The anomaly detection for communication networks is significant for improve the quality of communication services and network reliability. However, traditional communication monitoring methods lack proactive monitoring and real-time alerts and the prediction effect of a single machine learning model on communication data containing multiple features is not ideal. To solve the problem, A prediction-then-detection anomaly detection method was proposed, and quantitative assessment of network anomalies was developed. Specifically, anomaly-free data was obtained by eliminating outliers, and the long short-term memory (LSTM) and autoregressive integral moving average (ARIMA) were combined via residual weighting to predict the future state of the key performance indicators (KPI) without outliers. Anomalies were identified using the error comparison between the prediction and actual values, and the network condition was quantified using the scoring method. It is observed that the proposed LSTM-ARIMA hybrid model has better prediction effect, which can well represent the performance of KPIs of the future state, and the prediction-then-detection anomaly detection method has excellent performance on both precision and recall.

With the cloud computing is becoming mature, many of its characteristics for meteorological data processing is extremely important. Since HDFS is designed for reading and writing large files, it’s difficult to be taken advantage of small meteorological files. In this paper, an improved approach on HDFS is proposed for small meteorological files, small files are to be merged, indexed, and blocks are compressed, the pressure of memory on master node occupied by metadata is relieved, the speed of reading and writing small files is increased, read speed is increased by 50%, and write speed is up to 3-4 times of the original, saving about 2/3 of storage space and computing performance has also been improved. Thus, meteorological data processing can make use of cloud computing platform more closely.

In this paper we present a new method for three-dimensional (3D) computerized modeling of geological objects from sets of intersected cross-sections. Geometric reconstruction from a set of topological and polygonal cross-sections is used in many cases of geological reconstruction now. However, most of current methods require that cross-sections must be parallel. Our method is based on sets of intersected cross-sections. With the help of cross-sections at another direction, this method can effectively solve the correspondence problem, matching components in adjacent sections. Besides, many complex geological phenomena can be reconstructed precisely, for example, stratum dying out position can be well defined in another directional sections. Last but not least, because sections at different directions are used to construct geological surfaces, model quality is greatly improved. The major steps and key algorithms in this method are all discussed in detail. A 3-D software platform has been developed based on this method. A case study in An Shan, Liao Ning Province, P.R. China shows that the method can be applied to practical and complex geological areas.

With the increase of public concern about health and smoking, the authorities have gradually tightened the control of tar content in cigarettes, making reconstituted tobacco a growing concern for tobacco companies. Tobacco stems are used as the main raw material for reconstituted tobacco, but they contain a large number of small broken impurities mainly from cigarette butts, which are difficult to remove efficiently by air selection and manual methods. Detection schemes for cigarette butt impurities based on computer vision and deep learning are still difficult. The scarcity of images containing foreign impurities in cigarette butts and the small size of impurities limit the efficient application of deep learning algorithms. In view of the small impurities' characteristics, this paper optimizes the model structure of the YOLOv7 algorithm, and only retains the two detection head structures with high feature resolution, which reduces the model parameters by 29.68%. Using online data augmentation and transfer learning, the difficulty of small sample datasets is overcome. After the CutMix, Mosaic, Affine transformation, Copy-paste data augmentation in this paper, the model precision is increased by 6.95%, and the recall rate is increased by 10.51%. Detection FPS has been increased from 99 FPS to 111 FPS. Precision and recall rate reached 97.21% and 92.11%. Compared with YOLOv4_csp, the precision is in-creased by 11.58%, and the recall rate is increased by 0.48%. It shows that the improved YOLOv7xs model has the potential for wide application in small target recognition. At the same time, it has shown the potential to avoid the harm of toxic substances produced by cigarette impurities in the combustion process and promotes the application of computer vision and deep learning in industrial production.

Explosion is the most dangerous accident during direct combustion of lowconcentration gas. Few investigations have focused on explosion risk assessment of low-concentration gas safe combustion system. This study aims to analyze explosion risk of above system. Due to insufficient accident data for calculating the failure probability of explosion, expert evaluation method and fuzzy algorithm were adopted to obtain the failure probability of basic events (BEs) and Fuzzy Analytic Hierarchy (FAHP) Process method used to improve the calculation of expert weights. Then qualitative and quantitative analysis was carried out using fault tree analysis. Results show that the failure probability of system explosion is 0.085%. Events with high-risk probability are mainly on equipment management and personnel behavior norms. Safe combustion system has excellent performance on tempering prevention, explosion resistance and explosion relief. The outcomes of this study are expected to help the safety professionals while formulating safety management measures for low-concentration gas safety combustion system.