kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Discharge prediction for rectangular sharp-crested weirs by machine learning techniques
KTH, School of Architecture and the Built Environment (ABE), Civil and Architectural Engineering, Concrete Structures.ORCID iD: 0000-0002-5239-6559
KTH, School of Architecture and the Built Environment (ABE), Sustainable development, Environmental science and Engineering, Resources, Energy and Infrastructure.ORCID iD: 0000-0002-4242-3824
KTH, School of Architecture and the Built Environment (ABE), Civil and Architectural Engineering, Concrete Structures.ORCID iD: 0000-0001-8336-1247
(English)Manuscript (preprint) (Other academic)
Abstract [en]

The stage-discharge relationship of a weir is essential for a posteriori calculations of flow discharges. Conventionally, it is determined by regression methods, which is time-consuming and may subject to limited prediction accuracy. With the intention to provide better estimate, the machine learning models, artificial neural network (ANN), support vector machine (SVM) and extreme learning machine (ELM), are assessed for prediction of discharges of rectangular sharp-crested weirs. A large number of experimental data sets are adopted to develop and calibrate these models. Different input scenarios and data management strategies are employed for optimization of the models, for which performance is evaluated in the light of statistical criteria. The results show that all three models are capable of predicting the discharge coefficient with high accuracy, but the SVM exhibits somewhat better performance. Its maximum and mean relative error are respectively 5.44 and 0.99%, and 99% of the predicted data show an error below 5%. The coefficient of determination and root mean square error are 0.95 and 0.01, respectively. The model sensitivity is examined, indicative of the dominant roles of weir Reynolds number and contraction ratio in discharge estimation. The existing empirical formulas are assessed and compared against the machine learning models. It is found that the relationship proposed by Vatankhah exhibits a highest accuracy. However, it is still less accurate than the machine learning approaches. The study is intended to provide reference for discharge determination of overflow structures including spillways.

National Category
Civil Engineering
Identifiers
URN: urn:nbn:se:kth:diva-289586OAI: oai:DiVA.org:kth-289586DiVA, id: diva2:1525588
Note

QC 20210209

Available from: 2021-02-04 Created: 2021-02-04 Last updated: 2022-06-25Bibliographically approved
In thesis
1. Numerical simulations of flow discharge and behaviours in spillways
Open this publication in new window or tab >>Numerical simulations of flow discharge and behaviours in spillways
2021 (English)Licentiate thesis, comprehensive summary (Other academic)
Abstract [en]

A spillway is an important component of a dam and serves as a flood release structure. It achieves controlled discharge of water and protects the dam from overtopping. The majority of the hydropower dams were built before the 1980s, and many spillways are undersized in light of the present design flood guidelines. Another issue that arises in connection with the high design floods is the energy dissipation capacity. Many existing energy-dissipating arrangements are insufficient or construed only for a design flood standard at the time of dam construction. The increment in the flood discharges requires that the energy dissipation should be improved to obtain sufficient capacity or higher efficiency. In addition, the high-velocity flow is a major concern in the design of spillways. If the flow velocity exceeds approximately 20 m/s, the risk of cavitation may arise. In Sweden, many dams belong to this category. To address these issues, an assessment of their discharge behaviours is required. Innovative engineering solutions for better energy dissipation and cavitation mitigation are also necessary for safe operation. This thesis presents machine learning based methods for discharge estimation. Three data-driven models are developed to study the discharge behaviours of the overflow weirs. Their reliability is validated through the comparison with the experimental and empirical results. These models are capable of giving accurate predictions and show superiority over the conventional approaches. With high accuracy and adaptability, data-driven models are an effective and fast alternative for spillway discharge prediction. This research also focuses on the hydraulic design of stepped spillways, aiming to devise innovative engineering solutions to enhance energy dissipation and reduce cavitation risks. Consequently, several unconventional step layouts are conceived and their hydraulic behaviours are investigated. The modified configurations include steps with chamfers and cavity blockages, non-linear steps and inclined steps. This part attempts to gain insight into the effects of the step geometries on the spillway hydraulics via computational fluid dynamics, which provides references for engineering applications.

