Automatic mass spectra recognition for Ultra High Vacuum systems using multilabel classification

Mateo, Fernando; Garcés-Iniesta, Juan José; Jenninger, B.; Gómez-Sanchís, Juan; Soria-Olivas, Emilio; Chiggiato, Paolo

doi:10.1016/j.eswa.2021.114959

Cited by 3 publications

(13 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The samples used for the comparative metrics have been obtained by synthetic generation, as described in a previous work [2]. Ideally, one should be able to use true residual gas samples obtained from RGA in UHV systems, but the availability of these is very limited, so this is an effective and reliable method for the controlled generation of a large number of samples.…”

Section: Synthetic Data Generation and Normalizationmentioning

confidence: 99%

“…Such contaminants are deposited (mainly during the manufacturing process) on the inner surface of the vacuum chambers and hinder the pump-down process and, consequently, the generation of the required pressure. In particular, the presence of some contaminants may have effects on many other surface properties, such as the surface energy and secondary electron yield [1,2].…”

Section: Introductionmentioning

confidence: 99%

“…However, reconstruction is a more complicated task than recognition. Recent developments in automatic MS reconstruction methods are able to either provide a probability score based on machine learning (ML) algorithms that indicates a degree of presence in the gas sample [2] or provide a set of candidate gases based on some criteria and then produce a reconstructed MS with estimated partial pressures for each one of the candidate species [8]. Each one of these methods has their advantages and drawbacks.…”

Section: Introductionmentioning

confidence: 99%

“…Each one of these methods has their advantages and drawbacks. The proposal in [2] explores a multilabel classification technique based on XGBoost trees that provides high classification accuracy (>89%) even for complex species. However, this technique does not provide an estimation of the relative contribution of each species, and hence, does not provide a way to reconstruct the original MS. On the other hand, the technique proposed in [8] provides an estimated reconstruction based on an iterative algorithm that sequentially adds gases at their optimal pressure to the sample.…”

Section: Introductionmentioning

confidence: 99%

“…The goal of this work is to propose a novel way to add the MS reconstruction stage to the pre-selection of gases obtained by the method proposed in [2] using genetic algorithms (GAs), and thus, enable not only the accurate identification of contaminants, but also the quantification of their relative contribution. We compare the reconstruction error obtained by our method to the one obtained by the iterative reconstruction method proposed by [8] and assess the computational times required by both methods.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Reconstruction of Ultra-High Vacuum Mass Spectra Using Genetic Algorithms

et al. 2021

Self Cite

View full text Add to dashboard Cite

In ultra-high vacuum systems, obtaining the composition of a mass spectrum is often a challenging task due to the highly overlapping nature of the individual profiles of the gas species that contribute to that spectrum, as well as the high differences in terms of degree of contribution (several orders of magnitude). This problem is even more complex when not only the presence but also a quantitative estimation of the contribution (partial pressure) of each species is required. This paper aims at estimating the relative contribution of each species in a target mass spectrum by combining a state-of-the-art machine learning method (multilabel classifier) to obtain a pool of candidate species based on a threshold applied to the probability scores given by the classifier with a genetic algorithm that aims at finding the partial pressure at which each one of the species contributes to the target mass spectrum. For this purpose, we use a dataset of synthetically generated samples. We explore different acceptance thresholds for the generation of initial populations, and we establish comparative metrics against the most novel method to date for automatically obtaining partial pressure contributions. Our results show a clear advantage in terms of the integral error metric (up to 112 times lower for simpler spectra) and computational times (up to 4 times lower for complex spectra) in favor of the proposed method, which is considered a substantial improvement for this task.

show abstract

Section: Synthetic Data Generation and Normalizationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Reconstruction of Ultra-High Vacuum Mass Spectra Using Genetic Algorithms

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

Enhancement of DNN-based multilabel classification by grouping labels based on data imbalance and label correlation

Chen

Wang

2022

Pattern Recognition

View full text Add to dashboard Cite

Analysis for Time-Dependent Behavior of Soft Rock Through a Reinforced Learning Fusion Constitutive Model

Tan,

Chen,

Tian

et al. 2024

Int. J. Appl. Mechanics

View full text Add to dashboard Cite

The long-term stability of tunnel structures is significantly influenced by the time-dependent behavior of the surrounding rock. Existing constitutive models often deviated from surrounding deformation due to the anisotropic nature of rock mass. In response, this study introduces a novel reinforced learning fusion constitutive model to accurately capture the time-dependent behaviors of soft rock. The framework and methodologies are first outlined, followed by the instantiation of the constitutive model of Burgers and creep parameters using laboratory testing data. To enhance accuracy, an XGBoost model is trained to reinforce the results of the constitutive model. The reliability of the proposed model is then validated against the original constitutive model and other representative machine learning models. Experimental findings demonstrate the superior characterization ability and stability of the presented reinforced model, where the calculation error reduces by 7.2E−06 at least, and [Formula: see text] score is improved by at least 1% to others. Consequently, the proposed model is reliable, offering a promising approach to capturing the actual time-dependent behaviors of tunnel surroundings in practical field applications.

show abstract

Automatic mass spectra recognition for Ultra High Vacuum systems using multilabel classification

Cited by 3 publications

References 28 publications

Reconstruction of Ultra-High Vacuum Mass Spectra Using Genetic Algorithms

Reconstruction of Ultra-High Vacuum Mass Spectra Using Genetic Algorithms

Enhancement of DNN-based multilabel classification by grouping labels based on data imbalance and label correlation

Analysis for Time-Dependent Behavior of Soft Rock Through a Reinforced Learning Fusion Constitutive Model

Contact Info

Product

Resources

About