Early warning method and system for discharge voltage fault of ignition electric nozzle of engine

Abstract

The invention discloses an engine ignition electric nozzle discharge voltage fault early warning method and system, and the method comprises the steps: obtaining the sampling data of the discharge of an ignition electric nozzle, and the sampling data comprise a voltage signal and a current signal; de-noising processing is carried out on the sampled data according to a set filtering rule, and the set filtering rule comprises the steps that low-pass filtering or band-pass filtering processing is carried out on the sampled data through a multiply-accumulate operation mode by adopting a set transfer function; analyzing and processing the de-noised data to obtain each spark performance parameter, wherein each spark performance parameter comprises spark energy, ignition voltage and ignition frequency; and each spark performance parameter carries out fault early warning on the ignition electric nozzle through the fault prediction model. The discharge voltage signal and current signal of the ignition electric nozzle are analyzed and processed to obtain each spark performance parameter, fault early warning of the ignition electric nozzle is carried out according to each spark performance parameter, and the problems of each spark performance test and fault early warning of the ignition electric nozzle are solved.

Description

Technical Field

[0001] The invention belongs to the technical field of aviation engine fault detection, and in particular relates to an engine ignition nozzle discharge voltage fault early warning method and system. Background Art

[0002] In the field of engines, the ignition nozzle is an important part of the engine ignition system and an accessory used to generate electric sparks. The performance of the ignition nozzle directly affects whether the engine ignition system can ignite successfully. The discharge voltage of the ignition nozzle is an important parameter to test its performance. The discharge voltage of the ignition nozzle refers to the minimum input voltage that breaks down the ignition nozzle to generate a discharge spark. At present, the detection device of the discharge voltage of the ignition nozzle of the aircraft engine is a simple voltage detection, which cannot test the performance of various sparks and analyze faults. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for early warning of discharge voltage fault of an engine ignition nozzle, which solves the problem of testing the spark performance of the ignition nozzle and early warning of faults.

[0004] The present invention is achieved through the following technical solutions:

[0005] An engine ignition nozzle discharge voltage fault early warning method, comprising:

[0006] Acquire sampling data of ignition nozzle discharge, the sampling data including voltage signal and current signal;

[0007] The sampled data is subjected to denoising according to a set filtering rule, wherein the set filtering rule includes: using a set transfer function to perform low-pass filtering or band-pass filtering on the sampled data through a multiplication-accumulation operation method;

[0008] Analyze and process the denoised data to obtain various spark performance parameters, including spark energy, ignition voltage, and ignition frequency;

[0009] Each spark performance parameter is used to provide ignition nozzle fault warning through a fault prediction model.

[0010] In some embodiments, when low-pass filtering is performed, the transfer function used includes:

[0011] y(k)=ay(k-1)+bx(k)-bx(kN).

[0012] In some embodiments, when bandpass filtering is performed, the transfer function used includes:

[0013] y(k)=ay(k-1)+bx(k)+bx(kN).

[0014] In some embodiments, the training of the fault prediction model includes:

[0015] A large number of spark performance parameters are fed into the machine learning model;

[0016] The first loss function is constructed using the spark performance data and historical fault data output by the machine learning model;

[0017] The first loss function iteratively updates the parameters of the initial machine learning model until the preset conditions are met to obtain a fault prediction model.

[0018] In some embodiments, the warning mechanism of the fault warning includes an audio signal, a visual signal and a control panel warning.

[0019] The present invention also relates to an engine ignition nozzle discharge voltage fault warning system, comprising:

[0020] The sampling data acquisition module is used to obtain the sampling data of the ignition nozzle discharge, and the sampling data includes voltage signal and current signal;

[0021] The data processing module is used to perform denoising on the sampled data according to the set filtering rules;

[0022] A data sending module, used for sending the denoised sampled data to the calculation control module through a bus;

[0023] A calculation control module, used for analyzing and processing the denoised sampled data to obtain various spark performance parameters, by denoising the sampled data according to a set filtering rule, and analyzing and processing the denoised data to obtain various spark performance parameters, wherein the set filtering rule includes: using a set transfer function, performing low-pass filtering or band-pass filtering on the sampled data by a multiplication-accumulation operation mode, and the various spark performance parameters include spark energy, ignition voltage, and ignition frequency;

[0024] The fault warning module is used to provide fault warning of the ignition nozzle through the fault prediction model of various spark performance parameters.

[0025] In some embodiments, the sampling data acquisition module includes a photoelectric isolation probe.

[0026] In some embodiments, the sampling data acquisition module further includes a Rogowski current measurement coil.

[0027] In some embodiments, the data sending module uses a multi-channel data acquisition card.

[0028] In some embodiments, the computing control module adopts a PLC module.

[0029] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0030] 1. Obtain the voltage signal and current signal of the ignition nozzle discharge, and analyze and process them to obtain various spark performance parameters. Each spark performance parameter is used to provide fault warning for the ignition nozzle through a fault prediction model, ensuring the performance of the ignition nozzle and improving the safety of the ignition nozzle.

