Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Grouped-neural network modeling for model predictive control.

Jing Ou1, R Russell Rhinehart

  • 1School of Chemical Engineering, Oklahoma State University, Stillwater 74078-5021, USA.

ISA Transactions
|June 20, 2002
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Genome-wide analysis identifies LliNAC80 as a potential key regulator of abiotic stress resilience in endangered mangrove Lumnitzera littorea.

BMC plant biology·2026
Same author

Preparation and application of a novel dendrimer modified silica reversed phase/hydrophilic interaction mixed-mode stationary phase.

Mikrochimica acta·2026
Same author

Low-Speed Permanent Magnet Synchronous Motor Rotor Position Estimation Using Structural Vibration Modal Phase Carrier.

Sensors (Basel, Switzerland)·2026
Same author

A novel model based on apparent diffusion coefficient and signal intensity on MRI to quantify subtle internal difference between tumor-adjacent and -distant liver tissues in T<sub>3</sub>-staged resectable gallbladder carcinoma.

BMC medical imaging·2026
Same author

Baseline dynamic contrast-enhanced magnetic resonance imaging derived parameters and magnetic resonance signal intensity based logistic regression model to predict response of esophageal squamous cell carcinoma to neoadjuvant immunochemotherapy.

Diseases of the esophagus : official journal of the International Society for Diseases of the Esophagus·2026
Same author

Chromosome-scale genome of Woonyoungia septentrionalis provides insights into sex differentiation and genomic evolution in Magnoliaceae.

Genomics·2025
Same journal

Hybrid vehicle state estimation using closed-form liquid neural networks and nonlinear Kalman filtering.

ISA transactions·2026
Same journal

Cross-coupled synchronization control strategy for rebar binding robots based on impedance control.

ISA transactions·2026
Same journal

Gas flow tracking for electronic pressure control system in gas chromatography under state constraints and hysteresis：An innovative fuzzy adaptive control approach.

ISA transactions·2026
Same journal

Stackelberg differential game-based fuzzy adaptive hierarchical optimal control for a nonlinear system with unknown dynamics.

ISA transactions·2026
Same journal

Composite fault-tolerant predictive control strategy for PMSM demagnetization faults.

ISA transactions·2026
Same journal

Bias-compensated Q-learning for optimal tracking control under denial-of-service attacks.

ISA transactions·2026
See all related articles

This study introduces a novel model-based predictive control (MPC) scheme using parallel neural networks (NNs) for dynamic predictions. This approach reduces training complexity and eliminates error propagation, enhancing control system performance.

Area of Science:

  • Control Engineering
  • Artificial Intelligence
  • Nonlinear Systems

Background:

  • Model-based predictive control (MPC) is crucial for complex dynamic systems.
  • Traditional MPC often relies on single, complex models, leading to high computational load and potential error propagation.
  • Neural networks (NNs) offer powerful modeling capabilities but integrating them into MPC presents challenges.

Purpose of the Study:

  • To develop a novel nonlinear MPC scheme utilizing a group of parallel feed-forward neural networks (NNs) as the dynamic prediction model.
  • To address the challenges of training complexity and error propagation inherent in cascaded NN models.
  • To introduce a new strategy for compensating process-model mismatch within the grouped-NN framework.

Main Methods:

Related Experiment Videos

  • A group of independent, parallel feed-forward neural networks (NNs) was employed for dynamic process output prediction.
  • The parallel structure facilitated independent training and implementation, reducing overall complexity.
  • A novel compensation strategy was developed to manage process-model mismatch specific to the grouped-NN architecture.

    • Main Results:

    • The parallel NN structure significantly decreased training complexity and eliminated compounded error propagation.
    • The developed compensation strategy effectively addressed process-model mismatch.
    • Simulation results demonstrated the scheme's effectiveness as a general nonlinear MPC.

    Conclusions:

    • The proposed grouped-NN model structure offers an efficient and robust approach for nonlinear MPC.
    • This parallel NN strategy mitigates key limitations of traditional MPC and cascaded NN integration.
    • The method shows promise for advanced control applications requiring accurate dynamic predictions.