Test run method based on semi-physical simulation technology

Abstract

The invention discloses a test run method based on a semi-physical simulation technology, and belongs to the technical field of aero-engine test run. In the aero-engine test run process, test run personnel need to monitor a large amount of test run data and can encounter various sudden faults, and the personnel need to find the faults in the first time and find related data in the large amount of test run data at the moment of the faults, then judge and analyze the faults and make a normal processing mode to guarantee the safety of an engine. Some test run faults often need to be judged through a plurality of test run data and fault phenomena, and different processing modes are made, so that the ability of test run personnel is tested. According to the invention, a set of test run system is developed by using a semi-physical simulation technology, training of test run personnel is realized, the fault processing capability is improved, and the engine test run safety is guaranteed.

Description

Technical Field

[0001] The invention belongs to the technical field of aviation engine test, and in particular relates to a test method based on semi-physical simulation technology. Background Art

[0002] During the aircraft engine test process, test personnel need to monitor a large amount of test data and may encounter various sudden failures. At the moment of failure, personnel need to discover it immediately and find relevant data in a large amount of test data, and then make judgments, analyses, and make normal handling methods to ensure engine safety. Some test failures often require judgment based on multiple test data and failure phenomena, and different handling methods are made. However, the existing historical data is not effectively used, and the engine working model is not formed. Using physical engines for training cannot reproduce all test failures. On the one hand, this is because some test failures cause damage to the engine, and on the other hand, specific conditions are required to reproduce the failure. Generally, additional conditions are required to simulate the actual working conditions of the engine, which increases costs. Summary of the Invention

[0003] The purpose of the present invention is to provide a test method based on semi-physical simulation technology to solve the low data utilization efficiency and increased engine test cost of the traditional method, as described below:

[0004] A test method based on semi-physical simulation technology, the test method includes:

[0005] S1: Acquire test data and establish an engine test data database based on the test data, wherein the test data includes historical status data and historical operating data of the engine installed on the test bench;

[0006] S2: using a threshold method to distinguish fault data from non-fault data in the test data, wherein the threshold method is a pre-set data screening method;

[0007] S3: The fault data and non-fault data are input into a neural network model for machine learning or training to establish an engine operating model, which can be used to identify engine fault data;

[0008] S4: Real-time acquisition of actual test data of the engine under test, and input into the engine working model to obtain theoretical result data. The theoretical result data is manually calibrated in a human-computer interaction manner, and an indicator is assigned to indicate whether the data meets the standard.

[0009] S5: The engine working model marked with a qualified identifier receives the specified or preset fault data input by the operator in a human-computer interaction manner, and outputs the matching test data.

[0010] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0011] The method of the present invention is stable in operation and simple in operation, can effectively simulate engine test runs, can quickly reproduce test run faults during the test, improves data utilization efficiency, and reduces test costs. At the same time, it effectively trains personnel during the test or test run process, solves the problems of long test run personnel training cycle and high cost, and can also ensure the safety of engine test runs. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0013] Figure 1 It is a flow chart of the technical solution;

[0014] Figure 2 Define the diagram for the over-speed fault;

[0015] Figure 3 This is a diagram of the low oil level signal light failure and its treatment. DETAILED DESCRIPTION

[0016] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0017] In addition, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that aspects can be practiced without these specific details. In order to enable those skilled in the art to better understand the present invention, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments. The terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise stated, "multiple" means two or more.

[0018] like Figures 1 to 3 The test method based on semi-physical simulation technology shown in the figure includes:

[0019] S1: Acquire test data and build a database of engine test data based on the test data. The test data includes historical status data and historical working data of the engine installed on the test bench. Specifically,

[0020] The historical status data is the data fed back by the sensors installed on the engine itself, and the historical working data is the data fed back by the collection sensors added to the engine. It is converted and noise processed by the trolley intelligent software. For example, the engine's current, voltage or frequency and other signals are obtained and converted into temperature, pressure or speed, etc. through the trolley intelligent software. The test data is collected by the data collection software. The data collection software can set the startup conditions. For example, when the test data has a certain number of samples, it can meet the prerequisites for machine learning, which depends on the number of trolley starts, such as the number, type, and data type of the car to meet the preset standards, such as 30 engines of different models as the conditions for starting data collection.

