CN112529248A - Data-driven intelligent flight space-ground mirror image system of carrier rocket - Google Patents
Data-driven intelligent flight space-ground mirror image system of carrier rocket Download PDFInfo
- Publication number
- CN112529248A CN112529248A CN202011240138.0A CN202011240138A CN112529248A CN 112529248 A CN112529248 A CN 112529248A CN 202011240138 A CN202011240138 A CN 202011240138A CN 112529248 A CN112529248 A CN 112529248A
- Authority
- CN
- China
- Prior art keywords
- rocket
- fault
- data
- fault diagnosis
- arrow
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0751—Error or fault detection not based on redundancy
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
-
- 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
Abstract
The invention discloses a data-driven intelligent flying space and ground mirror image system of a carrier rocket, which comprises: the data transceiver module is used for storing the received remote measurement parameters on the rocket to the platform database; meanwhile, the remote measurement parameters on the arrow are sent to a fault diagnosis module; the fault diagnosis module is used for carrying out fault diagnosis according to the remote measurement parameters on the arrow and storing fault information into the platform database; the trend prediction module is used for storing a flight trend prediction result to the platform database according to the subsequent flight trend of the predicted rocket; and the visual display module is used for displaying the rocket telemetry parameters, the fault information and the flight trend prediction results acquired from the platform database. The invention truly reproduces the running state of the rocket, so that the launcher can better master the task progress status; meanwhile, the computing resources on the ground are fully utilized, potential faults can be detected on the arrow and the follow-up trend can be predicted, and manual uplink interference instructions can be carried out if necessary.
Description
Technical Field
The invention belongs to the technical field of fault diagnosis of carrier rockets, and particularly relates to a data-driven intelligent flying space-ground mirror image system of a carrier rocket.
Background
At present, advanced carrier rocket models, particularly manned models, are arranged outside China, and the fault detection and reconstruction functions are arranged on the rocket, which is a necessary trend for the development of future aerospace in China. However, due to the limitation of hardware resources, complicated operations are difficult to be carried out on the arrows, and the threshold judgment method is mainly used, so that the detection and prediction capabilities are insufficient. Meanwhile, the ground can realize more complex and accurate calculation and prediction by relying on strong computing power and historical data advantages.
With the development of the internet of things, big data and artificial intelligence, the digital twin technology has gained more and more extensive attention in recent years. The digital twin body is used as a virtual model, and the behavior of the physical entity in the real environment is simulated through means of virtual-real interaction feedback, data fusion analysis, decision iterative optimization and the like, so that new capability is developed for the physical entity. By using data generated by various sensors on the rocket and transmitted to the ground through a telemetry system, a digital twin can be created to synchronize the state of the real-time rocket, detect potential faults and give an auxiliary decision.
Disclosure of Invention
The technical problem of the invention is solved: the intelligent flying space-ground mirror system of the data-driven carrier rocket is integrated with a digital twinning technology concept, and realizes synchronous work of a virtual rocket and a real rocket on the ground through a data transceiver module, a fault diagnosis module, a trend prediction module and a visual display module, so that the running state of the rockets is truly reproduced, and a launcher can better master the task progress condition; meanwhile, the computing resources on the ground are fully utilized, potential faults can be detected on the arrow and the follow-up trend can be predicted, and manual uplink interference instructions can be carried out if necessary.
In order to solve the technical problem, the invention discloses a data-driven intelligent flying space and ground mirror system of a carrier rocket, which comprises:
the data transceiver module is used for receiving the on-arrow telemetering parameters sent by the telemetering system ground detection station and storing the received on-arrow telemetering parameters into the platform database; meanwhile, sending the received remote measurement parameters on the arrow to a fault diagnosis module;
the fault diagnosis module is used for carrying out fault diagnosis according to the rocket telemetry parameters sent by the data transceiver module and storing the fault information obtained by diagnosis into the platform database;
the trend prediction module is used for acquiring the rocket telemetering parameters from the platform database, predicting the subsequent flight trend of the rocket according to the rocket telemetering parameters and storing the obtained flight trend prediction result into the platform database;
the visual display module is used for acquiring the rocket telemetry parameters, the fault information and the flight trend prediction result from the platform database; and displaying the acquired on-rocket telemetering parameters, fault information and flight trend prediction results in two modes of three-dimensional vision and data browsing.
