Weitao Lu

Characteristics of unconnected upward leaders initiated from tall structures observed in Guangzhou

[1]xa0Forty-five unconnected upward leaders (UULs) occurred in 19 downward negative flashes are analyzed. Each observed UUL is initiated by a downward stepped leader before a new strike point is struck. For each UUL, several parameters are determined when possible mainly by using high-speed images: inception height, inception time prior to return stroke (RS), horizontal distance from the flashs strike point, two-dimensional (2D) distance between the nearest downward leader branch tip and the UULs inception point at its inception time, 2D length, and 2D average propagation velocity. Their values range from 40 to 503xa0m (number of samples: 45), <0.1 to 1.32xa0ms (38), 20xa0m to 1.3xa0km (38), 99 to 578xa0m (21), 0.48 to 399xa0m (45), and 5.79 to 33.8 × 104xa0m s−1 (22), respectively. 86% (19/22) of the velocities are smaller than 1.7 × 105xa0m s−1. No UUL with an inception time prior to RS greater than 0.5xa0ms is initiated from a structure lower than 300xa0m. Those UULs with inception heights lower than 300xa0m seldom exhibit lengths longer than 50xa0m and only can be initiated by flashes within approximately 600xa0m, while those higher than 400xa0m can even reach several hundred meters and be initiated by flashes over 1xa0km away. The maximum distances for the downward leaders to attract the UULs with inception heights from 100 to 200xa0m, 200 to 300xa0m, and over 400xa0m are approximately 350xa0m, 450xa0m, and 600xa0m, respectively.

Performance Evaluation for a Lightning Location System Based on Observations of Artificially Triggered Lightning and Natural Lightning Flashes

AbstractPerformance evaluation for the lightning location system (LLS) of the power grid in Guangdong Province, China, was conducted based on observation data of the triggered lightning flashes obtained in Conghua, Guangzhou, during 2007–11 and natural lightning flashes to tall structures obtained in Guangzhou during 2009–11. The results show that the flash detection efficiency and stroke detection efficiency were about 94% (58/62) and 60% (97/162), respectively. The arithmetic mean and median values for location error were estimated to be about 710 and 489 m, respectively, when more than two reporting sensors were involved in the location retrieval (based on 87 samples). After eliminating one obviously abnormal sample, the absolute percentage errors of peak current estimation were within 0.4%–42%, with arithmetic mean and median values of about 16.3% and 19.1%, respectively (based on 21 samples).

A hybrid method based on extreme learning machine and k-nearest neighbor for cloud classification of ground-based visible cloud image

A classification scheme based on extreme learning machine and k nearest neighbor is proposed for cloud classification. In this work, 21 characteristic parameters of texture features, color features and shape features are selected from four different sky conditions (cumulus, stratus, cirrus and clear sky) for classification. The results show that the new scheme using texture features, color features and shape features together can get better performance than using these features alone or any two of them together. When all 21 features are used for classification, the accurate identification rates of cumulus, stratus, cirrus and clear sky are 84.56%, 78.06%, 76.67% and 100.00% respectively, with an average of 84.82%. The proposed model can benefit from the merits of the k-nearest neighbor and the extreme learning machine through its novel structure with high robustness particularly for cloud classification. The simulation results demonstrate that the proposed model in this work is practical for cloud classification and outperforms extreme learning machine (ELM) models, artificial neural network (ANN), k-nearest neighbor (KNN), hybrid method based on KNN and ANN ( KNN - ANN ), and support vector machine (SVM).

An Automated Cirrus Cloud Detection Method for a Ground-Based Cloud Image

Cloud detection is a basic research for achieving cloud-cover state and other cloud characteristics. Because of the influence of sunlight, the brightness of sky background on the ground-based cloud image is usually nonuniform, which increases the difficulty for cirrus cloud detection, and few detection methods perform well for thin cirrus clouds. This paper presents an effective background estimation method to eliminate the influenceofvariableilluminationconditionsandproposesabackgroundsubtractionadaptivethresholdmethod (BSAT) to detect cirrus clouds in visible images for the small field of view and mixed clear‐cloud scenes. The BSAT algorithm consists of red-to-blue band operation, background subtraction, adaptive threshold selection, and binarization. The experimental results show that the BSAT algorithm is robust for all types of cirrus clouds, and the quantitative evaluation results demonstrate that the BSAT algorithm outperforms the fixed threshold (FT) and adaptive threshold (AT) methods in cirrus cloud detection.

