Predicting Discharges in Sewer Pipes Using an Integrated Long Short-Term Memory and Entropy A-TOPSIS Modeling Framework

Abstract

Predicting discharges in sewage systems play an essential role in reducing sewer overflows and impacts on the environment and public health. Choosing a suitable model to predict discharges in these systems is essential to realizing these aforementioned goals. Long Short-Term Memory (LSTM) has been proposed as a robust technique for predicting discharges in wastewater networks. This study explored the potential application of an LSTM model to predict discharges using 3-month data set in a sewer network in Ålesund city, Norway. Different sequence-to-sequence LSTMs were investigated using various input and output datasets. The impact of data aggregation (10-min and 30-min intervals) was examined and compared to original sensor data (5-min intervals) to evaluate the performance of the LSTM model. The results show that 50-neuron LSTM architecture performed better (MAPE = 0.09, RMSE = 0.0008, R2 = 0.8) in predicting discharges for the study area. The study indicates that using the same sequence length for the prior and the forecast can improve the effectiveness of the LSTM model. Based on the results, using a 10-min aggregated discharge dataset reduces energy consumption, transmission bandwidth, and storage capacity. Additionally, it improves prediction performance compared to an original 5-min interval data in Ålesund city.

Early Flood Monitoring and Forecasting System Using a Hybrid Machine Learning-Based Approach

“…First introduced by Hochreiter and Schmidhuber [73], LSTM stands as an advanced variation of the RNN architecture, because it possesses a remarkable capability to capture both long-term and short-term dependencies, with its memory cell playing a crucial role in storing and retaining cell states [74]. LSTM networks employ gates to enhance their performance [75]. These gates, including input, output, and forget gates, play a crucial role in remembering and learning from past information, thereby facilitating accurate time series prediction of sequential data [72,76].…”

“…In this paper LSTM models were implemented on the Keras library with TensorFlow backend [85]. Keras is a Python interface developed by Google that offers an open-source software for artificial neural networks and deep learning [75,86]. Additionally, the implementation utilized the Pandas, NumPy, Scikit Learn, and Matplotlib libraries to address specific data-driven modeling requirements.…”

Optimal Control Strategy of a Sewer Network

“…The control method gives good results even in rainfall conditions. In [16], neural networks were used to predict the discharges in the sewer systems. Different neural networks were investigated using different sets of input and output data, resulting in the neural network with 50 neurons giving the best performance.…”

A Novel Calculation Method of Hydrodynamic Pressure Based on Polyhedron SBFEM and Its Application in Nonlinear Cross-Scale CFRD-Reservoir Systems

“…Considering the important influence of hydrodynamic pressure on dam response, it is an urgent problem, which until now remains to be solved, for a cross-scale dynamic dam analysis system based on a polyhedron SBFEM. In the analysis of the interaction between fluid and structure [34][35][36][37], the calculation method of hydrodynamic pressure on dams has always been one of the hot research topics. At present, much research on the numerical analysis of the dam-reservoir dynamic interaction under earthquake conditions has emerged.…”

“…However, many nodal DOFs need to be introduced, especially for large scale 3D models of actual projects, which can dramatically increase the calculation amount when simulating dynamic coupling of dam-reservoir systems. Lin et al [50] realized an efficient SBFEM-based solution of hydrodynamic pressure in a 3D reservoir by only discretizing the two-dimensional (2D) interfaces between the reservoir water and the dam's upstream face, thus saving many DOFs, improving In the analysis of the interaction between fluid and structure [34][35][36][37], the calculation method of hydrodynamic pressure on dams has always been one of the hot research topics. At present, much research on the numerical analysis of the dam-reservoir dynamic interaction under earthquake conditions has emerged.…”