In the data-driven intelligent flying space and ground mirror system of the launch vehicle, the data transceiver module is further configured to: when a fault occurs, fault information is obtained from the platform database, the obtained fault information is forwarded to the fault detection processor on the arrow, and the fault detection processor judges the reasonability and then sends the fault information to the arrow machine for reconstruction processing on the arrow.
In the data-driven intelligent flying space-ground mirror image system of the carrier rocket, when the fault diagnosis module carries out fault diagnosis according to the rocket telemetering parameters sent by the data transceiver module, the fault diagnosis method comprises the following steps: and according to the on-arrow telemetering parameters sent by the data transceiver module, fault diagnosis is carried out through any one of a dynamic model, an engine rule and a data correlation model.
In the data-driven intelligent flying space-ground mirror image system of the carrier rocket, the data correlation model is a fault diagnosis model based on a neural network; the fault diagnosis model based on the neural network takes a feature vector formed by the on-arrow telemetry parameters at the current moment and the on-arrow telemetry parameters at 4 time nodes before the current moment as input; the output of the fault diagnosis model based on the neural network is the probability distribution of the rocket flight fault at the current moment.
In the above data-driven intelligent flying space and ground mirror system of a launch vehicle, the fault diagnosis model based on the neural network comprises: the hidden layer comprises a first hidden layer, a second hidden layer, a third hidden layer and a softmax layer;
adopting tanh activation functions between the input of the fault diagnosis model based on the neural network and the first hidden layer and between the first hidden layer and the second hidden layer;
a relu activation function is adopted between the second hidden layer and the third hidden layer;
and after the output of the third hidden layer passes through the softmax layer, obtaining the probability distribution of the rocket flight fault at the current moment.
In the data-driven intelligent flying space and ground mirror image system of the carrier rocket, the trend prediction module predicts the subsequent flying trend of the rocket according to the remote measurement parameters on the rocket, and the method comprises the following steps: and predicting the subsequent flight trend of the rocket through a dynamic model and/or a flight trend model based on a neural network according to the on-rocket telemetry parameters.
In the data-driven intelligent flying space and ground mirroring system of the launch vehicle, the contents displayed by the visual display module comprise: the system comprises rocket motion, rocket flight attitude, rocket orbital transfer process, engine motion, rocket flight state-time curve, rocket engine thrust-time curve, video monitoring information, parameter information and fault alarm information.
The invention has the following advantages:
(1) the invention discloses a data-driven intelligent flying space-ground mirror image system of a carrier rocket.A three-dimensional visual picture displayed by a visual display module is driven by real-time telemetering data, so that the current state of the rocket can be truly reproduced.
(2) The invention discloses a data-driven intelligent flying space-ground mirror image system of a carrier rocket.A fault diagnosis module integrates multiple fault diagnosis methods based on data, models, neural networks and the like, and improves the detection rate.
(3) The invention discloses a data-driven intelligent flying space-ground mirror image system of a carrier rocket, which can predict the subsequent flying trend in real time according to the preorder flying data.
(4) The invention discloses a data-driven intelligent flying space-ground mirror image system of a carrier rocket, which adopts a universal design idea, can be quickly applied to different models, and can also quickly increase a fault mode and a diagnosis algorithm; the ground health management system can be applied to ground health management systems of carrier rockets of any types.
