CN112130543B - Carrier rocket electrical system based on FC-AE-1553 high-speed bus - Google Patents
Carrier rocket electrical system based on FC-AE-1553 high-speed bus Download PDFInfo
- Publication number
- CN112130543B CN112130543B CN202010704225.0A CN202010704225A CN112130543B CN 112130543 B CN112130543 B CN 112130543B CN 202010704225 A CN202010704225 A CN 202010704225A CN 112130543 B CN112130543 B CN 112130543B
- Authority
- CN
- China
- Prior art keywords
- rocket
- data
- speed bus
- long
- time memory
- 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.)
- Active
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
- G05B23/0208—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the configuration of the monitoring system
- G05B23/0213—Modular or universal configuration of the monitoring system, e.g. monitoring system having modules that may be combined to build monitoring program; monitoring system that can be applied to legacy systems; adaptable monitoring system; using different communication protocols
-
- 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
- G05B23/0254—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 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural 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
- G06N3/04—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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Automation & Control Theory (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
A test method for a carrier rocket electrical system based on an FC-AE-1553 high-speed bus comprises the following steps: s1, starting a test program, initializing a ground host into a network controller of the FC-AE-1553 high-speed bus, and initializing other equipment of an electric system into network terminal equipment of the FC-AE-1553 high-speed bus; s2, after the test indexes of each piece of equipment in the rocket area are qualified, the ground host computer gives the control right to the comprehensive control computer; s3, the ground host sends a primary engine ignition instruction to the comprehensive actuator, and the comprehensive control computer broadcasts the takeoff time to the FC-AE-1553 high-speed bus globally after receiving the takeoff and enters navigation guidance and rocket attitude control; s4, the data management center collects the measurement data of the carrier rocket, then evaluates the health degree of the rocket, and sends the evaluation result to the comprehensive control computer.
Description
Technical Field
The invention relates to a carrier rocket electrical system based on an FC-AE-1553 high-speed bus, in particular to a carrier rocket electrical system based on high-speed bus information interaction, and belongs to the technical field of carrier rocket electrical system design.
Background
At present, the competition in the field of space launch is increasingly intense, low-cost carrier rockets are released or developed by all aerospace big countries, the rockets have high automation degree, short development period and strong universality, the launch period and the cost of the rockets are greatly reduced by adopting an advanced intelligent development and production mode, and the method has great advantages compared with the traditional development mode of the carrier rockets in China. The electric system is one of the most important systems of the carrier rocket, and whether high reliability, health management and rapid test launching can be realized becomes an important factor for determining the quality of the electric system. The traditional carrier rocket electrical system has the defects of large scale, complex structure, low integration degree, complex bus type, low bus throughput and the like, seriously influences the carrying capacity, brings great inconvenience to function upgrading, maintenance and testing of the carrier rocket, and needs high sampling rate and high bus throughput rate support due to high power output and high dynamic characteristics of a rocket engine when meeting the requirements of rocket fault diagnosis and health monitoring functions under new conditions, so that the traditional rocket electrical system architecture is difficult to meet, and a new generation carrier rocket electrical system with high-speed information interaction, high reliability and high integration needs to be developed.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the defects of the prior art are overcome, and the test method for the electric system of the launch vehicle based on the FC-AE-1553 high-speed bus is provided and comprises the following steps: s1, starting a test program, initializing a ground host into a network controller of the FC-AE-1553 high-speed bus, and initializing other equipment of an electric system into network terminal equipment of the FC-AE-1553 high-speed bus; s2, after the test indexes of each piece of equipment in the rocket area are qualified, the ground host computer gives the control right to the comprehensive control computer; s3, the ground host sends a primary engine ignition instruction to the comprehensive actuator, and the comprehensive control computer broadcasts the takeoff time to the FC-AE-1553 high-speed bus globally after receiving the takeoff and enters navigation guidance and rocket attitude control; s4, the data management center collects the measurement data of the carrier rocket, then evaluates the health degree of the rocket, and sends the evaluation result to the comprehensive control computer.
The purpose of the invention is realized by the following technical scheme:
a test method for a carrier rocket electrical system based on an FC-AE-1553 high-speed bus is disclosed, wherein the electrical system at least comprises a ground host, a comprehensive control computer, a comprehensive actuator and a data management center, and the test method comprises the following steps:
s1, starting a test program, initializing a ground host into a network controller of the FC-AE-1553 high-speed bus, and initializing other equipment of an electric system into network terminal equipment of the FC-AE-1553 high-speed bus;
s2, after the test indexes of each piece of equipment on the rocket ground are qualified, the ground host sends a switching control right instruction to the comprehensive control computer, and the comprehensive control computer receives the switching control right instruction and then re-initializes the network controller of the FC-AE-1553 high-speed bus to receive the network control right;
s3, the ground host sends a primary engine ignition instruction to the comprehensive actuator, and the comprehensive actuator performs ignition on the engine according to the ignition timing of the engine; the integrated control computer dynamically monitors a rocket takeoff signal, broadcasts the takeoff time to the FC-AE-1553 high-speed bus globally after receiving the takeoff, finishes the time correction of the whole system, and enters navigation guidance and rocket attitude control;
s4, the data management center collects the measurement data of the carrier rocket, then evaluates the health degree of the rocket, and sends the evaluation result to the comprehensive control computer.
Preferably, in S4, the method for testing the electric system of the launch vehicle based on the FC-AE-1553 high-speed bus adopts a long-time memory-based self-encoder to evaluate the health degree of the launch vehicle, and the evaluation method includes the following steps:
s41, measuring and collecting original data in the starting process of the carrier rocket engine, and then preprocessing the original data;
s42, dividing the preprocessed data into a training set and a test set;
s43, establishing a long-time memory self-encoder model according to the dimension of the original data, and initializing model parameters;
s44, performing multi-generation training on the long-time and short-time memory self-encoder model by using a training set, so that the loss of the long-time and short-time memory self-encoder model is converged, and obtaining a trained long-time and short-time memory self-encoder model after a preset training generation number is reached;
and S45, calculating the output of the test set by utilizing the trained long-time memory autoencoder model, and then calculating the health degree of the launch vehicle engine according to the test set and a preset threshold value.
