Conference PaperPDF Available

PV Advancements & Challenges: Forecasting Techniques, Real Applications, and Grid Integration for a Sustainable Energy Future

June 2023

June 2023

DOI:10.1109/EEEIC/ICPSEurope57605.2023.10194796

Conference: 2023 IEEE International Conference on Environment and Electrical Engineering and 2023 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe)

Authors:

Michał Jasiński

Wroclaw University of Science and Technology

Show all 5 authorsHide

Content uploaded by Radomir Gono

Content may be subject to copyright.

PV Advancements & Challenges: Forecasting

Techniques, Real Applications, and Grid Integration

for a Sustainable Energy Future

Michał Jasiński

Wrocław University of Science and

Technology & VSB Technical

University of Ostrava

Wrocław Poland, Ostrava Czech

Republic

0000-0002-0983-2562

Luigi Martirano

Department of Astronautical, Electrical

and Energy Engineering (DIAEE),

Sapienza University of Rome

Rome, Italy

0000-0003-0784-265X

Zbigniew Leonowicz

Wrocław University of Science and

Technology & VSB Technical

University of Ostrava

Wrocław Poland, Ostrava Czech

Republic

0000-0002-2388-3710

Radomir Gono

Wrocław University of Science and

Technology & VSB Technical

University of Ostrava

Wrocław Poland, Ostrava Czech

Republic

0000-0003-1125-3305

Jan Jasiński

School No. 97 in Wroclaw

Wrocław, Poland

0009-0000-2962-8123

Abstract— The rapid increase in solar power generation

requires accurate forecasting of photovoltaic (PV) energy

production for effective grid integration and energy market

participation. This paper presents state-of-the-art forecasting

techniques and models, including statistical and time series

models, machine learning models, deep learning models,

probabilistic forecasting models, and data-driven feature

engineering and selection. In addition, applications of PV energy

production forecasting in grid integration and energy market

operations such as load balancing, generation planning,

transmission and distribution planning, energy storage

optimisation, bidding strategies, and renewable energy

certificate trading are presented. Furthermore, challenges

persist in PV power generation forecasting are presented,

including data quality and availability, model uncertainty,

forecast horizon and time resolution, and integration of multiple

data sources and models. Finally, this article focusses on

providing insight into the current PV energy production

forecasting landscape and guiding future research and

development efforts to address these challenges, enhance

forecasting capabilities, and facilitate the global transition to a

more sustainable and low-carbon energy future.

Keywords—photovoltaic energy production, forecasting

techniques, grid integration, energy markets, renewable energy

I. INTRODUCTION TO PHOTOVOLTAIC ENERGY PRODUCTION

AND IMPORTANCE OF ACCURATE FORECASTING

The increasing deployment of photovoltaic (PV) systems

around the world is driven by the need for clean and

sustainable energy sources to combat climate change and

reduce dependence on fossil fuels [1]. As PV systems become

more prevalent, accurately forecasting their energy production

is essential for effective grid management, participation in the

energy market, and system optimisation. This section provides

an introduction to photovoltaic energy production and

highlights the importance of accurate forecasting in the

context of the integration of renewable energy and grid

stability [2]. Accurate forecasting of photovoltaic energy

production plays a vital role in several areas:

• Grid management: As the share of photovoltaic energy

in the grid increases, accurate forecasts become

essential to maintain grid stability and ensure the

balance between electricity supply and demand. This

helps grid operators make informed decisions about

power generation scheduling, reserve management,

and transmission capacity allocation [3].

• Energy market participation: In liberalised energy

markets, accurate forecasts of PV energy production

are crucial for market participants, including power

producers and traders, to optimise their bidding

strategies, minimise imbalance penalties, and

maximise revenues [4].

• System optimisation: Accurate forecasting enables

owners and operators of photovoltaic systems to

optimise the operation and maintenance of their

systems, plan energy storage and demand response

strategies, and improve the overall efficiency and

reliability of the system [5].

• Integration with other renewable energy sources:

Accurate forecasting of photovoltaic energy

production, in combination with forecasts for other

renewable energy sources such as wind and hydro, can

facilitate the efficient integration of these sources into

the energy mix, improving the resilience and

sustainability of the overall energy system [6].

In addition, a significant number of research studies have

investigated the issue of PV forecasting. Based on the

SCOPUS database, the following observation can be made

(information was presented using data indicated in 16 May

2023 in the Scopus database [7]):

• There are 4775 documents that have been represented

by key words.

