Electronic system diagnostic support system, electronic system diagnostic support method, and electronic system diagnostic support program

Abstract

To provide a technique which supports efficient deterioration diagnosis of an electronic system.SOLUTION: An electronic system diagnosis support system which supports diagnosis of an electronic system includes a memory which stores therein a software program and data and a processor which executes the software program by using the data. The processor performs simulation of circuit operations in a design of the electronic system and performs simulation of circuit operations with components deteriorated by simulation of a prescribed deterioration and takes an output item in which a variation in output due to the deterioration satisfies a prescribed condition, as a candidate for observation point, on the basis of results of the simulations and learns a relation between the deterioration and the variation to create a component deterioration model for estimating the deterioration from the variation.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a technology for assisting in the diagnosis and / or design of electronic systems.

Background Art

[0002] Various electronic systems with various functions are installed in various products such as automobiles and industrial equipment. As products become more highly functional, electronic systems also become more highly functional, and the scale of the circuits of electronic systems is increasing. When components that make up the circuit of an electronic system deteriorate, the performance of the electronic system decreases, so a monitor circuit for detecting component deterioration is provided. Patent Document 1 discloses a method for performing deterioration diagnosis using a model showing the deterioration tendency of components.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] As the device to be diagnosed becomes more complex and large-scale, the number of electronic systems installed therein increases, and the number of components installed in the electronic systems also increases. As a result, for diagnosis such as deterioration state estimation, it is necessary to arrange a large number of monitor circuits, collect a large amount of data thereby, and process the large amount of data, increasing the processing load for diagnosing the electronic system.

[0005] One object included in the present disclosure is to provide a technology for assisting in efficient deterioration diagnosis of electronic systems.

Means for Solving the Problems

[0006] An electronic system diagnostic support system according to one aspect of the present disclosure is an electronic system diagnostic support system that supports the diagnosis of an electronic system, comprising: a memory for storing software programs and data; and a processor for executing the software programs using the data, wherein the processor performs a simulation of the circuit operation of the electronic system in design and a simulation of the circuit operation when components have deteriorated by simulating predetermined deterioration, and based on the results of the simulation, selects output items whose output fluctuations due to the deterioration satisfy predetermined conditions as candidates for observation locations, learns the relationship between the deterioration and the fluctuations, and generates a component deterioration model that estimates deterioration from the fluctuations. [Effects of the Invention]

[0007] According to the present invention, it becomes possible to support the efficient deterioration diagnosis of electronic systems. [Brief explanation of the drawing]

[0008] [Figure 1] This is a schematic block diagram of the electronic system diagnostic support system in Example 1. [Figure 2] This is a block diagram showing the configuration of a computer. [Figure 3] This is a flowchart of the component degradation model generation process. [Figure 4] This figure shows an example of output item information. [Figure 5] This figure shows an example of degradation factors and degradation parameter data. [Figure 6] This is a flowchart for the process of extracting potential monitoring locations. [Figure 7] This is a diagram to explain the output fluctuation rate data. [Figure 8] This figure shows an example of candidate information for monitoring locations. [Figure 9] This is a block diagram showing the configuration of an electronic system. [Figure 10] This is a flowchart for the degradation monitoring process. [Figure 11]This is a schematic block diagram of the electronic system diagnostic support system in Example 2. [Figure 12] This figure shows an example of design change candidate data. [Modes for carrying out the invention]

[0009] Embodiments of the present invention will be described below with reference to the drawings. [Examples]

[0010] Figure 1 is a schematic block diagram of the electronic system diagnostic support system in Example 1.

[0011] Referring to Figure 1, the electronic system diagnostic support system 1 includes a computer 10 and an electronic system 20.

[0012] The electronic system 20 is a computer that is embedded in or connected to edge devices such as automobiles, home appliances, manufacturing equipment, inspection equipment, and industrial robots, and comprises memory for storing software programs and data, and a processor for executing software programs using the data. In addition to functions related to the operation of edge devices, the electronic system 20 also has a function for diagnosing the deterioration of parts that constitute the functions related to the operation of edge devices. The parts that constitute the functions of the edge device are not shown in Figure 1. The parts that constitute the deterioration diagnosis function of the electronic system 20 include an internal monitor unit 21, a deterioration estimation unit 22, a deteriorated component determination unit 24, and a performance deterioration alert unit 25. Each of these parts is realized by the processor executing a software program. The functions of each part will be described later.

[0013] The computer 10 is connectable to the electronic system 20 via the communication network 91, is a computer that manages the electronic system 20, and may be integrally configured with a control system (not shown). The computer 10 has a deterioration model generation unit 12 and an internal monitor candidate output unit 13. These units are realized by a processor executing a software program. The functions of each unit will be described later.

[0014] The deterioration model generation unit 12 generates a component deterioration model 23, which is a machine learning model for estimating the deterioration of the components of the electronic system 20 based on the deterioration factor - deterioration parameter data 11 representing the factors that deteriorate the characteristics of each component and the changes in characteristics during deterioration, and transmits it to the electronic system 20. The component deterioration model 23 is stored in the electronic system 20 and is used for the processes described later. The internal monitor candidate output unit 13 enumerates and outputs candidates for observation locations where output signals should be measured in order to appropriately diagnose the deterioration of the electronic system 20. The candidates for observation locations are used when determining the observation locations of the electronic system 20. The detailed configurations and operations of the deterioration model generation unit 12 and the internal monitor candidate output unit 13 will be described later.

