A data filling methodology for time series based on CNN and (Bi)LSTM neural networks

Tzoumpas, Kostas; Estrada, Aaron R.; Miraglio, Pietro; Zambelli, Pietro

doi:10.48550/arxiv.2204.09994

Cited by 3 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the latest algorithmic frameworks, based on deep-learning architectures, can be deployed on low-cost commercial hardware architectures (even those not integrating GPUs), in our proposed approach three DL architectures were considered for performance comparison on the forecasting model: (1) a one-dimensional Convolutional Neural Network (1D-CNN) [21], (2) Long Short-Term Memory (LSTM) [22] and (3) a Recurrent Neural Network (RNN) [23]. Each of the 1D-CNN, LSTM and RNN models often show good performance on multivariate time-series data [24][25][26][27].…”

Section: Methodsmentioning

confidence: 99%

Multi-Sensor Platform for Predictive Air Quality Monitoring

Rescio,

Manni,

Caroppo

et al. 2023

Sensors

View full text Add to dashboard Cite

Air quality monitoring is a very important aspect of providing safe indoor conditions, and carbon dioxide (CO2) is one of the pollutants that most affects people’s health. An automatic system able to accurately forecast CO2 concentration can prevent a sudden rise in CO2 levels through appropriate control of heating, ventilation and air-conditioning (HVAC) systems, avoiding energy waste and ensuring people’s comfort. There are several works in the literature dedicated to air quality assessment and control of HVAC systems; the performance maximisation of such systems is typically achieved using a significant amount of data collected over a long period of time (even months) to train the algorithm. This can be costly and may not respond to a real scenario where the habits of the house occupants or the environment conditions may change over time. To address this problem, an adaptive hardware–software platform was developed, following the IoT paradigm, with a high level of accuracy in forecasting CO2 trends by analysing only a limited window of recent data. The system was tested considering a real case study in a residential room used for smart working and physical exercise; the parameters analysed were the occupants’ physical activity, temperature, humidity and CO2 in the room. Three deep-learning algorithms were evaluated, and the best result was obtained with the Long Short-Term Memory network, which features a Root Mean Square Error of about 10 ppm with a training period of 10 days.

show abstract

Section: Methodsmentioning

confidence: 99%

Multi-Sensor Platform for Predictive Air Quality Monitoring

Rescio,

Manni,

Caroppo

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…They reported that the CNN model showed better performance than traditional statistical models. Tzoumpas et al [ 18 ] proposed a data-filling methodology and used CNN and LSTM models to predict indoor temperature. They reported improved accuracy compared to traditional statistical models.…”

Section: Related Workmentioning

confidence: 99%

Performance Analysis of Time Series Deep Learning Models for Climate Prediction in Indoor Hydroponic Greenhouses at Different Time Intervals

Eraliev

Lee

2023

Plants

View full text Add to dashboard Cite

Indoor hydroponic greenhouses are becoming increasingly popular for sustainable food production. On the other hand, precise control of the climate conditions inside these greenhouses is crucial for the success of the crops. Time series deep learning models are adequate for climate predictions in indoor hydroponic greenhouses, but a comparative analysis of these models at different time intervals is needed. This study evaluated the performance of three commonly used deep learning models for climate prediction in an indoor hydroponic greenhouse: Deep Neural Network, Long–Short Term Memory (LSTM), and 1D Convolutional Neural Network. The performance of these models was compared at four time intervals (1, 5, 10, and 15 min) using a dataset collected over a week at one-minute intervals. The experimental results showed that all three models perform well in predicting the temperature, humidity, and CO2 concentration in a greenhouse. The performance of the models varied at different time intervals, with the LSTM model outperforming the other models at shorter time intervals. Increasing the time interval from 1 to 15 min adversely affected the performance of the models. This study provides insights into the effectiveness of time series deep learning models for climate predictions in indoor hydroponic greenhouses. The results highlight the importance of choosing the appropriate time interval for accurate predictions. These findings can guide the design of intelligent control systems for indoor hydroponic greenhouses and contribute to the advancement of sustainable food production.

show abstract

“…In [11,12], the authors used a combination of LSTM and transfer learning to perform data imputation in the fields of building energy and water quality. In [13], a model combining CNN and Bi-LSTM was successfully used to effectively interpolate temperature data inside an apartment. And in [14], an improved Bi-LSTM algorithm was used to improve the filling problem of pregnancy examination data for predicting Hypertensive Disorders of Pregnancy (HDP).…”

Section: Introductionmentioning

confidence: 99%

Enhancing Predictive Maintenance in Airborne Weapon Systems through Advanced Frequency Augmentation Techniques

Kang,

Jeong,

Lee

et al. 2024

Preprint

View full text Add to dashboard Cite

This study investigates the application of frequency augmentation techniques to improve condition-based maintenance (CBM+) systems for airborne weapon systems, focusing on the predictive accuracy of velocity and acceleration parameters. Utilizing MIL-STD-1553B MUX data from 138 sorties of fixed-wing aircraft, we explored six augmentation methods, including traditional interpolations (linear, quadratic, cubic spline) and advanced machine learning models (K-Nearest Neighbor, LSTM, and Bi-LSTM). Our findings indicate that while traditional methods like linear interpolation are more effective for velocity parameters, advanced ML techniques, particularly Bi-LSTM, provide better results for acceleration parameters, which exhibit more complex and rapid variability. The study underscores the necessity of tailoring augmentation approaches to specific parameter characteristics and highlights the potential of ML models in capturing intricate patterns within time-series data. These insights are critical for advancing CBM+ systems in military avionics, enhancing reliability and operational efficiency through improved data processing strategies. Future work should explore more performant augmentation techniques, such as by integrating multi-dimensional datasets, and explore ways to develop automated tools for a more powerful predictive maintenance framework.

show abstract

A data filling methodology for time series based on CNN and (Bi)LSTM neural networks

Cited by 3 publications

References 25 publications

Multi-Sensor Platform for Predictive Air Quality Monitoring

Multi-Sensor Platform for Predictive Air Quality Monitoring

Performance Analysis of Time Series Deep Learning Models for Climate Prediction in Indoor Hydroponic Greenhouses at Different Time Intervals

Enhancing Predictive Maintenance in Airborne Weapon Systems through Advanced Frequency Augmentation Techniques

Contact Info

Product

Resources

About