Artificial Neural Network Models for Rainfall Prediction

Abstract

Rainfall prediction is an important meteorological problem that can greatly affect humanity in areas such as agriculture production, flooding, drought, and sustainable management of water resources. The dynamic and nonlinear nature of the climatic conditions have made it impossible for traditional techniques to yield satisfactory accuracy for rainfall prediction. As a result of the sophistication of climatic processes that produced rainfall, using quantitative techniques to predict rainfall is a very cumbersome task. The paper proposed four non-linear techniques such as Artificial Neural Networks (ANN) for rainfall prediction. ANN has the capacity to map different input and output patterns. The Feed Forward Neural Network (FFNN), Cascade Forward Neural Network (CFNN), Recurrent Neural Network (RNN), and Elman Neural Network (ENN) were used to predict rainfall. The dataset used for this work contains some meteorological variables such as temperature, wind speed, humidity, rainfall, visibility, and others for the year 2015-2019. Simulation results indicated that of all the proposed Neural Network (NN) models, the Elman NN model produced the best performance. We also found out that Elman NN has the best performance for the year 2018 (having the lowest RMSE, MSE, and MAE of 6.360, 40.45, and 0.54 respectively). The results indicated that NN algorithms are robust, dependable, and reliable algorithms that can be used for daily, monthly, or yearly rainfall prediction.

Comparison of Adaptive Holt-Winters Exponential Smoothing and Recurrent Neural Network Model for Forecasting Rainfall in Malang City

“…For rainfall data in Malang City, the RNN model provides a better model than adaptive Holts-Winters exponential smoothing model. On the other hand, these results are in line with Krichene, Masmoudi, Alimi, Abraham, and Chabchoub (2017) and Dada, Yakubu, and Oyewola (2021) showing that the RNN method outperforms other models.…”

Prediction of Electric Power Demand of HVACs for Operating Rooms in Case of Dynamic Set Points of Temperature: A Case Study

“…The performance of the ANN is dependent on it, thus choosing the best value for neurons is crucial. If the network's number of neurons is less than the ideal, the network will not train appropriately, and the results will be inaccurate [37]. Furthermore, if a large number of neurons is used in comparison to the optimum amount, poor interpolation quality might arise, which is known as an over-fitting problem [38], [39].…”

Precipitation forecast using RNN variants by analyzing Optimizers and Hyperparameters for Time-series based Climatological Data

“…ML algorithms, such as Regression [15], Support Vector Machine (SVM) [7], Decision Trees (DT) [16,17], Naive Bayes [18] and K-Nearest Neighbors (KNN) [19] have been effectively used to construct precipitation prediction or classification models in a variety of domains. DL algorithms, such as Artificial Neural Networks (ANN) [20], Recurrent Neural Networks (RNN) [21,22], Convolutional Neural Networks (CNN) [23] and Generative Adversarial Networks (GAN) [24] play an essential role in processing and analyzing massive amounts of precipitation data to deliver meaningful information. Venkatesh, et al (2021) [25] constructed a precipitation prediction system using GAN with a CNN upon time-series annual precipitation data of 36 subdivisions in India from 1901 to 2015.…”

“…The LSTM model accurately predicted the precipitation with RMSE values of 5.343, 6.280, and 7.706 for the three regions respectively. Dada et al [21] proposed four Neural Network models: Feed Forward Neural Network (FFNN), RNN, Elman Neural Network (ENN) and Cascade Forward Neural Network (CFNN) for predicting precipitation using India's precipitation data from the Kaggle repository. Additionally, it is clear from the statistical findings that the ENN model outperformed the other three models.…”

“…Time-series data-based precipitation forecast can aid in the decision-making processes associated with flood and disaster to manage, flood, control floods and plan safety preparations [31,32]. Also, based on a survey of related work [21,[25][26][27][28][29][30][31], it has been determined that articles for predicting precipitation use DL neural network models such as LSTM, ANN, GRU, FFNN, RNN, ENN, and CFNN. Also, measures such as RMSE and MAE are used to assess the expected precipitation.…”