Drawings
FIG. 1 is a block diagram of a data-driven intelligent flying space and ground mirror system of a launch vehicle according to an embodiment of the invention;
FIG. 2 is a block diagram of a neural network based fault diagnosis model according to an embodiment of the present invention;
fig. 3 is a structural block diagram of a flight trend model based on a neural network in an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The invention discloses a data-driven intelligent flying space-ground mirror image system of a carrier rocket, which mainly comprises a data transceiver module, a fault diagnosis module, a trend prediction module and a visual display module, and can receive real-time rocket-mounted telemetering parameters downloaded by a rocket through a telemetering system as system input; the real-time rocket telemetry parameters are simultaneously transmitted to the fault detection module and the visual display module, the current flight attitude and the flight trend of the rocket can be visually presented on the ground, when a fault occurs, detection can be completed faster and more accurately by an algorithm based on data, a model and a neural network in the fault detection module, a fault diagnosis result and a fault initial state are transmitted to the trend prediction module, the subsequent development state of the simulated fault is accelerated, and whether a reconstruction strategy is feasible or not is verified. The system adopts a universal design, and can provide ground service in a flight phase for different models by loading different dynamic models, three-dimensional models and fault diagnosis prediction models, and the system is used as a powerful support for fault detection and reconstruction.
Referring to fig. 1, in the present embodiment, the data-driven intelligent flying space and ground mirroring system of a launch vehicle comprises:
the data transceiver module 101 is used for receiving the on-arrow telemetry parameters sent by the ground detection station of the telemetry system and storing the received on-arrow telemetry parameters into the platform database; meanwhile, sending the received remote measurement parameters on the arrow to a fault diagnosis module; and when a fault occurs, acquiring fault information from the platform database, forwarding the acquired fault information to the fault detection processor on the arrow, and sending the fault information to the arrow machine for reconstruction processing on the arrow after the fault detection processor judges the reasonability.
And the fault diagnosis module 102 is used for performing fault diagnosis according to the rocket telemetry parameters sent by the data transceiver module, and storing the diagnosed fault information in the platform database.
In this embodiment, the fault diagnosis module 102 may load different fault diagnosis algorithms according to the interface convention, and has a relatively strong extensibility. For example, fault diagnosis algorithms include, but are not limited to: a dynamic model based fault diagnosis algorithm, an engine rule based fault diagnosis algorithm, and a data correlation model based fault diagnosis algorithm. The fault diagnosis algorithm may be integrated in the fault diagnosis module 102 in a plug-in form, and when performing fault diagnosis in the fault diagnosis module 102, any one or more plug-in of the fault diagnosis algorithm may be called according to actual requirements to perform fault diagnosis.
Preferably, the data correlation model may be a neural network based fault diagnosis model. Due to the difference of rocket operating environments, the fault type not only depends on the operating state of the rocket when the fault occurs, but also needs to judge a time sequence before the fault occurrence time. Therefore, the fault diagnosis model based on the neural network takes the feature vector formed by the on-arrow telemetering parameters at the current moment and the on-arrow telemetering parameters at 4 time nodes before the current moment as input; the output of the fault diagnosis model based on the neural network is the probability distribution of the rocket flight fault at the current moment.
See FIG. 2, t1、t2、……、tnEtc. represent the state vector of the rocket running at a certain moment. During training, the state vectors at 5 continuous moments are combined into a one-dimensional tensor which is used as the input of the model. For example, if the state vector at each time contains 18 parameters, the input to the neural network is a 90-dimensional vector. The fault diagnosis model based on the neural network adopts three hidden layers (a first hidden layer, a second hidden layer and a third hidden layer), and the number of nodes of the hidden layers is 2-3 times of that of input vectors. And the output result of the hidden layer passes through a softmax layer to obtain probability distribution maps of different failure modes. And selecting the fault mode with the highest possibility as the fault positioning result at the current moment according to the maximum likelihood estimation and outputting the fault mode.
