Hybrid renewable energy integration (HREI) system for subtropical climate in Central Queensland, Australia
Introduction
The world's total net electricity consumption as well as its electricity generation is increasing day by day and most of the energy is generated from coal fired power plants which cause greenhouse gas (GHG) emissions as well as global warming [1], [2], [3]. GHG emissions from electricity generation are approximately 40% of the world's total emissions as most industries uses fossil fuels, particularly coal and oil, hence are a leading contributor to global energy-related CO2 emissions [4]. In 2012–13, Australia's energy production (including exports) was dominated by coal, which accounted for 64% of total Australian energy production in energy content terms, followed by natural gas with a share of 21%, renewable energy with a share of 13% and others with a share of 2% [5]. Australia's abundance of coal has helped to keep energy prices low; however, reliance on coal-fired power makes it one of the world's highest per-capita GHG emissions producing countries [6].
In contrast to fossil fuels, renewable energy (RE) offers alternative sources of energy which are in general pollution free, technologically effective and environmentally sustainable and these are expected to play a leading role in meeting future electricity demands. Solar and wind energy are the most promising RE sources that encourage interest in increasing their use worldwide. Australia has favourable weather conditions for both solar PV and wind energy, considered as the most promising RE sources. It is therefore a fundamental priority today to be able to bring higher percentages of renewable electricity into the energy mix to build an environmentally friendly and sustainable power system. The Australian government has taken initiatives to encourage utilities, industries and household consumers to increase RE generation [4], [5], [6].
In line with the Government initiatives to integrate large-scale RE sources into the Australian energy mix, Shafiullah et al. [7] conducted a feasibility study in which it is evident that Australia has significant potential for RE generation, in particular solar and wind energy. It also shows that RE sources not only reduce the cost of energy generation, but also reduce GHG emissions significantly which plays a key role in developing a sustainable climate friendly environment. Based on the performance metrics, optimisation results and sensitivity analysis, it has been observed that many locations in Australia have potential for solar and wind energy generation [7]. However, only a few locations perform extremely well for wind energy and only a few other locations are highly promising for solar energy generation. It was shown that Tasmania is the most suitable State in which to install large scale wind generation plants; on the other hand, the Northern Territory is the most suitable place to install solar plants. However, simulation results clearly indicated that Queensland has enormous potentialities for both solar and wind energy sources due to its subtropical climate. Therefore, this study explores the potentialities of both small-scale and large-scale RE deployment in the subtropical climate at Queensland, in particular in Central Queensland (CQ).
Currently, most of the electricity generated in CQ comes from coal-fired power stations due to the availability of abundant coal. Burning of coal emits GHGs into the atmosphere and changes the climate continuously. Climate change adversely influences the community, business, industry, agriculture, mining and tourism. CQ is generally one of the most promising and growing regions in Queensland, Australia due to large-scale availability of mining and the nearby national and international ecological icons of Great Keppel Island (GKI) and the Great Barrier Reef (GBR) [8]. In response, the Queensland Government is moving forward to become a world-leading ‘guilt-free’ tourism destination in the Capricornia region and it has introduced attractive financial incentives to encourage the utilities and consumers to deploy large-scale RE systems [9]. Considering availability of both solar and wind sources, the growing energy demand and ongoing strengthening of the socio-economic conditions of regional Australia, this study will concentrate on facilitating integration of RE sources into the electricity distribution system in the subtropical climate.
To integrate a large amount of electricity from RE sources into the grid, the main concern needing to be addressed is the variability of energy from these sources as they are intermittent in nature and strongly dependent on weather conditions. It is difficult to predict the power outputs available at certain times of the day from these sources which are essential data required by the utilities for adequate management of consumer load demand. Moreover, the effective utilisation of wind and solar energy entails having a detailed knowledge of the wind and solar characteristics at the particular location, as the distribution of wind speeds and solar irradiation is important for the design of wind farms and solar plants, and also the associated power generators [10]. Therefore, for the effective deployment of large-scale RE facilities, the issues that need to be considered are: intermittent nature as well as unpredictable generation from RE sources; precise deployment information such as sizes of RE sources, net present costs and cost of energy generation considering all sources of costs, and finally, the socio-techno-economic viability of the facility utilising a load management approach. Several research initiatives have already been undertaken throughout the world to facilitate large-scale RE integration into the grid by introducing forecasting modelling to assess electricity generation from RE sources in advance, techno-economic modelling to investigate the economic and environmental prospects of RE integration, and energy management and optimisation approaches to manage the consumer load demand efficiently.
