Pattern Extraction of Water Quality Prediction Using Machine Learning Algorithms of Water Reservoir

Lerios, Jefferson L.; Programs, Manila Graduate; Villarica, Mia V.

doi:10.18178/ijmerr.8.6.992-997

Cited by 23 publications

(8 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Holzkämper et al (2012) used a Bayesian network approach to produce an integrated 'meta-model' from loose numerical and knowledge based sub-models that each specialised in a different aspect of catchment management, such as flood risk, land and sewage management. Lerios & Villarica (2019) used multiple machine learning algorithms to predict a state-based index of water quality and to choose an overall 'best' model rather than constructing a meta-model using all algorithms. Similarly, Lu & Ma (2020) used multiple machine learning models for shortterm water-quality prediction but did not use a meta-model to combine the strengths of each individual model.…”

Section: Discussionmentioning

confidence: 99%

Short-term prediction of stream turbidity using surrogate data and a meta-model approach

Bhargav¹,

Caleb²,

Kandanaarachchi³

et al. 2022

Preprint

View full text Add to dashboard Cite

Many water-quality monitoring programs aim to measure turbidity to help guide effective management of waterways and catchments, yet distributing turbidity sensors throughout networks is typically cost prohibitive. To this end, we built and compared the ability of dynamic regression (ARIMA), long short-term memory neural nets (LSTM), and generalized additive models (GAM) to forecast stream turbidity one step ahead, using surrogate data from relatively low-cost in-situ sensors and publicly available databases. We iteratively trialled combinations of four surrogate covariates (rainfall, water level, air temperature and total global solar exposure) selecting a final model for each type that minimised the corrected Akaike Information Criterion. Cross-validation using a rolling time-window indicated that ARIMA, which included the rainfall and water-level covariates only, produced the most accurate predictions, followed closely by GAM, which included all four covariates. However, according to the no-free-lunch theorems in machine learning, no single model has an advantage over all other models for all instances. Therefore, we constructed a meta-model, trained on time-series features of turbidity, to take advantage of the strengths of each model over different time points and predict the best model (that with the lowest forecast error one-step prior) for each time step. The meta-model outperformed all other models, indicating that this methodology can yield high accuracy and may be a viable alternative to using measurements sourced directly from turbidity-sensors where costs prohibit their deployment and maintenance, 1

show abstract

Section: Discussionmentioning

confidence: 99%

Short-term prediction of stream turbidity using surrogate data and a meta-model approach

Bhargav¹,

Caleb²,

Kandanaarachchi³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Since the recurrent neural network was already proven effective in time series analysis [14], [18], it was chosen in this project as the deep learning algorithm to use in knowledge development. RNN is thought for managing sequential statistics.…”

Section: Knowledge Developmentmentioning

confidence: 99%

“…Machine learning was used in different discipline such as fire incidents, health, education and even in environmental modelling. With this technology, it is possible to provide decision making through pattern recognition and time series analysis [18]- [20]. Recurrent neural networks (RNNs) are designed to operate upon sequences of data.…”

Section: Introductionmentioning

confidence: 99%

Intelligent aquaculture system for pisciculture simulation using deep learning algorithm

Sapin

Alibudbud

Molleno

et al. 2022

IJEECS

View full text Add to dashboard Cite

The project aims to develop an intelligent system for simulating pisciculture in Taal Lake in the Philippines through geographical information system and deep learning algorithm. Records of 2018-2020 from the database of Bureau of fisheries and aquatic resources IV-A-protected area management board (BFAR IVA-PAMB) was collected for model development. Deep learning algorithm model was developed and integrated to the system for time series analysis and simulation. Different technologies including tensorflow.js were used to successfully developed the intelligent system. It is found on this paper that recurrent neural network (RNN) is a good deep learning algorithm for predicting pisciculture in Taal lake. Further, it is also shown in the initial visualization of the system that barangay Sampaloc in Taal has highest rate of fish production in Taal while Tilapia nilotica sp. is the major product of the latter.

show abstract

“…Lerios and Villarica (2019) used multiple machine learning algorithms to predict a state-based index of water quality and to choose an overall 'best' model rather than constructing a meta-model using all algorithms. Similarly,Lu and Ma (2020) used multiple machine learning models for short-term water-quality prediction but did not use a meta-model to combine the strengths of each individual model.…”

mentioning

confidence: 99%

Short‐term prediction of stream turbidity using surrogate data and a meta‐model approach: A case study

Bhargav

Caleb

Kandanaarachchi

et al. 2023

Hydrological Processes

View full text Add to dashboard Cite

Many water-quality monitoring programs aim to measure turbidity to help guide effective management of waterways and catchments, yet distributing turbidity sensors throughout networks is typically cost prohibitive. To this end, we built and compared the ability of dynamic regression (auto-regressive integrated moving average [ARIMA]), long short-term memory neural nets (LSTM), and generalized additive models (GAM) to forecast stream turbidity one step ahead, using surrogate data from relatively low-cost in-situ sensors and publicly available databases. We iteratively trialled combinations of four surrogate covariates (rainfall, water level, air temperature and total global solar exposure) selecting a final model for each type that minimized the corrected Akaike information criterion. Cross-validation using a rolling time-window indicated that ARIMA, which included the rainfall and water-level covariates only, produced the most accurate predictions, followed closely by GAM, which included all four covariates. However, according to the nofree-lunch theorems in machine learning, no single model has an advantage over all other models for all instances. Therefore, we constructed a meta-model, trained on time-series features of turbidity, to take advantage of the strengths of each model over different time points and predict the best model (that with the lowest forecast error one-step prior) for each time step. The meta-model outperformed all other models, indicating that this methodology can yield high accuracy and may be a viable alternative to using measurements sourced directly from turbidity-sensors where costs prohibit their deployment and maintenance, and when predicting turbidity across the short term. Our findings also indicated that temperature and lightassociated variables, for example underwater illuminance, may hold promise as cost-effective, high-frequency surrogates of turbidity, especially when combined with other covariates, like rainfall, that are typically measured at coarse levels of spatial resolution.

show abstract

Pattern Extraction of Water Quality Prediction Using Machine Learning Algorithms of Water Reservoir

Cited by 23 publications

References 2 publications

Short-term prediction of stream turbidity using surrogate data and a meta-model approach

Short-term prediction of stream turbidity using surrogate data and a meta-model approach

Intelligent aquaculture system for pisciculture simulation using deep learning algorithm

Short‐term prediction of stream turbidity using surrogate data and a meta‐model approach: A case study

Contact Info

Product

Resources

About