Circuit life prediction device, circuit life prediction system and circuit life prediction method

Abstract

To provide a technology for conveniently predicting life of a power conversion circuit.SOLUTION: A circuit life prediction device comprises one or more processors and one or more memory resources. Each processor comprises: a monitor data acquisition unit which acquires monitor data of power conversion efficiency of a power conversion circuit of an electronic system comprising the power conversion circuit for life prediction; an approximate expression calculation unit which predicts future variation of the power conversion efficiency of the power conversion circuit by using the monitor data; and a remaining life calculation unit which calculates time until the power conversion efficiency reaches a predetermined life threshold as a remaining life.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a circuit life prediction device, a circuit life prediction system, and a circuit life prediction method. [Background technology]

[0002] There is a growing trend to reduce the environmental impact of industrial systems during their life cycles. Specifically, there is a growing demand for technology to estimate the lifespan of power conversion circuits commonly used in electronic systems, such as AC / DC converters, DC / AC converters, and DC / DC converters. If the lifespan of power conversion circuits can be estimated, it can be used as a quantitative basis for making decisions when reusing electronic systems.

[0003] Furthermore, it will be possible to abolish the currently widely practiced time-based maintenance and switch to condition-based maintenance, which will enable safe reuse without waste.Technologies related to predicting the lifespan of power conversion circuits include predicting the lifespan of electrolytic capacitors, which are products with a finite lifespan, and detecting deterioration from fluctuations in the control parameters of power conversion circuits.

[0004] Patent Document 1 describes a technique for detecting degradation by monitoring the duty ratio of the switching time, which is a control parameter of the target switching power conversion circuit. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2019-62689 [Non-patent literature]

[0006] [Non-Patent Document 1] Abe Masaya and Zaitsu Toshiyuki, "Fundamentals of Switching Power Supply Control Design," Nikkei BP Publishing, March 15, 2015, pp. 19-22 Summary of the Invention [Problem to be solved by the invention]

[0007] However, the technology described in Patent Document 1 determines whether deterioration leading to a warning or shutdown occurs based on whether a change in the duty ratio of the switching time exceeds a threshold, and is therefore unable to predict the lifespan. Furthermore, since the power conversion circuit in question is a switching power supply controlled by PWM (Pulse Width Modulation), the technology only refers to the duty ratio, and is therefore unable to directly apply to switching power conversion circuits controlled by PFM (Pulse Frequency Modulation) or the like, in which parameters other than the duty ratio are used as control parameters. The object of the present invention is to provide a technology for easily predicting the lifespan of a power conversion circuit. [Means for solving the problem]

[0008] The present application includes multiple means for solving at least part of the above-mentioned problems, examples of which are as follows: A circuit life prediction device according to one aspect of the present invention for solving the above-mentioned problems is a circuit life prediction device having one or more processors and one or more memory resources, wherein the processor comprises: a monitor data acquisition unit that acquires monitor data of the power conversion efficiency of a power conversion circuit of an electronic system including the power conversion circuit to be the target of life prediction; an approximation equation calculation unit that predicts future changes in the power conversion efficiency of the power conversion circuit using the monitor data; and a remaining life calculation unit that calculates the time until the power conversion efficiency reaches a predetermined life threshold as the remaining life. [Effects of the Invention]

[0009] According to the present invention, it is possible to provide a technology for easily predicting the life of a power conversion circuit. Problems, configurations, and effects other than those described above will become apparent from the following description of the preferred embodiment of the present invention. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a circuit life prediction system. [Figure 2] FIG. 2 is a diagram illustrating an example of a hardware configuration of an electronic system to be diagnosed. [Figure 3] FIG. 1 illustrates an example of a hardware configuration of a circuit life prediction device. [Figure 4] FIG. 2 is a diagram illustrating an example of a data structure of information included in a memory resource. [Figure 5] FIG. 10 is a diagram illustrating an example of a processing flow of remaining life prediction processing. [Figure 6] FIG. 1 is a diagram illustrating an example of a power conversion circuit. [Figure 7] FIG. 10 is a diagram illustrating an example of power conversion efficiency relative to output current. [Figure 8] FIG. 10 is a diagram showing an example of aging of a resistor. [Figure 9] FIG. 10 is a diagram illustrating an example of acquisition of monitor data and trend data. [Figure 10] FIG. 10 is a diagram illustrating an example of remaining life prediction using an approximation formula. [Figure 11] FIG. 10 is a diagram illustrating a configuration example of a modified example of a circuit life prediction system. [Figure 12] FIG. 10 is a diagram illustrating an example of a processing flow of remaining life prediction processing (maximum efficiency standard). [Figure 13] FIG. 10 is a diagram illustrating an example of acquiring temporary monitor data and maximum monitor data. [Figure 14] FIG. 10 is a diagram illustrating an example of the configuration of another modified example of the circuit life prediction system. [Figure 15] FIG. 10 is a diagram illustrating an example of a processing flow of remaining life prediction processing (under load). [Figure 16] FIG. 10 is a diagram showing an example of deterioration over time of power conversion efficiency at a high load current. [Figure 17] FIG. 10 is a diagram illustrating an example of a remaining life prediction result screen. DETAILED DESCRIPTION OF THE INVENTION

[0011] First Embodiment Hereinafter, embodiments of the present invention will be described with reference to the drawings. The examples are illustrative of the present invention, and for clarity of explanation, appropriate omissions and simplifications have been made. The present invention can be implemented in various other forms. Unless otherwise specified, each component may be singular or plural.

