Elizabeth Marques Duarte Pereira

Using the NextClosure algorithm to extract rules from trained neural networks application in solar energy systems

Due to their capability of dealing with nonlinear problems, artificial neural networks (ANN) is widely used with several purposes. Once trained, they are capable to solve unprecedented situations, keeping tolerable errors in their outputs. However, humans cannot assimilate the knowledge kept by those nets, since such knowledge is implicitly represented by their connections weights. So, in order to facilitate the extraction of rules that describe the knowledge of ANN, formal concept analysis (FCA) and the NextClosure algorithm have been used. Such method is presented in this work, combining ANN, FCA and the NextClosure algorithm to compute the minimal implication base (Stem Base). As an example, solar energy systems are the domain application considered here, due to their importance as substitutes of traditional energy systems.

Techniques for Training Sets Selection in the Representation of a Thermosiphon System Via ANN

The research of alternative forms of energy production became more important nowadays in a context where the natural resources are scarce. In this sense, thermosiphon systems have been developed as an alternative way of energy economy for the water heating process using a renewable energy source: the sun. A thermosiphon system is greatly influenced by several parameters: the ambient temperature, the input water temperature, the solar irradiance, the flow rate, the inclination of the solar collector, the height of the water storage tank and mainly by the manufacturing process. Nowadays, there are interests in the development of analytical models that consider parameters of installation such as: height of the water storage tank and inclination of the solar collector. These analytical models can be complex and nonlinear. In the last decades, ANNs (i.e. artificial neural networks) have been used to represent many kinds of industrial processes, dealing with the complexity and non-linearity of them. Moreover, ANNs are capable of dealing with aspects of manufacturing not considered by the analytical models but that are important in determining the efficiency of the real thermosiphon system. A trained ANN can eliminate the necessity of new laboratory experiments for real and new conditions of installation. A better modeling of the process by means of ANN depends on a representative training set. In order to define the training set better, statistical ways and clustering techniques have been proposed and compared. Results of both techniques have been discussed in this work.

Hybrid thermal model for swimming pools based on artificial neural networks for southeast region of Brazil

Nowadays, the usage of systems based on solar energy have been largely stimulated. The correct designing and efficiency of these systems are highly dependent of the seasonal climatic characteristics of the regions where they will be installed. In this work, we propose a hybrid structure to simulate the thermodynamic behavior of pools, which uses neural computational models to incorporate the climatic information of the regions being analyzed. The neural models have as input variables data of geographic position such as: elevation, latitude and longitude, what permits to delineate the climatic profile of the region being considered. The human activity is another factor that directly influences the thermodynamic behavior of pools and, therefore, is also considered. In this work, changes of volume are estimated in order to track losses due to the human activity.

Parametric Analysis of Solar Collectors Through Sensitivity Factors Via Artificial Neural Networks

Since solar collectors have been presented as an alternative way of energy producing, many researches have been working with these systems. The solar collectors are greatly influenced by the operation parameters: ambient temperature (Tamb), input water temperature (Tin), solar irradiance (G) and mainly by manufacture process. Those parameters are important in order to know the quality and the efficiency of a specific solar collector. The efficiency of those systems can be influenced by manufacture process and this condition is not considered in mathematical models of collectors. On other hand, due its facility in solving nonlinear problems, experimental data based, in this paper, artificial neural networks (ANN) have been proposed as alternative to represent and to compare solar collectors. In the classification of solar collectors, is important to know how Tamb, Tin, G influence the output water temperature (Tout) (strongly associated to the system efficiency) for each collector considered. These influences may be obtained through the sensitivity analysis of the parameters in relation to Tout. So, through differentiation of a previously trained net, the sensitivity factors can be obtained. The sensitivity factors show how much the input variables influence the output variables. In this paper, the sensitivity analysis via ANN, to compare and classify solar collectors is applied and discussed.

Neural representation of a solar collector with statistical optimization of the training set

One of the difficulties of using Artificial Neural Networks (ANNs) to estimate atmospheric temperature is the large number of potential input variables available. In this study, four different feature extraction methods were used to reduce the input vector to train four networks to estimate temperature at different atmospheric levels. The four techniques used were: genetic algorithms (GA), coefficient of determination (CoD), mutual information (MI) and simple neural analysis (SNA). The results demonstrate that of the four methods used for this data set, mutual information and simple neural analysis can generate networks that have a smaller input parameter set, while still maintaining a high degree of accuracy.

Neural models to insert climatic information in thermal model of swimming pools — simulation for different climate conditions and profile of human activity

In this paper, a hybrid structure is proposed to simulate thermal behaviors of pools. The structure uses several neural models to represent the climatic data and a parametric estimation algorithm to determine the variation in volume due to the human activity. The new structure allows adapting the thermal dynamic model for pools, considering variations over time for different regional weather conditions. As a case study, the region of the state of Minas Gerais (MG) - Brazil was considered.

Climatic data neural representation for large territorial extensions: Case study for the State of Minas Gerais

It is possible to observe that for large areas the number of meteorological stations is small or they are improperly distributed. In environments or systems whose climatic variables impact directly or indirectly in the production, it is necessary to know or at least be able to estimate climate data to improve the production of the processes. To meet this demand, in this paper a representation of weather data for large areas through artificial neural networks (ANN) is proposed. All the procedures adopted are detailed which allow to be used to represent other regions. The main input variables of the neural model are the latitude, longitude and altitude.

HYBRID STRUCTURE TO SIMULATE THE THERMAL BEHAVIOR OF POOLS FOR DIFFERENT CLIMATE CONDITIONS AND PROFILE OF HUMAN ACTIVITY

In this work is proposed a hybrid structure to simulate thermal behaviors of pools. The structure uses neural representation to model the climatic data and parametric estimation to determine the variation in volume due the human activity. The new structure allows adapting the theoretical dynamic models, with variations over time for specific regional weather conditions. As a case of study, data of state of Minas Gerais (MG) – Brazil were used.

A Framework for Photovoltaic and Thermosiphon Systems

Pessimistic forecasts are growing in the Brazilian energy scenario demanding the use of renewable sources of energy such as the solar one. As metropolitan regions have become more populous, private and public companies have developed new technologies based on renewable energy sources. In order to supply such demand, new computer techniques have to be developed. This paper presents a framework to assist the developer to model new components and simulate solar energy applications. By applying the framework concepts, such as source code reuse, one can create a complete environment to evaluate solar energy data. The framework supports software development and tool implementation to be used in photovoltaic and thermosiphon processes.

SolarEnergy: A framework for photovoltaic systems - A case study

Pessimistic forecasts are growing in the Brazilian energy scenario demanding the use of renewable sources of energy such as the solar one. As it use grows, new technologies and tools must be developed. This paper describes a photovoltaic process for simulating the behavior of photovoltaic systems. By applying the framework concepts, such as source code reuse, one can create a complete environment to evaluate solar energy data. This work focuses in the software development and tools to be used in the photovoltaic energy generation processes.