Assembly and method for predicting the remaining service life of a machine

A machine and life-span technology, applied in reasoning methods, neural learning methods, instruments, etc., to achieve the effect of strong predictive power

Pending Publication Date: 2021-03-16
SIEMENS AG
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Neural network outputs weights for remaining useful life predictions

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
  • Assembly and method for predicting the remaining service life of a machine
  • Assembly and method for predicting the remaining service life of a machine

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] figure 1 A flowchart for predicting the remaining useful life of a machine is shown. First, the sensor detects sensor data of the operation of the machine. Sensor measurements of vibrations are of particular importance in many cases. A machine is for example an engine in a production facility. The sensor data are collected decentralized in the peripheral units and forwarded to the memory-programmable control unit.

[0022] For monitoring the engine, for example two to three sensors are placed, such as vibration sensors and acceleration sensors, which can be supplemented, if necessary, with temperature sensors and strain gauges. The sensor is preferably arranged on the drive side on the bearing shield, as close as possible to the bearing or the shaft. The measuring direction of the vibration sensor is preferably oriented transversely to the axis.

[0023] exist figure 1 A state detection module, not shown in more detail, derives state data A from the sensor data, f...

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

On the basis of condition data (A) of a machine, a plurality of basic simulations (B) independent of one another are carried out, which determine respective remaining service life predictions (P1, P2,P3) for the machine. The remaining service life predictions and characteristic data (D), which describe properties of the machines, are fed to a neural network (E). The neural network outputs weights(F) for the remaining service life predictions. A final prediction (G) is calculated from the remaining service life predictions with computer assistance, by weighting the remaining service life predictions relative to one another in accordance with the weights. A hybrid model is thus produced, which results from the combination of the basic simulations with the neural network. Unlike before, theremaining service life can be predicted not only for a small number of machines for which a specific simulation model has been manually created. Instead, the hybrid model also enables condition monitoring for any further types and configurations of machines that merely belong to the same machine class. The basic simulations can therefore also be applied to previously unknown machines. It is no longer necessary to spend time manually creating an FEM simulation and possibly a CAD model. The neural network decides, in a data-driven manner, how strongly each of the basic simulations should be used as machine experts.

Description

Background technique [0001] For predicting the remaining useful life of machines in industrial applications (in English "remaining usefullife"), known from the field of condition monitoring in general (in English "Condition Monitoring") and from in particular the so-called structural Ways and means of Structural Health Monitoring. By monitoring the mechanical load of a machine, the expected remaining service life of the machine can be predicted, whereby repair times and downtimes of the machine can be minimized. The service life of the machine is increased and maintenance intervals can be better planned. [0002] In order to predict the remaining service life, with the help of the finite element method (FEM (Finite-Elemente-Method)-Simulation (simulation)) of the machine, experts in the corresponding field usually create a CAD model as well as a structural mechanics model. Subsequently, state data of the machine are determined, which are obtained, for example, from sensor da...

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
IPC IPC(8): G05B23/02
CPCG05B23/0283G05B2219/32234G05B23/0243G06N3/08G06N5/04G06N3/042
Inventor C·贝格斯M·希尔德布兰特M·卡利尔S·莫戈雷亚努S·什亚姆孙德
Owner SIEMENS AG
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