[0012] In order to facilitate understanding of the invention, the position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings.

[0013] Examples of various types of information may be described using expressions such as "table," "list," and "queue," but the various types of information may be expressed using data structures other than these. For example, various types of information such as "XX table," "XX list," and "XX queue" may be expressed as "XX information." When describing identification information, expressions such as "identification information," "identifier," "name," "ID," and "number" are used, but these are interchangeable. Furthermore, in the embodiments, identification information described using these expressions is expressed using symbols, numbers, natural language, or a combination thereof, but the identification information may be in a format other than these.

[0014] When there are multiple components with the same or similar functions, they may be described using the same reference numeral with different subscripts. When there is no need to distinguish between these multiple components, the subscripts may be omitted.

[0015] In the embodiments, processing performed by executing a program may be described. Here, a computer executes the program using a processor (e.g., a CPU or a GPU) and performs processing defined by the program using storage resources (e.g., a memory) and interface devices (e.g., a communication port). Therefore, the entity performing the processing by executing the program may be the processor. Similarly, the entity performing the processing by executing the program may be a controller, device, system, computer, or node having a processor. The entity performing the processing by executing the program may be any computing unit, and may include a dedicated circuit that performs specific processing. Here, the dedicated circuit may be, for example, an FPGA (Field Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), or a CPLD (Complex Programmable Logic Device).

[0016] A program may be installed on a computer from a program source. The program source may be, for example, a program distribution server or a computer-readable storage medium. When the program source is a program distribution server, the program distribution server may include a processor and a storage resource for storing the program to be distributed, and the processor of the program distribution server may distribute the program to be distributed to other computers. In addition, in an embodiment, two or more programs may be realized as one program, or one program may be realized as two or more programs.

[0017] Furthermore, although the present invention is typically realized by an information processing device, it may also be realized as a platform having the functions of the present invention.

[0018] The circuit life prediction system 1 is used during operation and maintenance of an electronic system, or in preparation for reuse and redistribution of the electronic system. Such an electronic system is typically composed of electronic devices, and therefore electronic devices are mainly exemplified in this embodiment. However, the system is not limited to electronic devices, and may be applied in various environments (such as mobile objects powered by reciprocating engines, etc.).

[0019] 1 is a diagram showing an example of the configuration of a circuit life prediction system. The circuit life prediction system 1 includes a circuit life prediction device 100. A plurality (N) of electronic systems 200 to be diagnosed by the circuit life prediction system 1 exist via a network 50.

[0020] For example, the circuit life prediction device 100 is an information processing device including a memory resource 110, a processor 120, a UI (User Interface) device 130, and an NI (Network Interface) device 140.

[0021] The memory resource 110 includes a life prediction program 111 , monitor data 112 , trend data 113 , life threshold data 114 , and remaining life data 115 .

[0022] The electronic system 200 to be diagnosed includes at least a power conversion circuit 201, a power source 202, a load 203, and a monitor module 204. The electronic system 200 to be diagnosed is a variety of devices or edge devices, such as IT equipment, medical equipment, EVs (Electric Vehicles), various industrial equipment such as elevators, and renewable energy power generation conversion systems.

[0023] 2 is a diagram showing an example of the hardware configuration of an electronic system to be diagnosed. The power conversion circuit 201 is, for example, an AC / DC conversion circuit that uses an AC power supply as input and outputs DC. However, the power conversion circuit 201 is not limited to this, and may be, for example, a DC / DC conversion circuit (a step-up circuit or a step-down circuit) that uses a DC power supply as input and outputs DC of a different voltage, or a DC / AC conversion circuit that uses a DC power supply as input and outputs AC.

[0024] The power source 202 is a power source in the electronic system under diagnosis 200. For example, the power source 202 is an AC circuit that supplies power to the system. The load 203 is a component, such as a computing unit of the electronic system under diagnosis 200, that performs computations to realize the functions of the electronic system under diagnosis 200 using power supplied from the power conversion circuit 201.

[0025] The monitor module 204 measures the power conversion efficiency of the power conversion circuit 201. The monitor module 204 includes an input current sensor 204a, an input voltage sensor 204b, an output current sensor 204c, an output voltage sensor 204d, a plurality of AD converters 204e, a nonvolatile memory 204f, a CPU 204g, and a data transmission circuit 204h.

[0026] The AD converter 204e converts analog information of the output voltages from the input current sensor 204a, input voltage sensor 204b, output current sensor 204c, and output voltage sensor 204d into digital information and passes the digital information to the CPU 204g. The CPU 204g calculates the power conversion efficiency η using the information received from the AD converter 204e and outputs the calculated power conversion efficiency η to the nonvolatile memory 204f and the data transmission circuit 204h in association with time information. The nonvolatile memory 204f stores data including the power conversion efficiency η output from the CPU 204g, and the data transmission circuit 204h transmits the data including the power conversion efficiency η output from the CPU 204g to the circuit life prediction device 100 via the network 50. At this time, the CPU 204g calculates the power conversion efficiency η using the following equation (1):

[0027]

number

[0028] Here, V O , I O , V I , I I are the output voltage, the output current, the input voltage, and the input current, respectively. Note that the CPU 204g may not calculate the power conversion efficiency η, but may have the processor 120 of the circuit life prediction device 100 calculate it. The CPU 204g may also transmit the power conversion efficiency η or information received from the AD converter 204e to the circuit life prediction device 100 as needed, or may thin out the data and transmit it at a predetermined timing (for example, calculate and transmit a representative value every hour or every day). Alternatively, the CPU 204g may obtain data from the nonvolatile memory 204f at a predetermined timing and transmit it.

[0029] In this way, since the monitor module 204 includes the non-volatile memory 204f, even when the electronic system 200 to be diagnosed is reused and applied to another system, the monitor data 112 can be continuously stored in the other system.

[0030] The life prediction program 111 is a software program that causes the processor 120 to execute a remaining life prediction process, which will be described later. The monitor data 112 is information obtained from the monitor module 204 of the electronic system 200 to be diagnosed via the NI device 140, and stores the voltage value, current value, and power conversion efficiency calculated from these values ​​applied to the power conversion circuit 201 included in the electronic system 200 to be diagnosed.

[0031] The trend data 113 is data obtained by extracting features of time-series data of the monitored situation obtained by calculation from the monitor data 112. Specifically, the trend data 113 is data obtained by accumulating the monitor data 112 for a predetermined period and extracting features that indicate long-term trends. For example, the trend data 113 is the arithmetic average, moving average, or exponentially smoothed moving average, maximum value, or minimum value of the monitor data 112 that indicates a long-term trend.

[0032] The life threshold data 114 is a threshold for determining the end of life for the future operating state predicted from the trend data 113. For example, the life threshold data 114 is a threshold for power conversion efficiency.

[0033] The remaining life data 115 is the period until the end of the life predicted using the trend data 113 and the life threshold data 114. Specifically, the remaining life data 115 is the period until the future value of the power conversion efficiency predicted using the trend data 113 falls below the life threshold data 114.

[0034] The processor 120 performs a remaining life prediction process, which will be described later, using the life prediction program 111. Specifically, the processor 120 includes a monitor data acquisition unit 121 that performs a process of acquiring monitor data 112 of the power conversion efficiency of the power conversion circuit 201 output from the monitor module 204. The processor 120 also includes a remaining life calculation unit 122 that performs a process of predicting future changes in the power conversion efficiency of the power conversion circuit 201 using the monitor data 112 and calculating the time until the efficiency reaches life threshold data 114 as the remaining life.

[0035] The processor 120 also includes a trend calculation unit 123 that accumulates the monitor data 112 for a predetermined period and calculates trend data that indicates a long-term trend. The processor 120 also includes an approximate equation calculation unit 124 that calculates a regression equation that approximates the trend data 113 to a predetermined degree or more. In the processing of the remaining life calculation unit 122, the processor 120 predicts future changes in the power conversion efficiency of the power conversion circuit 201 using the regression equation.

[0036] Furthermore, the trend calculation unit 123 calculates an average value of the monitor data 112 over a predetermined period as trend data 113. Furthermore, in the process of calculating the trend data 113, the trend calculation unit 123 calculates maximum, minimum, or average data from among a predetermined number of accumulated sample values ​​of the monitor data 112 as trend data 113. Furthermore, in the process of calculating the trend data 113, the trend calculation unit 123 calculates an arithmetic average, moving average, or exponentially smoothed moving average data from among a predetermined number of accumulated sample values ​​of the monitor data 112 as trend data 113.

[0037] In addition, the monitor data acquisition unit 121 may acquire the power conversion efficiency output from the monitor module 204 as monitor data 112 when the voltage specified by the output voltage or the current specified by the output current is equal to or greater than a predetermined value.

[0038] The UI device 130 accepts input of various instructions from a user (operator). The input contents include at least information about the electronic system 200 to be diagnosed. For example, the UI device 130 accepts designation of a device in the electronic system 200 to be diagnosed via an input device such as a mouse or keyboard. Note that the UI device 130 accepts input of information about diagnosis, not limited to the device in the electronic system 200 to be diagnosed. For example, the UI device 130 accepts input of life threshold data 114, etc.

[0039] The UI device 130 displays at least the remaining life of the electronic system 200 to be diagnosed on a screen via a display device. For this reason, the UI device 130 may be configured integrally with an input device such as a touch panel. Furthermore, the UI device 130 may be configured in a separate housing from the circuit life prediction device 100. In this case, the UI device 130 may be realized by a separate terminal device, and input information and output information may be transferred via a network.

