CN111177975A - Aviation equipment availability prediction method based on artificial intelligence - Google Patents

Aviation equipment availability prediction method based on artificial intelligence Download PDF

Info

Publication number
CN111177975A
CN111177975A CN201911353973.2A CN201911353973A CN111177975A CN 111177975 A CN111177975 A CN 111177975A CN 201911353973 A CN201911353973 A CN 201911353973A CN 111177975 A CN111177975 A CN 111177975A
Authority
CN
China
Prior art keywords
neural network
layer
artificial neural
network model
training sample
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201911353973.2A
Other languages
Chinese (zh)
Inventor
朱威仁
樊西龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xian Aircraft Design and Research Institute of AVIC
Original Assignee
Xian Aircraft Design and Research Institute of AVIC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xian Aircraft Design and Research Institute of AVIC filed Critical Xian Aircraft Design and Research Institute of AVIC
Priority to CN201911353973.2A priority Critical patent/CN111177975A/en
Publication of CN111177975A publication Critical patent/CN111177975A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Testing And Monitoring For Control Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention belongs to the aviation maintenance support technology, and relates to an aviation equipment availability prediction method based on artificial intelligence. The method comprises the following steps: acquiring various input data of the aviation equipment to be tested, wherein the input data comprises equipment mean fault interval time, spare part turnover time, mean repair time, regular inspection period and overhaul period; inputting various input data of the aviation equipment to be tested into the three-layer artificial neural network model to obtain the availability index of the aviation equipment; the three-layer artificial neural network model is trained according to training sample data of the conventional aviation equipment.

