AGRICULTURAL EQUIPMENT

Author: FENOLLOSA RIBERA MARÍA LORETO.
Year: 2005.
University: POLITÉCNICA DE VALENCIA [www.upv.es].
Place of defense: UNIVERSIDAD POLITÉCNICA DE VALENCIA.
Place of preparation: UNIVERSIDAD POLITÉCNICA DE VALENCIA.

Summary: The main thesis aims to obtain, for econometric methods, mathematical models to estimate the value of tractors and harvesters of first and second hand, in Spain and Italy. This will define the variables that influence the value of each type of machine, and the evolution of communications, new and used over time. It is also proposed a classification of tractors by groups of power, to provide models of valuation and depreciation differentiated behaviors that help professionals in the appraisal. Finally, the theoretical models are contrasted with the amortization of empirical models, and set the weights they have in the depreciation of machinery, the two effects: wear and obsolescence.

Author: LÓPEZ ÁLVAREZ PAZ.
Year: 2006.
University: SANTIAGO DE COMPOSTELA [www.usc.es].
Place of defense: ESCUELA POLITÉCNICA SUPERIOR.
Place of preparation: ESCUELA POLITÉCNICA SUPERIOR.

Summary: Galicia is the autonomous Spanish region with more development of wind energy as corroborates that at the end of the year 2005 there is an installed capacity of 2450 MW. It is an area of high wind potential for its geographical features, but has a very complex topography makes complicated his characterization wind. The correct identification of this potential is essential to optimize use systems of wind energy. Currently Galicia areas now have greater use of wind farms in operation, on the other where it is smaller, which is crucial adequate estimate of the wind speed to ensure the suitability of the location of new wind power plants. Moreover, despite the immense growth of wind power in Galicia in the past 10 years, there are hardly any small power plants in the middle agroforestry. In this thesis proposes an alternative method to conventional antennas to estimate the wind speed at a site that does not have its weather station to measure. It determines the annual average wind speed at a location through neural networks, using as input for the training data wind stations nearby and exit some measures wind in that location by measuring mobile equipment. The wind data used as input to the neural network have been submitted by MeteoGalicia body of the Galician regional government. These are data from 20 stations located in different parts of the region, near the coast, inside and various altitudes above sea level. The number of inputs to be used by the neural network was discussed in detail and concluded that sufficient data speed and direction of a single station, the address is a very important input in complex terrain locations. The neural network used is the type perceptrónmulticapa with one hidden layer with 15 neurons. The training algorithm used is the regularization Bayes, which is a method that ensures the generalization, in our case and we intend to measure needed only 60 days and from these obtain the average annual rate. Moreover addition to the annual wind speed to estimate the average power available at a site, it is necessary to know the parameter k of the role of Wwibull, which can be calculated directly on the extent of the 60 days distributed in every month of year or calculate a new neural network for v3 with 20 neurons in the hidden layer when the 60 days are not distributed throughout the year. Also discusses the different types of wind turbines for small commercial power for use in rural areas and an estimated energy from the simulation speed averages annulled, the parameter k cornering power of the wind turbines. In models of wind turbines analyzed who provided the best results, are all technologies, in the absence of a model that offers the best benefits. In addition, there are on the market in diameter wind turbines far superior to standard, up more than double, to a determination power, which are able to have high performance for low wind speeds. In small wind turbines this method allows for a faster and less expensive alternative to the current measurement using conventional antennas. In addition to large wind turbines, considering that measures 6 locations in the same year with a single mobile equipment, it could be an alternative to commercial software simulation as is the WAsP, leading to errors in the simulation of high land complex. Well to hear the appeal in the exact locations of the turbines, either to make corrections to the simulations WAsP. In addition to face-to 8 an esti 316 mation appeal this method could be used to implement a wind map of Galicia.