On robust estimation of the location parameter

This paper reviews the current state of the art on reinforcement learning (RL)-based feedback control solutions to optimal regulation and tracking of single and multiagent systems. Existing RL solutions to both optimal $\mathcal {H}_{2}$ and $\mathcal {H}_\infty $ control problems, as well as graphical games, will be reviewed. RL methods learn the solution to optimal control and game problems online and using measured data along the system trajectories. We discuss Q-learning and the integral RL algorithm as core algorithms for discrete-time (DT) and continuous-time (CT) systems, respectively. Moreover, we discuss a new direction of off-policy RL for both CT and DT systems. Finally, we review several applications.

Optimal and Autonomous Control Using Reinforcement Learning: A Survey

Recently, multilevel inverters (MLIs) have gained lots of interest in industry and academia, as they are changing into a viable technology for numerous applications, such as renewable power conversion system and drives. For these high power and high/medium voltage applications, MLIs are widely used as one of the advanced power converter topologies. To produce high-quality output without the need for a large number of switches, development of reduced switch MLI (RS MLI) topologies has been a major focus of current research. Therefore, this review paper focuses on a number of recently developed MLIs used in various applications. To assist with advanced current research in this field and in the selection of suitable inverter for various applications, significant understanding on these topologies is clearly summarized based on the three categories, i.e., symmetrical, asymmetrical, and modified topologies. This review paper also includes a comparison based on important performance parameters, detailed technical challenges, current focus, and future development trends. By a suitable combination of switches, the MLI produces a staircase output with low harmonic distortion. For a better understanding of the working principle, a single-phase RS MLI topology is experimentally illustrated for different level generation using both fundamental and high switching frequency techniques which will help the readers to gain the utmost knowledge for advance research.

/pdf/recently-developed-reduced-switch-multilevel-inverter-for-4lxrfejht6.pdf

Recently Developed Reduced Switch Multilevel Inverter for Renewable Energy Integration and Drives Application: Topologies, Comprehensive Analysis and Comparative Evaluation

This paper proposes a novel single-dc-source multilevel inverter called Packed E-Cell (PEC) topology to achieve nine levels with noticeably reduced components count, while dc capacitors are actively balanced. The nine-level PEC (PEC9) is composed of seven active switches and two dc capacitors that are shunted by a four-quadrant switch to from the E-cell, and it makes use of a single dc link. With the proper design of the corresponding PEC9 switching states, the dc capacitors are balanced using the redundant charging/discharging states. Since the shunted capacitors are horizontally extended, both capacitors are simultaneously charged or discharged with the redundant states, so only the auxiliary dc-link voltage needs to be sensed and regulated to half of the input dc source voltage, and consequently, dc capacitors' voltages are inherently balanced to one quarter of the dc bus voltage. To this end, an active capacitor voltage balancing integrated to the level-shifted half-parabola carrier PWM technique has been designed based on the redundant charging/discharging states to regulate the dc capacitors voltages of PEC9. Furthermore, using the E-cell not only reduces components count but also the proposed topology permits multi ac terminal operation. Thus, five-level inverter operation can be achieved during the four-quadrant switch fault, which confers to the structure high reliability. The theoretical analysis as well as the experimental results are presented and discussed, showing the basic operation, multi-functionality, as well as the superior performance of the proposed novel PEC9 inverter topology.

/pdf/packed-e-cell-pec-converter-topology-operation-and-39rdsn1m79.pdf

Packed E-Cell (PEC) converter topology operation and experimental validation

An efficient and cost-effective power converter is a pre-requisite for the modern power applications. With the evolvement of matured medium power self-commutated switching devices, multilevel inverters (MLIs) are emerged as a promising solution for high-power medium-voltage applications. Though, MLIs are performing a promising role in industrial applications, their high device count, size, cost and control complexities have restricted their market penetration. To address the disadvantages of MLIs, researchers are continuously contributing to new generation topologies under the name of reduced switch count (RSC) MLIs. From the past decade, numerous RSC-MLIs topologies have been reported for various applications. Therefore, this paper presents a comprehensive review and classification of RSC-MLI topologies, in terms of their structure, features, limitations, suitability and selection for specific applications.