Description

Aviation equipment availability prediction method based on artificial intelligence
Technical Field
The invention belongs to the aviation maintenance support technology, and relates to an aviation equipment availability prediction method based on artificial intelligence.
Background
In the process of developing aviation equipment, the availability index of the equipment needs to be subjected to predictive analysis so as to develop design and development in a targeted manner, wherein the availability index of the equipment represents the probability that the equipment is in an intact and available state at any moment. However, the traditional availability calculation model is rough, the needed parameters are mostly statistical data in equipment use, and in the early development stage of novel aviation equipment, a corresponding equipment availability prediction method is lacked to predict equipment availability and optimize equipment design.
Disclosure of Invention
The purpose of the invention is as follows: the aviation equipment availability prediction method based on artificial intelligence is provided, and prediction analysis of equipment availability indexes is achieved.
The invention provides an aviation equipment availability prediction method based on artificial intelligence, which comprises the following steps:
acquiring various input data of the aviation equipment to be tested, wherein the input data comprises equipment mean fault interval time, spare part turnover time, mean repair time, regular inspection period and overhaul period;
inputting various input data of the aviation equipment to be tested into the three-layer artificial neural network model to obtain the availability index of the aviation equipment; the three-layer artificial neural network model is trained according to training sample data of the conventional aviation equipment.
Further, the artificial neural network model comprises an input layer, a hidden layer and an output layer, wherein the number of input nodes of the input layer is 5, and the number of nodes of the hidden layer is 3.
Further, the method further comprises:
collecting and counting average fault interval time, spare part turnover time, average repair time, regular inspection period and overhaul period of the conventional aviation equipment as training sample data, and using availability data as a training result sample;
training an initial artificial neural network model through training sample data to train an available artificial neural network.
Further, training the initial artificial neural network model by training sample data to train an available artificial neural network, including:
dividing training sample data and corresponding training sample results into n groups;
correcting the initial artificial neural network model according to each set of training sample data and corresponding training sample results in sequence by taking a group as a unit to obtain a first weight correction factor from a hidden layer to an output layer and a second weight correction factor from an input layer to the hidden layer;
according to the weight, the first weight correction factor and the second weight correction factor, correcting the parameters of the corresponding layer of the artificial neural network model; wherein the weight is composed of a group number and a group number used for training.
The invention provides an aviation equipment availability prediction device based on artificial intelligence, which comprises:
the acquisition module is used for acquiring various input data of the aviation equipment to be detected, wherein the input data comprises equipment average fault interval time, spare part turnover time, average repair time, regular inspection period and overhaul period;
the input module is used for inputting various input data of the aviation equipment to be tested into the three-layer artificial neural network model to obtain the availability index of the aviation equipment; the three-layer artificial neural network model is trained according to training sample data of the conventional aviation equipment.
Further, the artificial neural network model comprises an input layer, a hidden layer and an output layer, wherein the number of input nodes of the input layer is 5, and the number of nodes of the hidden layer is 3.
Further, the apparatus further comprises:
the collection module is used for collecting and counting average fault interval time, spare part turnover time, average repair time, regular inspection period and overhaul period of the conventional aviation equipment as training sample data, and using availability data as training result samples;
and the training module is used for training the initial artificial neural network model through training sample data to train an available artificial neural network.
Further, the training module comprises:
the grouping unit is used for dividing the training sample data and the corresponding training sample results into n groups;
the correction unit is used for correcting the initial artificial neural network model according to each group of training sample data and the corresponding training sample result in sequence by taking a group as a unit to obtain a first weight correction factor from the hidden layer to the output layer and a second weight correction factor from the input layer to the hidden layer; according to the weight, the first weight correction factor and the second weight correction factor, correcting the parameters of the corresponding layer of the artificial neural network model; the weight is composed of a group number and a group number used for training.
The invention has the advantages that: the method can accurately predict and analyze the availability index of the equipment in the early stage of development of novel aviation equipment, and supports development of equipment index demonstration and design optimization.
Drawings
FIG. 1 is a flow chart of an artificial intelligence based aviation equipment availability prediction method.
FIG. 2 is a schematic diagram of a three-layer artificial neural network model.
Detailed description of the invention
Artificial neural networks are artificial intelligence techniques developed in recent years that simulate biological processes of the nervous system. The method is characterized in that a complex nonlinear system is formed by interconnection of a large number of simple-structured neurons, and the nonlinear mapping capability is strong. The availability of the aviation equipment is a nonlinear model, so that the artificial neural network has a wide application prospect in the field of equipment availability prediction. The input signal and the expected output can be continuously provided to the network in the training process of the neural network, and the network is considered to complete the learning when the neural network can generate the actual output which is similar to the expected output for various input signals provided by learning the statistical data of the previous model equipment, so that the specific work task can be executed, and the usability prediction of the model equipment in research can be carried out. The neural network algorithm has the advantages of no need of complex nonlinear modeling work, high calculation speed and strong adaptability and stability.
The invention provides an aviation equipment availability prediction method based on artificial intelligence, which comprises the following steps of:
the method comprises the following steps: and establishing a three-layer artificial neural network model as shown in figure 2. Wherein xi(i 1, 2.. n.) is the input to the network, yj(j ═ 1, 2.. times, p) is the output of the hidden layer, and o is the output of the output layer, which is equipment availability. v. ofijRepresenting the weight between the ith node of the input layer and the jth node of the hidden layer, wjRepresenting the jth node and output layer of the hidden layerWeight between nodes. ThetajRepresents the threshold of the jth neuron node of the hidden layer, and gamma represents the threshold of the output node.
This patent is selected and is equipped mean fault interval time, spare part turnover time, mean repair time, regular maintenance cycle, overhaul period totally 5 parameters as the input, sets for input number n promptly and equals 5, sets for implicit layer node number p and equals 3.
Step two: collecting and counting average fault interval time, spare part turnover time, average repair time, regular inspection period, overhaul period and availability data of past aviation equipment as training sample data.
initializing each weight value of the neural network as a random number, setting a learning rate η as a decimal between 0 and 1, and determining a threshold theta of each neuron node of the hidden layerj(j ═ 1, 2,. p), an output layer node threshold γ is set.
Step four: inputting training sample data of past equipment, wherein d represents the availability value of the past equipment. And calculating the output of each layer of the neural network according to a formula.
And calculating the output of each node of the hidden layer:
Figure RE-GDA0002419425860000041
calculating output layer node output:
Figure RE-GDA0002419425860000042
calculating a weight correction factor from a hidden layer to an output layer:
Δwf＝η(d-o)o(1-α)yi；
calculating a weight correction factor from an input layer to a hidden layer:
Δvij＝η(d-o)o(1-o)wfyj(1-yj)xi；
step five: and updating all weights according to the weight correction factors:
Figure RE-GDA0002419425860000043
k is the group number of the samples, and K is the total group number of the samples;
Figure RE-GDA0002419425860000044
step six: and (4) checking whether all training samples are trained, if so, executing the step seven, and if not, selecting untrained samples and returning to the step four.
Step seven: inputting various input data of the aviation equipment needing availability prediction, and calculating through a neural network to obtain an output value, namely the predicted availability value of the aviation equipment.

Claims (8)

