Energies

Editorial

Jump to: Research, Review

8 pages, 243 KiB

Open AccessEditorial

Machine Learning for Energy Systems Optimization

by Insu Kim, Beopsoo Kim and Denis Sidorov

Energies 2022, 15(11), 4116; https://doi.org/10.3390/en15114116 - 03 Jun 2022

Cited by 7 | Viewed by 3716

Abstract

This editorial overviews the contents of the Special Issue “Machine Learning for Energy Systems 2021” and review the trends in machine learning (ML) techniques for energy system (ES) optimization [...] Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

Research

Jump to: Editorial, Review

13 pages, 3364 KiB

Open AccessArticle

Enhancement of the Technique for Calculation and Assessment of the Condition of Major Insulation of Power Transformers

by Olga Melnikova, Alexandr Nazarychev and Konstantin Suslov

Energies 2022, 15(4), 1572; https://doi.org/10.3390/en15041572 - 21 Feb 2022

Cited by 7 | Viewed by 1845

Abstract

The findings of the analysis of data on the accident rate of power transformers indicate that one of the main causes of their failures is a decrease in the dielectric strength of the insulation. To reduce failures and extend the service life of [...] Read more.

The findings of the analysis of data on the accident rate of power transformers indicate that one of the main causes of their failures is a decrease in the dielectric strength of the insulation. To reduce failures and extend the service life of power transformers in operation, the issue of enhancing the techniques for assessing the condition of their internal insulation becomes relevant. Currently, when selecting the major insulation of transformers, one takes into account the dependency of the dielectric strength of the oil passage on its width. Experts discuss the issues involved in the choice of major insulation while taking into account the effect of the generalized factor being the volume of the oil passage. The solution to that problem largely depends on the study of the statistical characteristics of the dielectric strength of oil passages of different volumes and the effect rated parameters of transformers have on them. The efficiency of the application of such diagnostic characteristics depends on the extent of studies available on them and the establishment of their standardized parameters. The paper proposes a method for estimating the change in the transformer oil volume in stressed oil passages of major insulation of high-voltage power transformers and statistical characteristics of the dielectric strength of these passages while taking into account the effect of the rated values of capacity and voltage of transformers. It is shown that the degree of effect of transformer technical parameters on the statistical characteristics of the dielectric strength of oil passages depends on the quality of transformer oil, which undergoes a change in operating conditions. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Figure 1

16 pages, 22445 KiB

Open AccessArticle

Low-Cost Sensors for Indoor PV Energy Harvesting Estimation Based on Machine Learning

by Bastien Politi, Alain Foucaran and Nicolas Camara

Energies 2022, 15(3), 1144; https://doi.org/10.3390/en15031144 - 03 Feb 2022

Cited by 8 | Viewed by 1850

Abstract

With the number of communicating sensors linked to the Internet of Things (IoT) ecosystem increasing dramatically, well-designed indoor light energy harvesting solutions are needed. A first step in this direction would be to be able to accurately estimate the harvestable energy in a [...] Read more.

With the number of communicating sensors linked to the Internet of Things (IoT) ecosystem increasing dramatically, well-designed indoor light energy harvesting solutions are needed. A first step in this direction would be to be able to accurately estimate the harvestable energy in a specific light environment. However, inside, this energy varies in spectral composition and intensity, depending on the emission source as well as the time of day. These challenging conditions mean that it has become necessary to obtain accurate information about these variations and determine their impact on energy recovery performance. In this context, this manuscript presented a method to apply an innovative energy harvesting estimation method to obtain practical and accurate insight for the design of energy harvesting systems in indoor environments. It used a very low-cost device to obtain spectral information and fed it to supervised machine learning classification methods to recognize light sources. From the recognized light source, a model developed for flexible GaAs solar cells was able to estimate the harvestable energy. To validate this method in real indoor conditions, the estimates were compared to the energy harvested by an energy harvesting prototype. The mean absolute error percentage between estimates and the experimental measurements was less than 5% after more than 2 weeks of observation. This demonstrated the potential of this low-cost estimation system to obtain reliable information to design energetically autonomous devices. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Figure 1

