Hydrate can easily form and cause plugging in offshore oil and gas pipelines. As the managing strategy moves from full avoidance to dynamic control, antiagglomerants (AAs) could be an economical and efficient option for hydrate management. Searching for new efficient hydrate AAs is always needed. In this work, the antiagglomerating performances of 57 commercial chemicals, including cationic surfactants, zwitterionic surfactants, nonionic surfactants, and polymers, to methane hydrate in an oil–gas–water system were evaluated by using a rocking-cell apparatus. The slider-moving trajectories were recorded during hydrate formation. The performances of AAs were evaluated based on the slider-moving profiles at various hydrate fractions. It was found that cocamidopropyl dimethylamine, a nonionic surfactant that is labeled as an A class AA, could fully avoid hydrate plugging when the hydrate fraction was under 18.76%. The slider-moving trajectory in the full range of the cell indicated that the hydrate was well-dispersed in the liquid phase. Besides, 23 chemicals mainly from cationic and zwitterionic surfactants, which are labeled as B class AAs, could partially avoid hydrate plugging at various hydrate fractions (10.96–19.92%). The slider could move in a partial range of the cell, indicating that the hydrate may agglomerate and accumulate at the end of the cell. The rest of the chemicals were labeled as class C AAs, with which the slider was fully stuck. The working mechanism is discussed for an insightful understanding on the antiagglomerating performances of the screened AAs. Chemicals with a combined presence of an amide group and an amine group exhibit great potential for hydrate antiagglomeration, while the hydrogen-bonding ability of the hydrophilic amine headgroup may be adversely affected by the incorporation of anions.

Crime prediction is of great significance to the formulation of policing strategies and the implementation of crime prevention and control. Machine learning is the current mainstream prediction method. However, few studies have systematically compared different machine learning methods for crime prediction. This paper takes the historical data of public property crime from 2015 to 2018 from a section of a large coastal city in the southeast of China as research data to assess the predictive power between several machine learning algorithms. Results based on the historical crime data alone suggest that the LSTM model outperformed KNN, random forest, support vector machine, naive Bayes, and convolutional neural networks. In addition, the built environment data of points of interests (POIs) and urban road network density are input into LSTM model as covariates. It is found that the model with built environment covariates has better prediction effect compared with the original model that is based on historical crime data alone. Therefore, future crime prediction should take advantage of both historical crime data and covariates associated with criminological theories. Not all machine learning algorithms are equally effective in crime prediction.

No abstract

No abstract

No abstract