Prediction of methane emission and electricity generation from landfills: Deep learning approach

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F24%3A10256906" target="_blank" >RIV/61989100:27240/24:10256906 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.sciencedirect.com/science/article/pii/S2352484724007467" target="_blank" >https://www.sciencedirect.com/science/article/pii/S2352484724007467</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1016/j.egyr.2024.11.025" target="_blank" >10.1016/j.egyr.2024.11.025</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Prediction of methane emission and electricity generation from landfills: Deep learning approach

Popis výsledku v původním jazyce

This paper presents a novel deep learning (DL) framework for predicting methane emissions from landfills by analyzing images of solid waste mixtures. The generation of renewable and sustainable electricity from landfills is a rapidly developing field that has gained considerable interest due to its potential to reduce greenhouse gas (GHG) emissions. Methane is a potent GHG produced during the decomposition of solid waste in landfills, significantly contributing to climate change. Precise estimation of methane emissions is crucial for efficient landfill management and the development of mitigation strategies. The framework is composed of four phases. During the initial phase, object detection identifies individual items within the landfill waste images. Subsequently, a hybrid Inception-ResNet-V2 DL method is suggested for categorizing the resulting waste image data into four classes: organic, paper and textile, wood, and food waste. In the next phase, the total weight of waste for each of the four classes is calculated utilizing a suggested method called waste weight estimation (WWE). Following that, methane production is estimated in the third phase using a proposed method (CH4PE), which relies on the weight values of waste mixtures. Finally, the fourth phase entails estimating electricity production using the electricity production estimation (EPE) method, which is based on the obtained methane weight. The validation findings revealed that the classification accuracy attained 93 %. Additionally, the framework&apos;s findings confirm a total estimated methane weight of 39,695.03 kg in the landfill case study, along with an estimated electricity production of 474,449.7 kW/month. This framework presents numerous advantages over conventional methods for predicting landfill-based methane emissions. It eliminates the necessity for labour-intensive manual waste classification and offers scalability, enabling application on large-scale landfill sites. This scalability facilitates the identification of high-emission areas and optimization of gas collection systems, fostering green and sustainable electricity production. © 2024 The Authors

Název v anglickém jazyce

Prediction of methane emission and electricity generation from landfills: Deep learning approach

Popis výsledku anglicky

This paper presents a novel deep learning (DL) framework for predicting methane emissions from landfills by analyzing images of solid waste mixtures. The generation of renewable and sustainable electricity from landfills is a rapidly developing field that has gained considerable interest due to its potential to reduce greenhouse gas (GHG) emissions. Methane is a potent GHG produced during the decomposition of solid waste in landfills, significantly contributing to climate change. Precise estimation of methane emissions is crucial for efficient landfill management and the development of mitigation strategies. The framework is composed of four phases. During the initial phase, object detection identifies individual items within the landfill waste images. Subsequently, a hybrid Inception-ResNet-V2 DL method is suggested for categorizing the resulting waste image data into four classes: organic, paper and textile, wood, and food waste. In the next phase, the total weight of waste for each of the four classes is calculated utilizing a suggested method called waste weight estimation (WWE). Following that, methane production is estimated in the third phase using a proposed method (CH4PE), which relies on the weight values of waste mixtures. Finally, the fourth phase entails estimating electricity production using the electricity production estimation (EPE) method, which is based on the obtained methane weight. The validation findings revealed that the classification accuracy attained 93 %. Additionally, the framework&apos;s findings confirm a total estimated methane weight of 39,695.03 kg in the landfill case study, along with an estimated electricity production of 474,449.7 kW/month. This framework presents numerous advantages over conventional methods for predicting landfill-based methane emissions. It eliminates the necessity for labour-intensive manual waste classification and offers scalability, enabling application on large-scale landfill sites. This scalability facilitates the identification of high-emission areas and optimization of gas collection systems, fostering green and sustainable electricity production. © 2024 The Authors

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Energy Reports

ISSN

2352-4847

e-ISSN

—

Svazek periodika

12

Číslo periodika v rámci svazku

Dec

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

11

Strana od-do

5462-5472

Kód UT WoS článku

001361944800001

EID výsledku v databázi Scopus

2-s2.0-85209542954