Review of machine learning applications to power systems studies

The complexity of electric power networks from generation, transmission and distribution stations in modern times has resulted to generation of big and more complex data that requires more technical and mathematical analysis because it deals with monitoring, supervisory control and data acquisition all in real time. This has necessitated the need for more accurate analysis and predictions in power systems studies especially under transient, uncertainty or emergency conditions without interference of humans. This is necessary so as to minimize errors with the aim targeted towards improving the overall performance and the need to use more technical but very intelligent predictive tools has become very relevant. Machine learning (ML) is a powerful tool which can be utilized to make accurate predictions about the future nature of data based on past experiences. ML algorithms operate by building a model (mathematical or pictorial) from input examples to make data driven predictions or decisions for the future. ML can be used in conjunction with big data to build effective predictive systems or to solve complex data analytic problems. Electricity generation forecasting systems that could predict the amount of power required at a rate close to the electricity consumption have been proposed in several works. This study seeks to review machine learning applications to power system studies. This paper reviewed applications of ML tools in power systems studies.

Method to determine fact and place of commercial losses in distribution networks

The improvement of methods to register the commercial losses in electrical distribution networks, and especially in low voltage networks, is one of the most important tasks for power supply providers. It is rather difficult to correctly register the fact of occurrence of such losses in the network. It is objectively impossible to analyze the state of the networks based on data obtained from various points of the specified network with the required accuracy. In this regard, at present no methods have been developed for remote detection of the fact and determination of the place of commercial losses in distribution networks, that could work in the mode of integration with automated information-measuring system of fiscal electricity metering. To solve this problem a method is to be developed that allows us to establish accurately for practical purposes the volume of commercial losses in the network and determine the place of their occurrence. During the research, methods of electric power networks modeling have been used. The assumption has been made about no flow of capacitive leakage currents to ground in the network, about full compliance of the line parameters with their calculated (nominal) values, as well as the basic laws of electrical engineering science. A unique method is proposed to determine the fact and the place of commercial losses in distribution networks. In contrast to the prototypes, it is based on the analysis of data obtained from metering devices, based on the key laws of electrical engineering and it allows us to get reliable arithmetically rigorous results without using fuzzy logic. The authors have proved theoretically and practically the effectiveness of the proposed solutions, and the possibility of their application. A calculation has been made to determine the place of commercial losses in the network using an example. The proposed method to determine the fact and place of commercial losses in distribution networks of low and medium voltage levels solves the problem of inability to effectively identify the points of occurrence of commercial losses in distribution networks. The reliability of the results obtained is confirmed by mathematical rigor of the method and algorithmic nature of the procedure for analyzing the distribution network.

Modeling Grounding Systems for Electromagnetic Compatibility Analysis

In recent years, the development of smart grids for power distribution and the increasing usage of 5G communication networks have played a large impact on the resilience and reliability of grounding systems. Unexpected electromagnetic coupling between a communication tower and the one used for the electric power networks may pose a threat to the suitable performance of either system as one must assure that electromagnetic compatibility together with unexpected transient issues is within reasonable parameters. This requires wideband modeling of a grounding system, typically carried out using numerical approaches based on the Method of Moments. This modeling is implied in numerous segments to represent the conductors involved and the numerical solution of a double integral for each one of these segments. The modified nodal formulation used to obtain system voltages and branch currents is first solved in the frequency domain, leading to a heavy computational burden and a time-consuming simulation. This chapter briefly reviews the procedure used to model grounding grids and presents some results to illustrate the typical behavior. Afterward, a more complex system comprising a case of electromagnetic coupling is then analyzed to illustrate the impact of nearby grounding grids.

An Assessment of Spatio-Temporal Characteristics of Lightning Occurrence with Elevation, a Case of Uganda

Lightning has received a lot of attention in scientific literature during the recent decade, not only because it is an impressive atmospheric phenomenon but also its associations with severe storms that cause unprecedented damages to agriculture, electric power networks, property, and life. This study assessed the Spatio-temporal characteristics of lightning occurrence with elevation in Uganda using lightning flash and elevation datasets for a period of fifteen years (1998-2013). Datasets used in this study included daily lightning flashes as captured by Lightning Imaging Sensor (LIS) aboard on Tropical Rainfall Measurement Mission (TRMM) satellite and elevation data in form of Digital Elevation Model (DEM) obtained from the Shuttle Radar Topography Mission (SRTM). Spatio-temporal results indicated that ~80% of areas with an elevation that ranges from 800-1200 m above mean sea level (masl) in Uganda had severe lightning occurrences and ~20% of areas with an elevation greater than 1200 m (masl) had severe lightning occurrences. The country received an enhanced number of lighting events with the highest number in 1999. Subsequently, a reduced trend was observed from 2002 to 2007 followed by an increment in the number of lightning events in (2010, 2011, 2012, and 2013). The intensity of the events decreased gradually though two peaks were observed, (1998-2001) and (2010-2013). Furthermore, results indicate escalations in the frequency and duration of lightning events from 60 times in 1998 to approximately 200 times in 2013 and from 1000 microseconds in 1998 to more than 2000 microseconds in 2013. Generally, the country experienced an enhanced increase in lighting occurrences over the study period which therefore calls for urgent actions to combat the root cause and also provide effective measures to reduce the impacts of lightning strikes.

Exploring Electric Transmission Operation with Cognitive Work Analysis Models

Electric power networks are regarded as complex socio-technical systems as they incorporate a wide number of tasks and dynamic data that needs to be monitored continuously. This preliminary study investigated the work environment of electric power transmission and the related control tasks performed by the system operators. These findings shall be incorporated into the design of an ecological interface to improve situation awareness, reduce cognitive workload, and especially support novice operators in training.

An improved technique for power transformer protection using fuzzy logic protective relaying

The three-phase power transformer in the transmission or distribution substations represents one of the essential devices on electric power networks. Losing this devise cause a disconnection of the power utility to a large number of electrical loads. The robust protection system must be designed to protect the device during abnormal operations. A complete protection system for a poly-phase power transformer for one of the Karbala transmission networks (East Karbala substation) is modeled and simulated, adopting a fuzzy logic protective relaying using MATLAB/SIMULINK environment. This study discusses fuzzy logic-based relaying for a power transformer safety, as well as internal faults that are clearly identified. Two principles of operation are used to protect the transformer; differential relay and overcurrent relay. The differential relay is proposed as the unit protection, while the overcurrent is backup protection. The proposed fuzzy logic controller (FLC) is used to detect abnormal operation; it is also modeled to organize the operation between unit and backup protection. The numerical results clarify that the proposed model can perform fast, rigorous, and authoritative protection for the transformer. Also, modeling of the protection mode decreases the complexity of designing various subsystem and combining them in one controller.

Allocation of active power losses to generators in electric power networks

This paper presents a method for allocating active power losses in electric power networks to generators. A technique that uses current distribution factor is used to allocate losses to generator nodes. The core of the allocation scheme is based on graph theory and flows distribution in a network. Losses are only allocated based on the segment of a network that is used for power evacuation. Models of IEEE 14, 39, 57 and 118 test systems in PYPOWER 5.12 were used to test the scheme. It was observed that although the total network losses is minimised when optimal power flow is used for scheduling generation, however that does not translate to minimisation of loss allocation to some generators. The results obtained show that, the scheme can be used to allocate transmission network losses to generation nodes in electric power networks in a fair manner.

System integration of processes of ensuring electric power networks safety under the conditions of impact of meteorological factors

The problem of the influence of meteorological factors on electric power networks continues to be relevant. For the safe fulfillment of their purpose, electric power networks as an element of the electric power system must guarantee the safety, stability and efficiency of the electric power supply process. To solve the above problems, it is necessary to make decisions based on meteorological observations, since the safety of electric power networks depends on the timely forecasting of meteorological factors. The study examines the possibility of system integration of the process of ensuring the safety of electric power networks and the process of monitoring meteorological factors. In order to increase the level of safety of electric power networks, it is proposed to use a geo information system that collects, processes and transfers meteorological data.