Method for evaluating service life of instrument control board card under combined stress and aging experiment device

Abstract

The invention discloses a method for evaluating the service life of an instrument control board card under composite stress and an aging experiment device, and the method comprises the steps: carrying out the simulation aging test of a key sensitive device of the instrument control board card under multi-stress composite, obtaining the performance parameters of the device, taking the aging time as an independent variable, taking the performance data of the device as a dependent variable, constructing a device service life model, and obtaining the service life of the instrument control board card. Predicting the service lives of the devices, fusing the service lives of all the key sensitive devices in the instrument control board card, and taking the minimum value of the service life prediction values of all the key sensitive devices as the service life evaluation value of the board card; the device comprises a programmable power supply, a protective resistor, a pulse current sensor, an air thermal aging oven, a component clamp, a high-voltage probe, an oscilloscope and an industrial personal computer. According to the invention, composite stress such as high temperature, cold and hot impact, current, moisture and vibration of the board card and the device in service is simulated through tests, a device life model under different stress fusion is established, and the accuracy of life prediction is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of nuclear power plant operation license extension (OLE), and particularly relates to a method for evaluating the life of an instrument control board card under composite stress and an aging experiment device. BACKGROUND

[0002] The instrument control board card is one of important devices commonly used in the instrument control system of a nuclear power plant, which can play the roles of power supply, control, signal processing and communication, and is generally composed of various types of electronic components, such as resistors, capacitors and semiconductor devices. The board card type device is prone to degradation with the increase of running time, which affects the reliability of the instrument control system. There are many factors affecting the life of the board card, such as device failure, circuit welding failure and line failure, etc. However, through the summary and analysis of industry operation experience, most of the board card failures are caused by the aging failure of components. The degradation of components is a regular degradation phenomenon, and the aging state of components can be accurately evaluated through the degradation trajectory of the components. Therefore, a technical means is needed to study the degradation law of components under composite stress, so as to evaluate the life of components in the actual service scene, and then obtain the life of the instrument control board card.

[0003] The life calibration of key components of the instrument control board card at home and abroad is mainly based on standard data calculation. According to the basic component failure rate data, combined with service temperature, voltage stress and other parameters, the average failure-free time of components can be calculated. However, this pure reliability calculation method deviates from the actual aging degradation mechanism of components, and there is a certain gap between the obtained result and the actual service life of components. Therefore, only by simulating the composite stress of high temperature, cold and hot impact, current, humidity and vibration of the board card and components in service, establishing a component life model under different stress fusion, can the aging life evaluation result more consistent with the actual scene be calculated. SUMMARY

[0004] One of the purposes of the present application is to provide an aging life evaluation method for an instrument control board card under composite stress, which overcomes the problem that the aging life evaluation result of key components of the instrument control board card based on standard data cannot accurately reflect the aging of components under actual complex working conditions, and improves the accuracy of the aging life evaluation of the instrument control board card.

[0005] The technical scheme of the present application is as follows: An aging life evaluation method for an instrument control board card under composite stress, comprising the following steps: Step S1, analyzing and determining the key sensitive components of the instrument control board card; Step S2, performing a simulation aging test on the key sensitive components of the instrument control board card under multiple stress combinations, and continuously collecting the performance parameters of the key sensitive components; Step S3, taking the aging time as the independent variable and the device performance parameter as the dependent variable, constructing a device life model to predict the device life; the step S3 further comprises: S3.1, constructing a device life model formula, the model formula includes a linear function, an exponential function, and a polynomial function; for each performance parameter of each device, the above three model formulas are used for fitting respectively; S3.2, identifying the model parameters by using a particle swarm optimization algorithm; the objective function of the particle swarm optimization algorithm is the root mean square error between the model output performance parameter value obtained by parameter fitting and the actual performance parameter value of the device obtained by aging test; S3.3, for each performance parameter of the device, comparing the objective function values obtained by the particle swarm optimization algorithm under different model formulas, and selecting the model formula with the smallest error as the performance parameter life model; S3.4, for each performance parameter of the device, comparing the life predicted by the performance parameter life model determined in S3.3 with the life corresponding to the performance parameter cutoff criterion of the device, and taking the smaller value as the expected life of the performance parameter; S3.5, comparing the life predicted by the life model of each performance parameter of the device at the same time, and taking the minimum value as the life prediction value of the device; Step S4, taking the minimum value of the life prediction values of all key sensitive devices of the board card as the life evaluation value of the board card.

[0006] Preferably, during the simulation aging test, the composite stress is continuously applied to the key sensitive devices of the instrument control board card, and the composite stress includes but is not limited to electric stress, high humidity stress, high temperature stress, cold and hot impact stress, and vibration stress.

