Evaluation Model of Power Operation and Maintenance Based on Text Emotion Analysis

The application of data mining technology in power field mainly focuses on the application of power defect text and dispatching text. However, the power operation and maintenance data contains a lot of information about power equipment suppliers. Taking the operation and maintenance text involving power equipment suppliers as an example, this paper summarizes the theme of operation and maintenance text and studies the evaluation model of power equipment suppliers. The next sentence prediction analysis model of single round dialogue text based on transformer bidirectional encoder prediction and cosine similarity weighting is proposed, which can effectively divide the topic of dialogue text. Aiming at the semantic richness and complexity of power operation and maintenance text, a supplier evaluation model based on text emotion analysis is proposed. Based on the expansion of the entries and attributes of the existing power ontology dictionary, the dialogue emotion analysis rules are established to realize the normal evaluation of power equipment suppliers.

Neuromorphic Dynamics at the Nanoscale in Silicon Suboxide RRAM

Resistive random-access memories, also known as memristors, whose resistance can be modulated by the electrically driven formation and disruption of conductive filaments within an insulator, are promising candidates for neuromorphic applications due to their scalability, low-power operation and diverse functional behaviors. However, understanding the dynamics of individual filaments, and the surrounding material, is challenging, owing to the typically very large cross-sectional areas of test devices relative to the nanometer scale of individual filaments. In the present work, conductive atomic force microscopy is used to study the evolution of conductivity at the nanoscale in a fully CMOS-compatible silicon suboxide thin film. Distinct filamentary plasticity and background conductivity enhancement are reported, suggesting that device behavior might be best described by composite core (filament) and shell (background conductivity) dynamics. Furthermore, constant current measurements demonstrate an interplay between filament formation and rupture, resulting in current-controlled voltage spiking in nanoscale regions, with an estimated optimal energy consumption of 25 attojoules per spike. This is very promising for extremely low-power neuromorphic computation and suggests that the dynamic behavior observed in larger devices should persist and improve as dimensions are scaled down.

Batteryless, Miniaturized Implantable Glucose Sensor Using a Fluorescent Hydrogel

We propose a biomedical sensor system for continuous monitoring of glucose concentration. Despite recent advances in implantable biomedical devices, mm sized devices have yet to be developed due to the power limitation of the device in a tissue. We here present a mm sized wireless system with backscattered frequency-modulation communication that enables a low-power operation to read the glucose level from a fluorescent hydrogel sensor. The configuration of the reader structure is optimized for an efficient wireless power transfer and data communication, miniaturizing the entire implantable device to 3 × 6 mm 2 size. The operation distance between the reader and the implantable device reaches 2 mm with a transmission power of 33 dBm. We demonstrate that the frequency of backscattered signals changes according to the light intensity of the fluorescent glucose sensor. We envision that the present wireless interface can be applied to other fluorescence-based biosensors to make them highly comfortable, biocompatible, and stable within a body.

Prediction of divertor heat flux width for ITER Pre-Fusion Power Operation using BOUT++ transport code

Prediction of divertor heat flux width is performed for the first and the second Pre-Fusion Power Operation (PFPO) phases specified in the new ITER Research Plan using BOUT++ transport code [Li N.M. et al 2018 Comput. Phys. Commun. 228 69–82]. The initial plasma profiles inside the separatrix are taken from CORSICA scenario studies. Transport coefficients in transport code are calculated by inverting the plasma profiles inside the separatrix and are assumed to be constants in the scrape-off-layer (SOL). An anomalous thermal diffusivity scan is performed with E×B and magnetic drifts. The results in two scenarios identify two distinct regimes: a drift dominant regime when diffusivity is smaller than the respective critical diffusivity χc and a turbulence dominant regime when diffusivity is larger than it. The Goldston heuristic drift model and the ITPA multi-machine experimental scaling yield a lower limit of the width λq. From transport simulations, we obtain the critical diffusivity χc = 0.5 m2⁄ s in 5MA/1.77T PFPO-1 scenario and χc = 0.3 m2⁄ s in 7.5MA/2.65T PFPO-2 scenario. Separatrix temperature and collisionality also have a significant impact on the heat flux width in the drift dominant regime. The investigation clearly yields a scaling for critical thermal diffusivity χc ∝ A½ ⁄ ((Z(1+Z)½ Bp 2)) using ITER scenarios with fixed safety factor q95, major radius R, aspect ratio R/a, and the separatrix temperature T, as well as established the connection with CFETR and C-Mod discharges. This scaling implies that for a given tokamak device with q95, R, R/a, and T fixed, a reduction of poloidal magnetic field by a factor of 3 leads to a 9 times higher critical value of thermal diffusivity χc, possibly yielding a transition from turbulence to drift dominant regime.

Research on the computer intelligent recognition of electric appliance by Yolo V5 algorithm

In the process of power operation, the correct identification of tools can lay a foundation for the detection of violations in power operation. In order to realize the recognition of power instruments, based on the current Yolo V5 algorithm, a detection algorithm for power instruments is proposed by improving Yolo V5 algorithm. Firstly, the model of Yolo V5 algorithm is introduced. Then the establishment of the power tool database and the process of model training are analysed. Finally, the test results are analysed and evaluated. The models generated after training were accelerated by TensorRT and then deployed on Jetson Xavier NX.

Research on Visual Detection Algorithm of 500kV High Voltage Live Operating Robot

The 500kV transmission line is exposed to the outdoor for a long time. It is affected by complex climate charge change and other factors which leads to line connection parts bolt loosening, wire breaking and fixture damage and other line failures. In order to ensure the stable transmission of electric energy, power operators need to wear shielding suits and work in the high-risk and high-voltage environment. Use of electric power operation robot instead of manual operation is an effective way to liberate maintenance staff labor. But robots still have some problems such as low degree of automation and low efficiency of power operation. Machine vision detection technology in recent years has been widely used in major areas including deep learning as emerging visual detection technology shows excellent performance. In this paper, the vision detection algorithm is studied respectively for the bolt fastening end working device and wire repairing end working device of the live transmission line robot to improve the operating efficiency of the robot.