Satellite soft error communication process modeling and simulation method based on Finite State Machine
Technical field
The present invention relates to satellite soft errors to propagate modeling and simulation technology field, in particular to a kind of to be based on finite state machine
Theoretical satellite soft error communication process modeling and simulation method.
Background technique
At present in safety satellite field, overwhelming majority research establishes mould both for the safe operation state of system
Type, and what the Development Communication process of few pairs of soft errors was modeled.Soft error refers between high energy particle and element silicon
Interaction and in the semiconductors caused by random, interim state change or transition, that is, cause satellite to interrupt, influence satellite can
Therefore an important factor for property, studies the method for carrying out modeling and simulating to satellite soft error communication process and just seems very necessary.
Satellite structure is divided into service platform and payload two parts by function, and service platform part mainly includes power supply system
System, control system, radio measurement and control system, payload portions and task performed by satellite are directly related, payload portion
Point and it includes system be undertake satellite function main body and soft error influence propagate main study subject.
So far, Safety Analysis Method has had been developed that multinomial technology, but there are still some shortcomings when it is used for satellite.
Common fault propagation process emulation mode is the purely logical emulation based on probability distribution mostly, is had some limitations, base
In fault tree, the Simulation Application Monte Carlo theory of event tree-model, the accident analysis and security risks for being able to achieve system are commented
Estimate, but solve the problems, such as comprising the time, process variable, people the dynamic factors such as operation behavior on there is also certain deficiencies.
In Simulink, (Visual Simulation Tools are one to mathematics library tools build based on Finite State Machine
Block diagram design environment of the kind based on MATLAB) on, discrete event system is imitated by state process and event-driven realization
Very.Finite State Machine provides description and the very powerful method of logic is applied in control, and many complicated logics can be with
It is stated in a small chart, simple in rule, readable and verifiability is very strong.In level of practice, Stateflow is (based on having
State machine and flow chart are limited to construct combination and sequential logic decision model and the environment emulated) realize finite state machine
Code automatically generates, and user only need to draw state transition diagram and process identification in the design interface of Stateflow, will
Stateflow generates Simulink simulation code, while can intuitively observation state conversion process.
Summary of the invention
The satellite soft error communication process based on Finite State Machine that the object of the present invention is to provide a kind of is modeled and is imitated
True method, the method are based on Finite State Machine, using the state block diagram in graphical tools Stateflow to satellite
The circulation way of failure mode and all parts failure of each component influenced in system by soft error is described, and is divided into vertical
To with laterally two propagations levels, composition soft error propagation model can be intuitive by being emulated to soft error propagation model
The dynamic communication process of soft error is observed on ground, and obtains soft error propagation path.The step of specific modeling and simulation method, is such as
Under:
Step 1. determines navigation satellite structure and research object;
Navigation satellite structure is different levels according to subordinate relation, such as system-level, subsystem irrespective of size, component-level.By function
It is divided into service platform and payload two parts, service platform part mainly includes power-supply system, control system, radio measurement and control
System, payload portions and task performed by satellite are directly related, are the parts influenced vulnerable to soft error, therefore will be effective
Payload segment is as research object.Specifically, the research object can also or payload portions subsystems.
Step 2. is based on FMEA (failure model and effect analysis) analysis method and obtains each portion in soft error communication process
The circulation way of failure mode and all parts failure of part;
According to the information flow and functional structure of Satellite Payloads, each level is obtained by soft error using FMEA analysis method
Failure mode caused by accidentally and its circulation way of failure.There are two types of the communication modes: the first is failure in same layer
Lateral propagation in grade between different components, second is the longitudinal propagation failed between different levels.
Step 3. is based on Finite State Machine and constructs satellite soft error propagation model;
Based on Finite State Machine, using its graphical tools Stateflow state block diagram on being influenced by soft error
The failure mode of each component is described, and the structure of each component is described with hierarchical states nesting and the form of parallel state
Functional relationship.Longitudinal propagation process (such as component-level → subsystem irrespective of size → system-level) is in soft error propagation model definition mark
Conditional jump triggering is laterally propagated (components A → part B in such as component-level) and is realized conditional jump, structure using direct event
Propagation relationship is built, soft error propagation model is eventually formed.Since module and component count are very huge in entire satellite system, therefore
It is named respectively using the unified format of text all over Britain, avoids the later period due to naming bring modeling software lack of standardization to report an error,
It improves work efficiency.And annotated with Chinese, it is easy to understand.
Step 4. process model emulates to obtain soft error propagation path;
The state control logic of Simulink environment and Stateflow is combined, to the soft error propagation model of building
Input and output are defined, it, can be intuitive by emulating by state that blue highlight show (becoming blue) into the state being currently active
The process that each block diagram is successively activated in Stateflow is observed on ground, obtains the failure state transfer occurred by certain timing,
That is soft error propagation path;To thrashing sample carry out reasonable assumption (such as extract one obedience Weibull distribution failure
Time), the average time between failures MTBF of each component can be calculated according to the time randomly selected in emulation, be satellite
Soft error communication process and safety analysis verification technique provide new approaches and new method.
Compared with prior art, the beneficial effects of the present invention are:
(1) multiclass failure mode caused by soft error and its failure effect can be described in the transmitting of each level of satellite, and
It is divided into longitudinal propagation between different levels by its invalid characteristic and is propagated with the transverse direction between level, analysis is made more to be bonded practical feelings
Condition, and not only rely on engineering experience.
(2) can describe soft error influences the stochastic behaviour propagated in the system structure of entire satellite.
(3) based on Finite State Machine establish soft error communication process model have design it is simple and clear, should be readily appreciated that,
Visual feature can construct corresponding simulation model for the specific structure of different satellite systems, have certain versatility,
Support reliability design analysis work.
Detailed description of the invention
Fig. 1 is the modeling and simulating step schematic diagram of satellite soft error communication process model provided by the invention.
Fig. 2 is navigation subsystem hierarchical structure nested model.
Fig. 3 is the propagation relation schematic diagram of each component failure.
Specific embodiment
The present invention provides a kind of satellite soft error communication process modeling and simulation method based on Finite State Machine, under
Face is described in detail with attached drawing in conjunction with the embodiments.
A kind of satellite soft error communication process modeling and simulation method based on Finite State Machine provided by the invention,
Process as shown in Figure 1, the specific steps are as follows:
Step 1. navigation satellite structure is divided into service platform and payload two parts by function, and service platform part is main
Including power-supply system, control system, radio measurement and control system, payload portions include navigation subsystem and antenna subsystem,
The soft error wherein caused vulnerable to influences such as single-particles is predominantly located in navigation subsystem, thereby determines that research object for navigation point
System.The navigation subsystem is mainly by uplink injection device unit, atomic clock/reference frequency synthesis unit, navigation task
Processing unit and navigation signal broadcast the composition such as unit, belong to subsystem level.
Step 2. propagates soft error using FMEA analysis method according to the information flow and functional structure of Satellite Payloads
Process is analyzed, and the failure mode as caused by soft error and its circulation way in navigation subsystem are obtained.
Device level soft error, which propagates to single machine grade total, in embodiment, in navigation subsystem may cause 6 kinds of single machine failures
Mode, i.e. uplink inject losing lock, data transmission fault, text output error, downlink signal without output, time reference mistake, phase
Potential difference measurement is abnormal, and finally generates the whole star of 4 classes and interrupt and influence, i.e., precision orbit determination accuracy decline, downlink signal interrupt,
Uplink and downlink signals interrupt simultaneously, star ranging it is incorrect.
Step 3. builds the state of each unit in navigation subsystem using the Stateflow state block diagram in finite state machine
Found corresponding illustraton of model, including system hierarchy nested model such as Fig. 2, navigation subsystem include uplink injection device unit,
Atomic clock/reference frequency synthesis unit, navigation task processing unit and navigation signal broadcast four units of unit, four units
Block diagram be represented by dashed line, illustrate that they are concurrency relations.
The communication mode of failure mode and all parts failure to each component of navigation subsystem is described, example
Such as Fig. 3, eventually forming soft error influences propagation model.By taking a navigation task processing unit in navigation subsystem as an example, lead
The task processing unit that navigates includes a normal condition (normal) and three failure states (failure1-failure3), from
It is (flag_232==1) realized by the mark in conditional jump, i.e. 232 components hair that normal, which is transferred to failure1,
Failure has been given birth to, has been led to that navigation task processing unit is transferred to first failure state, embodies the longitudinal propagation process of failure.And
Being transferred to failure2, failure3 then from normal is to shift (event_FCC_NTP) Lai Shixian state by direct event
Conversion, NTP (Navigation task processing unit) refers to navigation task processing unit.
Step 4. combines the state control logic of Simulink environment and Stateflow, and it is defeated that soft error influences model
Enter including soft error flow of event, soft error detection, system reset, switching on and shutting down, cut machine Annual distribution.It can be intuitively by emulation
It observes that each block diagram successively becomes blue process in Stateflow, obtains the failure state transfer occurred by certain timing, i.e., it is soft
Error propagation path;From engineering experience, each component in model is influenced for soft error and defines obedience Weibull point
The random number of cloth is extracted a random number by way of grab sample, is defined as the out-of-service time of component, when the system is operated
Between be equal to extract the component failure time when, which is transferred to failure state by normal condition.It is random according to system in emulation
The average time between failures MTBF of each component can be calculated in out-of-service time caused by the soft error of extraction, be satellite soft error
Accidentally communication process and safety analysis verification technique provide new approaches and new method.