[0007] Preferably, before the simulation aging test of the device, the stress data under the actual service condition of the nuclear power plant instrument control board card is collected; during the aging test, the stress data under the actual service condition is used as the test input to apply the composite stress to the device.

[0008] Preferably, the key sensitive devices of the instrument control board card are selected according to the failure characteristics of the device itself and the circuit environment in which the device is located.

[0009] Preferably, the selection standard of the key sensitive devices of the instrument control board card is that the ohmic heat temperature rise of the device itself under normal service condition is greater than or equal to 5℃.

[0010] Preferably, the key sensitive devices of the instrument control board card include but are not limited to electrolytic capacitor, high-power diode, optical coupler, triode, and field effect transistor.

[0011] The capacity life model of the 47uf direct insertion type aluminum electrolytic capacitor is y=-0.6336t+45.7281, the equivalent series resistance ESR life model is , the on-resistance life model of the Schottky diode BAT54 is , the on-resistance life model of the MOSFET tube W14NK50Z is , and the reverse leakage current model of the BYV72EW diode is .

[0012] Preferably, the steps of parameter identification of the life model by using the particle swarm optimization algorithm are as follows: (1) Initialize the particle swarm Randomly generate N particles, the dimension of each particle is the same as the number of model parameters to be identified; each particle is randomly initialized with a position x and a speed v; the position x of each particle represents a possible solution, that is, a coefficient group of the life model; Record the historical optimal position P i and the global optimal position g of each particle; (2) Define the objective function The objective function is the root mean square error between the model output performance parameters obtained by parameter fitting and the actual performance parameter values of the device obtained by aging test; (3) Iterative update For each particle i and each iteration t: 1) Update the speed: Wherein, w is the inertia weight, c1 and c2 are the individual and group learning factors respectively, r1 and r2 are random numbers, P i is the individual optimal position, g is the global optimal position, x i is the current position of the particle, and v i is the current speed of the particle; 2) Update the position: 3) Update the individual optimal After updating the position each time, the position of the new particle is brought into the objective function to calculate the fitness value of the particle, and the particle with smaller fitness value is better, if the new position is better than P i , then P i =x i (t+1); 4) Update the global optimal If the new position is better than g, then g=x i (t+1); Repeat the above steps until the termination condition is met.

[0013] Preferably, when initializing the particle swarm, the dimension of each particle is 2, 3 or 4; wherein, for a linear function, the particle dimension is 2, and the identified model parameters are the slope k and the intercept b of the linear function; for a polynomial function, the particle dimension is 3 or 4, and the identified model parameters are the coefficients p1, p2, p3, p4 of the polynomial function; for an exponential function, the particle dimension is 4, and the identified model parameters are the coefficient a, the exponential coefficient h, the horizontal translation d and the vertical translation c of the exponential function; when initializing the particle swarm, the value range of each particle position and velocity is set to [-1, 1]; and the termination condition of the particle swarm optimization algorithm is to reach 150 iterations.

[0014] Preferably, before the aging test, the device degradation mechanism analysis is carried out to obtain the key performance parameters related to the performance degradation of the device, and the parameter data of these performance parameters are collected during the aging test.

[0015] In addition, the second object of the present application is to provide an aging test device for an instrument control board card under composite stress, so as to simulate the composite stress under actual working conditions, simulate the field service working conditions of the device, and solve the technical problem of low accuracy of life prediction of the instrument control board card caused by the fact that the performance degradation trajectory of the components and devices does not match the performance degradation trend under the actual service working conditions.

[0016] The technical scheme of the present application is: An aging test device for an instrument control board card under composite stress, comprising a programmable power supply, a protective resistor, a pulse current sensor, an air thermal aging oven, a component and device clamp, a high-voltage probe, an oscilloscope and an industrial computer. The programmable power supply, the protective resistor, the component and device clamp and the pulse current sensor constitute a main test loop and are in a series connection in the main test loop, and the main test loop provides electrical stress for the components and devices; the protective resistor is used for protecting the components and devices and the programmable power supply from being burnt out; and the pulse current sensor is used for monitoring the current level in the main loop. The air thermal aging oven is used for providing high temperature, cold and hot impact and humidity stress for the components and devices, and the component and device clamp in the air thermal aging oven is used for clamping the components and devices during the aging test. The high-voltage probe is connected in parallel at both ends of the component and device clamp and is used for collecting the output voltage performance parameters of the components and devices in the main loop. The oscilloscope is connected in parallel at both ends of the component and device clamp and is used for collecting the voltage waveform of the components and devices in the main loop. The industrial computer is connected with the programmable power supply, the air thermal aging oven, the oscilloscope, the current sensor and the high-voltage probe, is responsible for the control logic of the whole system, issues control instructions, and collects device test data.

[0017] Preferably, the air thermal aging oven should select a 750V and below power frequency single-phase alternating current power supply as the input.

[0018] Preferably, the service working condition data is input in the industrial computer, the industrial computer automatically sets the temperature and humidity parameters of the air thermal aging box, and sets the voltage and current size of the program-controlled power supply, and carries out the combined aging test.

[0019] The beneficial effects of the present application are: (1) The present application simulates the combined stresses of high temperature, cold and hot impact, current, humidity and vibration in service through experiments, and establishes a device life model integrating different stresses, greatly improves the accuracy of the device life model through extrapolation calculation between the actual service working condition and the aging test working condition. In order to further improve the accuracy of the life model, the device degradation mechanism analysis is carried out before the aging test, the key performance parameters related to the performance degradation of the device are obtained, and the number of device performance parameters collected in the aging test is reduced; then, the particle swarm optimization algorithm is used to fit different device life models, and the performance parameters are screened according to the root mean square error of the model as an index, and finally a number of performance parameters strongly related to the aging state of the device are determined for life modeling, on the basis of which the device aging test is carried out, the test cost is greatly reduced on the basis of ensuring the effectiveness of the data. Compared with the actual life of the typical components and devices in the field service, the life deviation of the typical components and devices predicted by the present method is within 85%, which can better guide the replacement cycle optimization of the field instrument control board card components and devices, and improve the service stability of the board card.

[0020] (2) The present application determines the key sensitive devices of the instrument control board card through the degradation mechanism analysis of the card components of the instrument control board card. The key sensitive devices are mainly selected according to the failure characteristics of the devices themselves and the circuit environment in which they are located. When analyzing the failure characteristics of the devices, the devices prone to aging or failure are selected according to the material properties, working principle and failure mechanism of the devices themselves; when analyzing the circuit environment, the circuit environment in which the devices are located is considered, including working voltage, current, temperature, humidity, electromagnetic interference and other factors. The main reason for selecting the key sensitive devices for testing instead of conducting comprehensive testing on the entire instrument control board card is that the key sensitive devices are the core components that affect the performance and life of the board card, and their aging characteristics can directly reflect the overall reliability of the board card. Testing these devices separately can quickly evaluate their performance degradation and life limit in a shorter time and at a lower cost, thereby providing key basis for the reliability evaluation of the entire instrument control board card, and avoiding the waste of resources caused by redundant testing of a large number of non-key devices.

[0021] (3) The application adopts linear function, exponential function, polynomial function and other possible model structures, in the parameter identification scene, the parameters correspond to the parameters of the device life model, such as the slope and intercept in the linear life model, in the training process of the model parameters using the particle swarm optimization (PSO) algorithm, the various parameter identification problems of the device life model can be effectively solved, and when the PSO algorithm completes the training, the minimum root mean square error trained under different model structures is compared, and the model structure with the minimum error is selected as the device life model, which can greatly improve the accuracy of the device life model and avoid poor fitting effect caused by improper model structure selection.

[0022] (4) The simulation aging test device of the application realizes free combination of multi-dimensional stress aging acceleration, maximally simulates the service working condition of the device, so as to ensure that the performance degradation trajectory of the component device is maximally restored to the performance degradation trend in actual service, and greatly improves the accuracy of life prediction. Through the composite stress aging test, the service working condition of the device is maximally simulated; then the device aging life model is developed according to the test data, the actual service life of the device is extrapolated and calculated combined with the performance parameter failure criterion of the device, the obtained result is compared with the life failure time statistically on site, and the life deviation of the typical device is less than 10%. The life of the key device on the board is fused and analyzed, so as to evaluate the life prediction value of the whole board, and the life deviation of the board is less than 15% compared with the average failure time of the card on site. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the present application, the following will briefly introduce the drawings needed to be used in the technical description.

[0024] Figure 1 The flow chart of the aging life evaluation method of the instrument control board card under the composite stress of the application; Figure 2 The triode IV curve of the application; Figure 3 The diode IV curve of the application; Figure 4 The structure schematic diagram of the simulation aging test device of the instrument control board card under the composite stress of the application. DETAILED DESCRIPTION

[0025] The following will be further described in detail through specific embodiments.

[0026] The following will be further described in detail through specific embodiments.

[0027] The main aging stresses of the instrument control board card during service come from the high temperature, cold and hot impact, humidity, vibration and other stresses of the external environment. Among them, the thermal aging of the board card will cause the failure of the semiconductor characteristics, the PN junction and the field effect tube cannot work normally, and the circuit welding will also appear deterioration with thermal aging; too high temperature will affect the precision, stability and inductance of resistance value and other parameters; too high environmental humidity will cause sensitive components to rust, short circuit, insulation failure and other problems. In addition, the impact voltage, surge and pulse magnetic field from the internal circuit will also affect the electrical signal output characteristics of the components.

