Project and Development of a Reinforcement Learning Based Control Algorithm for Hybrid Electric Vehicles

Hybrid electric vehicles are, nowadays, considered as one of the most promising technologies for reducing on-road greenhouse gases and pollutant emissions. Such a goal can be accomplished by developing an intelligent energy management system which could lead the powertrain to exploit its maximum energetic performances under real-world driving conditions. According to the latest research in the field of control algorithms for hybrid electric vehicles, Reinforcement Learning has emerged between several Artificial Intelligence approaches as it has proved to retain the capability of producing near-optimal solutions to the control problem even in real-time conditions. Nevertheless, an accurate design of both agent and environment is needed for this class of algorithms. Within this paper, a detailed plan for the complete project and development of an energy management system based on Q-learning for hybrid powertrains is discussed. An integrated modular software framework for co-simulation has been developed and it is thoroughly described. Finally, results have been presented about a massive testing of the agent aimed at assessing for the change in its performance when different training parameters are considered.

Intelligent on-demand design of phononic metamaterials

With the growing interest in the field of artificial materials, more advanced and sophisticated functionalities are required from phononic crystals and acoustic metamaterials. This implies a high computational effort and cost, and still the efficiency of the designs may be not sufficient. With the help of third-wave artificial intelligence technologies, the design schemes of these materials are undergoing a new revolution. As an important branch of artificial intelligence, machine learning paves the way to new technological innovations by stimulating the exploration of structural design. Machine learning provides a powerful means of achieving an efficient and accurate design process by exploring nonlinear physical patterns in high-dimensional space, based on data sets of candidate structures. Many advanced machine learning algorithms, such as deep neural networks, unsupervised manifold clustering, reinforcement learning and so forth, have been widely and deeply investigated for structural design. In this review, we summarize the recent works on the combination of phononic metamaterials and machine learning. We provide an overview of machine learning on structural design. Then discuss machine learning driven on-demand design of phononic metamaterials for acoustic and elastic waves functions, topological phases and atomic-scale phonon properties. Finally, we summarize the current state of the art and provide a prospective of the future development directions.

Interaction of excitons with Cherenkov radiation in WSe2 beyond the non-recoil approximation

Cherenkov radiation from electrons propagating in materials with a high refractive index have applications in particle-detection mechanisms and could be used for high-yield coherent electron beam-driven photon sources. However, the theory of the Cherenkov radiation has been treated up to now using the non-recoil approximation, which neglects the effect of electron deceleration in materials. Here, we report on the effect of electron-beam deceleration on the radiated spectrum and exciton-photon interactions in nm-thick 〖WSe〗_2 crystals. The calculation of the Cherenkov radiation is performed by simulating the kinetic energy of an electron propagating in a thick sample using the Monto Carlo method combined with the Lienard-Wiechert retarded potential. Using this approach, we numerically investigate the interaction between the excitons and generated photons (Cherenkov radiation) beyond the non-recoil approximation and are able to reproduce experimental cathodoluminescence spectra. Our findings pave the way for an accurate design of particle scintillators and detectors, based on the strong-coupling phenomenon.

Prolonged XPO1 inhibition is essential for optimal anti-leukemic activity in NPM1-mutated AML

NPM1 encodes for a nucleolar multifunctional protein and is the most frequently mutated gene in adult acute myeloid leukemia (AML). NPM1 mutations cause the aberrant accumulation of mutant NPM1 (NPM1c) in the cytoplasm of leukemic cells, that is mediated by the nuclear exporter Exportin-1 (XPO1). Recent work has demonstrated that the interaction between NPM1c and XPO1 promotes high homeobox (HOX) genes expression, which is critical for maintaining the leukemic state of NPM1-mutated cells. However, the XPO1 inhibitor Selinexor administered once or twice/week in early-phase clinical trials did not translate into clinical benefit for NPM1-mutated AML patients. Here, we demonstrate that this dosing strategy results in only temporary disruption of the XPO1-NPM1c interaction and transient HOX genes downregulation, limiting the efficacy of Selinexor in the context of NPM1-mutated AML. Since second-generation XPO1 inhibitors can be administered more frequently, we compared intermittent (twice/week) versus prolonged (5 days/week) XPO1 inhibition in NPM1-mutated AML models. Integrating in vitro and in vivo data, we show that only prolonged XPO1 inhibition results in stable HOX downregulation, cell differentiation and remarkable anti-leukemic activity. This study lays the groundwork for the accurate design of clinical trials with second-generation XPO1 inhibitors in NPM1-mutated AML.

Algorithm for Compression Design Allowable Determination of Composite Laminates with Initial Delaminations

With the increasing demands for detailed design of composite aircraft structures, the method of covering all damages with low design allowables cannot meet the current requirements for aircraft structure design. Herein, this paper proposes a novel algorithm for design allowable determination of composite laminates by combining the damage distribution with damage factor model of design allowable, so as to provide different structures with more accurate design allowables based on their initial damages. For the composite laminates with initial delaminations, a model describing the effect of delamination size and depth position on the compression design allowable is developed and the compression design allowable of different aircraft structures are individually determined by employing abundant initial delamination statistics. Compared with the design allowable offered by the single-point method, the design allowable based on the initial damage can be increased by at least 5% to 20%, greatly improving the economic benefits of the aircraft structures and providing an important support for the damage tolerance design of the composite structures.

Near-Field Route Optimization-Supported Polar Ice Navigation via Maritime Radar Videos

The accurate design of ship routing plans in arctic areas is not easy, considering that navigation conditions (e.g., weather, visibility, and ice thickness) may change frequently. A ship’s crew identifies sea ice in arctic channels with the help of radar echoes, and ship maneuvering decisions are made to avoid navigation interference. Ship officials must manually and consistently change the ship’s route of travel, which is time-consuming and tedious. To address this issue, we propose a near-field route optimization model for the purpose of automatically selecting an optimal route with the help of radar echo images. The ship near-field route optimization model uses a multiobjective optimal strategy considering factors of minimum navigation risk and steaming distance. We verified the model’s performance with the support of the Xuelong voyage dataset. This research finding can help a ship’s crew to design more reasonable navigation routes in polar channels.

Support–Activity Relationship in Heterogeneous Catalysis for Biomass Valorization and Fine-Chemicals Production

Heterogeneous catalysts are progressively expanding their field of application, from high-throughput reactions for traditional industrial chemistry with production volumes reaching millions of tons per year, a sector in which they are key players, to more niche applications for the production of fine chemicals. These novel applications require a progressive utilization reduction of fossil feedstocks, in favor of renewable ones. Biomasses are the most accessible source of organic precursors, having as advantage their low cost and even distribution across the globe. Unfortunately, they are intrinsically inhomogeneous in nature and their efficient exploitation requires novel catalysts. In this process, an accurate design of the active phase performing the reaction is important; nevertheless, we are often neglecting the importance of the support in guaranteeing stable performances and improving catalytic activity. This review has the goal of gathering and highlighting the cases in which the supports (either derived or not from biomass wastes) share the worth of performing the catalysis with the active phase, for those reactions involving the synthesis of fine chemicals starting from biomasses as feedstocks.

Extracting Dielectric Permittivity with a Cross-Like Stripline

Parametric retrieval of electromagnetic properties is important for both new materials characterization and an accurate design of devices. While quite a few techniques have been developed over the years, precise mapping of high-permittivity samples remain challenging. Here we advance a so-called micro-strip technique, where transmission coefficients of a waveguide system with an analyte on top are used to extract electromagnetic parameters of the later. Our cross-like strip line configuration has a split ring resonator on one edge and an open circuit termination on another. This design allows performing a simultaneous test of cylindrical and rectangular samples. Our new post-processing scheme was tested on a water-filled container and showed 96.3% accuracy, assessed by comparing our results with tabulated data.

Optimization of Charge Pump Based on Piecewise Modeling of Output-Voltage Ripple

This work proposes a piecewise modeling of output-voltage ripple for linear charge pumps. The proposed modeling can obtain a more accurate design expression of power-conversion efficiency. The relationship between flying and output capacitance, as well as switching frequency and optimize power-conversion efficiency can be calculated. The proposed modeling is applied to three charge-pump circuits: 1-stage linear charge pump, dual-branch 1-stage linear charge pump and 4× cross-coupled charge pump. Circuit-level simulation is used to verify the accuracy of proposed modeling.