[0040] The NI device 140 is connected to an external device so as to be able to communicate with it via a communication path such as a network 50 that is any one of, or a combination of, a public network such as the Internet, a communication network that uses in part or in whole a LAN (Local Area Network), a WAN (Wide Area Network), a VPN (Virtual Private Network), etc. Note that the network 50 may be a wireless communication network such as Wi-Fi (registered trademark) or 5G (Generation).

[0041] [Hardware Configuration] Figure 3 is a diagram showing an example of the hardware configuration of a circuit life prediction device. The circuit life prediction device 100 can be realized as a general information processing device 900 including a processor 901, hardware memory 902 such as RAM (Random Access Memory), storage 903 such as a hard disk drive (HDD) or SSD (Solid State Drive), a reading device 905 for reading information from a portable storage medium 904 such as a CD (Compact Disk) or DVD (Digital Versatile Disk), an input device 906 such as a keyboard, mouse, barcode reader, or touch panel, an output device 907 such as a display, and a communication device 908 for communicating with other computers via a communication network such as a LAN or the Internet, or as a network system including a plurality of such information processing devices 900. Note that the reading device 905 may be capable of not only reading but also writing to the portable storage medium 904.

[0042] The processor 901 is, for example, a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). The processor 901 performs various processes by executing various predetermined programs loaded from the storage 903 into the memory 902. The programs are, for example, application programs that can be executed on an OS (Operating System) program. The programs may be installed in the storage 903 from a portable storage medium 904 via a reading device 905, or may be downloaded from a network via a communication device 908 and executed by the processor 901.

[0043] For example, the monitor data acquisition unit 121, remaining life calculation unit 122, trend calculation unit 123, and approximate expression calculation unit 124 can be realized by loading a program stored in storage 903 into memory 902 and executing it on processor 901.

[0044] The UI device 130 can be realized by the processor 901 using the input device 906, the output device 907, and the communication device 908. The memory resource 110 can be realized by the processor 901 using the memory 902 or the storage 903. The NI device 140 can be realized by the processor 901 using the communication device 908.

[0045] [Data Description] Figure 4 shows an example of the data structure of information contained in the memory resource 110. For each target circuit 110a, the memory resource 110 includes a fitting approximation formula 110b, and monitor data 112 and trend data 113 associated with a date and time 110c.

[0046] The target circuit 110a is information of an identifier or name that identifies the electronic system 200 to be diagnosed. The fitting approximation formula 110b is information that identifies a function (a polynomial or an exponential function) that most closely approximates the trend data 113. The date and time 110c is information that identifies the date and time. The monitor data 112 is monitor data such as power conversion efficiency acquired from the monitor module 204 at the date and time identified by the date and time 110c. The trend data 113 is trend data obtained from the monitor data such as power conversion efficiency acquired from the monitor module 204 at the date and time identified by the date and time 110c.

[0047] [Explanation of Operation] Fig. 5 is a diagram showing an example of the processing flow of the remaining life prediction process. The remaining life prediction process is started, for example, when the UI device 130 receives an instruction to execute the remaining life prediction process from the user.

[0048] First, the monitor data acquisition unit 121 samples the power conversion efficiency of the electronic system 200 to be diagnosed (step S01). Specifically, the monitor data acquisition unit 121 acquires information on the power conversion efficiency, which is monitor data, from the monitor module 204 to be diagnosed, and stores the information in the memory resource 110 as the monitor data 112 in association with the date and time 110c.

[0049] Then, the trend calculation unit 123 calculates a moving average (step S02). Specifically, the trend calculation unit 123 calculates a moving average for a predetermined period of the monitor data 112 to obtain the trend data 113. However, the trend calculation unit 123 is not limited to calculating a moving average for the predetermined period of the monitor data 112, and may calculate an arithmetic average, an exponentially smoothed moving average, a maximum value, or a minimum value to obtain the trend data 113.

[0050] Then, the trend calculation unit 123 determines whether the number of accumulated trend data 113 exceeds a predetermined number of data N (N is an integer) (step S03). This is because a certain amount of data accumulation is necessary to predict long-term trends. If the number of data is N or less ("No" in step S03), the trend calculation unit 123 returns control to step S01 to continue repeatedly acquiring data. However, even if the number of data exceeds N, the monitor data acquisition unit 121 continues to accumulate monitor data, and the trend calculation unit 123 continues to calculate trend data.

[0051] Then, if the amount of trend data 113 accumulated is greater than N ("Yes" in step S03), the approximate expression calculation unit 124 performs function fitting using polynomial and exponential regression on the trend data 113 (step S04). Specifically, the approximate expression calculation unit 124 determines an approximate expression for the time series data, and the variables define an objective function for time t. P In order to select an approximate equation that can accurately represent the data up to now, the approximate equation calculation unit 124 may simultaneously fit a plurality of equations (for example, quadratic equations, exponential functions, etc.). The approximate equation calculation unit 124 selects the approximate equation to be applied by using the coefficient of determination R 2 It can be the expression with the largest value.

