Short, medium and long term load forecasting model and virtual load forecaster based on radial basis neural networks

Artificial neural networks (ANNs) can be easily applied to short-term load forecasting (STLF) models for electric power distribution applications. However, they are not typically used in medium and long term load forecasting (MLTLF) electric power models because of the difficulties associated with collecting and processing the necessary data. Virtual instrument (VI) techniques can be applied to electric power load forecasting but this is rarely reported in the literature. In this paper, we investigate the modelling and design of a VI for short, medium and long term load forecasting using ANNs. Three ANN models were built for STLF of electric power. These networks were trained using historical load data and also considering weather data which is known to have a significant affect of the use of electric power (such as wind speed, precipitation, atmospheric pressure, temperature and humidity). In order to do this a V-shape temperature processing model is proposed. With regards MLTLF, a model was developed using radial basis function neural networks (RBFNN). Results indicate that the forecasting model based on the RBFNN has a high accuracy and stability. Finally, a virtual load forecaster which integrates the VI and the RBFNN is presented.

Is China's fifth-largest inland lake to dry-up? Incorporated hydrological and satellite-based methods for forecasting Hulun Lake water levels

Hulun Lake, China's fifth-largest inland lake, experienced severe declines in water level in the period of 2000-2010. This has prompted concerns whether the lake is drying up gradually. A multi-million US dollar engineering project to construct a water channel to transfer part of the river flow from a nearby river to maintain the water level was completed in August 2010. This study aimed to advance the understanding of the key processes controlling the lake water level variation over the last five decades, as well as investigate the impact of the river transfer engineering project on the water level. A water balance model was developed to investigate the lake water level variations over the last five decades, using hydrological and climatic data as well as satellite-based measurements and results from land surface modelling. The investigation reveals that the severe reduction of river discharge (-364±64 mm/yr, ∼70% of the five-decade average) into the lake was the key factor behind the decline of the lake water level between 2000 and 2010. The decline of river discharge was due to the reduction of total runoff from the lake watershed. This was a result of the reduction of soil moisture due to the decrease of precipitation (-49±45 mm/yr) over this period. The water budget calculation suggests that the groundwater component from the surrounding lake area as well as surface run off from the un-gauged area surrounding the lake contributed ∼ net 210 Mm3/yr (equivalent to ∼ 100 mm/yr) water inflows into the lake. The results also show that the water diversion project did prevent a further water level decline of over 0.5 m by the end of 2012. Overall, the monthly water balance model gave an excellent prediction of the lake water level fluctuation over the last five decades and can be a useful tool to manage lake water resources in the future.

Low-frequency electrical response to microbial induced sulfide precipitation

We investigated the sensitivity of low-frequency electrical measurements to microbe-induced metal sulfide precipitation. Three identical sand-packed monitoring columns were used; a geochemical column, an electrical column and a control column. In the first experiment, continuous upward flow of nutrients and metals in solution was established in each column. Cells of Desulfovibrio vulgaris (D. vulgaris) were injected into the center of the geochemical and electrical columns. Geochemical sampling and post-experiment destructive analysis showed that microbial induced sulfate reduction led to metal precipitation on bacteria cells, forming motile biominerals. Precipitation initially occurred in the injection zone, followed by chemotactic migration of D. vulgaris and ultimate accumulation around the nutrient source at the column base. Results from this experiment conducted with metals show (1) polarization anomalies, up to 14 mrad, develop at the bacteria injection and final accumulation areas, (2) the onset of polarization increase occurs concurrently with the onset of lactate consumption, (3) polarization profiles are similar to calculated profiles of the rate of lactate consumption, and (4) temporal changes in polarization and conduction correlate with a geometrical rearrangement of metal-coated bacterial cells. In a second experiment, the same biogeochemical conditions were established except that no metals were added to the flow solution. Polarization anomalies were absent when the experiment was replicated without metals in solution. We therefore attribute the polarization increase observed in the first experiment to a metal-fluid interfacial mechanism that develops as metal sulfides precipitate onto microbial cells and form biominerals. Temporal changes in polarization and conductivity reflect changes in (1) the amount of metal-fluid interfacial area, and (2) the amount of electronic conduction resulting from microbial growth, chemotactic movement and final coagulation. This polarization is correlated with the rate of microbial activity inferred from the lactate concentration gradient, probably via a common total metal surface area effect.

The effect of co-precipitation on cadmium(II) adsorption on hydrous aluminium(III) hydroxide in the presence of a range of chelates

The adsorption of cadmium(II) on freshly precipitated aluminium(III) hydroxide in the presence of a range of chelates has been investigated. By precipitating the metal, chelate and adsorbent together it is possible to change the pH variation of the metal-complex adsorption from anionic, ligand-like, binding to cationic binding. This is a general phenomenon and is explained by the formation of a ternary Al-O-Cd-L surface species. As a consequence of the preparation method, the pH edge is found to shift to lower pH values in the presence of the chelate which gives rise to an apparent increase in adsorption of Cd2+. This increase is, in general, most pronounced at [chelate] / [metal] > 1. Computer modelling shows that the observed trends result from the competition between Al-O-Cd-L and Al-L for the available aluminium( III) binding sites. The enhanced adsorption in the presence of phenylenediaminetetraacetate is anomalous since it is observed at a [ chelate] / [metal] approximate to 0.1 and cannot be interpreted by the simple competition model.