pith. sign in

arxiv: 2605.01090 · v2 · submitted 2026-05-01 · 📡 eess.SY · cs.SY

Sampled-data Robust Control of Electrically Stimulated Engineered Cell Factories

Papri Dey , Ksenia Zoblina , Nicholas A. Rondoni , Marcella M. Gomez This is my paper

Pith reviewed 2026-05-09 18:26 UTC · model grok-4.3

classification 📡 eess.SY cs.SY
keywords engineered cell factoriesbioelectronic controlsampled-data controladaptive PID controllerrobust controlthyroid hormone T4input-to-state stabilityintracellular delay
0
0 comments X

The pith

A sampled-data robust adaptive PID controller achieves stable regulation of T4 hormone production in electrically stimulated engineered cells despite delays and uncertainties.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a framework for closed-loop control of hormone secretion in engineered cells that respond indirectly to electric fields. It combines a reduced mechanistic model of the T4 pathway with an EF-responsive Hill function and an Erlang cascade for intracellular delays, then designs an adaptive PID controller with filtering, anti-windup, and limits, plus a robust extension using scenario-based updates. Under local Lyapunov conditions the sampled tracking error stays ultimately bounded by a disturbance-dependent constant. Simulations show the approach maintains target T4 levels even with noise, bias, jitter, and rhythmic disturbances.

Core claim

We design a sampled-data adaptive proportional-integral-derivative (PID) controller with derivative filtering, anti-windup, saturation and rate limits, and hysteretic band-locking, together with a robust adaptive extension that accounts for parameter mismatch, sensor noise and bias, actuator mismatch, delay/jitter, and exogenous rhythmic disturbance through a scenario-based risk-aware update. We provide local sampled-data input-to-state stability interpretations for both APID and RAPID, showing that, under standard local Lyapunov and bounded-disturbance conditions, the sampled tracking error is ultimately bounded by a disturbance-dependent constant. In silico experiments demonstrate Sustaine

What carries the argument

The sampled-data adaptive PID controller and its robust extension (RAPID) that uses scenario-based risk-aware parameter updates to enforce local input-to-state stability.

If this is right

  • The sampled tracking error remains ultimately bounded by a constant determined by the size of bounded disturbances.
  • Sustained regulation of extracellular T4 can be maintained at multiple prescribed target levels.
  • The controller handles the combination of delayed nonlinear dynamics, sparse measurements, and burst-constrained electric actuation.
  • Stability holds locally whenever standard Lyapunov conditions are met and disturbances remain bounded.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The modeling and control structure could be applied to regulation of other secreted proteins in different engineered cell types.
  • Physical wet-lab tests would be needed to check whether the ODE model's delay distribution matches observed cell behavior.
  • The risk-aware update rule suggests a route for incorporating online learning to refine disturbance bounds during operation.

Load-bearing premise

The control-oriented ODE model of the T4 pathway, Hill module, and Erlang cascade sufficiently captures the real intracellular dynamics of the engineered cells.

What would settle it

If laboratory experiments on the actual engineered thyroid-like cells show that tracking error grows without bound under the same disturbance levels used in the simulations, the local stability claim would be falsified.

Figures

Figures reproduced from arXiv: 2605.01090 by Ksenia Zoblina, Marcella M. Gomez, Nicholas A. Rondoni, Papri Dey.