[0052] Then, the remaining life calculation unit 122 calculates the threshold value and the approximate formula F P The date and time T when the difference with (t) becomes zero L Specifically, the remaining life calculation unit 122 calculates the approximate formula F obtained in step S04. P Since the future trend (predicted value) can be obtained by (t), the threshold value-F P The date and time T when (t) = 0 L Calculate.

[0053] Then, the remaining life calculation unit 122 calculates the remaining life (step S06). Specifically, the remaining life calculation unit 122 calculates the remaining life from the date and time T LIt is calculated by subtracting the current date and time from the

[0054] Then, the remaining life calculation unit 122 records the remaining life (step S07). Specifically, the remaining life calculation unit 122 stores the remaining life calculated in step S06 as remaining life data 115.

[0055] The above is an example of the processing flow of the remaining life prediction process. According to the example of the processing flow of the remaining life prediction process, the remaining life of the power conversion circuit 201 can be predicted by using the monitor data 112 of the power conversion efficiency.

[0056] Note that, in step S04 of the processing flow of the remaining life prediction process, function fitting is performed, but the present invention is not limited to this, and for example, a model may be constructed by machine learning or the like, with the remaining life as the objective variable and the actual values ​​of trend data obtained from the power conversion circuits 201 of multiple electronic systems 200 to be diagnosed as explanatory variables, and the remaining life calculation unit 122 may apply this model to infer the remaining life. In this way, the remaining life can be predicted more simply and accurately.

[0057] 6 is a diagram illustrating an example of a power conversion circuit. The power conversion circuit 201 can be, for example, a step-down switching DC-DC converter 201a. In the step-down switching DC-DC converter 201a, V I =12V, V O =5V, L=10µF, R L =0.1, C=300 microF, R C =0.2 ohm, S: Nch-MOSFET, D: Diode, R SON = 0.05 ohms (Initial), R DON = 0.1 ohms (Initial), Switching frequency: 100 kHz.

[0058] 7 is a diagram showing an example of power conversion efficiency versus output current. For example, in a step-down switching DC-DC converter 201a, the power conversion efficiency varies depending on the magnitude of the load 203. An example of the power conversion efficiency versus such output current (load) is shown in the diagram as graph 201b. Here, to analytically calculate the power conversion efficiency, the equation already derived in Non-Patent Document 1 is applied.

[0059] First, power conversion efficiency can be calculated simply from the output power, the loss due to Joule heat generated in each element, and the switching loss using the following formula (2):

[0060]

number

[0061] Each, P SR is the ON resistance R of the active switch element S such as a MOSFET SON Loss in P DR is the ON resistance R of the diode D DON Loss in P L is the parasitic resistance R of the inductor L L Loss in P C is the parasitic resistance R of the capacitor C C Loss in P SW is the switching loss. Each loss is calculated using the following equation (3).

[0062]

number

[0063] where D is the duty ratio of the control signal for the switching element, and f SW is the switching frequency, and the current flowing through the switch I Srms and the current flowing through the diode I Drms are the effective currents, respectively.

[0064] Graph 201b shown in Fig. 7 is an example of power conversion efficiency calculated in this manner. As can be seen from graph 201b, power conversion efficiency depends on the current required by load 203, and therefore it is difficult to measure the degree of degradation by simply monitoring power conversion efficiency without considering load 203. However, it is generally difficult to predict the load to be monitored. For example, in the case of a power conversion circuit 201 for renewable energy, the load changes depending on the amount of sunlight, and in the case of a power conversion circuit 201 for a server, the load changes depending on the user's request (resource demand). For this reason, in this embodiment, a random load such as a Gaussian distribution or a Poisson distribution is assumed to occur.

[0065] FIG. 8 is a diagram showing an example of resistance degradation over time. Example 300 of aging of the ON resistance of a MOSFET and a diode is an example that simulates the degradation of the ON resistance of a switching element (MOSFET) and a diode. Example 300 of aging of the ON resistance of a MOSFET and a diode shows that the ON resistance increases exponentially with operating time. However, example 300 of aging of the ON resistance of a MOSFET and a diode is merely a simulation of degradation, and the rate of degradation and characteristics of actual components will vary depending on the model number and individual components.

[0066] FIG. 9 is a diagram showing an example of monitor data and trend data acquisition. In the example 400 of monitor data and trend data acquisition, the load is randomly changed every second according to a Gaussian distribution, and monitor data 112 (power conversion efficiency) is shown when element degradation over time is taken into account. A moving average line is also displayed as a white line. This moving average line is the moving average of one day's worth of data (60s x 60m x 24h), and can be considered trend data 113 for the monitor data 112.

[0067] FIG. 10 is a diagram showing an example of remaining life prediction using an approximation formula. In the example 500 of remaining life prediction using an approximation formula, the results of polynomial (quadratic) and exponential function fitting are shown using the data up to the present time of the trend data 113, which is a moving average line. It can be seen that the exponential function fitting approximation formula 110b matches well with the moving average line (R 2 (It can be seen from the values ​​that there is a high degree of agreement.) Therefore, when the exponential approximation formula is used, the time until the intersection with the value set as the life threshold data 114 (75% in the figure) is the remaining life.

