Synergistic Method for Machine-Learning-Based Diagnosis and Gradient-Based Optimization of Lightning Damage Mechanisms

Summary form only given. Lightning induce voltages have been seriously reconsidered by power utilities and researchers due to the increasing demand of good quality by customers in power supply. The contribution reviews the standard procedures to evaluate the flashover rate vs basic insulation level (BIL) curves (IEEE 1410 Guide-Distribution Lightning Performance) and describes another methodology used to estimate such curves by means of a statistical program based on the Monte Carlo method for the evaluation of the lightning performances of distribution lines. The program provides the number of lightning events that exceed, for a certain ground flash density, a given value which, in turn, allows estimation of the frequency distribution of lightning-induced voltages for a given line, as well as the flashover rate vs. BIL curves. The computation of the induced voltages is performed using the LIOV code (lightning induced over-voltage), developed within the framework of an international research collaboration involving the University of Bologna Department of Electrical Engineering, the Swiss Federal Institute of Technology (Power Systems Laboratory) and the University of Rome La Sapienza (Department of Electrical Engineering). The influence of the main parameters of interest is examined (ground resistivity, lightning current peak amplitude and rise time, return-stroke velocity) along with the influence of the various critical distance expressions used in the analysis.

As overhead distribution lines and their associated equipments are most prone to insulation failure due to lightning and very much affected in terms of reliability, a study was intended to determine the line lightning performance and to analyze available mitigation options focusing on medium voltage overhead distribution line. This paper describes the simulation studies in determining the lightning performance of the chosen line using commercially available software. A Monte Carlo statistical method is applied for the simulation of lightning activity, while a three-dimensional Electro-geometric model is adopted for the determination of stroke termination. Shielding from the nearby objects is also taken into account. Linear and non-linear tower footing resistance representation, counterpoises, leader propagation flashover model are among standard features used in the simulations. Simulations are carried out to study the effects of critical flashover voltage, tower footing resistance and ground flash density on the line lightning performance. In addition, several mitigation steps to improve the lightning performance are also recommended.

By use of ATP-EMTP software, built the simulation of 220kV double-circuit transmission line. Then discussed the factors that affect the lightning performance of each tower, such as the height of tower, grounding resistance, insulation configurations, grounding wires and so on. The simulation results show that the tower height has a great impact on the lightning performance of double-circuit, so we should minimize tower height considering the actual condition; reduce the ground resistance can improve the lightning impulse withstand level effectively, especially in the low resistance. For tension tower, using an unbalanced insulation is the most effective way to reducing double tripping. For straight-line tower, increasing the piece of glass insulator is also an effective way to reducing double tripping rate. If the conditions allow, a coupling grounding wire can reduce the double-line trip obviously by use of its shunt and coupling function.

The purpose of this research is to analyze and improve the technical approaches to designing overhead power lines with self-supporting insulated conductors during the transition from bare conductors in low- and medium-voltage networks. The applied methods include analytical evaluation of mechanical and climatic loads using the limit state design approach, comparative assessment of conductor sizing methods based on economic current density and economic current intervals, and modeling of lightning impact using simplified expressions for critical lightning current and insulator flashover probability. The results show that enhancing the reliability and electrical safety of overhead lines requires detailed consideration of the predominance of horizontal loads over vertical ones, the influence of wind pressure, conductor and ground wire weight, insulators and mounting equipment on the dimensions of supports and foundations. The study examines the selection of conductor cross-sections in overhead and cable lines, verifying thermal limits and voltage drop for various voltage levels and network configurations. Particular attention is paid to sizing phase conductors of self-supporting insulated cables considering allowable operating temperatures of thermoplastic and cross-linked polyethylene insulation, the annual duration of maximum load, mechanical strength, and voltage drop constraints. The main lightning impact scenarios are analyzed, including shielding failure, direct lightning strokes to the tower with back-flashover, and lightning strokes to the shield wire in mid-span causing flashover of the shield-to-phase air gap. The research confirms the importance of tower height, grounding resistance, and the inductances of the tower and shield wire for evaluating lightning performance. Structural and technological aspects of erecting insulated overhead lines are presented, including standardization of concrete and steel poles, methods of pole footing in various soils, installation of hooks and cross-arms, conductor stringing using running blocks, pilot ropes and pulling grips, and compression joints for phase conductors. The role of nonlinear metal-oxide surge arresters as a primary means of protecting equipment and insulation in the voltage range from 0.38 to 110 kV is substantiated; the requirements for environmental resistance, pollution performance, mechanical strength, and design of polymeric and porcelain-housed arresters are summarized. The conclusions emphasize that the coordinated use of structural design methods, conductor sizing techniques, lightning protection systems and modern erection technologies significantly increases the reliability, operational safety and energy efficiency of overhead lines employing self-supporting insulated conductors.

