The water quality monitoring data reflects the ecological environment of the river and its changing trend. The accurate analysis and prediction of water quality is of great significance to improve the efficiency of river management and maintenance, ensure the safety and stability of the ecological environment of the river basin. The traditional water quality prediction model combines hydrodynamic theory and water quality change process mechanism, which can represent the overall change trend of water quality. However, this method is mostly aimed at a fixed watershed, with poor portability ability and difficult to accurately predict key areas. Although some new mathematical statistical methods and machine learning methods can achieve high accuracy in the prediction of water quality data from a single station, the spatial correlation characteristics between multiple stations are ignored in the comprehensive water quality prediction of the basin, resulting in a decline in accuracy. In this paper, based on the spatial distribution and graph structure characteristics of multiple water quality monitoring stations in the basin, the temporal graph convolutional neural network is used to analyse the time series characteristics of water quality, and realize the prediction of water quality data from multiple monitoring stations. The analysis of water quality monitoring data of a river located in the Guangdong Province of China proves that, compared with the commonly used model LSTM, the T-GCN model can better fit the spatial correlation characteristics and time series characteristics of water quality monitoring data, and achieve a more accurate water quality predictive analysis. Based on the water quality monitoring data of multiple stations, T-GCN model can reflect the changing trend of water quality in the entire basin, and accurately predict the water quality changing in specific locations. The research results can provide a more accurate and effective reference for river operation and management.
Non-contact identification of objects can be achieved using radio frequency technology (RFID). This paper proposes a location method based on Support Vector Machine (SVM), which harnesses RFID technology to locate objects. By analysing the shortcomings of standard SVM algorithm in dealing with significantly misclassified samples and multi-classification problems, a fuzzy iterative OVO-FSVM algorithm is proposed. Combined with RFID hardware equipment, a RFID-based location recognition system was designed. The experimental results show that the proposed location system demonstrated an accuracy of 77.8%, which can meet the needs of system positioning.
Surface subsidence is a major engineering geological issue in the construction of major cities worldwide, posing a serious threat to the safety of the general public. Therefore, achieving precise monitoring of surface deformation holds significant value in practical applications. Interferometric Synthetic Aperture Radar (InSAR) offers advantages such as all-day, all-weather, large-area coverage, and high precision, making it widely used in the monitoring of broad-scale surface deformation. This study is based on the SBAS-InSAR technique, which effectively addresses the limitations of InSAR, such as temporal and spatial decorrelation, and the difficulty in obtaining accurate deformation information. Taking the Pearl River Bridge in the Nansha District of Guangzhou as the research object, Sentinel-1 radar satellite images from June 2015 to July 2019 were acquired to conduct surface deformation monitoring in the Nansha District of Guangzhou. This resulted in the formation of a dataset for monitoring surface subsidence and performing subsidence analysis of the Pearl River Bridge. The research demonstrates that the SBAS-InSAR technique has strong capabilities in the sequential monitoring of nonlinear surface deformation, and the extracted deformation results can provide scientific evidence for analysis and decision-making by production units and government departments.
High resolution remote sensing image resource is a very important geographic information recording medium. It has great utilization value and can be digitized, informationized and transmitted through network in space. This paper mainly introduces the idea and implementation process of forming a high resolution remote sensing image resource classification system based on the combination of E-Cognition high resolution remote sensing image receiving technology and blind area recognition theory. And it also briefly analyzes the application status of this method. In addition, the landscape planning platform is designed later, and the performance of the platform is tested. The test results show that the landscape planning platform based on eCognition high-resolution remote sensing images can accurately locate user needs, with user accuracy of more than 80%.
Abstract: or the purpose of preventing misoperation from damaging the automobile engine and saving the teaching cost, this paper uses MR technology to carry out relevant research on the holographic teaching application system of automobile engine structure, disassembly and maintenance. Use UG and 3ds-max and other software to model the 3D model of the automobile engine at a scale of 1:1, and the model is consistent with the rendering effect of the real engine. Use Unity 3D as the MR development platform, and use C # language programming to realize the interaction and animation functions of the system. The developed system includes six functional modules, including exploded view, disassembly and assembly, working principle, transparent sectioning, structure introduction and free movement. The system combines MR technology with engine structure cognition, working principle analysis and disassembly and maintenance, realizes the integration of virtual equipment, interactive animation and real space equipment, and solves the practical problems such as high purchase cost due to high engine price and complex maintenance and disassembly, which are easy to damage the engine by misoperation [1].