In the test method of the launch vehicle electrical system based on the FC-AE-1553 high-speed bus, preferably, the preprocessing described in S41 adopts a min-max standardization method, that is:
wherein x' is the normalized data sample, x is the data sample before normalization, xminIs the minimum of x, xmaxIs the maximum value in x.
Preferably, the method for testing the electric system of the launch vehicle based on the FC-AE-1553 high-speed bus comprises the following steps: and selecting normal original data as a training set, and selecting data with measurement errors as a test set.
Preferably, the long-time memory self-encoder model comprises a plurality of encoding layers and a plurality of decoding layers; the coding layer is used for extracting the characteristics of the data samples; the decoding layer is used to restore the extracted features to a matrix consistent with the input sample format.
Preferably, in the test method for the carrier rocket electrical system based on the FC-AE-1553 high-speed bus, in S44, when training for multiple generations, each generation divides a verification set from a training set, and the verification set is used for checking the training effect of the long-time and short-time memory self-encoder model and judging whether the long-time and short-time memory self-encoder model is over-fitted.
Preferably, the trained long-time and short-time memory self-encoder model is calculated to calculate the output of the test set, the absolute value of the difference value between the output of the test set and the output of the test set is subtracted from the absolute value of the difference value of the test set, and finally the health degree of the launch vehicle engine is calculated.
A carrier rocket electrical system based on an FC-AE-1553 high-speed bus comprises an rocket comprehensive control computer, inertia measurement equipment, a data management center, a comprehensive actuator, a servo controller, a driving mechanism and a ground host; the carrier rocket electrical system adopts an FC-AE-1553 high-speed bus for data communication;
the rocket comprehensive control computer is used for autonomously controlling the rocket, is used as a network controller of an FC-AE-1553 high-speed bus, and controls a comprehensive actuator, a servo controller and a driving mechanism;
the system comprises an inertial measurement device, a comprehensive actuator, a servo controller, a driving mechanism, a rocket engine nozzle, a ground host and a control system, wherein the inertial measurement device is used for measuring inertial parameters, the comprehensive actuator is used for controlling the working time sequence of the rocket, the servo controller and the driving mechanism are used for controlling the rocket engine nozzle to swing, and the ground host is used for rocket-ground information interaction, control right switching, ignition sending and real-time data interpretation;
the inertia measurement equipment, the servo controller and the driving mechanism send data to the arrow comprehensive control computer;
the data management center is used for collecting data of the carrier rocket and modulating a telemetering signal, and meanwhile, the state of the rocket is diagnosed and evaluated by the self-encoder based on long-time and short-time memory, and an evaluation result is sent to the comprehensive control computer.
Preferably, the carrier rocket electrical system based on the FC-AE-1553 high-speed bus adopts a long-time memory-based self-encoder to evaluate the health degree of the rocket, and the evaluation method comprises the following steps:
s101, measuring and collecting original data in the starting process of a carrier rocket engine, and then preprocessing the original data;
s102, dividing the preprocessed data into a training set and a test set;
s103, establishing a long-time memory self-encoder model according to the dimension of the original data, and initializing model parameters;
s104, performing multi-generation training on the long-time and short-time memory self-encoder model by using a training set, so that the loss of the long-time and short-time memory self-encoder model is converged, and obtaining a trained long-time and short-time memory self-encoder model after a preset training generation number is reached;
and S105, calculating the output of the test set by utilizing the trained long-time and short-time memory autoencoder model, and then calculating the health degree of the engine of the carrier rocket according to the test set and a preset threshold value.
Preferably, the inertia measurement device, the comprehensive actuator, the servo controller and the driving mechanism are all used as network terminal devices of the FC-AE-1553 high-speed bus; the ground host serves as a network controller of an FC-AE-1553 high-speed bus.
Compared with the prior art, the invention has the following beneficial effects:
(1) the invention adopts the switched bus architecture situation, information is uniformly transmitted through the FC-AE-1553 bus, the number of buses on the arrow is reduced, a single machine interface is simple, and the system architecture is clear;
(2) compared with a conventional rocket control system, the rocket control system is additionally provided with a data management center, the device inherits the functions of the original telemetering acquisition center, utilizes the advantages of data of the device, adopts expert knowledge and an artificial intelligence method to carry out comprehensive health diagnosis on the rocket multi-system, can provide an emergency method according to the fault type and the fault grade when an abnormality occurs, and increases the rocket emergency capacity;
(3) the thermal redundancy is realized at the bottom layer of the FC-AE-1553 high-speed bus, the hardware double redundancy and triple redundancy schemes of the MIL-STD-1553B standard scheme are avoided in use, the system design difficulty is reduced, emergency treatment measures are added on the system level, the bus abnormal probability is reduced, and the bus reliability is further improved;
(4) by adopting the technical scheme of the FC-AE-1553 high-speed bus, the information interaction rate is greater than 1Gbps, and is greatly improved compared with the 1Mbps of the MIL-STD-1553B standard used by the active rocket, so that the problem of insufficient information interaction capacity of the carrier rocket is solved, and the functional implementation of health diagnosis, autonomous planning, high-speed measurement and the like of the rocket is guaranteed;
(5) the invention is based on FC-AE-1553 high-speed bus design, adopts optical fiber to replace copper shaft bus or analog cable, reduces the sensitivity of communication to external electromagnetic signals, and the system communication error rate can be 10 of the bus of MIL-STD-1553B standard-7Down to 10-12And the reliability of rocket information interaction is greatly improved.