Machine learning techniques have been used in previous research to predict solar irradiation and wind speed [11], [12], [13], [14], [15], [16], [17]. Seme et al. [11] have developed an Artificial Neural Network (ANN) model to predict half hourly solar irradiation. The model considers extra-terrestrial solar irradiation on a horizontal surface, solar zenithal angle, day in the year, temperature, and relative air pressure. The experimental results demonstrated that daily distribution of the global solar irradiance predicted by the trained ANN is acceptable for clear days [11]. However, the same is not the case for cloudy days. Azadeh et al. [12] have used an ANN for predicting global solar irradiation in Iran. Though the results derived from these models show a considerable accuracy, e.g., root mean squared error of 3.33 and coefficient of determination of 0.94, these are not optimum for this particular type of application [12]. Al-Alawi and Al-Hinai [13], Mohandes et al. [14], and Lopez et al. [15] have also developed models that utilise ANNs to predict solar irradiation. Most of the models achieved a forecasting accuracy of around 93%. However, ANNs require significant computational time, processing power and memory. Ji et al. [16] have proposed a mean hourly wind speed forecasting model in a wind farm using a Support Vector Machine (SVM) technique. Experimental results showed that their proposed approach is acceptable as it can forecast mean hourly wind speed with a low mean square error and mean absolute percentage error. More and Deo [17] have employed a technique to forecast daily, weekly as well as monthly wind speeds at two coastal locations in India using ANNs that have been trained with past data in an auto-regressive manner using back propagation and cascade correlation algorithms. Forecast results showed that their technique has high correlation with and low deviations from actual observations. Considering flexibility and robustness, in addition to ANN and SVM, this current research study examines the use of ten regression algorithms to forecast hourly distribution of solar irradiation and wind speed and then selects the most suitable model.
There are a number of research publications currently available worldwide that explore the techno-economic prospects of integration of RE sources into electricity distribution grids. Most of the studies prove that applying hybrid renewable energy systems in such applications is cost-effective, saves energy resources and is climate-friendly, though the systems are dependent on geographical and climatic conditions. Some of the recent studies only focused on the combination of solar and storage with grid-connected or off-grid systems, while a few studies considered both wind and solar generation. Some other studies have considered large-scale RE integration while a few studies considered only small-scale RE integration such as roof-top PV systems. Hybrid renewable energy systems have been investigated for both developing and developed countries as well as for sub-tropical, tropical and hot-arid climatic conditions and investigated the potentialities of RE sources in electricity generation [18], [19], [20], [21], [22].
Dihrab and Sopian [20] developed a grid-connected hybrid system for three different locations in Iraq in which it has seen that a hybrid system can generate enough power for villages in the desert and rural areas though the plant location strongly affects the plant performance. Shafiqur et al. [21] developed a wind-PV-diesel hybrid power system for a village in Saudi Arabia and, from the results, it was shown that the system with 35% renewable energy contributions and 65% diesel power contribution was the most economical power system with a cost of energy (COE) of 0.212 US$/kWh at a diesel price of 0.2 US$/l, while the COE for a diesel only system was 0.232 US$/kWh. Saheb-Koussa [22] developed a wind/PV/diesel hybrid system for Algeria with battery backup in which it is evident that the hybrid system is the best suitable option for all the sites considered. Techno-economic performance of a residential grid-connected PV system was studied by Liu et al. [18] for the subtropical climate of Queensland, Australia in which it was shown that, for both the high price and low price scenarios, the 6 kW PV system is able to reduce more than half of the residential electricity consumption in all the cities. From the study, it was also evident that Rockhampton, Queensland is a promising location for PV generation with a ROI of approximately 15%. Baghdadi et al. [23] proposed a hybrid renewable energy system with energy management and operational strategies in which it was seen that 43% of the total electricity demand was supplied from wind turbines while 26% was supplied from PV panels. An optimal operational strategy for a hybrid renewable energy system was proposed by Hassiba et al. [24] from which it is evident that a combination of solar and wind energy provides an optimum solution in electricity generation. Authors also investigated possible power equipment or control mechanisms to manage the generation mix and load demand.
