DUBAI, UNITED ARAB EMIRATES - A Yemen rebel drone strike this week on a critical Saudi oil pipeline shows that the otherwise-peaceful sandy reaches of the Arabian Peninsula now are at risk of similar assault, including an under-construction nuclear power plant and Dubai International Airport, among the world's busiest. U.N. investigators said the Houthis' new UAV-X drone, found in recent months during the Saudi-led coalition's war in Yemen, likely has a range of up to 1,500 km (930 miles). That puts the far reaches of both Saudi Arabia and the United Arab Emirates, the two main opponents of the Iranian-allied Houthi rebels in Yemen, within reach of drones difficult to detect and track. Their relatively simple design, coupled with readily available information online, makes targeting even easier, analysts say. "These installations are easily findable, like on Google Earth," said Tim Michetti, an expert on illicit weapons technology with experience in Yemen.
For scale, consider the Statue of Liberty, standing 305 feet tall. At 466 feet, the average wind turbine in the U.S. dwarfs Lady Liberty by more than half. And when GE's next-generation monster wind turbine, the Haliade-X, hits the market in 2021, it will nearly double that size to 877 feet, just shy of the Eiffel Tower. A single Haliade-X rotor blade will stretch 315 feet, longer than a football field. As a general rule of thumb, when it comes to energy and energy exploration, bigger is better: the larger the machinery, the deeper the dig, the greater the production yield.
Forecasting multivariate time series data, such as prediction of electricity consumption, solar power production, and polyphonic piano pieces, has numerous valuable applications. However, complex and non-linear interdependencies between time steps and series complicate the task. To obtain accurate prediction, it is crucial to model long-term dependency in time series data, which can be achieved to some good extent by recurrent neural network (RNN) with attention mechanism. Typical attention mechanism reviews the information at each previous time step and selects the relevant information to help generate the outputs, but it fails to capture the temporal patterns across multiple time steps. In this paper, we propose to use a set of filters to extract time-invariant temporal patterns, which is similar to transforming time series data into its "frequency domain". Then we proposed a novel attention mechanism to select relevant time series, and use its "frequency domain" information for forecasting. We applied the proposed model on several real-world tasks and achieved the state-of-the-art performance in all of them with only one exception. We also show that to some degree the learned filters play the role of bases in discrete Fourier transform.
DNV GL operates globally across a range of industries to provide trust and help stakeholders understand and manage risks. The company has over 2,200 experts providing advice for renewable generation, transmission, distribution and energy management and efficiency. The work they do ranges from testing high voltage grid components to bankability assessments to allow solar farms to be financed. Their latest research showcases the insight related to how drone technology and computer vision can help with inspections while also providing a global outlook to 2050. We wanted to learn more about the energy innovations DNV GL is exploring and utilizing, so we talked with Elizabeth Traiger, Senior Researcher, Power & Renewables DK & GB at DNV GL – Energy.