• Since 2014 it can be noticing the huge increase in

interest in the topic (please see Fig. 1)

2023 IEEE International Conference on Environment and Electrical Engineering and 2023 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe) | 979-8-3503-4743-2/23/$31.00 ©2023 IEEE | DOI: 10.1109/EEEIC/ICPSEUROPE57605.2023.10194796

Authorized licensed use limited to: Technical University of Ostrava. Downloaded on December 28,2023 at 22:46:21 UTC from IEEE Xplore. Restrictions apply.

• The main interest of the researcher is indicated in

counties like China, the United States, India, Japan,

and Italy (Figure 2)

In conclusion, accurate forecasting of PV energy

production is a critical aspect of managing the integration of

photovoltaic systems into energy systems and markets. The

following sections of this paper will dive into the key factors

that influence the forecasting of PV energy production, the

various techniques and models used in forecasting, and the

applications and challenges associated with forecasting in grid

integration and energy markets.

II. S

OLAR

RRADIANCE AND

EATHER

ATA

ACTORS

NFLUENCING

NERGY

RODUCTION

ORECASTING

A. Overview

Solar irradiance and weather conditions have a significant

role in determining the energy performance of photovoltaic

systems, making them key factors in forecasting PV energy

production. This section discusses the key solar and weather

factors that affect PV energy production and the data sources

used to estimate them in forecasting models.

B. Solar Irradiance

Solar irradiance, the power of sunlight per unit area, is the

primary driver of PV energy production. The three main

components of solar irradiance that influence the energy

output of a photovoltaic system are [8]:

• Global Horizontal Irradiance (GHI): The total amount

of solar radiation received on a horizontal surface,

including both direct and diffuse components.

• Direct Normal Irradiance (DNI): The amount of solar

radiation received in a direct beam, perpendicular to

the sun's rays, and unobstructed by clouds or other

atmospheric factors.

• Diffuse horizontal irradiance (DHI): The portion of

solar radiation received on a horizontal surface after

being scattered by the atmosphere.

C. 2.3 Weather Data and Influencing Factors

Weather conditions can significantly affect the

performance of photovoltaic systems, making them an

essential consideration in forecasting PV energy production.

Key weather factors include[9]:

• Cloud cover: Clouds can block and diffuse sunlight,

reducing the availability of direct solar radiation and

leading to fluctuations in PV energy output.

• Temperature: The performance of the PV system is

typically inversely related to temperature; as the

temperature increases, the efficiency of the solar cells

decreases, reducing the overall energy output.

• Humidity: High levels of humidity can lead to

increased cloud cover and atmospheric scattering,

affecting the amount of solar radiation that enters the

photovoltaic system.

• Snow and dust: The accumulation of snow or dust on

solar panels can reduce their efficiency and energy

output by obstructing sunlight.

Fig. 1. Documents by year that can be found in the Scopus database [7].

Fig. 2. Documents by country or territory that can be found in the Scopus

database [7].

D. Data Sources for Solar Irradiance and Weather

Estimation

Various data sources can be used to estimate solar

irradiance and weather conditions for forecasting PV energy

production, including [10]:

• Ground-based measurements: Ground-based sensors,

such as pyranometers and pyrheliometers, provide

local measurements of solar irradiance and weather

data. However, their spatial coverage is limited.

• Satellite data: Satellite-based remote sensing offers a

broader spatial and temporal coverage of solar

irradiance and weather data. Several satellite-derived

datasets, such as NASA's POWER, ESA's CM SAF,

and NOAA's NSRDB, provide historical and near-

real-time information on solar irradiance and weather

conditions.

• Numerical weather prediction (NWP) models: NWP

models, such as ECMWF, GFS, and WRF, offer

weather forecasts at various spatial and temporal

resolutions, providing essential input for forecasting

PV energy production.

E. 2.5 Conclusion

Solar irradiance and weather data are critical factors that

affect the forecasting of PV energy production. Accurate

estimation of these factors, utilizing ground measurements,

satellite data, and numerical weather forecasting models, is

essential to developing reliable and robust forecasts. In the

sections below, this paper will explore the various techniques

and models used in PV energy production forecasting, as well

as their applications and challenges in grid integration and

energy markets.

Authorized licensed use limited to: Technical University of Ostrava. Downloaded on December 28,2023 at 22:46:21 UTC from IEEE Xplore. Restrictions apply.