In the neural network-based fault diagnosis model, there are:
the input of the fault diagnosis model of the neural network and the first hidden layer and the second hidden layer adopt tanh activation functions, and the mathematical expression is as follows:
a relu activation function is adopted between the second hidden layer and the third hidden layer to reduce the complexity of the model and reduce overfitting, and a mathematical expression is as follows:
and after the output of the third hidden layer passes through the softmax layer, obtaining the probability distribution of the rocket flight fault at the current moment, wherein the mathematical expression is as follows:
wherein, yiIs the output vector of the hidden layer, n is the dimension of the output vector of the hidden layer, and is also the total number of categories in the probability distribution graph output by the softmax layer.
It should be noted that the basic principle of the fault diagnosis algorithm based on the dynamic model in the fault diagnosis module 102 is as follows: and (3) reversely deducing the current (engine and servo mechanism) fault state information by using the navigation state quantity and rocket control instruction information and using a carrier rocket flight dynamics model. It should be noted that the dynamic model-based fault diagnosis algorithm directly diagnoses the fault type and the fault value, and performs fault diagnosis based on multi-model comparison, and the diagnosis process does not need specific criteria.
And the trend prediction module 103 is used for acquiring the rocket telemetry parameters from the platform database, predicting the subsequent flight trend of the rocket according to the rocket telemetry parameters, and storing the obtained flight trend prediction result in the platform database.
In this embodiment, the trend prediction module 103 may load different trend prediction algorithms according to the interface convention, and has a strong extension type. Preferably, the trend prediction algorithm may include: 1) trend prediction algorithm based on dynamic model: taking flight profile parameters and fault information (if existing) at a certain moment as input, simulating and quickly calculating a subsequent flight state, and if reconstruction operation occurs, predicting a reconstructed state according to a reconstruction model in the dynamic model. 2) Model-based trend prediction algorithms: the model in the model-based trend prediction algorithm may specifically refer to a flight trend model based on a neural network, and the subsequent flight trend is predicted after the input parameters are subjected to pattern recognition (in a normal state or in a certain fault state) by using the flight trend model based on the neural network obtained through training in the early stage.
Preferably, the neural network-based flight trend model uses a long-short term memory neural network (LSTM) model to implement the prediction work of the flight trajectory: 1) determining a training task: through the rocket flight state parameters of the first 30 moments (30 seconds), the flight path of the rocket at the next 30 moments (30 seconds) is predicted, namely the position x, the position y and the position z of the rocket in the launching coordinate system in the next 30 seconds. 2) The state of each time in the recurrent neural network is selectively influenced by information through a "gate" structure. 3) The Sigmoid activation function will output a value between 0 and 1, describing how much information the current input can go through the structure, and will function like a door.
The core of the long-short term memory neural network is an input gate, a forgetting gate and a state vector representing memory. The forgetting gate determines how much the previous memory needs to be forgotten according to the current input and the output at the last moment. If f takes 0 to indicate all forgetting, f takes 1 to indicate all remaining. The input gate determines how much new information needs to be added to the memory based on the current input and the output at the previous time. Through the input gate and the forgetting gate, the long-short term memory neural network can decide which information is forgotten and which information is reserved.
The forward propagation of the long-short term memory neural network is relatively complex, and the forward propagation process shown in fig. 3 can be expressed by using the following formula:
z＝tanh(Wz[hi-1,xt])
i＝sigmoid(Wi[hi-1,xt])
f＝sigmoid(Wf[hi-1,xt])
o＝sigmoid(Wo[hi-1,xt])
ci＝f·ci-1+i·z
hi＝o·tanh(ci)
wherein h isi-1Is the output of the previous moment, xtAs current input, ci-1For the memory of the previous moment, ciIs the memory of the current moment.
In addition, the trend prediction algorithm based on the dynamic model integrated in the trend prediction module 103 obtains the subsequent flight trend by taking the parameters at the current moment as the input profile of the dynamic model and performing accelerated iterative operation. And if the reconstruction occurs, replacing the model in the normal state with the reconstructed dynamic model.