[0015] In the electronic system 20, the internal monitor unit 21 is a circuit that measures signals at the observation locations provided in each part of the electronic system 20 and acquires the output values of the output items. The components constituting the electronic system 20 include resistors and capacitors. The internal monitor unit 21 measures the voltage value, current value, etc. of the signals at the terminal parts of the components. The deterioration estimation unit 22 estimates the deterioration of each component from the measurement results of the internal monitor unit 21 using the component deterioration model 23. The deteriorated component determination unit 24 determines whether or not deterioration has occurred in each component based on the estimation results by the deterioration estimation unit 22. The performance deterioration alert unit 25 issues a warning if deterioration has occurred in the component. The warning is presented to the user 92 by means of lamp lighting, output of an alarm sound, display on the screen, etc. The detailed configuration and operation of the deterioration estimation unit 22 will be described later.

[0016] Although one electronic system 20 is shown as an example in FIG. 1, there may be a plurality of them. Also, although an example where the computer 10 and the electronic system 20 are connected via the communication network 91 is shown, the computer 10 and the electronic system 20 may be configured to be connected without going through the communication network 91. Further, the communication network 91 is a communication network that provides a wired line, a wireless line, or a communication line combining them, and the physical medium and the communication method are not limited.

[0017] FIG. 2 is a block diagram showing the configuration of the computer. In the computer 10, the degradation model generation unit 12 includes a degradation simulation parameter extraction unit 34, a simulator unit 35, an output calculation unit 36, a parameter normalization unit 37, an output normalization unit 49, and a machine learning unit 40, and the internal monitor candidate output unit 13 includes a contribution rate calculation unit 42. As a hardware configuration not shown, the computer 10 has a memory for storing software programs and data, and a processor for executing software programs using the data. The degradation simulation parameter extraction unit 34, the simulator unit 35, the output calculation unit 36, the parameter normalization unit 37, the output normalization unit 49, the machine learning unit 40, and the contribution rate calculation unit 42 are realized by the processor executing a software program. The operations of these units will be described in the description of the operation of the entire computer 10 using a flowchart.

[0018] Hereinafter, the operation of the computer 10 will be described.

[0019] FIG. 3 is a flowchart of the component degradation model generation process. The component degradation model generation process is a process in which the degradation model generation unit 12 generates a component degradation model 23.

[0020] The degradation model generation unit 12 previously acquires a list of components to be degraded 31, circuit data 32, and output item information 33, and records them in the memory.

[0021] The list of components to be detected for degradation 31 is a list of components in the electronic system 20 that are subject to degradation detection. The circuit data 32 is data describing the circuit configuration of the electronic system 20 using a circuit diagram and netlist, etc. The output item information 33 is information indicating the output items whose output values ​​are measured in the electronic system 20. The output items may be those whose output values ​​are measured directly, or they may be calculated from measured values. For example, voltage values, current values, or their RMS values ​​and integral values ​​measured at the input or output terminals of component nodes in the circuit diagram can be used as output items.

[0022] Figure 4 shows an example of output item information. In this embodiment, the electronic system 20 has several operating patterns, A, B, C, etc., and the output items whose output values ​​are measured differ depending on the operating pattern. The output values ​​of the output items can be measured by the electronic system 20, and can also be calculated by simulation of circuit operation and / or by analytical formulas such as output transient analysis (time series waveform) or AC analysis (frequency response).

[0023] Referring to Figure 4, the output item information 33 records the item and operation pattern associated with each output item. The "Item" indicates the name of the output item. The "Operation Pattern" lists the operation pattern in which the output value of the output item is measured. For example, in operation pattern A, the maximum voltage, minimum voltage, and average voltage are measured. In operation pattern B, the maximum voltage, minimum voltage, average voltage, and rise time are measured. In operation pattern C, the maximum voltage, minimum voltage, average voltage, and bandwidth (cutoff) are measured. Bandwidth (cutoff) is the bandwidth of the frequency components contained in the signal. This bandwidth can be expressed, for example, as the cutoff frequency of a filter circuit. Note that the output items are not limited to voltage values ​​such as maximum voltage, minimum voltage, and average voltage shown in Figure 4, but may also include current values ​​such as maximum current, minimum current, and average current as output items.

[0024] Furthermore, the degradation model generation unit 12 can refer to the degradation factor and degradation parameter data 11.

[0025] Figure 5 shows an example of degradation factors and degradation parameter data.

[0026] Referring to Figure 5, the degradation factor / degradation parameter data 11 records degradation factors and influencing parameter information for each component. "Degradation factors" are listed as factors that degrade the characteristics of the component. "Influencing parameter information" shows information indicating the change in characteristics during degradation. For example, a degradation factor for resistors is burnout due to surges, and the degradation of resistors caused by burnout due to surges is listed as an increase in resistance. Another degradation factor for resistors is alteration due to overload, and the degradation of resistors caused by alteration due to overload is listed as a decrease in resistance. Furthermore, a degradation factor for capacitors is thermal stress, and the degradation of capacitors caused by thermal stress is listed as a decrease in capacitance and an increase in ESR (equivalent series resistance). Another degradation factor for capacitors is overvoltage, and the degradation of capacitors caused by overvoltage is listed as a decrease in capacitance and an increase in ESR. The information for each component registered in the degradation factor / degradation parameter data 11 may be obtained and used from information published by the component manufacturer, or it may be used from in-house information on FTA / FMEA or assets accumulated in the past.

[0027] In Figure 3, in step S101, the simulator unit 35 simulates the circuit operation of the electronic system 20 using the design values ​​(e.g., catalog values) of each component based on the circuit data 32. For example, a simulator such as SPICE can be used to simulate the circuit operation. Alternatively, machine learning using surrogate models may be used to simulate the circuit operation.

[0028] Next, in step S102, the output calculation unit 36 ​​calculates the output value of each output item shown in the output item information 33 based on the simulation results from the simulator unit 35. Hereinafter, the output value calculated here will also be called the "non-degraded output value". The output calculation unit 36 ​​records the non-degraded output value of each output item as the nominal output result 38.

[0029] Next, in step S103, the degradation simulation parameter extraction unit 34 extracts components to be detected for degradation based on the circuit data 32 and the list of components to be detected for degradation 31. Component degradation manifests as changes in parameters that indicate the performance of the component. For example, the degradation of a resistor manifests as a change in its resistance value, which is one of the resistor's parameters. For example, since the degradation factor / degradation parameter data 11 shows a degradation pattern in which the resistance value of a resistor increases due to burnout caused by a surge, the resistors on the circuit diagram in the circuit data 32 can be designated as locations to be detected for degradation.

[0030] Next, in step S104, the degradation simulation parameter extraction unit 34 extracts parameter values ​​that simulate degradation due to various degradation factors in the component to be detected, based on the degradation factor / degradation parameter data 11. Hereinafter, the parameter values ​​that simulate degradation will also be called degradation simulation parameter values. As described above, the degradation of a component appears as a change in the parameters of that component, so the degradation of a component can be simulated by changing the parameters of the component. For example, the degradation of a resistor can be simulated by increasing or decreasing its resistance value by 10%. Since the degradation factor / degradation parameter data 11 shows a type of degradation in which the resistance value of a resistor increases due to burnout caused by a surge, if a resistor is used as the component to be detected for degradation, it is possible to extract resistance values ​​of the resistor that have been increased in various ways to simulate degradation.

[0031] Next, in steps S105 to S106, the simulator unit 35 simulates the circuit operation of each component targeted for degradation detection, degraded by its respective degradation factor, until the simulation using all the degradation simulation parameter values ​​extracted in step S104 is completed.

[0032] After step S106, in step S107, the output calculation unit 36 ​​calculates the output value of each output item based on the simulation results. Hereinafter, the output value calculated here will also be called the "degraded output value".

[0033] Next, in step S108, the output normalization unit 39 normalizes the output values ​​of the output items when they are degraded, and the parameter normalization unit 37 normalizes the parameter values ​​that simulate degradation. For example, the output values ​​when they are degraded can be normalized by the output values ​​when they are not degraded, and the parameter values ​​can be normalized by the design values ​​of each component.

[0034] In step S105, when the simulator unit 35 completes the simulation using all the degradation simulation parameter values ​​extracted in step S104, in step S109, the machine learning unit 40 learns the relationship between the fluctuation of component parameters due to degradation factors and the fluctuation of the output value from the non-degraded output value to the degraded output value, and generates a component degradation model 23 that estimates the degradation factors from the output fluctuations. The learning here is machine learning, such as reinforcement learning or deep learning.

[0035] The normalized degradation output value and the parameter value simulating the normalized degradation, generated in the process of generating the component degradation model 23 by the degradation model generation unit 12 as described above, are recorded as output fluctuation rate data 41 in the internal monitor candidate output unit 13 and used in the processing of the internal monitor candidate output unit 13.

[0036] Figure 6 is a flowchart of the monitoring location candidate extraction process. The monitoring location candidate extraction process is a process in which the internal monitoring location candidate output unit 13 extracts monitoring location candidate information 43. The monitoring location candidate information 43 is useful information for determining the locations where output values ​​should be measured in order to determine the deterioration of parts.

[0037] The internal monitor candidate output unit 13 pre-generates output fluctuation rate data 41 from the normalized degradation output values ​​of the output items calculated by the output normalization unit 39 and the normalized degradation simulation parameter values ​​calculated by the parameter normalization unit 37, and records it in memory.

[0038] Figure 7 is a diagram illustrating the output fluctuation rate data.

[0039] Referring to Figure 7, the upper part of explanatory diagram 50 shows an example of output fluctuation rate data 41. The output fluctuation rate data 41 records the component name, degradation simulation, and output item fluctuation rate in association. "Component name" shows the name of each component targeted for degradation detection in the electronic system 20. In the example in Figure 7, capacitor C1 and resistor R1 are shown. "Degradation simulation" records the degradation simulation parameter value for the component. In the example in Figure 7, the degradation simulation parameter value for capacitor C1 is shown to be a capacitance value reduced by 10% from the design value. Also, the degradation simulation parameter values ​​for resistor R1 are shown to be a resistance value reduced by 10% from the design value and a resistance value increased by 10% from the design value. "Output item fluctuation rate" records the fluctuation rate of each output item. The fluctuation rate is an index that represents the ratio of the amount of change in the output item value to the amount of change in the parameter when the parameter of the component is changed from the design value to the degradation simulation parameter value. The larger the absolute value of the fluctuation rate, the more significantly the output item changes in response to component degradation.

[0040] In the middle section of explanatory figure 50, an RC filter circuit 51 is shown as an example of the target circuit, and in the lower section, waveforms 52 of the input and output voltages of the RC filter circuit are shown.

[0041] The RC filter circuit 51 is an RC low-pass filter consisting of a resistor R1 and a capacitor C1, with the output voltage Vo measured by resistor Ro. As shown in the table above, resistor R1 and capacitor C1 are the components targeted for degradation detection. The design values ​​for the parameters of each component are R1 = 160Ω and C1 = 1μF. Also, resistor Ro = 1kΩ.

[0042] Here, as shown by the dashed line in waveform 52, an input voltage Vi with a period of 0.001 sec, a duty cycle of 0.5, and an amplitude of 1 V is assumed. While only one operating pattern is shown here as an example, multiple patterns are possible.

[0043] The maximum and minimum voltages are the maximum and minimum values ​​of the output voltage Vo shown in the circled portion of waveform 52. RMS(0~10ms) is the effective value of the output voltage Vo between 0msec and 10msec. The integral value(0~0.5ms) is the integral value of the output voltage Vo between 0msec and 0.5msec. For example, when the capacitance of capacitor C1 is reduced by 10%, the fluctuation rate of the maximum voltage is 0.86%, the fluctuation rate of the minimum voltage is -32.90, the RMS is 1.45%, and the integral value is -0.01%.

[0044] In Figure 6, in step S201, the contribution rate calculation unit 42 refers to the output fluctuation rate data 41 and divides the fluctuation amount of each output item for each component in relation to the simulated deterioration of that output item into a set of fluctuation amounts with positive values ​​and a set of fluctuation amounts with negative values. Hereinafter, the set of fluctuation amounts with positive values ​​may be called the positive set, and the set of fluctuation amounts with negative values ​​may be called the negative set.

[0045] Next, in step S202, the contribution rate calculation unit 42 calculates the contribution rate of each fluctuation belonging to the positive set. Hereinafter, this contribution rate may be referred to as the degradation detection contribution rate. The contribution rate of each fluctuation belonging to the positive set is calculated as the ratio of that fluctuation to the total value of the fluctuations in the positive set.

[0046] Next, in step S203, the contribution rate calculation unit 42 calculates the contribution rate of each fluctuation belonging to the negative set. The contribution rate of each fluctuation belonging to the negative set is calculated as the ratio of that fluctuation to the total value of the fluctuations in the negative set.

[0047] Next, in step S204, the contribution rate calculation unit 42 calculates the average absolute value fluctuation of each output item. The average absolute value fluctuation is the average of the absolute values ​​of the fluctuation rate of the output item. In the example of output fluctuation rate data 41 shown in Figure 7, the average absolute value fluctuation of the maximum voltage is 1.74, which is the average of 0.86, 2.17, and 2.19. Here, the average absolute value fluctuation is calculated as an example, but as another example, the total absolute value fluctuation may be calculated. The total absolute value fluctuation is the sum of the absolute values ​​of the fluctuation rate of the output item. Both the total absolute value fluctuation and the average absolute value fluctuation are indicators of the magnitude of the fluctuation of the output item. The table of output fluctuation rate data 41 in Figure 7 shows both the total absolute value fluctuation and the average absolute value fluctuation.

[0048] Returning to Figure 6, in step S205, the contribution rate calculation unit 42 sorts the output items in descending order of average fluctuation amount and outputs the data calculated in steps S202 to S204 as candidate monitoring location information 43. Furthermore, the contribution rate calculation unit 42 may output the output items selected based on the candidate monitoring location information 43 as candidate monitoring locations.

[0049] Figure 8 shows an example of candidate information for monitoring locations.

[0050] Referring to Figure 8, the monitor location candidate information 43 records the average absolute value fluctuation and the degradation detection contribution rate for each degradation simulation parameter value for each output item, in descending order of average absolute value fluctuation. The minimum voltage has the highest average absolute value fluctuation, followed by RMS, maximum voltage, and integral value. For example, looking at the minimum voltage, the average absolute value fluctuation is 30.56%. Furthermore, looking at the minimum voltage, the degradation detection contribution rate for the degradation simulation parameter value that reduces the capacitance of capacitor C1 by 10% is 53.65%, the degradation detection contribution rate for the degradation simulation parameter value that reduces the resistance of resistor R1 by 10% is 46.35%, and the degradation detection contribution rate for the degradation simulation parameter value that increases the resistance of resistor R1 by 10% is 100%.

[0051] By monitoring output items that exhibit large output fluctuations in response to changes in simulated degradation parameter values, the probability of detecting degradation can be increased. Furthermore, the closer the degradation detection contribution rate is to 100%, the higher the probability of estimating the degraded component and parameter when degradation is detected. Therefore, in the example in Figure 8, by monitoring the minimum voltage with the highest average absolute fluctuation, which is an index value indicating the degree of output value fluctuation in response to changes in simulated degradation parameter values, the probability of detecting degradation can be increased. Since the degradation detection contribution rate is 100% when the resistance value of resistor R1 at the minimum voltage increases by 10%, the 10% increase in the resistance value of resistor R1 can be estimated with high probability by measuring the minimum voltage. From this, let's first select the minimum voltage of the output voltage Vo as the monitoring point and the output item to be monitored. In that case, since the degradation detection contribution rate when the capacitance of the capacitor C1 with the lowest voltage decreases by 10% and the degradation detection contribution rate when the resistance of resistor R1 decreases by 10% are both approximately 50%, in order to uniquely estimate the degraded component and parameter with a high probability, it is necessary to further select the integral value of the output voltage Vo as the monitoring point and the output item to be monitored.

