Master Degree / Yüksek Lisans Tezleri
Permanent URI for this collectionhttps://hdl.handle.net/11147/3008
Browse
2 results
Search Results
Master Thesis Control, Optimization and Monitoring of Portland Cement (pc 42.5) Quality at the Ball Mill(Izmir Institute of Technology, 2006) Avşar, Hakan; Doymaz, Fuat; Doymaz, FuatIn this study, artificial neural networks (ANN) and fuzzy logic models were developed to model relationship among cement mill operational parameters. The response variable was weight percentage of product residue on 32-micrometer sieve (or fineness), while the input parameters were revolution percent, falofon percentage, and the elevator amperage (amps), which exhibits elevator charge to the separator. The process data collected from a local plant, Cimenta Cement Factory, in 2004, were used in model construction and testing. First, ANN (Artificial Neural Network) model was constructed. A feed forward network type with one input layer including 3 input parameters, two hidden layer, and one output layer including residue percentage on 32 micrometer sieve as an output parameter was constructed. After testing the model, it was detected that the model.s ability to predict the residue on 32-micrometer sieve (fineness) was successful (Correlation coefficient is 0.92). By detailed analysis of values of parameters of ANN model.s contour plots, Mamdani type fuzzy rule set in the fuzzy model on MatLAB was created. There were three parameters and three levels, and then there were third power of three (27) rules. In this study, we constructed mix of Z type, S type and gaussian type membership functions of the input parameters and response. By help of fuzzy toolbox of MatLAB, the residue percentage on 32-micrometer sieve (fineness) was predicted. Finally, It was found that the model had a correlation coefficient of 0.76. The utility of the ANN and fuzzy models created in this study was in the potential ability of the process engineers to control processing parameters to accomplish the desired cement fineness levels. In the second part of the study, a quantitative procedure for monitoring and evaluating cement milling process performance was described. Some control charts such as CUSUM (Cumulative Sum) and EWMA (Exponentially Weighted Moving Average) charts were used to monitor the cement fineness by using historical data. As a result, it is found that CUSUM and EWMA control charts can be easily used in the cement milling process monitoring in order to detect small shifts in 32-micrometer fineness, percentage by weight, in shorter sampling time interval.Master Thesis Artificial Neural Networks and Fuzzy Logic Applications in Modeling the Compressive Strength of Portland Cement(Izmir Institute of Technology, 2004) Can, Sever; Akkurt, SedatPortland cement production is a complex process that involves the effect of several processing parameters on the quality control of 28-day cement compressive strength (CCS). There are some chemical parameters like the C3S, C2S, C3A, C4AF, and SO3 contents in addition to the physical parameters like Blaine (surface area) and particle size distribution. These factors are all effective in producing a single quantity of 28-day CCS. The long duration of 28 day CCS test provided the motivation for research on predictive models. The purpose for these studies was to be able to predict the strength instead of waiting for 28 days for the test to be complete. In this thesis, artificial intelligence (AI) methods like artificial neural networks (ANNs) and fuzzy logic were used in the modeling of the 28-day CCS. The two models were compared for their quality of fit and for the ease of application.Quality control data from a local cement plant were used in the modeling studies. The data were separated randomly into two parts: the first one contained 100 data points to be used in training and the second part had 50 data points to be used in testing stages of the models. In this study, four different AI models were created and tested (3 ANN, 1 fuzzy logic). One of the ANN models (Model A) had 20 input parameters in 20x20x1 architecture with testing average absolute percentage error (AAPE) of 2.24%. The other ANN model (Model B) had four input parameters (SO3, C3S, Blaine and total alkali amount) in 4x4x1 architecture with AAPE of 2.41%. Both of the Model A and the Model B were created in the MatLAB® environment by writinga custom computer code. The last ANN model (Model C) actually refers to 72 differentANN models created in the MatLAB® neural networks toolbox. In order to obtain a model with the lowest error, different learning algorithms, training functions and architectures in combinations were tested. The lowest AAPE among these models appeared to be 2.31%. The fuzzy logic model (Model D) which had four input parameters (SO3, C3S, Blaine and total alkali amount) was created in the MatLAB fuzzy logic toolbox. In order to write the fuzzy rules, the sensitivity analysis of the Model B was utilized. The AAPE of the Model D was 2.69%. The model was compared with the ANN models for its error levels and ease of application. The results indicated that through the application of fuzzy logic algorithm, a more user friendly andmore explicit model than the ANNs could be produced within successfully low error margins.