[0021] S2: Use the threshold method to distinguish the fault data and non-fault data in the test data. The threshold method is a pre-set data screening method. The specific method of the threshold method is:

[0022] The overspeed fault data setting includes: when the engine high-pressure compressor speed N2 on the test bench exceeds the preset warning speed X1 r / min (different engine models correspond to different overspeed speeds), the speed warning signal is triggered; when the engine high-pressure compressor speed N2 exceeds the preset overspeed speed X2 r / min, the speed alarm signal is triggered and marked as overspeed fault data;

[0023] Setting of over-temperature fault data, wherein, during engine startup, when the actual temperature T5 of the engine's turbine gas exceeds the warning temperature A1°C, a temperature warning signal is triggered; when the actual temperature T5 of the engine's turbine gas exceeds the alarm temperature A2°C, a temperature alarm signal is triggered, marked as temperature fault data during engine startup; and, when the actual temperature T6 of the engine's turbine gas at maximum state exceeds the warning temperature A3°C, a temperature warning signal is triggered; when the actual temperature T6 of the engine's turbine gas at maximum state exceeds the alarm temperature A4°C, a temperature alarm signal is triggered, marked as temperature fault data during engine maximum state;

[0024] Setting of icing fault data: The engine is equipped with an icing temperature sensor. When the engine is frozen, the sensor will be blocked. After the blockage, the icing signal changes from 0 to 1. When the icing temperature sensor outputs 1 at this temperature, it means that ice has formed. Generally, when the icing signal changes from 0 to 1, the engine's icing signal light will light up, which is considered an icing fault. When the icing signal changes from 0 to 1, it is marked as icing fault data.

[0025] The setting of metal chip fault data, wherein a metal chip signal sensor is provided on the transmission system of the engine. When the metal chip alarm signal changes from 0 to 1, it is marked as metal chip fault data. The transmission system of the engine, for example, the metal chips generated by the grinding of the gears, triggers the alarm signal when the metal chip signal sensor of the engine is blocked. When the metal chip alarm signal changes from 0 to 1, the metal chip alarm signal light is on, which is defined as a metal chip fault.

[0026] S3: Fault data and non-fault data are input into the neural network model for machine learning or training to establish an engine operating model. The engine operating model can be used to identify engine fault data.

[0027] S4: Real-time acquisition of actual test data of the engine in the current test, and input into the engine working model to obtain theoretical result data. The theoretical result data is manually calibrated in a human-computer interaction manner, and an identification is given to indicate whether it meets the standard. Specifically,

[0028] Manual calibration includes,

[0029] If a fault has been manually determined and the theoretical result data output by the engine working model is also a fault, a qualified mark is given;

[0030] If a fault has been manually determined and the theoretical result data output by the engine working model has not reported a fault, corrections are made through human-computer interaction and the data is stored;

[0031] If it is manually judged that no fault has occurred and the theoretical result data output by the engine working model is an error, corrections are made and the data is stored in a human-computer interaction manner.

[0032] The marking of whether the standard is met includes that if the percentage of qualified marks in the tests under the same batch is greater than 80-90% (the ratio of the number of grid marks to the number of read data items), a qualified mark is given.

[0033] S5: The engine working model marked with a qualified mark receives the specified or preset fault data input by the operator in a human-computer interaction manner, and outputs matching test data. Generally, the engine working model with a qualified mark inputs the relationship between the engine parameters and the angle change of the throttle lever to simulate the normal engine test. By inputting the specified or preset fault data, the working model can output the test data corresponding to the fault data, which can be used by experienced staff to train new personnel. The test fault inside the software is called out, and the software will change the fault-related data according to the fault mode during the simulated test process to meet the requirements of reproducing the fault during the test process. This makes it possible to reproduce all test faults during training using a real engine, and will not cause damage to the engine due to some test faults. It also avoids the situation where specific conditions are required to reproduce the fault and incur additional costs. For example, using a real engine for training effectively reduces fuel consumption and shortens the personnel training cycle.

