A method for optimal placement of sensors based on deep reinforcement learning

A reinforcement learning and sensor technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as high-dimensional, non-convex objective functions, and achieve the effect of reducing time-consuming and improving training efficiency

Active Publication Date: 2020-10-16
哈尔滨工业大学人工智能研究院有限公司
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problem of sensor arrangement in the prior art
In the sensor arrangement of the actual engineering structure, the method of the present invention can effectively solve the problem of non-convex and high-dimensional objective function when the sensor arrangement of the complex engineering structure is arranged, and the output result of the method can realize {0, 1} discrete representation (0 Indicates that no sensor is arranged, 1 indicates that a sensor is arranged), so as to clearly provide decision support for whether a certain position sensor is arranged

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A method for optimal placement of sensors based on deep reinforcement learning
  • A method for optimal placement of sensors based on deep reinforcement learning
  • A method for optimal placement of sensors based on deep reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0093] combine Figure 4 , optimize the arrangement of sensors for the eight-story frame structure, and use the optimization algorithm to find two optimal positions for sensor arrangement. The mass matrix M and stiffness matrix K information of the engineering structure are as follows:

[0094]

[0095]

[0096] Next, use the optimization algorithm based on deep reinforcement learning in the present invention to find the optimal arrangement position of the sensor:

[0097] Said step one is specifically: based on the design information of the engineering structure, establishing relevant structural parameters (including the structural natural frequency ω 0 , Rayleigh damping coefficients α and β, amplitude and frequency a of the external force on the engineering structure 0 and the prior probability distribution of ω):

[0098] ω 0 ~lnN(·|μ=2π,σ=0.25)

[0099] α~lnN(|μ=0.1,σ=0.01)

[0100] β~lnN(|μ=10 -4 ,σ=10 -5 )

[0101] a 0 ~N(|μ=0,σ=0.4g)

[0102] ω~lnN(·|μ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The present invention proposes a sensor optimization layout method based on deep reinforcement learning, which relates to the fields of structural health monitoring and vibration testing. Efficiently find the optimal solution of the objective function, that is, the optimal sensor arrangement position of the structure. The present invention can more quickly and effectively find the optimal arrangement of sensors for engineering structures, and the optimization algorithm based on deep reinforcement learning can simultaneously utilize the computing power of deep neural networks and the decision-making ability of reinforcement learning, and can effectively solve the problem of complex engineering structure sensor layout. The objective function is non-convex and high-dimensional. At the same time, the output of this method can realize {0, 1} discrete representation, where 0 means no sensor is arranged, and 1 means that the sensor is arranged, so as to clearly provide a decision for whether to arrange the sensor at a certain position support.

Description

technical field [0001] The invention relates to the technical field of structural health monitoring and vibration testing, in particular to a sensor optimization arrangement method based on deep reinforcement learning. Background technique [0002] In structural health monitoring in the field of civil engineering, how to arrange sensors reasonably and effectively while meeting the requirements of economy plays a very important role in ensuring the safety of structures. With the continuous development of structural engineering, more and more complex structures have emerged, such as long-span bridges and super high-rise buildings. Most of these structures are huge in size and have many degrees of freedom. Many measuring points need to be arranged to obtain more structural dynamic information. However, as the number of measuring points increases, the number of sensors required and the supporting acquisition equipment increase accordingly, and the cost of instruments and the wo...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06N3/04G06N3/063G06N3/08
CPCG06N3/063G06N3/08G06N3/045
Inventor 黄永李惠孟元旭
Owner 哈尔滨工业大学人工智能研究院有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap