The MULTIVARIATE TIME-SERIES DATA PREPARATION FOR WATER QUALITY FORECASTING IN LIVING LAB SMART AQUACULTURE SYSTEM
Abstract
Water quality monitoring in biofloc aquaculture systems is still commonly performed through manual observation, which limits the ability to detect short-term fluctuations in environmental parameters. Although Internet of Things (IoT)-based sensors enable continuous data acquisition, raw sensor datasets often contain missing values, noise, and inconsistent temporal structures that reduce their suitability for time-series forecasting applications. This study proposes a structured preprocessing pipeline for multivariate water quality sensor data consisting of temperature, pH, and total dissolved solids (TDS) to improve dataset readiness for predictive modeling. The preprocessing stages include data filtering, interpolation, outlier detection, normalization using Min–Max scaling, and sliding window transformation to construct supervised multi-step forecasting sequences with a 144-timestep input–output horizon. Experimental validation using a Long Short-Term Memory (LSTM) model demonstrates that the transformed dataset supports stable forecasting performance across multiple parameters. The proposed preprocessing framework contributes to improving the reliability of IoT-based aquaculture monitoring systems and supports the development of intelligent early warning mechanisms for water quality management.
Downloads
References
B. G. K. Yudistira, C. Hapsari, G. D. W. Adnyana, W. Nath, and I. P. R. M. Putra, “Smart Fisheries: Real-Time Water Quality Management and Automated Feeding System Design for Tilapia Farming using ESP32 Micro Controller,” Journal of Artificial Intelligence and Software Engineering (J-AISE), vol. 5, no. 2, pp. 827–832, Jun. 2025, doi: 10.30811/jaise.v5i2.7288.
Gandh, R. D., Harigovindan, V. P., Rasheed Abdul Haq, K. P., & Bhide, A. (2024). Attention-driven LSTM and GRU deep learning techniques for precise water quality prediction in smart aquaculture. Aquaculture International. https://doi.org/10.1007/s10499-024-01574-5
P. Liu, J. Wang, A. K. Sangaiah, Y. Xie, and X. Yin, “Analysis and Prediction of Water Quality Using LSTM Deep Neural Networks in IoT Environment,” Sustainability, vol. 11, no. 2058, pp. 2–14, Apr. 2019, doi: 10.3390/su11072058.
S. C. M. Sundararajan et al., “IoT-based prediction model for aquaponic fish pond water quality using multiscale feature fusion with convolutional autoencoder and GRU networks,” Sci. Rep., vol. 15, no. 1, Dec. 2025, doi: 10.1038/s41598-024-84943-7.
R. Krishnamurthi, A. Kumar, D. Gopinathan, A. Nayyar, and B. Qureshi, “An overview of iot sensor data processing, fusion, and analysis techniques,” Nov. 01, 2020, MDPI AG. doi: 10.3390/s20216076.
K. P. Rasheed Abdul Haq and V. P. Harigovindan, “Water Quality Prediction for Smart Aquaculture Using Hybrid Deep Learning Models,” IEEE Access, vol. 10, pp. 60078–60098, 2022, doi: 10.1109/ACCESS.2022.3180482.
R. Baena-Navarro, Y. Carriazo-Regino, F. Torres-Hoyos, and J. Pinedo-López, “Intelligent Prediction and Continuous Monitoring of Water Quality in Aquaculture: Integration of Machine Learning and Internet of Things for Sustainable Management,” Water (Switzerland), vol. 17, no. 82, pp. 1–25, Jan. 2025, doi: 10.3390/w17010082.
E. Antony, N. S. Sreekanth, R. K. Sunil Kumar, and T. Nishanth, “Data preprocessing techniques for handling time series data for environmental science studies,” International Journal of Engineering Trends and Technology, vol. 69, no. 5, pp. 196–207, May 2021, doi: 10.14445/22315381/IJETT-V69I5P227.
E. Eze, S. Kirby, J. Attridge, and T. Ajmal, “Aquaculture 4.0: Hybrid Neural Network Multivariate Water Quality Parameters Forecasting Model,” Res. Sq., pp. 1–21, Mar. 2023, doi: 10.21203/rs.3.rs-2711537/v1.
P. Martín-Calzada, P. Martín Sánchez, F. J. Rodríguez-Sánchez, C. Santos-Pérez, and J. Ballesteros, “Optimized Sensor Data Preprocessing Using Parameter-Transfer Learning for Wind Turbine Power Curve Modeling,” Sensors, vol. 25, no. 17, Sep. 2025, doi: 10.3390/s25175329.
J. Wang, W. Jiang, Z. Li, and Y. Lu, “A new multi-scale sliding window lstm framework (Mssw-lstm): A case study for gnss time-series prediction,” Remote Sens. (Basel)., vol. 13, no. 16, Aug. 2021, doi: 10.3390/rs13163328.
N. Okafor and D. Delaney, “MISSING DATA IMPUTATION ON IOT SENSOR NETWORKS: IMPLICATIONS FOR ON-SITE SENSOR CALIBRATION,” Jan. 26, 2021, techrxiv by IEEE. doi: 10.36227/techrxiv.13633529.v1.
R. Chandra, S. Goyal, and R. Gupta, “Evaluation of Deep Learning Models for Multi-Step Ahead Time Series Prediction,” IEEE Access, vol. 9, pp. 83105–83123, 2021, doi: 10.1109/ACCESS.2021.3085085.
K. W. Palihakkara and M. N. Jayakody, “A Comparative Study on Deep Learning Models for Time-Series Forecasting,” Jan. 26, 2026, preprints.org. doi: 10.20944/preprints202601.1962.v1.
X. Ji et al., “Long-term multivariate water quality forecasting for sustainable aquaculture management,” Water Res. X, vol. 29, Dec. 2025, doi: 10.1016/j.wroa.2025.100402.
Z. Li, S. Qi, Y. Li, and Z. Xu, “Revisiting Long-term Time Series Forecasting: An Investigation on Linear Mapping,” ArXiv, May 2023, [Online]. Available: http://arxiv.org/abs/2305.10721
J. Brownlee Disclaimer, Deep Learning for Time Series Forecasting Predict the Future with MLPs, CNNs and LSTMs in Python. Machine Learning Mastery, 2018.
A. Jaffar, N. M. Thamrin, M. S. A. M. Ali, M. F. Misnan, and A. I. M. Yassin, “WATER QUALITY PREDICTION USING LSTM-RNN: A REVIEW,” J. Sustain. Sci. Manag., vol. 17, no. 7, pp. 204–225, Jul. 2022, doi: 10.46754/jssm.2022.07.015.
I. P. R. Mahaputra, B. G. K. Yudistira, J. Shiddiq, G. D. W. Adnyana, and P. Z. E. S. Nugraha, “Short-term time-series forecasting of hydroponics water quality parameters using XGBoost based IoT sensor data,” JITET (Jurnal Informatika dan Teknik Elektro Terapan), vol. 14, no. 2, 2026, doi: 10.23960/jitet.v14i2.9458.
Y. Yang, P. Zhang, and Y. Wang, “An attention-based parallel model with sliding window decomposition algorithm for water quality prediction,” Journal of Water Process Engineering, vol. 78, p. 108751, 2025, doi: 10.1016/j.jwpe.2025.108751.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
