a method and system for modeling industrial processes with closed loop feedback.
- The system includes a sensor to measure the input for the industrial process and a processor to receive the measurement.
- The processor implements a hybrid neural network model that includes neural network blocks and a first-principle block with a dynamic model.
- The dynamic model uses an ordinary differential equation to define the rate of change of the output over time as a function of the input.
- The processor uses the neural network model to output a derivative of the output at the input time.
- The derivative is then used as input for an ordinary differential equation solver to predict the output at a subsequent time.
- The predicted output is then outputted using the measured input.
Potential Applications of this Technology:
- Industrial process control and optimization
- Predictive maintenance in manufacturing
- Energy management and optimization in power plants
- Quality control in production processes
Problems Solved by this Technology:
- Accurate modeling and prediction of industrial processes
- Improved control and optimization of closed loop feedback processes
- Enhanced efficiency and productivity in manufacturing and power generation
Benefits of this Technology:
- Improved process control and optimization leading to cost savings and increased productivity
- Enhanced predictive maintenance capabilities, reducing downtime and maintenance costs
- More accurate energy management, leading to energy savings and reduced environmental impact
- Improved quality control, reducing defects and waste in production processes
Original Abstract Submitted
a method and system for modelling industrial processes, including closed loop feedback processes. the system includes a sensor for measuring an input for the industrial process; and a processor configured to receive a measurement of the input from the sensor at an input time. the processor is also configured to implement a hybrid neural network model to output a derivative of the output at the input time, wherein the neural network model incorporates at least one neural network block and a first-principle block incorporating dynamic model having an ordinary differential equation defining the rate of change over time of the output as a function of the or each associated input; to input the derivative to an ordinary differential equation solver to predict the output at a subsequent time; and to output the prediction of the output at the subsequent time using the measured input.
