公开(公告)号：US20250010958A1

公开(公告)日：2025-01-09

申请号：US18279737

申请日：2022-12-05

Applicant: DALIAN UNIVERSITY OF TECHNOLOGY ,  DONGHAI LABORATORY

Inventor： Chongwei ZHANG ,  Jindie WEI ,  Xunhao ZHU ,  Dezhi NING

IPC: B63B39/03 ,  B63B1/12 ,  B63B21/50 ,  B63B35/44 ,  F03D13/25

Abstract: An air-liquid dual control anti-rolling control system for a floating offshore wind turbine in offshore deep sea comprises an equipment compartment and three closed TLCD loop units. The equipment compartment is arranged above box girders; and each TLCD loop unit mainly forms a closed loop by a liquid tank, an air tube and a liquid tube and is embedded into the structure of a floating foundation. The system of the present invention has simple structure, easy installation, detachability, easy replacement and convenient use. The system of the present invention has universality, the designed TLCD loop units are independent of each other, and the start and stop of each TCLD loop unit is entirely coordinated and scheduled by a control module, which is easy to expand. The system of the present invention can realize intelligent autonomous control. By analyzing the measured motion parameters of the floating offshore wind turbine foundation, the control module autonomously determines the TLCD loop unit to be started according to the swing direction of the floating offshore wind turbine foundation, and determines to start an air-control module or liquid-control module of the TLCD loop unit according to the swing frequency of the floating offshore wind turbine foundation, as well as the resistance value of a slide rheostat in an air-control module or the rotational speed of motors of water-turbine sets in a liquid-control module.

公开(公告)号：US12044799B2

公开(公告)日：2024-07-23

申请号：US18451818

申请日：2023-08-17

Applicant: ZHEJIANG UNIVERSITY ,  DONGHAI LABORATORY

Inventor： Chunyi Song ,  Zhihui Cao ,  Zhiwei Xu ,  Yuying Song ,  Fuyuan Ai ,  Jingxuan Wu

IPC: G01S7/41 ,  G01S13/53 ,  G01S13/536 ,  G01S13/72 ,  G06N3/084

CPC classification number: G01S7/417 ,  G01S13/536 ,  G01S13/726 ,  G06N3/084

Abstract: The embodiment of the present disclosure provides a deep neural network (DNN)-based multi-target constant false alarm rate (CFAR) detection method. The method includes: obtaining target values to be measured based on radar IF (IF) signals to be detected, the target values to be measured including a measured frequency value and a measured intensity value of the radar IF signals; obtaining peak sequences based on the target values to be measured; generating a target detection result by processing the peak sequences based on a DNN detector, the DNN detector being a machine learning model; generating approximated maximum likelihood estimation (AMLE) of a scale parameter based on an approximated maximum likelihood estimator; generating a false alarm adjustment threshold based on a preset false alarm rate and the AMLE; and generating a constant false alarm detection result by processing the target detection result based on the false alarm adjustment threshold.

公开(公告)号：US20240355105A1

公开(公告)日：2024-10-24

申请号：US18762595

申请日：2024-07-02

Applicant: DONGHAI LABORATORY ,  ZHEJIANG UNIVERSITY

Inventor： Chunyi SONG ,  Fuyuan AI ,  Hussain AMJAD ,  Zecheng LI ,  Yuying SONG ,  Zhiwei XU

IPC: G06V10/80 ,  G01J5/00 ,  G01J5/08 ,  G06T7/20 ,  G06T7/55 ,  G06V10/143 ,  G06V10/774 ,  G06V10/82 ,  G06V20/58

CPC classification number: G06V10/806 ,  G01J5/0859 ,  G06T7/20 ,  G06T7/55 ,  G06V10/143 ,  G06V10/774 ,  G06V10/82 ,  G06V20/58 ,  G01J2005/0077 ,  G06T2207/10028 ,  G06T2207/10048 ,  G06T2207/20016 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30261 ,  G06V2201/07

Abstract: A method for target detection based on a visible camera, an infrared camera, and a LiDAR is provided. The method designates visible light images, infrared images, and LiDAR point clouds, which are synchronously acquired, as inputs, and generates an input pseudo-point cloud using visible light images and infrared images, to realize alignment of multimodal information in a three-dimensional space and fusion feature extraction. Then the method adopts a cascade strategy to output more accurate target detection results step by step. In the present disclosure, different characteristics of multi-sensors are complemented, which improves and extends traditional target detection algorithms, improves the accuracy and robustness of target detection, and realizes multi-category target detection in a road scene.

公开(公告)号：US20240004032A1

公开(公告)日：2024-01-04

申请号：US18451818

申请日：2023-08-17

Applicant: ZHEJIANG UNIVERSITY ,  DONGHAI LABORATORY

Inventor： Chunyi SONG ,  Zhihui CAO ,  Zhiwei XU ,  Yuying SONG ,  Fuyuan AI ,  Jingxuan WU

IPC: G01S7/41 ,  G01S13/72 ,  G01S13/536 ,  G06N3/084

CPC classification number: G01S7/417 ,  G01S13/726 ,  G01S13/536 ,  G06N3/084

Abstract: The embodiment of the present disclosure provides a deep neural network (DNN)-based multi-target constant false alarm rate (CFAR) detection method. The method includes: obtaining target values to be measured based on radar IF (IF) signals to be detected, the target values to be measured including a measured frequency value and a measured intensity value of the radar IF signals; obtaining peak sequences based on the target values to be measured; generating a target detection result by processing the peak sequences based on a DNN detector, the DNN detector being a machine learning model; generating approximated maximum likelihood estimation (AMLE) of a scale parameter based on an approximated maximum likelihood estimator; generating a false alarm adjustment threshold based on a preset false alarm rate and the AMLE; and generating a constant false alarm detection result by processing the target detection result based on the false alarm adjustment threshold.