Time difference of Arrival (TDOA)-based localization method, although used widely, calls for a fast and accurate solution owing to its time inefficiency and sensitivity to time delay estimation. In order to speed up the solution for nonlinear TDOA equations, while guaranteeing the location accuracy, this paper presents a hybrid approach namely multi-deep neural network model based on a virtual measurement method (MDNNM-VMM). Data consisting of multiple time difference values, resulting from a virtual measurement method (VMM), are fed to a pretrained multi-deep neural network model (MDNNM). These ‘n’ number of virtually generated sequences of time delays are obtained from a single set of TDOA equations, while conforming with a uniform distribution. The multi-DNN model using these data, outputs the required partial discharge (PD) coordinates that help determine accurate PD location. While applying a measurement error of 24 ns, the average error values $\Delta \text{r}$ , $\Delta \theta $ , $\Delta \Phi $ , and $\Delta \text{d}$ for the proposed method, compared to a multi-DNN method, see a significant percentage decrease of 32 %, 24 %, 39 %, and 44 %, respectively. Additionally, varying simulated error, different array designs, and certain other parameters are studied to make the PD localization process more efficient and multifaceted.

/pdf/partial-discharge-localization-in-3-d-with-a-multi-dnn-model-1qilnfbse7.pdf

Partial Discharge Localization in 3-D With a Multi-DNN Model Based on a Virtual Measurement Method

The traditional method of using electric field sensors to realize early warning of electric power safety distance cannot measure the distance of dangerous sources. Therefore, aiming at the electric field with a frequency of 50 to 60 Hz (AC electric field), a new method for localization of aerial AC target by the capacitive one-dimensional spherical electric field sensor circular array is studied. This method can directly calculate the distance, elevation, and azimuth of the detector from the dangerous source. By combining the measurement principle of the spherical electric field sensor and the plane circular array theory, a mathematical model for the localization of aerial targets in an AC electric field is established. An error model was established using Gaussian noise and the effects of different layout parameters on the localization error were simulated. Based on mutual interference between sensors, minimum induced charge, and localization error, an optimal model for sensor layout was established, and it was solved by using genetic algorithms. The optimization results show that when the number of sensors is 4, the array radius is 20 cm, and the sensor radius is 1.5 cm, the ranging error is 8.4%. The detector was developed based on the layout parameters obtained from the optimization results, and the localization method was experimentally verified at 10 and 35 kV alarm distances. The experimental results show that when the detector is located at 10 kV alarm distance, the distance error is 0.18 m, the elevation error is 6.8°, and the azimuth error is 4.57°, and when it is located at 35 kV alarm distance, the distance error is 0.2 m, the elevation error is 4.8°, and the azimuth error is 5.14°, which meets the safety distance warning requirements of 10 and 35 kV voltage levels.

https://www.mdpi.com/1424-8220/20/6/1585/pdf

Method for Localization Aerial Target in AC Electric Field Based on Sensor Circular Array.

Partial discharge (PD) is a common defect to determine the quality of high-voltage power equipment insulation systems. Localization of PD is very important for equipment maintenance and risk assessment. A localization algorithm of received signal strength (RSS) using demodulating PD optical signal is proposed. This is based on the optical effects caused by PD, which influence the optical signal that is transmitted through an optical fiber sensor placed inside the equipment. Compared with plastic fluorescent fiber, SiO2 cladding YAP:Ce crystal core fluorescent fiber (SYCF) with high luminous efficiency, low loss, and strong anti-electromagnetic interference capability is selected as the sensing unit to detect the received optical signal strength of PD, through the all-fiber optic sensing system to achieve the PD location in 3-D space. Based on the optical RSS localization method, the range of localization deviation of PD under 37-kV applied voltage is measured and studied. The average localization deviation and the error are 6.9 cm and 7.4%, respectively. Comparing the optical signal location method based on RSS with other methods of PD localization, the advantages and accuracy of the optical location method proposed in this article are outstanding. Therefore, the all-fiber system with SYCF via optical signal has a wide application prospect in the spatial localization of PD.

