Wyniki wyszukiwania - BazTech

1

An attempt of optimization of zinc production line

Jarosz P., Kusiak J., Małecki S., Morkisz P., Oprocha P., Pietrucha W., Sztangret Ł.

Archives of Civil and Mechanical Engineering

|

2018

|

Vol. 18, no. 4

1116--1122

EN

The goal of the research is an attempt of optimization of the hydrometallurgy-based zinc production line, consisting of three stages: mixing of raw materials, oxidative roasting and leaching. The output product of one stage is an input to the next stage. Goal of mixing is preparation of zinc concentrates mix on the basis of zinc concentrates originated from different mines. The output semi-product of the next stage, the oxidative roasting process, is calcine, which is the input of the leaching. The result of the leaching is zinc sulfate solution and the goal of leaching is to carry out the maximum amount of zinc to solution. The preliminary step of any optimization is modeling of the analyzed processes. Modeling of considered three stages of zinc production line, based on the real industrial data of one of zinc production plants, was performed using different techniques. The elaborated models were the basis of the optimization for given objective functions of each of the production stages. The optimization methodology of multi-stage processes developed by the authors was applied. Obtained results of modeling and optimization are presented.

2

Short review of dimensionality reduction methods for failure detection

Pocha A., Misztal K., Morkisz P.

Schedae Informaticae

|

2017

|

Vol. 26

69--78

EN

Size of a dataset is often a challenge in real-life applications. Especially, when working with time series data, when the next sample is produced every few milliseconds and can include measurements from hundreds of sensors, one has to take dimensionality of the data into consideration. In this work, we compare various dimensionality reduction methods for time series data and check their performance on a failure detection task. We work on sensory data coming from existing machines.

3

Failures prediction based on performance monitoring of a gas turbine: a binary classification approach

Mulewicz B., Marzec M., Morkisz P., Oprocha P.

Schedae Informaticae

|

2017

|

Vol. 26

9--21

EN

This paper is dedicated to employ novel technique of deep learning for machines failures prediction. General idea of how to transform sensor data into suitable data set for prediction is presented. Then, neural network architecture that is very successful in solving such problems is derived. Finally, we present a case study for real industrial data of a gas turbine, including results of the experiments.