Controller performance benchmarking and tuning using generalised minimum variance control

Grimble, M.J.

doi:10.1016/s0005-1098(02)00141-3

Cited by 134 publications

(56 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The calculations require the loop time delays which for the level and temperature loops were 2 s and 8 s respectively. The regulatory performances for Tests 1 to 8 were determined from 2000 samples from regulatory operation using equations (8) and (9). Performance indexes were also determined for regulatory operation in Tests 9 to 15 and for the industrial data.…”

Section: Discussionmentioning

confidence: 99%

“…These features can be seen in lower middle panel of Figure 3 where the most vigorous mv action is in Test 5 and 7. [8][9][10], and elsewhere in this Special Issue, where the need for an enhanced controller performance measure that pays attention to the mv has been identified.…”

Section: Discussionmentioning

confidence: 99%

“…The upper left hand panel of Figure 8 shows that the tests dominated by deterministic flow disturbance (10, 14, 15) had a low η while those dominated by the random bubble disturbance (9,11,12) had a high η . Thus the η value during regulatory control responds to the nature of the disturbance even though the controller tuning setting did not change.…”

Section: Discussionmentioning

confidence: 99%

“…There is thus an incentive to relax the minimum variance requirement. Kadali and Huang [8] and Grimble [9] have described benchmarks for Linear Quadratic Gaussian and Generalised Minimum Variance control that take into account the manipulated variable (i.e. the input into the controlled system) as well as the controlled variable in order to strike a balance between variability in the controlled variable and wear on the actuator.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Controller performance assessment in set point tracking and regulatory control

Thornhill

Huang

Shah

2003

Adaptive Control & Signal

View full text Add to dashboard Cite

Recent critiques of minimum variance benchmarking for single-input-single-output control loops have focused on the need for assessment of performance during set point changes and also on the need to pay attention to the movements in the manipulated variable. This paper examines factors that influence the minimum variance performance measure of a SISO control loop. It discusses the reasons why performance during set point changes differs from the regulatory performance during operation at a constant set point. The results demonstrate how regulatory performance is influenced by the nature of a disturbance, and that correlation of signals within a control loop can indicate whether the disturbance is random or deterministic. The paper is illustrated with simulated, experimental and industrial examples.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Controller performance assessment in set point tracking and regulatory control

Thornhill

Huang

Shah

2003

Adaptive Control & Signal

View full text Add to dashboard Cite

show abstract

“…The GMV control philosophy has characteristics midway between the MV and LQG design philosophies. Both methods provide real advantages for benchmarking, particularly in comparison with MV-based algorithms [102,103,105,106].…”

Section: Benchmarkingmentioning

confidence: 99%

Introduction to Optimal and Robust Control

Grimble

2006

Robust Industrial Control Systems

View full text Add to dashboard Cite

Limits of control performance for distributed networked control systems in presence of communication delays

Yan

Liu

Huang

2018

Adaptive Control & Signal

View full text Add to dashboard Cite

SummaryIn terms of computational complexity and fault tolerance, distributed networked control systems (DNCSs) is favorable for large‐scale processes. However, it poses additional limitations on the achievable control performance, especially when communication delay is present. The conventional minimum variance (MV) benchmarks mainly consider the limitations caused by the system itself and can give overly estimates of achievable performance when applied to the systems under distributed networked control. This paper proposes a solution to the MV benchmark for DNCSs considering both system time delays and time‐invariant communication delays. Furthermore, lower and upper bounds of the MV benchmark are proposed to assess the performance of DNCSs when there are time‐varying communication delays. These results are useful for evaluating the potential performance improvement if a DNCS is implemented to replace a decentralized control system. The proposed results are illustrated by a simulation example.

show abstract

Controller performance benchmarking and tuning using generalised minimum variance control

Cited by 134 publications

References 9 publications

Controller performance assessment in set point tracking and regulatory control

Controller performance assessment in set point tracking and regulatory control

Introduction to Optimal and Robust Control

Limits of control performance for distributed networked control systems in presence of communication delays

Contact Info

Product

Resources

About