The visual display module 104 is used for acquiring the rocket telemetry parameters, the fault information and the flight trend prediction result from the platform database; and displaying the acquired on-rocket telemetering parameters, fault information and flight trend prediction results in two modes of three-dimensional vision and data browsing.
In this embodiment, the visual display module 104 may include: the system comprises a three-dimensional view unit and a data browsing unit.
The three-dimensional view unit can be displayed by driving two-dimensional/three-dimensional views: the configuration of the mapping relation between external data and the three-dimensional view is realized through a TCP/IP interface and a configuration file, and the display effect of the carrier rocket in the three-dimensional view is configured in a three-dimensional CAD digital-analog importing mode; by operating a mouse or a keyboard, the definition, the configuration and the view angle switching of the view can be realized. The specific display content comprises:
rocket action: driving the rocket to perform corresponding action display according to the data, comprising the following steps: ignition, booster separation, primary separation, secondary separation, fairing separation, star-arrow separation and the like.
The flight attitude of the rocket is as follows: and driving the rocket to fly according to the data, and displaying the positions, the speeds and the postures of the rocket in the active section and the free flight section in real time.
The rocket orbital transfer process: and dynamically displaying the rocket orbital transfer process taking the earth as a reference object under the condition of abnormal engine.
The engine acts: and (5) driving to swing the rocket engine and move the servo mechanism according to the data, and animating the failure of the servo mechanism.
Scene interaction: the visual angle switching can be realized by operating through a mouse, a keyboard and the like.
The data browsing unit displays the content and comprises the following steps:
the curves show that: rocket flight states (position, speed, attitude, engine pivot angle, high wind area dynamic pressure, remaining shutdown time) -time curves to be displayed can be dynamically configured; the thrust-time curve of the rocket engine (primary/secondary) that needs to be displayed can be dynamically configured.
Video monitoring information: and a plurality of video floating windows are supported, and the configuration of newly adding, closing, dragging, stopping and transparency can be realized.
Parameter browsing: and displaying the parameter information.
And (5) fault alarm information.
In summary, the data-driven intelligent flight space-ground mirror system of the launch vehicle described in this embodiment adopts a universal design, and can customize different contents for different rocket models: rocket model, fault diagnosis algorithm, preset trajectory, action time sequence on the rocket, and the like. The rocket model, the preset trajectory and the action time sequence on the rocket are mainly used for visual display, and loading is realized under an initialization page of the visual display module. In addition, a new fault diagnosis function needs to be loaded in the inference engine, and for a data-driven diagnosis algorithm, a diagnosis model needs to be retrained and packaged based on newly-added application model simulation data.
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.
Those skilled in the art will appreciate that the invention may be practiced without these specific details.
Claims (7)
1. A data-driven intelligent flying space and ground mirror system of a launch vehicle is characterized by comprising:
the data transceiver module is used for receiving the on-arrow telemetering parameters sent by the telemetering system ground detection station and storing the received on-arrow telemetering parameters into the platform database; meanwhile, sending the received remote measurement parameters on the arrow to a fault diagnosis module;
the fault diagnosis module is used for carrying out fault diagnosis according to the rocket telemetry parameters sent by the data transceiver module and storing the fault information obtained by diagnosis into the platform database;
the trend prediction module is used for acquiring the rocket telemetering parameters from the platform database, predicting the subsequent flight trend of the rocket according to the rocket telemetering parameters and storing the obtained flight trend prediction result into the platform database;
the visual display module is used for acquiring the rocket telemetry parameters, the fault information and the flight trend prediction result from the platform database; and displaying the acquired on-rocket telemetering parameters, fault information and flight trend prediction results in two modes of three-dimensional vision and data browsing.
2. The data-driven space-ground intelligent flying mirror system for a launch vehicle of claim 1, wherein the data transceiver module is further configured to: when a fault occurs, fault information is obtained from the platform database, the obtained fault information is forwarded to the fault detection processor on the arrow, and the fault detection processor judges the reasonability and then sends the fault information to the arrow machine for reconstruction processing on the arrow.
