scholarly journals Assessing Feature Importance for Short-Term Prediction of Electricity Demand in Medium-Voltage Loads

Energies ◽  
2022 ◽  
Vol 15 (2) ◽  
pp. 549
Author(s):  
Giuliano Armano ◽  
Paolo Attilio Pegoraro
Keyword(s):  
Distribution Networks ◽  
Multilayer Perceptrons ◽  
Smart Meters ◽  
Medium Voltage ◽  
Data Encoding ◽  
Feature Importance ◽  
Importance Analysis ◽  
Hidden Layer ◽  
Electrical Loads ◽  
Short Term Prediction

The design of new monitoring systems for intelligent distribution networks often requires both real-time measurements and pseudomeasurements to be processed. The former are obtained from smart meters, phasor measurement units and smart electronic devices, whereas the latter are predicted using appropriate algorithms—with the typical objective of forecasting the behaviour of power loads and generators. However, depending on the technique used for data encoding, the attempt at making predictions over a period of several days may trigger problems related to the high number of features. To contrast this issue, feature importance analysis becomes a tool of primary importance. This article is aimed at illustrating a technique devised to investigate the importance of features on data deemed relevant for predicting the next hour demand of aggregated, medium-voltage electrical loads. The same technique allows us to inspect the hidden layers of multilayer perceptrons entrusted with making the predictions, since, ultimately, the content of any hidden layer can be seen as an alternative encoding of the input data. The possibility of inspecting hidden layers can give wide support to researchers in a number of relevant tasks, including the appraisal of the generalisation capability reached by a multilayer perceptron and the identification of neurons not relevant for the prediction task.

scholarly journals Detecting the Fault Section in the Distribution Network with Distributed Generators Based on Optimal Placement of Smart Meters

2019 ◽  
Vol 19 (2) ◽  
pp. 28-34 ◽  
Author(s):  
Majid Dashtdar ◽  
Masoud Dashtdar
Keyword(s):  
Fault Location ◽  
Voltage Drop ◽  
Distribution Network ◽  
Distribution Networks ◽  
Optimal Number ◽  
Optimal Placement ◽  
Voltage Distribution ◽  
Smart Meters ◽  
Good Efficiency ◽  
Medium Voltage

AbstractOne of the most important issues in employing distribution networks is detecting the fault location in medium-voltage distribution feeders. Due to the vastness of distribution networks and growing distributed generation (DG) sources in this network, detection is difficult with the common methods. The aim of this paper is to present a method based on voltage distributed meters in a medium-voltage distribution network (by smart meters installed along the feeder) in order to detect the fault location in the presence of DG sources. Due to vastness of distribution network and cost of installing smart meters, it is not economically possible to install meters in all the Buses of the network. That’s why in this article, combination of genetic and locating algorithms and fault-based on voltage drop has been used to suggest a method to optimize the meter locations. In order to evaluate the efficiency of the method suggested, first we determine the optimal number and location of the meters and then we apply the fault that has been simulated in different Buses of the sample network, using PSCAD/EMTDC software. After results analysis, the fault location is estimated by MATLAB. Simulation results show that the fault locating method by optimal number of meters has good efficiency and accuracy in detecting faults in different spots and in different resistance ranges.

scholarly journals A convolution based computational approach towards DNA N6-methyladenine site identification and motif extraction in rice genome

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Chowdhury Rafeed Rahman ◽  
Ruhul Amin ◽  
Swakkhar Shatabda ◽  
Md. Sadrul Islam Toaha
Keyword(s):  
Operating Characteristic ◽  
Adenine Nucleotide ◽  
Characteristic Curve ◽  
Rice Genome ◽  
Plant Genome ◽  
Feature Importance ◽  
Importance Analysis ◽  
Motif Extraction ◽  
Site Identification ◽  
By Products

AbstractDNA N6-methylation (6mA) in Adenine nucleotide is a post replication modification responsible for many biological functions. Automated and accurate computational methods can help to identify 6mA sites in long genomes saving significant time and money. Our study develops a convolutional neural network (CNN) based tool i6mA-CNN capable of identifying 6mA sites in the rice genome. Our model coordinates among multiple types of features such as PseAAC (Pseudo Amino Acid Composition) inspired customized feature vector, multiple one hot representations and dinucleotide physicochemical properties. It achieves auROC (area under Receiver Operating Characteristic curve) score of 0.98 with an overall accuracy of 93.97% using fivefold cross validation on benchmark dataset. Finally, we evaluate our model on three other plant genome 6mA site identification test datasets. Results suggest that our proposed tool is able to generalize its ability of 6mA site identification on plant genomes irrespective of plant species. An algorithm for potential motif extraction and a feature importance analysis procedure are two by products of this research. Web tool for this research can be found at: https://cutt.ly/dgp3QTR.

scholarly journals Real Fault Location in a Distribution Network Using Smart Feeder Meter Data