(6) According to the method, the model with the long-time and short-time memory self-encoder is adopted to train the ground data, the obtained model can be used for engine data diagnosis in the rocket flight process, historical flight data is verified, the measurement fault can be effectively eliminated, the diagnosis accuracy rate is higher than 95%, the false alarm rate is lower than 5%, and the health degree evaluation can be accurately realized.
Drawings
FIG. 1 is a block diagram of an electric system of a launch vehicle based on an FC-AE-1553 high-speed bus.
FIG. 2 is a flowchart of the test procedure of the FC-AE-1553-based high-speed bus of the present invention.
FIG. 3 is a flow diagram of the information for the device of the present invention based on the FC-AE-1553 high speed bus standard.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
In order to solve the technical problem, the invention provides a carrier rocket electrical system based on an FC-AE-1553 high-speed bus, which comprises an rocket comprehensive control computer, inertia measurement equipment, a data management center, a comprehensive actuator, a servo controller, a driving mechanism and a ground host computer, wherein the rocket comprehensive control computer is connected with the inertia measurement equipment;
the FC-AE-1553 high-speed bus is a bottom-layer communication protocol developed based on the FC-AE-1553 standard, has a communication code rate of 1Gbps, realizes A/B dual redundancy per se, and supports a multi-Network Controller (NC) mode and a multi-Network Terminal (NT) mode.
The rocket comprehensive control computer is used for autonomously controlling a rocket flight test and controlling global communication as an FC-AE-1553 Network Controller (NC).
The inertia measurement equipment is a ten-meter optical fiber inertia unit which comprises five accelerometers and five rate gyroscopes, realizes triple redundant output by the inertia measurement equipment, is accessed to an FC-AE-1553 bus and is used as network terminal equipment (NT);
the data management center is used for collecting whole rocket data, modulating telemetering signals, performing functions of online fault diagnosis, online health management trajectory planning and the like on the flight state of the rocket, and accessing the data management center to an FC-AE-1553 bus as network terminal equipment (NT);
the comprehensive actuator is used for executing rocket time sequence control, comprises engine switch control, initiating explosive device detonation, interstage separation and the like, is used as FC-AE-1553 network terminal equipment (NT), receives a rocket comprehensive computer control instruction and executes corresponding actions;
the servo controller and the driving mechanism execute the swinging work of the rocket engine spray pipe, are accessed to an FC-AE-1553 bus to serve as FC-AE-1553 network terminal equipment (NT), receive an instruction of controlling a swinging angle of the rocket comprehensive computer, control the driving mechanism to swing the engine, and feed back rudder deflection angle information to the comprehensive control computer under the control of the comprehensive control computer (NC);
and the comprehensive control computer (NC) acquires inertia measurement information of the inertia measurement equipment (NT), feeds back rudder deflection angle information by a servo controller in the previous beat, iteratively calculates the current attitude, speed and position of the rocket, simultaneously outputs the swing angle information to the servo controller, and outputs rocket time sequence control information to the comprehensive actuator according to a theoretical orbit.
The ground host is used for realizing arrow-ground information interaction, switching control power, sending ignition, real-time data interpretation and the like and is used as one of FC-AE-1553 Network Controllers (NC) with multiple NCs.
A test method for a carrier rocket electrical system based on an FC-AE-1553 high-speed bus is disclosed, wherein the electrical system at least comprises a ground host, a comprehensive control computer, a comprehensive actuator and a data management center, and the test method comprises the following steps:
s1, starting a test program, initializing a ground host into a network controller of the FC-AE-1553 high-speed bus, and initializing other equipment of an electric system into network terminal equipment of the FC-AE-1553 high-speed bus;
s2, after the test indexes of each piece of equipment on the rocket ground are qualified, the ground host sends a switching control right instruction to the comprehensive control computer, and the comprehensive control computer receives the switching control right instruction and then re-initializes the network controller of the FC-AE-1553 high-speed bus to receive the network control right;
s3, the ground host sends a primary engine ignition instruction to the comprehensive actuator, and the comprehensive actuator performs ignition on the engine according to the ignition timing of the engine; the integrated control computer dynamically monitors a rocket takeoff signal, broadcasts the takeoff time to the FC-AE-1553 high-speed bus globally after receiving the takeoff, finishes the time correction of the whole system, and enters navigation guidance and rocket attitude control;
s4, the data management center collects the measurement data of the carrier rocket, then evaluates the health degree of the rocket, and sends the evaluation result to the comprehensive control computer.
Example (b):
taking a two-stage rocket as an example, a carrier rocket electrical system based on an FC-AE-1553 high-speed bus adopts an exchange structure, rocket secondary equipment comprises a comprehensive control computer, inertia measurement equipment, a comprehensive actuator (secondary), a servo controller and a driving mechanism (secondary), a data management center, a rocket primary comprises a comprehensive actuator (primary), a servo controller and a driving mechanism (primary), and telemetering, collecting and editing equipment (primary); the ground equipment comprises a ground host. As shown in FIGS. 1-3.
The integrated control computer 201, the inertia measurement device 202, the integrated actuator (secondary) 203, the servo controller and driving mechanism (secondary) 204, and the data management center 205 are connected to the dual-redundancy optical fiber switch 206 through optical fibers, the optical fibers led out by the integrated actuator (primary) 101, the servo controller and the driving mechanism (primary) 102 are connected to the dual-redundancy optical fiber switch 206 through optical fiber separation plugs, and the ground host 104 is further connected to the dual-redundancy optical fiber switch 206 through optical fiber unplugging plugs and optical fiber separation plugs.