Figure 1
Figure 1. Figure 1: System-level view of the proposed closed-loop thyroid-hormone regulation platform. view at source ↗
Figure 2
Figure 2. Figure 2: Schematic of T4 production pathways in the engineered cell culture. Tg – thyroglob￾ulin, I – iodide, I0 – iodine, N – Na/I transporter. P – pendrin, X – thyroid peroxidase, Ti – iodinated thyroglobulin, S – conjugated complex, T4 – thyroxine, T3 – triiodothyronine, D2 – deiodinase 2, M – MCT8 transporter. evant to extracellular hormone release. We then complement this mechanistic structure with data-inform… view at source ↗
Figure 3
Figure 3. Figure 3: T4 production pathway in engineered cell culture: schematic of the reduced model. 2.3 Effect of Electric Field on T4 production Model In the reduced model, T4 production is represented as a single-step reaction between thyroglobu￾lin and iodide. While this simplification captures the overall output, we know that in reality, the process involves multiple sequential stages, each regulated by various factors,… view at source ↗
Figure 4
Figure 4. Figure 4: EF-driven T4 production cascade in engineered thyroid cells. Electric field (EF) affects the synthetic EF-responsive promoter (uf ) and the multi-stage gene expression cascade (u1, . . . , u10) leading to effective transcription factor activity (uNK). It modulates the kinetic parameter (k1) and ultimately controls T4 synthesis. This cumulative cascade introduces a natural total delay in the system, often r… view at source ↗
Figure 6
Figure 6. Figure 6: Additionally, view at source ↗
Figure 5
Figure 5. Figure 5: Schematic of the amplitude map Am versus the controller command um. The two parameter orderings Amin < thr < Amax and thr < Amin < Amax are qualitatively similar, though not exactly identical: in both cases, the map is zero below thr, then rises with slope kA, and finally saturates at Amax. Windowwise controller command um Amplitude map um 7→ Am Signal fed into ODE model Real actuator waveform EF Window 1 … view at source ↗
Figure 6
Figure 6. Figure 6: Hardware-to-model actuator mapping. A windowwise command view at source ↗
Figure 7
Figure 7. Figure 7: Burst-averaged EF supplied to the plant over a zoomed time interval. Narrow pulse view at source ↗
Figure 8
Figure 8. Figure 8: Open-loop trajectories of T ext 4 (t) under fixed EF amplitudes A ∈ {0, 0.005, 0.01, 0.015, 0.02}. Increasing amplitude shifts the response to higher oscillatory regimes, while also changing the transient rise and oscillation magnitude, highlighting the sen￾sitivity of the output to EF amplitude. toward the constant target T des 4 . The plant evolves continuously, while the controller updates only at the w… view at source ↗
Figure 9
Figure 9. Figure 9: High-level schematic of the APID controller. The feedback loop generates the EF view at source ↗
Figure 10
Figure 10. Figure 10: Closed-loop performance of APID with band-lock holding for view at source ↗
Figure 11
Figure 11. Figure 11: Closed-loop trajectories for multiple setpoints view at source ↗
Figure 12
Figure 12. Figure 12: Comparison of a sampled-data PID baseline with fixed gains and the adaptive PID view at source ↗
Figure 13
Figure 13. Figure 13: Comparison of measured T ext 4 trajectories for adaptive PID control with and without the band-lock mechanism at T des 4 = 25 (a.u.). The shaded region denotes the ±5% tolerance band around the target. The band-lock controller exhibits improved near-setpoint retention, spends more time inside the tolerance band, and achieves a final measured T4 value closer to the target view at source ↗
Figure 14
Figure 14. Figure 14: Closed-loop performance of RAPID under five simultaneous perturbation classes for view at source ↗
Figure 15
Figure 15. Figure 15: Measured-output comparison of APID and RAPID for view at source ↗
Figure 16
Figure 16. Figure 16: Time-resolution comparison for dt = 0.1 s and dt = 0.008333 s. Top: measured T ext 4 trajectories. Bottom: burst-averaged EF signals supplied to the plant. 7 Supplementary Materials This section collects additional numerical checks supporting the main results. We report the actuator time-resolution study, discuss the measurement/update interval, and provide a RAPID setpoint sweep. These studies document i… view at source ↗
Figure 17
Figure 17. Figure 17: Sweep of the RAPID controller over multiple target setpoints view at source ↗
read the original abstract

Closed-loop bioelectronic regulation of engineered secretory cell systems is challenging because electric-field (EF) stimulation acts indirectly through transcription-factor activation, in the presence of delayed, nonlinear, and noisy intracellular dynamics, sparse measurements, and constrained burst-based actuation. We develop a framework for robust closed-loop endocrine regulation in electrically stimulated engineered cell factories, illustrated through extracellular thyroid hormone \(T_4\) production in engineered thyroid-like cells. The plant is modeled by a control-oriented ODE formulation combining a reduced mechanistic \(T_4\) pathway, an EF-responsive Hill module, and a linear-chain Erlang cascade representing distributed intracellular delay. On this basis, we design a sampled-data adaptive proportional-integral-derivative (PID) controller with derivative filtering, anti-windup, saturation and rate limits, and hysteretic band-locking, together with a robust adaptive extension that accounts for parameter mismatch, sensor noise and bias, actuator mismatch, delay/jitter, and exogenous rhythmic disturbance through a scenario-based risk-aware update. We provide local sampled-data input-to-state stability interpretations for both APID and RAPID, showing that, under standard local Lyapunov and bounded-disturbance conditions, the sampled tracking error is ultimately bounded by a disturbance-dependent constant. In silico experiments demonstrate sustained regulation of extracellular \(T_4\) across prescribed targets despite significant uncertainty.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 3 minor