III. FORECASTING TECHNIQUES AND MODELS FOR

PHOTOVOLTAIC ENERGY PRODUCTION

A. Overview

Numerous forecasting techniques and models have been

established for predicting PV energy production, ranging from

simple statistical methods to advanced machine learning

algorithms. This section reviews the most popular techniques

and models used in PV energy production forecasting,

highlighting their key features, advantages and limitations.

B. Statistical and Time Series Models

Statistical and time series models are widely used for

short-term PV energy production forecasting, primarily due to

their simplicity and ease of implementation. Key models in

this category include[11]:

• Persistence models: Persistence models assume that

the current PV energy output will persist for a short

period in the future. Although these models are simple

and easy to implement, their accuracy tends to

decrease rapidly with increasing forecast horizons.

• Autoregressive Integrated Moving Average (ARIMA)

models: ARIMA models are a popular time series

analysis technique that captures the temporal

dependencies in the data. These models are relatively

easy to implement but may not be suitable for

capturing complex relationships between solar

irradiance, weather conditions, and PV energy

production.

C. Machine Learning Models

Machine learning models have gained popularity in PV

energy production forecasting due to their ability to learn

complex relationships between input variables and energy

output. Key models in this category include[12]:

• Artificial neural networks (ANNs): ANNs are inspired

by the structure and function of biological neural

networks and can model complex, nonlinear

relationships between input variables and PV energy

production. Although ANNs can achieve high

accuracy, they may require large amounts of data for

training and can be computationally intensive.

• Support Vector Machines (SVMs): SVMs are a

powerful supervised learning technique that can be

used for regression tasks, such as PV energy

production forecasting. SVMs are relatively robust to

noise and outliers but may require careful selection of

hyperparameters and kernel functions.

• Random Forests (RFs): RFs are an ensemble learning

method that combines multiple decision trees to

improve prediction accuracy and reduce overfitting.

RFs are known for their robustness and ability to

handle large data sets with multiple input variables,

making them suitable for forecasting PV energy

production.

D. Probabilistic Forecasting Models

Probabilistic forecasting models provide not only point

forecasts, but also uncertainty estimates, which can be crucial

for decision-making in grid management and energy markets.

Key approaches in this category include[13]:

• Quantile regression: Quantile regression models

estimate the conditional quantiles of the response

variable, providing a more comprehensive view of the

forecast distribution and associated uncertainties.

• Bayesian methods: Bayesian methods combine prior

knowledge with observed data to generate

probabilistic forecasts. These methods can provide

uncertainty estimates and accommodate different

sources of information, making them suitable for

forecasting PV energy production.

E. Data-Driven Feature Engineering and Selection

Feature engineering and selection play a vital role in the

performance of PV energy production forecasting models, as

they can help capture the most relevant information from solar

irradiance and weather data[14]. Key approaches in this

category include the following:

• Domain-specific feature engineering: Creating new

features based on domain knowledge, such as

aggregating solar irradiance data over different time

scales or calculating clear-sky indices, can help

improve the predictive power of forecasting models.

• Data-driven feature selection: Techniques such as

recursive feature elimination, LASSO regression, and

mutual information-based methods can help identify

the most relevant features for PV energy production

forecasting, reducing model complexity, and

improving prediction accuracy.

F. Hybrid and Ensemble Models

Hybrid and ensemble models combine multiple

forecasting techniques or models to improve prediction

accuracy and reduce the impact of individual model

limitations. Key approaches in this category include[15]:

• Model stacking: Model stacking involves training

multiple base models and using their predictions as

input for a meta-model that generates the final forecast.

This approach can help capture the strengths of

different models and improve overall prediction

accuracy.

• Weighted averaging: Weighted averaging combines

the predictions of multiple models by assigning

weights to each model based on its performance. This

approach can help reduce the impact of poor-

performing models and improve overall forecast

accuracy.

G. Conclusion

Various forecasting techniques and models are available

for PV energy production forecasting, each with its strengths

and limitations. Choosing the most suitable model depends on

factors such as the forecast horizon, data availability, and the

desired level of accuracy. In the next section, this paper will

discuss the applications and challenges of PV energy

production forecasting in grid integration and energy markets.

Authorized licensed use limited to: Technical University of Ostrava. Downloaded on December 28,2023 at 22:46:21 UTC from IEEE Xplore. Restrictions apply.

IV. APPLICATIONS AND CHALLENGES OF PV ENERGY

PRODUCTION FORECASTING IN GRID INTEGRATION AND

ENERGY MARKETS

A. Overview

Accurate PV energy production forecasting is crucial for

the successful integration of solar power into the energy grid

and for effective participation in energy markets. This section

discusses the various applications of PV energy production

forecasting and the challenges faced in grid integration and

energy market operations.