Further, the test program starts, the ground host 104 is initialized to NC, other devices are initialized to NT, the ground host 104 sequentially completes device initialization handshake in the FC-AE-1553 bus network, binds test data of the integrated control computer 201, sends test instructions, and after test indexes of various devices in the rocket ground are qualified, the ground host 104 sends out a control right switching instruction in the FC-AE-1553_ NCNT mode, the integrated control computer 201 receives the control right switching instruction, re-initializes itself to be new NC, receives the FC-AE-1553 bus network control right, the ground host 104 sends out a primary engine ignition instruction to the integrated actuator (primary) 101 in the FC-AE-1553_ NCNT mode, the integrated actuator (primary) 101 receives the ignition instruction, and ignites the engine according to the engine ignition timing sequence, the integrated control computer 201 dynamically monitors a rocket takeoff signal, broadcasts the takeoff time globally in the FC-AE-1553 bus network after receiving the takeoff signal, finishes the time correction of the whole system, and enters the navigation guidance and the rocket attitude control.
Further, the comprehensive control computer 201 executes FC-AE-1553 bus protocol primitive synchronization once every 1s period, and the synchronization delay is less than 10 us.
Further, every 5ms period, the integrated control computer 201 retrieves inertial measurement data from the inertial measurement device 202 in FC-AE-1553_ NTNC mode;
further, the integrated control computer 201 calculates the arrow body 6DOF information every 10ms period, performs timing calculations such as a rocket engine shutdown equation and cowling separation, and outputs a shutdown type or timing to the integrated actuator (primary) 101 or the integrated actuator (secondary) 203 in an FC-AE-1553_ NCNT mode to perform shutdown or timing control if a shutdown or timing condition is satisfied.
Further, every 20ms period, the integrated control computer 201 obtains a program angle according to navigation operation, obtains an attitude angle through actual measurement calculation, and brings the attitude angle into a known differential equation, calculates a swing angle signal, outputs the swing angle signal to the servo controller and driving mechanism (primary) 102 or the servo controller and driving mechanism (secondary) 204 in an FC-AE-1553_ NCNT mode to execute swing angle control, and meanwhile, the integrated control computer 201 obtains rudder feedback information of the servo controller and driving mechanism (primary) 102 or the servo controller and driving mechanism (secondary) 204 in the FC-AE-1553_ NTNC mode, updates a differential equation coefficient, and achieves a closed-loop control effect.
Further, every 5ms period, the data management center 205 performs, in an NM (monitor terminal) mode, monitoring bus data, collecting rocket 6DOF information, controlling an actuating mechanism to telemeter information, sampling overload sensor information and temperature sensor information, sampling engine pipeline and valve pressure information at 1KHz frequency, the health level of the rocket is comprehensively evaluated by the test data on information such as engine process parameter diagnosis, rocket thrust, axial overload, rocket body posture and the like, in particular to a rocket health degree evaluation method based on a long-time memory self-encoder, evaluating the acquired information, namely the health degree of the rocket, performing remedial judgment according to the abnormal condition and the abnormal grade when the abnormality is found, performing online track planning according to the current state and the residual propellants if necessary, and uploading the diagnosis result to the comprehensive control computer 201 in an FC-AE-1553_ NTNC mode under the control of the comprehensive control computer 201 for subsequent decision making.
Further, every 20ms period, the data management center 205 performs telemetry code modulation and downloading of the received data message.
Preferably, every 100ms period, the integrated control computer 201 obtains a rocket real-time GNSS signal, performs combined navigation, and readjusts a rocket navigation result.
In particular, due to the high reliability and non-intervention characteristics of the launch vehicle, the following emergency treatment measures are adopted for the FC-AE-1553 high-speed bus electrical system:
when receiving data, the NC device performs data confirmation comparison on the A/B dual redundant channels, does not process the data when the data are consistent, and can start a fault retransmission mechanism to retransmit the last bus session when the data are inconsistent. When the session process of FC-AE-1553_ NTNC, FC-AE-1553_ NCNT and the like is overtime (a response state frame is not received within a specified time), a retransmission mechanism is started (retransmission is carried out twice), a bus protocol layer state machine heartbeat signal is detected, if the state machine is abnormal, the current network state, namely a recording state, is recorded, NC is reinitialized, remote initialization is started for access bus network equipment (according to the recording state), and network communication is continued.
When the NT equipment receives data, data confirmation comparison is carried out on the A/B dual redundant channels, when the data are consistent, processing is not carried out, when the data are inconsistent, a fault retransmission mechanism can be applied, and the NC equipment retransmits a request bus conversation after receiving the request through timing polling. If the NT end receives or sends data periodically, if the time exceeds 2 times of the period and the data is not received or sent externally, the NT end requests the remote end to initialize, detects a heartbeat signal of a state machine of a bus protocol layer, if the state machine is abnormal, and reinitializes the NT end to continue communication.
The carrier rocket electrical system based on the FC-AE-1553 high-speed bus has the characteristics of reliable information interaction, large interaction capacity, high equipment integration level, support of a health diagnosis system with high sampling rate telemetry and the like, and has stronger advancement.
In the data management center 205, the rocket health degree evaluation method based on the long-time and short-time memory self-encoder is adopted, and the evaluation of the rocket health degree comprises the following steps:
s1, measuring and collecting original data in the starting process of the carrier rocket engine, and then preprocessing the original data;
s2, dividing the preprocessed data into a training set and a test set;
s3, establishing a long-time memory self-encoder model according to the dimension of the original data, and initializing model parameters;
s4, performing multi-generation training on the long-time and short-time memory self-encoder model by using a training set, so that the loss of the long-time and short-time memory self-encoder model is converged, and obtaining a trained long-time and short-time memory self-encoder model after a preset training generation number is reached;
and S5, calculating the output of the test set by utilizing the trained long-time memory autoencoder model, and then calculating the health degree of the launch vehicle engine according to the test set and a preset threshold value.
The invention adopts a rocket health degree evaluation method based on a long-time memory self-encoder, and the preferred scheme for evaluating the rocket health degree is as follows:
step 1: and inputting the original engine performance data obtained by measurement, standardizing, reserving standardized parameters, and changing a data format to meet the subsequent input requirements.