Summary. The manuscript develops a sampled-data robust control framework for electrically stimulated engineered cell factories, illustrated on extracellular T4 regulation in engineered thyroid-like cells. The plant is modeled via a control-oriented ODE combining a reduced mechanistic T4 pathway, an EF-responsive Hill module, and a linear-chain Erlang cascade for distributed intracellular delay. An adaptive PID controller (APID) is designed with derivative filtering, anti-windup, saturation, rate limits, and hysteretic band-locking; a robust adaptive extension (RAPID) further accounts for parameter mismatch, sensor noise/bias, actuator mismatch, delay/jitter, and rhythmic disturbance via scenario-based risk-aware updates. Local sampled-data input-to-state stability interpretations are provided for both controllers under standard local Lyapunov and bounded-disturbance conditions, claiming that the sampled tracking error is ultimately bounded by a disturbance-dependent constant. In silico experiments demonstrate sustained regulation across targets despite significant uncertainty.

Significance. If the local ISS interpretations and simulation results hold, the work would offer a concrete control-theoretic contribution to bioelectronic regulation of synthetic secretory systems, addressing indirect actuation, delays, noise, and sparse measurements through practical sampled-data adaptive designs. Strengths include the explicit incorporation of implementation constraints (saturation, anti-windup) and the scenario-based robustness mechanism. The local Lyapunov-based ISS analysis provides a standard but useful stability lens. Significance is limited by the purely in silico nature of the validation and the absence of checks confirming that trajectories remain inside the local domain invoked by the stability claims.

major comments (1)
  1. [Abstract / stability analysis] Abstract and stability analysis section: the local sampled-data ISS claim states that 'under standard local Lyapunov and bounded-disturbance conditions, the sampled tracking error is ultimately bounded by a disturbance-dependent constant.' The in silico experiments (with parameter mismatch, sensor noise, actuator mismatch, delay/jitter, and rhythmic disturbance) provide no post-simulation verification that closed-loop trajectories remained inside the region of validity of the local Lyapunov function for either APID or RAPID. If trajectories exit this domain, the ISS bound does not apply and the sustained-regulation result rests only on the specific numerical runs rather than the stated stability interpretation.
minor comments (3)
  1. [In silico experiments] Simulation results lack error bars or statistical summaries across repeated runs with randomized disturbances.
  2. [Methods / results] Full numerical parameter values for the plant model, controller gains, and scenario-based update are not provided in a table or supplementary material, hindering reproducibility.
  3. [Plant modeling] The model is not validated against any real experimental data; the transfer of the stability interpretation therefore depends entirely on the untested assumption that the reduced ODE captures the dominant intracellular dynamics.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading and constructive feedback on our manuscript. We address the major comment below and will revise the manuscript accordingly to strengthen the connection between the theoretical claims and the numerical results.

read point-by-point responses

  1. Referee: [Abstract / stability analysis] Abstract and stability analysis section: the local sampled-data ISS claim states that 'under standard local Lyapunov and bounded-disturbance conditions, the sampled tracking error is ultimately bounded by a disturbance-dependent constant.' The in silico experiments (with parameter mismatch, sensor noise, actuator mismatch, delay/jitter, and rhythmic disturbance) provide no post-simulation verification that closed-loop trajectories remained inside the region of validity of the local Lyapunov function for either APID or RAPID. If trajectories exit this domain, the ISS bound does not apply and the sustained-regulation result rests only on the specific numerical runs rather than the stated stability interpretation.

    Authors: We agree that the absence of explicit post-simulation verification of the local domain represents a gap in substantiating the local ISS claims. The manuscript currently presents the stability interpretation under the stated local Lyapunov and bounded-disturbance conditions but does not confirm that the simulated trajectories remained inside the relevant region for either controller. In the revised manuscript we will add a dedicated verification subsection in the simulation results. This will include an estimate of the local region (derived from the Lyapunov function level sets or state bounds used in the analysis) and explicit checks confirming that all reported trajectories, across the tested uncertainty scenarios, satisfy the local conditions. These checks will be presented alongside the existing performance plots. We believe this addition directly addresses the concern and allows the stability interpretation to apply to the numerical evidence. revision: yes

Circularity Check

0 steps flagged

No circularity detected; stability claims rest on external standard Lyapunov theory applied to the given model

full rationale

The paper's central stability result invokes standard local Lyapunov and bounded-disturbance conditions to conclude ultimate boundedness of the sampled tracking error for the APID and RAPID controllers. This is an application of established input-to-state stability theory rather than a reduction of any prediction or bound to quantities defined by the same data fits or self-citations. The model itself is presented as a control-oriented ODE formulation whose parameters are not claimed to be derived from the stability analysis. In silico experiments are numerical demonstrations under uncertainty, not predictions forced by construction from the stability derivation. No self-definitional loops, fitted inputs renamed as predictions, or load-bearing self-citations appear in the provided derivation chain. The result is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the control-oriented ODE model being adequate and on standard Lyapunov-based stability results from control theory; no new entities are postulated.

axioms (1)
  • standard math standard local Lyapunov and bounded-disturbance conditions suffice for input-to-state stability
    Invoked directly for the sampled tracking error bound in the stability interpretation.

pith-pipeline@v0.9.0 · 5551 in / 1317 out tokens · 21029 ms · 2026-05-09T18:26:33.162925+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages

  1. [1]

    Alon.An Introduction to Systems Biology: Design Principles of Biological Circuits

    U. Alon.An Introduction to Systems Biology: Design Principles of Biological Circuits. Chapman & Hall/CRC, 2006

    work page 2006

  2. [2]

    Carmona-Hidalgo, E

    B. Carmona-Hidalgo, E. Herrera-Ramos, R. Rodr´ ıguez-L´ opez, L. Nou-Fontanet, J. C. Moreno, J. A. Blasco-Amaro, J. L´ eger, J. D. Ortigoza-Escobar, and NKX2-1-Related Dis- orders Guideline Working Group. Systematic review of thyroid function in nkx2-1-related disorders: Treatment and follow-up.PLoS ONE, 19(10):e0309064, 2024. 29

    work page 2024

  3. [3]

    Del Vecchio and R

    D. Del Vecchio and R. M. Murray.Biomolecular Feedback Systems. Princeton University Press, 2018

    work page 2018

  4. [4]

    K. Famm, B. Litt, K. J. Tracey, E. S. Boyden, and M. Slaoui. A jump-start for electroceu- ticals.Nature, 496:159–161, 2013

    work page 2013

  5. [5]

    Gomez, Mehdi Sadeghpour, Matthew R

    Marcella M. Gomez, Mehdi Sadeghpour, Matthew R. Bennett, G´ abor Orosz, and Richard M. Murray. Stability of systems with stochastic delays and applications to genetic regulatory networks.SIAM Journal on Applied Dynamical Systems, 15(4):1844–1873, 2016

    work page 2016

  6. [6]

    Hurtado and Adam S

    Paul J. Hurtado and Adam S. Kirosingh. Generalizations of the Linear Chain Trick: incorporating more flexible dwell time distributions into mean field ODE models.Journal of Mathematical Biology, 79(5):1831–1883, 2019

    work page 2019

  7. [7]

    A feedback control architecture for bioelectronic devices with applications to wound healing.Journal of the Royal Society Interface, 18(185):20210497, 2021

    Bashir Hosseini Jafari, Ksenia Zlobina, Giovanny Marquez, Mohammad Jafari, John Sel- berg, Manping Jia, Marco Rolandi, and Marcella Gomez. A feedback control architecture for bioelectronic devices with applications to wound healing.Journal of the Royal Society Interface, 18(185):20210497, 2021

    work page 2021

  8. [8]

    Merging machine learning and bioelectronics for closed-loop control of biological systems and homeostasis.Cell Reports Physical Science, 4(8):101535, 2023

