知搜-全球专利检索

  1. 登录
  2. 注册

搜索结果

389 条记录 (耗时 0.02 秒)

    Anomaly detection method and apparatus for multi-type data
    4.
    发明授权
    Anomaly detection method and apparatus for multi-type data 有权

    公开(公告)号：US11423260B1

    公开(公告)日：2022-08-23

    申请号：US17589888

    申请日：2022-01-31

    申请人： Harbin Institute of Technology (Shenzhen) (Shenzhen Institute of Science and Technology Innovation, Harbin Institute of Technology) ,  Dongguan University of Technology

    发明人： Qing Liao ,  Binxing Fang ,  Ye Ding

    IPC分类号： G06K9/62

    摘要： The present disclosure provides an anomaly detection method and apparatus for multi-type data. According to the anomaly detection method for multi-type data, an adversarial learning network is trained, so that a generator in the adversarial learning network fits a distribution of a normal training sample and learns a potential mode of the normal training sample, to obtain an updated adversarial learning network, an anomaly evaluation function in the updated adversarial learning network is constructed according to a reconstruction error generated during training, and the updated adversarial learning network is constructed into an anomaly detection model, to perform anomaly detection on inputted detection data by the anomaly detection model, to obtain an anomaly detection result. A mode classifier is introduced to effectively resolve difficult anomaly detection when a distribution of detected data is similar to that of normal data, further improving the accuracy of anomaly detection.

    Learning control system and method for nano-precision motion stage
    8.
    发明授权
    Learning control system and method for nano-precision motion stage 有权

    公开(公告)号：US12124228B1

    公开(公告)日：2024-10-22

    申请号：US18731174

    申请日：2024-05-31

    申请人： Harbin Institute of Technology

    发明人： Fazhi Song ,  Kaixin Liu ,  Shuaiqi Chen ,  Yang Liu ,  Jiubin Tan

    IPC分类号： G05B13/02 ,  G05B19/19

    CPC分类号： G05B13/0265 ,  G05B19/19 ,  G05B2219/34048 ,  G05B2219/41439 ,  G05B2219/41453

    摘要： A learning control system for a nano-precision motion stage comprises a closed-loop feedback section including a motion trajectory generator, a feedback controller, a motion stage, and a first Fourier transformer; and a feedforward section including a second Fourier transformer, a learning controller, an iteration backward shift operator, and a Fourier inverse transformer. An iteration experiment count j is initialized as j=1, and a j-th frequency domain feedforward signal is initialized to 0; the system is run to collect a frequency domain error signal and a frequency domain position measurement signal; a (j+1)-th frequency domain feedforward signal is updated; and an iteration experiment count j is incremented by 1. The present disclosure can effectively suppress the influence of external noise and disturbances, and improve convergence performance. Moreover, the present disclosure requires less computation, achieves simple determination of learning gains and strong robustness, and is convenient for engineering applications.

    USER-DISTINGUISHED FINITE-FIELD RESOURCE CONSTRUCTION METHOD AND FINITE-FIELD MULTIPLE ACCESS SYSTEM
    9.
    发明公开
    USER-DISTINGUISHED FINITE-FIELD RESOURCE CONSTRUCTION METHOD AND FINITE-FIELD MULTIPLE ACCESS SYSTEM 审中-公开

    公开(公告)号：US20240322931A1

    公开(公告)日：2024-09-26

    申请号：US18612803

    申请日：2024-03-21

    申请人： HARBIN INSTITUTE OF TECHNOLOGY

    发明人： Qiyue YU ,  Jiangxuan LI ,  Shu LIN

    IPC分类号： H04J13/12 ,  H04J13/16

    CPC分类号： H04J13/12 ,  H04J13/16

    摘要： The present disclosure relates to the field of communication technologies and in particular to a user-distinguished finite-field resource construction method and a finite-field multiple access system. In order to solve the problem of the limitation of the multiple access resource in the current communication field, the present disclosure employs a user-distinguished finite-field resource construction method to construct a basic-field resource and/or extension-field resource, i.e. finite-field resource. During the use of the finite-field resource, each user sending a binary sequence is assigned one codebook marking symbols that 0 and 1 are respectively mapped into a finite field. The transmitter sends a corresponding finite-field symbol sequence. At the receiver, based on the received finite-field symbols, a finite-field symbol sent by each user can be determined uniquely and thus, a binary symbol sent by each user can be decoded. The present disclosure is applied to the finite-field multiple access system.

    Method of Determining River Nitrous Oxide Emission based on Land-River-Atmosphere Simulation
    10.
    发明公开
    Method of Determining River Nitrous Oxide Emission based on Land-River-Atmosphere Simulation 审中-公开

    公开(公告)号：US20240321403A1

    公开(公告)日：2024-09-26

    申请号：US18605776

    申请日：2024-03-14

    申请人： Harbin Institute of Technology

    发明人： Yu TIAN ,  Huihang SUN ,  Haoran ZHANG ,  Yiming MENG ,  Nanqi REN

    IPC分类号： G16C20/20 ,  G06Q10/04

    CPC分类号： G16C20/20 ,  G06Q10/04

    摘要： A method of determining nitrous oxide emission of a river based on land-river-atmosphere simulation, includes the steps of: obtaining nitrogen emission from land in each region; dividing the nitrogen emission into a prediction set and a test set; using nitrogen emission prediction set, and geographical variables and climate variables under the nitrogen emission prediction set to process RF regression model training to obtain a trained RF regression model, using nitrogen emission test set, and geographical variables and climate variables under the nitrogen emission test set to process RF regression model training to obtain a trained RF regression model, and outputting a river water quality concentration of each sub-basin in each region; obtaining river hydrological parameters of each sub-basin, inputting the river hydrological parameters and river water quality concentration of each sub-basin to an air-water interface gas exchange model to obtain a total river N2O emission in each sub-basin.