[0068] The above is an example of the circuit life prediction system 1 according to the first embodiment of the present invention. The circuit life prediction system 1 according to the first embodiment makes it possible to simply predict the life of a power conversion circuit.

[0069] The present invention is not limited to the above-described embodiment, but includes various modifications. For example, the above-described embodiment has been described in detail to clearly explain the present invention, and is not necessarily limited to an embodiment including all of the described configurations. It is possible to replace part of the configuration of an embodiment with another configuration, and it is also possible to add the configuration of another embodiment to the configuration of an embodiment. It is also possible to delete part of the configuration of an embodiment.

[0070] For example, in the above embodiment, the circuit life prediction device 100 acquires monitor data 112 and calculates trend data 113 using a moving average, but this is not limited to this. A window of a predetermined period may be set in the monitor data 112, and the maximum value of the data within the window may be recorded to create trend data 113. In this way, it becomes possible to track the maximum value of power conversion efficiency as a trend, and to acquire a trend that is independent of the load fluctuation distribution. Such a modification will be described below as a second embodiment.

[0071] Second Embodiment The circuit life prediction system 1' according to the second embodiment is basically the same as the circuit life prediction system 1 according to the first embodiment. Therefore, the following description will focus on the differences.

[0072] 11 is a diagram showing a configuration example of a modified example of a circuit life prediction system. In modified example 1′ of the circuit life prediction system, a circuit life prediction device 100′ stores temporary monitor data 116 and maximum monitor data 117 in a memory resource 110′.

[0073] 12 is a diagram showing an example of the processing flow of the remaining life prediction process (maximum efficiency standard). The remaining life prediction process (maximum efficiency standard) is basically the same as the remaining life prediction process of the first embodiment, but there are some differences in the processing. The different processing will be described below.

[0074] First, the monitor data acquisition unit 121 samples the power conversion efficiency of the electronic system 200 to be diagnosed (step S11). Specifically, the monitor data acquisition unit 121 acquires information on the power conversion efficiency, which is monitor data, from the monitor module 204 to be diagnosed, and stores the information in the memory resource 110′ as temporary monitor data 116 in association with the date and time 110c.

[0075] Then, the monitor data acquiring unit 121 determines whether the number of data in the temporary monitor data 116 has accumulated up to a predetermined number of windows (step S12). The predetermined number of windows to accumulate can be any number, for example, a number of windows equivalent to one day's worth. If not accumulated ("No" in step S12), the monitor data acquiring unit 121 returns control to step S11. However, even if the number of data in the temporary monitor data 116 exceeds the predetermined number of windows ("Yes" in step S12), the monitor data acquiring unit 121 continues accumulating the temporary monitor data 116.

[0076] Then, the trend calculation unit 123 extracts the value of the power conversion efficiency that is the maximum value within the window (step S13). Specifically, the trend calculation unit 123 identifies the maximum value within the window for the most recent data of a predetermined number of windows from the temporary monitor data 116, and sets this as maximum monitor data 117. Note that when the maximum value is identified, the monitor data acquisition unit 121 deletes the temporary monitor data 116, which can contribute to reducing the capacity of the memory resource 110, etc.

[0077] Then, the approximate expression calculation unit 124 performs function fitting using polynomial and exponential regression on the maximum monitor data 117 (step S14). Specifically, the approximate expression calculation unit 124 determines an approximate expression for the time series data, and the variables define an objective function for time t. P In order to select an approximate equation that can accurately represent the data up to now, the approximate equation calculation unit 124 may simultaneously fit a plurality of equations (for example, quadratic equations, exponential functions, etc.). The approximate equation calculation unit 124 selects the approximate equation to be applied by using the coefficient of determination R 2 It can be the expression with the largest value.

[0078] The above is an example of the processing flow of the remaining life prediction process (maximum efficiency standard). According to the example of the processing flow of the remaining life prediction process (maximum efficiency standard), it becomes possible to obtain a trend that is not dependent on the load fluctuation distribution, thereby improving the accuracy of the remaining life prediction.

[0079] FIG. 13 is a diagram showing an example of acquiring temporary monitor data and maximum monitor data. In an example 400′ of acquiring temporary monitor data and maximum monitor data, the maximum value of the monitor data is extracted for each predetermined number of windows, and a trend based on the maximum monitor data 117 can be displayed. The trend based on the maximum monitor data 117 differs from the trend data 113 in the first embodiment, which is composed of values ​​obtained by moving averages, but it is possible to obtain a trend with maximum efficiency, thereby enabling monitoring that is not dependent on load behavior. In addition, because the temporary monitor data 116 can be deleted as needed, the total data capacity of the memory resource 110′ can be reduced. The above is an example of a circuit life prediction system 1′ according to the second embodiment of the present invention.

[0080] Furthermore, in the first embodiment described above, the circuit life prediction device 100 acquires monitor data 112 and calculates trend data 113 using a moving average, but this is not limited to this. For example, the monitor data 112 may be limited to data on power conversion efficiency near the maximum load. Since it is known that deterioration over time in power conversion efficiency becomes more pronounced under high load conditions, this makes it possible to detect deterioration at an earlier stage. Such a modification will be described below as a third embodiment.

