Search

Accurate solar photovoltaic (PV) generation forecast is critical to the reliable and economic operation of a modern power system. In practice, due to various faulty issues in the sensor, communication, or database system, the historical and online measurement data may not be always complete, and the missing data could dramatically degrade the forecasting model's accuracy. To solve this problem, this paper proposes an integrated missing-data tolerant model for probabilistic PV power generation forecasting. Taking historical PV generations as input, this model is based on a recursive long short-term memory network (Rec-LSTM), which can provide multi-step ahead forecasting of the probability distribution of PV generation. The unobserved input data will be imputed recursively based on the model output at the previous time step. During the training process, the imputations and forecasting values are iteratively updated by the negative log-likelihood loss function. As a salient advantage, this method can deal with data missing scenarios at both offline and online stages. Numerical experiments are conducted on two one-year datasets from Australia and Singapore, respectively. Probabilistic forecasting for both large-scale and small-scale building-level PV power generation is tested at the time resolution of 15 mins. Testing results show the proposed method can achieve superior probabilistic prediction accuracy as well as strong robustness under various data missing scenarios, compared to other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Ammonia (NH 3) is carbon-free and is known as a promising renewable fuel to achieve carbon reduction and carbon neutrality. However, ammonia presents much lower fuel reactivity compared to hydrocarbon fuels. Thus, ammonia is often blended with high reactivity fuels in practical combustion facilities, and natural gas is a good candidate due to its low carbon emissions and massive reserves. Methane is often used to represent natural gas in previous experimental and simulation studies, while the larger alkanes in natural gas can affect the fuel reactivity of ammonia/natural gas blends. This study aims to explore the effects of fuel components on the auto-ignition behaviors of ammonia/natural gas blends. As methane and ethane are the most important species in natural gas, a blend of 90 % methane and 10 % ethane by volume is used to represent natural gas in this study. A low-pressure shock tube was used to measure the ignition delay times (IDTs) of ammonia/methane, ammonia/ethane, and ammonia/natural gas blends in 'air' mixtures, with different ammonia blending ratios, at stoichiometric condition, at 1 atm and over the temperature range 1200 – 1800 K. The experimental results show that the IDTs of fuel blends increase with ammonia blending ratios. Meanwhile, 10 % ethane in natural gas can influence the fuel reactivity of ammonia/natural gas blends. A detailed kinetic model proposed in our previous study is used to simulate these new IDTs in addition to experimental data available in the literature. Overall, the kinetic model can predict well the auto-ignition behavior of ammonia/methane, ammonia/ethane, and ammonia/natural gas blends over a wide range of experimental conditions. The simulation and sensitivity analyses results show that the interaction reaction between ammonia and ethane, C 2 H 6 + ṄH 2 〈=〉 Ċ 2 H 5 + NH 3 , is important for predicting the fuel reactivity of ammonia/ethane blends below 1600 K at the studied conditions. [ABSTRACT FROM AUTHOR]