The raw data includes: main turbopump speed (nt), kerosene primary pump outlet pressure (Pepf1), thrust chamber ignition circuit pressure (Pigc), ignition conduit forward pressure (Piti).
The storage format of the raw data is shown in table 1.
TABLE 1
| Name of field | Letter code | Data type | Unit of |
| Main turbine pump speed | Nt | Float | r/m |
| Outlet pressure of first-stage kerosene pump | Pepf1 | Float | MPa |
| Pressure of ignition path of thrust chamber | Pigc | Float | MPa |
| Ignition tube front pressure | Piti | Float | MPa |
The raw data examples are shown in table 2.
TABLE 2
| Pepf1 | Nt | Piti | Pigc |
| 0.36 | 0 | 0.54 | 0.16 |
| 0.36 | 0 | 0.53 | 0.16 |
| 0.36 | 0 | 0.53 | 0.16 |
| 0.36 | 0 | 0.53 | 0.16 |
| 0.37 | 0 | 0.53 | 0.16 |
The normalization step employs a min-max normalization method, namely:
wherein x' is the normalized data sample, x is the data sample before normalization, xminIs the minimum of x, xmaxIs in xA maximum value.
The normalized parameter is xminAnd xmaxThereafter, the normalization process for the sample to be measured also uses these two values instead of the corresponding values of the sample to be measured.
The modified data format specifically includes: the original input data format is (n,4), n is the number of samples, 4 is the characteristic dimension, namely four parameter indexes of the main turbine pump rotating speed, the kerosene first-stage pump outlet pressure, the thrust chamber ignition circuit pressure and the ignition guide pipe front pressure. The input format required by the long and short term memory unit is (samples, time, input _ dim), wherein: samples is the number of samples, namely n; input _ dim is the feature dimension, i.e., 4; the time is a time step and can be set according to requirements, the time step is 1, namely the original data format is changed into (n,1,4), if a longer time step needs to be set, the required time step is only needed to be transmitted when the function in the data input module is called, the corresponding data format is (n-t +1, t,4), wherein t is the time step.
Assuming that the input data is [ [1,2], [3,4], [5,6], [7,8] ], and the required time step is 2, the data after changing the data format is: [[[1,2],[3,4]],[[3,4],[5,6]],[[5,6],[7,8]]].
Step 2: all original data are divided into a training set and a test set, normal original data are selected as the training set, and data with measurement errors are selected as the test set. If all the data are normal data, the training and testing sets can be divided according to a certain proportion.
And step 3: an automatic long-time and short-time memory encoder model is established according to an input data format (the number of original data indexes), matrix weights are initialized randomly, an average absolute error is selected as a loss function, and an Adam optimizer is selected as an optimizer.
The long and short time memory automatic encoder model structure and the parameter number are shown in Table 3
TABLE 3
In the model, the lstm (long short term memory) layer adopts a Relu function as an activation function, and the formula is as follows:
where x is the output of the last long-short memory cell and y is the input passed to the next long-short memory cell.
The calculation formula of each value in the long-time and short-time memory unit structure is as follows:
ft＝σ(Wf·[ht-1,xt]+bf) ⑷
it＝σ(Wi·[ht-1,xt]+bi) ⑸
ot＝σ(Wo·[ht-1,xt]+bo) ⑺
ht＝ot×tanh(Ct) ⑻
wherein f istIs a forgetting gate, is a vector in which each element is located at [0,1 ]]Within the range for Ct-1Which features of C are to be used in calculating Ct. Using sigmoid function as activation function (i.e.. sigma.)
Representing cell state update values, from input data xtAnd hidden node ht-1Activation function for updating cell state obtained via a neural network layerTanh (. cndot.) is often used. i.e. itCalled entry gate, again with each element at [0,1 ]]Vectors within range for control
Which features in C are used to update Ct。otIs an output gate, again with each element being located at [0,1 ]]Vectors within the range, which are related to CtCombined with the output h for calculating the long and short time memory cellt。
The loss function is:
where num is the number of samples, y1 is the final model output, and x1 is the raw data, i.e., the model input
The number of training parameters required for each LSTM layer in table 3 satisfies the following formula:
Param＝4×((input_dim+output_dim)×output_dim+output_dim) ⑽
wherein input _ dim is an input feature dimension, output _ dim is an output feature dimension, the first LSTM layer is used as an example, the input dimension is 4, the output dimension is 16, and the number of parameters required to be trained is as follows:
4×((4+16)×16+16)＝1344
in Table 3, InputLayer is the input layer, representing the input data, so there are no training parameters.
The Repeatvector layer in table 3 is a layer for sorting data format, so there is no training parameter.
In table 3, the timedistributed layer is essentially a full connection layer, and the calculation formula of the number of parameters is as follows:
Param＝input_dim×timestep×output_dim+output_dim×timestep ⑾
the method specifically comprises the following steps:
16×4×1+4×1＝68
and 4, step 4: training the constructed model by using the processed original data training as a training sample, and calculating the output of the model to the original data.
The training process adopts batch training, the batch size is 32, namely 32 samples are taken for training each time, and the corresponding data format is (32,1, 4).
In the training process, the model is trained for multiple generations, the default training generation is 5 generations, and after 3 generations of training, model loss is converged.
The training process uses a back propagation algorithm, which can be expressed as
Where W is a weight matrix: the result is the evaluation operation, eta is the learning rate,
is the gradient of the loss function to the weight matrix.
And 5: the absolute error of the final output from the original input is calculated, i.e. loss is calculated.
Step 6: and (3) calculating each neuron (the long and short time memory unit is essentially a multilayer neural network) by using a back propagation algorithm to update each neuron matrix parameter, judging whether a preset training algebra is completed, if so, entering a step 7, and otherwise, entering a step 4.
And 7: and (5) calculating the output y of the original data to be measured by applying the model trained in the steps 4 to 6.