Energies ◽  
2021 ◽  
Vol 14 (11) ◽  
pp. 3242
Author(s):  
Hamid Mirshekali ◽  
Rahman Dashti ◽  
Karsten Handrup ◽  
Hamid Reza Shaker
Keyword(s):  
Fault Location ◽  
Network Reliability ◽  
Distribution Network ◽  
Estimation Method ◽  
Electrical Energy ◽  
Distribution Networks ◽  
Active Power ◽  
Financial Loss ◽  
Smart Meters ◽  
The Real

Distribution networks transmit electrical energy from an upstream network to customers. Undesirable circumstances such as faults in the distribution networks can cause hazardous conditions, equipment failure, and power outages. Therefore, to avoid financial loss, to maintain customer satisfaction, and network reliability, it is vital to restore the network as fast as possible. In this paper, a new fault location (FL) algorithm that uses the recorded data of smart meters (SMs) and smart feeder meters (SFMs) to locate the actual point of fault, is introduced. The method does not require high-resolution measurements, which is among the main advantages of the method. An impedance-based technique is utilized to detect all possible FL candidates in the distribution network. After the fault occurrence, the protection relay sends a signal to all SFMs, to collect the recorded active power of all connected lines after the fault. The higher value of active power represents the real faulty section due to the high-fault current. The effectiveness of the proposed method was investigated on an IEEE 11-node test feeder in MATLAB SIMULINK 2020b, under several situations, such as different fault resistances, distances, inception angles, and types. In some cases, the algorithm found two or three candidates for FL. In these cases, the section estimation helped to identify the real fault among all candidates. Section estimation method performs well for all simulated cases. The results showed that the proposed method was accurate and was able to precisely detect the real faulty section. To experimentally evaluate the proposed method’s powerfulness, a laboratory test and its simulation were carried out. The algorithm was precisely able to distinguish the real faulty section among all candidates in the experiment. The results revealed the robustness and effectiveness of the proposed method.

scholarly journals AB0652 MACHINE LEARNING TO PREDICT EARLY TNF INHIBITOR USERS IN PATIENTS WITH ANKYLOSING SPONDYLITIS

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 1620.1-1621
Author(s):  
J. Lee ◽  
H. Kim ◽  
S. Y. Kang ◽  
S. Lee ◽  
Y. H. Eun ◽  
...  
Keyword(s):  
Machine Learning ◽  
Ankylosing Spondylitis ◽  
Tnf Inhibitors ◽  
Tnf Inhibitor ◽  
Ann Model ◽  
Learning Models ◽  
Feature Importance ◽  
Importance Analysis ◽  
Baseline Characteristics ◽  
Machine Learning Models

Background:Tumor necrosis factor (TNF) inhibitors are important drugs in treating patients with ankylosing spondylitis (AS). However, they are not used as a first-line treatment for AS. There is an insufficient treatment response to the first-line treatment, non-steroidal anti-inflammatory drugs (NSAIDs), in over 40% of patients. If we can predict who will need TNF inhibitors at an earlier phase, adequate treatment can be provided at an appropriate time and potential damages can be avoided. There is no precise predictive model at present. Recently, various machine learning methods show great performances in predictions using clinical data.Objectives:We aim to generate an artificial neural network (ANN) model to predict early TNF inhibitor users in patients with ankylosing spondylitis.Methods:The baseline demographic and laboratory data of patients who visited Samsung Medical Center rheumatology clinic from Dec. 2003 to Sep. 2018 were analyzed. Patients were divided into two groups: early TNF inhibitor users treated by TNF inhibitors within six months of their follow-up (early-TNF users), and the others (non-early-TNF users). Machine learning models were formulated to predict the early-TNF users using the baseline data. Additionally, feature importance analysis was performed to delineate significant baseline characteristics.Results:The numbers of early-TNF and non-early-TNF users were 90 and 509, respectively. The best performing ANN model utilized 3 hidden layers with 50 hidden nodes each; its performance (area under curve (AUC) = 0.75) was superior to logistic regression model, support vector machine, and random forest model (AUC = 0.72, 0.65, and 0.71, respectively) in predicting early-TNF users. Feature importance analysis revealed erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and height as the top significant baseline characteristics for predicting early-TNF users. Among these characteristics, height was revealed by machine learning models but not by conventional statistical techniques.Conclusion:Our model displayed superior performance in predicting early TNF users compared with logistic regression and other machine learning models. Machine learning can be a vital tool in predicting treatment response in various rheumatologic diseases.Disclosure of Interests:None declared

scholarly journals Machine Learning and GIS Approach for Electrical Load Assessment to Increase Distribution Networks Resilience

