Enhanced performance of biochar-biosensor applied to heavy metals detection in constructed wetlands and biological mechanisms

Journal of Cleaner Production, Journal Year: 2023, Volume and Issue: 436, P. 140339 - 140339

Published: Dec. 24, 2023

Language: Английский

---

Causality-Driven Feature Selection for Calibrating Low-Cost Airborne Particulate Sensors Using Machine Learning

Sensors, Journal Year: 2024, Volume and Issue: 24(22), P. 7304 - 7304

Published: Nov. 15, 2024

With escalating global environmental challenges and worsening air quality, there is an urgent need for enhanced monitoring capabilities. Low-cost sensor networks are emerging as a vital solution, enabling widespread affordable deployment at fine spatial resolutions. In this context, machine learning the calibration of low-cost sensors particularly valuable. However, traditional models often lack interpretability generalizability when applied to complex, dynamic data. To address this, we propose causal feature selection approach based on convergent cross mapping within pipeline build more robustly calibrated networks. This in optical particle counter OPC-N3, effectively reproducing measurements PM1 PM2.5 recorded by research-grade spectrometers. We evaluated predictive performance these causally optimized models, observing improvements both while reducing number input features, thus adhering Occam's razor principle. For model, proposed reduced mean squared error test set 43.2% compared model with all SHAP value-based only achieved reduction 29.6%. Similarly, led 33.2% error, outperforming 30.2% selection. By integrating advanced techniques, advances urban quality monitoring, fostering deeper scientific understanding microenvironments. Beyond current cases, method holds potential broader applications other applications, contributing development interpretable robust models.

Language: Английский

---

Polygon Simplification for the Efficient Approximate Analytics of Georeferenced Big Data

Sensors, Journal Year: 2023, Volume and Issue: 23(19), P. 8178 - 8178

Published: Sept. 29, 2023

The unprecedented availability of sensor networks and GPS-enabled devices has caused the accumulation voluminous georeferenced data streams. These streams offer an opportunity to derive valuable insights facilitate decision making for urban planning. However, processing managing such is challenging, given size multidimensionality these data. Therefore, there a growing interest in spatial approximate query depending on stratified-like sampling methods. solutions, as number strata increases, response time grows, thus counteracting benefits sampling. In this paper, we originally show design realization novel online geospatial solution called GeoRAP. GeoRAP employs front-stage filter based Ramer-Douglas-Peucker line simplification algorithm reduce study area coverage; thereafter, it method that minimizes strata, increasing throughput minimizing time, while keeping accuracy loss check. Our applicable various batch workloads, including complex geo-statistics, aggregation queries, generation region-based aggregate geo-maps choropleth maps heatmaps. We have extensively tested performance our prototyped with real-world big data, paper shows can outperform state-of-the-art baselines by order magnitude terms statistically obtaining results good accuracy.

Language: Английский

---

Smart calibration and monitoring: leveraging artificial intelligence to improve MEMS-based inertial sensor calibration

Complex & Intelligent Systems, Journal Year: 2024, Volume and Issue: 10(6), P. 7451 - 7474

Published: July 18, 2024

Abstract Micro-electro-mechanical systems (MEMS)-based sensors endure complex production processes that inherently include high variance. To meet rigorous client demands (such as sensitivity, offset noise, robustness against vibration, etc.). products must go through comprehensive calibration and testing procedures. All undergo a standardized sequential process with predetermined number of steps, even though some may reach the correct value sooner. Moreover, traditional method faces challenges due to specific operating conditions resulting from manufacturing discrepancies. This not only extends duration but also introduces rigidity inefficiency. tackle issue variances elongated time enhance efficiency, we provide novel quasi-parallelized framework aided by an artificial intelligence (AI) based solution. Our suggested utilizes supervised tree-based regression technique statistical measures dynamically identify optimize appropriate working point for each sensor. The objective is decrease total while ensuring accuracy. findings our investigation show reduction 23.8% calibration, leading substantial cost savings in process. In addition, propose end-to-end monitoring system accelerate incorporation into production. guarantees prompt execution solution enables identification modifications or data irregularities, promoting more agile adaptable

Language: Английский

---

Detecting and quantifying PM2.5 and NO2 contributions from train and road traffic in the vicinity of a major railway terminal in Dublin, Ireland

Environmental Pollution, Journal Year: 2024, Volume and Issue: 361, P. 124903 - 124903

Published: Sept. 6, 2024

Language: Английский

---