Abstract: A training method based on an improved protograph neural decoder includes the following steps: a to-be-trained decoding network is constructed based on an initial variable sub-network layer, an initial check sub-network layer and a preset shuffled belief-propagation (BP) sub-network layer; the initial variable sub-network layer, the initial check sub-network layer and the preset shuffled BP sub-network layer are updated and trained by calculating log-likelihood ratio (LLR) based on a preset mean square error loss function and a preset decoder objective function to obtain a target protograph neural decoder; and the preset mean square error loss function is configured to calculate a loss value between output information of the check sub-network layer and the preset shuffled BP sub-network layer. The target protograph neural decoder includes an optimized variable sub-network layer, an optimized check sub-network layer and an optimized shuffled BP sub-network layer. A training device is also provided.

Abstract: A method for preparing immobilized cells for producing mannose, and a method using same for producing mannose, comprising: fermenting to separately obtain fermentation broths of Escherichia coli or Bacillus subtilis expressing a-glucan phosphorylase, phosphoglucomutase, glucose phosphoisomerase, mannose-6-phosphate isomerase and mannose-6-phosphate phosphatase, and mixing the fermentation broths to obtain a mixed fermentation broth.

Abstract: A training method based on an improved protograph neural decoder includes the following steps. A to-be-trained decoding network is constructed based on an initial variable sub-network layer, an initial check sub-network layer and a preset shuffled belief-propagation (BP) sub-network layer. The initial variable sub-network layer, the initial check sub-network layer and the preset shuffled BP sub-network layer are updated and trained by calculating log-likelihood ratio (LLR) based on a preset mean square error loss function and a preset decoder objective function to obtain a target protograph neural decoder. The preset mean square error loss function is configured to calculate a loss value between output information of the check sub-network layer and the preset shuffled BP sub-network layer. The target protograph neural decoder includes an optimized variable sub-network layer, an optimized check sub-network layer and an optimized shuffled BP sub-network layer. A training device is also provided.

Abstract: Provided are a Bacillus subtilis genetically engineered bacterium for producing tagatose and a method for preparing tagatose. The genetically engineered bacterium comprises constructing thermostable ?-glucan phosphorylases, thermostable glucose phosphomutases, thermostable glucose phosphate isomerases, thermostable 6-tagatose phosphate epimerases, and thermostable 6-tagatose phosphate phosphatases which are independently expressed or co-expressed. The usage of the genetically engineered bacterium can effectively convert starch into tagatose. Compared with existing methods for producing tagatose, the method has advantages such as suitability for whole-cell recycling, high safety, high yield, simple production process, low cost, and easiness in large-scale preparation.

Abstract: The invention discloses a circumferential laser dynamic display device, including an inner frame, having an inner frame pattern with multiframe microstructures made on surface; a sheet metal part, at least provided with first and second through holes, wherein the inner frame pattern is directly irradiated by a first beam of laser through the first through hole and by a second beam of laser through the second through hole; and a rotating shaft, with one end fixedly connected with a driving shaft, and the other end rotatably connected with the inner frame pattern. The rotating shaft can drive the inner frame pattern to do translational movement when rotating; wherein the translational movement of the inner frame enables the inner frame pattern to do circumferential movement relative to laser. The dynamic display device disclosed by the invention has simple structure, reasonable design and low production cost, and achieves circumferential dynamic image display.

Abstract: The present invention discloses a laser animation display method. A display system includes three parts, i.e., a laser source, a diffraction optical element and a mechanical driving device. The display method includes the following steps: S1: the laser source emits laser at first, and the laser is caused to be incident on the pattern of a first microstructure on the diffraction optical element; S2: the diffraction optical element is driven by the mechanical driving device to translate to irradiate the pattern of a second microstructure of the diffraction optical element with the laser, and the patterns on other microstructures are sequentially irradiated with the laser according to the same method; and S3: cyclic movement is kept, thereby achieving an animation effect on a screen. The present invention implements laser animation display on the premise of no additional motion control unit and almost no increase of the system size and cost.

