Method for evaluating gamma radiation resistance of robot sensing system

Abstract

The invention discloses a method for evaluating gamma radiation resistance of a robot sensing system. The method comprises the following steps: dividing the sensing system into three levels of system-module-device; obtaining radiation sensitive performance parameters of the devices, and performing a module irradiation experiment to obtain data of the performance parameters of the devices changing along with the irradiation accumulated dose; fitting a degradation process of the performance parameters of the device influenced by irradiation by adopting a Wiener process; constructing correlation of performance parameters of each device in the module by adopting a Copula function, and calculating a change curve of the reliability of the module along with irradiation accumulative measurement; on the basis of the reliability of each module, calculating the reliability of the sensing system by adopting a dynamic Bayesian network; and taking the corresponding accumulated dose when the reliability of the sensing system is 50% as the anti-gamma radiation performance index of the robot sensing system. According to the method, the degradation data of the performance parameters when the devices are affected by irradiation are fully utilized, the correlation of reliability changes of the devices in the module is considered, and the gamma radiation resistance of the robot sensing system is evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of circuit system radiation resistance performance, and mainly to a method for evaluating the gamma radiation resistance performance of a robot perception system. Background Art

[0002] Nuclear industry robots usually work in radiation environments for a long time. Their internal perception systems are susceptible to radiation, which can cause system material damage and performance degradation, affecting system efficiency and reliability, and further affecting the robot's life and mission reliability. In order to avoid radiation-induced performance degradation or functional failure of the perception system, it is necessary to use radiation-resistant devices or radiation-resistant reinforcement measures to enable it to have the required radiation resistance. Before performing radiation-resistant reinforcement, it is necessary to evaluate the radiation resistance of the perception system to select appropriate reinforcement measures. However, due to the constraints of experimental funds and conditions, the robot perception system can only conduct a small amount of radiation resistance assessment, and it is even difficult to conduct system-level radiation tests, making it difficult to obtain reliable radiation resistance performance evaluation results. With the new development of nuclear industry robots, new changes in robot development models, design technologies, components and materials have put forward new requirements for the radiation resistance reinforcement technology of complex circuit systems.

[0003] However, in the existing technology, it is difficult to meet the performance requirements of complex circuit systems in radiation environments. The existing technology is unable to carry out small-sample multi-factor complex system radiation performance evaluation methods and application research in response to the small sample size and difficult testing of system-level radiation tests. As a result, it is impossible to reasonably verify and evaluate the system's radiation resistance, as well as analyze and improve the system's radiation resistance reinforcement design. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a method for evaluating the radiation resistance performance of a nuclear industry robot perception system that solves the above-mentioned problem.

[0005] Specifically, the present invention proposes a method for evaluating the anti-γ radiation performance of a robot perception system, and the method for evaluating the anti-γ radiation performance of a robot perception system comprises the following steps:

[0006] Step 1: Analyze the radiation-sensitive devices in the perception system and divide the perception system into three levels: system-module-device;

[0007] Step 2, obtaining performance parameters of radiation sensitive devices and data on their changes with cumulative radiation measurements;

[0008] Step 3, using the Wiener process to fit the degradation process of the device characteristic parameters affected by irradiation, and further obtaining the marginal distribution value F(Y) of the characteristic parameter value;

[0009] Step 4: Use the Copula function to construct the correlation of the performance parameters of each device in the module, and calculate the reliability curve of the module with the cumulative measurement of radiation;

[0010] Step 5: Based on the reliability of each module, the reliability of the perception system is calculated using a dynamic Bayesian network.

[0011] Furthermore, in step 1, considering the impact of device reliability changes on the reliability of the perception system and the mutual influence between the functions of each device, the devices that will influence each other are divided into the same module to form a three-level system-module-device.

[0012] Furthermore, in step 3, the following steps are included:

[0013] Step 31, using the Wiener process to fit the degradation process of the device characteristic parameters affected by radiation, and obtaining the unknown parameter values ​​in the model;

[0014] Step 32: According to the BIC value of each model, the Wiener process model with the best fitting effect is selected, and the marginal distribution value F(Y) of the characteristic parameter value is measured.

[0015] Furthermore, in step 31, three Wiener process models in equations M1 to M3 are used to fit the degradation process of the device characteristic parameters affected by irradiation:

[0016] M1:

[0017] M2:

[0018] M3:

[0019] Where:

[0020] For cumulative radiation measurement, is the measured value of the performance parameter, a is the drift coefficient, σ is the diffusion coefficient, b is the unknown parameter indicating that the characteristic parameter degenerates into a nonlinear process, is the standard Brownian motion, N is the symbol of normal distribution, μ a is the mean of the normal distribution, σ a is the standard deviation of the normal distribution.

[0021] Furthermore, in step 32, the BIC value of each model is calculated:

[0022] BIC = -2log(L) + M log(m)

[0023] Where:

[0024] L is the maximum likelihood estimate of the Wiener process model, M is the number of unknown parameters in the Wiener process model, and m is the number of samples.

[0025] Furthermore, in step 4, the correlation of the performance parameters of each device in the module is constructed using the Copula functions in equations M4 to M6 respectively:

[0026] M4:C(F(Y1),F(Y2);θ)=exp{-[(-ln F(Y1)) θ +(-lnF(Y2)) θ ] 1 / θ}

[0027] M5:C(F(Y1),F(Y2);θ)=max{(F(Y1) -θ +F(Y2) -θ -1) 1 / θ ,0}

[0028] M6:

[0029] Where:

[0030] F(Y1) is the marginal distribution function of the first characteristic parameter, F(Y2) is the marginal distribution function of the second characteristic parameter, θ is the unknown parameter in the Copula function, and C is the symbol of the Copula function;

[0031] The maximum likelihood estimation method is used to solve the unknown parameter values ​​in the Copula function, and the Copula function that best describes the correlation of the performance parameters of each device is selected based on the principle of minimizing the BIC value.

[0032] Furthermore, in step 5, a system fault tree is established according to the connection relationship between the functions of each module in the system, and the dynamic fault tree is converted into a dynamic Bayesian network according to the conversion principle of the dynamic fault tree into a dynamic Bayesian network, and the reliability of the perception system is calculated.

[0033] Furthermore, in step 5, the cumulative dose corresponding to when the reliability of the perception system is 50% is used as the anti-γ radiation performance index of the robot perception system.

[0034] The beneficial effects achieved by the present invention are:

[0035] The present invention is not limited by the simulation model of the device, and evaluates the radiation resistance performance of the circuit system based on experimental data. It proposes a radiation resistance performance evaluation method based on Coupla function that takes into account the mutual influence between device reliability in a radiation environment. Compared with the existing data-based radiation resistance performance evaluation method, the evaluation result of the present invention is more accurate. It also proposes to use a dynamic Bayesian network to evaluate the radiation resistance performance of the circuit system, so that the scope of application of the present invention is wider. In addition to conventional static circuits, it can also be applied to dynamic circuits including circuits and sequentially related circuits. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic diagram of a method for evaluating the anti-γ radiation performance of a robot perception system provided by an embodiment of the present invention;

[0037] Figure 2 It is a schematic diagram of the hierarchical division of the perception system;

[0038] Figure 3 It is a schematic diagram of the degradation process of the characteristic parameters of the Wiener process fitting device affected by irradiation;

[0039] Figure 4 This is a schematic diagram of the module reliability calculation process based on the Copula function. DETAILED DESCRIPTION

[0040] The technical solution of the present invention is described in more detail below in conjunction with the accompanying drawings. The present invention includes but is not limited to the following embodiments.

[0041] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods in the embodiments.

[0042] As attached Figure 1 As shown, a schematic diagram of a method for evaluating the anti-γ radiation performance of a robot perception system provided by an embodiment of the present invention is provided. This embodiment provides a method for evaluating the anti-γ radiation performance of a robot perception system based on a dynamic Bayesian network and a Copula function, and the method includes:

[0043] Step 1: Perception system hierarchical division;

[0044] As attached Figure 2 As shown, the radiation-sensitive devices in the perception system are analyzed. For devices that are not sensitive to radiation, such as resistors and capacitors, the impact of their reliability changes on the reliability of the perception system is not considered; considering the mutual influence between the functions of each device (such as the reduction in the output voltage of the power chip will affect the functions of digital devices such as Flash), the devices that will influence each other are divided into the same module. Based on this, the perception system is divided into three levels: system-module-device.

[0045] Step 2: Analyze and obtain performance parameters of radiation sensitive devices;

[0046] Consult literature or conduct experiments to analyze the performance parameters of devices in the perception system that are sensitive to radiation, conduct module irradiation tests, and obtain data on the changes in performance parameters of each device as the cumulative measurement of radiation increases.

[0047] Step 3: Use the Wiener process to fit the degradation process of device performance parameters affected by radiation, and further obtain the marginal distribution value F(Y) of the characteristic parameter value;

[0048] Step 31, using the Wiener process to fit the degradation process of the device characteristic parameters affected by radiation, and obtaining the unknown parameter values ​​in the model;

[0049] As attached Figure 3 As shown, the Wiener process is used to fit the degradation process of the device characteristic parameters affected by radiation; the three Wiener process models shown in formulas M1 to M3 are used to fit the degradation process of the device characteristic parameters affected by radiation, and the maximum likelihood estimation method is used to solve the unknown parameter values ​​in the model.

[0050] M1:

[0051] M2:

[0052] M3:

[0053] Where:

[0054] For cumulative radiation measurement, is the measured value of the performance parameter, a is the drift coefficient, σ is the diffusion coefficient, b is the unknown parameter indicating that the characteristic parameter degenerates into a nonlinear process, is the standard Brownian motion, N is the symbol of normal distribution, μ a is the mean of the normal distribution, σ a is the standard deviation of the normal distribution.

[0055] Step 32, according to the BIC value of each model, select the Wiener process model with the best fitting effect, and measure the marginal distribution value F(Y) of the characteristic parameter value;

[0056] The BIC value of each model is calculated according to the following formula. The Wiener process model with the best fitting effect is selected based on the principle of minimizing the BIC value. Based on the fitting results of the Wiener process model, the distribution function of the characteristic parameters under any cumulative dose can be obtained, and further the marginal distribution value F(Y) of the characteristic parameter value measured by the experiment can be obtained.

[0057] BIC = -2log(L) + M log(m)

[0058] Where:

[0059] L is the maximum likelihood estimate of the Wiener process model, M is the number of unknown parameters in the Wiener process model, and m is the number of samples.

[0060] Step 4: Use the Copula function to construct the correlation of the performance parameters of each device in the module, and calculate the reliability curve of the module with the cumulative measurement of radiation;

[0061] As attached Figure 4 As shown, three commonly used Copula functions shown in equations M4 to M6 are used to construct the correlation of the performance parameters of each device in the module, and the maximum likelihood estimation method is used to solve the unknown parameter values ​​in the Copula function. The Copula function that best describes the correlation of the performance parameters of each device is selected based on the principle of minimizing the BIC value.

[0062] M4:C(F(Y1),F(Y2);θ)=exp{-[(-lnF(Y1)) θ +(-ln F(Y2)) θ ] 1 / θ}

[0063] M5:C(F(Y1),F(Y2);θ)=max{(F(Y1) -θ +F(Y2) -θ -1) 1 / θ ,0}

[0064] M6:

[0065] Where:

[0066] F(Y1) is the marginal distribution function of the first characteristic parameter, F(Y2) is the marginal distribution function of the second characteristic parameter, θ is the unknown parameter in the Copula function, and C is the symbol of the Copula function.

[0067] After determining the Copula function that best describes the correlation of each device performance parameter, the cumulative dose is The random variables F(Y1) and F(Y2) whose joint distribution is C(F(Y1), F(Y2); θ) are used to generate sample points of each performance parameter by combining F(Y1) and F(Y2) with the edge degradation model; the probability that each performance parameter does not reach the failure domain value is counted, and the connection relationship calculation module is used to calculate the cumulative dose when the cumulative dose is The reliability of the time, gradually increasing the cumulative dose The reliability of the module can be obtained as the cumulative dose The change curve.

[0068] Step 5: Based on the reliability of each module, the reliability of the perception system is calculated using a dynamic Bayesian network.

[0069] According to the connection relationship between the functions of each module in the system, a system fault tree is established. According to the transformation principle of dynamic fault tree into dynamic Bayesian network, the dynamic fault tree is converted into a dynamic Bayesian network, and the reliability of the perception system is calculated. The cumulative dose corresponding to the reliability of the perception system is 50% as the anti-γ radiation performance index of the robot perception system.

[0070] The present invention is not limited to the above-mentioned specific implementation modes. A person skilled in the art may implement the present invention in various other specific implementation modes according to the embodiments and the disclosure of the drawings. Therefore, any design that adopts the design structure and concept of the present invention and makes some simple transformations or changes shall fall within the scope of protection of the present invention.

Claims

1. A method for evaluating the anti-γ radiation performance of a robot perception system, characterized in that: The method for evaluating the anti-γ radiation performance of the robot perception system comprises the following steps: Step 1: Analyze the radiation-sensitive devices in the perception system and divide the perception system into three levels: system-module-device; Step 2, obtaining performance parameters of radiation sensitive devices and data on their changes with cumulative radiation measurements; Step 3, using the Wiener process to fit the degradation process of the device characteristic parameters affected by irradiation, and further obtaining the marginal distribution value F(Y) of the characteristic parameter value; Step 4: Use the Copula function to construct the correlation of the performance parameters of each device in the module, and calculate the reliability curve of the module with the cumulative measurement of radiation; Step 5: Based on the reliability of each module, the reliability of the perception system is calculated using a dynamic Bayesian network.

2. The method for evaluating the anti-γ radiation performance of a robot perception system according to claim 1 is characterized in that: In step 1, considering the impact of device reliability changes on the reliability of the perception system and the mutual influence between the functions of each device, the devices that will influence each other are divided into the same module to form a three-level system-module-device.

3. The method for evaluating the anti-γ radiation performance of a robot perception system according to claim 1 is characterized in that: In step 3, the following steps are included: Step 31, using the Wiener process to fit the degradation process of the device characteristic parameters affected by radiation, and obtaining the unknown parameter values ​​in the model; Step 32: According to the BIC value of each model, the Wiener process model with the best fitting effect is selected, and the marginal distribution value F(Y) of the characteristic parameter value is measured.

4. The method for evaluating the anti-γ radiation performance of a robot perception system according to claim 3 is characterized in that: In step 31, three Wiener process models in equations M1 to M3 are used to fit the degradation process of device characteristic parameters affected by irradiation: Where: For cumulative radiation measurement, is the measured value of the performance parameter, a is the drift coefficient, σ is the diffusion coefficient, b is the unknown parameter indicating that the characteristic parameter degenerates into a nonlinear process, is the standard Brownian motion, N is the symbol of normal distribution, μ a is the mean of the normal distribution, σ a is the standard deviation of the normal distribution.

5. The method for evaluating the anti-γ radiation performance of a robot perception system according to claim 3 is characterized in that: In step 32, the BIC value of each model is calculated: BIC = -2log(L) + Mlog(m) Where: L is the maximum likelihood estimate of the Wiener process model, M is the number of unknown parameters in the Wiener process model, and m is the number of samples.

6. The method for evaluating the anti-γ radiation performance of a robot perception system according to claim 1, characterized in that: In step 4, the Copula functions in equations M4 to M6 are used to construct the correlation of the performance parameters of each device in the module: M4:C(F(Y1),F(Y2);θ)=exp{-[(-ln F(Y1)) θ +(-ln F(Y2)) θ ] 1 / θ } M5:C(F(Y1),F(Y2);θ)=max{(F(Y1) -θ +F(Y2) -θ -1) 1 / θ ,0} Where: F(Y1) is the marginal distribution function of the first characteristic parameter, F(Y2) is the marginal distribution function of the second characteristic parameter, θ is the unknown parameter in the Copula function, and C is the symbol of the Copula function; The maximum likelihood estimation method is used to solve the unknown parameter values ​​in the Copula function, and the Copula function that best describes the correlation of the performance parameters of each device is selected based on the principle of minimizing the BIC value.

7. The method for evaluating the anti-γ radiation performance of a robot perception system according to claim 1, characterized in that: In step 5, a system fault tree is established according to the connection relationship between the functions of each module in the system. According to the transformation principle of dynamic fault tree into dynamic Bayesian network, the dynamic fault tree is transformed into a dynamic Bayesian network, and the reliability of the perception system is calculated.

8. The method for evaluating the anti-γ radiation performance of a robot perception system according to claim 1, characterized in that: In step 5, the cumulative dose corresponding to a reliability of 50% of the perception system is used as an indicator of the robot perception system's anti-γ radiation performance.

Images