[0081] Third Embodiment The circuit life prediction system 1'' according to the third embodiment is basically the same as the circuit life prediction system 1 according to the first embodiment. Therefore, the following description will focus on the differences.

[0082] 14 is a diagram showing an example of the configuration of another modified circuit life prediction system. In modified circuit life prediction system 1'', a circuit life prediction device 100'' stores load threshold data 118 in a memory resource 110''. It is desirable that the load threshold data 118 stores in advance the maximum value of the current (or voltage) determined by the load 203, or a value approximate to that (for example, a value that is 80% of the maximum value).

[0083] 15 is a diagram showing an example of the processing flow of the remaining life prediction process (under load). The remaining life prediction process (under load) is basically the same as the remaining life prediction process of the first embodiment, but there are some differences in the processing. The different processing will be explained below.

[0084] First, the monitor data acquisition unit 121 detects the load current (or load voltage) of the electronic system 200 to be diagnosed (step S21). Specifically, the monitor data acquisition unit 121 acquires the value of the output current (or output voltage) as monitor data from the monitor module 204 to be diagnosed.

[0085] Then, the monitor data acquiring unit 121 determines whether the load is equal to or greater than a predetermined load threshold (step S22). Specifically, the monitor data acquiring unit 121 determines whether the output current (or output voltage) acquired in step S21 is equal to or greater than the load threshold defined by the load threshold data 118. In other words, the monitor data acquiring unit 121 determines whether or not the load is in a high load state.

[0086] If the load is not equal to or greater than the predetermined load threshold ("No" in step S22), the monitor data acquiring unit 121 returns control to step S21. If the load is equal to or greater than the predetermined load threshold ("Yes" in step S22), the monitor data acquiring unit 121 proceeds to step S01. However, even if the load is equal to or greater than the predetermined load threshold ("Yes" in step S22), the monitor data acquiring unit 121 continues to detect the load current (or output voltage).

[0087] The approximate expression calculation unit 124 also performs function fitting using polynomial and exponential regression on the monitor data 112 (step S23). Specifically, the approximate expression calculation unit 124 determines an approximate expression for the time series data, and the variables define an objective function for time t. PIn order to select an approximate equation that can accurately represent the data up to now, the approximate equation calculation unit 124 may simultaneously fit a plurality of equations (for example, quadratic equations, exponential functions, etc.). The approximate equation calculation unit 124 selects the approximate equation to be applied by using the coefficient of determination R 2 It can be the expression with the largest value.

[0088] The above is an example of the processing flow of the remaining life prediction process (under load). According to this example of the processing flow of the remaining life prediction process (under load), by sampling the power conversion efficiency only when the load is equal to or greater than a certain value, it becomes possible to obtain a trend that is not dependent on the load fluctuation distribution, and it is possible to use monitor data during high loads, when degradation is more noticeable, thereby improving the accuracy of the remaining life prediction.

[0089] FIG. 16 is a diagram showing an example of deterioration over time of power conversion efficiency at a high load current. In an example 600 of deterioration over time of power conversion efficiency at a high load current, the difference in power conversion efficiency versus load current between initial operation and deterioration is graphed. Comparing the power conversion efficiency at low load and high load (the minimum and maximum load currents in the range of +-3σ in the Gaussian distribution in the figure) between initial operation and deterioration, the decrease in power conversion efficiency is greater at high load. Therefore, by limiting the use of monitor data 112 to a range of large load currents, it becomes possible to use only data at points where the efficiency difference during deterioration is large.

[0090] In the example of this graph, since there is a region in the low load range where the power conversion efficiency is extremely poor, it can be said that it is better to adopt the monitor data 112 in the higher load range. However, if adopting the monitor data 112 in the lower load range results in a larger efficiency difference during deterioration, it is better to determine deterioration using the monitor data 112 in a low load state. The above is an example of a circuit life prediction system 1'' according to the third embodiment of the present invention. The circuit life prediction system 1'' according to the third embodiment makes it possible to predict the life of a power conversion circuit at an early stage.

[0091] FIG. 17 is a diagram showing an example of a remaining life prediction result screen. The remaining life prediction result screen is a screen that displays the results of the remaining life prediction process according to the first to fourth embodiments described above. The remaining life prediction result screen is displayed on the UI device 130. An example 700 of the remaining life prediction result screen displays a status display 710 for each power conversion circuit 201. The status display 710 displays data stored in the memory resource 110 (model number (model number of the power conversion circuit 201), application device (diagnosed electronic system 200), approximate formula (F) fitted by regression analysis, etc. P The fitted approximation formula 110b of (t), the life threshold data 114, the remaining life data 115, the current time, etc. 711 are displayed. P Included is a graph 712 that visually displays remaining life data up to the intersection of the time (t) and the life threshold data 114.

