Algoritmi strojnog učenja u ekoinženjerstvu

undergraduate thesis

• Title (english): Machine learning algorithms in environmental engineering
• Author: Matej Viljevac
• Mentor: Miroslav Jerković (mentor)
• Committee member: Miroslav Jerković (predsjednik povjerenstva)
• Committee member: Marija Vuković Domanovac (član povjerenstva)
• Committee member: Erna Begović Kovač (član povjerenstva)
• Granter: University of Zagreb; Faculty of Chemical Engineering and Technology; Zagreb
• Defense date and country: 2021-09-22, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

Otpadne vode jedan su od glavnih problema kojima se bave ekoinženjeri. One su često kontaminirane raznim tvarima (opasnim ili neopasnim) nastalim iz različitih izvora, a najčešće ljudskim aktivnostima. Ispuštanje otpadnih voda koje prethodno nisu pročišćene, odnosno čiji parametri nisu zadovoljavajući, značajno doprinose onečišćenju okoliša. Pokazatelji kakvoće otpadnih voda kao što su biokemijska potrošnja kisika (BPK5), kemijska potrošnja kisika (KPK), ukupne suspendirane čestice, nitrati, nitriti, ukupni dušik, amonij i ukupne otopljene tvari daju mjeru glavnih onečišćujućih tvari, a određivanje tih parametara zahtijeva skupe postupke analize. Kako bi se smanjili troškovi, a zadovoljili kriteriji, mogu se koristiti algoritmi strojnog učenja koji će nam s određenom točnošću procijeniti vrijednost nepoznatih parametara. U ovome radu koristit će se programski jezik R koji je jedan od pogodnijih za implementaciju raznih algoritama. Cilj je ovoga istraživanja unaprijediti tehnologije u analizi otpadnih voda i smanjiti uporabu opasnih otapala koja se koriste u postupcima određivanja parametara kakvoće otpadnih voda, zamjenjujući ih približno točnim procjenama parametara na osnovi primjene algoritama strojnog učenja.

Abstract (english)

One of the main issues environmental engineers are dealing with are waste waters. They are often contaminated with different substances (hazardous and non hazardous), emerged from different sources, most frequently from people. Discharge of waste water that is not previously purified, in other words, whose parameters are not satisfactory, significantly contributes to environmental pollution. Indicators of the waste water quality such as biochemical oxygen demand (BOD), chemical oxygen demand (COD), total suspended particles and total dissolved solids measure main pollutants and the determination of those parameters requires expensive anaylsis procedures. In order to reduce costs and satisfy criteria, it is possible to use machine learning algorithms that will, with certain accuracy, estimate the value of unknown parameters. A programming language R, which is one of the most adequate for implementation of different algorithms, will be used in this work. The aim of this research is to improve technologies in the waste water analysis and to reduce the usage of hazardous solvents that are used within procedures of determining parameters of waste water quality by replacing them with approximate assessments of parameters on the basis of machine learning algorithms.

• Keywords: strojno učenje; otpadne vode; programski jezik R; ekoinženjerstvo
• Keywords (english): machine learning; waste water; programming language R; environmental engineering
• Language: croatian
• URN:NBN: urn:nbn:hr:149:814604
• Study programme: Title: Environmental Engineering - Undergraduate study; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka ekoinženjerstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka ekoinženjerstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Access restricted to students and staff of home institution
• Created on: 2023-04-26 09:14:17