Reinforcement learning model for the operation of conjunctive use schemes

F. J C Bouchart, H. Chkam

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    A Reinforcement Learning (RL) model is proposed to identify operational strategies for conjunctive use schemes. This model is based on neuron-like adaptive elements that learn on-line to avoid system failures. The strength of this approach is that the resulting operational strategy for the water supply scheme reflects the need to respond and adapt to discrete failure events. A second advantage of the RL methodology is the inherent ability of control elements to operate in a distributed manner. By responding to local state inputs combined with a combination of local and global performance signals, the individual control elements are capable of operating effectively with only a limited set of state variables. The implication of such localized control is the avoidance of the curse of dimensionality commonly exhibited in other methodologies. Application of the RL model is demonstrated using the Burncrooks reservoir complex in Scotland. The model learns to effectively avoid failures, resulting in improved operational reliability.

    Original languageEnglish
    Title of host publicationInternational Conference on Hydraulic Engineering Software, Hydrosoft, Proceedings
    Subtitle of host publicationProceedings of the 1998 7th International Conference on Hydraulic Engineering Software, HYDROSOFT; Villa Olmo, Italy; ; 1 September 1998 through 1 September 1998
    Pages319-329
    Number of pages11
    Publication statusPublished - 1998
    EventProceedings of the 1998 7th International Conference on Hydraulic Engineering Software, HYDROSOFT - Villa Olmo, Italy
    Duration: 1 Sep 19981 Sep 1998

    Conference

    ConferenceProceedings of the 1998 7th International Conference on Hydraulic Engineering Software, HYDROSOFT
    CityVilla Olmo, Italy
    Period1/09/981/09/98

    Fingerprint

    Reinforcement learning
    Water supply
    Neurons

    Cite this

    Bouchart, F. J. C., & Chkam, H. (1998). Reinforcement learning model for the operation of conjunctive use schemes. In International Conference on Hydraulic Engineering Software, Hydrosoft, Proceedings: Proceedings of the 1998 7th International Conference on Hydraulic Engineering Software, HYDROSOFT; Villa Olmo, Italy; ; 1 September 1998 through 1 September 1998 (pp. 319-329)
    Bouchart, F. J C ; Chkam, H. / Reinforcement learning model for the operation of conjunctive use schemes. International Conference on Hydraulic Engineering Software, Hydrosoft, Proceedings: Proceedings of the 1998 7th International Conference on Hydraulic Engineering Software, HYDROSOFT; Villa Olmo, Italy; ; 1 September 1998 through 1 September 1998. 1998. pp. 319-329
    @inproceedings{c8014c4cc62649a1a28b26a42274888e,
    title = "Reinforcement learning model for the operation of conjunctive use schemes",
    abstract = "A Reinforcement Learning (RL) model is proposed to identify operational strategies for conjunctive use schemes. This model is based on neuron-like adaptive elements that learn on-line to avoid system failures. The strength of this approach is that the resulting operational strategy for the water supply scheme reflects the need to respond and adapt to discrete failure events. A second advantage of the RL methodology is the inherent ability of control elements to operate in a distributed manner. By responding to local state inputs combined with a combination of local and global performance signals, the individual control elements are capable of operating effectively with only a limited set of state variables. The implication of such localized control is the avoidance of the curse of dimensionality commonly exhibited in other methodologies. Application of the RL model is demonstrated using the Burncrooks reservoir complex in Scotland. The model learns to effectively avoid failures, resulting in improved operational reliability.",
    author = "Bouchart, {F. J C} and H. Chkam",
    year = "1998",
    language = "English",
    pages = "319--329",
    booktitle = "International Conference on Hydraulic Engineering Software, Hydrosoft, Proceedings",

    }

    Bouchart, FJC & Chkam, H 1998, Reinforcement learning model for the operation of conjunctive use schemes. in International Conference on Hydraulic Engineering Software, Hydrosoft, Proceedings: Proceedings of the 1998 7th International Conference on Hydraulic Engineering Software, HYDROSOFT; Villa Olmo, Italy; ; 1 September 1998 through 1 September 1998. pp. 319-329, Proceedings of the 1998 7th International Conference on Hydraulic Engineering Software, HYDROSOFT, Villa Olmo, Italy, 1/09/98.

    Reinforcement learning model for the operation of conjunctive use schemes. / Bouchart, F. J C; Chkam, H.

    International Conference on Hydraulic Engineering Software, Hydrosoft, Proceedings: Proceedings of the 1998 7th International Conference on Hydraulic Engineering Software, HYDROSOFT; Villa Olmo, Italy; ; 1 September 1998 through 1 September 1998. 1998. p. 319-329.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    TY - GEN

    T1 - Reinforcement learning model for the operation of conjunctive use schemes

    AU - Bouchart, F. J C

    AU - Chkam, H.

    PY - 1998

    Y1 - 1998

    N2 - A Reinforcement Learning (RL) model is proposed to identify operational strategies for conjunctive use schemes. This model is based on neuron-like adaptive elements that learn on-line to avoid system failures. The strength of this approach is that the resulting operational strategy for the water supply scheme reflects the need to respond and adapt to discrete failure events. A second advantage of the RL methodology is the inherent ability of control elements to operate in a distributed manner. By responding to local state inputs combined with a combination of local and global performance signals, the individual control elements are capable of operating effectively with only a limited set of state variables. The implication of such localized control is the avoidance of the curse of dimensionality commonly exhibited in other methodologies. Application of the RL model is demonstrated using the Burncrooks reservoir complex in Scotland. The model learns to effectively avoid failures, resulting in improved operational reliability.

    AB - A Reinforcement Learning (RL) model is proposed to identify operational strategies for conjunctive use schemes. This model is based on neuron-like adaptive elements that learn on-line to avoid system failures. The strength of this approach is that the resulting operational strategy for the water supply scheme reflects the need to respond and adapt to discrete failure events. A second advantage of the RL methodology is the inherent ability of control elements to operate in a distributed manner. By responding to local state inputs combined with a combination of local and global performance signals, the individual control elements are capable of operating effectively with only a limited set of state variables. The implication of such localized control is the avoidance of the curse of dimensionality commonly exhibited in other methodologies. Application of the RL model is demonstrated using the Burncrooks reservoir complex in Scotland. The model learns to effectively avoid failures, resulting in improved operational reliability.

    UR - http://www.scopus.com/inward/record.url?scp=0032277442&partnerID=8YFLogxK

    M3 - Conference contribution

    SP - 319

    EP - 329

    BT - International Conference on Hydraulic Engineering Software, Hydrosoft, Proceedings

    ER -

    Bouchart FJC, Chkam H. Reinforcement learning model for the operation of conjunctive use schemes. In International Conference on Hydraulic Engineering Software, Hydrosoft, Proceedings: Proceedings of the 1998 7th International Conference on Hydraulic Engineering Software, HYDROSOFT; Villa Olmo, Italy; ; 1 September 1998 through 1 September 1998. 1998. p. 319-329