B. 4.2 Grid Integration Applications

Reliable photovoltaic energy production forecasting plays

a key role in several aspects of grid integration, including [16]:

• Load balancing and generation scheduling: Accurate

PV energy production forecasts help grid operators

balance supply and demand by effectively scheduling

generation resources, ensuring grid stability and

minimising operational costs.

• Transmission and distribution planning: The forecasts

for PV energy production inform the operators of the

transmission and distribution systems of the expected

solar generation, allowing them to manage the network

more efficiently and accommodate higher levels of

solar penetration.

• Energy storage optimization: Forecasting solar energy

production helps optimise the operation of energy

storage systems, determining the optimal charging and

discharging schedules to maximise the value of stored

energy.

C. Energy Market Applications

PV energy production forecasting is essential for solar

power producers participating in energy markets, as it can[17]:

• Improve bidding strategies: Accurate forecasts of

photovoltaic energy production enable solar power

producers to submit more competitive offers in energy

markets, increasing their revenues and market share.

• Reduce imbalance penalties: Solar power producers

that deviate from their scheduled generation may be

subject to financial penalties. Accurate PV energy

production forecasts help minimise these penalties by

reducing the discrepancies between the scheduled and

actual generation.

• Facilitate the trading of renewable energy certificates

(REC). Accurate forecasting of solar energy

production can improve the transparency and

efficiency of REC trading, providing a better

understanding of the renewable energy generation

landscape.

D. Challenges in PV Energy Production Forecasting

Despite advances in forecasting techniques and models,

several challenges remain in forecasting photovoltaic energy

production, including [18]:

• Data quality and availability: Accurate solar irradiance

and weather data are critical for forecasting PV energy

production. However, data quality and availability

problems, such as missing values, measurement errors,

and spatial and temporal resolution limitations, can

affect the accuracy of forecasting models.

• Model uncertainty: All forecasting models are subject

to inherent uncertainties, which can stem from model

assumptions, parameter estimation errors, or

limitations in capturing complex relationships between

input variables and PV energy production.

• Forecast horizon and temporal resolution: Forecast

accuracy tends to decrease as the forecast horizon and

temporal resolution increase, making it challenging to

provide accurate long-term and high-resolution

forecasts.

Integration of multiple data sources and models:

Combining data from different sources and integrating

multiple forecasting models can be challenging due to

differences in data formats, spatial and temporal resolutions,

and model output.

E. Conclusion

PV energy production forecasting is essential for grid

integration and energy market operations, with applications

ranging from load balancing and generation scheduling to

bidding strategies and REC trading. While significant

progress has been made in developing accurate forecasting

models and techniques, challenges related to data quality and

availability, model uncertainty, forecast horizon, and data and

model integration persist. Future research and development

efforts should focus on addressing these challenges and

improving the overall reliability and accuracy of PV energy

production forecasts.

V. CONCLUSION

In this article, we discussed the importance of forecasting

PV power generation for successful integration of solar power

into the grid and effective participation in energy markets. We

provided an overview of various forecasting techniques and

models, including statistical and time series models, machine

learning models, deep learning models, probabilistic

forecasting models, and data-driven feature engineering and

selection. In addition, we explored the applications and

challenges of forecasting solar photovoltaic production in grid

integration and energy market operations. As solar energy

continues to play an increasingly important role in the global

energy mix, the need for accurate forecasting of PV energy

production becomes even more critical. Although significant

progress has been made in developing and refining forecasting

techniques and models, there are still challenges to be

addressed, such as data quality and availability, model

uncertainty, forecast horizon and temporal resolution, and

integration of multiple data sources and models. Future R&D

efforts should focus on addressing these challenges,

improving the overall reliability and accuracy of PV power

generation forecasts, and leveraging advances in data

acquisition, processing and analysis to enhance forecasting

capabilities. In this way, we can facilitate the effective

integration of solar energy into the power grid and energy

markets, supporting the global transition to a more sustainable

and low-carbon energy future.

Authorized licensed use limited to: Technical University of Ostrava. Downloaded on December 28,2023 at 22:46:21 UTC from IEEE Xplore. Restrictions apply.

Fig. 3. PV system in children point of view – own elaboration of Jan

Jasiński – co-author of paper.

Furthermore, promotion of the topic of PV forecasting

should be included in the next generation. Fig. 3 is an example

of work done by school pupil who shows how the PV

installation should look like in any green area that could be

use to put RES energy. As the view of new installation become

more and more popular not only on roofs the presented PV

system is independently located in green area.