[0028] The embodiment provides an aging life evaluation method of an instrument control board card under composite stress, and the flowchart is shown in the following figure. Figure 1 The method comprises the following steps: Step S1, analyzing and determining the key sensitive components of the instrument control board card; Step S2, performing a simulation aging test on the key sensitive components of the instrument control board card under multiple stress combinations to obtain a performance parameter data set of the components; Step S3, constructing a component life model with aging time as the independent variable and component performance parameters as the dependent variable to predict the life of the components; Step S4, fusing the lives of all the key sensitive components in the instrument control board card, and taking the minimum value of the life prediction values of all the key sensitive components as the life evaluation value of the board card.

[0029] The specific implementation process of the above steps is as follows: S1, analyzing and determining the key sensitive components of the instrument control board card.

[0030] Before the simulation aging test on the key sensitive components of the instrument control board card under multiple stress combinations, the key sensitive components of the instrument control board card need to be analyzed and determined. The key sensitive components are selected for the test instead of the overall test of the instrument control board card, mainly because the key sensitive components are the core components that affect the performance and life of the board card, and their aging characteristics can directly reflect the overall reliability of the board card. The test on these components alone can quickly evaluate the performance degradation and life limit in a shorter time and at a lower cost, thereby providing a key basis for the reliability evaluation of the entire instrument control board card, and avoiding the waste of resources caused by the redundant test on a large number of non-key components.

[0031] By analyzing the mechanism of the card of the instrument control board, the key sensitive devices are determined. In this embodiment, the sensitive devices are mainly screened according to the failure characteristics of the devices themselves and the circuit environment in which they are located. When analyzing the failure characteristics of the devices, the devices prone to aging or failure are screened according to the material characteristics, working principle and failure mechanism of the devices themselves; when analyzing the circuit environment, the circuit environment in which the devices are located is considered, including factors such as working voltage, current, temperature, humidity, electromagnetic interference, etc. For example, the electrolytic capacitor slowly increases the electrolyte with the increase of the running time, resulting in the decrease of the capacitance value and the increase of the equivalent series resistance, and finally completely fails; for example, the high-power diode plays the role of voltage division and rectification, and works in a high-current environment for a long time, which causes serious heating and accelerates the failure process; therefore, the electrolytic capacitor and the high-power diode are the key sensitive devices of the instrument control board.

[0032] As a preferred way of screening key sensitive devices, the ohmic heat temperature rise of the device itself under normal service conditions is greater than or equal to 5℃ as an important standard for screening.

[0033] S2, the key sensitive devices of the instrument control board are subjected to simulation aging test under multiple stress combinations, and the performance parameters of the devices are continuously collected.

[0034] When carrying out the simulation aging test under multiple stress combinations, the devices of the same manufacturer, same model specification and same packaging form as the instrument control board are selected as the objects of the device aging test.

[0035] The aging stress simulates the actual service conditions, and considers the coupling of multiple factors such as electric stress, high humidity stress, high temperature stress, cold and hot impact stress and vibration stress. During the continuous application of the combined stress to the device, the performance parameters of the device are collected every period (such as 80 hours), and this is repeated for multiple periods, so that the performance parameter data of the device at different aging stages can be obtained. The device performance parameter data set includes resistance value, capacitance value, equivalent series resistance of capacitor, forward conduction voltage of diode, reverse leakage current of diode, current gain of transistor, saturation voltage drop of field effect transistor, etc.

[0036] The termination condition of the aging test can adopt the fixed-time truncation method and the fixed-number truncation method. In this embodiment, the fixed-number truncation method is adopted, and when the output performance parameters of the device reach the termination criterion during the aging test process, the test is terminated. The termination criterion of the performance parameters of the device refers to the performance parameter value when the device completely fails, which is generally referred to from literature, reports, national standards and industry standards. For example, the termination criterion of the electrolytic capacitor is that the capacitance value degrades to 80% of the nominal value, and the equivalent series resistance ESR increases to 200% of the initial value.

[0037] The stress data of the nuclear power plant instrument control board card in the actual service condition can be obtained by sensors, monitoring equipment or other data acquisition means, and the stress data includes but is not limited to temperature, humidity, vibration, voltage fluctuation, current change and the like. The stress data is cleaned, noise is removed, missing data is filled, and abnormal data points are identified and processed; if the data comes from multiple sensors (such as temperature sensors, strain sensors, vibration sensors), the timestamps of them need to be aligned to ensure the consistency of the data in time, and the data collected by different sensors is converted into a unified unit.

[0038] Figure 2 The current-voltage characteristic curve of the triode is shown, and the triode current gain, saturation region inflection point current, cutoff current and other parameters can be obtained from the figure, Figure 3 The current-voltage characteristic curve of the diode is shown, and the diode reverse leakage current and other parameters can be obtained from the figure.

