ABSTRACT The performance of regional groundwater level (GWL) prediction model hinges on understanding intricate spatiotemporal correlations among monitoring wells. In this study, a graph convolutional network (GCN) with a long short-term memory (LSTM) (GCN–LSTM) model is introduced for GWL prediction utilizing data from 16 wells located in the northeastern Xiangtan City, China. This model is designed to account for both the hybrid temporal dependencies and spatial autocorrelations among wells. It consists of two parts: the spatial part employs GCNs to extract spatial characteristics from a spatial self-similarity weight matrix and an attribute self-similarity weight matrix among wells; the temporal part utilizes a LSTM module to capture the temporal patterns of GWL sequences, along with monthly precipitation and temperature data. This model dynamically predicts changes in groundwater levels, achieving higher accuracy on average compared to single-well predictions using LSTM. By incorporating both temporal dependencies and spatial autocorrelations, the GCN–LSTM model demonstrated an average improvement in goodness-of-fit of approximately 11.21% over the LSTM-based model for individual wells. Its application holds significant reference value for the sustainable utilization and development of groundwater resources in Xiangtan City.
A reliable management system based on RFID was introduced to ensure the more efficient and effective operation of laboratory.In this system, reader/writer is on the basis of MFRC522;P89LPC932 is used as access controller; control nodes communicate with center server based on RS485 and TCP/IP ; development platform is based on Visual Studio2010 and database is established on SQL Server2008. This system has high reliability and good real-time capability because of the distributed structure and centralized control.
The paper analyzes the construction and basic characters of COM/DCOM component technology,and also analyzes the procedure that base on the DCOM′s multilayer application.COM/DCOM is a mainstream discreteness system which Microsoft Corporation advanced,COM provids a set of interface that allow client-side and server-side in a computer to communicate.DCOM is the extend of COM,the application,components,tools and knowledge based on the COM can be migrated to the area of standard distributing calculation.DCOM can be used to deal with the details of the low layer in the network protocol.The presented example which based on the COM/DCOM multilayer distributing model gets good effect in the application.
A wide range of product lifecycle management (PLM) maturity models are proposed to assess the relative position of companies on their road to complete PLM implementation. However, it is a tough job for the company to dynamically evaluate the gradual process of PLM maturity by using existing values and accurately make decisions of improving PLM maturity by selecting the optimum alternative. A fuzzy PLM components maturity model (PCMA) is presented to build the internal logical relationship between maturity levels and existing values that can automatically predict the unknown PLM maturity levels. A fuzzy AHP–VIKOR methodology is used to make a decision among option PLM strategies. The weights of the criteria are determined by fuzzy pairwise comparison matrices (PCM). The weights of alternatives with respect to criteria are calculated by fuzzy VIKOR. The fuzzy AHP–VIKOR is a compromise solution and has the ability of transfer subjective and implicit linguistics into objective and transparent data. A numerical example illustrates and clarifies the running steps of the proposed methodology.
In order to construct concept lattice without rebuilding the whole structure, we introduced some basic notions in FCA and Graph, such as concept, cover, maxmod and so on. In Section 2, we elaborated the procedures of a proposed algorithm for computing the covers by adding non-dominating maxmods to the intent of a given concept. After that we analyzed time complexity with comparison to other relevant researches and proved that our algorithm is superior to other algorithms. In the end, we discussed a few open issues.
Magnetic resonance (MR) images are usually limited by low spatial resolution, which leads to errors in post-processing procedures. Recently, learning-based super-resolution methods, such as sparse coding and super-resolution convolution neural network, have achieved promising reconstruction results in scene images. However, these methods remain insufficient for recovering detailed information from low-resolution MR images due to the limited size of training dataset.To investigate the different edge responses using different convolution kernel sizes, this study employs a multi-scale fusion convolution network (MFCN) to perform super-resolution for MRI images. Unlike traditional convolution networks that simply stack several convolution layers, the proposed network is stacked by multi-scale fusion units (MFUs). Each MFU consists of a main path and some sub-paths and finally fuses all paths within the fusion layer.We discussed our experimental network parameters setting using simulated data to achieve trade-offs between the reconstruction performance and computational efficiency. We also conducted super-resolution reconstruction experiments using real datasets of MR brain images and demonstrated that the proposed MFCN has achieved a remarkable improvement in recovering detailed information from MR images and outperforms state-of-the-art methods.We have proposed a multi-scale fusion convolution network based on MFUs which extracts different scales features to restore the detail information. The structure of the MFU is helpful for extracting multi-scale information and making full-use of prior knowledge from a few training samples to enhance the spatial resolution.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)