REFERENCES

[1] P. N. L. Mohamad Radzi, M. N. Akhter, S. Mekhilef, and N. Mohamed

Shah, “Review on the Application of Photovoltaic Forecasting Using

Machine Learning for Very Short- to Long-Term Forecasting,”

Sustainability, vol. 15, no. 4, p. 2942, Feb. 2023, doi:

10.3390/su15042942.

[2] J. Faraji, H. Hashemi-Dezaki, and A. Ketabi, “Stochastic operation and

scheduling of energy hub considering renewable energy sources’

uncertainty and N-1 contingency,” Sustain. Cities Soc., vol. 65, 2021,

doi: 10.1016/j.scs.2020.102578.

[3] M. Bilal, F. Ahmad, and M. Rizwan, “Techno-economic assessment of

grid and renewable powered electric vehicle charging stations in India

using a modified metaheuristic technique,” Energy Convers. Manag.,

vol. 284, p. 116995, May 2023, doi: 10.1016/j.enconman.2023.116995.

[4] E. Chomać-Pierzecka, H. Gąsiński, J. Rogozińska-Mitrut, D. Soboń,

and S. Zupok, “Review of Selected Aspects of Wind Energy Market

Development in Poland and Lithuania in the Face of Current

Challenges,” Energies, vol. 16, no. 1, p. 473, Jan. 2023, doi:

10.3390/en16010473.

[5] M. Ali et al., “A Comparison of Grid-Connected Local Hospital Loads

with Typical Backup Systems and Renewable Energy System Based

Ad Hoc Microgrids for Enhancing the Resilience of the System,”

Energies, vol. 16, no. 4, p. 1918, Feb. 2023, doi: 10.3390/en16041918.

[6] Y. Li, R. Wang, Y. Li, M. Zhang, and C. Long, “Wind power

forecasting considering data privacy protection: A federated deep

reinforcement learning approach,” Appl. Energy, vol. 329, p. 120291,

Jan. 2023, doi: 10.1016/j.apenergy.2022.120291.

[7] Scopus. Database, “Analyze search results TITLE-ABS-KEY ( pv

AND forecasting ).”

https://www.scopus.com/term/analyzer.uri?sid=2c2c2152b975ad27a3

096b382166199a&origin=resultslist&src=s&s=TITLE-ABS-

KEY%28PV+Forecasting%29&sort=plf-

f&sdt=b&sot=b&sl=29&count=4775&analyzeResults=Analyze+resul

ts&txGid=25a4eac90e19cb9fe0179952cc6caaec.

[8] M. Jasiński et al., “Influence of Measurement Aggregation Algorithms

on Power Quality Assessment and Correlation Analysis in Electrical

Power Network with PV Power Plant,” Energies, vol. 12, no. 18, p.

3547, Sep. 2019, doi: 10.3390/en12183547.

[9] P. P. Kumar, V. Suresh, M. Jasinski, and Z. Leonowicz, “Off-Grid

Rural Electrification in India Using Renewable Energy Resources and

Different Battery Technologies with a Dynamic Differential Annealed

Optimization,” Energies, vol. 14, no. 18, p. 5866, Sep. 2021, doi:

10.3390/en14185866.

[10] M. Tamil Selvan and K. Malar, “Challenges faced for installation of

emergency restoration system (ERS) in the major natural disaster hit

high voltage transmission lines in India,” Disaster Adv., vol. 12, no. 6,

pp. 1–8, 2019, [Online]. Available:

https://www.scopus.com/inward/record.uri?eid=2-s2.0-

85068618265&partnerID=40&md5=cbd69eea6af5de55e834977a086

3307b.

[11] P. K. Jain, W. Quamer, and R. Pamula, “Electricity Consumption

Forecasting Using Time Series Analysis BT - Advances in Computing

and Data Sciences,” 2018, pp. 327–335.

[12] E. Shahrabi, S. M. Hakimi, A. Hasankhani, G. Derakhshan, and B.

Abdi, “Developing optimal energy management of energy hub in the

presence of stochastic renewable energy resources,” Sustain. Energy,

Grids Networks, vol. 26, 2021, doi: 10.1016/j.segan.2020.100428.

[13] O. Lindberg, D. Lingfors, J. Arnqvist, D. van der Meer, and J.

Munkhammar, “Day-ahead probabilistic forecasting at a co-located

wind and solar power park in Sweden: Trading and forecast

verification,” Adv. Appl. Energy, vol. 9, p. 100120, Feb. 2023, doi:

10.1016/j.adapen.2022.100120.

[14] G. Ramesh, J. Logeshwaran, T. Kiruthiga, and J. Lloret, “Prediction of

Energy Production Level in Large PV Plants through AUTO-Encoder

Based Neural-Network (AUTO-NN) with Restricted Boltzmann

Feature Extraction,” Futur. Internet, vol. 15, no. 2, p. 46, Jan. 2023,

doi: 10.3390/fi15020046.

[15] M. Yu et al., “Short-term photovoltaic power point-interval forecasting

based on double-layer decomposition and WOA-BiLSTM-Attention

and considering weather classification,” Energy, p. 127348, Apr. 2023,

doi: 10.1016/j.energy.2023.127348.

[16] Blaabjerg, Y. Yang, K. A. Kim, and J. Rodriguez, “Power Electronics

Technology for Large-Scale Renewable Energy Generation,” Proc.

IEEE, vol. 111, no. 4, pp. 335–355, Apr. 2023, doi:

10.1109/JPROC.2023.3253165.

[17] Ø. S. Klyve, M. M. Nygård, H. N. Riise, J. Fagerström, and E. S.

Marstein, “The value of forecasts for PV power plants operating in the

past, present and future Scandinavian energy markets,” Sol. Energy,

vol. 255, pp. 208–221, May 2023, doi: 10.1016/j.solener.2023.03.044.

[18] T. Polasek and M. Čadík, “Predicting photovoltaic power production

using high-uncertainty weather forecasts,” Appl. Energy, vol. 339, p.

120989, Jun. 2023, doi: 10.1016/j.apenergy.2023.120989.

Authorized licensed use limited to: Technical University of Ostrava. Downloaded on December 28,2023 at 22:46:21 UTC from IEEE Xplore. Restrictions apply.

ResearchGate has not been able to resolve any citations for this publication.

Power Electronics Technology for Large-Scale Renewable Energy Generation

Article

Full-text available

Mar 2023

Power electronics is the enabling technology for the grid-integration of large-scale renewable energy generation, which provides high controllability and flexibility to energy generation, conversion, transmission, and utilization. However, power electronics also brings several challenges to the conventional power grids, e.g., reducing the system inertia. Advanced control strategies have been developed to enhance the energy conversion process. In this article, the grid-integration structure and control for renewable energy are discussed with the focus on large-scale wind, solar photovoltaic, and energy storage systems. Future research and development trends for these technologies are also presented. ABSTRACT | Grid-integration of renewable energy requires efficient and reliable power conversion stages, particularly with an increasing demand of high controllability and flexibility seen from the grid side. Underpinned by advanced control and information technologies, power electronics converters play an essential role in large-scale renewable energy generation. However, the use of power converters has also exposed several challenges in conventional power grids, e.g., reducing the system inertia. In this article, the grid integration using power electronics is presented for large-scale renewable energy generation. Technical issues and requirements are discussed with a special focus on grid-connected wind, solar photovoltaic, and energy storage systems. Additionally, the core of the energy generation and conversion-control for individual power converters (e.g., general current control) as well as for the system level (e.g., coordinated operation of large-scale energy systems) is briefly discussed. Future research perspectives are then presented that further advance large-scale renewable energy generation technologies through incorporating more power electronics systems.

A Comparison of Grid-Connected Local Hospital Loads with Typical Backup Systems and Renewable Energy System Based Ad Hoc Microgrids for Enhancing the Resilience of the System

Article

Full-text available

Feb 2023

Extreme weather conditions and natural disasters (ND) are the main causes of power outages in the electric grid. It is necessary to strengthen the electrical power system’s resilience during these catastrophic occurrences, and microgrids may be seen as the best way to achieve this goal. In this paper, two different energy system scenarios were proposed for increasing the resiliency of the electric power system during random outages. In the first scenario, a diesel generator (DG) was used to deliver energy to key loads during grid disruptions, in conjunction with a utility electric grid (UEG) and local electric load (ELL). A grid-connected ad hoc microgrid (MG) with a photovoltaic (PV) system, a battery energy storage (BES) system, and local electric loads made up the second scenario. The PV system and the BES system were used to supply the key loads with electricity during the outage. The major aim of this research was to compare the two resilient-based systems from the perspectives of technology, economics, and the environment. Given that it requires greater resilience than the other loads during severe weather, a hospital load on Indonesia’s Lombok Island was chosen as the critical load. The objective function considers the system’s predefined constraints to reduce the overall net present cost (NPC) and the cost of energy in order to maximize the system resilience (COE). The Optimization of Multiple Energy Resources (HOMER) Grid simulated a 3-day outage in August 2021, and the results demonstrated that the resiliency enhancement for both scenarios was nearly identical. The first scenario resulted in fewer CO2 emissions; however, the second scenario delivered lower operating costs and COE. The simulation’s findings showed that system 1 created an annual emission of 216.902 kg/yr while system 2 only produced an emission of 63.292 kg/yr. This study shows that since RES-based MGs don’t burn fossil fuels to generate power, they are more environmentally friendly resources.