A self-adaptive gradient-based particle swarm optimization algorithm with dynamic population topology

Modeling and control of V/f controlled induction motor using genetic-ANFIS algorithm

This study describes the design of a 10-kW solar tower for the Adrar region in southwest Algeria. The heliostats were arranged according to an offset radial pattern, safety distances between adjacent heliostats, surface density criteria, and the blocking factor of solar rays. The heliostat field was designed for a tower height of 18 m with an optical efficiency of 69%, and individual heliostat surfaces of 2.25 m². To model the optical behavior of the heliostat field and receiver, a code in FORTRAN was developed. The results show that a reflective surface of 81 m² of the 36 heliostats is necessary to reach the target power of 10 kW. This approach offers a comprehensive method for heliostat field design, optimizing solar radiation collection, and minimizing losses due to shading and blocking. The developed simulation tool also allows for quick evaluation and adjustment of system parameters, facilitating the design process for similar solar tower systems in other regions.

The results of experimental investigation of acoustic emission (AE) source characterization in 2-D frames is reported in this paper. Determination of the AE source location and magnitude from the displacement field is an inverse problem. It is known that humans have remarkable sensory abilities to undo a large number of demanding inverse problems with little effort. Here, nature is modeled by implementing intelligent signal processing algorithms to solve the problems of the AE source location and magnitude determination. The signal processing algorithms implemented are; FFNN trained by a modified back-propagation algorithm, and a linear system called an auto-associative processor (AAP). The standard feature of these algorithms is the use of a set of pre-processed, measured signals to form a system memory. The pre-processed signals used by the AAP algorithm are tutor augmented, while in the case of the FFNN the pre-processed signals are nonlinearly mapped to the training vectors provided by the tutor. The system memory is consequently utilized to process experimental signals to determine some of the AE source properties. [Work partially supported by NSF and ONR.]

In complex industrial processes (CIPs), due to technical and economic limitations, key performance indicators (KPIs), especially the chemical content-related KPIs, are often difficult to measure in real time, which hinders the propagation of advanced process control technologies. This paper presents a soft sensor-based online KPI inference scheme by a state transition algorithm (STA)-optimized adaptive pre-sparse neuro-fuzzy inference system model, called STA-APSNFIS. It introduces a pre-sparse neural network to the traditional adaptive neuro-fuzzy inference system (ANFIS) model to establish an adaptive pre-sparse neuro-fuzzy inference system (APSNFIS) model to alleviate the adverse effects of data redundancy and noise interference in the detectable process monitoring data, which can effectively reduce the complexity of neuro-fuzzy inference system (NFIS) and speed up its convergence. Successively, to avoid being trapped at a local optimum, the STA-based optimization algorithm is adopted to replace the traditional gradient-based optimization approach to achieve an optimal APSNFIS model. Extensive validation and comparative experiments on nonlinear numeric simulation systems, benchmark Tenessee Eastman (TE) process and a real industrial bauxite flotation process demonstrated that the proposed STA-APSNFIS performed favorably against traditional ANFIS model as well as its variants, e.g., PSO-ANFIS, GA-ANFIS, and some other soft sensor-based KPI inference models.

