Artificial neural network simulation of combined humic substance coagulation and membrane filtration

Mohammed Al-Abri, Nidal Hilal

    Research output: Contribution to journalArticle

    Abstract

    Backpropagation artificial neural network (BPNN) was utilized to predict membrane performance. The network was used to predict and compare humic substance (HS) retention and membrane fouling with previously obtained experimental data. BPNN simulation results show high network reliability, if the network is implemented correctly. The difference between the predicted and experimental data was lower than 5%. Low number of training data input has been shown to hinder the learning process. A high number of training data input has lead to over-fitting or memorization of the training data set, reducing the networks predictability. The number of neurons in the hidden layers needs to be chosen carefully to obtain a reliable network. This paper shows that a lower number of neurons result in low reliability, while a higher number of neurons leads to data over-fitting. The best performance was obtained with 2-10 neurons for HS and heavy metals agglomeration and 5-15 neurons for HS coagulation with and without heavy metals.

    Original languageEnglish (US)
    Pages (from-to)27-34
    Number of pages8
    JournalChemical Engineering Journal
    Volume141
    Issue number1-3
    DOIs
    StatePublished - Jul 15 2008

    Fingerprint

    Humic Substances
    humic substance
    Coagulation
    coagulation
    artificial neural network
    Neurons
    membrane
    Neural networks
    Membranes
    simulation
    Heavy Metals
    Backpropagation
    Heavy metals
    heavy metal
    Membrane fouling
    agglomeration
    fouling
    Agglomeration
    learning

    Keywords

    • Artificial neural network
    • Membrane separation
    • Prediction

    ASJC Scopus subject areas

    • Chemical Engineering(all)
    • Environmental Engineering

    Cite this

    Artificial neural network simulation of combined humic substance coagulation and membrane filtration. / Al-Abri, Mohammed; Hilal, Nidal.

    In: Chemical Engineering Journal, Vol. 141, No. 1-3, 15.07.2008, p. 27-34.

    Research output: Contribution to journalArticle

    @article{84895d9ae3c64b6f91847b6227bbba9e,
    title = "Artificial neural network simulation of combined humic substance coagulation and membrane filtration",
    abstract = "Backpropagation artificial neural network (BPNN) was utilized to predict membrane performance. The network was used to predict and compare humic substance (HS) retention and membrane fouling with previously obtained experimental data. BPNN simulation results show high network reliability, if the network is implemented correctly. The difference between the predicted and experimental data was lower than 5{\%}. Low number of training data input has been shown to hinder the learning process. A high number of training data input has lead to over-fitting or memorization of the training data set, reducing the networks predictability. The number of neurons in the hidden layers needs to be chosen carefully to obtain a reliable network. This paper shows that a lower number of neurons result in low reliability, while a higher number of neurons leads to data over-fitting. The best performance was obtained with 2-10 neurons for HS and heavy metals agglomeration and 5-15 neurons for HS coagulation with and without heavy metals.",
    keywords = "Artificial neural network, Membrane separation, Prediction",
    author = "Mohammed Al-Abri and Nidal Hilal",
    year = "2008",
    month = "7",
    day = "15",
    doi = "10.1016/j.cej.2007.10.005",
    language = "English (US)",
    volume = "141",
    pages = "27--34",
    journal = "Chemical Engineering Journal",
    issn = "1385-8947",
    publisher = "Elsevier",
    number = "1-3",

    }

    TY - JOUR

    T1 - Artificial neural network simulation of combined humic substance coagulation and membrane filtration

    AU - Al-Abri, Mohammed

    AU - Hilal, Nidal

    PY - 2008/7/15

    Y1 - 2008/7/15

    N2 - Backpropagation artificial neural network (BPNN) was utilized to predict membrane performance. The network was used to predict and compare humic substance (HS) retention and membrane fouling with previously obtained experimental data. BPNN simulation results show high network reliability, if the network is implemented correctly. The difference between the predicted and experimental data was lower than 5%. Low number of training data input has been shown to hinder the learning process. A high number of training data input has lead to over-fitting or memorization of the training data set, reducing the networks predictability. The number of neurons in the hidden layers needs to be chosen carefully to obtain a reliable network. This paper shows that a lower number of neurons result in low reliability, while a higher number of neurons leads to data over-fitting. The best performance was obtained with 2-10 neurons for HS and heavy metals agglomeration and 5-15 neurons for HS coagulation with and without heavy metals.

    AB - Backpropagation artificial neural network (BPNN) was utilized to predict membrane performance. The network was used to predict and compare humic substance (HS) retention and membrane fouling with previously obtained experimental data. BPNN simulation results show high network reliability, if the network is implemented correctly. The difference between the predicted and experimental data was lower than 5%. Low number of training data input has been shown to hinder the learning process. A high number of training data input has lead to over-fitting or memorization of the training data set, reducing the networks predictability. The number of neurons in the hidden layers needs to be chosen carefully to obtain a reliable network. This paper shows that a lower number of neurons result in low reliability, while a higher number of neurons leads to data over-fitting. The best performance was obtained with 2-10 neurons for HS and heavy metals agglomeration and 5-15 neurons for HS coagulation with and without heavy metals.

    KW - Artificial neural network

    KW - Membrane separation

    KW - Prediction

    UR - http://www.scopus.com/inward/record.url?scp=44849141066&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=44849141066&partnerID=8YFLogxK

    U2 - 10.1016/j.cej.2007.10.005

    DO - 10.1016/j.cej.2007.10.005

    M3 - Article

    VL - 141

    SP - 27

    EP - 34

    JO - Chemical Engineering Journal

    JF - Chemical Engineering Journal

    SN - 1385-8947

    IS - 1-3

    ER -