Localization of Partial Discharge Using All-Fiber System Based on Crystal Fluorescent Fiber

The current technological world is growing rapidly and each aspect of life is being transformed toward automation for human comfort and reliability. With autonomous vehicle technology, the communication gap between the driver and the traditional vehicle is being reduced through multiple technologies and methods. In this regard, state-of-the-art methods have proposed several approaches for advanced driver assistance systems (ADAS) to meet the requirement of a level-5 autonomous vehicle. Consequently, this work explores the role of textual cues present in the outer environment for finding the desired locations and assisting the driver where to stop. Firstly, the driver inputs the keywords of the desired location to assist the proposed system. Secondly, the system will start sensing the textual cues present in the outer environment through natural language processing techniques. Thirdly, the system keeps matching the similar keywords input by the driver and the outer environment using similarity learning. Whenever the system finds a location having any similar keyword in the outer environment, the system informs the driver, slows down, and applies the brake to stop. The experimental results on four benchmark datasets show the efficiency and accuracy of the proposed system for finding the desired locations by sensing textual cues in autonomous vehicles.

/pdf/an-intelligent-system-to-sense-textual-cues-for-location-1r57ub5q.pdf

An Intelligent System to Sense Textual Cues for Location Assistance in Autonomous Vehicles

Accurate partial discharge localisation using a multi-deep neural network model trained with a novel virtual measurement method

The localization of partial discharges in air-insulated substations using ultra-high frequency technology is widely studied for power equipment early warning purposes. Ultra-high frequency partial discharge localization systems are usually based on the time-difference of electromagnetic wave signals. However, the large size of test equipment and the need for a high sampling rate and time synchronization accuracy limit their practical application. To address this challenge, this paper proposes a power-based partial discharge direction of arrival method using a received signal strength indicator from an ultra-high frequency wireless sensor array. Furthermore, the Gaussian mixture model is used for noise suppression, and the Gaussian process classifier is used for line of sight received signal strength indicator data identification. Laboratory tests are performed and the results show the average error of direction of arrival is less than 5°. The results verify the effectiveness of the proposed partial discharge localization system.

/pdf/rssi-power-based-direction-of-arrival-estimation-of-partial-34ig0kewqz.pdf

RSSI-Power-Based Direction of Arrival Estimation of Partial Discharges in Substations

Partial Discharge Angle of Arrival Estimation Using UHF Wireless Sensor Array

This paper presents a novel method for the dynamic positioning of an unmanned underwater vehicle (UUV) with unknown trajectories based on an autonomous tracking buoy (PUVV-ATB) that indirectly positions the UUV using ultra-short baseline measurements. The method employs a spatial location geometric model and divides the positioning process into four steps, including data preprocessing to detect geometric errors and apply mean filtering, direction capture, position tracking, and position synchronization. To achieve these steps, a new adaptive tracking control algorithm is proposed that does not require trajectory prediction and is applied to the last three steps. The algorithm is deployed to the buoy for tracking simulation and sea trial experiments, and the results are compared with those of a model predictive control algorithm. The autonomous tracking buoy based on the adaptive tracking control algorithm runs more stably and can better complete the precise tracking task for the UUV with a positioning error of less than 10 cm. This method breaks the premise of trajectory prediction based on traditional tracking control algorithms, providing a new direction for further research on UUV localization. Furthermore, the conclusion of this paper has important reference value for other research and application fields related to UUV.

/pdf/positioning-of-unmanned-underwater-vehicle-based-on-9jbtgmvc.pdf

Positioning of Unmanned Underwater Vehicle Based on Autonomous Tracking Buoy

Location Based Service (LBS) is becoming increasingly valuable in many indoor application scenarios. However, existing indoor positioning technologies are either too costly or have high computational volumes. It is a challenge to achieve positioning both at low cost and low computation. To address this problem, we propose an indoor positioning algorithm based on computer vision, which can achieve localization at a low-cost and low computational complexity. In the proposed method, the lightweight MobileNet network is used to build an image classification model to train and optimize the collected dataset, then the predicted positions are obtained by inputting the images of the locations to be measured into the trained model. The experimental results demonstrate that the MobileNet model has lower computational effort than the classical convolutional neural network CNN. The designed computer vision-based method can successfully recognize images for indoor positioning, which is highly practical. ∗Corresponding author Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org. APIT 2022, January 14–16, 2022, Virtual Event, Thailand © 2022 Association for Computing Machinery. ACM ISBN 978-1-4503-9557-1/22/01. . . $15.00 https://doi.org/10.1145/3512353.3512378 CCS CONCEPTS •Computingmethodologies→Artificial intelligence; Computer vision.

Anqing Sun

Papers

RSSI-Power-Based Direction of Arrival Estimation of Partial Discharges in Substations

Partial Discharge Angle of Arrival Estimation Using UHF Wireless Sensor Array

Positioning of Unmanned Underwater Vehicle Based on Autonomous Tracking Buoy

Low-cost and lightweight indoor positioning based on computer vision