Abstract: A multi-task joint perception network model and detection method for traffic road surface information can simultaneously detect a lane line and a drivable area. A coordinate attention mechanism is integrated into a traditional feature extraction network to ensure that a feature extraction effect is enhanced while a calculated amount is not increased. In a neck network, a dilated convolution residual module is proposed to enhance performance of prediction of details by the network, and a decoder part shares features of the drivable area into lane line detection to enhance a lane line detection effect under complex road conditions. In a training stage, there is provided a alternating optimization training method to improve integral segmentation performance of the model. The multi-task joint perception network model and detection method realizes quite high accuracy and excellent speed performance in a challenging BDD100K dataset.

Abstract: A gradient-structured ultra-fine bainitic low-alloy rail steel and a preparation method thereof are provided in the present application, belonging to the technical field of steel metallurgy. The rail steel to be treated is subjected to three thermal processing treatments so that the gradient-structured ultra-fine bainitic rail steel has a gradient structure consisting of a ferrite-pearlite dual phase microstructure and an ultra-fine bainite microstructure. The gradient structure of the present application enables the matrix of the rail to be a ferrite-pearlite composite phase structure, and only the surface layer of the service surface to be ultrafine bainite.

Abstract: A gradient-structured ultra-fine bainitic low-alloy rail steel and a preparation method thereof are provided in the present application, belonging to the technical field of steel metallurgy. The rail steel to be treated is subjected to three thermal processing treatments so that the gradient-structured ultra-fine bainitic rail steel has a gradient structure consisting of a ferrite-pearlite dual phase microstructure and an ultra-fine bainite microstructure. The gradient structure of the present application enables the matrix of the rail to be a ferrite-pearlite composite phase structure, and only the surface layer of the service surface to be ultrafine bainite.

Abstract: The disclosure relates to the technical field of batteries. Disclosed are a battery compartment, a battery pack and an electric device. The battery compartment includes a battery shell and a battery cell assembly; the battery shell is provided with a male port opening and a female port opening; the battery cell assembly is disposed in the battery shell, and includes a connection bar and a plurality of battery cells; the connection bar is configured to electrically connect the plurality of battery cells, and is provided with a male port pole and a female port pole; and the male port pole protrudes from the male port opening, and the female port pole protrudes from the female port opening.

Abstract: Disclosed are an iron-aluminum alloy and its preparation method. The iron-aluminum alloy comprises, by weight, 50% to 80% of iron and the balance of aluminum. The method comprises: adding metal aluminum or molten aluminum to a container, wherein the temperature of the molten aluminum is between 700° C. and 800° C.; adding a metal iron raw material to the molten aluminum, closing a furnace cover, measuring the pressure, and introducing argon to ensure that the interior of a magnetic induction furnace is in a positive-pressure state, and stirring the mixture with a graphite stirring head; powering on and heating so that the metal aluminum or the molten aluminum is heated to 1000° C. or above and molten, and holding the temperature between 1000° C. and 1500° C.; and after alloying is completed, cooling to about 1000 t, opening the furnace cover, and taking the iron-aluminum alloy out.

Abstract: Disclosed are a silicon-aluminum alloy and its preparation method. The method comprises: adding aluminum metal or molten aluminum into a container, wherein the temperature of the molten aluminum is between 700° C. and 800° C.; adding a semi-metallic silicon raw material to the molten aluminum, closing a furnace cover, carrying out vacuumization, and introducing argon, to ensure that the interior of a magnetic induction furnace is in a positive-pressure state, and stirring the aluminum metal or molten aluminum with a graphite stirring head; powering on and heating so that the aluminum metal or molten aluminum is heated to 1000° C. or above and molten, and holding the temperature between 1000° C. and 1500° C.; and after alloying is completed, cooling the molten aluminum to 1000° C. or below, opening the furnace cover, pouring the silicon-aluminum alloy into a corresponding mold, and cooling for molding.

Abstract: Disclosed are a manganese-aluminum alloy and its preparation method. The manganese-aluminum alloy comprises, by weight, 5% to 90% of manganese and the balance of aluminum. The method comprises: adding metal aluminum or molten aluminum to a container, the temperature of the molten aluminum being between 700° C. and 800° C.; adding a metal manganese raw material to the molten aluminum, closing a furnace cover, measuring the pressure, and introducing argon to ensure that the interior of a magnetic induction furnace is in a positive-pressure state, and stirring the mixture with a graphite stirring head; powering on and heating the metal aluminum or the molten aluminum to 1000° C. or above, melting, and holding the temperature between 1000° C. and 1500° C.; and after alloying is completed, cooling to 850° C. or below, opening the furnace cover, and taking a manganese-aluminum alloy out.