Time filter

Source Type

Turku, Finland

Bulsari A.,Nonlinear Solutions Oy | Keife H.,Aurubis Sweden AB | Geluk J.,Aurubis Netherlands BV
Advanced Materials and Processes

A research conducted at the Aurubis Zutphen plant illustrates how nonlinear modeling is an efficient approach for relating composition and process variables of annealing with resulting grain size. Nonlinear models were developed entirely from production data, and were implemented in suitable software for use by plant operators. The quality of the nonlinear model of grain size was quite good considering it was developed from plain production data, and measurements of grain size were done manually, making them subjective to some extent. The nonlinear model showed correct effects of input variables, and the correlation coefficient was above 86%. The standard deviation of the prediction error was about 4.2 μm. Nonlinear models were implemented in software suitable for use in metals industries. Models were tested by Aurubis and found to be quite good and useful. Source

Bulsari A.,Nonlinear Solutions Oy
Materials World

Abhay Bulsari examines the benefits of nonlinear modeling for hardness and thermal conductivity. Copper alloys are commonly used in applications that require high electrical and thermal conductivity, high hardness and good resistance to softening at elevated temperatures. Mathematical modeling can be performed in different ways, and a variety of these are suitable for different situations. A total of 13 experiments have been carried out in a laboratory oven at five temperatures for periods of up to three hours, resulting in 66 observations. Nonlinear models for hardness and conductivity have been developed from the experimental data using NLS 020 software. The root mean square (rms) error was 1 .39HV for hardness and 2.16% International Annealed Copper Standard (IACS) for conductivity, which amount to correlations of 99.01% and 98.05% respectively. Source

Bulsari A.,Nonlinear Solutions Oy | Wemberg A.,Fortum | Anttila A.,Fortum | Multas A.,Fortum
Energy and Environment

Coal fired power plants should be operated in such a way that the emissions are kept clearly below desired limits and the combustion efficiency is as high as can be achieved. This requires a lot of quantitative knowledge of the effects of the process variables and fuel characteristics on the emissions and efficiency. Mathematical models can be developed with different approaches. Physical models are too slow to be used for on-line process guidance, and require too many assumptions and simplifications. It is feasible to develop empirical or semi-empirical models from normal production data of the power plant. This technical communication explains with an example of a coal fired power plant how nonlinear models are an effective means of determining the best operating conditions at any given load for a given type of coal. Source

Bulsari A.,Nonlinear Solutions Oy | Vuoristo I.,Luvata Oy | Koppinen I.,Luvata Oy
Advanced Materials and Processes

Nonlinear models for hardness and conductivity were developed from experimental data. The models had three input variables, such as heat treatment time, temperature, and initial conductivity. The nonlinear models were in the form of feed-forward neural networks with sigmoidal activation functions on the hidden layer. The nonlinear models show very good statistical characteristics. The rms error was 1.993 for hardness and 1.587 for conductivity, which amount to correlations of 98.44% and 99.05% respectively. At higher temperatures, the conductivity rises faster and to higher levels. Higher initial conductivities lead to higher conductivities after aging, but this is not desirable from the point of view of hardness. It is observed that systematic experimentation followed by nonlinear modeling of precipitation hardening in certain copper alloys makes it possible to find the best operating conditions with less effort than trial and error experimentation. Source

Bulsari A.,Nonlinear Solutions Oy | Ilomaki J.,Nokian Tyres PLC | Lahtinen M.,Nokian Tyres PLC | Perkio R.,Nokian Tyres PLC
Rubber World

Nonlinear modeling is empirical or semi-empirical modeling which takes at least some nonlinearities into account. Feed-forward neural networks have several features which make them better tools for nonlinear empirical modeling. Besides their universal approximation capability it is usually possible to produce nonlinear models with some extrapolation capabilities with feed-forward neural networks. In the recent work of Nokian Tyres, an experimental approach was preferred. The experiments were planned such that sufficient information on nonlinearities can be extracted from the experimental data. The raw experimental data were analyzed and preprocessed, after which nonlinear models were developed and tested for several material properties over a period of a few months using the NLS 020 software. Nonlinear models were developed for Mooney viscosity before vulcanization and for several material properties of the compounded rubbers after vulcanization. Source

Discover hidden collaborations