Stochastic Computing Emulation of Memristor Cellular Nonlinear Networks

Cellular Nonlinear Networks (CNN) are a concept introduced in 1988 by Leon Chua and Lin Yang as a bio-inspired architecture capable of massively parallel computation. Since then, CNN have been enhanced by incorporating designs that incorporate memristors to profit from their processing and memory capabilities. In addition, Stochastic Computing (SC) can be used to optimize the quantity of required processing elements; thus it provides a lightweight approximate computing framework, quite accurate and effective, however. In this work, we propose utilization of SC in designing and implementing a memristor-based CNN. As a proof of the proposed concept, an example of application is presented. This application combines Matlab and a FPGA in order to create the CNN. The implemented CNN was then used to perform three different real-time applications on a 512 × 512 gray-scale and a 768 × 512 color image: storage of the image, edge detection, and image sharpening. It has to be pointed out that the same CNN was used for the three different tasks, with the sole change of some programmable parameters. Results show an excellent capability with significant accompanying advantages, such as the low number of needed elements further allowing for a low cost FPGA-based system implementation, something confirming the system’s capacity for real time operation.

Tracking and Monitoring of the Alignment System Used for Nuclear Physics Experiments

The ELI-NP (Extreme Light Intensity—Nuclear Physics) project, developed at the Horia Hulubei National Institute for RD in Physics and Nuclear Engineering (IFIN-HH), has included one component dedicated to the study of interactions between brilliant gamma-ray and matter, with applications in nuclear physics and the science of materials. The paper is focused on the interaction chamber, an important part of the facility which hosts the experiment’s samples. The interaction chamber is endowed with a mobile sample support (holder), which automatically tracks the γ-ray beam. The γ-ray radiation source presents a slight variation of the direction of the emitted radiation in time. The built system ensures the permanent collimation between the γ-ray beam and the sample that is being investigated. This is done with two electric motors, which have a symmetrical movement with respect to the center of a rectangle. The specific measures taken by the design and implementation that permit to reach performances of tracking system are emphasized in the paper. The methodology considers the relative displacement between the detectors with which the laboratory is equipped and the absolute position in space of the sample boundary. The control of this motion is designed to respect the symmetry of the system. Both facets of the project (hardware and software) are detailed, emphasizing the way in which the designers ensured compliance with the system of real-time operation conditions of the tracking and monitoring system.

Research on Identification of Power Grid Weakness Based on Bayesian Inference

With the gradual reform and development of the power grid, it is of great significance to study how to effectively identify and evaluate the weak links of the power grid for the actual planning, construction, and operation of the power grid. This paper analyzed the power grid’s historical component data and real-time operation state parameters. We established a weak link identification model based on Bayesian reasoning. Firstly, we constructed the node branch Bayesian network according to the network topology relationship. The power transmission distribution factor is modified according to the historical operation load of the grid components, and the conditional probability table is calculated based on the grid structure; finally, we used the maximum possible explanation algorithm in the Bayesian network. The weakness degree of all components in the network is calculated, and the maximum probability weak link sequence is obtained. The correctness and effectiveness of the proposed method are verified by IEEE 39 bus simulation and regional power grid data.

A numerical typhoon track prediction model and its application on the real-time operation

ABSTRACT. Recently a nested numerical model for typhoon track prediction (MTP) with a higher horizontal resolution and a complex physical package has been developed by the National Meteorological Center (NMC, Beijing). For its improvement of initialization, a modified typhoon bogus scheme (Iwasaki 1987) also has been applied. The design of the M1TP had been determined by the end of 1992 and the forecast capability of this model was tested firstly with a series of experiments for the selected typhoon cases in 1993 (Wang and Li 1994). After it was examined in real-time forecast in the next year, further improvements were made in 1995 including a higher horizontal resolution increased from about 100 to 50 km, a package of complex physics instead of the simple one and a scheme for removal of analyzed vortex. With the improved forecast capability, the MTP run in quasi-operation from the date of 1 June 1995. Its products were also used by the forecasters during the past two years and the results were very encouraging.

