2results about How to "Fast judgment" patented technology

Embodiments of the present application disclose a grid-connected inverter control method, device, equipment and storage medium. The grid-connected inverter control method comprises: calculating a grid voltage prediction value in a two-phase stationary coordinate system in a current period according to a grid voltage sample value in the two-phase stationary coordinate system in a previous period, a variation angle of a three-phase grid voltage space vector and a preset numerical relationship; judging whether a grid voltage drop occurs according to a numerical relationship between an absolute value of a difference between the grid voltage sample value in the two-phase stationary coordinate system in the current period and the grid voltage prediction value and a preset voltage value; and controlling the grid-connected inverter according to a judgment result. The technical scheme of the embodiments of the present application is helpful to improve the judgment speed of the grid voltage drop, thereby relieving the problem of the grid voltage drop and improving the adaptability of the grid-connected inverter to the grid voltage drop.

The application provides a method and device for identifying a shell of an Android application and a storage medium. The method comprises the following steps: obtaining a training Android application, and obtaining a dex file after decompression; counting the number of occurrences of each string in the dex file, and calculating the frequency of character occurrence based on the number of occurrences of the string; obtaining the string with the highest difference in frequency of character occurrence as feature data according to the frequency of character occurrence of the shell Android application and the frequency of character occurrence of the unshell Android application; training a shell identification model through the feature data; obtaining a to-be-identified Android application, obtaining a to-be-identified dex file after decompression, and identifying the to-be-identified dex file through the trained shell identification model, so as to obtain an identification result. The application provides a method for quickly determining whether an Android application is shelled, and uses a machine learning model to identify the shell, thereby further improving the efficiency and accuracy of identification.