[0039] S3. Constructing a device life model with aging time as the independent variable and device performance parameters as the dependent variable to predict the life of the device. The following steps are included: S3.1. Constructing a device life model formula, which includes a linear function, an exponential function and a polynomial function; for each performance parameter of each device, the above three model formulas are used for fitting respectively; The model parameters of the linear model are the slope k and the intercept b of the linear function, and the model formula is The model parameters of the exponential model are the coefficient a, the exponential coefficient h, the horizontal translation d and the vertical translation c of the exponential function, and the model formula is The model parameters of the polynomial model are the coefficients p1, p2, p3,... p n (n=3 or 4), and the model formula is In the above model formulas, t represents the aging time, and y represents the key sensitive device performance parameter.

[0040] S3.2. Identifying the model parameters using a particle swarm optimization algorithm; the objective function of the particle swarm optimization algorithm is the root mean square error between the model output performance parameter value obtained by parameter fitting and the actual performance parameter value of the device obtained by aging test; S3.3. For each performance parameter of the device, compare the objective function values identified by the particle swarm optimization algorithm under different model formulas, and select the model formula with the smallest error as the performance parameter life model; S3.4. For each performance parameter of the device, compare the life predicted by the performance parameter life model determined in S3.3 with the life corresponding to the performance parameter cutoff criterion, and take the smaller value as the expected life of the performance parameter; S3.5, compare the lifetimes predicted by each performance parameter lifetime model of the device at the same time, and take the minimum value as the lifetime prediction value of the device.

[0041] Particle Swarm Optimization (PSO) algorithm is a kind of optimization algorithm based on swarm intelligence, which can effectively solve the problem of identifying various parameters of device lifetime model. The principle of identifying parameters of device lifetime model by particle swarm optimization algorithm is as follows: particle swarm optimization algorithm simulates the behavior of bird foraging, particles move in the solution space, and constantly update their positions through information exchange between each other to find the optimal solution. Each particle represents a potential solution, and its position is composed of multiple parameters. In the context of parameter identification, these parameters correspond to the parameters of the device lifetime model, such as the slope and intercept in the linear lifetime model. In this embodiment, the meanings of the parameters of the particle swarm algorithm are as follows: position x represents the parameters to be optimized in the model, velocity v represents the speed of change of x, optimal position P i is the local optimal solution of x, and global optimal position g is the global optimal solution of x.

[0042] The specific steps of identifying parameters of device lifetime model by particle swarm algorithm are as follows: (1) Initialize the particle swarm Randomly generate N particles (candidate solutions), N is 70 in this embodiment; the dimension of each particle is the same as the number of model parameters to be identified, and the lifetime model parameters to be identified in this embodiment are: slope k and intercept b (linear function); coefficient a, exponential coefficient h, horizontal translation d and vertical translation c (exponential function); polynomial coefficients p1, p2, p3,... p n (n=3 or 4). Therefore, the dimension of each particle in this embodiment is 2, 3 or 4; each particle is randomly initialized with a position x and a velocity v, where the values of position and velocity are affected by the manually set model parameter interval and identification speed range. In this embodiment, the interval of all parameters is set to [-1, 1]. The position x of each particle represents a possible solution, i.e. a set of parameters of the lifetime model.

[0043] Record the historical optimal position P i and the global optimal position g of each particle.

[0044] (2) Define the objective function The purpose of the PSO algorithm in this embodiment is to optimize the parameters of the device lifetime model, so that more device test data points are distributed on the model curve. Therefore, the objective function is the root mean square error between the performance parameter (i.e. the value of the dependent variable calculated by the model, and the independent variable is the aging time, such as the performance parameters of the capacitor including capacitance and ESR) obtained by parameter fitting and the actual performance parameter value of the device obtained by aging test.

[0045] (3) Iterative update For each particle i and each iteration t: 1) Update velocity: Where w is the inertia weight, c1 and c2 are the individual and group learning factors, respectively, and r1 and r2 are random numbers, P i is the individual optimal position, g is the global optimal position, x i is the current position of the particle, v i is the current velocity of the particle.

[0046] 2) Update position: 3) Update individual optimal After each update of the position, the new particle position is brought into the objective function to calculate the fitness value of the particle, and the particle with a smaller fitness value is better. If the new position is better than P i , then P i = x i (t+1).

[0047] 4) Update global optimal If the new position is better than g, then g = x i (t+1).

[0048] Repeat the above steps until the termination condition is met. The termination condition of the embodiment is to reach 150 iterations.

[0049] After the particle swarm algorithm is calculated, the parameters of the device life model are optimized, and the completed device life model is obtained.