[0031] 2. Use the set filtering rules to perform denoising, enhance the effective components in the signal, reduce the errors caused by environmental factors, and improve the accuracy of fault warning.

[0032] 3. Use various forms of early warning mechanisms to provide fault warning prompts and multi-faceted fault warnings to speed up the staff's discovery of fault characteristics, reduce the failure rate of the ignition nozzle, reduce maintenance costs, and improve safety.

[0033] 4. Using photoelectric isolation probes to collect voltage signals can greatly reduce coupling interference and improve the accuracy of voltage signals.

[0034] 5. The Rogowski current measurement coil is used to collect current signals without direct contact with the circuit being measured, which greatly reduces the electromagnetic interference of the circuit and improves the accuracy of the current signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 The figure is a flow chart of the method for early warning of discharge voltage failure of an engine ignition nozzle in an embodiment of the present invention.

[0037] Figure 2 Schematic diagram of an engine ignition nozzle discharge voltage fault warning system according to an embodiment of the present invention DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0039] Example 1

[0040] Reference Figure 1 In some embodiments, a method for early warning of an engine ignition nozzle discharge voltage failure includes:

[0041] Acquire sampling data of ignition nozzle discharge, the sampling data including voltage signal and current signal;

[0042] The sampled data is subjected to denoising according to a set filtering rule, wherein the set filtering rule includes: using a set transfer function to perform low-pass filtering or band-pass filtering on the sampled data through a multiplication-accumulation operation method;

[0043] Analyze and process the denoised data to obtain various spark performance parameters, including spark energy, ignition voltage, and ignition frequency;

[0044] Each spark performance parameter is used to provide ignition nozzle fault warning through a fault prediction model.

[0045] Among them, spark energy is an important indicator to measure the ignition ability of the ignition nozzle. For example, when the spark energy generated by the ignition nozzle is lower than 0.05, it indicates that the ignition ability of the ignition nozzle is relatively weak.

[0046] Voltage requirement: Sparks can only be generated when the voltage of the ignition nozzle reaches a preset threshold. For example, when the voltage of the ignition nozzle reaches 10kV to 18kV, the gas insulation layer between the electrodes is broken down to generate the arc required for ignition.

[0047] Ignition frequency is also an important performance parameter. For example, when the spark frequency reaches 3.5Hz, it means that the ignition nozzle can generate sparks multiple times in a short period of time, which can improve the efficiency and reliability of ignition.

[0048] The voltage signal and current signal of the ignition nozzle discharge are obtained, and the spark performance parameters are obtained by analysis and processing. The spark performance parameters are used to provide fault warning for the ignition nozzle through the fault prediction model, ensuring the performance of the ignition nozzle and improving the safety of the ignition nozzle.

[0049] In some embodiments, when low-pass filtering is performed, the transfer function used includes: y(k)=

[0050] ay(k-1)+bx(k)-bx(kN).

[0051] In some embodiments, when performing bandpass filtering, the transfer function used includes: y(k)=

[0052] ay(k-1)+bx(k)+bx(kN).

[0053] The set filtering rules are used for denoising, which can enhance the effective components in the signal, reduce the errors caused by environmental factors, and improve the accuracy of fault warning.

[0054] In some embodiments, the fault prediction model comprises a machine learning model.

[0055] In some embodiments, the training of the fault prediction model includes:

[0056] A large number of spark performance parameters are fed into the machine learning model;

[0057] The first loss function is constructed using the spark performance data and historical fault data output by the machine learning model;

[0058] The first loss function iteratively updates the parameters of the initial machine learning model until the preset conditions are met to obtain a fault prediction model.

[0059] In some embodiments, the warning mechanism of the fault warning includes an audio signal, a visual signal and a control panel warning.

[0060] Audio signal: when the fault prediction model outputs that the ignition nozzle is faulty, a preset audio signal is used to give a sound warning.

[0061] Visual signal: when the fault prediction model outputs that the ignition nozzle is faulty, a preset red light flashes continuously to provide a visual warning prompt;

[0062] Control panel warning means that when the fault prediction model outputs that the ignition nozzle is faulty, the fault prompt information is displayed on the control panel to provide information warning to the staff.

[0063] Use various forms of early warning mechanisms to provide fault warning prompts and multi-faceted fault warnings to speed up the staff's discovery of fault characteristics. Through the analysis of fault characteristics, the ignition system can be adjusted and maintained in a timely manner to reduce the failure rate of the ignition nozzle, reduce maintenance costs, and improve safety.

[0064] The present invention also relates to an engine ignition nozzle discharge voltage fault warning system, referring to Figure 2 ,include:

[0065] The sampling data acquisition module is used to obtain the sampling data of the ignition nozzle discharge, and the sampling data includes voltage signal and current signal;

[0066] The data processing module is used to perform denoising on the sampled data according to the set filtering rules;

[0067] A data sending module, used for sending the denoised sampled data to the calculation control module through a bus;

[0068] A calculation control module, used for analyzing and processing the denoised sampled data to obtain various spark performance parameters, by denoising the sampled data according to a set filtering rule, and analyzing and processing the denoised data to obtain various spark performance parameters, wherein the set filtering rule includes: using a set transfer function, performing low-pass filtering or band-pass filtering on the sampled data by a multiplication-accumulation operation mode, and the various spark performance parameters include spark energy, ignition voltage, and ignition frequency;

[0069] The fault warning module is used to provide fault warning of the ignition nozzle through the fault prediction model of various spark performance parameters.

