Abstract: A method for training a contrastive learning model, acquiring an optimization item, and pushing information includes: obtaining sample data including first category information and query information; predicting a correlation between the first category information and the query information using the contrastive learning model; establishing a loss function based on the prediction results, wherein the loss function is configured to characterize accuracy of the prediction results; optimizing the loss function based on a semantic relationship between the first category information and second category information, wherein the first category information and the second category information correspond to different semantic information in a category tree, and the semantic relationship is related to relative positions of the first category information and the second category information within the category tree; and training the contrastive learning model based on the optimized loss function.

Abstract: A composition, and a preparation method therefor and an application thereof. The composition comprises the following raw materials in percentage by weight: 2.5%-80% of an inactivated Kluyveromyces cell, 1%-50% of edible oil, and 18%-96.5% of water. The composition is stable and uniform, has high nutritional components such as protein and dietary fiber, and can be used in beverages and foods.

Abstract: This application relates to thermal management systems and vehicles. An example thermal management system includes a compressor, a condenser, a first heat exchanger, a second heat exchanger, and a first electronic expansion valve. The compressor includes an input port and an output port. The compressor is configured to compress a refrigerant from the input port and output the compressed refrigerant through the output port. The condenser includes a first interface and a second interface. The second interface is connected to the output port. The first heat exchanger includes a third interface and a fourth interface. The third interface is connected to the output port, and the fourth interface is connected to the input port. The second heat exchanger includes a fifth interface and a sixth interface. The sixth interface is connected to the input port.

Abstract: A fast measurement method for micro-nano deep groove structure based on white light interference, including: establishing a white light interference system, using the white light interference system to measure the structure of the groove, the CCD camera collects and obtains multiple groups of groove interferograms and the serial number corresponding to each groove interferogram in each group; processing each group of groove interferograms of the groove sample to obtain the maximum contrast of each group of groove interferograms and the 3D reconstruction diagram of the local structure; extracting the interface reconstruction diagram in the 3D reconstruction diagram of the local structure according to each group of the groove interferograms; after splicing the interface reconstruction diagrams corresponding to all groups of groove interferograms, obtaining a 3D structural reconstruction diagram of the groove sample, and measuring the depth and width of the groove sample according to the 3D structural reconstruc

Abstract: A fast measurement method for micro-nano deep groove structure based on white light interference, including: establishing a white light interference system, using the white light interference system to measure the structure of the groove, the CCD camera collects and obtains multiple groups of groove interferograms and the serial number corresponding to each groove interferogram in each group; processing each group of groove interferograms of the groove sample to obtain the maximum contrast of each group of groove interferograms and the 3D reconstruction diagram of the local structure; extracting the interface reconstruction diagram in the 3D reconstruction diagram of the local structure according to each group of the groove interferograms; after splicing the interface reconstruction diagrams corresponding to all groups of groove interferograms, obtaining a 3D structural reconstruction diagram of the groove sample, and measuring the depth and width of the groove sample according to the 3D structural reconstruc

Abstract: The invention claims an engine surge fault prediction system and method based on fusion neural network model, belonging to the technical field of time series data prediction. The system comprises a prediction module, used for generating prediction time series with a specified length through 3D structure time series data of the engine; a feature extraction module, used for extracting the local features of the prediction time series, semantic relations among data, and overall trend features of the series; a classification module, used for judging if there is a surge fault according to the local features of the prediction time series, semantic relations among data, and overall trend features of the series.

Abstract: The invention discloses a pulsating chlorination machine for purification of city water. The machine includes a body, two ends of which are respectively a water inlet and outlet end between which a converging pipe, a throat pipe and a diverging pipe are connected in turn; and a liquefied chlorine input port located below a middle part of the throat pipe and connecting with the throat pipe via a cone-shaped valve seat. A normally closed cone-shaped valve rod is mounted over the cone-shaped valve seat and inserted into the throat pipe through a vertical circular hole on a top of the body, and moves up and down under the control of a pulsating electromagnetic executive mechanism to open/close the liquefied chlorine input port. Through the liquefied chlorine input port the liquefied chlorine flows into the throat pipe and out from the diverging pipe and then into a clear-water reservoir.

Abstract: A direct chlorination system and device characterized by feeding of liquefied chlorine into water without vaporization in a process that liquefied chlorine from a source liquefied chlorine cylinder is led to a liquefied chlorine settlement cylinder for removal of solid impurities, a metering pump for a pressurization and flow rate control at the metering pump and a liquefied chlorine feeder for mixing with and dissolving in pressurized water to become a high concentration chlorine solution which is then led to a water injector for feeding into a rapid water filter, a mixing basin and a clearing water reservoir for reaction and disinfection before city water is sent to a city water distribution network, and possessing the following merits: upgrading potable water quality, minimizing initial investment, lowering operating cost, and providing a simply but reliable, easy-to-control device that can facilitate water works automation.