There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

This article gives an overview of the artificial intelligence (AI) applications for power electronic systems. The three distinctive life-cycle phases, design, control, and maintenance are correlated with one or more tasks to be addressed by AI, including optimization, classification, regression, and data structure exploration. The applications of four categories of AI are discussed, which are expert system, fuzzy logic, metaheuristic method, and machine learning. More than 500 publications have been reviewed to identify the common understandings, practical implementation challenges, and research opportunities in the application of AI for power electronics. This article is accompanied by an Excel file listing the relevant publications for statistical analytics.

/pdf/an-overview-of-artificial-intelligence-applications-for-6kbutwbxvl.pdf

An Overview of Artificial Intelligence Applications for Power Electronics

As the Internet of Things (IoT) continues its run as one of the most popular technology buzzwords of today, the discussion really turns from how the massive data sets are collected to how value can be derived from them, i.e., how to extract knowledge out of such (big) data. IoT devices are used in an ever-growing number of application domains (see Figure 1), ranging from sports gadgets (e.g., Fitbits and Apple Watches) or more serious medical devices (e.g., pacemakers and biochips) to smart homes, cities, and self-driving cars, to predictive maintenance in missioncritical systems (e.g., in nuclear power plants or airplanes). Such applications introduce endless possibilities for better understanding, learning, and informedly acting (i.e., situational awareness and actionable information in government lingo). Although rapid expansion of devices and sensors brings terrific opportunities for taking advantage of terabytes of machine data, the mind-boggling task of understanding growth of data remains and heavily relies on artificial intelligence and machine learning [1], [2].

Deep Learning and Reconfigurable Platforms in the Internet of Things: Challenges and Opportunities in Algorithms and Hardware

Condition monitoring is a proactive measure to realize operation optimization, predictive maintenance, and high availability of Power Electronic Systems (PES). It is demanded by reliability-, safety-, or availability-critical applications. The core of condition monitoring is a prediction based on historical and present information. Artificial Intelligence (AI) could play a role in addressing optimization, regression, and classification problems in predicting the operation or health status of PES. Besides AI algorithms, quality data collection, objective formulation, and result validation require an in-depth understanding of the PES. The nexus between PES and AI expects to create overarching effects in the condition monitoring area. This article presents exploratory efforts in the data-driven condition monitoring of PES in the view of existing challenges and emerging opportunities.

Enabling Data-Driven Condition Monitoring of Power Electronic Systems With Artificial Intelligence: Concepts, Tools, and Developments

Complex control structures are required for the operation of photovoltaic electrical energy systems. In this paper, a general review of the controllers used for photovoltaic systems is presented. This review is based on the most recent papers presented in the literature. The control architectures considered are complex hybrid systems that combine classical and modern techniques, such as artificial intelligence and statistical models. The main contribution of this paper is the synthesis of a generalized control structure and the identification of the latest trends. The main findings are summarized in the development of increasingly robust controllers for operation with improved efficiency, power quality, stability, safety, and economics.

A Review of Control Techniques in Photovoltaic Systems

A photovoltaic (PV) health diagnostic system for solar power systems is presented. The system consists of two levels of embedded platforms, including the Data Acquisition Module (DAM) and the Control Module (CM). Each DAM is connected to two series-connected PV panels under test. When testing the PV panels, it will inject large-signal disturbances into their panel voltages. Then, the terminal voltage and current of each panel are sampled and, thus, the dynamic current-voltage characteristics of the PV panel are obtained. The CM has a Real-coded Jumping Gene Genetic Algorithm (RJGGA) programmed and a dedicated Field Programmable Gate Array (FPGA) accelerator designed to evaluate objective functions. Its function is to extract the intrinsic parameters of the panels with the dynamic current-voltage characteristics. Panel degradation can thus be observed with the variation of the estimated intrinsic parameters. Prototypes designed for diagnosing four 80W PV panels have been built and evaluated on panels with different degradation levels.

Photovoltaic Panel Health Diagnostic System for Solar Power Plants

Particle Swarm Optimization (PSO) is a widely used algorithm to solve complex optimization problems with non-linear objective functions. PSO usually requires powerful and expensive computers to achieve reasonable execution times. Sometimes the price or size of the computing system is unacceptable, forcing designers to simplify the objective function or to discard PSO. To overcome this limitation, this paper proposes the implementation of PSO in Field Programmable Systems-on-Chip (FPSoCs). FPSoC devices combine in the same chip powerful processors and reconfigurable logic (FPGA fabric). Experimental results are presented demonstrating that the proposed system achieves a performance similar to that of a desktop computer for a fraction of cost and size. It can be clearly concluded that the proposed system is a good option for running PSO both at design and final application deployment levels.

Implementation of Particle Swarm Optimization in FPSoC devices

Diagnostic module for series-connected photovoltaic panels

The power output of a solar panel depends on the level of degradation and external factors, such as temperature and solar irradiation. The changes in the power output caused by temperature and irradiation fluctuations coincide with the changes due to degradation. Therefore, a common way of degradation determination requires the knowledge of external factors. This work investigates a modeling of the panel power output at the reference state without the knowledge of the external factors. Firstly, a sparse regression modeling is used to identify relevant lumped equivalent circuit parameters, which capture the reference state of the solar panel. The p-n junction capacitance is found to be an important indicator of external factors tied to the panel’s state of health. Secondly, a set of regression models for modelling the solar panel reference state are tested to distinguish a permanent 5% drop in power output from a regular temperature and irradiation fluctuations, without the information of the temperature and irradiation.

Martin Garaj

Papers

Photovoltaic Panel Health Diagnostic System for Solar Power Plants

Implementation of Particle Swarm Optimization in FPSoC devices

Diagnostic module for series-connected photovoltaic panels

Analysis of solar panel’s lumped equivalent circuit parameters using LASSO

A granular modeling method for non-uniform panel degradation based on I–V characterization and electroluminescence imaging