CN111580498A - Aircraft environmental control system air cooling equipment robust fault diagnosis method based on random forest - Google Patents
Aircraft environmental control system air cooling equipment robust fault diagnosis method based on random forest Download PDFInfo
- Publication number
- CN111580498A CN111580498A CN202010379466.2A CN202010379466A CN111580498A CN 111580498 A CN111580498 A CN 111580498A CN 202010379466 A CN202010379466 A CN 202010379466A CN 111580498 A CN111580498 A CN 111580498A
- Authority
- CN
- China
- Prior art keywords
- control system
- air cooling
- random forest
- environmental control
- cooling equipment
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64D—EQUIPMENT FOR FITTING IN OR TO AIRCRAFT; FLIGHT SUITS; PARACHUTES; ARRANGEMENTS OR MOUNTING OF POWER PLANTS OR PROPULSION TRANSMISSIONS IN AIRCRAFT
- B64D45/00—Aircraft indicators or protectors not otherwise provided for
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64D—EQUIPMENT FOR FITTING IN OR TO AIRCRAFT; FLIGHT SUITS; PARACHUTES; ARRANGEMENTS OR MOUNTING OF POWER PLANTS OR PROPULSION TRANSMISSIONS IN AIRCRAFT
- B64D45/00—Aircraft indicators or protectors not otherwise provided for
- B64D2045/0085—Devices for aircraft health monitoring, e.g. monitoring flutter or vibration
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
Abstract
The invention discloses a robust fault diagnosis method for air cooling equipment of an aircraft environmental control system based on random forests, which comprises the following specific steps: the method comprises the following steps: acquiring and normalizing time sequence data; step two: constructing a random forest model and a training sample subset; step three: training each decision tree and finally finishing the training of the random forest model; step four: robust fault diagnosis of air cooling equipment of an aircraft environmental control system. The method provided by the invention adopts the random forest model to replace a threshold value constructed based on expert knowledge, extracts a logical association relation with the thermal fault of the air cooling equipment from the operation parameters of the aircraft environmental control system, and realizes the high-robustness thermal fault diagnosis of the air cooling equipment by means of the joint detection of multiple decision trees in the random forest model. According to the method, the detection risk caused by a single sensor and a single threshold is avoided by excavating the incidence relation between the parameters of the environmental control system and the fault signals of the air cooling equipment, and the fault detection rate is improved by avoiding the limitation of a single classifier on the fitting degree. Therefore, the method has higher engineering application value.
Description
Technical Field
The invention relates to the field of fault detection of an aircraft environmental control system, in particular to a robust fault diagnosis method for air cooling equipment of the aircraft environmental control system based on a random forest.
Background
An aircraft environmental control system, which is one of the important airborne systems, is responsible for providing a comfortable air environment for the onboard personnel. The environment control system provides adequate living and working environment for the driver and the airborne equipment by controlling parameters such as temperature, humidity, flow rate, pressure and the like of air in the control cabin.
In the process of executing the task by the airplane, the airborne electronic equipment releases a large amount of heat energy while running to increase the temperature of the equipment compartment, and the excessive temperature can cause the airborne electronic equipment of the airplane to be abnormal, thereby seriously influencing the execution of the task by the airplane. Therefore, the aircraft environmental control system extracts high-temperature and high-pressure gas from the engine through an air cooling loop, the high-temperature and high-pressure gas is subjected to primary cooling through a series of heat exchangers, is subjected to expansion cooling through a cooling turbine, is condensed and mixed with hot and cold air, and then is changed into cold air with proper temperature and is conveyed to an equipment cabin, so that air cooling and heat dissipation are provided for airborne equipment. The airborne electronic equipment is a key part for controlling the flight of the airplane, so that the normal working environment of the airplane needs to be guaranteed. Therefore, effective detection of air cooling equipment failure is an important part of aircraft maintenance and security work.
The existing air cooling equipment fault diagnosis method of the airplane environmental control system mainly uses a sensor to monitor an airplane equipment cabin, a temperature threshold value is artificially set through expert knowledge, and when an actual temperature measurement value exceeds the threshold value, the system sends out an air cooling equipment fault alarm. The fault detection method combining sensor monitoring with a fixed threshold has the problems of poor flexibility, easiness in disturbance of environmental factors and the like, so that a large number of false alarms and fault failure situations occur in the actual use process, and the correct judgment of the state of the airplane by a driver is seriously influenced.
