Model predictive control for systems with fast dynamics using inverse neural models

Stogiannos, Marios; Alexandridis, Alex; Sarimveis, Haralambos

doi:10.1016/j.isatra.2017.09.016

Cited by 40 publications

(15 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a result, a quadratic optimisation task is obtained in place of a nonlinear one. (b) A neural approximator may be used to find the initial solution of the MPC optimisation problem, which speeds up calculations [58,59]. (c) Neural networks are able to approximate the MPC control law [60][61][62].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Computationally Efficient Nonlinear Model Predictive Control Using the L1 Cost-Function

Ławryńczuk

Nebeluk

2021

Sensors

View full text Add to dashboard Cite

Model Predictive Control (MPC) algorithms typically use the classical L2 cost function, which minimises squared differences of predicted control errors. Such an approach has good numerical properties, but the L1 norm that measures absolute values of the control errors gives better control quality. If a nonlinear model is used for prediction, the L1 norm leads to a difficult, nonlinear, possibly non-differentiable cost function. A computationally efficient alternative is discussed in this work. The solution used consists of two concepts: (a) a neural approximator is used in place of the non-differentiable absolute value function; (b) an advanced trajectory linearisation is performed on-line. As a result, an easy-to-solve quadratic optimisation task is obtained in place of the nonlinear one. Advantages of the presented solution are discussed for a simulated neutralisation benchmark. It is shown that the obtained trajectories are very similar, practically the same, as those possible in the reference scheme with nonlinear optimisation. Furthermore, the L1 norm even gives better performance than the classical L2 one in terms of the classical control performance indicator that measures squared control errors.

show abstract

Section: Introductionmentioning

confidence: 99%

“…A neural approximator may be used to find the initial solution of the MPC optimisation problem, which speeds up calculations [ 58 , 59 ].…”

Section: Introductionmentioning

confidence: 99%

Computationally Efficient Nonlinear Model Predictive Control Using the L1 Cost-Function

Ławryńczuk

Nebeluk

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…The advantage of replacing the static control with a machine learning based control, such as an ANN, is the possibility of learning continuously, online, and adapting according to uncertainties or system variations, allowing the design of adaptive controllers [2], [30]. The direct and inverse system identification, by ANN, allow control strategies based on models, such as observed-based control, internal model control (IMC), model predictive control (MPC), model reference control (MRC), among others [2], [11], [33], [34].…”

Section: B Ann In Identification For Controlmentioning

confidence: 99%

“…Noise, disturbances, and uncertainties are inherent aspects of real-world problems description [9], not usually considered by the state estimation methods for nonlinear control techniques [10]. The ANN attractiveness in identification and control derives from its intrinsic ability to use experimental data to model unknown systems, even with perturbations or uncertainties, enabling a notable alternative in situations where conventional methods could fail in determining the appropriate control laws [11].…”

mentioning

confidence: 99%

Rotary Inverted Pendulum Identification for Control by Paraconsistent Neural Network

et al. 2021

View full text Add to dashboard Cite

Artificial neural networks (ANNs) have been used over the last few decades to perform tasks by learning with comparisons. Fitting input-output models, system identification, control, and pattern recognition are some fields for ANN applications. However, problems involving uncertain situations could be challenging for them. The family of paraconsistent logics (PL) is a powerful tool that can deal with uncertainty and contradictory information, so getting attention from researchers for its implications and applications in artificial intelligence. This investigation describes a novel activation function reasoned on the paraconsistent annotated logic by two-value annotations (PAL2v) rules, a variation of PL, allowing the design of a new paraconsistent neural net (PNN), applied in model identification for control (I4C) of a closed-loop rotary inverted pendulum (RIP) system.

show abstract

“…The RBF neural network can calculate future prediction value faster and improve the operation speed of MPC. Many scholars have used RBF neural networks in MPC and good control effects have been achieved [25], [26].…”

Section: Introductionmentioning

confidence: 99%

Multi-Step Predictive Current Control for NPC Grid-Connected Inverter

Wang

2019

IEEE Access

View full text Add to dashboard Cite

Aiming at improving the quality of output current when the inverter is connected to the grid, this paper proposes a control strategy of neutral point clamped (NPC) grid-connected inverter based on radial basis function neural network (RBFNN) multi-step predictive control. Firstly, the predictive control model for output current of NPC inverter is constructed by αβ coordinate transformation. Then the future values of output current are predicted for each voltage vectors, using the RBFNN prediction model. After the predicted values are obtained, a cost function f is calculated for each voltage vectors and the optimal voltage vector which minimizes cost function is selected. The switch state corresponding to the optimal voltage vector is applied to the inverter in the next sampling period. Thereby the control of grid-connected output current can be achieved. The simulation results show that the RBFNN multi-step predictive control method improves the capability of current tracking and reduces the harmonic rate of output current. The proposed method can balance output current under the condition of grid voltage imbalance. INDEX TERMS Inverter, RBF neural network, predictive control, optimal voltage vector, three-phase current balance.

show abstract

Model predictive control for systems with fast dynamics using inverse neural models

Cited by 40 publications

References 53 publications

Computationally Efficient Nonlinear Model Predictive Control Using the L1 Cost-Function

Computationally Efficient Nonlinear Model Predictive Control Using the L1 Cost-Function

Rotary Inverted Pendulum Identification for Control by Paraconsistent Neural Network

Multi-Step Predictive Current Control for NPC Grid-Connected Inverter

Contact Info

Product

Resources

About