And 8: calculating the absolute value of the difference between the output and the original data to be measured, namely:
difference1＝|y-x| ⒀
wherein x and y are both (n,1,4)
And step 9: subtracting the preset threshold value from the difference1 to obtain the difference between the preset threshold value and the threshold value, namely:
diffenrece2＝difference1-threshold ⒁
the threshold value is a division threshold value for abnormal data and normal data, and represents tolerance to data fluctuation, and the smaller the value, the more sensitive the data fluctuation is.
The threshold value in this embodiment is obtained by fitting the output of the normal data. The method specifically comprises the following steps: after the model training is finished, obtaining the output of all normal data, and solving the mean value mu and the variance sigma2. The threshold is then fitted using the following formula.
In this embodiment, the more 0.01423 the value calculation result is.
Step 10: the difference2 is passed into the health calculation function to obtain the health of the original data.
The health degree calculation function is a modified sigmoid function, and because data are standardized and have small values, the finally calculated absolute value of the error and the difference value between the absolute value of the error and the threshold are also small, scaling processing is performed before the health degree is calculated by the function. The scaling processing function is as follows:
target＝(1(difference2)*w+b)*(difference2) ⒂
w and b are preset weight matrixes, and data below the threshold and data above the threshold can be respectively subjected to expansion processing by combining with 1(difference2 is less than 0). Wherein 1(difference2 is less than 0) is specifically as follows:
the final health calculation function is as follows:
wherein difference2 has the shape of (n,1,4), and i, j, and w are subscripts of three dimensions.
Step 11: and (3) utilizing matplotlib and tkater to make a UI interface to display the curve of the original data to be tested in real time and evaluate the real-time health.
After the model training is completed, the results of the test using a group of normal data and two groups of different abnormal data are shown in table 4 (test data average health summary table) and table 5.
TABLE 4
Table 5 the model serves as a summary table of the classifier prediction accuracy, i.e. it is verified using normal data and two kinds of abnormal data, respectively. The data with the health degree lower than 90 is taken as abnormal data, the accuracy of the abnormal data and the accuracy of the normal data judged according to the method are respectively tested, and the specific values are shown in the following table.
TABLE 5
Those skilled in the art will appreciate that those matters not described in detail in the present specification are well known in the art.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the present invention, and those skilled in the art can make variations and modifications of the present invention without departing from the spirit and scope of the present invention by using the methods and technical contents disclosed above.
Claims (7)
1. A test method of a carrier rocket electrical system based on an FC-AE-1553 high-speed bus is characterized in that the electrical system at least comprises a ground host, a comprehensive control computer, a comprehensive actuator and a data management center, and the test method comprises the following steps:
s1, starting a test program, initializing a ground host into a network controller of the FC-AE-1553 high-speed bus, and initializing other equipment of an electric system into network terminal equipment of the FC-AE-1553 high-speed bus; the ground host sequentially completes initialization handshake of equipment accessed into the FC-AE-1553 high-speed bus network, binding of test data of the comprehensive control computer 201 and sending of test instructions;
s2, after the test indexes of each piece of equipment on the rocket ground are qualified, the ground host sends a switching control right instruction to the comprehensive control computer, and the comprehensive control computer receives the switching control right instruction and then re-initializes the network controller of the FC-AE-1553 high-speed bus to receive the network control right;
s3, the ground host sends a primary engine ignition instruction to the comprehensive actuator, and the comprehensive actuator performs ignition on the engine according to the ignition timing of the engine; the integrated control computer dynamically monitors a rocket takeoff signal, broadcasts the takeoff time to the FC-AE-1553 high-speed bus globally after receiving the takeoff, finishes the time correction of the whole system, and enters navigation guidance and rocket attitude control;
s4, the data management center collects the measurement data of the carrier rocket, then evaluates the health degree of the rocket, and sends the evaluation result to the comprehensive control computer;
in S4, a long-time memory-based self-encoder is adopted to evaluate the health degree of the rocket, and the evaluation method comprises the following steps:
s41, measuring and collecting original data in the starting process of the carrier rocket engine, and then preprocessing the original data;
s42, dividing the preprocessed data into a training set and a test set;
s43, establishing a long-time memory self-encoder model according to the dimension of the original data, and initializing model parameters;
s44, performing multi-generation training on the long-time and short-time memory self-encoder model by using a training set, so that the loss of the long-time and short-time memory self-encoder model is converged, and obtaining a trained long-time and short-time memory self-encoder model after a preset training generation number is reached;
s45, calculating the output of the test set by utilizing the trained long-time memory autoencoder model, and then calculating the health degree of the launch vehicle engine according to the test set and a preset threshold value;
the system comprises a data management center, a data processing center and a data processing center, wherein the data management center monitors bus data to collect rocket information in a monitoring terminal mode, controls an actuating mechanism to remotely measure the information, samples overload sensor information and temperature sensor information, samples engine pipeline and valve pressure information at a certain frequency, comprehensively evaluates the health level of the rocket through test data on engine process parameter diagnosis, rocket thrust, axial overload and rocket body posture, specifically evaluates the acquired information, namely the health level of the rocket by adopting a rocket health degree evaluation method based on a long-time memory self-encoder, carries out remediation judgment according to abnormal conditions and abnormal levels when an abnormality is found, and carries out online track planning according to the current state and residual propellants;
when receiving data, the network controller performs data confirmation comparison on the A/B dual redundant channel, does not process the data when the data are consistent, and starts a fault retransmission mechanism to retransmit the last bus session when the data are inconsistent; overtime happens in the session process of FC-AE-1553_ NTNC and FC-AE-1553_ NCNT, a retransmission mechanism is started, a state machine heartbeat signal of a bus protocol layer is detected, if the state machine is abnormal, the current network state is recorded, the recording state is called as the recording state for short, a network controller is reinitialized, remote initialization is started for access bus network equipment, and network communication is continued.