3. The intelligent flying space-ground mirror system of data-driven carrier rocket according to claim 1, wherein the fault diagnosis module comprises, when performing fault diagnosis according to the rocket telemetry parameters sent by the data transceiver module: and according to the on-arrow telemetering parameters sent by the data transceiver module, fault diagnosis is carried out through any one of a dynamic model, an engine rule and a data correlation model.
4. The data-driven launch vehicle intelligent flying space-ground mirroring system of claim 3 wherein the data correlation model is a neural network based fault diagnosis model; the fault diagnosis model based on the neural network takes a feature vector formed by the on-arrow telemetry parameters at the current moment and the on-arrow telemetry parameters at 4 time nodes before the current moment as input; the output of the fault diagnosis model based on the neural network is the probability distribution of the rocket flight fault at the current moment.
5. The data-driven launch vehicle intelligent flying space-ground mirroring system of claim 4, wherein the neural network based fault diagnosis model comprises: the hidden layer comprises a first hidden layer, a second hidden layer, a third hidden layer and a softmax layer;
adopting tanh activation functions between the input of the fault diagnosis model based on the neural network and the first hidden layer and between the first hidden layer and the second hidden layer;
a relu activation function is adopted between the second hidden layer and the third hidden layer;
and after the output of the third hidden layer passes through the softmax layer, obtaining the probability distribution of the rocket flight fault at the current moment.
6. The data-driven vehicle rocket intelligent flying space and ground mirror system according to claim 1, wherein the trend prediction module predicts the subsequent flying trend of the rocket according to the on-rocket telemetry parameters, and comprises the following steps: and predicting the subsequent flight trend of the rocket through a dynamic model and/or a flight trend model based on a neural network according to the on-rocket telemetry parameters.
7. The data-driven intelligent flying space and ground mirroring system for launch vehicles according to claim 1, wherein the content displayed through the visual display module comprises: the system comprises rocket motion, rocket flight attitude, rocket orbital transfer process, engine motion, rocket flight state-time curve, rocket engine thrust-time curve, video monitoring information, parameter information and fault alarm information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011240138.0A CN112529248A (en) | 2020-11-09 | 2020-11-09 | Data-driven intelligent flight space-ground mirror image system of carrier rocket |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011240138.0A CN112529248A (en) | 2020-11-09 | 2020-11-09 | Data-driven intelligent flight space-ground mirror image system of carrier rocket |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112529248A true CN112529248A (en) | 2021-03-19 |
Family
ID=74980700
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011240138.0A Pending CN112529248A (en) | 2020-11-09 | 2020-11-09 | Data-driven intelligent flight space-ground mirror image system of carrier rocket |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112529248A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115077308A (en) * | 2022-05-18 | 2022-09-20 | 中国人民解放军61191部队 | Rocket space-time position simulation method and device based on rocket launching time sequence |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130090831A1 (en) * | 2010-06-14 | 2013-04-11 | Inha-Industry Partnership Institute | System for fault detection and diagnosis of aircraft engine and method thereof |
| RU2481251C1 (en) * | 2011-12-01 | 2013-05-10 | Федеральное государственное унитарное предприятие "Государственный космический научно-производственный центр имени М.В. Хруничева" | Method of safe launching of rocket with multiengine first stage |
| CN104008674A (en) * | 2014-06-16 | 2014-08-27 | 中国民航大学 | Terminal area approach flight time forecasting method based on pattern matching |
| CN105486526A (en) * | 2015-11-30 | 2016-04-13 | 北京宇航系统工程研究所 | Multi-strategy fault diagnosis system for carrier rocket test launching process |
