Model Predictive Control Research Articles

To solve the problems of the lack of economic efficiency and the short driving range of electric commercial vehicles, a hybrid system was developed in this work that uses fuel cells as a range extender. In addition, a method to solve the problem of multi-power energy management was proposed using the model predictive control as a framework. In the state of charge maintenance interval, a quadratic utility function was used to calculate the output power of the fuel cell and battery. The unknown parameters in the quadratic utility function were solved using the model prediction control. Speed prediction was performed using long short-term memory and particle swarm optimization. The demanded power sequence within the prediction horizon was calculated based on the predicted speed. The dynamic programming algorithm was used to solve the power demand sequence within the prediction horizon length, and the unknown parameters in the utility function were deduced inversely. The simulation results show that the proposed energy management strategy (EMS) is superior to conventional EMS in improving component durability and vehicle economy.

Abstract - The study showcases the latest ideas regarding architectural and control strategies that juxtapose Artificial Intelligence and Machine Learning with fuel cell technologies to increase the efficiency, range, and reliability of electric vehicles and advance toward the next generation of vehicles. A new hybrid framework is proposed that unifies physics-based models of fuel cells and batteries with data-driven predictive algorithms. Within this framework, a hierarchical Energy Management System (EMS) is designed in which a reinforcement learning layer signals power distribution set points, while a model predictive controller ensures safe operation of the system in real time! The proposed system, using simulation results within multiple driving scenarios, shows improvement in energy efficiency and hydrogen consumption as compared to traditional EMS along with an overall reduction in battery wear. The study considers computing challenges while running ML models in Battery Management Systems (BMS), predictive maintenance for life extension of components, and deployment challenges regarding hydrogen infrastructure and safety certification. The study ends with suggestions for pilot implementations and possible pathways for future enhancements. Keywords: Electric Vehicles, Fuel Cell, AI, Machine Learning, Energy Management System (EMS), Predictive Maintenance

Abstract The study proposes a torque motor control method based on the improved model predictive control algorithm, aiming to solve the core problems such as insufficient dynamic adaptability, weak anti-interference ability, and difficulty in multi-objective collaborative optimization in the traditional mechatronics optimization technology. Through the improved model predictive control algorithm in the torque motor control algorithm, the load is reasonably distributed and the status of mechatronics integration is monitored in real-time, thereby optimizing mechatronics integration. The results showed that the improved control algorithm performed significantly better than the other two algorithms in optimizing mechatronics technology. Under different external interference conditions, the improved algorithm had strong anti-interference ability and could optimize mechatronics technology without being affected by external interference factors. The optimization accuracy was always maintained within the range of 1.00–1.05. At 1.0032 min, the accuracy curve of the algorithm for optimizing mechatronics technology has already flattened. When comparing the optimization effects of the three algorithms, the improved algorithm maintained an optimization error value within the range of 0.00–0.01%, significantly lower than the optimization error values of other algorithms. Overall, the algorithm achieves high-precision real-time control and multi-objective collaborative optimization through rolling time-domain optimization and dynamic load allocation, providing theoretical support and technical paths for the efficient and stable operation of complex electromechanical systems.

This paper introduces the model predictive direct power control (MPDPC) method, which integrates a passive resistor–inductor–capacitor (RLC) filter and a parallel capacitor (C) filter to improve inverter performance in off-grid photovoltaic (PV) systems. The system features a two-level, three-phase inverter connected to a series RLC filter and a parallel capacitor filter, supplying an RL load. During the night and on cloudy days, when the system voltage level drops, the battery system compensates for this voltage drop and supplies the output load power. Since standalone PV systems cannot guarantee a stable energy supply without storage, a battery system is added in parallel with a boost converter to stabilize the direct current bus voltage and enhance system reliability. The proposed MPDPC strategy and the filter with a series resistor–inductance–capacitor and parallel capacitor structure (RLCǁC) ensure high power quality and system stability. Simulation results from MATLAB software verify the method's effectiveness, highlighting excellent performance in steady-state and dynamic conditions. The system's response to short-circuit faults is analyzed, demonstrating robust fault tolerance. The proposed control strategy also significantly reduces output current harmonics, achieving a total harmonic distortion of only 0.05%.

An integrated energy management strategy for plug-in hybrid electric buses based on receding horizon control and TD3 algorithm

Frequency-robust inductance identification in model predictive control via adaptive weighting

High-fidelity microreactor load follow simulations with model predictive control

Research on optimization strategy for steel strip temper rolling elongation based on model predictive control

A crash-injury-aware driving safety field for real-time risk assessment and its application in autonomous vehicle motion planning.

Development of condensation-free operation strategy for thermally activated building systems using model predictive control

Artificial intelligence in the non-clinical laboratory: enhancing good laboratory and documentation practices.

Learning-based real-time model predictive tracking control for autonomous vehicles with path-pattern adaptability

Event-triggered tube-based model predictive anti-rollover control for liquid tank trucks considering time-varying parameters

A Deep Learning Density Shaping Model Predictive Gust Load Alleviation Control of a Compliant Wing Subjected to Atmospheric Turbulence

A model predictive trajectory tracking control strategy for heavy-duty unmanned tracked vehicle using deep Koopman operator

Load Frequency Control of Multiarea Interconnected Power System Based on Distributed Economic Model Predictive Control With Guaranteed Stability

Melt viscosity control in polymer extrusion using nonlinear model predictive control with neural state space modelling and soft sensor feedback

A data-driven neural model predictive controller for multi-layer nonlinear vibration isolation system

An optimal model predictive control based on Hammerstein model considering fatigue load reduction for wind turbines

Distributed nonlinear model predictive control of an array of wave energy converters