6results about How to "Improve global convergence" patented technology

The invention discloses a tumor postoperative treatment staged planning method based on artificial intelligence. The method comprises the following steps: S1, constructing a multi-modal medical data set; s2, performing standardization processing on the acquired multi-modal medical data set to form a standardized data set which can be used for feature extraction; s3, generating a unified patient feature vector; s4, according to the characteristics of tumor postoperative treatment, dividing the whole treatment process into a postoperative immediate recovery stage, a postoperative recovery middle stage and a long-term follow-up visit stage, and defining a corresponding optimization target for each stage; s5, based on the patient feature vector and the optimization target of each stage, generating an initial treatment plan by using a deep ant colony algorithm; and S6, outputting a staged treatment scheme. According to the method, the professional and technical thresholds required by traditional modeling are reduced, and the universality and convenience of three-dimensional modeling are improved.

The invention discloses an AI-based cloud platform network security threat intelligent detection system, which belongs to the technical field of cloud platform security, detects cloud platform network security threats by introducing an AI model so as to realize high-speed detection, and compared with the prior art, adopts a spatial multi-point interaction algorithm to train the AI model so as to realize high-speed detection. The global convergence capability of the algorithm can be effectively improved, and the convergence capability of the algorithm can also be effectively improved, so that the detection accuracy of the network security threats of the cloud platform is improved, and the security protection capability of the cloud platform is effectively improved.

The present application provides a kind of physical constraint converter steelmaking multi-physical field soft measurement method and system of teacher-student network, it is related to industrial process control technical field.First, build the mechanism model of converter steelmaking thermal fluid dynamics including parameterized dynamic source term, including thermal fluid dynamics control equation, parameterized dynamic source term;Teacher network is constructed and trained to obtain fluid dynamics benchmark;Student network is then constructed, student network executes selective knowledge distillation mechanism to carry out fluid dynamics calculation and thermochemical reaction dynamics inversion;The student network is parallel with teacher network, and the velocity field and pressure field of knowledge distillation teacher network are distilled, and the final physical output of student network forward propagation is expanded to five-dimensional physical quantity;Finally, unknown thermal fluid reaction source term and student network weight are jointly optimized and parameter inversion.The method greatly improves the self-adaptive ability of model when facing the highly fluctuating reaction kinetics in actual industry.

The invention discloses a probe layout optimization method, device and system for an air interface test and a storage medium, and the method comprises the steps: training a preset training model according to a reference test data set obtained in advance, and obtaining a reference model; obtaining antenna characteristic data and probe characteristic data of the current detected equipment; inputting the antenna feature data and the probe feature data into a reference model to obtain target probe layout information; performing instruction conversion according to the target probe layout information to obtain an axial control instruction; and sending an axial control instruction to a mechanical movement mechanism, and driving at least one test probe to move to a corresponding target position in the probe layout information. And by using the physical information neural network, the calculation efficiency and global convergence of probe position optimization are improved. And the condition number of the channel matrix is reduced from the physical level, the ill-conditioned matrix problem is effectively improved, and the stability and the measurement precision of wireless cable construction are improved. The probe position is changed by driving a mechanical mechanism, and the testing efficiency is improved.

This application discloses an automated vehicle window ventilation and temperature regulation method and system based on fuzzy control, belonging to the field of automated vehicle window control technology. The method includes: real-time acquisition of in-vehicle and out-of-vehicle environmental data, preprocessing to obtain preprocessed in-vehicle and out-of-vehicle environmental data; inputting the preprocessed in-vehicle and out-of-vehicle environmental data into a fuzzy control system to construct a fuzzy rule base for the fuzzy control system; optimizing the fuzzy control system based on the Zebra optimization algorithm and the preprocessed in-vehicle and out-of-vehicle environmental data to obtain an optimized fuzzy control system; and dynamically adjusting the opening and closing degree of the vehicle window through a window motor actuator using the optimized fuzzy control system based on subsequently obtained in-vehicle and out-of-vehicle environmental data. This application, based on in-vehicle and out-of-vehicle environmental data and combined with a fuzzy control system for optimized control, overcomes the shortcomings of existing technologies in complex environment adaptability, multi-objective optimization capability, and global convergence performance, achieving efficient and precise vehicle window ventilation and temperature regulation.

The invention discloses a while-drilling stratum electromagnetic imaging method and device. The method comprises the steps that an inversion stratum model of while-drilling stratum electromagnetic imaging is determined; constructing an objective function with a regularization item for the inversion stratigraphic model, and calculating the gradient of the objective function with the regularization item to obtain a gradient function with the regularization item; utilizing a pre-gradient matrix obtained by pre-training of a gradient function, iteratively updating stratum parameters of the inversion stratum model according to the number of inversion iterations and the pre-gradient matrix, determining stratum parameters for constructing an inversion graph of the current stratum based on a target function, and obtaining the inversion graph of the current stratum; the pre-gradient matrix corresponds to the number of inversion iterations. The inversion precision and the global convergence capability are improved by combining the inversion with the stratum model of machine learning. The pre-gradient matrix obtained through pre-training is directly applied to inversion, the inversion speed and real-time performance are remarkably improved, and the inversion accuracy of complex stratum parameters is improved.