A group of researchers from Switzerland's ETH Zurich – Swiss Federal Institute of Technology and Germany's University of Bamberg has developed a techno-economic simulation model based on a machine learning algorithm, which is aimed at optimizing configuration and profitability of residential solar-plus-storage power systems. In the paper, Economic assessment of photovoltaic battery systems based on household load profiles, the research team has created its model on the basis of real-world energy consumption data from 4,190 Swiss households, which were taken under current electricity rates and weather conditions in Zurich. The authors of the study stressed, however, that their algorithm is based only on a limited set of features, and on shorter measurement time-frames of smart-meter data. Several cost scenarios were presented in the research, the most optimistic of which envisages that the installation of a residential photovoltaic-battery (PVB) system, with a mean installed PV power of 4.4 kW and a mean battery size of 9.6 kWh, will be profitable for 99.9% of Swiss households. Under this scenario, the cost of residential solar is expected to be under €1,000 per kW of PV installed, while that of the battery will not exceed €250 per kWh.

There was much celebrating in America last month when the US Department of Energy unveiled Summit, the world's fastest supercomputer. Now the race is on to achieve the next significant milestone in processing power: exascale computing. This involves building a machine within the next few years that's capable of a billion billion calculations per second, or one exaflop, which would make it five times faster than Summit (see chart). Every person on Earth would have to do a calculation every second of every day for just over four years to match what an exascale machine will be able to do in a flash. This phenomenal power will enable researchers to run massively complex simulations that spark advances in many fields, from climate science to genomics, renewable energy, and artificial intelligence.

With the launch of its Smart Energy subscription plan this year in Philadelphia, Inspire fulfilled a major part of its vision, explained CTO Mike Durst from the company's Center City office. "Our mission is all about creating this brighter energy future," he said. "A pillar of that is getting as many people as possible on clean energy." The subscription plan combines a seamless sign-up, flat supply price, cash rewards (when you use less energy than predicted) and energy-saving smart devices to outfit your home. Also, Inspire's Smart Home app allows you to control the devices and monitor your energy use, all from your smartphone.

Less than a quarter of UK consumers claim to be aware of the term'smart cities', according to a report by out-of-home and location marketing specialists, Posterscope. Smart in the City surveyed more than 5,500 consumers about their opinion on smart cities and their key features. Deeper analysis into the research report, however, shows there is more understanding and interest when respondents were questioned about specific initiatives. When asked which smart city features were deemed to be the most useful, those initiatives that provide a real-life benefit that people can easily relate to were rated the highest: smart water (89 per cent), smart construction (85 per cent), smart energy (81 per cent) and smart health (79 per cent). Features considered the least useful were smart tourism and leisure (59 per cent), smart retail (57per cent) and smart finance (57 per cent).

As commerce and society feel the impact of digitization, automation proliferates along assembly lines and deployment of robotics gathers pace, an influential U.K. academic at the operating helm of the country's engineering academy says the time has come to ditch old clichés and have an apolitical holistic dialogue about future engineering skills. Meet Dr Hayaatun Sillem, Chief Executive of the Royal Academy of Engineering, who reckons such broader changes make it a "very interesting time" for engineers with the so-called Industry 4.0 or the next industrial revolution now firmly among us. "People should look at the ongoing transformation from a prism of not how many jobs will go, but rather at the changing nature and scope of roles and tasks. We should be optimistic that there would be many new jobs created partly through the fact that technology would enable us to do things we could not previously do." Dr Hayaatun Sillem, Chief Executive of the Royal Academy of Engineering, says the broader societal changes we are witnessing make it a "very interesting time" for engineers.Calum McCarron Looking at the size of the opportunity in a global context, engineers are not only among those feeling the first effects of Industry 4.0 but also the drivers of the shift towards a technologically astute low carbon economy, she adds.

A new approach to operating a building's cooling system using machine learning techniques and Internet of Things (IoT) data can help to drive down energy consumption and costs, as the global demand for energy increases. The buildings sector is one of the largest energy-consuming entities, accounting for a staggering 40 percent of global energy consumption today1. What makes this statistic grim is that building energy consumption is projected to increase by 50 percent by 2050, unless energy efficiency strategies are actively embraced to curb the growth. To put this in perspective, the 50 percent increase is equivalent to the combined energy consumption of Russia and India today2. Global pressure for improving environmental sustainability, combined with increasing electricity prices across several nations worldwide, are pushing corporations to reduce the energy consumption incurred in operating their buildings.

Collision avoidance is a critical task in many applications, such as ADAS (advanced driver-assistance systems), industrial automation and robotics. In an industrial automation setting, certain areas should be off limits to an automated vehicle for protection of people and high-valued assets. These areas can be quarantined by mapping (e.g., GPS) or via beacons that delineate a no-entry area. We propose a delineation method where the industrial vehicle utilizes a LiDAR {(Light Detection and Ranging)} and a single color camera to detect passive beacons and model-predictive control to stop the vehicle from entering a restricted space. The beacons are standard orange traffic cones with a highly reflective vertical pole attached. The LiDAR can readily detect these beacons, but suffers from false positives due to other reflective surfaces such as worker safety vests. Herein, we put forth a method for reducing false positive detection from the LiDAR by projecting the beacons in the camera imagery via a deep learning method and validating the detection using a neural network-learned projection from the camera to the LiDAR space. Experimental data collected at Mississippi State University's Center for Advanced Vehicular Systems (CAVS) shows the effectiveness of the proposed system in keeping the true detection while mitigating false positives.

Deep learning usually requires big data, with respect to both volume and variety. However, most remote sensing applications only have limited training data, of which a small subset is labeled. Herein, we review three state-of-the-art approaches in deep learning to combat this challenge. The first topic is transfer learning, in which some aspects of one domain, e.g., features, are transferred to another domain. The next is unsupervised learning, e.g., autoencoders, which operate on unlabeled data. The last is generative adversarial networks, which can generate realistic looking data that can fool the likes of both a deep learning network and human. The aim of this article is to raise awareness of this dilemma, to direct the reader to existing work and to highlight current gaps that need solving.

In this work we offer a framework for reasoning about a wide class of existing objectives in machine learning. We develop a formal correspondence between this work and thermodynamics and discuss its implications.

Oak Ridge National Laboratory unveiled Summit, the world's smartest, most powerful supercomputer. To put it into perspective, the universe is only half an exa seconds old. This supercomputer is designed to tackle challenges that will drive new breakthroughs further than ever before with AI and HPC applications. Summit's ability to combine HPC and AI techniques will give researchers the ability to automate, accelerate, and drive advancements in fields such as health, energy, and engineering. Summit runs 8 times faster than it's previous supercomputer, Titan.