From the literature, it can be concluded that hybrid renewable energy systems play a key role in managing load growth, reducing energy costs and global warming worldwide, though the potentialities of such systems mostly depend on a few attributes such as geographical location, climatic conditions, weather patterns, availability of resources, socio-economic factors and population growth. There is little or no research available for hybrid renewable energy models/systems that consider both the prediction model and techno-economic model that facilitate the integration of RE sources into the energy mix. Moreover, most of the available research was carried out primarily in the USA, Asia and Europe. However, the characteristics of Australian distribution networks are different in many ways compared to those of other developed countries due to its geographical and climatic conditions. Research conducted by other countries could not simply be adopted without further research into the Australian context. Therefore, considering the problems identified from the existing research, this current study proposes a full-fledged hybrid renewable energy integration (HREI) system that comprises a prediction model, a techno-economic model and a load management system to facilitate the integration of RE into the distribution grid in the subtropical climate of CQ. Finally, this study considered both small-scale roof-top PV systems comprising grid-connected only and PV/grid-connected systems, and large-scale RE systems comprising grid-connected only, PV/grid-connected, wind/grid-connected and PV/wind/grid-connected systems and compared their performances based on performance metrics and identified optimised small-scale and large-scale RE systems.
Section snippets
Hybrid renewable energy integration (HREI) systems
This study proposes the development of a HREI system for the integration of large quantities of RE in the subtropical climate of CQ, in particular, the Capricornia region. The proposed HREI system analyses the characteristics and availability of unpredictable RE sources, in particular wind and solar, and explores techno-economic viability of RE generation in CQ with a load management system that ensures efficient and reliable climate-friendly power to the community. The proposed HREI system
Prediction model: forecasting energy generation
Precise solar irradiation and wind flow estimation techniques are critical in the design of RE systems. Forecasting and real-time monitoring is needed in order to have comprehensive knowledge and understanding of the geographic areas where the resource exists and the total MW possible to be generated under given scenarios [10]. The variability of RE sources must be known, particularly the relevant long-term weather patterns which can be used to develop better procedures and capabilities to
Techno-economic model
This part of the study explores the cost-economic and environmental potential for deployment of both small-scale and large-scale renewable energy generation in the subtropical climate of CQ to meet the growing energy demand and to reduce GHG emissions. Yeppoon and Rockhampton have significant potentialities in solar energy generation due to their geographical locations and solar exposure. However, wind energy generation is not significant for either Yeppoon or Rockhampton, though the coastal
Load management system
An optimised load management system is needed with which the utilities can manage consumers' load demand intelligently based on the available generation from RE sources, and that maximises the RE usage and minimises the grid supply. Finally, the proposed hybrid model analyses the energy generation from PV, wind and battery storage, and compares that generation with the load demand of the Capricornia region to determine the amount of electricity needed to be provided from the grid. The following
Conclusions and discussion
Renewable energy integration in the subtropical climate of CQ is an indispensable requirement for both environmental sustainability and meeting the growing energy demand. To reduce the energy crisis as well as minimise global warming, a hybrid renewable energy integration system was developed to facilitate renewable energy integration in the Capricornia region of CQ. Initially, a forecasting approach was developed that predicts the hourly distribution of solar irradiation and wind speed as well
References (44)
- et al.
Social, economical and environmental impacts of renewable energy systems
Renew. Energy
(2009) - et al.
Prospects of renewable energy – a feasibility study in the Australian context
Renew. Energy
(2012) - et al.
An Integrated Artificial Neural Networks approach for predicting global radiation
J. Energy Convers. Manag.
(2009) - et al.
An ANN approach for predicting global radiation in locations with no direct measurement instrumentation
Renew. Energy
(1998) - et al.
Estimation of global solar radiation using artificial neural networks
J. Renew. Energy
(1998) - et al.
Estimation of hourly global photo synthetically active radiation using artificial neural network models
J. Agric. For. Meteorology
(April, 2001) - et al.
Forecasting wind with neural networks
J. Mar. Struct.
(2003) - et al.
Techno-economic simulation and optimization of residential grid-connected PV system for the Queensland climate
Renew. Energy
(2012) - et al.
Pre-feasibility study of stand-alone hybrid energy systems for applications in Newfoundland
Renew. Energy
(2005) - et al.
Electricity generation of hybrid PV/wind systems in Iraq
Renew. Energy
(2010)