Optical progression characteristics of an interesting natural downward bipolar lightning flash

Using high-speed cameras, Lightning Attachment Process Observation Systems, and fast and slow electrical antennas, we documented a downward bipolar lightning flash that contained one first positive stroke with a peak current of 142u2009kA and five subsequent negative strokes hitting on a 90u2009m tall structure on 29 July 2010 in Guangzhou City, China. All the six strokes propagated along the same viewed channel established by the first positive return stroke. The leader which preceded the positive return stroke propagated downward without any branches at a two-dimensional (2-D) speed of 2.5u2009×u2009106u2009m/s. An upward connecting leader with a length of about 80u2009m was observed in response to the downward positive leader. The 10–90% risetimes of the return strokes optical pulses ranged from 2.2u2009µs to 3.2u2009µs, while the widths from the 10% wavefront to the 50% wave tail ranged from 56.5u2009µs to 83.1u2009µs, and the half peak widths ranged from 53.4u2009µs to 81.6u2009µs. All the return strokes exhibited similar speeds, ranging from 1.0u2009×u2009108u2009m/s to 1.3u2009×u2009108u2009m/s. Each of the return strokes was followed by a continuing current stage (CC). The first positive stroke CC lasted more than 150u2009ms, much larger than all the subsequent negative stroke CC, ranging from 13u2009ms to 70u2009ms.

Characteristics of the initial stage and return stroke currents of rocket-triggered lightning flashes in southern China

This study investigates the initial stage (IS) and return stroke (RS) currents of 50 triggered lightning flashes (TLFs) that were conducted in southern China. The IS of the negative TLFs has a longer duration, and larger average current, charge transfer, and action integral than those reported elsewhere, with geometric means (GMs) of 347.9 ms, 132.5 A, 45.1 C, and 10.0 × 103 A2 s, respectively. Two positive TLFs containing no RS have much greater average currents, charge transfers, and action integrals in the IS when compared with the negative TLFs. The RS has a greater peak current (17.2 kA; GM, same to below), charge transfer within 1 ms (1.3 C), and action integral within 1 ms (5.8 × 103 A2 s), and shorter 10% to 90% rise time (0.4 μs) than elsewhere. The peak current is prominently correlated with the rate of rise, charge transfer within 1 ms, and action integral within 1 ms. Furthermore, when the total duration of the RS and any following continuing currents is longer than 40 ms, the peak current, charge transfer within 1 ms, and action integral within 1 ms of the RS are seldom greater than 25 kA, 2.6 C, and 15 × 103 A2 s, respectively. It is indicated that TLFs containing RSs tend to have a longer duration but a smaller charge transfer during the IS than those without RS. The peak current of the RS is weakly correlated with its preceding silence period when there was no channel base current.

Influence of the Canton Tower on the cloud-to-ground lightning in its vicinity

This paper analyzed the influence of the Canton Tower on cloud-to-ground (CG) lightning in its vicinity based on the data obtained by Lightning Location System (LLS) of the Guangdong Power Grid Corporation from 1999 to 2015. The LLS data obtained before (1999-2005) and after (2010-2015) the erection of the Canton Tower were compared. The flash/stroke density showed a significant increase in the immediate vicinity around the Canton Tower within a radius of 1u2009km, and a clear decrease in the annular vicinity with a radius from 1u2009km to 4u2009km. The arithmetic mean and median LLS-inferred peak current of strokes occurred in 1u2009km radius area around the Canton Tower exhibited a significant increase, while that of the strokes occurred beyond 1u2009km did not show a clear change. The percentage of negative flashes (ratio of negative flashes to total flashes) showed a slight increase in the immediate vicinity around the Canton Tower, which could be caused by the upward lightning (often negative) initiated from the tower. Due to the existence of the Canton Tower, the lightning multiplicity presented a decreasing trend to some extent within 4u2009km radius around the tower compared with the arithmetic mean value within 10u2009km radius. It is speculated that the Canton Tower induced lots of upward flashes and attracted some downward flashes around it within several kilometers radius to itself.

Cloud Type Classification of Total-Sky Images Using Duplex Norm-Bounded Sparse Coding

Cloud type classification plays an essential role in ground-based cloud observation. However, it is a challenge to accurately identify the cloud categories that involve a large variety of structural patterns and visual appearances. Image representation and classifier are the crucial factors for cloud classification, though they are individually investigated in the literature. This paper proposes a new cloud type classification method using duplex norm-bounded sparse coding (DNSC), which designs image representation and classifier under the same framework, i.e., norm-bounded sparse coding (NSC). NSC is not only used to encode local descriptors, but also well explored to develop an effective classifier. NSC takes both locality and sparseness into account, and it can be benefit to capture discriminative patterns for image representation and have discriminative power for classifier. Furthermore, NSC has closed-form solution and can be computed efficiently. More specifically, DNSC first extracts local descriptor from an input cloud image, and then DNSC forms a holistic representation leveraging NSC and max-pooling strategy. Finally, a classifier is built on the holistic representation using NSC. The proposed DNSC is evaluated on the total-sky cloud image set, and the experimental results demonstrate that DNSC outperforms the state-of-the-art methods and its accuracy increases by about

Low-frequency E-field Detection Array (LFEDA)—Construction and preliminary results

7%

A Vision-Based Precipitation Sensor for Detection and Classification of Hydrometeors

compared with baselines. In addition, the categorywise performance improvement is particularly pronounced over the complex categories, such as Cirriform and Mixed cloudiness.