1. An aviation equipment availability prediction method based on artificial intelligence is characterized by comprising the following steps:
acquiring various input data of the aviation equipment to be tested, wherein the input data comprises equipment mean fault interval time, spare part turnover time, mean repair time, regular inspection period and overhaul period;
inputting various input data of the aviation equipment to be tested into the three-layer artificial neural network model to obtain the availability index of the aviation equipment; the three-layer artificial neural network model is trained according to training sample data of the conventional aviation equipment.
2. The method of claim 1, wherein the artificial neural network model comprises an input layer, a hidden layer and an output layer, wherein the number of inputs of the input layer is 5, and the number of nodes of the hidden layer is 3.
3. The method of claim 1, further comprising:
collecting and counting average fault interval time, spare part turnover time, average repair time, regular inspection period and overhaul period of the conventional aviation equipment as training sample data, and using availability data as training result samples;
training an initial artificial neural network model through training sample data to train an available artificial neural network.
4. The method of claim 3, wherein training the initial artificial neural network model with training sample data to train an available artificial neural network comprises:
dividing training sample data and corresponding training sample results into n groups;
correcting the initial artificial neural network model according to each set of training sample data and corresponding training sample results in sequence by taking a group as a unit to obtain a first weight correction factor from a hidden layer to an output layer and a second weight correction factor from an input layer to the hidden layer;
according to the weight, the first weight correction factor and the second weight correction factor, correcting the parameters of the corresponding layer of the artificial neural network model; wherein the weight is composed of a group number and a group number used for training.
5. An artificial intelligence based aviation equipment availability prediction device, comprising:
the acquisition module is used for acquiring various input data of the aviation equipment to be detected, wherein the input data comprises equipment average fault interval time, spare part turnover time, average repair time, regular inspection period and overhaul period;
the input module is used for inputting various input data of the aviation equipment to be tested into the three-layer artificial neural network model to obtain the availability index of the aviation equipment; the three-layer artificial neural network model is trained according to training sample data of the conventional aviation equipment.
6. The apparatus of claim 5, wherein the artificial neural network model comprises an input layer, a hidden layer and an output layer, wherein the number of inputs of the input layer is 5, and the number of nodes of the hidden layer is 3.
7. The apparatus of claim 5, further comprising:
the collection module is used for collecting and counting average fault interval time, spare part turnover time, average repair time, regular inspection period and overhaul period of the conventional aviation equipment as training sample data, and using the availability data as a training result sample;
and the training module is used for training the initial artificial neural network model through training sample data to train an available artificial neural network.
8. The apparatus of claim 7, wherein the training module comprises:
the grouping unit is used for dividing the training sample data and the corresponding training sample results into n groups;
the correction unit is used for correcting the initial artificial neural network model according to each set of training sample data and the corresponding training sample result in sequence by taking a set as a unit to obtain a first weight correction factor from the hidden layer to the output layer and a second weight correction factor from the input layer to the hidden layer; according to the weight, the first weight correction factor and the second weight correction factor, correcting the parameters of the corresponding layer of the artificial neural network model; wherein the weight is composed of a group number and a group number used for training.
CN201911353973.2A 2019-12-24 2019-12-24 Aviation equipment availability prediction method based on artificial intelligence Pending CN111177975A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911353973.2A CN111177975A (en) 2019-12-24 2019-12-24 Aviation equipment availability prediction method based on artificial intelligence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911353973.2A CN111177975A (en) 2019-12-24 2019-12-24 Aviation equipment availability prediction method based on artificial intelligence

Publications (1)

Publication Number Publication Date
CN111177975A true CN111177975A (en) 2020-05-19

Family

ID=70657931

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911353973.2A Pending CN111177975A (en) 2019-12-24 2019-12-24 Aviation equipment availability prediction method based on artificial intelligence

Country Status (1)

Country Link
CN (1) CN111177975A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112347560A (en) * 2020-11-09 2021-02-09 张铮 Aviation equipment availability prediction method and device based on artificial intelligence

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112347560A (en) * 2020-11-09 2021-02-09 张铮 Aviation equipment availability prediction method and device based on artificial intelligence

Similar Documents

Publication Publication Date Title
CN112149316B (en) Aero-engine residual life prediction method based on improved CNN model
CN109308522B (en) GIS fault prediction method based on recurrent neural network
CN103226741B (en) Public supply mains tube explosion prediction method
CN106779151B (en) A kind of line of high-speed railway wind speed multi-point multi-layer coupling prediction method
CN102707256B (en) Fault diagnosis method based on BP-Ada Boost nerve network for electric energy meter
CN110807554B (en) Generation method and system based on wind power/photovoltaic classical scene set
CN113609596B (en) Aircraft aerodynamic characteristic prediction method based on neural network
CN111160520A (en) BP neural network wind speed prediction method based on genetic algorithm optimization
CN107590565A (en) A kind of method and device for building building energy consumption forecast model
CN112685504B (en) Production process-oriented distributed migration chart learning method
CN110070228B (en) BP neural network wind speed prediction method for neuron branch evolution
CN101814158A (en) Method for analyzing and processing experimental data based on artificial neural network
CN113468817A (en) Ultra-short-term wind power prediction method based on IGOA (optimized El-electric field model)
CN112580588A (en) Intelligent flutter signal identification method based on empirical mode decomposition
CN108537336B (en) Air quality prediction method based on deep neural network
CN113449072A (en) Construction method of excavator fault knowledge map based on deep learning
Altun et al. Identification of dynamic loads on structural component with artificial neural networks
CN112763215B (en) Multi-working-condition online fault diagnosis method based on modular federal deep learning
CN114295967A (en) Analog circuit fault diagnosis method based on migration neural network
CN111177975A (en) Aviation equipment availability prediction method based on artificial intelligence
CN116432359A (en) Variable topology network tide calculation method based on meta transfer learning
CN110909922B (en) Water resource efficiency detection and prediction method
CN111291490B (en) Nonlinear mapping intelligent modeling method for structure multi-scale heterogeneous response
CN106341837A (en) Inferior quality telephone traffic proportion value prediction method and device
CN113675844B (en) Multi-node power load prediction method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination