Search

A large proportion of photovoltaic (PV) power generation is connected to the power grid, and its volatility and stochasticity have significant impacts on the power system. Accurate PV power forecasting is of great significance in optimizing the safe operation of the power grid and power market transactions. In this paper, a novel dual-channel PV power forecasting method based on a temporal convolutional network (TCN) is proposed. The method deeply integrates the PV station feature data with the model computing mechanism through the dual-channel model architecture; utilizes the combination of multihead attention (MHA) and TCN to extract the multidimensional spatio-temporal features between other meteorological variables and the PV power; and utilizes a single TCN to fully extract the temporal constraints of the power sequence elements. The weighted fusion of the dual-channel feature data ultimately yields the ideal forecasting results. The experimental data in this study are from a 26.52 kW PV power plant in central Australia. The experiments were carried out over seven different input window widths, and the two models that currently show superior performance within the field of PV power forecasting: the convolutional neural network (CNN), and the convolutional neural network combined with a long and short-term memory network (CNN_LSTM), are used as the baseline models. The experimental results show that the proposed model and the baseline models both obtained the best forecasting performance over a 1-day input window width, while the proposed model exhibited superior forecasting performance compared to the baseline model. It also shows that designing model architectures that deeply integrate the data input method with the model mechanism has research potential in the field of PV power forecasting. [ABSTRACT FROM AUTHOR]

This paper introduces an innovative framework for wind power prediction that focuses on the future of energy forecasting utilizing intelligent deep learning and strategic feature engineering. This research investigates the application of a state-of-the-art deep learning model for wind energy prediction to make extremely short-term forecasts using real-time data on wind generation from New South Wales, Australia. In contrast with typical approaches to wind energy forecasting, this model relies entirely on historical data and strategic feature engineering to make predictions, rather than relying on meteorological parameters. A hybrid feature engineering strategy that integrates features from several feature generation techniques to obtain the optimal input parameters is a significant contribution to this work. The model's performance is assessed using key metrics, yielding optimal results with a Mean Absolute Error (MAE) of 8.76, Mean Squared Error (MSE) of 139.49, Root Mean Squared Error (RMSE) of 11.81, R-squared score of 0.997, and Mean Absolute Percentage Error (MAPE) of 4.85%. Additionally, the proposed framework outperforms six other deep learning and hybrid deep learning models in terms of wind energy prediction accuracy. These findings highlight the importance of advanced data analysis for feature generation in data processing, pointing to its key role in boosting the precision of forecasting applications. [ABSTRACT FROM AUTHOR]

A circular economy involves managing and reducing the environmental and social impacts of products and materials throughout their entire lifecycle, from production to end of life, including clean energy technologies. The remarkable growth of wind turbine (WT) deployment in Australia, as a clean energy source, is promising, with over 10 gigawatts (GW) installed by 2023. Responsible management of wind turbines throughout the entire supply chain, including their end of life, is crucial to prevent potential environmental issues caused by significant waste volumes and to identify opportunities for resource recovery. This study offers a comprehensive overview of current and future WT waste through material flow analysis (MFA) under five national wind energy deployment scenarios, considering various wind turbine technologies. The results indicate that the projected cumulative WT installation capacity will range from 13 to 38 GW by 2041. Consequently, the cumulative WT waste volume is expected to range between 6.69 and 19.76 million tonnes in 2060, depending on the scenario, with the "slow change" scenario producing the least waste and the "step change" scenario generating the most. The estimated waste stream will see a rapid increase from about 2028, encompassing a variety of materials, primarily concrete at 10.20 million tonnes, followed by 3.21 million tonnes of steel and 35.41 kt of copper by 2060. Additionally, valuable materials such as rare earth elements (REEs) and composites, despite their smaller quantities, have significant environmental, economic, and supply chain security implications. This substantial waste material presents an opportunity for resource recovery and underscores the importance of adopting a circular economy approach for wind energy systems. [ABSTRACT FROM AUTHOR]

Australia has abundant volumes of forest residues that are a potential feedstock for supplying biomass as a renewable carbon carrier to the market. However, there remains an underutilization of this resource, even in mature bioeconomy markets. Several existing or perceived barriers can be attributed to the underdeveloped, forest-based bioeconomy in Australia. One of these is the limited understanding of feedstock supply costs. In this study, two ranking approaches were applied to identify the optimal biomass feedstock supply chain from field to conversion plant gate. A panel of experts embedded in the Australian bioeconomy were employed to first assign ranks to biomass supply chain items by cost intensity. Then, a layer of analytic hierarchical process (AHP) was used to weigh and rank various biomass supply pathways by efficiency. The results reveal that biomass extraction ranks the highest and biomass feedstock storage ranks the lowest, relative to other supply chain costs. Extracting and chipping material in the field attracted the most support from the experts in terms of efficiency, followed by transporting and chipping at the roadside and, finally, transporting and chipping at the conversion plant. This study provides insights for designers of the forest-based bioeconomy in Australia into relative cost drivers that may be applied to investment and industry decisions. It also provides a framework to support further investigations into forest biomass development and the management of biomass as a renewable carbon carrier at a time when Australia is transitioning from an energy policy focused on fossil fuels to a renewable energy strategy. [ABSTRACT FROM AUTHOR]

The increased uptake of variable renewable energy sources has increased electricity price volatility in many energy pool markets, providing an opportunity for storage systems to profit through energy arbitrage. Comparison between the cost or value of storage systems engaging in energy arbitrage should be performed on a levelised basis due to differences in system lifetime. Existing energy arbitrage models with bid/offer curves and imperfect forecasting are typically computationally expensive and are impractical for calculating lifetime levelised cost metrics. In this work, an open-source modular energy arbitrage model with bid and offer curve inputs was developed for a lithium-ion battery energy storage system (BESS) and pumped hydro system (PHS) to analyse lifetime levelised cost and revenue. The mixed integer linear program scheduling module included a new piece-wise linearised description of PHS charging behaviour for rapid optimisation. A one-at-a-time sensitivity analysis indicated that levelised cost and revenue were highly sensitive to discharging efficiency. In a case study based on Australia's National Electricity Market, imperfect forecasting with no risk hedging was found to increase levelised costs by up to 24% and decrease levelised revenue by up to 50% relative to perfect price forecasting, despite 95% of prices being forecast to be within $35/MWh of the actual trading price. BESS levelised costs were more significantly correlated with consistent low risk bids (Kendall Tau-b of 0.75), since the undiscounted capital costs contribute to a larger proportion of the overall costs than in the PHS systems. [ABSTRACT FROM AUTHOR]