UTAR Institutional Repository

Restoration Of Missing Meteorological Data For Long Term Monitoring Using Machine Learning

Lee, Shuan Chung (2021) Restoration Of Missing Meteorological Data For Long Term Monitoring Using Machine Learning. Final Year Project, UTAR.

[img] PDF
Restricted to Repository staff only until 31 December 2023.

Download (3335Kb)


    Given the growing demand for comprehensive meteorological data for further studies and project management, various machine learning methods for the restoration of missing data have been proposed in recent years, all of which are progressively more advanced than the conventional methods to fill the missing value in a data. Six different models involving feedforward neural networks such as multilayer perceptron (MLP) and radial basis function (RBF), and swarm intelligence optimization such as particle swarm optimization (PSO) and artificial bee colony (ABC), have been proposed to restore missing meteorological data. Proposed models are applied to complete a fifteen-year time series data of minimum temperature, maximum temperature, mean temperature, relative humidity, average wind speed, and evaporation data at twelve different weather stations in Malaysia. The models are validated by the statistical performance measures of root mean square error, mean absolute error and mean absolute percentage error. From the results, the MLP model has generated lower errors as compared to the RBF model. When comparing the single model with the hybrid model, the hybrid models, namely PSO-MLP, ABC-MLP, PSO-RBF, and ABC-RBF, have outperformed the single model and replaced the single model as the standout model for most meteorological parameters and weather stations. The performance of the models has also varied among the different locations. Results for the stations at the northern region of West Malaysia have generated higher errors, while the results at East Malaysia have generated lower errors. It can be concluded that the proposed models have shown suitability for imputing missing meteorological data and are thus recommended for other uses in other sectors where the missing values are of concern.

    Item Type: Final Year Project / Dissertation / Thesis (Final Year Project)
    Subjects: T Technology > TA Engineering (General). Civil engineering (General)
    Divisions: Lee Kong Chian Faculty of Engineering and Science > Bachelor of Engineering (Honours) Civil Engineering
    Depositing User: Sg Long Library
    Date Deposited: 10 Dec 2021 22:08
    Last Modified: 10 Dec 2021 22:08
    URI: http://eprints.utar.edu.my/id/eprint/4243

    Actions (login required)

    View Item