[0050] In the process of training the model parameters using the PSO algorithm, various possible model structures are used, including linear functions, exponential functions, and polynomial functions. When the PSO algorithm completes the training, the minimum root mean square error of the training under different model structures is compared, and the model structure with the smallest error is selected as the device life model. This method can greatly improve the accuracy of the device life model and avoid poor fitting effect caused by improper model structure selection.

[0051] Through the above training, the typical life model obtained is as follows: the typical linear function life model has the capacitance life model of electrolytic capacitor, for example, the capacitance model of 47uf electrolytic capacitor is y=-0.6336t+45.7281. The typical exponential function life model has the equivalent series resistance (ESR) life model of electrolytic capacitor and the on-resistance life model of transistor, for example, the ESR model of 47uf electrolytic capacitor is , the on-resistance lifetime model of BAT54S Schottky diode is , the on-resistance lifetime model of W14NK50Z MOS is . The typical polynomial function lifetime model is the reverse leakage current lifetime model of the transistor, for example, the reverse leakage current model of BYV72EW diode is .

[0052] After the device lifetime model is constructed, the remaining effective lifetime value of the device can be evaluated according to the current performance parameters of the device. For example, the current performance of the 47uf electrolytic capacitor is 42uf, according to the model y=-0.6336t+45.7281, the current service time of the capacitor is 5.88 years, and the service time corresponding to the cutoff criterion 37.6uf is 12.83 years, so the remaining lifetime is 12.83-5.88=6.95 years.

[0053] S4. The minimum value of the lifetime prediction value of all key sensitive devices in the instrument control board card is obtained, that is, the lifetime evaluation value of the board card is obtained.

[0054] In the above process, in order to further improve the accuracy of the lifetime model, the degradation mechanism analysis of the device is carried out before the aging test, and the key performance parameters related to the performance degradation of the device are obtained, so as to reduce the number of device performance parameters collected in the aging test; then, the particle swarm optimization algorithm is used to fit the lifetime model of different devices respectively, and the performance parameters are screened according to the root mean square error of the model as an index, and finally a number of performance parameters strongly related to the aging state of the device are determined for lifetime modeling. On this basis, the device aging test is carried out, which greatly reduces the test cost on the basis of ensuring the effectiveness of the data.

[0055] The device degradation mode and degradation influence obtained by the device degradation mechanism analysis are shown in Table 1.

[0056] Table 1 Degradation rule of key sensitive devices of instrument control board card The method simulates the on-site service working condition of the device to the greatest extent through the composite stress aging test; then, the device aging lifetime model is developed according to the test data, the actual service lifetime of the device is extrapolated and calculated in combination with the performance parameter failure criterion of the device, the obtained result is compared with the lifetime failure time statistically obtained on site, and the lifetime deviation of the typical device is less than 10%. The lifetime of the key devices on the board card is fused and analyzed, so as to evaluate the lifetime prediction value of the whole board card, and the lifetime deviation of the board card is less than 15% compared with the average failure time of the card statistically obtained on site.

[0057] The embodiment also proposes a simulation aging test device for simulating the composite stress aging test, as shown in Figure 4 .

[0058] The device can simulate the actual service conditions of the instrument control board of a nuclear power plant, continuously apply complex stresses such as electrical stress, high humidity stress, high temperature stress, cold and hot impact stress, and vibration stress to components, and regularly collect component performance parameters. The device is composed of a program-controlled power supply 1, a protection resistor 2, a pulse current sensor 7, an air thermal aging box 3, a component clamp 5, a high-voltage probe 4, an oscilloscope 6, and an industrial computer 8.

[0059] The main test circuit is composed of the program-controlled power supply 1, the protection resistor 2, the component clamp 5, and the pulse current sensor 7, and the above devices are in series in the main circuit, which mainly provides electrical stress for the components. The protection resistor 2 is used to protect the components from excessive current and voltage, and to protect the power supply from being burned out in the case of short circuit failure of the components. The pulse current sensor 7 is used to monitor the current level in the main circuit.

[0060] The air thermal aging box 3 provides high temperature, cold and hot impact, humidity, and other stresses for the components. During the aging test, the component clamp 5 is clamped on the component clamp 5 in the air thermal aging box 3, and the main test circuit line enters and exits through the air thermal aging box vent. The air thermal aging box 3 can be selected from market products, and a 750V and below power frequency single-phase alternating current power supply should be selected as the input and good grounding should be maintained.

[0061] The component clamp 5 is used to fix and support the components, and to connect the components well in the main circuit. The component clamp 5 is placed in the air thermal aging box 3, and the components are clamped by the clamp. The clamp is made of steel material and includes a base and end screws. The base is used to rest the components, and the end screws are used to tighten the components and power on.

[0062] The high-voltage probe 4 is connected in parallel across the component clamp, and is used to collect the output voltage performance parameters of the components in the main circuit, such as voltage average value, voltage effective value, and voltage ripple value. The high-voltage probe 4 can be selected from market products.

[0063] The oscilloscope 6 is connected in parallel across the component clamp, and is used to collect the voltage waveform of the components in the main circuit in real time, and is used to calculate the voltage ripple value, rise time, fall time, and protection time. The oscilloscope 6 can be selected from market products.

[0064] The industrial computer 8 is connected with the program-controlled power supply 1, the air thermal aging box 3, the oscilloscope 6, the current sensor 7 and the high-voltage probe 4, is responsible for the control logic of the whole system, and issues control instructions to the device, including calling the program-controlled power supply 1 to output a certain level of voltage and current signal, calling the air thermal aging box 3 to generate a corresponding temperature and humidity environment according to the input service condition data; meanwhile, the device test data are collected, including calling the oscilloscope to collect the voltage signal (image and time sequence data) in the main loop, calling the current sensor to collect the current signal in the main loop, calling the high-voltage probe to collect the high-voltage signal in the main loop, and performing data analysis and calculation.

[0065] The steps of the aging test performed by using the above device are as follows: S201. Test device assembly. The device to be tested is fixed in the air thermal aging box through the component fixture, the bolt and nut structure is tightened, and it is ensured that the device will not shake and fall when various stresses are applied.

[0066] S202. Configuration of control module and data acquisition module. The two together can be called the main test loop. First, the program-controlled power supply 1, the protection resistor 2, the component fixture 5 and the pulse current sensor 7 are connected in series to form the main loop, then the industrial computer is connected with the program-controlled power supply 1 and the air thermal aging box 3 to form the control module, and then the industrial computer is connected with the oscilloscope 6, the current sensor 7 and the high-voltage probe 4 to form the data acquisition module.

[0067] S203. Module parameter setting. The service condition data are input in the industrial computer, the industrial computer automatically sets the temperature and humidity parameters of the air thermal aging box 3, and sets the voltage and current size of the program-controlled power supply, and performs joint aging test. During the test, the voltage waveform and characteristic parameters of the device are recorded on the oscilloscope, the current parameters passing through the device are recorded on the current sensor, and the high-voltage information in the loop is recorded on the high-voltage probe.

[0068] S204. Data acquisition. The data acquisition module returns the test data collected to the industrial computer.

[0069] The simulated aging test device realizes free combination of multi-dimensional stress aging acceleration, maximally simulates the field service condition of the device, thereby ensuring that the performance degradation trajectory of the component is maximally restored to the performance degradation trend in the actual service, and greatly improving the accuracy of life prediction.

[0070] The above merely describes specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which shall be covered within the protection scope of the present application.

Claims

1. A method for evaluating the aging life of an instrument control board card under complex stress, characterized in that, The method comprises the following steps: S1, analyzing and determining the key sensitive components of the instrument control board card; S2, performing a simulation aging test on the key sensitive components of the instrument control board card under multiple stresses, and continuously collecting performance parameters of the key sensitive components; S3, constructing a device life model with aging time as the independent variable and device performance parameters as the dependent variable, and predicting the device life; S3 specifically comprises: S3.1, constructing a device life model formula, which includes a linear function, an exponential function, and a polynomial function; for each performance parameter of each device, the three model formulas are used for fitting, respectively; S3.2, identifying the model parameters using a particle swarm optimization algorithm; the objective function of the particle swarm optimization algorithm is the root mean square error between the model output performance parameter value obtained by parameter fitting and the actual performance parameter value of the device obtained by the aging test; S3.3, for each performance parameter of the device, comparing the objective function values obtained by the particle swarm optimization algorithm under different model formulas, and selecting the model formula with the smallest error as the performance parameter life model; S3.4, for each performance parameter of the device, comparing the life predicted by the performance parameter life model determined in S3.3 with the life corresponding to the performance parameter cutoff criterion, and taking the smaller value as the expected life of the performance parameter; S3.5, comparing the life predicted by the life model of each performance parameter of the device at the same time, and taking the minimum value as the life prediction value of the device; S4, taking the minimum value of the life prediction values of all key sensitive components as the life evaluation value of the board card.

2. The method of claim 1, wherein the method further comprises: During the simulation aging test, the key sensitive components of the instrument control board card are continuously subjected to composite stress, including electrical stress, high humidity stress, high temperature stress, cold and hot impact stress, and vibration stress.

3. The aging life assessment method for instrumentation and control boards under composite stress as described in claim 1, characterized in that, Before the simulation aging test of the device, stress data under the actual service conditions of the nuclear power plant instrument control board card is collected; during the aging test, the stress data under the actual service conditions is used as the test input to apply composite stress to the device.

