Examining the Applicability of Wavelet Packet Decomposition on Different Forecasting Models in Annual Rainfall Prediction

Accurate precipitation prediction can help plan for different water resources management demands and provide an extension of lead-time for the tactical and strategic planning of courses of action. This paper examines the applicability of several forecasting models based on wavelet packet decomposition (WPD) in annual rainfall forecasting, and a novel hybrid precipitation prediction framework (WPD-ELM) is proposed coupling extreme learning machine (ELM) and WPD. The works of this paper can be described as follows: (a) WPD is used to decompose the original precipitation data into several sub-layers; (b) ELM model, autoregressive integrated moving average model (ARIMA), and back-propagation neural network (BPNN) are employed to realize the forecasting computation for the decomposed series; (c) the results are integrated to attain the final prediction. Four evaluation indexes (RMSE, MAE, R, and NSEC) are adopted to assess the performance of the models. The results indicate that the WPD-ELM model outperforms other models used in this paper and WPD can significantly enhance the performance of forecasting models. In conclusion, WPD-ELM can be a promising alternative for annual precipitation forecasting and WPD is an effective data pre-processing technique in producing convincing forecasting models.

Download Full-text

Improving annual precipitation prediction using extreme learning machine with wavelet packet decomposition

10.21203/rs.3.rs-317538/v1 ◽

2021 ◽

Author(s):

Hua Wang ◽

Wenchuan Wang ◽

Yu-jin Du ◽

Dong-mei Xu ◽

Yi-duo Zhang

Keyword(s):

Extreme Learning Machine ◽

Wavelet Packet ◽

Moving Average ◽

Back Propagation ◽

Back Propagation Neural Network ◽

Annual Precipitation ◽

Wavelet Packet Decomposition ◽

Precipitation Forecasting ◽

Precipitation Prediction ◽

Learning Machine

Abstract Accurate precipitation prediction can help plan for different water resources management demands and provide an extension of lead-time for the tactical and strategic planning of courses of action as well as activity. In this paper, a novel merged precipitation prediction framework (ELM-WPD) is proposed on the Extreme learning machine (ELM) with wavelet packet decomposition (WPD). The model can be described as the following: (a) we use the WPD to decompose the original precipitation data into several sub-layers; (b) ELM model is employed to complete the forecasting calculation for the decomposed series; (c) the results are integrated to complete the final prediction. Four quantitative indexes (RMSE, MAE, R and NSEC) are employed for the comparison criteria. The results are compared with Back-propagation neural network (BPNN), autoregressive integrated moving average model (ARIMA), ELM, BPNN-WPD model, ARIMA-WPD, indicating that the ELM-WPD model has better performance than other models used in this paper. Hence, the proposed method can provide higher accuracy and reliability for annual precipitation forecasting and can be extended to similar situations for precipitation forecasting.

Download Full-text

A Comparison of BPNN, GMDH, and ARIMA for Monthly Rainfall Forecasting Based on Wavelet Packet Decomposition

Water ◽

10.3390/w13202871 ◽

2021 ◽

Vol 13 (20) ◽

pp. 2871

Author(s):

Wenchuan Wang ◽

Yujin Du ◽

Kwokwing Chau ◽

Haitao Chen ◽

Changjun Liu ◽

...

Keyword(s):

Wavelet Packet ◽

Arima Model ◽

Back Propagation ◽

Forecast Model ◽

Monthly Rainfall ◽

Ensemble Prediction ◽

Group Method ◽

Wavelet Packet Decomposition ◽

Rainfall Forecasting ◽

Bpnn Model

Accurate rainfall forecasting in watersheds is of indispensable importance for predicting streamflow and flash floods. This paper investigates the accuracy of several forecasting technologies based on Wavelet Packet Decomposition (WPD) in monthly rainfall forecasting. First, WPD decomposes the observed monthly rainfall data into several subcomponents. Then, three data-based models, namely Back-propagation Neural Network (BPNN) model, group method of data handing (GMDH) model, and autoregressive integrated moving average (ARIMA) model, are utilized to complete the prediction of the decomposed monthly rainfall series, respectively. Finally, the ensemble prediction result of the model is formulated by summing the outputs of all submodules. Meanwhile, these six models are employed for benchmark comparison to study the prediction performance of these conjunction methods, which are BPNN, WPD-BPNN, GMDH, WPD-GMDH, ARIMA, and WPD-ARIMA models. The paper takes monthly data from Luoning and Zuoyu stations in Luoyang city of China as the case study. The performance of these conjunction methods is tested by four quantitative indexes. Results show that WPD can efficiently improve the forecasting accuracy and the proposed WPD-BPNN model can achieve better prediction results. It is concluded that the hybrid forecast model is a very efficient tool to improve the accuracy of mid- and long-term rainfall forecasting.

