Water Level Prediction Model Applying a Long Short-Term Memory (LSTM)–Gated Recurrent Unit (GRU) Method for Flood Prediction

Abstract

The damage caused by floods is increasing worldwide, and if floods can be predicted, the economic and human losses from floods can be reduced. A key parameter of flooding is water level data, and this paper proposes a water level prediction model using long short-term memory (LSTM) and a gated recurrent unit (GRU). As variables used as input data, meteorological data, including upstream and downstream water level, temperature, humidity, and precipitation, were used. The best results were obtained when the LSTM–GRU-based model and the Automated Synoptic Observing System (ASOS) meteorological data were included in the input data when experiments were performed with various model structures and different input data formats. As a result of the experiment, the mean squared error (MSE) value was 3.92, the Nash–Sutcliffe coefficient of efficiency (NSE) value was 0.942, and the mean absolute error (MAE) value was 2.22, the highest result in all cases. In addition, the test data included the historical maximum water level of 3552.38 cm in the study area, and the maximum water level error was also recorded as 55.49, the lowest result. Through this paper, it was possible to confirm the performance difference according to the composition of the input data and the time series prediction model. In a future study, we plan to implement a flood risk management system that can use the predicted water level to determine the risk of flooding, and evacuate in advance.

Quantitative Detection of Defects in Multi-Layer Lightweight Composite Structures Using THz-TDS Based on a U-Net-BiLSTM Network

“…The BiLSTM network structure is illustrated in Figure 5. In Figure 5, the LSTM layer incorporates three key gates: the forget, input, and output gates [26]. The forget gate is represented by 𝑓 ; the input gate is represented by 𝑖 ; and the output gate is represented by 𝑜 .…”

“…The input gate controls how much information about the current state is input into the memory cell, and the output gate regulates the extent of output based on the current cell state 𝑐 . The specific implementation formulas are presented as follows: In Figure 5, the LSTM layer incorporates three key gates: the forget, input, and output gates [26]. The forget gate is represented by f t ; the input gate is represented by i t ; and the output gate is represented by o t .…”

Physical Reservoir Computing Based on Nanoscale Materials and Devices

“…In RNN, a node contains only one neuron, while LSTMs and GRUs introduce an internal mechanism named “gate” that modulates the flow of previous information. [ 40,41 ] The fundamental concept of LSTM is the cell state and its gates. [ 42 ] The cell state plays the role of a pathway that transfers information along the sequence chain, as the memory in the network.…”

A Spatial-Reduction Attention-Based BiGRU Network for Water Level Prediction

“…The former assigns different weights to the hidden layer outputs at different times and then uses weighted summation to obtain an RNN context vector. The latter can be thought of as assigning different attention weights to various dimensions of the output vector [38,39]. Thus, this paper proposes the attention mechanism based on BiGRU to address the issues of reduced prediction accuracy due to the extended time span in the water-level-prediction task and insufficient spatial information utilization in the water-level-prediction task.…”

“…This experiment took the average of the water levels at six points surrounding the empty item, such as x 1 , x 2 , x 3 , x 4 , x 5 , and x 6 , where x 4 was the empty item, and the filling of x 4 was (x 1 + x 2 + x 3 + x 5 + x 6 + x 7 )/6. Finally, data outliers such as extremely large and small water-level-observation values that clearly deviate from the average level of the series were eliminated and filled in using the data missing processing method [39]. Furthermore, to hasten the convergence of the proposed model and improve its accuracy, maximum-minimum normalization was used so that all values are compressed within the interval [0, 1] [30].…”