24 pages, 4254 KiB

Open AccessArticle

Towards the Flexible Distribution Networks Design Using the Reliability Performance Metric

by Ilia Shushpanov, Konstantin Suslov, Pavel Ilyushin and Denis N. Sidorov

Energies 2021, 14(19), 6193; https://doi.org/10.3390/en14196193 - 28 Sep 2021

Cited by 20 | Viewed by 1606

Abstract

At present, the entire world is moving towards digitalization, including in the electric power industry. Digitalization is in its heyday and a lot of articles and reports are devoted to this topic. At the same time, the least digitalized of the electrical networks [...] Read more.

At present, the entire world is moving towards digitalization, including in the electric power industry. Digitalization is in its heyday and a lot of articles and reports are devoted to this topic. At the same time, the least digitalized of the electrical networks are distribution networks that account for a very large share in electric power systems. The article proposes a methodology for creating a flexible distribution network based on the use of digital technology. Additionally, we elaborate a methodology with the identification and collection of the necessary information to create digital networks, develop ways to adapt the required equipment, and suggest methods of recognition of some short circuits. Furthermore, we address the issue of reliability of the information obtained from digital devices, develop a technique for arranging the devices to cover the entire network as required to improve the power system protection of electrical power distribution networks. The above measures make it possible to ensure the flexibility of the active distribution network, as well as to adjust the parameters of the actuation of power system protection depending on changes in external conditions and in the event of emergencies. We propose a technique for controlling the distribution network, based on the factoring-in of the type of damage during an emergency in real time, as well as a technique for arranging the measuring devices and the creation of an information and communication network. We provide recommendations for the design and operation of electric power distribution networks with digital network control technology. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Figure 1

25 pages, 9233 KiB

Open AccessArticle

The Selection of the Most Cost-Efficient Distributed Generation Type for a Combined Cooling Heat and Power System Used for Metropolitan Residential Customers

by Jaemin Park, Haesung Jo and Insu Kim

Energies 2021, 14(18), 5606; https://doi.org/10.3390/en14185606 - 07 Sep 2021

Cited by 6 | Viewed by 1831

Abstract

Distributed generation (DG) using renewable energy sources is of widespread interest. For example, modern centralized conventional fossil fuel power generation commonly adds DG using renewable energy resources to the grid. Therefore, in these changes, it is necessary to optimize renewable energy systems to [...] Read more.

Distributed generation (DG) using renewable energy sources is of widespread interest. For example, modern centralized conventional fossil fuel power generation commonly adds DG using renewable energy resources to the grid. Therefore, in these changes, it is necessary to optimize renewable energy systems to increase energy efficiency and reduce emissions. In previous studies, meta-heuristic algorithms were used to optimize DG location and capacity, but different types of DG systems and integrated energy hub conditions were not considered. Determining the most effective DG type for an integrated energy hub is critical. Accordingly, this study presented a methodology for selecting the most cost-efficient DG for metropolitan residential customers of energy hubs. In this paper, we model energy hubs for residential customers and the most cost-efficient DG type using MATLAB and HOMER software, considering microturbine (MT), photovoltaic (PV), wind turbine, and fuel cell (FC) power sources. For this purpose, the energy hub was modeled as a combined cooling heat and power (CCHP) system and selected a specific metropolitan area as a testbed (Atlanta, USA). For practical simulation, the total active power of the Atlanta community was measured by multiplying the average load profile data of residential houses collected by open energy information (OpenEI). The first case study showed that optimal-blast MTs without absorption chillers (AbCs) were the most cost-efficient compared to other optimal-blast DG systems without AbCs. Additional second case studies for optimal and full-blast MTs with AbCs were performed to verify the results for energy consumption, costs, and emissions savings. As a result, full-blast MTs with AbCs comprise the most cost-efficient DG type in the CCHP system for metropolitan residential customers, reducing energy consumption, cost, and emissions. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Graphical abstract

20 pages, 11112 KiB

Open AccessArticle

Environmentally Constrained Optimal Dispatch Method for Combined Cooling, Heating, and Power Systems Using Two-Stage Optimization

by Haesung Jo, Jaemin Park and Insu Kim

Energies 2021, 14(14), 4135; https://doi.org/10.3390/en14144135 - 08 Jul 2021

Cited by 6 | Viewed by 1738

Abstract

The reliance on coal-fired power generation has gradually reduced with the growing interest in the environment and safety, and the environmental effects of power generation are now being considered. However, it can be difficult to provide stable power to end-users while minimizing environmental [...] Read more.