[0034] For example:

[0035] When the low oil level signal changes from 0 to 1 during the test run, the low oil level signal light will be on, and the software will report a low oil level signal light fault. When the test run personnel are handling the issue, they need to check other oil system parameters for further confirmation to see if the actual oil level is abnormal. If the oil level is <X1 L, the software will report a low oil level fault. If the oil level is ≥X1 L, it means that there is no abnormality in the oil level. It may be a circuit problem or a problem with the accessory itself. The test run can continue after monitoring the oil level without affecting the test run safety. If the oil level is <X1 L, the software will report a low oil level fault, indicating that the low oil level is a real signal, and further judgment needs to be made by confirming the oil supply pressure. When the oil supply pressure is ≥X2 MPa, we can pull the engine back to slow speed for cooling and then stop normally. If the oil supply pressure is <X2 MPa, the software reports a low oil pressure fault. At this time, the engine can only be shut down urgently and the engine can be started and cooled after the fault is eliminated within 15 minutes. If the fault cannot be eliminated or cannot be started within 15 minutes, the engine needs to be disassembled and inspected.

[0036] The above is a detailed introduction to the product provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only intended to help understand the core ideas of the present invention. It should be pointed out that, for those skilled in the art, without departing from the principles of the invention, several improvements and modifications can be made to the invention, and these improvements and modifications also fall within the scope of protection of the invention claims.

Claims

1. A test method based on semi-physical simulation technology, characterized in that: The test method shown includes, S1: Acquire test data and establish an engine test data database based on the test data, wherein the test data includes historical status data and historical operating data of the engine installed on the test bench; S2: using a threshold method to distinguish fault data from non-fault data in the test data, wherein the threshold method is a pre-set data screening method; S3: The fault data and non-fault data are input into a neural network model for machine learning or training to establish an engine operating model, which can be used to identify engine fault data; S4: Real-time acquisition of actual test data of the engine under test, and input into the engine working model to obtain theoretical result data. The theoretical result data is manually calibrated in a human-computer interaction manner, and an indicator is assigned to indicate whether the data meets the standard. S5: The engine working model marked with a qualified identifier receives the specified or preset fault data input by the operator in a human-computer interaction manner, and outputs the matching test data.

2. The test method according to claim 1, characterized in that: The historical status data in S1 is the data fed back by the sensors installed on the engine itself, and the historical working data is the data fed back by the acquisition sensors added to the engine, which are converted and processed by the trolley intelligent software; The test data is collected by data collection software, and the data collection software can set the start conditions.

3. The test method according to claim 1, characterized in that: The threshold method used in S2 to distinguish the fault data and non-fault data in the test data includes: The setting of overspeed fault data, wherein when the high-pressure compressor speed N2 of the engine on the test bench exceeds the preset warning speed X1r / min, a speed warning signal is triggered; when the high-pressure compressor speed N2 of the engine exceeds the preset overspeed speed X2r / min, a speed alarm signal is triggered and marked as overspeed fault data; Setting of over-temperature fault data, wherein, during engine startup, when the actual temperature T5 of the engine's turbine gas exceeds the warning temperature A1°C, a temperature warning signal is triggered; when the actual temperature T5 of the engine's turbine gas exceeds the alarm temperature A2°C, a temperature alarm signal is triggered, marked as temperature fault data during engine startup; and, when the actual temperature T6 of the engine's turbine gas at maximum state exceeds the warning temperature A3°C, a temperature warning signal is triggered; when the actual temperature T6 of the engine's turbine gas at maximum state exceeds the alarm temperature A4°C, a temperature alarm signal is triggered, marked as temperature fault data during engine maximum state; Setting of icing fault data, wherein an icing temperature sensor is set on the engine, and when the icing signal changes from 0 to 1, it is marked as icing fault data; The setting of metal chip fault data, wherein a metal chip signal sensor is provided on the transmission system of the engine. When the metal chip alarm signal changes from 0 to 1, it is marked as metal chip fault data.

4. The test method according to claim 1, characterized in that: Manual calibration in S4 includes, If a fault has been manually determined and the theoretical result data output by the engine working model is also a fault, a qualified mark is given; If a fault has been manually determined and the theoretical result data output by the engine working model has not reported a fault, corrections are made through human-computer interaction and the data is stored; If it is manually judged that no fault has occurred and the theoretical result data output by the engine working model is an error, corrections are made and the data is stored in a human-computer interaction manner.

5. The test method according to claim 4, characterized in that: The indicators of whether the standards are met in S4 include: In the tests under the same batch, if the percentage of qualified marks is greater than 80-90%, the qualified mark will be given.

6. The test method according to claim 1, characterized in that: S5 includes, The engine working model that has been assigned a qualified mark inputs the relationship between various engine parameters and the angle change of the throttle lever to simulate the normal engine test.

Images