Machine learning based adaptive fault diagnosis considering hosting capacity amendment in active distribution network

Sahu, Sourav Kumar; Roy, Millend; Dutta, Soham; Ghosh, Debomita; Mohanta, Dusmanta Kumar

doi:10.1016/j.epsr.2022.109025

Cited by 18 publications

(10 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Depending on the dataset of real systems in South Australia, the PV system curtailment and its influences on the demand side response were analyzed to increase the HC using the data-driven methods [221]. In [225], ML techniques have been used for fault prediction diagnosis (type and location) on reconfigured IEEE-33 bus ADNs developed in Typhoon HIL's real-time environment. Moreover, three DL algorithms were applied to increase the accuracy during calculating the HC through DSs (IEEE 34, 123-bus feeders) using CYME in [222].…”

Section: B Tools For Data-driven Hc Methodsmentioning

confidence: 99%

A Comprehensive Review on Recent Developments of Hosting Capacity Estimation and Optimization for Active Distribution Networks

Mousa,

Mahmoud,

Lehtonen

2024

IEEE Access

View full text Add to dashboard Cite

Recently, several types of distributed energy resources (DERs) have been developed to reduce the environmental impact and support the global demand for electrical energy. However, the continuous penetration of the DERs into modern power systems (MPSs) may cause several adverse impacts in terms of operation performance indices (PIs) and power quality issues, especially in low-voltage distribution systems (LVDSs). To cope with these serious impacts and achieve optimal control, the hosting capacity (HC) of DERs must be accurately estimated and optimally optimized. However, it requires an extensive communication infrastructure, which is hardly offered without clear financial benefits. In this regard, this article investigates the historical developments of HC definitions as well as the recent developments in terms of operational performance indices, and estimation methods for active distribution networks (ADNs) with the highpenetration level existence of the DERs, energy storage systems (ESSs), electric vehicle (EV) charging systems, sector coupling, hydrogen technologies, and multi-carrier energy systems (MESs) to deal with electrical, thermal, and cooling demands. In this regard, this review article is intended to exhibit an appropriate reference for comprehensive research trends in HC estimation and optimization based on ADNs. Additionally, it involves and covers most current research HC topics in detail compared to other published review articles. Moreover, the authors deliberate the recent approaches for evaluating and improving the HC, especially concerning data-driven methods, with the aid of various software for simulating real systems. Moreover, modern research trends and main factors of MPS operations are deliberated with current energy market developments. Also, prominent challenges, current status, and future aspects are discussed.

show abstract

Section: B Tools For Data-driven Hc Methodsmentioning

confidence: 99%

A Comprehensive Review on Recent Developments of Hosting Capacity Estimation and Optimization for Active Distribution Networks

Mousa,

Mahmoud,

Lehtonen

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The data obtained are voluminous as 2 million samples from the internal signal scope of the simulator is obtained per second. The whole model is explained in detail in [40].…”

Section: Test System and Datasetmentioning

confidence: 99%

A customised artificial neural network for power distribution system fault detection

Bhagwat,

Dutta,

Jadoun

et al. 2024

IET Generation Trans & Dist

View full text Add to dashboard Cite

Machine learned fault detection approaches are being increasingly used for fault detection in distribution grid. However, the performance of the models can be improved by customizing the models. In this regard, a customised artificial neural network (CANN) for fault detection in a distribution grid is proposed in this paper. The proposed work develops a CANN that combines the “up‐pyramid” and “down‐pyramid” model of ANN into a “custom‐pyramid” model. As a result, the same model can be used both for determining the types of fault as well as its location. The data needed to train the model has been taken from a reconfigured IEEE‐33 bus distribution system developed in Typhoon HIL real‐time simulator. Spectral‐kurtosis is utilized for extraction of features of the faulted transient signals which are used as input data to develop the CANN. The result showcases that the reduction of input features reduces computational complexity without compromising its accuracy. The proposed model classifies fault location with an accuracy of 95.43%. The proposed method also identifies fault type with an accuracy of 96.08%. Several test cases have been developed to test the method. The method proved to be able to perform in most of the cases.

show abstract

“…One is the gray-headed fault diagnosis methods that have been used for a long time [4], such as expert systems Petri networks and diagnostic methods knowledge-based analysis. Another category includes pattern recognition diagnostic models and composite fault diagnostic models combining traditional fault diagnostic methods and artificial intelligence algorithms [5].…”

Section: Introductionmentioning

confidence: 99%

Novel anomaly distribution network data feature recognition method based on double convolutional neural

Wang,

Zhu,

Cui

et al. 2024

Eighth International Conference on Energy System, Electricity, and Power (ESEP 2023)

View full text Add to dashboard Cite

Electricity energy is more and more important to the survival and development of human society. The distribution network is the key medium for delivering electricity energy. The auxiliary monitoring system and distribution secondary system are developing more rapidly. However, the current study suffers from the distortion of massive transient recorded waveform data. The efficiency of fault identification and incident handling in distribution networks is seriously affected. The paper proposes to recognize line fault types based on an accurate database of distribution networks. Firstly, to address the difficulty of extracting the feature volume of transient fault data, a fault data feature self-extraction model is established based on a one-dimensional convolutional self-encoder. Local feature self-extraction of various types of measured fault data is realized. Secondly, to address the problem of low accuracy of transient fault waveform recognition, the 1D convolutional self-encoder is used to obtain the error data features of the power grid. The data characteristics of the imported one dimensional convolution neural network model. Finally, a fault identification model based on dual convolutional neural network is established. The parameters of the convolutional model are adjusted to realize the accurate identification of transient error data types. The results of this paper show that the dual convolutional neural network model accurately recognizes the measured failure data. The recognition accuracy has been improved.

show abstract

Machine learning based adaptive fault diagnosis considering hosting capacity amendment in active distribution network

Cited by 18 publications

References 26 publications

A Comprehensive Review on Recent Developments of Hosting Capacity Estimation and Optimization for Active Distribution Networks

A Comprehensive Review on Recent Developments of Hosting Capacity Estimation and Optimization for Active Distribution Networks

A customised artificial neural network for power distribution system fault detection

Novel anomaly distribution network data feature recognition method based on double convolutional neural

Contact Info

Product

Resources

About