Forecasting Energy Consumption of Wastewater Treatment Plants with a Transfer Learning Approach for Sustainable Cities

Abstract

A major challenge of today’s society is to make large urban centres more sustainable. Improving the energy efficiency of the various infrastructures that make up cities is one aspect being considered when improving their sustainability, with Wastewater Treatment Plants (WWTPs) being one of them. Consequently, this study aims to conceive, tune, and evaluate a set of candidate deep learning models with the goal being to forecast the energy consumption of a WWTP, following a recursive multi-step approach. Three distinct types of models were experimented, in particular, Long Short-Term Memory networks (LSTMs), Gated Recurrent Units (GRUs), and uni-dimensional Convolutional Neural Networks (CNNs). Uni- and multi-variate settings were evaluated, as well as different methods for handling outliers. Promising forecasting results were obtained by CNN-based models, being this difference statistically significant when compared to LSTMs and GRUs, with the best model presenting an approximate overall error of 630 kWh when on a multi-variate setting. Finally, to overcome the problem of data scarcity in WWTPs, transfer learning processes were implemented, with promising results being achieved when using a pre-trained uni-variate CNN model, with the overall error reducing to 325 kWh.

Prediction Model of Wastewater Pollutant Indicators Based on Combined Normalized Codec

“…This experiment compares the CNC model with other classical prediction models and verifies the superiority of the improvement proposed in this paper compared with other models in applying actual wastewater treatment effect prediction. The comparison models include: ANN [37], DNN [38], LSTM [26], GRU [39], Attention_LSTM [40], Atten-tion_GRU [41] and Codec model [42]. The experiment uses the pollutant concentration index of the water inlet from time t-30 to t to predict the pollutant concentration index of the water outlet at time t+1, and the data set is divided into 90% training set and 10% test set.…”

“…Based on the data of pollutant concentration indicators in the actual brewery wastewater treatment process, the prediction accuracy of the proposed model and seven classical prediction models, including ANN [37], deep neural network (DNN) [38], LSTM [26], gated recurrent unit (GRU) [39], Attention_LSTM [40], Attention_GRU [41] and Codec [42] are compared.…”

Sustainable indicators for designing the wastewater treatment plant

“…The energy used in the WWTP continuously increased as clean water demand increased and eventually affected the total cost operational expenditure (OPEX) [10]. Besides that, [11] stated that most WWTP found low energy efficiency performance levels due to high water demand and higher discharge parameters in the treated effluent. In addition, the energy used from fossil fuels contributes to environmental issues such as the greenhouse effect and acid rain.…”

Advances in Public Transport Platform for the Development of Sustainability Cities

“…Other topics included in this special issue are energy consumption forecasting in sustainable cities [10] as well as the analysis of energy trading and the development of a trust model [11]. Security is also an important issue within public transpormation, in reference [12] the secure management of railway transportation systems has been analyzed.…”