[0052] Figure 9 is a block diagram showing the configuration of the electronic system.

[0053] The electronic system 20 includes an internal monitoring unit 21, a degradation estimation unit 22, a degradation component determination unit 24, and a performance degradation alert unit 25. The degradation estimation unit 22 includes a normalization output calculation unit 62 and a degradation probability inference unit 63. The operation of each of these units will be explained in the description of the degradation monitoring process using a flowchart.

[0054] Figure 10 is a flowchart of the degradation monitoring process. The degradation monitoring process is a process in which the electronic system 20 monitors the degradation of the components that make up the system.

[0055] The electronic system 20 has previously acquired the component degradation model 23 from the computer 10 and stored it in memory. Furthermore, the electronic system 20 assumes that there is no component degradation in its initial state.

[0056] In Figure 10, in step S301, the electronic system 20 in its initial state is activated. Subsequently, in step S302, the electronic system 20 operates according to a predetermined operating pattern, and the internal monitor unit 21 measures the signal at the observation point. Furthermore, in step S303, the normalization output calculation unit 62 acquires or calculates the output value of each output item based on the signal value measured in step S302. The output value acquired or calculated here may hereafter be referred to as the non-degraded measured output value. The normalization output calculation unit 62 records the non-degraded measured output value of each output item as the measured nominal output result 61.

[0057] Next, in step S304, the electronic system 20 operates according to a predetermined operating pattern, and the internal monitor unit 21 measures the signal at the observation point. From this point onward, the electronic system 20 continues to operate, so there is a possibility that some degradation factors may occur and some components may deteriorate. Next, in step S305, the normalization output calculation unit 62 acquires or calculates the output value of each output item based on the signal value measured in step S304. The output value acquired or calculated here may hereafter be referred to as the measured output value. Furthermore, in step S306, the normalization output calculation unit 62 normalizes the measured output value by the undegraded measured output value.

[0058] Next, in step S307, the degradation probability inference unit 63 inputs the normalized measured output values ​​into the component degradation model 23 to estimate the variation in the component's parameters. Here, the probability that a predetermined parameter variation, which signifies degradation of a given component, has occurred is obtained. The probability obtained here may hereafter be referred to as the degradation probability.

[0059] Next, in step S308, the degraded component determination unit 24 compares the degradation probability with a predetermined threshold. If the degradation probability is below the threshold, the process returns to step S304 and continues monitoring. If the degradation probability exceeds the threshold, in step S309, the degraded component determination unit 24 identifies the component corresponding to the degradation probability that exceeded the threshold. Hereafter, the degraded component identified here may be referred to as a degraded component. Next, in step S310, the performance degradation alert unit 25 outputs an alarm warning of the degradation of the degraded component.

[0060] As described above, according to Example 1, a component degradation model 23 is generated to estimate the degradation of components in the electronic system 20, and items to be monitored can be extracted from the output items of the electronic system 20. Therefore, it is not necessary to comprehensively process data from a large number of monitoring points, and the processing load for diagnosing the degradation of the electronic system 20 can be reduced.

[0061] Furthermore, by sending the component degradation model 23 from the computer 10 to the electronic system 20, the electronic system 20 can measure the values ​​of the output items at the monitoring locations within its own device and determine the degree of degradation of the component to be detected. In this case, the computer 10 only needs to collect the determination results from each electronic system 20, so it is possible to reduce the amount of information received from the electronic systems 20. [Examples]

[0062] In Example 2, the electronic system diagnostic support system, in addition to the functions of Example 1, is equipped with a function to extract candidate design changes for improving the electronic system 20 based on the results of the determination of deteriorated components.

[0063] Figure 11 is a schematic block diagram of the electronic system diagnostic support system in Example 2.

[0064] Referring to Figure 11, the electronic system diagnostic support system 1 of Example 2 has basically the same configuration as in Example 1 and includes a computer 10 and an electronic system 20.

[0065] The electronic system 20 of Example 2, in addition to the functions of the electronic system 20 of Example 1, includes a function in which the degraded component determination unit 24 transmits degraded component and degradation parameter information to the computer 10. For example, the degraded component and degradation parameter information includes information on degraded components and parameter fluctuations.

[0066] The computer 10 of Example 2, in addition to the configuration of the computer 10 of Example 1, has a deterioration location factor extraction unit 72 and stores deterioration countermeasure data 71 in advance. The deterioration countermeasure data 71 is data in which design change proposals are registered in advance in association with deteriorated parts and changes in the parameters of those parts. There may be multiple design change proposals registered for a set of deteriorated parts and changes in the parameters of those parts. The design change proposals registered in the deterioration countermeasure data 71 may, for example, be design changes that can reduce the impact of deterioration factors from assets accumulated in the past.

[0067] The deterioration location factor extraction unit 72 receives deteriorated component and deterioration parameter information from the electronic system 20, and based on the deteriorated components and parameter fluctuations indicated in the deteriorated component and deterioration parameter information, it extracts design change proposals by referring to the circuit data 32 and deterioration countermeasure data 71, and records them as design change candidate data 73.