Disclosure of Invention
In view of the above problems, the present application aims to provide a robust fault diagnosis method for an air cooling device of an aircraft environmental control system based on a random forest, which is implemented by establishing a random forest model including multi-decision tree integration, regarding an association relationship between ambient temperature changes of the air cooling device and changes of key device operation parameters in an air cooling loop, and using the operation parameters of the air cooling loop of the aircraft environmental control system as a criterion for judging the fault of the air cooling device.
The application discloses a robust fault diagnosis method for air cooling equipment of an aircraft environmental control system based on a random forest, which comprises the following steps:
the method comprises the following steps: time series data acquisition and normalization processing
The method comprises the steps of respectively obtaining state parameter signals of an aircraft environmental control system in a normal state and air cooling equipment in a fault state, constructing a normal state sample set and an air cooling equipment fault sample set, adjusting the sample quantities of the two types to be balanced, respectively marking fault labels, and carrying out normalization processing on all the sample sets to be used as training sample sets of a random forest model.
Step two: constructing random forest model and training sample subset
Setting parameters of a random forest model: the number of the decision trees, the depth of the decision trees and the termination condition are used as the basis to respectively construct the decision trees and finally form a random forest model. And (3) randomly extracting the training sample set obtained in the step one by adopting a re-sampling mode with replacement, ensuring the data quantity of the sample subsets to be consistent, and obtaining the random training sample subsets corresponding to the number of the decision trees.
Step three: training random forest model
And distributing the sample subset to each decision tree model in the random forest models, and independently training the decision tree models until each node of each decision tree model meets a termination condition, so that the training of the random forest models is finished.
Step four: robust fault diagnosis for air cooling equipment of aircraft environmental control system
Acquiring actual operation signals of the airplane environment control system, carrying out data normalization processing on the actual operation signals, wherein the data normalization processing is the same as that in the first step, obtaining a sample set to be detected of the airplane environment control system, respectively inputting the sample set into each decision tree in the random forest models trained in the fourth step, integrating prediction results of all the decision trees by the random forest models, and detecting whether the air cooling equipment fault exists in the current airplane environment control system.
The invention has the advantages and positive effects that:
(1) when the air cooling system fails, the monitoring parameters of the environmental control system are subjected to complex nonlinear change, the method utilizes a random forest algorithm to mine the incidence relation between the parameters of the environmental control system and the heat alarm signals of the air cooling equipment from the operation data, and further utilizes the parameters of the environmental control system to accurately detect the failure of the air cooling equipment, so that the problems of false alarm, missing alarm and the like caused by fixed single threshold in environmental disturbance are avoided;
(2) the multi-decision tree structure of the random forest model avoids the fitting upper limit of a single classifier, and the detection accuracy is higher in robust fault diagnosis of the air cooling equipment of the aircraft environmental control system;
(3) the fault diagnosis is carried out through a huge parameter set of the environmental control system, the detection risk caused by the abnormality of a single sensor is avoided, and meanwhile, the dynamic detection of the operation environment of the air cooling equipment is realized through the continuously changed parameters of the environmental control system.
Drawings
FIG. 1 is a flow chart of a robust fault diagnosis method for air cooling equipment of an aircraft environmental control system based on a random forest according to the invention;
FIG. 2 is a schematic diagram of a random forest model structure of the present invention;
FIG. 3 is initial input data in an embodiment of the invention;
FIG. 4 shows the result of normalization of the original signal parameters in an embodiment of the present invention;
fig. 5 is a result of robust fault diagnosis of the air cooling apparatus in the embodiment of the present invention;
Detailed Description
The robust fault diagnosis method for the air cooling equipment of the aircraft environmental control system based on the random forest mainly comprises the following steps:
the method comprises the following steps: time series data acquisition and normalization processing
And acquiring running parameter signals of the aircraft environmental control system corresponding to the normal state and the fault state of the air cooling equipment, and adjusting two sample data volumes to achieve sample balance. Suppose the acquired signal sequence is IN＝[X1，X2，...XN]The signal length is N, where [ X ]1，...XN/2]Is a normal state sample, [ X ]N/2+1，...XN]Are fault status samples. Then setting each parameter signal set XiIn which M operating parameters, i.e. X, are includedi＝[xi1，xi2，...xiM]. According to the fault state, is INAdd failure label Y ═ 0, 0, … 1, 1]. For original signal sequence INAnd (6) carrying out normalization processing.