2. The method for testing the electric system of the launch vehicle based on the FC-AE-1553 high-speed bus according to the claim 1, wherein the preprocessing in the S41 adopts a min-max standardization method, namely:
wherein x' is the normalized data sample, x is the data sample before normalization, xminIs the minimum value of the x values, and is,xmaxis the maximum value in x.
3. The test method for the electric system of the launch vehicle based on the FC-AE-1553 high-speed bus according to claim 1, wherein the method for dividing the training set and the test set in S42 comprises the following steps: and selecting normal original data as a training set, and selecting data with measurement errors as a test set.
4. The test method for the electric system of the launch vehicle based on the FC-AE-1553 high-speed bus according to any one of claims 1-3, wherein the long-time memory self-encoder model comprises a plurality of encoding layers and a plurality of decoding layers; the coding layer is used for extracting the characteristics of the data samples; the decoding layer is used to restore the extracted features to a matrix consistent with the input sample format.
5. A carrier rocket electrical system testing method based on an FC-AE-1553 high-speed bus according to any one of claims 1-3, characterized in that in S44, during multi-generation training, each generation is divided into verification sets from the training sets for checking the training effect of the long-time memory self-encoder model and for judging whether the long-time memory self-encoder model is over-fitted.
6. A carrier rocket electrical system testing method based on an FC-AE-1553 high-speed bus according to any one of claims 1-3, characterized in that the trained long-time memory self-encoder model is calculated to calculate the output of a test set, and after the absolute value of the difference between the output of the test set and the test set is subtracted, a preset threshold value is subtracted, and finally the carrier rocket engine health degree is calculated.
7. A carrier rocket electrical system based on an FC-AE-1553 high-speed bus is characterized by comprising an rocket comprehensive control computer, inertia measurement equipment, a data management center, a comprehensive actuator, a servo controller, a driving mechanism and a ground host; the carrier rocket electrical system adopts an FC-AE-1553 high-speed bus for data communication;
the rocket comprehensive control computer is used for autonomously controlling the rocket, is used as a network controller of an FC-AE-1553 high-speed bus, and controls a comprehensive actuator, a servo controller and a driving mechanism;
the system comprises an inertial measurement device, a comprehensive actuator, a servo controller, a driving mechanism, a rocket engine nozzle, a ground host and a control system, wherein the inertial measurement device is used for measuring inertial parameters, the comprehensive actuator is used for controlling the working time sequence of the rocket, the servo controller and the driving mechanism are used for controlling the rocket engine nozzle to swing, and the ground host is used for rocket-ground information interaction, control right switching, ignition sending and real-time data interpretation;
the inertia measurement equipment, the servo controller and the driving mechanism send data to the arrow comprehensive control computer;
the data management center is used for collecting data of the carrier rocket and modulating a telemetering signal, and meanwhile, a long-time memory-based self-encoder is adopted to diagnose and evaluate the state of the rocket and send an evaluation result to the comprehensive control computer;
a long-time memory-based self-encoder is adopted to evaluate the health degree of the rocket, and the evaluation method comprises the following steps:
s101, measuring and collecting original data in the starting process of a carrier rocket engine, and then preprocessing the original data;
s102, dividing the preprocessed data into a training set and a test set;
s103, establishing a long-time memory self-encoder model according to the dimension of the original data, and initializing model parameters;
s104, performing multi-generation training on the long-time and short-time memory self-encoder model by using a training set, so that the loss of the long-time and short-time memory self-encoder model is converged, and obtaining a trained long-time and short-time memory self-encoder model after a preset training generation number is reached;
s105, calculating the output of a test set by utilizing the trained long-time memory autoencoder model, and then calculating the health degree of the engine of the carrier rocket according to the test set and a preset threshold value;
the inertia measurement equipment, the comprehensive actuator, the servo controller and the driving mechanism are all used as network terminal equipment of an FC-AE-1553 high-speed bus; the ground host serves as a network controller of an FC-AE-1553 high-speed bus;
the system comprises a data management center, a data processing center and a data processing center, wherein the data management center monitors bus data to collect rocket information in a monitoring terminal mode, controls an actuating mechanism to remotely measure the information, samples overload sensor information and temperature sensor information, samples engine pipeline and valve pressure information at a certain frequency, comprehensively evaluates the health level of the rocket through test data on engine process parameter diagnosis, rocket thrust, axial overload and rocket body posture, specifically evaluates the acquired information, namely the health level of the rocket by adopting a rocket health degree evaluation method based on a long-time memory self-encoder, carries out remediation judgment according to abnormal conditions and abnormal levels when an abnormality is found, and carries out online track planning according to the current state and residual propellants;
when receiving data, the network controller performs data confirmation comparison on the A/B dual redundant channel, does not process the data when the data are consistent, and starts a fault retransmission mechanism to retransmit the last bus session when the data are inconsistent; overtime happens in the session process of FC-AE-1553_ NTNC and FC-AE-1553_ NCNT, a retransmission mechanism is started, a state machine heartbeat signal of a bus protocol layer is detected, if the state machine is abnormal, the current network state is recorded, the recording state is called as the recording state for short, a network controller is reinitialized, remote initialization is started for access bus network equipment, and network communication is continued.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010704225.0A CN112130543B (en) | 2020-07-21 | 2020-07-21 | Carrier rocket electrical system based on FC-AE-1553 high-speed bus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010704225.0A CN112130543B (en) | 2020-07-21 | 2020-07-21 | Carrier rocket electrical system based on FC-AE-1553 high-speed bus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112130543A CN112130543A (en) | 2020-12-25 |
