English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The article, taking neural network (NN) as the tool and electric actuator as study object, proposes a diagnosis method for electric actuator based on self-organization competitive neural network, by which state diagnosis is realized with discrimination capacity of nonlinear dynamic system of neural network as well as comparison of forecast value of the system and measured values of actual parameters. The system incorporated VB6.0 as development tool and SQL Server2000 as backstage database to realize intelligent diagnosis of states with electric actuator. Keywords—State Diagnosis; Electric Actuator; Artificial Intelligence (AI); Neural Network (NN) ISignal Acquisition System Based on Electric Actuator In traditional way, certain technical performances of electric actuator are analyzed unexceptionally through directly manual performance test of electric actuator; today, at some industrial workspace where electric actuator is used, most tests are performed by technical personnel based on their experience in combination with theoretical knowledge, which are time-consuming and undesirable in test effects. With increasing application of computer, it’s necessary to analyze and test these major performance indicators based on computer aid. The system acquires position feedback current signal of electric actuator with Keithley2000 digital multi-meter produced by American Keithley, inputs data to the computer via RS-232 serial interfaces of industrial personal computer, realizes acquisition of signal through VB 6.0 MSComm serial communication control programming, then draws curve of acquired signals by MatrixVB for VB software developed by MathWorks for fitting analysis, and then derives performance parameters reflecting system state, so as to realize effective diagnosis. Figure 1 is hard wiring diagram of data set, multi-circuit relay switch is connected to industrial personal computer via one ISA card, and then corresponding demand is sent from industrial personal computer to control opening/closing at relay point to realize function of controlling connection/disconnection to current signal source. In order to reduce the interference of additional position feedback at the moment of connection/disconnection to signal source, we’ve improved circuit connection, i.e., signal at input terminal of servo amplifier from current signal source is not open circuit when disconnecting to signal source, rather, relay switch is closed to cause short circuit, so as to reduce interference by external additional signals on the circuit. Fig. 1Wiring Diagram for Acquisition of Electric Actuator Signals IIIntelligent Diagnosis Based on Neural Network A. Network Learning Rules Self-organization competitive neural network selected by us is a kind of neural network without teaching of physical teacher and with self-organization function, which, by self-training, can automatically classify input mode and identify current mode from those of detected modes, make most close classification, that is, express classification result by winning competitive neuron nodes. Fig.2 Structure of Basic Competitive Neural Network RS-232

A Method for Diagnosing the State of Electric Actuator Based on Neural Network