Daniel G. McClement

Daniel G. McClement

Auto-generated profile, based on publicly available information

Daniel G. McClement

The University of British Columbia

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 January 2022 in Control Engineering Practice
Verified
Journal Article
Closed access
NLNathan P. Lawrence
Nathan P. Lawrence
MFMichael G. Forbes
Michael G. Forbes
PLPhilip D. Loewen
Philip D. Loewen
DMDaniel G. McClement
Daniel G. McClement
JBJohan U. Backström
Johan U. Backström
BGBhushan Gopaluni
Bhushan Gopaluni

17 September 2022 in Journal of Process Control
Verified
Journal Article
Closed access
DMDaniel G. McClement
Daniel G. McClement
NLNathan P. Lawrence
Nathan P. Lawrence
JBJohan U. Backström
Johan U. Backström
PLPhilip D. Loewen
Philip D. Loewen
MFMichael G. Forbes
Michael G. Forbes
RGR. Bhushan Gopaluni
R. Bhushan Gopaluni

24 September 2021
Verified
Preprint
Open access
PDF file
DMDaniel McClement
Daniel McClement
NLNathan P. Lawrence
Nathan P. Lawrence
PLPhilip Loewen
Philip Loewen
MFMichael Forbes
Michael Forbes
JBJohan Backstrom
Johan Backstrom
BGBhushan Gopaluni
Bhushan Gopaluni
Fixed structure controllers (such as proportional-integral-derivative controllers) are used extensively in industry. Finding a practical and versatile method to tune these controllers, particularly with imprecise process models and limited online computational resources, is an industrially relevant problem which could improve the efficiency of many plants. In this paper, we present two flexible neural network-based approaches capable of tuning any fixed structure controller for any control objective and process model and compare their advantages and disadvantages. The first approach is derived from supervised learning and classical optimization techniques, while the second approach applies techniques used in deep reinforcement learning. Both approaches incorporate model uncertainties when selecting controller parameters, reducing the need for costly experiments to precisely estimate model parameters in a plant. Both methods are also computationally efficient online, enabling their widespread usage.