The reliance on coal-fired power generation has gradually reduced with the growing interest in the environment and safety, and the environmental effects of power generation are now being considered. However, it can be difficult to provide stable power to end-users while minimizing environmental pollution by replacing coal-fired systems with combined cooling, heat, and power (CCHP) systems that use natural gas, because CCHP systems have various power output vulnerabilities. Therefore, purchasing power from external electric grids is essential in areas where CCHP systems are built; hence, optimal CCHP controls should also consider energy purchased from external grids. This study proposes a two-stage algorithm to optimally control CCHP systems. In Stage One, the optimal energy mix using the Lagrange multiplier method for state-wide grids from which CCHP systems purchase deficient electricity was calculated. In Stage Two, the purchased volumes from these grids were used as inputs to the proposed optimization algorithm to optimize CCHP systems suitable for metropolitan areas. We used case studies to identify the accurate energy efficiency, costs, and minimal emissions. We chose the Atlanta area to analyze the CCHP system’s impact on energy efficiency, cost variation, and emission savings. Then, we calculated an energy mix suitable for the region for each simulation period. The case study results confirm that deploying an optimized CCHP system can reduce purchased volumes from the grid while reducing total emissions. We also analyzed the impact of the CCHP system on emissions and cost savings. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Figure 1

19 pages, 3385 KiB

Open AccessArticle

Optimal Allocation of Large-Capacity Distributed Generation with the Volt/Var Control Capability Using Particle Swarm Optimization

by Donghyeon Lee, Seungwan Son and Insu Kim

Energies 2021, 14(11), 3112; https://doi.org/10.3390/en14113112 - 26 May 2021

Cited by 14 | Viewed by 2347

Abstract

Widespread interest in environmental issues is growing. Many studies have examined the effect of distributed generation (DG) from renewable energy resources on the electric power grid. For example, various studies efficiently connect growing DG to the current electric power grid. Accordingly, the objective [...] Read more.

Widespread interest in environmental issues is growing. Many studies have examined the effect of distributed generation (DG) from renewable energy resources on the electric power grid. For example, various studies efficiently connect growing DG to the current electric power grid. Accordingly, the objective of this study is to present an algorithm that determines DG location and capacity. For this purpose, this study combines particle swarm optimization (PSO) and the Volt/Var control (VVC) of DG while regulating the voltage magnitude within the allowable variation (e.g., ±5%). For practical optimization, the PSO algorithm is enhanced by applying load profile data (e.g., 24-h data). The objective function (OF) in the proposed PSO method considers voltage variations, line losses, and economic aspects of deploying large-capacity DG (e.g., installation costs) to transmission networks. The case studies validate the proposed method (i.e., optimal allocation of DG with the capability of VVC with PSO) by applying the proposed OF to the PSO that finds the optimal DG capacity and location in various scenarios (e.g., the IEEE 14- and 30-bus test feeders). This study then uses VVC to compare the voltage profile, loss, and installation cost improved by DG to a grid without DG. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Figure 1

15 pages, 1420 KiB

Open AccessArticle

Battery State of Health Estimation with Improved Generalization Using Parallel Layer Extreme Learning Machine

by Ethelbert Ezemobi, Andrea Tonoli and Mario Silvagni

Energies 2021, 14(8), 2243; https://doi.org/10.3390/en14082243 - 16 Apr 2021

Cited by 15 | Viewed by 2342

Abstract

The online estimation of battery state of health (SOH) is crucial to ensure the reliability of the energy supply in electric and hybrid vehicles. An approach for enhancing the generalization of SOH estimation using a parallel layer extreme learning machine (PL-ELM) algorithm is [...] Read more.

