Analisis kejadian hujan harian adalah penting, terutama sektor air yang berkaitan. Model hidrologi umumnya mengandungi parameter yang tidak dapat diukur secara langsung, tetapi boleh disimpulkan oleh kalibrasi terhadap hujan harian rekod lama. Model Markov Chain akan digunapakai untuk kajian ini. Kajian ini menggambarkan satu pendekatan analitikal bagi ciri-ciri proses curahan hujan harian lama untuk merancang data hujan masa hadapan. Stesen hujan harian yang terletak di Johor, Kuala Lumpur dan Pulau Pinang dipilih. Data akan dilengkapkan dengan mengisi kehilangan data curahan hujan, kemudian akan dianalisa dengan menganggarka mean, skewness, standard deviation dan lag-one dengan membandingkan pola data curah hujan harian rekod lama dan sintetik. Kaedah stokastik untuk transformasi data curah hujan harian menggunakan lag-satu model Markov Chain untuk menghasilkan sintetik data curah hujan harian. Perbandingan statik dibuat antara data yang dihasilkan (output model) dengan data rekod lama curah hujan harian untuk melihat persamaan antara satu sama lain. Statistik dari data yang dihasilkan adalah purata, skewness, standard deviasi dan lag-satu dibandingkan dengan data yang direkodkan, dan menunjukkan persamaan antara satu sama lain. Kesamaan statistik antara setiapnya menunjukan model adalah benar.
___________________________________________________________________________________
The analysis of the daily rainfall occurrence behavior is becoming more important, particularly in water related sectors. Hydrological models generally contain parameter that cannot be measured directly, but can only be meaningfully inferred by calibration against a historical record of input-output data. In this paper, recently adopted Markov chain model will be applied. This paper describes an analytical approach to the characterization of long-time daily rainfall process for planning of the future rainfall data. The point daily rainfall stations located at Johor, Kuala Lumpur and Pulau Pinang are chosen.
With 20 years of daily rainfall data was used in generated data. Data will be completed by fill up the missing data then analysis the data by estimating the mean, standard deviation, skewness and lag-one, the pattern of daily rainfall record data will be compared to synthetic data. A stochastic method for the transformation of equal data daily rainfall are using lag-one Markov Chain model is applied to in producing a synthetic generation of daily rainfall data. Statically comparisons were made between the generated data (output of the model) with the equal data of daily rainfall data. These comparisons are done to compare the statistics between the actual with the generated data to see whether they are similar with each other. The statistics of the generated data are mean, standard deviation, skew and lag-one were compared with the recorded data, and show a similarity between each other. The similarity of the statistics between each other means that the modeling is true.
Stocahastica rainfall data generation using lag-one markov chain modelo transformed data_Nurul Syahiddatina Mohd Ali_A9_2011_875004115_00003089671_NI