Research on time series characteristics of the gas drainage evaluation index based on lasso regression

The evaluation of the coal mine gas drainage effect is affected by many factors, such as flow rate, wind speed, drainage negative pressure, concentration, and temperature. This paper starts from actual coal mine production monitoring data and based on the lasso regression algorithm, features selection of multiple parameters of the preprocessed gas concentration time series to construct gas concentration feature selection based on the algorithm. The three-time smoothing index method is used to fill in the missing values. Aiming at the problem of different dimensions in the gas concentration time series, the MinMaxScaler method is used to normalize the data. The lasso regression algorithm is used to perform feature selection on the multivariable gas concentration time series, and the gas concentration time series selected by the lasso feature and the gas concentration time series without feature selection are input. The performance of the ANN algorithm for gas concentration prediction is compared and analyzed. The optimal α value and L1 norm are selected based on the grid search method to determine the strong explanatory gas concentration time series feature set of the working face, and an experimental comparison of the gas concentration prediction results before and after the lasso feature selection is performed. We verify the effectiveness of the algorithm.

Travel Time Modelling Using Time Series of Nitrate Concentration in Groundwater Improves Hydrological Process Understanding : A Case Study of a Small Agricultural Catchment in Brittany, France.

Rivers can act as mirrors to in-catchment processes, but integrated concentration-discharge dynamics might not be sufficient for constructing a well-posed solute travel time determination problem. One remedy is to look inside the catchment and see if the extra information provided by long-term time series of groundwater solutes constrains the problem or provides us with some additional insight on retrieving the processes which the stream is aggregating. To test this notion, we used data for Kerrien, a well-studied agriculture dominated small headwater catchment of the French Critical Zone Observatory in Brittany. It contains long-term nitrate concentration time-series from a network of piezometers as well as a stream outlet. In this study, a parsimonious, conceptual dual-permeability mixing model already developed for streams was adapted for piezometers along with detailed uncertainty and sensitivity analysis. We found out the nitrate flushing times of mid to upslope piezometers were consistently higher than the stream outlet. We further observed an asynchronicity in seasonal concentration-discharge dynamics between the piezometers and the stream. We hypothesize the reason behind this counterintuitive finding to be extensive riparian denitrification, vertical stratification of groundwater and disconnect between the stream and the deeper flowpaths that carry legacy contamination, evidenced by the non-closure of water budget at the stream outlet. As a consequence, we argue that in headwater catchments the stream signature might not fully reflect internal processes which can be revealed only by using piezometer data. This adapted conceptual framework could be of great interest for semi-arid catchments where groundwater monitoring could be used in combination or as an alternate to ephemeral streams in travel time determination.

Observations of atmospheric iodine-containing species during the MOSAiC expedition

<p>The Arctic is a unique region featuring many environmental variations from a season to another. For example, sea ice is highly dynamic, with varying thickness and homogeneity, ultimately leading to open sea with a boost of biological activity during the warmest month. This, in turn, affects the emissions of gas-phase chemicals, potentially impacting New Particle Formation (NPF) and subsequent aerosol growth.</p><p>Several chemical vapors such as sulfuric acid (SA) and methane sulfonic acid (MSA) are known to possibly contribute to NPF and/or particle growth. Additionally, halogenated compounds, such as iodic acid, have recently revealed to be important for the formation of aerosol particles, especially in coastal and Arctic sites.</p><p>Few studies exist regarding direct measurements of iodic acid in the high Arctic, and none of them report multi-seasonal continuous observations - especially during the polar-night when the extremely low temperatures and the absence of solar radiation would likely prohibit any synthesis of such chemical species.</p><p>Here, we present our observations of iodine-containing vapors, principally iodic acid, as the result of continuous on-line measurements with the Nitrate based Chemical Ionization Atmospheric Pressure interface Time Of Flight Mass Spectrometer (NO<sub>3</sub>-CI-APi-TOF-MS) during the whole Multidisciplinary Drifting Observatory of the Study of Arctic Climate (MOSAiC) expedition. In this study we combine and examine iodic acid multi-seasonal concentration time series in the central Arctic. In short, we aim at characterizing the observed iodic acid with the central Arctic environmental conditions (e.g., meteorological conditions, sea ice features and trace gases) and the linkage to NPF and particle growth.</p><p> </p>

A Simple Dendritic Neural Network Model-Based Approach for Daily PM2.5 Concentration Prediction

Air pollution in cities has a massive impact on human health, and an increase in fine particulate matter (PM2.5) concentrations is the main reason for air pollution. Due to the chaotic and intrinsic complexities of PM2.5 concentration time series, it is difficult to utilize traditional approaches to extract useful information from these data. Therefore, a neural model with a dendritic mechanism trained via the states of matter search algorithm (SDNN) is employed to conduct daily PM2.5 concentration forecasting. Primarily, the time delay and embedding dimensions are calculated via the mutual information-based method and false nearest neighbours approach to train the data, respectively. Then, the phase space reconstruction is performed to map the PM2.5 concentration time series into a high-dimensional space based on the obtained time delay and embedding dimensions. Finally, the SDNN is employed to forecast the PM2.5 concentration. The effectiveness of this approach is verified through extensive experimental evaluations, which collect six real-world datasets from recent years. To the best of our knowledge, this study is the first attempt to utilize a dendritic neural model to perform real-world air quality forecasting. The extensive experimental results demonstrate that the SDNN offers very competitive performance relative to the latest prediction techniques.

Sediment Bed Borehole Advection Method

This paper introduces and tests the Sediment Bed Borehole Advection Method (SBBAM), a low cost, point-measurement technique which utilizes a push-point probe to quantify the vertical direction and magnitude of Darcy flux at the surface water—groundwater sediment interface. The Darcy flux measurements are derived from the residence-time analysis of tracer arrival calculated from measured tracer concentration time-series data. The technique was evaluated in the laboratory using a sediment bed simulator tank at eight flow rates (1–90 cm/day). Triplicate test runs for each flow rate returned average errors between 4–20 percent; r2 = 0.9977.

Setting limits on neural network's predictive capacity in T1D blood glucose concentration

The ability to anticipate changes in blood glucose (BG) concentration would have a great impact on Type 1 diabetics (T1D). In order to create T1D treatment plans, patients collect a BG concentration time series. It has been demonstrated that various types of recurrent neural networks, such as Long Short Term Memory (LSTM), have success forecasting T1D BG concentrations. However, limited work has been done to characterize the T1D time series or set limits on neural network's predictive capacity. In this work, a T1D patient's 14 day BG concentration time series is studied. First, I test the time series stationarity. Then I use auto-correlation analysis, spectral analysis, and Gaussian process regression to characterize the T1D BG time series. Finally, the LSTM's prediction quality is quantified and interpreted at different prediction intervals. The success or failure of the LSTM's predictions are interpreted using the characterization of the time series.

Concentration Fluctuations from Localized Atmospheric Releases

We review the efforts made by the scientific community in more than seventy years to elucidate the behaviour of concentration fluctuations arising from localized atmospheric releases of dynamically passive and non-reactive scalars. Concentration fluctuations are relevant in many fields including the evaluation of toxicity, flammability, and odour nuisance. Characterizing concentration fluctuations requires not just the mean concentration but also at least the variance of the concentration in the location of interest. However, for most purposes the characterization of the concentration fluctuations requires knowledge of the concentration probability density function (PDF) in the point of interest and even the time evolution of the concentration. We firstly review the experimental works made both in the field and in the laboratory, and cover both point sources and line sources. Regarding modelling approaches, we cover analytical, semi-analytical, and numerical methods. For clarity of presentation we subdivide the models in two groups, models linked to a transport equation, which usually require a numerical resolution, and models mainly based on phenomenological aspects of dispersion, often providing analytical or semi-analytical relations. The former group includes: large-eddy simulations, Reynolds-averaged Navier–Stokes methods, two-particle Lagrangian stochastic models, PDF transport equation methods, and heuristic Lagrangian single-particle methods. The latter group includes: fluctuating plume models, semi-empirical models for the concentration moments, analytical models for the concentration PDF, and concentration time-series models. We close the review with a brief discussion highlighting possible useful additions to experiments and improvements to models.