The online estimation of battery state of health (SOH) is crucial to ensure the reliability of the energy supply in electric and hybrid vehicles. An approach for enhancing the generalization of SOH estimation using a parallel layer extreme learning machine (PL-ELM) algorithm is analyzed in this paper. The deterministic and stable PL-ELM model is designed to overcome the drift problem that is associated with some conventional machine learning algorithms; hence, extending the application of a single SOH estimation model over a large set of batteries of the same type. The PL-ELM model was trained with selected features that characterize the SOH. These features are acquired as the discrete variation of indicator variables including voltage, state of charge (SOC), and energy releasable by the battery. The model training was performed with an experimental battery dataset collected at room temperature under a constant current load condition at discharge phases. Model validation was performed with a dataset of other batteries of the same type that were aged under a constant load condition. An optimum performance with low error variance was obtained from the model result. The root mean square error (RMSE) of the validated model varies from 0.064% to 0.473%, and the mean absolute error (MAE) error from 0.034% to 0.355% for the battery sets tested. On the basis of performance, the model was compared with a deterministic extreme learning machine (ELM) and an incremental capacity analysis (ICA)-based scheme from the literature. The algorithm was tested on a Texas F28379D microcontroller unit (MCU) board with an average execution speed of 93 μs in real time, and 0.9305% CPU occupation. These results suggest that the model is suitable for online applications. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Figure 1

18 pages, 3528 KiB

Open AccessArticle

The Optimal Allocation of Distributed Generators Considering Fault Current and Levelized Cost of Energy Using the Particle Swarm Optimization Method

by Beopsoo Kim, Nikita Rusetskii, Haesung Jo and Insu Kim

Energies 2021, 14(2), 418; https://doi.org/10.3390/en14020418 - 13 Jan 2021

Cited by 10 | Viewed by 3221

Abstract

The power requirements of grids have risen as artificial intelligence and electric vehicle technologies have been used. Thus, the installation of distributed generators (DGs) has become an essential factor to streamline power grids. The objective of this study is to optimize the capacity [...] Read more.

The power requirements of grids have risen as artificial intelligence and electric vehicle technologies have been used. Thus, the installation of distributed generators (DGs) has become an essential factor to streamline power grids. The objective of this study is to optimize the capacity and location of DGs. For this purpose, an objective function was defined, which takes into account the fault current and the levelized cost of energy, and a modified particle swarm optimization method was applied. Then, we analyzed a case of a single line-to-ground fault with a test feeder (i.e., the IEEE 30 bus system) with no DGs connected, as well as a case where the DGs are optimally connected. The effect of the optimally allocated DGs on the system was analyzed. We discuss an optimal layout method that takes the economic efficiency of the DG installation into account. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Graphical abstract

13 pages, 764 KiB

Open AccessArticle

Probabilistic Forecasting Based Joint Detection and Imputation of Clustered Bad Data in Residential Electricity Loads

by Soyeong Park, Seungwook Yoon, Byungtak Lee, Seokkap Ko and Euiseok Hwang

Energies 2021, 14(1), 165; https://doi.org/10.3390/en14010165 - 30 Dec 2020

Cited by 4 | Viewed by 1775

Abstract

Residential electricity load data can include numerous types of bad data, even clustered bad data, as they that are typically captured by simple measurement instruments. For example, in the case of a time-series of Not-a-Number (NaN) errors, the values before or next to [...] Read more.

