Intelligent control of integrated on-board charger with improved power quality and reduced charging transients.

This paper proposes an intelligent control scheme for a two-stage integrated onboard electric vehicle (EV) battery charger connected to a single-phase household outlet which offers a close to ideal battery charging profile with power factor correction feature. Generally, the front-end AC-DC​ conversion stage is controlled by dual loop proportional-integral (PI) controllers, and tuning their gain constants is a difficult task. Furthermore, to achieve a close to ideal charging profile for an EV battery, the DC-DC conversion stage switches from constant current (CC) and constant voltage (CV) mode after a certain state of charge (SOC) which may lead to discontinuity in the charging current and voltage. This paper attempts to solve these issues by proposing an intelligent control scheme that includes the dynamic estimation of PI controller gain constants as well as provides a seamless mode transfer feature for battery charging. It is achieved by using fuzzy-PI-based control in the AC-DC conversion stage and Bayesian Regularization (BR) algorithm trained artificial neural network (ANN)-based control in the DC-DC conversion stage. The performance of the proposed control scheme is assessed both in steady-state and transient conditions in MATLAB® Simulink environment by comparing it against similar control schemes. The proposed intelligent control approach improves the dynamic response of DC link voltage, offers unity power factor operation and maintains the line current harmonics within IEEE 519 standards even during the switchover from CC to CV charging mode. Also, there is a decrease of 85% in the third harmonic component of the source current, 23.2% improvement in DC link voltage undershoot and 6.5% reduction in DC link voltage overshoot with reduced settling times using the proposed unified control scheme.

Deep Learning for Plant Stress Phenotyping: Trends and Future Perspectives.

Deep learning (DL), a subset of machine learning approaches, has emerged as a versatile tool to assimilate large amounts of heterogeneous data and provide reliable predictions of complex and uncertain phenomena. These tools are increasingly being used by the plant science community to make sense of the large datasets now regularly collected via high-throughput phenotyping and genotyping. We review recent work where DL principles have been utilized for digital image-based plant stress phenotyping. We provide a comparative assessment of DL tools against other existing techniques, with respect to decision accuracy, data size requirement, and applicability in various scenarios. Finally, we outline several avenues of research leveraging current and future DL tools in plant science.

Machine Learning for High-Throughput Stress Phenotyping in Plants.

Advances in automated and high-throughput imaging technologies have resulted in a deluge of high-resolution images and sensor data of plants. However, extracting patterns and features from this large corpus of data requires the use of machine learning (ML) tools to enable data assimilation and feature identification for stress phenotyping. Four stages of the decision cycle in plant stress phenotyping and plant breeding activities where different ML approaches can be deployed are (i) identification, (ii) classification, (iii) quantification, and (iv) prediction (ICQP). We provide here a comprehensive overview and user-friendly taxonomy of ML tools to enable the plant community to correctly and easily apply the appropriate ML tools and best-practice guidelines for various biotic and abiotic stress traits.