Weather Forecasting Using Merged Long Short-Term Memory Model (LSTM) and Autoregressive Integrated Moving Average (ARIMA) Model

Salman, Afan Galih; Heryadi, Yaya; Abdurahman, Edi; Suparta, Wayan

doi:10.3844/jcssp.2018.930.938

Cited by 37 publications

(15 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This shows that SVR is more suitable for long-term drought prediction than ARIMA model. LSTM has recently been used in many applications, including drought prediction (Poornima and Pushpalatha 2019), weather prediction (Salman et al 2018), and water table depth prediction (Zhang et al 2018), and has achieved good results because it can describe nonlinear relationships. Fig.…”

Section: Arima Svr and Lstm Model Predicts Resultsmentioning

confidence: 99%

Application of a hybrid ARIMA-LSTM model based on the SPEI for drought forecasting

Zhang

Ding

et al. 2021

Environ Sci Pollut Res

View full text Add to dashboard Cite

Drought forecasting can effectively reduce the risk of drought. We proposed a hybrid model based on deep learning methods that integrates an autoregressive integrated moving average (ARIMA) model and a long short-term memory (LSTM) model to improve the accuracy of short-term drought prediction. Taking China as an example, this paper compares and analyzes the prediction accuracy of six drought prediction models, ARIMA, support vector regression (SVR), LSTM, ARIMA-SVR, least square-SVR (LS-SVR) and ARIMA-LSTM, for SPEI. The performance of all the models was compared using measures of persistence, such as the Nash-Sutcliffe efficiency (NSE) and so on. The results show that all three hybrid models (ARIMA-SVR, LS-SVR and ARIMA-LSTM) had higher prediction accuracy than the single model. (ARIMA, SVR and LSTM), for a given lead time, at different scales. the NSEs of the hybrid ARIMA-SVR, LS-SVR and ARIMA-LSTM models for the predicted

show abstract

Section: Arima Svr and Lstm Model Predicts Resultsmentioning

confidence: 99%

Application of a hybrid ARIMA-LSTM model based on the SPEI for drought forecasting

Zhang

Ding

et al. 2021

Environ Sci Pollut Res

View full text Add to dashboard Cite

show abstract

“…The 96031station located with latitude and longitude coordinates are 3.61, 98.78. In this case, we focused on analyst the temperature data as time series recorded from each station in the Medan area and using LSTM to train, test, and predict the time series as a previous study [5]- [7]. In figure 1, as a brief depiction, the input of the LSTM cell could be a time arrangement set of information x that experiences a few sigmoid actuation doors σ.…”

Section: Methodsmentioning

confidence: 99%

“…As a type of recurrent neural network, LSTM can learn the order dependence between items in a sequence. It also had the potential of being able to learn the context required to make predictions in time series forecasting problems, rather than having this context prespecified and fixed [7], [8] The most objective of this consider to analyze middle factors, assess and anticipate the timeseries of normal temperature from 2008 until 2020 by utilizing the Long Short-Term Memory (LSTM) Demonstrate. The models were tried utilizing climate time arrangement information at each station which collected and compares the Root Cruel Square Mistake (RMSE) come about from the LTSM show.…”

Section: Introductionmentioning

confidence: 99%

Average Temperature Forecasting Based on Deli Serdang Station Using Long Short-Term Memory Model

Junaedi

Paramita

Sinaga

et al. 2021

JoCPES

View full text Add to dashboard Cite

An understanding of designs and gage of typical temperature joined of parameter climate and climate data for better water resource organization and orchestrating amid a bowl is uncommonly imperative. Examine climate designs utilizing ordinary and neighborhood every year typical temperatures, compare and make discernments. amid this consider, we'll analyze adjacent and conventional typical temperature data in 96031 Station backed recognition station input. the preeminent objective of this considers to appear the execution of the conventional temperature in an exceedingly single station and to predict the ordinary temperature data utilizing the Long memory Illustrate approach. bolstered the comes about of standard informatics of exploring temperature with adjacent temperature relationship, we got the appear of preparing bend, remaining plot, and thus the diffuse plot is showed up utilizing these codes. the decent execution of 96031 Station had a Mean Squared Error esteem of 0.01 and R squared esteem 0.98, concerning zero will speak to superior quality of the indicator.

show abstract

“…However, both the architectures have a similar root mean squared error (RMSE) indicating a scope of improvement in both the architectures Hu et al [39] presented a DL based LSTM approach for simulating rainfall-runoff process which outperforms the ANN models. In order to forecast the visibility at airports, Salman et al [40] presented a hybrid approach consisting of DL based LSTM and ARIMA model using three atmospheric variables viz. temperature, dew point and humidity.…”

Section: For Instances Gujarat and Maharashtra In 2005 Ladakhmentioning

confidence: 99%

Hybrid Deep Learning Approach for Multi-Step-Ahead Daily Rainfall Prediction Using GCM Simulations

Khan

Maity

2020

IEEE Access

View full text Add to dashboard Cite

Deep Learning (DL) is an effective technique for dealing with complex systems. This study proposes a hybrid DL approach, a combination of one-dimensional Convolutional Neural Network (Conv1D) and Multi-Layer Perceptron (MLP) (hereinafter referred to as hybrid Conv1D-MLP model), for multi-stepahead (1-day to 5-day in advance) daily rainfall prediction. Nine meteorological variables, closely associated with daily rainfall variation, are used as inputs to the hybrid model. The causal variables are obtained from a General Circulation Model (GCM). In general, simulation of meteorological variables from GCM is much better than rainfall estimate and observed records of meteorological variables is sparsely available, if not completely unavailable at many locations. Thus, proposed scheme helps to establish the effectiveness of the DL approach in augmenting the quality of rainfall prediction, exploiting the potential of GCM in simulating meteorological variables. The developed hybrid model is applied to twelve different locations in different climatic regimes in terms of daily precipitation characteristics. The proposed hybrid approach is compared with a DL approach namely, Multi-Layered Perceptron (deep MLP) and another machine learning approach namely, Support Vector Regression (SVR). It is also found that the performance of the model gradually decreases as the prediction lead time (in days) increases. Overall, this study establishes the fact that the hybrid Conv1D-MLP model is more effective in capturing the complex relationship between the causal variables and daily variation of rainfall. The benefit is due to the unification of potentials of individual approaches for extracting the hidden features of hydrometeorological association. INDEX TERMS Deep learning (DL), multi-layer perceptron (MLP), hybrid DL models (Conv1D-MLP), multi-step-ahead rainfall prediction, one-dimensional convolutional neural network (Conv1D), support vector regression (SVR).

show abstract

Weather Forecasting Using Merged Long Short-Term Memory Model (LSTM) and Autoregressive Integrated Moving Average (ARIMA) Model

Cited by 37 publications

References 8 publications

Application of a hybrid ARIMA-LSTM model based on the SPEI for drought forecasting

Application of a hybrid ARIMA-LSTM model based on the SPEI for drought forecasting

Average Temperature Forecasting Based on Deli Serdang Station Using Long Short-Term Memory Model

Hybrid Deep Learning Approach for Multi-Step-Ahead Daily Rainfall Prediction Using GCM Simulations

Contact Info

Product

Resources

About