Step two: constructing random forest model and training sample subset
According to the dimension M of the sample set, determining the number T of decision trees and the depth d of the decision trees, and setting a termination condition: and (3) constructing a random forest model by using the minimum sample number s on the node and the minimum information gain p on the node, wherein the structure diagram of the random forest model is shown in FIG. 2. With IN＝[X1，X2，...XN]And the corresponding fault label vector Y is the total sample set, the replaced random sampling is carried out, each random sampling is independent, the sampling times depend on the number T of decision trees, and the training sample set S is constructed as S1，S2，...ST]。
Step three: training random forest model
With a subset of samples SiAnd (4) as a sample of the root node of each decision tree in the random forest model, starting training from the root node, and judging whether the current node meets the termination condition in the step two. If the current node meets the termination condition in the second step, performing operation 1; and if the current node does not reach the termination condition, performing operation 2.
Operation 1: setting the current node as a leaf node, wherein the output of the leaf node is the mean value of all sample values in the sample set of the current node, and then continuing to train other non-leaf nodes.
Operation 2: and randomly and unreleased extracting M-dimensional features from the M-dimensional features of the training samples, searching the one-dimensional feature k with the best regression effect and the threshold th of the feature from the M-dimensional features, and if the kth-dimensional feature of the sample on the current node is smaller than th, dividing the sample into left nodes, otherwise, dividing the sample into right nodes.
And continuing to train other nodes, and growing each decision tree to the maximum extent. And continuing to train other nodes and other decision trees of the decision tree in the mode until all the decision trees are trained completely, and finishing training the random forest model.
Step four: robust fault diagnosis for air cooling equipment of aircraft environmental control system
Collecting actual operation signal of airplane environment control system
And (4) after normalization operation which is the same as the first step is carried out on the test sample, the test sample is input into a random forest model, the relation between the characteristic value of the test sample and a threshold th of the current node is judged from the root node of the current decision tree, if the characteristic value is smaller than th, the test sample enters a left node, otherwise, the test sample enters a right node until the test sample reaches a certain leaf node, and the predicted value of the decision tree is obtained. And repeating the operation in the step, voting all the decision trees to finally obtain the prediction result of the model, wherein when the output is 0, no air cooling equipment fault occurs, and when the output is 1, the air cooling equipment fault alarm is carried out.
Examples
In the embodiment, 29 operation parameters such as low-pressure input pressure, low-pressure output pressure and the like of a turbine of an aircraft environmental control system are taken as an example, the method is adopted to detect the faults of the air cooling equipment so as to explain the content of the invention, and the using process of the content of the invention is further explained.
Initial data as shown in fig. 3, according to the data recording situation, thermal failure of the air cooling equipment occurred between the 45856 th sampling point and 49824 th sampling point in the acquired operation parameter signal. Therefore, the data is divided into fault data, and the normal sampling point interval which is the same as the quantity of the fault data and is continuous is selected as normal data for model training.
The method comprises the following steps: time series data acquisition and normalization processing
Normal data [ X ]10000，...X13968]And fault data [ X ]45856，...X49824]Forming an original signal sequence, wherein the sample capacity of normal data and fault data is 29, carrying out normalization processing to obtain a random forest training sample I with the total length of 7938, and adding a fault label vector Y of [0, 0, … 1, 1 ]]The normalized result is shown in fig. 4.
Step two: constructing random forest model and training sample subset
The parameter settings for the random forest model are as follows:
number of decision trees: 100
Decision tree maximum depth: adaptive so that each leaf node has only one category
Leaf node minimum number of samples: 2
Minimum information gain: 0
And (4) performing replaced random sampling on the random forest training samples, wherein the random sampling is independent every time, the sampling times are 100 times, and a training sample subset is constructed.
Step three: training random forest model
Randomly selecting a certain training sample subset as a sample of a root node of the current decision tree, starting training from the root node, and continuing to train other nodes and other decision trees of the decision tree until all the decision trees are trained completely, and finishing training of the random forest model.
Step four: robust fault diagnosis for air cooling equipment of aircraft environmental control system
And acquiring actual operation data corresponding to the 29 parameters of the aircraft environment control system, carrying out normalization processing, and inputting each sampling point in each actual data into a trained random forest model respectively for fault detection to obtain a detection result as shown in the specification. As can be seen from the result chart of fig. 5, the fault detection result is consistent with the actual fault occurrence situation, which shows that the robust fault diagnosis of the aircraft air cooling device can be timely and accurately performed by the content of the present invention.
Unless defined otherwise, all technical and/or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The materials, methods, and examples set forth in this application are illustrative only and not intended to be limiting.
Although the present invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications, and variations will be apparent to those skilled in the art in light of the teachings of this application and yet remain within the scope of this application.
Claims (5)
1. A robust fault diagnosis method for air cooling equipment of an aircraft environmental control system based on a random forest is characterized by comprising the following steps:
the first step is as follows: acquiring and normalizing time sequence data;
the second step is that: constructing a random forest model and a training sample subset;
the third step: training a random forest model;
the fourth step: robust fault diagnosis of air cooling equipment of an aircraft environmental control system.
2. The robust fault diagnosis method for the air cooling device of the aircraft environmental control system according to claim 1, characterized in that:
the method comprises the steps of respectively obtaining state parameter signals of an aircraft environmental control system in a normal state and air cooling equipment in a fault state, constructing a normal state sample set and an air cooling equipment fault sample set, adjusting the sample quantities of the two types to be balanced, respectively marking fault labels, and carrying out normalization processing on all the sample sets to be used as training sample sets of a random forest model.
3. The robust fault diagnosis method for the air cooling device of the aircraft environmental control system according to claim 2, characterized in that:
setting parameters of a random forest model: the number of the decision trees, the depth of the decision trees and the termination condition are used as the basis to respectively construct the decision trees and finally form a random forest model. And (3) randomly extracting the training sample set obtained in the step one by adopting a re-sampling mode with replacement, ensuring the data quantity of the sample subsets to be consistent, and obtaining the random training sample subsets corresponding to the number of the decision trees.
4. The robust fault diagnosis method for the air cooling device of the aircraft environmental control system according to claim 3, characterized in that:
and C, distributing the sample subset obtained in the step II to each decision tree model in the random forest models, and independently training the decision tree models until each node of each decision tree model meets a termination condition, so that the training of the random forest models is finished.
5. The robust fault diagnosis method for the air cooling device of the aircraft environmental control system according to claim 4, characterized in that:
acquiring actual operation signals of the airplane environment control system, carrying out data normalization processing on the actual operation signals, wherein the data normalization processing is the same as the data normalization processing in the first step, obtaining a sample set to be detected of the airplane environment control system, respectively inputting the sample set into each decision tree in the trained random forest models in the third step, integrating prediction results of all the decision trees by the random forest models, and detecting whether the air cooling equipment fault exists in the current airplane environment control system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010379466.2A CN111580498A (en) | 2020-05-08 | 2020-05-08 | Aircraft environmental control system air cooling equipment robust fault diagnosis method based on random forest |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010379466.2A CN111580498A (en) | 2020-05-08 | 2020-05-08 | Aircraft environmental control system air cooling equipment robust fault diagnosis method based on random forest |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111580498A true CN111580498A (en) | 2020-08-25 |