Real-Time Edge Neuromorphic Tasting From Chemical Microsensor Arrays

Liquid analysis is key to track conformity with the strict process quality standards of sectors like food, beverage, and chemical manufacturing. In order to analyse product qualities online and at the very point of interest, automated monitoring systems must satisfy strong requirements in terms of miniaturization, energy autonomy, and real time operation. Toward this goal, we present the first implementation of artificial taste running on neuromorphic hardware for continuous edge monitoring applications. We used a solid-state electrochemical microsensor array to acquire multivariate, time-varying chemical measurements, employed temporal filtering to enhance sensor readout dynamics, and deployed a rate-based, deep convolutional spiking neural network to efficiently fuse the electrochemical sensor data. To evaluate performance we created MicroBeTa (Microsensor Beverage Tasting), a new dataset for beverage classification incorporating 7 h of temporal recordings performed over 3 days, including sensor drifts and sensor replacements. Our implementation of artificial taste is 15× more energy efficient on inference tasks than similar convolutional architectures running on other commercial, low power edge-AI inference devices, achieving over 178× lower latencies than the sampling period of the sensor readout, and high accuracy (97%) on a single Intel Loihi neuromorphic research processor included in a USB stick form factor.

Digital Drilling in Holographic World

Nowadays oil & gas industry is receiving a bulk of data than ever before from its onsite wells where may hundred miles away from operator's headquarter, which benefits us monitoring and analyzing those digital fortune in a data hub, saving a lot of expenditure and improving the efficiency compared to old-fashioned approach which requires senior engineers with rich experience working on wellsite. In this way, the oil & gas operators save money tremendously on human cost under the booming of drilling operations. While, could we do more to dig out further values from those data? Make our operations less dependable on limited resources, the senior drilling engineers, especially when the oil and gas industry face the chasm of human resources sustainability after the hit of downturn, also make the plain real-time data more intuitive and self-explanatory to the operation decision makers in an unprecedented way. What's more, could we make our drilling activities more visible and interactive? This paper is going to introduce using augmented reality technology to create an intuitive platform to integrate and present real-time operation parameters and data. Like any revolutionary method or technology, it could improve the industry efficiency in a non-negligible way, help us manage massive real-time data more effectively and efficiently. The 3D holographic projection presents dynamic models or systems based on the data stream and graphic algorithm, which evolves our industry from 2D world to 3D world, combining the reality environment with the digital world, creating a digital reflection of the real wellsite, bottom hole assembly (BHA), well trajectories, lithological layers, etc. Thanks to the visualization technologies and augmented reality, we can create a digital twin of physic world for those engineers, technicians, managers using holographic method to interact with, scale up and down, analyzing in a better awareness. In this paper, we will describe a digital drilling wellsite which is established on operator's Real Time Operation Center (RTOC)office to monitor and analyze live field operations, the operator could have an overview of their on-site operations, tracking the equipment performance, engineering parameters and downhole status to enhance the understanding and interaction with the on-going field operations. The wellbore trajectory model gives the team a superior knowledge by combining the engineering data or geological data. Not only help well placement in desired reservoir but also improve the anti-collision concept in direction drilling. This model is extreme meaningful when engineers need a discussion to optimize or change their drilling plan as it is 3D visible and able to interact with. We will continues digging out further more value of the real-time data collected from wellsite to educate us find the cost-saving ways which improve our performance and eliminate the complicated conditions that normally resulted in Non production time (NPT) event. For our oil & gas industry, we are just start to have a more adventure and prosperous journey in digitalizing transforming.

PHIL implementation of a decentralized online OPF for active distribution grids

This paper demonstrates a Power Hardware-in-the-Loop (PHIL) implementation of a decentralized optimal power flow (D-OPF) algorithm embedded into the operations of two microgrids connected by a tie line. To integrate the static behavior of the optimization model, a two layer control architecture is introduced. Underneath the dispatch commands from the D-OPF, a primary control scheme provides instantaneous reaction to the load dynamics. This setup is tested in the PHIL environment of the CoSES Lab in TU Munich. In the experiment, the two microgrids cooperatively optimize their operation through an ADMM based unbalanced D-OPF. The operations is then benchmarked against the exclusive use of primary control, without D-OPF. The decentralized approach outperforms, but also shows minor inefficiencies of integrating optimization methods into the real-time operation of the system.<br>