Gas supply system risk assessment is a serious and important problem in cities. Existing methods tend to manually build mathematical models to predict risk value from single-modal information, i.e., pipeline parameters. In this paper, we attempt to consider this problem from a deep-learning perspective and define a novel task, Urban Gas Supply System Risk Assessment (GSS-RA). To drive deep-learning techniques into this task, we collect and build a domain-specific dataset GSS-20K containing multi-modal data. Accompanying the dataset, we design a new deep-learning framework named GSS-RiskAsser to learn risk prediction. In our method, we design a parallel-transformers Vision Embedding Transformer (VET) and Score Matrix Transformer (SMT) to process multi-modal information, and then propose a Multi-Modal Fusion (MMF) module to fuse the features with a cross-attention mechanism. Experiments show that GSS-RiskAsser could work well on GSS-RA task and facilitate practical applications. Our data and code will be made publicly available.

With the rapid development of e-commerce based on the network, the logistics distribution support of e-commerce is becoming more and more obvious. The optimization of vehicle distribution routing can improve the economic benefit and realize the scientific of logistics [1]. Therefore, the study of logistics distribution vehicle routing optimization problem is not only of great theoretical significance, but also of considerable value of value. Particle swarm optimization algorithm is a kind of evolutionary algorithm, which is based on the random solution and the optimal solution by iteration, and the quality of the solution is evaluated through fitness. In order to obtain a more ideal logistics scheduling scheme, this paper proposes a logistics model based on particle swarm optimization algorithm.

Remote sensing is a technology for monitoring surface information in real time, remote sensing image interpretation has important application value in military and civil fields [1]. In this paper, aiming at the design of remote sensing image segmentation algorithm, we propose a remote sensing image segmentation method based on multi-Agent system and fuzzy clustering. The proposed method extracts and fuses the intensity, texture and edge information in collaboration and penalization, and it provides theoretical and technical support for remote sensing image segmentation.

In the seventies, the United States military organization for operational requirements developed wireless packet network, after decades of development and evolution. Wireless packet networks gradually became today's wireless self-organized network. Compared with the traditional network, the wireless self-organizing network with distributed, self-organizing, rapid deployment, low cost, etc. Nowadays the technology is not just a single application in military operations, it has been widely applied in social life practice, such as: sudden disaster communications, mobile office, etc. But because wireless self-organizing network development time is shorter, the imperfection of the related technology, wireless self-organizing network is faced with a lot of technical research In this paper, the wireless self-organized network comprehensively expounded, systematically introduces its technical characteristics, the present situation of the network structure, application and related research, etc.

Fuzzy similarity ratio method is a widely used method of fuzzy decision-making, not only to classification, but also to sort. In this paper, we constructed a geochemical anomaly evaluation model using the Fuzzy similarity ratio method; we also program it with C# language and applied the model to practical work. Practice has proved that the fuzzy decision-making applies to the evaluation of geochemical anomalies can effectively improve the accuracy of the evaluation.