| CN112130543B true CN112130543B (en) | 2022-02-08 |
Family
ID=73850514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010704225.0A Active CN112130543B (en) | 2020-07-21 | 2020-07-21 | Carrier rocket electrical system based on FC-AE-1553 high-speed bus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112130543B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112947041A (en) * | 2021-02-18 | 2021-06-11 | 上海航天控制技术研究所 | High-reliability dual-redundancy voting method for front-end and rear-end hosts of carrier rocket |
| CN113008270B (en) * | 2021-02-22 | 2022-12-27 | 上海航天控制技术研究所 | Ten-meter optical fiber inertia combination precision testing method based on three-bus interface |
| CN114826820A (en) * | 2022-03-31 | 2022-07-29 | 中国电子科技集团公司第五十四研究所 | FC-AE-1553 network service scheduling method and equipment under centralized control |
| CN114777580B (en) * | 2022-04-13 | 2022-09-02 | 东方空间技术(北京)有限公司 | Test method and device of rocket bus switcher and rocket test system |
| CN114992001A (en) * | 2022-05-25 | 2022-09-02 | 湖北三江航天红峰控制有限公司 | Control method and control circuit of aircraft engine safety mechanism |
| CN115065400B (en) * | 2022-06-13 | 2024-02-09 | 上海宇航系统工程研究所 | Down-frequency communication method and telemetry system for carrier rocket |
| CN115333988B (en) * | 2022-10-13 | 2023-01-24 | 东方空间技术(北京)有限公司 | Test method, system and equipment for rocket interstage communication signals |
| CN115540701B (en) * | 2022-11-07 | 2023-04-14 | 东方空间技术(山东)有限公司 | Carrier rocket distributed test system and test method based on 5G network |
| CN116045751A (en) * | 2023-03-07 | 2023-05-02 | 东方空间技术(山东)有限公司 | Carrier rocket state testing method, carrier rocket state testing system, computer equipment and carrier rocket state testing medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015076804A (en) * | 2013-10-10 | 2015-04-20 | セイコーエプソン株式会社 | Functional device, electronic apparatus, movable body, synchronous control system, operation method of functional device and synchronous control method |
| CN105868890A (en) * | 2016-03-24 | 2016-08-17 | 中国人民解放军海军航空工程学院 | Historical information-based health state assessment method for solid rocket engine |
| CN106598060A (en) * | 2015-10-14 | 2017-04-26 | 上海宇航系统工程研究所 | Carrier rocket information integration electrical system |
| CN110207997A (en) * | 2019-07-24 | 2019-09-06 | 中国人民解放军国防科技大学 | Liquid rocket engine fault detection method based on convolution self-encoder |
| CN111306997A (en) * | 2020-03-23 | 2020-06-19 | 北京中科宇航技术有限公司 | Electric system and electric control method of carrier rocket |
-
2020
- 2020-07-21 CN CN202010704225.0A patent/CN112130543B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015076804A (en) * | 2013-10-10 | 2015-04-20 | セイコーエプソン株式会社 | Functional device, electronic apparatus, movable body, synchronous control system, operation method of functional device and synchronous control method |
| CN106598060A (en) * | 2015-10-14 | 2017-04-26 | 上海宇航系统工程研究所 | Carrier rocket information integration electrical system |
| CN105868890A (en) * | 2016-03-24 | 2016-08-17 | 中国人民解放军海军航空工程学院 | Historical information-based health state assessment method for solid rocket engine |
| CN110207997A (en) * | 2019-07-24 | 2019-09-06 | 中国人民解放军国防科技大学 | Liquid rocket engine fault detection method based on convolution self-encoder |
| CN111306997A (en) * | 2020-03-23 | 2020-06-19 | 北京中科宇航技术有限公司 | Electric system and electric control method of carrier rocket |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112130543A (en) | 2020-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112130543B (en) | Carrier rocket electrical system based on FC-AE-1553 high-speed bus | |
| CN111290366B (en) | Multi-fault diagnosis method for attitude control system of spacecraft | |
| CN101697079B (en) | Blind system fault detection and isolation method for real-time signal processing of spacecraft | |
| CN108828944B (en) | Encoder fault diagnosis system and method based on improved PSO and SVM | |
| CN109902832A (en) | Training method, predicting abnormality method and the relevant apparatus of machine learning model | |
| CN114564000B (en) | Active fault tolerance method and system based on intelligent aircraft actuator fault diagnosis | |
| CN113311803B (en) | On-orbit spacecraft flywheel fault detection method based on kernel principal component analysis | |
| CN108540311B (en) | Fault detection deep learning network processing method and device of satellite actuating mechanism | |
| US20200066062A1 (en) | Method and system for vehicle analysis | |
| CN111523254B (en) | Vehicle verification platform with adjustable control characteristics and implementation method | |
| CN110412997B (en) | Spacecraft attitude control spray pipe fault intelligent diagnosis system and method based on neural network | |
| CN113378351B (en) | On-line intelligent field removing method for satellite attitude sensor measurement data | |
| KR102182226B1 (en) | Failure Detection-Diagnosis System and Method using Thereof | |
| CN113311714A (en) | Fault diagnosis and fault-tolerant control method and system for multi-joint bionic robot fish sensor | |
| CN111191770B (en) | Rocket system health state assessment method based on fuzzy neural network | |
| CN110287594B (en) | Aero-engine state diagnosis method based on neural network algorithm | |
| CN112766408B (en) | Aircraft micro fault diagnosis method based on principal component analysis and multilayer overrun learning machine | |
| CN113108779B (en) | Independent arrow measurement system based on 1553B bus triple redundancy | |
| CN114348291A (en) | Flight fault diagnosis method based on flight parameter data and simulation | |
| CN111857098A (en) | Fault diagnosis method of gas turbine electric actuator based on information statistical analysis | |
| CN112560252A (en) | Prediction method for residual life of aircraft engine | |
| Tran et al. | Developing an Approach for Fault Detection and Diagnosis of Angular Velocity Sensors | |
| CN111284497A (en) | Driving state recognition device | |
| CN116700203B (en) | Fault detection and isolation method for carrier rocket attitude control system | |
| CN116150683A (en) | Inertial measurement unit redundancy diagnosis method based on interaction multiple models |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |