Microsystem single event effect risk assessment method
By constructing a risk assessment model that combines device-level and system-level approaches, and employing fault tree and Bayesian network methods, the accuracy and comprehensiveness of single-event effect risk assessment for microsystems are addressed, thereby improving the reliability and safety of microsystems in space radiation environments.
Patent Information
- Application Number
- CN202511328444.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-17
- Publication Date
- 2026-02-06
AI Technical Summary
Existing technologies are insufficient to comprehensively and accurately assess the fault propagation patterns of microsystems under single-event effects in space radiation environments. Traditional methods lack consideration of the uncertainties in fault propagation and the complex dependencies between modules, resulting in assessment results that are highly subjective, inaccurate, and incomplete, thus affecting the reliability and safety of microsystems.
TCAD software is used to construct device-level models, and combined with fault tree models and Bayesian network models, system-level risk assessment is carried out. Through simulation and quantitative analysis, fault propagation paths and key nodes are identified, and a risk assessment model combining circuit-level and system-level approaches is constructed.
It enables a comprehensive and accurate assessment of the single-event effect risk of microsystems, identifies key failure points, provides a basis for design optimization, reduces assessment costs, and improves the reliability and safety of spacecraft in the space radiation environment.
Smart Images
Figure CN121480408A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of radiation risk assessment, and in particular relates to a method for assessing the risk of single-event effects in microsystems. Background Technology
[0002] With the continuous development of microelectronics technology, system-in-package (SiP) microsystems have been widely used in aerospace and other fields due to their advantages such as high integration and miniaturization. However, in the space radiation environment, high-energy particles can cause single-event effects in devices within the microsystem, leading to faults that propagate within the system and severely affecting its operational reliability. Therefore, accurately analyzing the propagation laws of single-event faults in microsystems and quantitatively assessing their failure risks under different orbital environments is of great significance for the design optimization and reliability improvement of microsystems.
[0003] Currently, research on single-event failures in microsystems largely focuses on single-event effect simulations of individual devices, lacking a comprehensive analysis of system-level fault propagation patterns. Furthermore, existing risk assessment methods often employ single models (such as fault trees), failing to adequately consider the uncertainties of fault propagation and the complex dependencies between modules, thus hindering the accuracy and comprehensiveness of assessment results. Aerospace microsystems are susceptible to single-event effects in the space radiation environment, leading to on-orbit anomalies. Existing risk assessment methods suffer from strong subjectivity, insufficient accuracy and comprehensiveness, and difficulty in co-designing with functional performance. Summary of the Invention
[0004] Purpose of the Invention: The purpose of this invention is to provide a method for assessing the risk of single-event effects in microsystems. This addresses the problems existing in the prior art and enables a comprehensive assessment of the operational risks of complex functional devices in space radiation environments.
[0005] Technical solution: The present invention provides a method for risk assessment of single-event effects in microsystems, comprising the following steps:
[0006] Step 1: Use TCAD software to build a device-level model, define process parameters, and write a simulation netlist file based on the circuit principle of the device to build a circuit-level model containing all devices and verify the working function of the model.
[0007] Step 2: Perform single-event effect simulation on the device to obtain the single-event transient current curves under different linear energy transfer (LET) values;
[0008] Step 3: Fit the single-event transient current curve using a double exponential model, inject it into the sensitive nodes of each module in the form of a current source, and obtain the LET threshold of each module.
[0009] Step 4: Establish a fault tree model based on the system architecture and perform quantitative analysis to calculate the failure rate, unavailability, and mean time between failures (MTTF) of each underlying event;
[0010] Step 5: Establish a fault propagation model based on Bayesian networks, conduct system risk assessment based on the model, define network nodes, establish conditional probability tables, determine the dependencies between nodes, identify and define the fault propagation path of the system, run Bayesian network inference, calculate the fault probability of each node, and identify key risk nodes.
[0011] Further, step 1 specifically involves: based on the process parameters in the device product manual, including materials, dimensions, doping concentration, and mesh parameters, using mainstream semiconductor process and device simulation tool TCAD to perform 3D modeling of the core unit MOS transistor in the device. First, the device structure is generated, then meshing is performed to generate a meshed device structure. Electrical modeling is then performed, and the substrate, drift region, and source region dimensions and concentration parameters of the device structure are optimized to ensure that the obtained electrical curves are consistent with those in the device product manual. Based on Spice software, according to the circuit principle of each device, the optimized parameter data is extracted into a library file. The library file is extracted and circuit modeling is performed, a netlist file is written, and functional verification is completed.
[0012] Further, step 2 specifically involves: performing single-event effect simulation on the device, defining the LET value, incident position, exit position, track radius, characteristic time parameter, and peak time parameter of the incident particle, performing single-event effect simulation on the device structure, and obtaining single-event transient current curves under different LET values.
[0013] Furthermore, step 3 specifically involves fitting the single-particle transient current curve I(t) obtained in step 2 using a double-exponential current source model. The formula for the double-exponential model is:
[0014]
[0015] Among them, each parameter represents the time t when the current rises. r Current rise time constant τ α The time t when the current decreases f Current fall time constant τ β Peak current I m The fitted parameters I m τ α τ β Defined as an exponential (EXP) current source in the Hspice netlist, the current source is injected in parallel into the sensitive nodes of the circuit-level model.
[0016] Furthermore, step 4 specifically involves: employing a top-down deductive analysis method to progressively decompose the top event into intermediate events, until the bottom event, which cannot be further decomposed, is reached. Logic gates are used to describe the causal relationships between events at each level, forming a complete fault tree model. The failure rate of the bottom event is the basis for quantitative analysis, and its calculation formula is as follows:
[0017]
[0018] σ represents the cross-section of the functional module for a specific single-event effect, obtained through circuit-level simulation and verification. The unavailability Q(t) is calculated based on the proton or heavy ion fluence rate predicted from the spacecraft mission orbit. Unavailability represents the probability that a component or system is in a faulty state within mission time t. For unrepairable systems, the calculation formula is as follows:
[0019] Q(t) = 1 - e -λt
[0020] Where t is the task time, and MTTF is calculated as follows: MTTF is the reciprocal of the failure rate.
[0021] MTTF is used to measure reliability: 1 / λ.
[0022] The present invention also discloses a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method of the present invention.
[0023] The present invention also discloses a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implements the steps of the method of the present invention.
[0024] The present invention also discloses a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the method of the present invention.
[0025] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:
[0026] This invention innovatively constructs a risk assessment model combining circuit-level and system-level approaches, enabling a comprehensive and accurate assessment of single-event effect risks in aerospace microsystems. Through fault tree modeling and Bayesian network modeling, it can deeply analyze fault propagation paths and influencing factors, identify critical fault nodes, and provide a scientific basis for microsystem design optimization and risk prevention. Simultaneously, this method effectively addresses the shortcomings of traditional assessment methods, such as strong subjectivity, insufficient accuracy and comprehensiveness, reduces reliance on expensive experimental resources, lowers assessment costs, and provides strong technical support for radiation resistance evaluation and hardening design of aerospace microsystems, contributing to improved reliability and safety of spacecraft in the space radiation environment. Attached Figure Description
[0027] Figure 1 This is a flowchart illustrating the implementation of the present invention.
[0028] Figure 2 This is a structural diagram of a SiP microsystem.
[0029] Figure 3 The graph shows the electrical characteristics of a 180nm NMOS.
[0030] Figure 4 The simulation waveform diagram for the DRAM memory cell.
[0031] Figure 5 The diagram shows the single-event transient current under different LET conditions.
[0032] Figure 6 This is a waveform diagram of a DRAM single-event flip.
[0033] Figure 7 This is a fault tree model diagram for a SiP system.
[0034] Figure 8 This is a diagram showing the inference results of the Bayesian network inside the GEO orbital module. Detailed Implementation
[0035] The technical solution of the present invention will be further described below with reference to the accompanying drawings.
[0036] like Figure 1 As shown, a method for risk assessment of single-event effects in microsystems includes the following steps:
[0037] Step 1: Based on the process parameters in the device datasheet, perform 3D modeling of the core unit, the MOSFET, using the mainstream semiconductor process and device simulation tool TCAD. Figure 2(For SiP circuit structure diagrams), firstly, the device structure is generated, then meshing is performed to generate a meshed device structure. Electrical modeling is then performed, and parameters such as substrate, drift region, source region size, and concentration are optimized to ensure that the resulting electrical curves are consistent with those in the device datasheet. Figure 3 The process involves generating electrical characteristic curves, extracting optimized parameter data into a library file, extracting the library file, performing circuit modeling, writing the netlist file, and completing functional verification, using a DRAM module as an example. Figure 4 This is a waveform diagram for DRAM simulation.
[0038] Step 2: Add single-event effect statements based on Step 2; define the LET value of the incident particle (Linear Energy Transfer, which physically represents the energy deposited per unit density and unit length, measured in MeV·cm). 2 By analyzing the incident position, exit position, track radius, characteristic time parameters, and peak time parameters, single-event effect simulations of the device structure can be performed to obtain single-event transient current curves under different LET conditions (e.g., [mg]). Figure 5 (as shown);
[0039] Step 3: Fit the transient current curve I(t) obtained in Step 2 using a double exponential current source model. The formula for the double exponential model is:
[0040]
[0041] Among them, each parameter represents the time of current rise (t) r ), current rise time constant (τ) α ), the time of current decrease (t) f ), current fall time constant (τ) β Peak current (I) m ), and the fitted parameters (I) m , τ α , τ β This is defined as an EXP current source in the Hspice netlist. This current source is injected in parallel into a sensitive node of the circuit-level model (such as the drain of an NMOS transistor in SRAM). Taking SRAM as an example, when LET = 0.8 MeV·cm... 2 At / mg, the SRAM output signal flipped (e.g., Figure 6 As shown in the figure, since no flipping occurred before, the LET threshold of this circuit is determined to be 0.9 MeV·cm. 2 / mg;
[0042] Step 4: Using a top-down deductive analysis method, the top event is decomposed level by level into intermediate events until it is decomposed to the basic events that cannot be further decomposed. Logic gates (AND gates, OR gates) are used to describe the causal relationships between events at each level, forming a complete fault tree model (e.g., ...). Figure 7 The failure rate of the lowest-level event is the foundation of quantitative analysis. Its calculation formula is: σ represents the cross-section of the functional module for a specific single-event effect, obtained through circuit-level simulation and verification in step 3 (unit: cm). 2 ). The proton or heavy ion fluence predicted based on spacecraft mission orbits (e.g., GEO, LEO, MEO, SSO) (unit: particles / (cm²)). 2 Unavailability (Q(t)) calculation: Unavailability represents the probability that a component or system is in a faulty state within task time t. For unrepairable systems, the calculation formula is: Q(t) = 1 - e -λt Where t is the mission duration (e.g., 10 years, or 87,600 hours). Mean Time To Failure (MTTF) calculation: Mean Time To Failure is the reciprocal of the failure rate and is used to measure reliability: MTTF = 1 / λ. Taking GEO orbit as an example, the failure rate, unavailability, and MTTF of SiP in GEO orbit are shown in Table 1.
[0043] Table 1. SiP's performance in GEO track failure rate, unavailability, and MTTF.
[0044]
[0045] Step 5: Using the failure rates of each underlying event calculated in Step 4 and the LET threshold obtained in Step 3 as input evidence, set them to the corresponding functional module nodes in the Bayesian network. Calculate node probabilities: Using the Bayesian network inference algorithm, perform probability propagation to calculate the failure probability of all sensitive nodes. Taking the GEO cabin as an example (…). Figure 8 The probability of SEU occurrence is 0.000304, the probability of data error (DE) is 0.000304, and the probability of normal system operation is 0.999696. Key nodes identified: FPGA:SRAM (highest probability of SEU), DC-DC converter (affects the entire system power supply). Protection recommendations: Use redundant design for DC-DC converter, and add EDAC (Error Detection and Correction) circuitry to FPGA:SRAM.
