Abstract:Large scale combustion process is complex and proposes challenges of optimizing its performance. Traditional approaches based on thermal dynamics have limitations on finding optimal operational regions due to time-shift nature of the process. Recent advances in information technology enable people collect large volumes of process data easily and continuously. The collected process data contains rich information about the process and, to some extent, represents a digital copy of the process over time. Although large volumes of data exist in industrial combustion processes, they are not fully utilized to the level where the process can be optimized. Data mining is an emerging science which finds patterns or models from large data sets. It has found many successful applications in business marketing, medical and manufacturing domains The focus of this dissertation is on applying data mining to industrial combustion processes, and ultimately optimizing the combustion performance. However the philosophy, methods and frameworks discussed in this research can also be applied to other industrial processes. Optimizing an industrial combustion process has two major challenges. One is the underlying process model changes over time and obtaining an accurate process model is nontrivial. The other is that a process model with high fidelity is usually highly nonlinear, solving the optimization problem needs efficient heuristics. This dissertation is set to solve these two major challenges. The major contribution of this 4-year research is the data-driven solution to optimize the combustion process, where process model or knowledge is identified based on the process data, then optimization is executed by evolutionary algorithms to search for optimal operating regions.

Optimization of Temporal Processes: a Model Predictive Control Approach

Dynamic Modeling And Nonlinear Predictive Control Based On Partitioned Model And Nonlinear Optimization

Multiobjective optimization of temporal processes.

Multi-rate Multivariable Model Predictive Control For Integrating Processes

Adaptive nonlinear model predictive control for a class of multivariable chemical processes

A Time-Delay Modeling Approach for Data-Driven Predictive Control of Continuous-Time Systems

Anticipatory Control Of Wind Turbines With Data-Driven Predictive Models

Learning-Based Neural Dynamic Surface Predictive Control for MMC

Predictive Control for Unknown Dynamics with Observation Loss: A Temporal Game-Theoretic Approach

Dynamic Control of Wind Turbines

Meta-control of combustion performance with a data mining approach

Design Of Model Predictive Controller Based On Neural Systems

Temporal Difference Learning for Model Predictive Control

A novel predictive control algorithm and robust stability criteria for integrating processes.

Virtual unmodeled dynamic and data-driven nonlinear robust predictive control

T-S Model-Based Predictive Control For Nonlinear Systems Based On Min-Max Optimization

An Adaptive Model Predictive Control Strategy for Discrete-Time Nonlinear System

Adaptive Predictive Control of a Kind of Processes Based on a Revised Wiener Model

Data-Driven Predictive Control for Continuous-Time Industrial Processes with Completely Unknown Dynamics

Synthesis of model predictive control based on data-driven learning

Predictive Control for Processes with Input Dynamic Nonlinearity