Abstract: According to a transformer-architecture-based method and system for IoT for and smart control of urban integrated energy, urban integrated energy IoT information acquired by a terminal from different areas is output to a server after being standardized into a sequence signal, where energy demand prediction is performed through learning of a deep learning network configured on the server, a smart control network is configured and built by using the deep learning network, real-time prediction processing is performed on the urban integrated energy IoT information acquired in real time, and a result is uploaded to a workstation for reviewing. Based on the transformer architecture, the deep learning network is combined with a fast Fourier transform algorithm and inverse transform on a sequence, and an FFT-Attention mechanism is proposed. Compared with a conventional transformer architecture, frequency domain information of a sequence is given more emphasis.

Abstract: A deep learning-based method for fusing multi-source urban energy data and a storage medium are provided to perform data fusion on multi-source urban energy data found in big data and perform multi-scale and multimodal information fusion by using a cross-modal transformer, thereby implementing cross-modal mutual fusion of multi-source heterogeneous types of data to obtain a fused feature for prediction of a quantity of energy that will be used in the future and a quantity of energy that needs to be produced. The present disclosure proposes a multi-scale cooperative multimodal transformer architecture to enhance an effect of representation learned from an unaligned multimodal sequence. Not only there is a higher degree of correlation in multi-source urban energy data fusion, but also a system becomes more lightweight.

Abstract: A deep learning-based method for fusing multi-source urban energy data and a storage medium are provided to perform data fusion on multi-source urban energy data found in big data and perform multi-scale and multimodal information fusion by using a cross-modal transformer, thereby implementing cross-modal mutual fusion of multi-source heterogeneous types of data to obtain a fused feature for prediction of a quantity of energy that will be used in the future and a quantity of energy that needs to be produced. The present disclosure proposes a multi-scale cooperative multimodal transformer architecture to enhance an effect of representation learned from an unaligned multimodal sequence. Not only there is a higher degree of correlation in multi-source urban energy data fusion, but also a system becomes more lightweight.

Abstract: According to a transformer-architecture-based method and system for IoT for and smart control of urban integrated energy, urban integrated energy IoT information acquired by a terminal from different areas is output to a server after being standardized into a sequence signal, where energy demand prediction is performed through learning of a deep learning network configured on the server, a smart control network is configured and built by using the deep learning network, real-time prediction processing is performed on the urban integrated energy IoT information acquired in real time, and a result is uploaded to a workstation for reviewing. Based on the transformer architecture, the deep learning network is combined with a fast Fourier transform algorithm and inverse transform on a sequence, and an FFT-Attention mechanism is proposed. Compared with a conventional transformer architecture, frequency domain information of a sequence is given more emphasis.

Abstract: A server receives a request from a first application to search for retail availability of a specific item near a defined location. The defined location may be provided by the user, of determined by the server. The server performs a search for the specific item in a database storing data records for a plurality of items, wherein the data records are populated by users of the first application. Each data record includes a unique identifier, an item type, a item sub-type, an item description, a retail outlet where the item is available, and the number of likes for the item at particular retail outlets, as posted by other users for the item at the retail outlet. The server generates search results as a list of different retail outlets ranked in descending order by the number of likes for the specific item at those different retail outlets.

Abstract: A method for treating boundary cells over a time-step in a computational fluid dynamic process employing a computational mesh representation of a fluid system characterized by governing equations and having at least one moving boundary comprises: identifying interior cells, boundary cell faces, boundary vertices, interior vertices and vertex locations at the beginning of the time step; applying a calculation process that includes determining cell volumes based on Lagrangian locations of the interior and boundary vertices and calculating the value of at least one system thermodynamic property; calculating at least one flux value across one or more boundary cell volumes by returning the interior vertices to their initial locations.

Abstract: A method and apparatus for accessing a data representation of a model associated with a fluid system, the data representation including at least one interior cell and at least one ghost cell, calculating a physical volume value and physical surface area value for at least one interior cell and at least one ghost cell, generating at least one control volume based on one or more physical volume values, generating at least one control surface based on one or more physical surface area values; substituting one or more of the at least one control volume parameter and the at least one surface area for corresponding elements of mathematical conservation equations representative of the fluid system, and solving the mathematical conservation equations representative of the fluid system.