Abstract [sv]

Ett utskov är en viktig komponent i en damm och fungerar som ett skydd mot översvämning. Det avbördar vatten på ett kontrollerat sätt och skyddar dammen från överströmning. Majoriteten av vattenkraftsdammarna byggdes före 1980-talet och många utskov är underdimensionerade i förhållande till de nuvarande riktlinjerna för utformning med avseende på dimensionerande flöden. En annan fråga som uppstår i samband med höga flöden är energiomvandlingskapaciteten. Många befintliga arrangemang för reducering omvandling av vattnets rörelseenergi är otillräckliga eller endast anpassade för det dimensionerande flöde som gällde vid tidpunkten för dammens uppförande. En avbördningsökning kräver i sin tur att energiomvandlingsförmågan förbättras för att uppnå tillräcklig kapacitet eller högre effektivitet. Dessutom är höghastighetsflödet ett stort bekymmer vid utformningen av utskov. Om flödeshastigheten överstiger t.ex. 20 m/s uppstår risk för kavitation i vattenvägar. I Sverige hör många dammar till denna kategori. För att lösa dessa problemställningar behöver en utvärdering av avbördningsanordningar göras. Innovativa tekniska lösningar som syftar till effektiv hantering av flödesenergi och kavitationsreducering, vilka utgör nödvändiga förutsättningar för säker drift av anläggningar.

Denna uppsats presenterar maskininlärningsbaserade metoder för att prognostisera avbördning i dammar. Tre datadrivna modeller har utvecklats för att studera avbördningsegenskaper hos överfallsdammarna. Deras tillförlitlighet valideras genom jämförelse med experimentella och empiriska resultat. Modellerna kan ge noggrann uppskattning, som kan användas som ett tillförlitligt alternativ för bestämning av avbördning. Forskningen fokuserar också på den hydrauliska utformningen av stegade bräddavlopp (s.k. stepped spillway), i syfte att utveckla innovativa tekniska lösningar för att åstadkomma hög energiförlust och minska kavitationsrisker. Flera okonventionella stegformade geometrier föreslås och deras hydrauliska egenskaper undersöks. Denna del syftar till att, via numerisk simulering, ge en inblick i vilka effekter olika steggeometrier har på avbördningshydrauliken, vilken tillhandahåller referens för tekniska applikationer.

Place, publisher, year, edition, pages
Stockholm: KTH Royal Institute of Technology, 2021. p. 48
Series
TRITA-ABE-DLT ; 2021:211
Keywords
Spillway, step geometry, discharge prediction, flow behaviour, machine learning, computational fluid dynamics, Utskov, steggeometri, avbördningsuppskattning, flödesbeteende strömningsegenskap, maskininlärning, flödesdynamisk beräkningsanalys numerisk simulering
National Category
Civil Engineering
Research subject
Civil and Architectural Engineering, Concrete Structures
Identifiers
urn:nbn:se:kth:diva-289591 (URN)978-91-7873-736-9 (ISBN)
Presentation
2021-03-01, Videolänk https://kth-se.zoom.us/j/65774256738, Du som saknar dator /datorvana kontakta Anders Ansell anders.ansell@byv.kth.se / Use the e-mail address if you need technical assistance., Stockholm, Stockholm, 13:00 (English)
Opponent
Supervisors
Note

QC 20210205

Available from: 2021-02-05 Created: 2021-02-04 Last updated: 2022-09-19Bibliographically approved

Open Access in DiVA

No full text in DiVA

Authority records

Li, ShichengYang, JamesAnsell, Anders

Search in DiVA

By author/editor
Li, ShichengYang, JamesAnsell, Anders
By organisation
Concrete StructuresResources, Energy and Infrastructure
Civil Engineering

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 168 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf