4results about How to "Does not affect the process" patented technology

This invention provides the preparation and application of a strain of *Lactiplantibacillus plantarum* INM3104 that reduces raffinose in whole grain dough. This strain, *Lactiplantibacillus plantarum* INM3104, which degrades raffinose, was deposited at the Guangdong Provincial Microbial Culture Collection Center on April 11, 2024, with accession number GDMCC NO:64512. The advantages are: it only reduces the raffinose content in whole grain products, while also improving the flavor and quality of the whole grain products; the whole grain products produced using this preparation do not affect the overall processing of the whole grain products after fermentation, not only improving the taste of the whole grain products but also reducing the adverse factors associated with whole grain products.

The application discloses an AI model-based RSS subscription source intelligent screening and automatic article generation method and related device, relates to the technical field of artificial intelligence, and comprises the following steps: acquiring a subscription configuration containing an RSS source basic URL, a push cycle and the like, processing tasks in batches through timed scheduling and monitoring states, and improving service reliability; intelligently identifying RSS sources and reducing user configuration thresholds; preprocessing heterogeneous data into a standard format and guaranteeing data flow efficiency; filtering expired content according to a time window calculated according to the push cycle and relieving information overload; dynamically grouping according to an AI screening model Token upper limit and improving AI call success rate; implementing structured screening by using a preset JSON Schema combined with screening prompt words, improving precision compared with a traditional scheme; secondarily grouping according to an AI generation model Token upper limit after recombining results and guaranteeing generation quality; automatically generating articles combined with generation prompt words, reducing the time consumed by manual work to the minute level; generating a matching title, saving a notification and improving experience. The method realizes the upgrade of RSS from passive aggregation to active intelligent processing and is suitable for multi-field content services.

The utility model discloses an internal mold supporting device for prefabricating a box girder, which comprises a mounting plate, a quick assembly is arranged at the top of the mounting plate, the quick assembly comprises a threaded rod and a first moving plate, the threaded rod is connected to the top of the mounting plate through a bearing, and the first moving plate is connected to the side wall of the threaded rod through threads. First U-shaped frames are fixedly installed on the two sides of the first movable plate respectively, and connecting rods are hinged to the inner side walls of the first U-shaped frames. According to the internal mold supporting device for prefabricating the box girder, the whole internal mold supporting device can be rapidly fixed in an internal mold through the arranged rapid assembly assembly, fixing is easy and convenient, fixing work can be easily completed by one worker, unnecessary workload is effectively reduced, the working efficiency of the worker is guaranteed, work can be completed within the shortest time, and the working efficiency of the worker is improved. And the working process is not affected, the using effect is improved accordingly, and application and popularization are facilitated.

The application discloses an AI model-based RSS subscription source intelligent screening and automatic article generation method and related device, relates to the technical field of artificial intelligence, and comprises the following steps: acquiring a subscription configuration containing an RSS source basic URL, a push cycle and the like, processing tasks in batches through timed scheduling and monitoring states, and improving service reliability; intelligently identifying RSS sources and reducing user configuration thresholds; preprocessing heterogeneous data into a standard format and guaranteeing data flow efficiency; filtering expired content according to a time window calculated according to the push cycle and relieving information overload; dynamically grouping according to an AI screening model Token upper limit and improving AI call success rate; implementing structured screening by using a preset JSON Schema combined with screening prompt words, improving precision compared with a traditional scheme; secondarily grouping according to an AI generation model Token upper limit after recombining results and guaranteeing generation quality; automatically generating articles combined with generation prompt words, reducing the time consumed by manual work to the minute level; generating a matching title, saving a notification and improving experience. The method realizes the upgrade of RSS from passive aggregation to active intelligent processing and is suitable for multi-field content services.