Claims
1. A method for assessing the risk of single-event effects in a microsystem, characterized in that, Includes the following steps: Step 1: Use TCAD software to build a device-level model, define process parameters, and write a simulation netlist file based on the circuit principle of the device to build a circuit-level model containing all devices and verify the working function of the model. Step 2: Perform single-event effect simulation on the device to obtain the single-event transient current curves under different linear energy transfer (LET) values; Step 3: Fit the single-event transient current curve using a double exponential model, inject it into the sensitive nodes of each module in the form of a current source, and obtain the LET threshold of each module. Step 4: Establish a fault tree model based on the system architecture and perform quantitative analysis to calculate the failure rate, unavailability, and mean time between failures (MTTF) of each underlying event; Step 5: Establish a fault propagation model based on Bayesian networks, conduct system risk assessment based on the model, define network nodes, establish conditional probability tables, determine the dependencies between nodes, identify and define the fault propagation path of the system, run Bayesian network inference, calculate the fault probability of each node, and identify key risk nodes.
2. The method for assessing the risk of single-event effects in a microsystem according to claim 1, characterized in that, Step 1 specifically involves: Based on the process parameters in the device product manual, including materials, dimensions, doping concentration, and mesh parameters, using mainstream semiconductor process and device simulation tools such as TCAD, to perform 3D modeling of the core unit MOS transistor in the device. First, the device structure is generated, then meshing is performed to generate a meshed device structure. Electrical modeling is then performed, and the substrate, drift region, and source region dimensions and concentration parameters of the device structure are optimized to ensure that the obtained electrical curves are consistent with those in the device product manual. Based on the Spice software, according to the circuit principle of each device, the optimized parameter data is extracted into a library file, a netlist file is written, circuit modeling is performed, and functional verification is completed.
3. The method for assessing the risk of single-event effects in a microsystem according to claim 1, characterized in that, Step 2 specifically involves: performing single-event effect simulation on the device, defining the LET value, incident position, exit position, track radius, characteristic time parameter, and peak time parameter of the incident particle, performing single-event effect simulation on the device structure, and obtaining the single-event transient current curves under different LET values.
4. The method for risk assessment of single-event effects in a microsystem according to claim 1, characterized in that, Step 3 specifically involves fitting the single-particle transient current curve I(t) obtained in Step 2 using a double-exponential current source model. The formula for the double-exponential model is as follows: Among them, each parameter represents the time t when the current rises. r Current rise time constant τ α The time t when the current decreases f Current fall time constant τ β Peak current I m The fitted parameters I m τ α τ β Defined as an exponential (EXP) current source in the Hspice netlist, the current source is injected in parallel into the sensitive nodes of the circuit-level model.
5. The method for assessing the risk of single-event effects in a microsystem according to claim 1, characterized in that, Step 4 specifically involves: using a top-down deductive analysis method, the top event is decomposed level by level into intermediate events, until it is decomposed to the bottom event, which cannot be further decomposed. Logic gates are used to describe the causal relationships between events at each level, forming a complete fault tree model. The failure rate of the bottom event is the basis for quantitative analysis, and its calculation formula is as follows: σ represents the cross-section of the functional module for a specific single-event effect, obtained through circuit-level simulation and verification. The unavailability Q(t) is calculated based on the proton or heavy ion fluence rate predicted from the spacecraft mission orbit. Unavailability represents the probability that a component or system is in a faulty state within mission time t. For unrepairable systems, the calculation formula is as follows: Q(t)=1-e -λt Where t is the task time, and the Mean Time Between Failures (MTTF) is calculated as follows: MTTF is the reciprocal of the failure rate and is used to measure reliability: MTTF = 1 / λ.
6. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method of claim 1.
7. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method of claim 1.
8. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method of claim 1.