A model for metal uptake by microorganisms based on surface adsorption has been developed, and then applied to the uptake of cadmium by Chlorella vulgaris. A linear equilibrium relationship between metal in the solution and that adsorbed on the cell surface is assumed and confirmed at low cadmium concentrations by short-term uptake experiments. When it incorporates a description of cell growth, the model predicts an initial rapid uptake and a subsequent slow uptake. Such behavior has often been observed in experiments with growing microorganisms. This indicates that the slow uptake, sometimes thought to be active or metabolic, could be due to the simultaneous effects of growth and surface adsorption. The model shows that initial metal uptake is fast and approaches equilibrium within a few seconds. This prediction is in agreement with experimental results in a batch system: Equilibrium is reached before the first samples are taken (at 10 min) and there is then no measurable change until growth provides a significant increase in cell surface (after several hours). Thus the equilibrium constant can be calculated from experimental results of uptake at 10 min. The equilibrium is found to be affected by phosphate concentration; the amount of cadmium adsorbed on the cell decreases as the concentration of phosphate is increased. Long-term uptake experiments in growing cultures show a greater metal accumulation than predicted by the adsorption model, suggesting the involvement in the slow long-term uptake of some mechanism other than adsorption. This is confirmed by experiments in which uptake in cultures exposed to cadmium throughout the growth period is compared with short-term uptake in similar cultures grown in the absence of cadmium. The modeling approach to metal adsorption provides a basis for further development. A model combining description of adsorption and of intracellular accumulation is necessary to provide a more complete description. Such a model, with precise definitions of system parameters and means of evaluating these parameters from experimental results, will be a powerful tool in investigation of metal uptake by microorganisms.

In this article, an artificial neural network (ANN) and a regression model are applied to forecast long term electricity consumption in Thailand. The inputs of both nonlinear models are gross domestic product, number of population. Maximum ambient temperature and electricity power demand are used as inputs in a neural network to predict electricity consumption. The results show that the ANN model can give more accurate estimations than regression model as indicated by the performance measures, namely coefficient of determination, mean absolute percentage error and root mean square error. Accoding to the forecasting results by the regression and ANN models of this study, the electricity consumption of the country in 2010, 2015, and 2020 will reach 160,136, 188,552, and 216,986 GWh, respectively, for the regression model while the ANN model will reach 155,917, 174,394, and 188,137 GWh, respectively.

No abstract

No abstract

No abstract