Abstract: The present invention discloses a laser animation display method. A display system includes three parts, i.e., a laser source, a diffraction optical element and a mechanical driving device. The display method includes the following steps: S1: the laser source emits laser at first, and the laser is caused to be incident on the pattern of a first microstructure on the diffraction optical element; S2: the diffraction optical element is driven by the mechanical driving device to translate to irradiate the pattern of a second microstructure of the diffraction optical element with the laser, and the patterns on other microstructures are sequentially irradiated with the laser according to the same method; and S3: cyclic movement is kept, thereby achieving an animation effect on a screen. The present invention implements laser animation display on the premise of no additional motion control unit and almost no increase of the system size and cost.

Abstract: The invention discloses a circumferential laser dynamic display device, including an inner frame, having an inner frame pattern with multiframe microstructures made on surface; a sheet metal part, at least provided with first and second through holes, wherein the inner frame pattern is directly irradiated by a first beam of laser through the first through hole and by a second beam of laser through the second through hole; and a rotating shaft, with one end fixedly connected with a driving shaft, and the other end rotatably connected with the inner frame pattern. The rotating shaft can drive the inner frame pattern to do translational movement when rotating; wherein the translational movement of the inner frame enables the inner frame pattern to do circumferential movement relative to laser. The dynamic display device disclosed by the invention has simple structure, reasonable design and low production cost, and achieves circumferential dynamic image display.

Abstract: This invention is directed to an improvement in a process for producing a fire suppressing mist comprised of finely divided water droplets and a fire suppressing gas in response to fires in an enclosed area. The improvement resides in the finding that one can reduce the size of water droplets generated in a nozzle system designed for generating said fire suppressing mist at low pressure by using deionized water as the water source. The fire suppressing mist can also include a low concentration of surfactant.

Abstract: An apparatus for treatment a substrate with an activated reactive gas which includes a processing chamber, an exhaust manifold, a conveyor adapted to sequentially introduce into the processing chamber untreated portions of the substrate for said treatment and to sequentially remove from the processing chamber treated portions of the substrate, wherein the length of the substrate exceeds a dimension of the inner volume of the processing chamber, a distribution conduit disposed in the processing chamber, wherein the length of the distribution conduit is approximately equal to the width of the substrate, and wherein the distribution conduit has a number (N) of openings, each opening has a cross sectional area (Ao), a cross sectional area of the distribution conduit (Ac), and wherein a maximum cross-sectional area (N*Ao) of the openings can be determined by the following expression: 1.0*Ac>N*Ao?0.1*Ac.

Abstract: A vessel for the storage and transportation of bulk volumes of a fluid is described herein and a method for using same. The vessel contains a plurality of dividers that apportion the internal volume into a number of sections. The dividers within the vessel aid in minimizing sloshing of the fluid contained therein during transport. In addition, the dividers optimize the fluid flow pattern thereby allowing for the continuous purge of the vessel without the need for the application of a partial or full vacuum.

Abstract: Disclosed herein is a system and method comprising same to supply oxygen having a purity of 30% or greater into containment vessels for the aquaculture of marine animals such as shrimp. In one embodiment of the present invention, the system comprises a plurality of oxygen generators in fluid communication with a plurality of oxygen injectors to supply the containment vessels having an aqueous medium and marine animals contained therein with oxygen to maintain the dissolved oxygen content above a certain level.

Abstract: A vessel for the storage and transportation of bulk volumes of a fluid is described herein and a method for using same. The vessel contains a plurality of dividers that apportion the internal volume into a number of sections. The dividers within the vessel aid in minimizing sloshing of the fluid contained therein during transport. In addition, the dividers optimize the fluid flow pattern thereby allowing for the continuous purge of the vessel without the need for the application of a partial or full vacuum.