| CN111259927A (en) * | 2020-01-08 | 2020-06-09 | 西北工业大学 | Rocket engine fault diagnosis method based on neural network and evidence theory |
-
2020
- 2020-11-09 CN CN202011240138.0A patent/CN112529248A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130090831A1 (en) * | 2010-06-14 | 2013-04-11 | Inha-Industry Partnership Institute | System for fault detection and diagnosis of aircraft engine and method thereof |
| RU2481251C1 (en) * | 2011-12-01 | 2013-05-10 | Федеральное государственное унитарное предприятие "Государственный космический научно-производственный центр имени М.В. Хруничева" | Method of safe launching of rocket with multiengine first stage |
| CN104008674A (en) * | 2014-06-16 | 2014-08-27 | 中国民航大学 | Terminal area approach flight time forecasting method based on pattern matching |
| CN105486526A (en) * | 2015-11-30 | 2016-04-13 | 北京宇航系统工程研究所 | Multi-strategy fault diagnosis system for carrier rocket test launching process |
| CN111259927A (en) * | 2020-01-08 | 2020-06-09 | 西北工业大学 | Rocket engine fault diagnosis method based on neural network and evidence theory |
Non-Patent Citations (2)
| Title |
|---|
| 孙成志，闫晓东: "基于神经网络和证据理论的火箭发动机故障诊断", 宇航总体技术, vol. 4, no. 4, pages 20 - 30 * |
| 陈新华，朱定一: "运载火箭姿控发动机故障诊断方法研究", 指挥技术学院学报, vol. 11, no. 3, pages 20 - 24 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115077308A (en) * | 2022-05-18 | 2022-09-20 | 中国人民解放军61191部队 | Rocket space-time position simulation method and device based on rocket launching time sequence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11600007B2 (en) | Predicting subject body poses and subject movement intent using probabilistic generative models | |
| JP6812538B2 (en) | Image depth prediction neural network | |
| CN110321603B (en) | Depth calculation model for gas path fault diagnosis of aircraft engine | |
| EP3776370A1 (en) | Graph neural network systems for behavior prediction and reinforcement learning in multple agent environments | |
| US11900645B2 (en) | Systems and methods for modeling and controlling physical dynamical systems using artificial intelligence | |
| CN112119409A (en) | Neural network with relational memory | |
| CN113826051A (en) | Generating digital twins of interactions between solid system parts | |
| Pan et al. | Data-driven vehicle modeling of longitudinal dynamics based on a multibody model and deep neural networks | |
| Bernstein et al. | Reinforcement learning for computer vision and robot navigation | |
| US20210103815A1 (en) | Domain adaptation for robotic control using self-supervised learning | |
| Chaffre et al. | Sim-to-real transfer with incremental environment complexity for reinforcement learning of depth-based robot navigation | |
| US20220092456A1 (en) | Controlling an agent to explore an environment using observation likelihoods | |
| CN112512755A (en) | Robotic manipulation using domain-invariant 3D representations predicted from 2.5D visual data | |
| CN113902159A (en) | Time series prediction using trained decoder models | |
| KR20230119023A (en) | Attention neural networks with short-term memory | |
| JP2004199692A (en) | System and method for implementing real-time application based on stochastic compute time algorithm | |
| CN113272829A (en) | Hardware accelerator extension for migratory learning-extend/complete training to the edge | |
| Dong et al. | Digital twin-assisted multiscale residual-self-attention feature fusion network for hypersonic flight vehicle fault diagnosis | |
| CN112529248A (en) | Data-driven intelligent flight space-ground mirror image system of carrier rocket | |
| Zhu et al. | Diff-lfd: Contact-aware model-based learning from visual demonstration for robotic manipulation via differentiable physics-based simulation and rendering | |
| Lu et al. | A hybrid CNN-LSTM architecture for path planning of mobile robots in unknow environments | |
| CN116595871A (en) | Vehicle track prediction modeling method and device based on dynamic space-time interaction diagram | |
| Du et al. | Imitation Learning for Autonomous Driving Based on Convolutional and Recurrent Neural Networks | |
| US20220138558A1 (en) | Deep simulation networks | |
| US20240051124A1 (en) | Simulation of robotics devices using a neural network systems and methods |
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 |