    Mohammad Jafari, Giovanny Marquez, Harika Dechiraju, Marcella Gomez, and Marco Rolandi. Merging machine learning and bioelectronics for closed-loop control of biological systems and homeostasis.Cell Reports Physical Science, 4(8):101535, 2023

    work page 2023

  9. [9]

    Input-to-state stability for discrete-time nonlinear systems.Automatica, 37(6):857–869, 2001

    Zhong-Ping Jiang and Yuan Wang. Input-to-state stability for discrete-time nonlinear systems.Automatica, 37(6):857–869, 2001

    work page 2001

  10. [10]

    A converse lyapunov theorem for discrete-time systems with disturbances.Systems & Control Letters, 45(1):49–58, 2002

    Zhong-Ping Jiang and Yuan Wang. A converse lyapunov theorem for discrete-time systems with disturbances.Systems & Control Letters, 45(1):49–58, 2002

    work page 2002

  11. [11]

    Koutsouras, George G

    Dimitrios A. Koutsouras, George G. Malliaras, and Geert Langereis. The rise of bioelec- tronic medicine.Bioelectronic Medicine, 10(1):19, 2024

    work page 2024

  12. [12]

    Kusakabe, A

    T. Kusakabe, A. Kawaguchi, N. Hoshi, R. Kawaguchi, S. Hoshi, and S. Kimura. Thyroid- specific enhancer-binding protein/nkx2.1 is required for the maintenance of ordered archi- tecture and function of the differentiated thyroid.Molecular Endocrinology, 20(8):1796– 1809, 2006

    work page 2006

  13. [13]

    Endogenous bioelectrics in development, cancer, and regeneration: Drugs and bioelectronic devices as electroceuticals for regenera- tive medicine.iScience, 22:519–533, Dec 2019

    Michael Levin, John Selberg, and Marco Rolandi. Endogenous bioelectrics in development, cancer, and regeneration: Drugs and bioelectronic devices as electroceuticals for regenera- tive medicine.iScience, 22:519–533, Dec 2019

    work page 2019

  14. [14]

    X. Li, S. Zhong, and J. Morizio. 16-channel biphasic current-mode programmable charge balanced neural stimulation.BioMedical Engineering OnLine, 16:104, 2017

    work page 2017

  15. [15]

    Liang, J

    J. Liang, J. Qian, L. Yang, X. Chen, X. Wang, X. Lin, X. Wang, and B. Zhao. Modeling human thyroid development by fetal tissue-derived organoid culture.Advanced Science, 9(9):e2105568, 2022

    work page 2022

  16. [16]

    Sampled-data semi-global robust output regulation for a class of nonlinear systems.Journal of Systems Science and Complexity, 34:1743–1765, 2021

    Wei Liu and Jie Huang. Sampled-data semi-global robust output regulation for a class of nonlinear systems.Journal of Systems Science and Complexity, 34:1743–1765, 2021

    work page 2021

  17. [17]

    Springer, Berlin, Heidelberg, 1978

    Norman MacDonald.Time Lags in Biological Models, volume 27 ofLecture Notes in Biomathematics. Springer, Berlin, Heidelberg, 1978

    work page 1978

  18. [18]

    Mansouri and M

    M. Mansouri and M. Fussenegger. Electrogenetics: Bridging synthetic biology and elec- tronics to remotely control the behavior of mammalian designer cells.Current Opinion in Chemical Biology, 68:102151, 2022. 30

    work page 2022

  19. [19]

    Hernandez, Mircea Teodorescu, Marco Rolandi, Min Zhao, and Marcella Gomez

    Giovanny Marquez, Mohammad Jafari, Manasa Kesapragada, Kan Zhu, Prabhat Baniya, Yao-Hui Sun, Hao-Chieh Hsieh, Cristian O. Hernandez, Mircea Teodorescu, Marco Rolandi, Min Zhao, and Marcella Gomez. Controlling cell migratory patterns under an electric field regulated by a neural network-based feedback controller.Bioengineering, 12(7):678, 2025

    work page 2025

  20. [20]

    Dragan Neˇ si´ c and Andrew R. Teel. A framework for stabilization of nonlinear sampled- data systems based on their approximate discrete-time models.IEEE Transactions on Automatic Control, 49(7):1103–1122, 2004

    work page 2004

  21. [21]

    Teel, and Daniele Carnevale

    Dragan Neˇ si´ c, Andrew R. Teel, and Daniele Carnevale. Explicit computation of the sam- pling period in emulation of controllers for nonlinear sampled-data systems.IEEE Trans- actions on Automatic Control, 54(3):619–624, 2009

    work page 2009

  22. [22]

    V. M. L. Ogundipe, A. H. Groen, N. Hosper, P. W. K. Nagle, J. Hess, H. Faber, A. L. Jellema, M. Baanstra, T. P. Links, K. Unger, J. T. M. Plukker, and R. P. Coppes. Gen- eration and differentiation of adult tissue-derived human thyroid organoids.Stem Cell Reports, 16(4):913–925, 2021

    work page 2021

  23. [23]

    P. S. Olofsson and K. J. Tracey. Bioelectronic medicine: technology targeting molecular mechanisms for therapy.Journal of Internal Medicine, 282:3–4, 2015

    work page 2015

  24. [24]

    Electrical stimulation: a novel therapeutic strategy to heal biological wounds.RSC Advances, 14(44):32142–32173, Oct 2024

    Subham Preetam, Arka Ghosh, Richa Mishra, Arunima Pandey, Debanjan Singha Roy, Sarvesh Rustagi, and Sumira Malik. Electrical stimulation: a novel therapeutic strategy to heal biological wounds.RSC Advances, 14(44):32142–32173, Oct 2024

    work page 2024

  25. [25]

    J. E. Purvis and G. Lahav. Encoding and decoding cellular information through signaling dynamics.Cell, 152:945–956, 2013

    work page 2013

  26. [26]

    Qian and D

    Y. Qian and D. Del Vecchio. Realizing ‘integral control’ in living cells: how to overcome leaky integration due to dilution?Journal of The Royal Society Interface, 15, 2018

    work page 2018

  27. [27]

    Machine learning-driven bioelectronics for closed-loop control of cells

    John Selberg, Mohammad Jafari, Juanita Mathews, Manping Jia, Pattawong Pansodtee, Harika Dechiraju, Chunxiao Wu, Sergio Cordero, Alexander Flora, Nebyu Yonas, Sophia Jannetty, Miranda Diberardinis, Mircea Teodorescu, Michael Levin, Marcella Gomez, and Marco Rolandi. Machine learning-driven bioelectronics for closed-loop control of cells. Advanced Intellig...

    work page 2020

  28. [28]

    C. R. Steinhardt, D. E. Mitchell, K. E. Cullen, et al. Pulsatile electrical stimulation creates predictable, correctable disruptions in neural firing.Nature Communications, 15:5861, 2024

    work page 2024

  29. [29]

    Springer, London, 2006

    Antonio Visioli.Practical PID Control. Springer, London, 2006

    work page 2006

  30. [30]

    The role of machine learning in advancing precision medicine with feedback control.Cell Reports Physical Science, 3(11):101149, 2022

    Ksenia Zlobina, Mohammad Jafari, Marco Rolandi, and Marcella Gomez. The role of machine learning in advancing precision medicine with feedback control.Cell Reports Physical Science, 3(11):101149, 2022

    work page 2022

  31. [31]

    K. J. ˚Astr¨ om and T. H¨ agglund.Advanced PID Control. ISA, 2006

    work page 2006

  32. [32]

    K. J. ˚Astr¨ om and B. Wittenmark.Computer-Controlled Systems: Theory and Design. Prentice Hall, 2010

    work page 2010

  33. [33]

    ˚Astr¨ om and Tore H¨ agglund

    Karl J. ˚Astr¨ om and Tore H¨ agglund. The future of PID control.Control Engineering Practice, 9(11):1163–1175, 2001. 31 0 10 20 30 40 50 60 70 80Time (h) 0 10 20 30Measured T4ext (a.u.) Measured tracking comparison Measured T4, dt = 0.1 s Measured T4, dt = 0.008333 sSetpoint 0 10 20 30 40 50 60 70 80Time (h) 0 0.002 0.004 0.006 0.008 0.01EF supplied Burs...

    work page 2001