Energies ◽  
2021 ◽  
Vol 14 (14) ◽  
pp. 4133
Author(s):  
Alessandro Bosisio ◽  
Matteo Moncecchi ◽  
Andrea Morotti ◽  
Marco Merlo
Keyword(s):  
Machine Learning ◽  
Distribution System ◽  
Low Voltage ◽  
Regression Tree ◽  
Distribution Networks ◽  
Machine Learning Algorithms ◽  
Extreme Weather Events ◽  
Medium Voltage ◽  
Weather Events ◽  
Building Area

Currently, distribution system operators (DSOs) are asked to operate distribution grids, managing the rise of the distributed generators (DGs), the rise of the load correlated to heat pump and e-mobility, etc. Nevertheless, they are asked to minimize investments in new sensors and telecommunication links and, consequently, several nodes of the grid are still not monitored and tele-controlled. At the same time, DSOs are asked to improve the network’s resilience, looking for a reduction in the frequency and impact of power outages caused by extreme weather events. The paper presents a machine learning GIS-based approach to estimate a secondary substation’s load profiles, even in those cases where monitoring sensors are not deployed. For this purpose, a large amount of data from different sources has been collected and integrated to describe secondary substation load profiles adequately. Based on real measurements of some secondary substations (medium-voltage to low-voltage interface) given by Unareti, the DSO of Milan, and georeferenced data gathered from open-source databases, unknown secondary substations load profiles are estimated. Three types of machine learning algorithms, regression tree, boosting, and random forest, as well as geographic information system (GIS) information, such as secondary substation locations, building area, types of occupants, etc., are considered to find the most effective approach.

scholarly journals An Equivalent Circuit for the Evaluation of Cross-Country Fault Currents in Medium Voltage (MV) Distribution Networks

Energies ◽  
2018 ◽  
Vol 11 (8) ◽  
pp. 1929 ◽  
Author(s):  
Fabio Gatta ◽  
Alberto Geri ◽  
Stefano Lauria ◽  
Marco Maccioni
Keyword(s):  
Distribution Networks ◽  
General Purpose ◽  
Simultaneous Occurrence ◽  
Medium Voltage ◽  
Electromagnetic Transient ◽  
High Impedance ◽  
Fault Currents ◽  
Cross Country ◽  
Return Path

A Cross-Country Fault (CCF) is the simultaneous occurrence of a couple of Line-to-Ground Faults (LGFs), affecting different phases of same feeder or of two distinct ones, at different fault locations. CCFs are not uncommon in medium voltage (MV) public distribution networks operated with ungrounded or high-impedance neutral: despite the relatively small value of LGF current that is typical of such networks, CCF currents can be comparable to those that are found in Phase-To-Phase Faults, if the affected feeder(s) consists of cables. This occurs because the faulted cables’ sheaths/screens provide a continuous, relatively low-impedance metallic return path to the fault currents. An accurate evaluation is in order, since the resulting current magnitudes can overheat sheaths/screens, endangering cable joints and other plastic sheaths. Such evaluation, however, requires the modeling of the whole MV network in the phase domain, simulating cable screens and their connections to the primary and secondary substation earth electrodes by suitable computer programs, such as ATP (which is the acronym for alternative transient program) or EMTP (the acronym for electromagnetic transient program), with substantial input data being involved. This paper presents a simplified yet accurate circuit model of the faulted MV network, taking into account the CCF currents’ return path (cable sheaths/screens, ground conductors, and earthing resistances of secondary substations). The proposed CCF model can be implemented in a general-purpose simulation program, and it yields accurate fault currents estimates: for a 20 kV network case study, the comparison with accurate ATP simulations evidences mismatches mostly smaller than 2%, and never exceeding 5%.

scholarly journals On the Use of LoRa Technology for Logic Selectivity in MV Distribution Networks

Energies ◽  
2018 ◽  
Vol 11 (11) ◽  
pp. 3079 ◽  
Author(s):  
Leopoldo Angrisani ◽  
Francesco Bonavolontà ◽  
Annalisa Liccardo ◽  
Rosario Schiano Lo Moriello
Keyword(s):  
Cost Effectiveness ◽  
Long Range ◽  
Distribution Networks ◽  
Fault Isolation ◽  
Optimized Design ◽  
Medium Voltage ◽  
Different Types ◽  
The Times ◽  
The Cost ◽  
Types Of Faults

In this paper, a logic selectivity system based on Long Range (LoRa) technology for the protection of medium-voltage (MV) networks is proposed. The development of relays that communicate with each other using LoRa allows for the combination of the cost-effectiveness and ease of installation of wireless networks with long-range coverage and reliability. The realized demonstrator to assess the proposed system is also presented in the paper; based on different types of faults and different locations, the times needed for clearing a fault and restoring the network were estimated from repeated experiments. The obtained results confirm that, with an optimized design of transmitted packets and of protocol characteristics, LoRa communication grants fault management that meets the criteria of logic selectivity, with fault isolation occurring within the maximum allowed time.

Sign in / Sign up

Export Citation Format

Share Document