Embedded PWM Predictive Control of DC-DC Power Converters Via Piecewise-Affine Neural Networks

Abstract

Predictive control is a flexible control methodology that can optimize performance while satisfying current and voltage constraints. Its application in the power electronics domain is however hampered by the high computational demands associated with it. In this paper, piecewise-affine neural networks are explored to greatly simplify these controllers and allow for an inexpensive implementation in commercial hardware. More specifically, we tackle the problem of enhancing the start-up transient response of a step-down dc-dc converter while also satisfying inductor current constraints. We analyze the neural network architecture, and detail its training and validation procedures. The learned controller is then embedded on an inexpensive 80-MHz microcontroller, and experimental results are provided showing that the whole control algorithm can be executed in under 30 microseconds.INDEX TERMS Model predictive control, embedded deployment, neural networks, dc-dc converters.

Application of Artificial Neural Networks in Chemical Process Control

“…It has been proved that artificial neural networks have played a significant role in the modeling and control of chemical processes, showing superior performance to the traditional control techniques for chemical processes. In particular, methods such as using stacked neural networks [33] and hybrid models to refine and complement neural networks may play a greater role in chemical processes. However, the structures of neural networks applied in chemical processes are relatively simple and still have great potential for development.…”

“…With the significant success of deep neural networks for tasks such as speech recognition and image classification [28,29], as well as the rapid increase in computer computing power and the accumulation of large-scale data, ANN has ushered in another climax of development and has a wide range of applications. The main application areas are intelligent driving [30,31], automatic control of power systems [32,33], signal processing [28,34], health care and medical treatment [35,36], process control and optimization [37][38][39], image processing [40,41]…”

Simplified Predictive Direct Power Control of Three-Phase Three-Level Four-Leg Grid Connected NPC Converter

“…As given in (15), the reactive power , g q includes the zero-sequence component of the line currents, therefore, its reference value * , g q must be forced to zero to remove the zero-sequence component of the line currents. Reactive power * g q can be set to zero to guarantee the unity power factor operation at the grid side.…”

Review of Artificial Neural Network and Its Application Research in Distillation

“…The unique structure and information processing method of ANN make it have obvious advantages in many aspects and have a wide range of applications. The main application areas are image processing [7][8][9][10], robot control, automatic control of power systems [11][12][13], signal processing [14][15][16], intelligent driving [17,18], health care and medical treatment [19][20][21], game theory [22,23], process control and optimization [24 -27], etc.…”