[0068] Figure 12 shows an example of design change candidate data. Figure 12 illustrates design change proposals for when the capacitance of capacitor C1 in the RC filter circuit 51 shown in Figure 7 decreases. The design change candidate data 73 is registered with the order, location, degradation cause, and design change proposal associated with it. The priority of the location to be changed and the degradation cause is indicated as the order. Degradation of capacitor C1 due to thermal stress is ranked first. Degradation of capacitor C1 due to overvoltage is ranked second. One or more design change proposals are registered for each rank. For example, design change proposal 1 for the first rank is to apply a high-temperature product with high heat resistance to capacitor C1. Design change proposal for the second rank is to add a cooling mechanism to cool capacitor C1.

[0069] As described above, according to Example 2, when component degradation is detected during the operation of the electronic system 20, appropriate design modification proposals can be obtained.

[0070] The embodiments described above are illustrative for explaining the present invention and are not intended to limit the scope of the invention to those embodiments only. Those skilled in the art can implement the present invention in various other forms without departing from the scope of the invention.

[0071] Furthermore, the above-described embodiments include the following items. However, the items included in the above-described embodiments are not limited to those listed below.

[0072] (Item 1) An electronic system diagnostic support system for assisting the diagnosis of an electronic system comprises a memory for storing software programs and data, and a processor for executing the software programs using the data. The processor performs simulations of the circuit operation of the electronic system in its design and simulations of the circuit operation when components have deteriorated due to a predetermined degradation simulation. Based on the results of the simulations, it selects output items whose output fluctuations due to the degradation satisfy predetermined conditions as candidate observation points, learns the relationship between the degradation and the fluctuations, and generates a component degradation model that estimates degradation from the fluctuations. This allows for the generation of a component degradation model that estimates the degradation of components in the electronic system, and also allows for the extraction of items to be monitored from the output items of the electronic system. This eliminates the need to comprehensively process data from a large number of monitoring points, thereby reducing the processing load for diagnosing the degradation of the electronic system. Furthermore, by sending the component degradation model to the electronic system, the electronic system can be configured to measure the values ​​of the output items at the monitoring points within its own device and determine the degree of degradation of the component to be detected. In this case, the computer only needs to collect the determination results from the electronic system, thus reducing the amount of data received from the electronic system.

[0073] (Item 2) In the electronic system diagnostic support system described in item 1, the memory stores a list of components to be detected for degradation in the electronic system, circuit data describing the circuit configuration of the electronic system, and output item information indicating output items for which output values ​​are measured in the electronic system. The processor acquires degradation factor / degradation parameter data representing the correspondence between degradation factors, which are factors that degrade the characteristics of the components to be detected for degradation shown in the list of components to be detected for degradation, and influence parameter information indicating the parameters that change due to the degradation of the components to be detected for degradation caused by the degradation factors and the manner in which the parameters change. Based on this, the circuit operation of the electronic system is simulated, the non-degraded output value which is the output value of the output item shown in the output item information is calculated, the circuit operation of the electronic system is simulated in which the components to be detected for degradation in the electronic system are degraded by simulating the degradation factors shown in the degradation factor / degradation parameter data, based on the circuit data, the list of components to be detected for degradation, the degradation factor / degradation parameter data is simulated, the degraded output value which is the output value of the output item is calculated, the relationship between the degradation factor and the fluctuation of the output value from the non-degraded output value to the degraded output value is learned, and a component degradation model that estimates the degradation factor from the output fluctuation is generated and output. According to this, it is possible to generate a component degradation model that estimates the degradation factor from the output fluctuation by learning the relationship between the degradation factor and the fluctuation of the output value of the output item.

[0074] (Item 3) In the electronic system diagnostic support system described in item 1, the memory stores a list of degradation detection target components, which are components mounted in the electronic system that are subject to degradation detection; circuit data describing the circuit configuration of the electronic system; and output item information indicating output items for which output values ​​are measured in the electronic system. The processor acquires degradation factor / degradation parameter data that represents the correspondence between degradation factors, which are factors that degrade the characteristics of the degradation detection target components shown in the list of degradation detection target components, and parameters that change due to the degradation caused by the degradation factors of the degradation detection target components, and influence parameter information that shows the manner in which the parameters change. Based on the circuit data and the output item information, the circuit operation of the electronic system is simulated, and the non-degraded output value, which is the output value of the output item, is calculated. Based on the circuit data, the list of components to be detected for degradation, the degradation factor / degradation parameter data, and the output item information, the circuit operation of the electronic system is simulated in which the components to be detected for degradation in the electronic system are degraded by simulating the degradation factors shown in the degradation factor / degradation parameter data, and the degraded output value, which is the output value of the output item, is calculated. The output items selected based on the non-degraded output value and the degraded output value are output as monitor candidates that are candidates for observation locations in the electronic system. This makes it possible to appropriately select output items to be candidates for observation locations in the electronic system based on the non-degraded output value and the degraded output value.

[0075] (Item 4) In the electronic system diagnostic support system described in item 3, the processor calculates an output fluctuation index value indicating the degree of fluctuation in the output value in response to changes in the parameter, and a degradation detection contribution rate which is the ratio of the fluctuation amount of each simulated parameter and its mode of change to the sum of the fluctuation amounts of each simulated parameter and its mode of change, based on the non-degraded output value and the degraded output value, and selects an output item to be monitored based on the output fluctuation index value and the degradation detection contribution rate. According to this, an appropriate monitor candidate can be selected based on the index value indicating the degree of fluctuation in the output value in response to changes in the parameter and the contribution rate of the fluctuation amounts of each simulated parameter and its mode of change.