Disaster response preparedness, risk mitigation, and effective post-event response are critical for national security, public health, and economic integrity. For areas impacted by extreme events, timely and accurate impact assessments of lifeline infrastructures are critical. Traditional disaster management and situational awareness platforms are not capable of handling the massive influx of available information from traditional and non-traditional sources. Next-generation analytic assessment approaches, where use of dynamically sourced, heterogeneous data with near real-time availability are integrated into a continually-updated analytics framework. The objective of this research was to develop a foundational multi-formalism modeling platform to integrate novel data sources for use in improving situational awareness after a disaster event. Specifically, to 1) provide an architecture for machine-learning based data fusion various multi-sensor datasets; 2) classify, geolocate, and temporally bin novel data sources for use in spatially-enabled machine-learning based data fusion; and 3) derive confidence-based observation data with machine-learning based prediction to fill spatial and temporal observation gaps. This project developed a reusable cloud-based pipeline architecture for ingesting, storing, and analyzing relevant data, executing a range of deep-learning models, and visualize the information geographically. The pipeline development was informed by previous PNNL investments (StreamKit) and built upon leading-edge technologies consisting of: 1) a raw datastore; 2) a data enrichment pipeline with custom and third-party analytics; 3) enriched data storage; and 4) a custom UI to display the enriched data. Machine learning/deep learning models were selected and tested to classify social media images for relevancy and presence of flood waters. Our custom trained relevance models were run on a validation dataset and the overall accuracy of the relevance classifiers ranged from 65% to 90%. We found that convolutional neural network (CNN) models were both faster and more accurate than more traditional machine learning models. We applied the relevance classifier to remove non-relevant images and then undertook a limited training of CNN models to detect flood on ~1,800 labeled images. The testing accuracy of the flood detection models were ~88%. To improve the geographic location of both social media and oblique aerial images we tested a multi-step workflow built on a pre-trained deep-learning model. To identify similarity features from test and reference scenes, we applied the Xception and Faiss models to a set of test images with inaccurate spatial locations, as well and precisely-located reference scenes from the general area. Matching reference scenes to the test scenes with the closest similarity index had limited success in improving the precision of test scene location and additional methods have been identified for testing. Data fusion between authoritative geographic data (satellite imaging) and non-authoritative crowdsourced data is an underlying theme throughout this research and methods are tested bring these data to a common geographic form through feature matching and transformations. The automated extraction of supporting information for disaster response from oblique aerial images or ground-level social media images is accomplished through semantic image segmentation and transformation. The non-authoritative data are used to validate the authoritative data collections and processing and optionally, t+0 flood models. In addition, the common form geographic data are combined with other static landscape features to develop spatially-explicit training vectors used in deep generative statistical inference models to develop complex non-linear relationships in static and dynamic data and ultimately derive a spatially continuous and probabilistic assessment of flood damage with the event domain.

Quality of single optical coherence tomography angiography (OCTA) images of myopic choroidal neovascularisation (mCNV) is poorer than in averaged images, although obtaining averaged images takes much time. This study evaluated the clinical usefulness of novel denoising process for depicting mCNV. This study included 20 eyes of 20 patients with mCNV. Ten en face images taken in a 3 × 3 mm macular cube were obtained from outer-retina-to-choriocapillaris layer. Three image types were prepared for analysis; single images before and after the denoising process accomplished deep learning (single and denoising groups, respectively) and up to 10 images were averaged (averaging group). Pairwise comparisons showed vessel density, vessel length density, and fractal dimension (FD) were higher; whereas, vessel density index (VDI) was lower in single group than in denoising and averaging groups. Detectable CNV indices, contrast-to-nose ratio, and CNV diagnostic scores were higher in denoising and averaging groups than in single group. No significant differences were detected in VDI, FD, or CNV diagnostic scores between denoising and averaging groups. The denoising process can utilise single OCTA images to provide results comparable to averaged OCTA images, which is clinically useful for shortening examination times with quality similar to averaging.

The expert information processing system was proposed for a traffic signal system at a multimodal intersection with three types of traffic users: public transport (tram), pedestrians, and general traffic. The system considers the dynamics characteristics of the tram after it is detected by the entry detector of the intersection and is based on a set of fuzzy decision-making rules for decision-making on the time parameters adjustment of the traffic signal plan based on the compatibility of traffic direction signals and the importance of traffic requests. For general traffic signals, the solution is to extend or terminate the permissive signal for a certain group of traffic directions controlled by this signal. Decisions for tram signals, depending on the proximity of the planned permissive signal and the importance of the associated request, can affect both the time parameters of the immediate traffic signal plan and change the overall signal sequence and/or their duration. The conditional priority involves taking into account possible conflicts be-tween the decisions from the public transport signals or from public transport and general traffic signals, which are handled by the corresponding production rules proposed in the paper. The traffic light plan is changed iteratively in an acyclic manner, considering the degree of confidence of the decision obtained by a two-level fuzzy inference model based on the Mamdani algorithm, the second level of which is responsible for determining the signal sequence and the first level for their time parameters. The information processing model is implemented and validated using SUMO urban mobility simulation tool for moderate and saturated transport demands for a non-trivial intersection with synthetic traffic demand. The experimental results demonstrate the applicability of the system in conditions of a saturated transport demand, allowing to reduce the negative impact on the general traffic flows without a significant increase in the time losses of tram passengers compared to the adaptive unconditional way of implementing tram traffic light priority.