[0070] In some embodiments, the sampling data acquisition module includes a photoelectric isolation probe.

[0071] Using photoelectric isolation probes to collect voltage signals can greatly reduce coupling interference and improve the accuracy of voltage signals.

[0072] In some embodiments, the sampling data acquisition module further includes a Rogowski current measurement coil.

[0073] The Rogowski current measurement coil is used to collect current signals without direct contact with the circuit being measured, which greatly reduces the electromagnetic interference of the circuit and improves the accuracy of the current signal.

[0074] In some embodiments, the data sending module uses a multi-channel data acquisition card.

[0075] In some embodiments, the computing control module adopts a PLC module.

[0076] The working process of the engine ignition nozzle discharge voltage fault warning system is as follows:

[0077] The engine ignition nozzle discharge voltage fault warning system collects the ignition nozzle discharge voltage signal through a photoelectric isolation probe;

[0078] At the same time, the engine ignition nozzle discharge voltage fault warning system collects the ignition nozzle discharge current signal through the Rogowski current measurement coil;

[0079] The data processing module performs denoising on the collected voltage and current signals according to the set filtering rules;

[0080] The multi-channel data acquisition card sends the denoised voltage and current signals to the PLC module through the bus;

[0081] The PLC module analyzes and processes the received voltage and current signals to obtain various spark performance parameters;

[0082] Each spark performance parameter is used to provide ignition nozzle fault warning through a fault prediction model.

[0083] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification or equivalent change made to the above embodiment based on the technical essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for early warning of discharge voltage failure of an engine ignition nozzle, characterized in that: include: Acquiring sampling data of ignition nozzle discharge, wherein the sampling data includes a voltage signal and a current signal; The sampled data is subjected to denoising according to a set filtering rule, wherein the set filtering rule includes: using a set transfer function to perform low-pass filtering or band-pass filtering on the sampled data by a multiplication-accumulation operation mode; Analyzing and processing the denoised data to obtain various spark performance parameters, wherein the various spark performance parameters include spark energy, ignition voltage, and ignition frequency; Each spark performance parameter is used to provide ignition nozzle fault warning through a fault prediction model.

2. The method for early warning of discharge voltage failure of an engine ignition nozzle according to claim 1, characterized in that: When low-pass filtering is performed, the transfer function used includes: y(k)=ay(k-1)+bx(k)-bx(kN).

3. The method for early warning of discharge voltage failure of an engine ignition nozzle according to claim 1, characterized in that: When performing bandpass filtering, the transfer functions used include: y(k)=ay(k-1)+bx(k)+bx(kN).

4. The method for early warning of discharge voltage failure of an engine ignition nozzle according to claim 1, characterized in that: The training of the fault prediction model includes: A large number of spark performance parameters are fed into the machine learning model; The first loss function is constructed using the spark performance data and historical fault data output by the machine learning model; The first loss function iteratively updates the parameters of the initial machine learning model until the preset conditions are met to obtain a fault prediction model.

5. The method for early warning of discharge voltage failure of an engine ignition nozzle according to claim 1, characterized in that: The warning mechanism of the fault warning includes audio signal, visual signal and control panel warning.

6. An engine ignition nozzle discharge voltage fault warning system, characterized in that: include: A sampling data acquisition module, used to obtain sampling data of ignition nozzle discharge, wherein the sampling data includes voltage signal and current signal; The data processing module is used to perform denoising on the sampled data according to the set filtering rules; A data sending module, used for sending the denoised sampled data to the calculation control module through a bus; A calculation control module, used for analyzing and processing the denoised sampled data to obtain various spark performance parameters, by denoising the sampled data according to a set filtering rule, and analyzing and processing the denoised data to obtain various spark performance parameters, wherein the set filtering rule includes: using a set transfer function, performing low-pass filtering or band-pass filtering on the sampled data by a multiplication-accumulation operation mode, and the various spark performance parameters include spark energy, ignition voltage, and ignition frequency; The fault warning module is used to provide fault warning of the ignition nozzle through the fault prediction model of various spark performance parameters.

7. The engine ignition nozzle discharge voltage fault warning system according to claim 6, characterized in that: The sampling data acquisition module includes a photoelectric isolation probe.

8. The engine ignition nozzle discharge voltage fault warning system according to claim 6, characterized in that: The sampling data acquisition module also includes a Rogowski current measurement coil.

9. The engine ignition nozzle discharge voltage fault warning system according to claim 6, characterized in that: The data sending module adopts a multi-channel data acquisition card.

10. The engine ignition nozzle discharge voltage fault warning system according to claim 6, characterized in that: The calculation control module adopts a PLC module.

Images