[0076] (Item 5) In the electronic system diagnostic support system described in item 1, the memory pre-stores degradation countermeasure data that associates recommended design change candidates with combinations of degraded components and degradation factors that caused the component to degrade. The processor acquires degraded component information, including components that have degraded in the electronic system and degradation factors that caused the component to degrade. The processor then extracts recommended design change candidates from the degradation countermeasure data for combinations of components that have degraded in the electronic system and degradation factors that caused the component to degrade, as indicated in the degraded component information. This allows for obtaining design change candidates.

[0077] (Item 6) In the electronic system diagnostic support system described in item 1, the processor acquires measured output values, which are the output values ​​of the output items measured at the observation points in the electronic system, and estimates the degradation of the components based on the measured output values ​​using the component degradation model. This makes it possible to estimate the degraded components from the output values ​​measured at appropriate observation points.

[0078] (Item 7) In the electronic system diagnostic support system described in item 2 or 3, the output item information indicates a combination of an output item and an operation pattern that defines the timing for measuring the output value of the output item, and the processor calculates the output value of the output item at the timing defined in the operation pattern as the non-degraded output value and the degraded output value. This makes it possible to generate an appropriate component degradation model considering the operation pattern of the electronic system.

[0079] (Item 8) In the electronic system diagnostic support system described in item 4, the output fluctuation index value is calculated based on a value obtained by normalizing the difference between the non-degraded output value and the degraded output value by the non-degraded output value. According to this, since an index value obtained by normalizing the output fluctuation by the non-degraded output value is used, the degree of fluctuation in relation to the degradation of various output items of various outputs can be captured with a certain standard.

[0080] (Item 9) A processor is installed in the management computer that performs a simulation of the circuit operation of the electronic system in its design and a simulation of the circuit operation when components have deteriorated by simulating predetermined deterioration, and based on the results of the simulation, selects output items whose output fluctuations due to deterioration satisfy predetermined conditions as candidates for observation locations, learns the relationship between the deterioration and the fluctuations, and generates a component deterioration model that estimates deterioration from the fluctuations. A memory that stores the non-deteriorated output values ​​of the output items measured at the observation locations, and a processor that acquires the measured output values, which are the output values ​​of the output items measured at the observation locations, calculates a measured output fluctuation index value based on the non-deteriorated output values ​​and the measured output values, determines whether the measured output fluctuation index value exceeds a predetermined output fluctuation deterioration threshold, and warns that deterioration has occurred in the component corresponding to the output item in which the measured output fluctuation index value exceeds the output fluctuation deterioration threshold. [Explanation of Symbols]

[0081] 1...Electronic system diagnostic support system, 10...Computer, 11...Degradation parameter data, 12...Degradation model generation unit, 13...Internal monitor candidate output unit, 20...Electronic system, 21...Internal monitor unit, 22...Degradation estimation unit, 23...Component degradation model, 24...Degraded component determination unit, 25...Performance degradation alert unit, 31...List of components to be detected for degradation, 34...Degradation simulation parameter extraction unit, 35...Simulator unit, 36...Output calculation unit, 37...Parameter normalization unit, 39...Output normalization unit, 40...Machine learning unit, 42...Contribution rate calculation unit, 43...Monitor location candidate information, 49...Output normalization unit, 51...RC filter circuit, 62...Normalized output calculation unit, 63...Degradation probability inference unit 72...Deterioration location factor extraction unit, 91...Communication network, 92...User

Claims

1. An electronic system diagnostic support system that assists in the diagnosis of electronic systems, A memory for storing software programs and data, The system includes a processor that executes the software program using the aforementioned data, The aforementioned processor, The following simulations are performed: a simulation of the circuit operation of the electronic system in its design, and a simulation of the circuit operation under conditions where components have deteriorated due to a predetermined degradation simulation. Based on the results of the simulation, output items whose output fluctuations due to degradation satisfy predetermined conditions are selected as candidate observation locations. The relationship between the aforementioned degradation and the aforementioned fluctuations is learned, and a component degradation model is generated that estimates degradation from the fluctuations. Electronic system diagnostic support system.

2. In the electronic system diagnostic support system according to claim 1, The memory stores a list of components to be detected for degradation, which are components in the electronic system whose degradation is to be detected; circuit data describing the circuit configuration of the electronic system; and output item information indicating output items for which output values ​​are measured in the electronic system. The aforementioned processor, Degradation factor / degradation parameter data is obtained that represents the correspondence between degradation factors, which are factors that degrade the characteristics of the components to be detected for degradation as shown in the list of components to be detected for degradation, and the parameters that change due to the degradation of the components to be detected for degradation caused by those degradation factors, and the influence parameter information that shows the manner in which those parameters change. Based on the circuit data, the circuit operation of the electronic system is simulated, and the non-degraded output value, which is the output value of the output item shown in the output item information, is calculated. Based on the circuit data, the list of components to be detected for degradation, the degradation factor / degradation parameter data, and the output item information, the circuit operation of the electronic system is simulated in which the components to be detected for degradation in the electronic system are degraded by simulating the degradation factors shown in the degradation factor / degradation parameter data, and the output value at the time of degradation, which is the output value of the output item, is calculated. The system learns the relationship between the degradation factors and the fluctuation in output values ​​from the non-degraded output value to the degraded output value, and generates and outputs a component degradation model that estimates the degradation factors from the output fluctuations. Electronic system diagnostic support system.