Download Full-text

Rainfall Forecasting Using Various Artificial Neural Network Techniques - A Review

International Journal of Scientific Research in Computer Science Engineering and Information Technology ◽

10.32628/cseit2173159 ◽

2021 ◽

pp. 506-526

Author(s):

Nisha Thakur ◽

Sanjeev Karmakar ◽

Sunita Soni

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Moving Average ◽

Weather Prediction ◽

Optimization Technique ◽

Back Propagation ◽

Rainfall Forecasting ◽

Rainfall Prediction ◽

Artificial Neural ◽

Review Reports

The present review reports the work done by the various authors towards rainfall forecasting using the different techniques within Artificial Neural Network concepts. Back-Propagation, Auto-Regressive Moving Average (ARIMA), ANN , K- Nearest Neighbourhood (K-NN), Hybrid model (Wavelet-ANN), Hybrid Wavelet-NARX model, Rainfall-runoff models, (Two-stage optimization technique), Adaptive Basis Function Neural Network (ABFNN), Multilayer perceptron, etc., algorithms/technologies were reviewed. A tabular representation was used to compare the above-mentioned technologies for rainfall predictions. In most of the articles, training and testing, accuracy was found more than 95%. The rainfall prediction done using the ANN techniques was found much superior to the other techniques like Numerical Weather Prediction (NWP) and Statistical Method because of the non-linear and complex physical conditions affecting the occurrence of rainfall.

Download Full-text

A rainfall forecasting method using machine learning models and its application to the Fukuoka city case

International Journal of Applied Mathematics and Computer Science ◽

10.2478/v10006-012-0062-1 ◽

2012 ◽

Vol 22 (4) ◽

pp. 841-854 ◽

Cited By ~ 33

Author(s):

S. Monira Sumi ◽

M. Faisal Zaman ◽

Hideo Hirose

Keyword(s):

Machine Learning ◽

Moving Average ◽

Principal Component ◽

Processing Technique ◽

Monthly Rainfall ◽

Multivariate Adaptive Regression Splines ◽

Support Vector ◽

Rainfall Forecasting ◽

Cross Validation Error ◽

Leave One Out

In the present article, an attempt is made to derive optimal data-driven machine learning methods for forecasting an average daily and monthly rainfall of the Fukuoka city in Japan. This comparative study is conducted concentrating on three aspects: modelling inputs, modelling methods and pre-processing techniques. A comparison between linear correlation analysis and average mutual information is made to find an optimal input technique. For the modelling of the rainfall, a novel hybrid multi-model method is proposed and compared with its constituent models. The models include the artificial neural network, multivariate adaptive regression splines, the k-nearest neighbour, and radial basis support vector regression. Each of these methods is applied to model the daily and monthly rainfall, coupled with a pre-processing technique including moving average and principal component analysis. In the first stage of the hybrid method, sub-models from each of the above methods are constructed with different parameter settings. In the second stage, the sub-models are ranked with a variable selection technique and the higher ranked models are selected based on the leave-one-out cross-validation error. The forecasting of the hybrid model is performed by the weighted combination of the finally selected models.

Download Full-text

Comparative Study of Daily Rainfall Forecasting Models Using Adaptive-Neuro Fuzzy Inference System (ANFIS)

Current World Environment ◽

10.12944/cwe.10.2.19 ◽

2015 ◽

Vol 10 (2) ◽

pp. 529-536 ◽

Cited By ~ 4

Author(s):

M. A Sojitra ◽

P. A Pandya

Keyword(s):

Relative Humidity ◽

Fuzzy Inference System ◽

Fuzzy Inference ◽

Moving Average ◽

Daily Rainfall ◽

Rainfall Forecasting ◽

Inference System ◽

Forecasting Models ◽

Mean Temperature ◽

Neuro Fuzzy

The study was carried out to develop rainfall forecasting Models. Adaptive Neuro-Fuzzy Inference System (ANFIS) was used for developing Models rainfall of Udaipur city. Two data sets were prepared using 35 year of weather parameters i.e. wet bulb temperature, mean temperature, relative humidity and evaporation of previous day and previous moving average week were used to prepare case I and case II respectively. Gaussian and Generalized Bell membership functions were used to prepare models. Statistical and hydrologic performance indices of ANFIS (Gaussian, 5) gave better performance among developed four models. The study showed that sensitivity analysis revealed wet bulb temperature is most sensible parameter followed by mean temperature, relative humidity and evaporation.

Download Full-text

Some statistical and CI models to predict chaotic high-frequency financial data

Journal of Intelligent & Fuzzy Systems ◽

10.3233/jifs-189107 ◽

2020 ◽

Vol 39 (5) ◽

pp. 6419-6430

Author(s):

Dusan Marcek

Keyword(s):

Time Series Data ◽

Moving Average ◽

Methodological Approach ◽

Back Propagation ◽

Large Data ◽

Series Data ◽

Data Set ◽

Training Time ◽

Optimal Population ◽

Forecast Time

To forecast time series data, two methodological frameworks of statistical and computational intelligence modelling are considered. The statistical methodological approach is based on the theory of invertible ARIMA (Auto-Regressive Integrated Moving Average) models with Maximum Likelihood (ML) estimating method. As a competitive tool to statistical forecasting models, we use the popular classic neural network (NN) of perceptron type. To train NN, the Back-Propagation (BP) algorithm and heuristics like genetic and micro-genetic algorithm (GA and MGA) are implemented on the large data set. A comparative analysis of selected learning methods is performed and evaluated. From performed experiments we find that the optimal population size will likely be 20 with the lowest training time from all NN trained by the evolutionary algorithms, while the prediction accuracy level is lesser, but still acceptable by managers.

Download Full-text