Review on the Application of Photovoltaic Forecasting Using Machine Learning for Very Short- to Long-Term Forecasting

Article

Full-text available

Feb 2023

Advancements in renewable energy technology have significantly reduced the consumer dependence on conventional energy sources for power generation. Solar energy has proven to be a sustainable source of power generation compared to other renewable energy sources. The performance of a photovoltaic (PV) system is highly dependent on the amount of solar penetration to the solar cell, the type of climatic season, the temperature of the surroundings, and the environmental humidity. Unfortunately, every renewable’s technology has its limitation. Consequently, this prevents the system from operating to a maximum or optimally. Achieving a precise PV system output power is crucial to overcoming solar power output instability and intermittency performance. This paper discusses an intensive review of machine learning, followed by the types of neural network models under supervised machine learning implemented in photovoltaic power forecasting. The literature of past researchers is collected, mainly focusing on the duration of forecasts for very short-, short-, and long-term forecasts in a photovoltaic system. The performance of forecasting is also evaluated according to a different type of input parameter and time-step resolution. Lastly, the crucial aspects of a conventional and hybrid model of machine learning and neural networks are reviewed comprehensively.

Prediction of Energy Production Level in Large PV Plants through AUTO-Encoder Based Neural-Network (AUTO-NN) with Restricted Boltzmann Feature Extraction

Article

Full-text available

Jan 2023

In general, reliable PV generation prediction is required to increase complete control quality and avoid potential damage. Accurate forecasting of direct solar radiation trends in PV power production could limit the influence of uncertainties on photovoltaics, enhance organizational dependability, and maximize the utilization factor of the PV systems for something such as an energy management system (EMS) of microgrids. This paper proposes an intelligent prediction of energy production level in large PV plants through AUTO-encoder-based Neural-Network (AUTO-NN) with Restricted Boltzmann feature extraction. Here, the solar energy output may be projected using prior sun illumination and meteorological data. The feature selection and prediction modules use an AUTO encoder-based Neural Network to improve the process of energy prediction (AUTO-NN). Restricted Boltzmann Machines (RBM) can be used during a set of regulations for development-based feature extraction. The proposed model’s result is evaluated using various constraints. As a result, the proposed AUTO-NN achieved 58.72% of RMSE (Root Mean Square Error), 62.72% of n-RMSE (Normalized Root Mean Square Error), 48.04% of Max-AE (Maximum Absolute Error), 48.66% of (Mean Absolute Error), and 46.76% of (Mean Absolute Percentage Error).

Review of Selected Aspects of Wind Energy Market Development in Poland and Lithuania in the Face of Current Challenges

Article

Full-text available

Jan 2023

The disruption of fossil fuel supply chains due to the war in Ukraine has resulted in the need for an urgent reorganisation of the energy supply system, the cost of which has created a substantial increase in electricity prices in many markets. In light of the above, the need for the development of a renewable energy market has become stronger than ever; hence, the authors of this study have oriented their efforts towards investigating the development of the renewable energy market in countries bordering the line of armed conflict in Ukraine, i.e., Poland—strongly dependent on traditional forms of energy production—and Lithuania. The primary objective of the paper is to review the literature on wind energy, which is necessary to establish the current role of this energy dimension in the renewable energy market in the energy systems of Poland and Lithuania. Therefore, this review paper is oriented towards a review and evaluation of the available thematic literature and industry studies, as well as conclusions related to the number and direction of research topics in the area of the explored issues. The basic finding of this review is that the reviewed literature and studies are most strongly oriented towards a general assessment of the ongoing energy transition in the world, in which the thread of the assessment of the energy situation in Poland and Lithuania, including the thread of the analysis of wind energy, is part of broader assessments, most often regarding EU countries. The wind energy of the countries included in the scope of the review is not discussed comprehensively. The gap identified in this respect relates in particular to the aspect of wind energy development potential concerning solutions targeted at the individual consumer. In quantitative terms, studies addressing wind energy in Lithuania represent a lower percentage of the thematic literature acquired for the review. In the area of noted niches, the need for research and analysis is recommended to increase the information supply for developing the renewable energy market in Poland and Lithuania. In doing so, it is important to explore the technical and technological solutions (with a focus on the individual customer) and the economic aspects of wind installations from a micro and macro perspective. In addition, there is a lack of sufficient studies revealing the position of public opinion regarding the development of this dimension of the RES market and the direction of its changes. This is an important problem—particularly in Poland, where the so-called distance law constantly blocks the development of this dimension of RES and where the need to develop energy from renewable sources is particularly urgent.