3. In the electronic system diagnostic support system according to claim 1, The memory stores a list of components to be detected for degradation, which are components mounted in the electronic system that are subject to degradation detection; circuit data describing the circuit configuration of the electronic system; and output item information indicating output items for which output values ​​are measured in the electronic system. The aforementioned processor, Degradation factor / degradation parameter data is obtained that represents the correspondence between degradation factors, which are factors that degrade the characteristics of the components to be detected for degradation as shown in the list of components to be detected for degradation, and the parameters that change due to the degradation of the components to be detected for degradation caused by those degradation factors, and the influence parameter information that shows the manner in which those parameters change. Based on the circuit data and the output item information, the circuit operation of the electronic system is simulated, and the non-degraded output value, which is the output value of the output item, is calculated. Based on the circuit data, the list of components to be detected for degradation, the degradation factor / degradation parameter data, and the output item information, the circuit operation of the electronic system is simulated in which the components to be detected for degradation in the electronic system are degraded by simulating the degradation factors shown in the degradation factor / degradation parameter data, and the output value at the time of degradation, which is the output value of the output item, is calculated. The output items selected based on the non-degraded output value and the degraded output value are output as monitor candidates that will serve as candidate observation locations in the electronic system. Electronic system diagnostic support system.

4. In the electronic system diagnostic support system described in claim 3, The aforementioned processor, Based on the non-degraded output value and the degraded output value, an output fluctuation index value indicating the degree of fluctuation of the output value in response to the change in the parameter and a degradation detection contribution rate, which is the ratio of the fluctuation amount of each simulated parameter and its change pattern to the sum of the fluctuation amounts of each simulated parameter and its change pattern, are calculated, and based on the output fluctuation index value and the degradation detection contribution rate, output items to be monitored are selected. Electronic system diagnostic support system.

5. In the electronic system diagnostic support system according to claim 1, The memory pre-stores degradation countermeasure data that associates recommended design change candidates with combinations of degraded components and the factors that caused the degradation of those components. The aforementioned processor, The aforementioned electronic system acquires deteriorated component information, including the component that has deteriorated and the deterioration factor that caused the deterioration of the component. From the degradation countermeasure data, candidates for recommended design changes are extracted for each combination of a component that has deteriorated in the electronic system and the degradation factor that caused the deterioration of that component, as shown in the deteriorated component information. Electronic system diagnostic support system.

6. In the electronic system diagnostic support system according to claim 1, The aforementioned processor, The aforementioned electronic system acquires the measured output value, which is the output value of the output item measured at the observation location. Using the aforementioned component degradation model, the degradation of the component is estimated based on the measured output values. Electronic system diagnostic support system.

7. In the electronic system diagnostic support system according to claim 2 or 3, The output item information indicates a combination of an output item and an operation pattern that defines the timing for measuring the output value of the output item. The processor calculates the output value of the output item at the timing defined in the operation pattern as the non-degraded output value and the degraded output value. Electronic system diagnostic support system.

8. In the electronic system diagnostic support system according to claim 4, The output fluctuation index value is calculated based on the difference between the non-degraded output value and the degraded output value, normalized by the non-degraded output value. Electronic system diagnostic support system.

9. In the electronic system diagnostic support system described in claim 6, A processor, mounted on a management computer, performs simulations of the circuit operation of the electronic system in its design state and simulations of the circuit operation when components have deteriorated by simulating predetermined deterioration. Based on the results of the simulations, it selects output items whose output fluctuations due to the deterioration satisfy predetermined conditions as candidate observation points, learns the relationship between the deterioration and the fluctuations, and generates a component deterioration model that estimates deterioration from the fluctuations. A memory for storing the non-degraded output value of the output item measured at the observation point, and a processor which, in the electronic system, acquires the measured output value which is the output value of the output item measured at the observation point, calculates a measured output fluctuation index value based on the non-degraded output value and the measured output value, determines whether the measured output fluctuation index value exceeds a predetermined output fluctuation degradation threshold, and warns that degradation has occurred in the component corresponding to the output item in which the measured output fluctuation index value exceeds the output fluctuation degradation threshold, are mounted in the electronic system. Electronic system diagnostic support system.

10. An electronic system diagnostic support method for supporting the diagnosis of electronic systems, A computer having memory for storing software programs and data, and a processor for executing the software programs using the data, The following simulations are performed: a simulation of the circuit operation of the electronic system in its design, and a simulation of the circuit operation under conditions where components have deteriorated due to a predetermined degradation simulation. Based on the results of the simulation, output items whose output fluctuations due to degradation satisfy predetermined conditions are selected as candidate observation locations. The relationship between the aforementioned degradation and the aforementioned fluctuations is learned, and a component degradation model is generated that estimates degradation from the fluctuations. Electronic system diagnostic support method.

11. An electronic system diagnostic support program for assisting in the diagnosis of electronic systems, A computer having a memory for storing software programs and data, and a processor for executing the software programs using the data, The following simulations are performed: a simulation of the circuit operation of the electronic system in its design, and a simulation of the circuit operation under conditions where components have deteriorated due to a predetermined degradation simulation. Based on the results of the simulation, output items whose output fluctuations due to degradation satisfy predetermined conditions are selected as candidate observation locations. The relationship between the aforementioned degradation and the aforementioned fluctuations is learned, and a component degradation model is generated that estimates degradation from the fluctuations. An electronic system diagnostic support program that enables the execution of this task.

Images