Family
ID=72112567
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010379466.2A Pending CN111580498A (en) | 2020-05-08 | 2020-05-08 | Aircraft environmental control system air cooling equipment robust fault diagnosis method based on random forest |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111580498A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113255546A (en) * | 2021-06-03 | 2021-08-13 | 成都卡莱博尔信息技术股份有限公司 | Diagnosis method for aircraft system sensor fault |
| CN113341927A (en) * | 2021-06-11 | 2021-09-03 | 江西洪都航空工业集团有限责任公司 | Flight control system servo actuator BIT fault detection method and device |
| CN113467230A (en) * | 2021-07-21 | 2021-10-01 | 珠海格力电器股份有限公司 | Magnetic bearing system and control method and device thereof, storage medium and processor |
| CN116520817A (en) * | 2023-07-05 | 2023-08-01 | 贵州宏信达高新科技有限责任公司 | ETC system running state real-time monitoring system and method based on expressway |
-
2020
- 2020-05-08 CN CN202010379466.2A patent/CN111580498A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113255546A (en) * | 2021-06-03 | 2021-08-13 | 成都卡莱博尔信息技术股份有限公司 | Diagnosis method for aircraft system sensor fault |
| CN113255546B (en) * | 2021-06-03 | 2021-11-09 | 成都卡莱博尔信息技术股份有限公司 | Diagnosis method for aircraft system sensor fault |
| CN113341927A (en) * | 2021-06-11 | 2021-09-03 | 江西洪都航空工业集团有限责任公司 | Flight control system servo actuator BIT fault detection method and device |
| CN113341927B (en) * | 2021-06-11 | 2022-12-02 | 江西洪都航空工业集团有限责任公司 | Flight control system servo actuator BIT fault detection method and device |
| CN113467230A (en) * | 2021-07-21 | 2021-10-01 | 珠海格力电器股份有限公司 | Magnetic bearing system and control method and device thereof, storage medium and processor |
| CN116520817A (en) * | 2023-07-05 | 2023-08-01 | 贵州宏信达高新科技有限责任公司 | ETC system running state real-time monitoring system and method based on expressway |
| CN116520817B (en) * | 2023-07-05 | 2023-08-29 | 贵州宏信达高新科技有限责任公司 | ETC system running state real-time monitoring system and method based on expressway |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111580498A (en) | Aircraft environmental control system air cooling equipment robust fault diagnosis method based on random forest | |
| CN109446187B (en) | Method for monitoring health state of complex equipment based on attention mechanism and neural network | |
| CN107168205B (en) | A kind of online health monitoring data collection and analysis method of civil aircraft air-conditioning system | |
| CN102246110B (en) | Standardization of data used for monitoring an aircraft engine | |
| US8112368B2 (en) | Method, apparatus and computer program product for predicting a fault utilizing multi-resolution classifier fusion | |
| CN109829468B (en) | Bayesian network-based civil aircraft complex system fault diagnosis method | |
| CN110567722A (en) | Civil engine starting system health monitoring method based on performance parameters | |
| CN104756029B (en) | A kind of system of the parts group of monitoring device | |
| CN113053171B (en) | Civil aircraft system risk early warning method and system | |
| CN105955214B (en) | Batch process fault detection method based on sample time-series and neighbour's affinity information | |
| CN104699050A (en) | Leaf-shred preparation segment on-line monitoring and fault diagnosing method for cigarette filament treatment driven by data | |
| CN106067032A (en) | Evaluation methodology for the sensor system of selection of data exception monitoring | |
| Lu et al. | Fault detection, diagnosis, and performance assessment scheme for multiple redundancy aileron actuator | |
| CN115320886A (en) | Real-time monitoring method and system for aircraft control surface fault | |
| US20090248363A1 (en) | Method of multi-level fault isolation design | |
| Zhong et al. | An improved correlation-based anomaly detection approach for condition monitoring data of industrial equipment | |
| CN114036998A (en) | Method and system for fault detection of industrial hardware based on machine learning | |
| CN116521406A (en) | Method for detecting anomaly of non-overrun flight parameter data of aero-engine based on residual gate GRU-VAE model | |
| Jiang et al. | Research recognition of aircraft engine abnormal state | |
| Gao et al. | Design requirements of PHM system fault diagnosis capability | |
| Arockia Dhanraj et al. | Increasing the Wind Energy Production by Identifying the State of Wind Turbine Blade | |
| Zhang et al. | Research on general aircraft cluster health assessment method | |
| CN113465930B (en) | Gas turbine multi-sensor fault detection method based on hybrid method | |
| CN110515365A (en) | A kind of industrial control system abnormal behaviour analysis method that Kernel-based methods excavate | |
| Chen et al. | A multimode anomaly detection method based on oc-elm for aircraft engine system |
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 |