Predicting photovoltaic power production using high-uncertainty weather forecasts

Article

Jun 2023
APPL ENERG

The value of forecasts for PV power plants operating in the past, present and future Scandinavian energy markets

Article

May 2023
SOL ENERGY

Techno-economic assessment of grid and renewable powered electric vehicle charging stations in India using a modified metaheuristic technique

Article

Apr 2023
ENERG CONVERS MANAGE

Renewable energy technologies have left an indelible impression on the global electrical market by supplying customers with uninterrupted electricity. Utilizing renewable energy sources to satisfy the load requirements of electric vehicle charging stations is one of the most significant areas of study in developing nations. This research investigates the technical and financial viability of grid and renewable-powered energy systems for an environmentally sustainable electric vehicle charging station in three distinct regions across India. The presented setup is subjected to an economic analysis for satisfying the charging requirement of the electric vehicle load. The system is designed in such a way that the power exchange takes place between the grid network and different components of the energy system. The assigned places are selected so that wind and solar prospects are contrary in these areas. The charging hours of electric vehicles are different in the specific sites. The optimal sizing of the system component is performed to minimize the levelized cost of electricity and total net present cost while keeping the probability of power supply failure within acceptable limits using a novel metaheuristic-based optimization algorithm i.e., modified salp swarm algorithm. The achievement of the designed system demonstrates that New Delhi has a lower total net present cost of $ 14853.63 and levelized cost of electricity of 0.0051 $/kWh. The optimum number of solar panels and wind turbines needed to meet the EV load demand in the New Delhi region is 120 with 325 W rated capacity and 310 with 650 W rated capacity. The optimal solution for the hybrid system consists of 64.5% wind energy, 33.5% solar energy and 1.94% grid purchase with annual electricity production is 5,91,117 kWh. Similarly, the optimum values of the total net present cost and levelized cost of electricity for the city of Ahmedabad are 222762.80 $ and 0.023 kWh, respectively. The total net present cost and levelized cost of electricity in the Madurai area are 37.44% and 8.17% higher than in the New Delhi region, respectively. The envisaged system could help to reduce the dependence on the overstrained grid, especially in developing nations.

Short-term photovoltaic power point-interval forecasting based on double-layer decomposition and WOA-BiLSTM-Attention and considering weather classification

Article

Apr 2023
ENERGY

Day-ahead probabilistic forecasting at a co-located wind and solar power park in Sweden: Trading and forecast verification

Article

Jan 2023

This paper presents a first step in the field of probabilistic forecasting of co-located wind and photovoltaic (PV) parks. The effect of aggregation is analyzed with respect to forecast accuracy and value at a co-located park in Sweden using roughly three years of data. We use a fixed modelling framework where we post-process numerical weather predictions to calibrated probabilistic production forecasts, which is a prerequisite when placing optimal bids in the day-ahead market. The results show that aggregation improves forecast accuracy in terms of continuous ranked probability score, interval score and quantile score when compared to wind or PV power forecasts alone. The optimal aggregation ratio is found to be 50%–60% wind power and the remainder PV power. This is explained by the aggregated time series being smoother, which improves the calibration and produces sharper predictive distributions, especially during periods of high variability in both resources, i.e., most prominently in the summer, spring and fall. Furthermore, the daily variability of wind and PV power generation was found to be anti-correlated which proved to be beneficial when forecasting the aggregated time series. Finally, we show that probabilistic forecasts of co-located production improve trading in the day-ahead market, where the more accurate and sharper forecasts reduce balancing costs. In conclusion, the study indicates that co-locating wind and PV power parks can improve probabilistic forecasts which, furthermore, carry over to electricity market trading. The results from the study should be generally applicable to other co-located parks in similar climates.