Residential electricity load data can include numerous types of bad data, even clustered bad data, as they that are typically captured by simple measurement instruments. For example, in the case of a time-series of Not-a-Number (NaN) errors, the values before or next to a NaN may appear as the sum of actual values during the times of the NaN series. To utilize load data that includes such erroneous data for prediction or data mining analysis, customized detection and imputation should be conducted. This study proposes a new joint detection and imputation method for handling clustered bad data in residential electricity loads. Examples of these data are known invalid data points, such as consecutive NaN or zero values followed by or being ahead of an outlier. The proposed joint detection and imputation scheme first investigates the neighbors of the invalid data points, using probabilistic forecasting techniques. These techniques are implemented by the next valid neighbors to determine whether there is an anomaly or not. Then, adaptive imputations are applied on the basis of the detection, the candidate point should be imputed simultaneously or not. To assess the potential of the newly proposed scheme to characterize the clustered bad data, we analyzed the electricity loads of 354 households. Moreover, joint detection and imputations are conducted to test with the randomly injected synthesized clustered bad data (containing NaNs of various lengths) that is followed by the summation of the actual NaN values. The proposed scheme succeeded in detecting clustered bad data with an accuracy of 95.5% and a false alarm rate of 3.6% for all households in the dataset. Outlier detection-assisted imputation schemes are evaluated for NaNs with optional outliers. Results demonstrate that these schemes improve the overall accuracy significantly compared to schemes without outlier detection. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Figure 1

13 pages, 2552 KiB

Open AccessArticle

Two-Stage Active and Reactive Power Coordinated Optimal Dispatch for Active Distribution Network Considering Load Flexibility

by Yu Zhang, Xiaohui Song, Yong Li, Zilong Zeng, Chenchen Yong, Denis Sidorov and Xia Lv

Energies 2020, 13(22), 5922; https://doi.org/10.3390/en13225922 - 13 Nov 2020

Cited by 12 | Viewed by 1691

Abstract

A high proportion of renewable energy connected to the power grid has caused power quality problems. Voltage-sensitive loads are extremely susceptible to voltage fluctuations, causing power system safety issues and economic losses. Considering the uncertainty factor and the time-varying characteristic, a linearized random [...] Read more.

A high proportion of renewable energy connected to the power grid has caused power quality problems. Voltage-sensitive loads are extremely susceptible to voltage fluctuations, causing power system safety issues and economic losses. Considering the uncertainty factor and the time-varying characteristic, a linearized random ZIP model (constant impedance (Z), constant current (I), and constant power (P)) with time-varying characteristics was proposed. In order to improve the voltage quality of the voltage-sensitive loads in the day-here stage in an active distribution network (ADN), a linearized two-stage active and reactive power coordinated stochastic optimization model was established. The day-ahead active and reactive power coordination optimization was to smooth the large voltage fluctuation and develop a reserve plan to eliminate the unbalanced power caused by the prediction error in the day-here optimization. In the day-here real-time redispatch, the voltage was further improved by the continuous reactive power compensation device. Finally, the simulation results on the IEEE-33 bus system showed that the control strategy could better eliminate the unbalanced power caused by the prediction error and obviously improve the voltage of sensitive loads in the real-time stage on the premise of maintaining economic optimality. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Figure 1

14 pages, 3309 KiB

Open AccessArticle

Predicting Temperature of Permanent Magnet Synchronous Motor Based on Deep Neural Network

by Hai Guo, Qun Ding, Yifan Song, Haoran Tang, Likun Wang and Jingying Zhao

Energies 2020, 13(18), 4782; https://doi.org/10.3390/en13184782 - 14 Sep 2020

Cited by 28 | Viewed by 5524

Abstract

The heat loss and cooling modes of a permanent magnet synchronous motor (PMSM) directly affect the its temperature rise. The accurate evaluation and prediction of stator winding temperature is of great significance to the safety and reliability of PMSMs. In order to study [...] Read more.

The heat loss and cooling modes of a permanent magnet synchronous motor (PMSM) directly affect the its temperature rise. The accurate evaluation and prediction of stator winding temperature is of great significance to the safety and reliability of PMSMs. In order to study the influencing factors of stator winding temperature and prevent motor insulation ageing, insulation burning, permanent magnet demagnetization and other faults caused by high stator winding temperature, we propose a computer model for PMSM temperature prediction. Ambient temperature, coolant temperature, direct-axis voltage, quadrature-axis voltage, motor speed, torque, direct-axis current, quadrature-axis current, permanent magnet surface temperature, stator yoke temperature, and stator tooth temperature are taken as the input, while the stator winding temperature is taken as the output. A deep neural network (DNN) model for PMSM temperature prediction was constructed. The experimental results showed the prediction error of the model (MAE) was 0.1515, the RMSE was 0.2368, the goodness of fit (R²) was 0.9439 and the goodness of fit between the predicted data and the measured data was high. Through comparative experiments, the prediction accuracy of the DNN model proposed in this paper was determined to be better than other models. This model can effectively predict the temperature change of stator winding, provide technical support to temperature early warning systems and ensure safe operation of PMSMs. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Graphical abstract

35 pages, 2405 KiB

Open AccessArticle

A Multi-Agent NILM Architecture for Event Detection and Load Classification

by André Eugenio Lazzaretti, Douglas Paulo Bertrand Renaux, Carlos Raimundo Erig Lima, Bruna Machado Mulinari, Hellen Cristina Ancelmo, Elder Oroski, Fabiana Pöttker, Robson Ribeiro Linhares, Lucas da Silva Nolasco, Lucas Tokarski Lima, Júlio Shigeaki Omori and Rodrigo Braun dos Santos

Energies 2020, 13(17), 4396; https://doi.org/10.3390/en13174396 - 26 Aug 2020

Cited by 10 | Viewed by 2737

Abstract

A multi-agent architecture for a Non-Intrusive Load Monitoring (NILM) solution is presented and evaluated. The underlying rationale for such an architecture is that each agent (load event detection, feature extraction, and classification) outperforms others of the same type in particular scenarios; hence, by [...] Read more.

A multi-agent architecture for a Non-Intrusive Load Monitoring (NILM) solution is presented and evaluated. The underlying rationale for such an architecture is that each agent (load event detection, feature extraction, and classification) outperforms others of the same type in particular scenarios; hence, by combining the expertise of these agents, the system presents an improved performance. Known NILM algorithms, as well as new algorithms, proposed by the authors, were individually evaluated and compared. The proposed architecture considers a NILM system composed of Load Monitoring Modules (LMM) that report to a Center of Operations, required in larger facilities. For the purposed of evaluating and comparing performance, five load event detect agents, five feature extraction agents, and five classification agents were studied so that the best combinations of agents could be implemented in LMMs. To evaluate the proposed system, the COOLL and the LIT-Dataset were used. Performance improvements were detected in all scenarios, with power-ON and power-OFF detection improving up to 13%, while classification accuracy improved up to 9.4%. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Figure 1

Review

Jump to: Editorial, Research

16 pages, 2691 KiB

Open AccessReview

Electric Power System Transformations: A Review of Main Prospects and Challenges

by Nikolai Voropai

Energies 2020, 13(21), 5639; https://doi.org/10.3390/en13215639 - 28 Oct 2020

Cited by 35 | Viewed by 3185

Abstract

The paper deals with the main prospects and challenges of radical transformations of electric power systems (EPSs) with changes in their structure and properties conditioned by wide use of innovative energy-related technologies and digitalization and intellectualization of system operation and control. Structural trends [...] Read more.

The paper deals with the main prospects and challenges of radical transformations of electric power systems (EPSs) with changes in their structure and properties conditioned by wide use of innovative energy-related technologies and digitalization and intellectualization of system operation and control. Structural trends of EPS development are the focus of the analysis. Consideration is given to changes in EPS properties driven by the use of new technologies, to the problems of system flexibility and to its enhancement. EPS “resiliency” and “survivability” notions are subjected to comparison. The main factors favoring the formation of future EPSs to cyber-physical systems are discussed. Objective trends of EPS control and protection system development are under consideration. Full article

(This article belongs to the Special Issue Machine Learning for Energy Systems 2021)

► Show Figures

Graphical abstract

Journal Menu

Journal Browser

Machine Learning for Energy Systems 2021

Share This Special Issue

Special Issue Editor

Special Issue Information

Keywords

Published Papers (14 papers)

Editorial

Research

Review

Further Information

Guidelines

MDPI Initiatives

Follow MDPI