4. The aging life assessment method for instrumentation and control boards under composite stress as described in claim 1, characterized in that, The key sensitive components of the instrument control board card are selected according to the failure characteristics of the device itself and the circuit environment in which it is located.

5. The method of claim 4, wherein the method further comprises: determining the aging life of the instrumented card under complex stress by using the aging life model. The key sensitive components of the instrument control board card include electrolytic capacitors, high-power diodes, optical couplers, transistors, and field effect tubes.

6. The method for evaluating the aging life of an instrument control board card under complex stress according to claim 1, wherein, 47uf direct insertion type aluminum electrolytic capacitor capacity life model is: y = -0.6336t + 45.7281, equivalent series resistance ESR life model is ; the on-resistance life model of the Schottky diode BAT54 is , the on-resistance life model of the MOSFET tube W14NK50Z is ; the reverse leakage current model of the BYV72EW diode is .

7. The method of claim 1 to 6, wherein, The steps of identifying the parameters of the life model using the particle swarm optimization algorithm are as follows: (1) Initialize the particle swarm Randomly generate N particles, each particle has the same dimension as the number of model parameters to be identified; each particle is randomly initialized with a position x and a velocity v; the position x of each particle represents a possible solution, i.e. a set of parameters of the life model; record the history of the optimal position P of each particle i and the global optimal position g; (2) Define the objective function The objective function is the root mean square error between the model output performance parameter obtained by parameter fitting and the actual performance parameter value of the device obtained by the aging test; (3) Iterative update For each particle i and each iteration t: 1) Update the velocity: where w is the inertia weight, c1, c2 are the individual and group learning factors, r1, r2 are random numbers, P i is the individual optimal position, g is the global optimal position, x i is the current position of the particle, v i is the current speed of the particle; 2) Update the position: 3) Update the individual optimum After each update of the position, the new position of the particle is brought into the objective function to calculate the fitness value of the particle, the smaller the fitness value is, the better the particle is, if the new position is better than P i , then P i = x i (t+1); 4) Update the global optimum If new position is better than g, then g = x i (t+1); Repeat the above steps until the termination condition is met.

8. The method of claim 7, wherein the method further comprises: determining the aging life of the instrumented card under complex stress by using the aging life model. The dimension of each particle is 2, 3 or 4 when the particle swarm is initialized; wherein, for a linear function, the dimension of the particle is 2, and the model parameters to be identified are the slope k and the intercept b of the linear function; for a polynomial function, the dimension of the particle is 3 or 4, and the model parameters to be identified are the coefficients p1, p2, p3 and p4 of the polynomial function; for an exponential function, the dimension of the particle is 4, and the model parameters to be identified are the coefficient a, the exponential coefficient h, the horizontal translation d and the vertical translation c of the exponential function; when the particle swarm is initialized, the value range of each particle position and speed is set as [-1, 1]; and the termination condition of the particle swarm optimization algorithm is that 150 iterations are reached.

9. The method of claim 7, wherein the method further comprises: determining the aging life of the instrumented card under stress by using the following equation: ###0002### wherein, T is the aging life of the instrumented card under stress, T0 is the aging life of the instrumented card under no stress, and K is the stress factor. Before the aging test, the device degradation mechanism analysis is carried out, the key performance parameters related to the device performance degradation are obtained, and the parameter data of these performance parameters are collected during the aging test.

10. An aging test device for an instrument control board card under complex stress, characterized in that, The system comprises a programmed power supply (1), a protective resistor (2), a pulse current sensor (7), an air thermal aging oven (3), a component fixture (5), a high-voltage probe (4), an oscilloscope (6) and an industrial computer (8). The programmed power supply (1), the protective resistor (2), the component fixture (5) and the pulse current sensor (7) constitute a main test loop, and are in series connection in the main test loop, and the main test loop provides electrical stress for the component; the protective resistor (2) is used for protecting the component and the programmed power supply from being burnt out; and the pulse current sensor (7) is used for monitoring the current level in the main loop. The air thermal aging oven (3) is used for providing high temperature, cold and hot impact and humidity stress for the component, and when the aging test is carried out, the component is clamped on the component fixture (5) in the air thermal aging oven (3); The high-voltage probe (4) is connected in parallel across the component fixture (5), and is used for collecting the output voltage performance parameter of the component in the main loop; The oscilloscope (6) is connected in parallel across the component fixture, and is used for collecting the voltage waveform of the component in the main loop; The industrial computer (8) is connected with the programmed power supply (1), the air thermal aging oven (3), the oscilloscope (6), the current sensor (7) and the high-voltage probe (4), is responsible for the control logic of the whole system, issues control instructions, and collects the device test data.

Images