Search

Summary: This paper investigates the operation of various optimised AC‐coupled hybrid power systems for an isolated community in South Australia. In addition to diesel generators, various combinations of distributed generators and storage technologies are considered to form different hybrid power systems. Optimal design of the hybrid power systems is implemented using particle swarm optimisation (PSO) algorithm. The performance of the PSO in optimizing the above hybrid systems is discussed in detail. Whilst meeting the power system operational constraints, the minimum levelised cost of electricity (LCOE) of all systems, with various combinations of system components, are calculated. Simulation results obtained for each optimized system are carefully described and analysed. The annual operations of various components in each system are then compared and discussed in detail. A sensitivity analysis by varying the cost of wind turbines and the photovoltaic is also conducted for all hybrid systems. It has been found that the inclusion of energy storage systems (battery and flywheel) can reduce the LCOE and CO2 emission significantly compared to the inclusion of renewable generations (solar and wind) alone. The diesel‐wind‐solar‐battery system is the eco‐friendliest with an annual CO2 emission of 451 t compared to the reference diesel system (1064 t). In addition, the hybrid system consisting of diesel, wind, solar, battery and flywheel is found to be the most economic with a LCOE of 58.94 ¢/kWh compared to the diesel system with a LCOE of 67.17 ¢/kWh. [ABSTRACT FROM AUTHOR]

Summary: The use of artificial neural networks (ANNs) for the modelling of water resources variables has increased rapidly in recent years. This paper addresses one of the important issues associated with artificial neural network model development; input variable selection. In this study, the partial mutual information (PMI) input selection algorithm is modified to increase its computational efficiency, while maintaining its accuracy. As part of the modification, use of average shifted histograms (ASHs) is introduced as an alternative to kernel based methods for the estimation of mutual information (MI). Empirical guidelines are developed to estimate the key ASH parameters as a function of sample size. The stopping criterion used with the original PMI algorithm is replaced with a more computationally efficient outlier detection technique based on the Hampel distance. The performance of the proposed PMI algorithm, in terms of computational efficiency and input selection accuracy, is first investigated by using it to identify significant variables for data series where dependencies of attributes are known a priori. The proposed ASH PMI input variable selection algorithm with the Hampel distance stopping criterion consistently selects the correct inputs, while being computationally efficient. The modified PMI algorithm is then applied to identify suitable inputs to forecast salinity in the River Murray at Murray Bridge, South Australia, with a lead time of 14 days using an ANN approach. The ANN models developed with the inputs selected with the modified PMI algorithm perform very well when compared with results obtained using ANN models with different input sets developed in previous studies. Furthermore, the proposed input variable selection algorithm results in more parsimonious ANN models. [Copyright &y& Elsevier]

Abstract: The SeaWiFS data set covering the period 1997–2007 is used to develop a framework for a comprehensive description of the inter-annual variations in chlorophyll a concentration (Chla). For each grid cell, the monthly Chla series is decomposed into seasonal, irregular and trend-cycle terms with the Census X-11 technique that is an iterative band-pass filter algorithm. This approach allows variations in the annual cycle, while the trend-term isolates the multi-annual evolution in the mean level of the signal. The patterns with relatively large inter-annual variations are selected using the variance due to the trend-term with respect to the total variance, and are compared with maps of monotonic trends derived by a non-parametric Kendall analysis. Most of these patterns are identified in the subtropical domain (30°S–30°N), even though there are patterns with strong variations at mid-latitudes, particularly in the Northeast Atlantic and South of Australia. The time series found within each pattern of interest are found coherent. Conversely, the ensemble of spatially averaged time series of Chla trend-terms shows a diversity of evolutions, with rather monotonic changes for all or part of the period, abrupt shifts or low-frequency oscillations, sometimes coupled with a modification in the amplitude of the annual cycle. Some of these series are correlated with climate indices, and those in subtropical regions usually show a negative correlation with the equivalent trend-term calculated for sea surface temperature. The identified inter-annual signals should be further monitored with longer time series and can serve as test cases for biogeochemical models. [Copyright &y& Elsevier]