/pdf/a-survey-on-reduced-switch-count-multilevel-inverters-1f1fm4shs8.pdf

A Survey on Reduced Switch Count Multilevel Inverters

A new cascaded multilevel inverter (MLI) is presented with the aim of utilising lesser number of switches, better modularity and reduced voltage stress. The new structure configured under symmetric and asymmetric mode, produces all odd and even voltage levels. This structure comprises semi-half-bridge cells connected in series with crisscross switches to generate any target level for synthesising the sinusoidal output voltage waveform. In extension to the proposed topology, subinverters derived from the proposed MLI are cascaded with an objective to produce more voltage levels with reduced standing voltage. Compared with the cascading H-bridge topology, the proposed MLI and the extended version uses lesser number of semi-conductor switches. The MATLAB R2013b-based simulation results along with the experimental results validate the proposed topology.

https://ietresearch.onlinelibrary.wiley.com/doi/pdf/10.1049/iet-pel.2016.0644

Crisscross switched multilevel inverter using cascaded semi-half-bridge cells

Nonlinear control of a boost converter using a robust regression based reinforcement learning algorithm

This paper presents a boost converter configuration, control scheme and design of single phase power factor controller for permanent magnet brushless DC motor (PMBLDCM) drive. PMBLDC motors are the latest choice of researchers, due to the high efficiency, silent operation, compact size, high reliability, and low maintenance requirements. The proposed Power Factor Controller topology improves power quality by improving performance of PMBLDCM drive, such as reduction of AC main current harmonics, near unity power factor. PFC converter forces the drive to draw sinusoidal supply current in phase with supply voltage. It uses a boost converter to obtain unity power factor with improved performance. The system includes a speed controller for PMBLDC drive and a voltage controller for boost converter‥The voltage or speed controllers can be realized using proportional integral (PI) controller. Simulations are done using MATLAB/ SIMULINK software.

PMBLDC motor drive with Power Factor correction controller

In process industries there is a high demand for design of multivariable control systems. The goal of this paper is to keep the liquid level in an Interacting Coupled Spherical Tank System (ICSTS) at a constant level. The system's mathematical model is based on fundamental principles such as mass balance and the energy conservation equation. The resulting mathematical model is linearized at an operational point for the design of conventional PI and fuzzy PI controllers. Ziegler–Nichol’s reaction curve tuning method is used to design the linear PI controller and the fuzzy PI controller parameters are obtained by optimizing the cost function using genetic algorithm. Simulation results acquired by conventional linear PI and fuzzy PI controllers are compared and contrasted.

Liquid Level Control of Interacting Coupled Spherical Tank System using PI and Fuzzy PI Controller

This study aims to design an improved stabilization controller for linear systems with time-varying delay. First, a proper augmented LKF has been constructed and then a tighter bound of the quadratic integral function in the LKF derivative is established by applying Wirtinger-based integral inequality (WBII) and extended reciprocally convex matrix inequality (RCMI). Congruence transformation has been performed to obtain the negative-definite condition for LKF derivative such that the closed-loop system becomes asymptotically stable in the sense of Lyapunov. Maximum permissible delay upper bound (MPDUB) and the corresponding control effort is computed using linear matrix inequality (LMI) convex optimization approach. A numerical example has been solved to verify the effectiveness of the proposed method using LMI control toolbox of MATLAB. 

N. Arun

Papers

Crisscross switched multilevel inverter using cascaded semi-half-bridge cells

Nonlinear control of a boost converter using a robust regression based reinforcement learning algorithm

PMBLDC motor drive with Power Factor correction controller

Liquid Level Control of Interacting Coupled Spherical Tank System using PI and Fuzzy PI Controller

An Improved Stabilization Criteria for Linear Systems with Time-Varying Delay Using a New Lyapunov–Krasovskii Functional