[0092] The circuit life prediction system according to the embodiment has been described above. In any of the circuit life prediction systems described above, assuming that the electronic system to be diagnosed 200 incorporating the power conversion circuit 201 to be predicted for its life is handled as a second-hand product, if the remaining life of the power conversion circuit 201 is sufficiently long for the electronic system to be diagnosed 200 to which it is to be applied, there is a possibility that a new life prediction will be performed at the application site.

[0093] In this case, since the monitor data 112 up to that point is stored in the circuit life prediction device 100, prediction is possible if the monitor data 112 is continuously stored in the next application. It is assumed that the power conversion efficiency of the power conversion circuit 201 changes depending on the operating environment (ambient temperature and humidity). Therefore, it is particularly effective when reusing the device in the electronic system 200 to be diagnosed, which is located in a clean room or server room where the environment is controlled and the operating environment is likely to be similar.

[0094] The present invention is not limited to the above-described embodiment, but includes various modifications. For example, the above-described embodiment has been described in detail to clearly explain the present invention, and is not necessarily limited to an embodiment including all of the described configurations. It is possible to replace part of the configuration of an embodiment with another configuration, and it is also possible to add the configuration of another embodiment to the configuration of an embodiment. It is also possible to delete part of the configuration of an embodiment.

[0095] Some or all of the above-described units, configurations, functions, processing units, etc. may be implemented in hardware, for example, by designing them as integrated circuits. Furthermore, the above-described units, configurations, functions, etc. may be implemented in software by a processor interpreting and executing a program that implements each function. Information such as the programs, tables, and files that implement each function can be stored in a memory, a recording device such as a hard disk, or a recording medium such as an IC card, SD card, or DVD.

[0096] It should be noted that the control lines and information lines in the above-described embodiments are those considered necessary for the explanation, and do not necessarily show all the control lines and information lines in the product. In reality, it can be considered that almost all components are interconnected. The present invention has been described above, focusing on the embodiments. [Explanation of symbols]

[0097] 1: Circuit life prediction system, 50: Network, 100: Circuit life prediction device, 110: Memory resource, 111: Life prediction program, 112: Monitor data, 113: Trend data, 114: Life threshold data, 115: Remaining life data, 120: Processor, 121: Monitor data acquisition unit, 122: Remaining life calculation unit, 123: Trend calculation unit, 124: Approximation formula calculation unit, 130: UI device, 140: NI device, 200: Electronic system to be diagnosed, 201: Power conversion circuit, 202: Power source, 203: Load, 204: Monitor module.

Claims

1. A circuit life prediction device having one or more processors and one or more memory resources, The processor: a monitor data acquisition unit that acquires monitor data of the power conversion efficiency of a power conversion circuit of an electronic system including the power conversion circuit whose life span is to be predicted; an approximate expression calculation unit that predicts a future change in the power conversion efficiency of the power conversion circuit using the monitor data; a remaining life calculation unit that calculates a time until the power conversion efficiency reaches a predetermined life threshold as a remaining life; A circuit life prediction device comprising:

2. 2. The circuit life prediction device according to claim 1, The processor: a trend calculation unit that accumulates the monitor data for a predetermined period of time and calculates trend data that indicates a long-term trend; the approximation equation calculation unit calculates a regression equation that approximates the trend data to a predetermined degree or more; the remaining life calculation unit calculates the remaining life using the regression equation. A circuit life prediction device characterized by:

3. 2. The circuit life prediction device according to claim 1, The electronic system includes: a monitor module is provided that acquires input voltage information, input current information, output voltage information, and output current information of the power conversion circuit and calculates the power conversion efficiency; the monitor data acquisition unit acquires the monitor data output from the monitor module; A circuit life prediction device characterized by:

4. 2. The circuit life prediction device according to claim 1, The electronic system includes: a monitor module is provided that acquires input voltage information, input current information, output voltage information, and output current information of the power conversion circuit and calculates the power conversion efficiency; the monitor module acquires a voltage using a voltage sensor to acquire the input voltage information and the output voltage information, and acquires a current using a current sensor to acquire the input current information and the output current information; the monitor data acquisition unit acquires the monitor data output from the monitor module; A circuit life prediction device characterized by:

5. 3. The circuit life prediction device according to claim 2, the trend calculation unit calculates an average value of the monitor data over a predetermined period as the trend data; A circuit life prediction device characterized by:

6. 3. The circuit life prediction device according to claim 2, the trend calculation unit calculates, as the trend data, data of a maximum value, a minimum value, or an average value from among a predetermined number of sample values ​​accumulated for the monitor data; A circuit life prediction device characterized by:

7. 3. The circuit life prediction device according to claim 2, the trend calculation unit calculates data of an arithmetic average, a moving average, or an exponentially smoothed moving average from among a predetermined number of accumulated sample values ​​of the monitor data as the trend data; A circuit life prediction device characterized by:

8. 2. The circuit life prediction device according to claim 1, The electronic system includes: a monitor module is provided that acquires input voltage information, input current information, output voltage information, and output current information of the power conversion circuit and calculates the power conversion efficiency; the monitor data acquisition unit acquires, as the monitor data, the power conversion efficiency output from the monitor module when the voltage specified by the output voltage information or the current specified by the output current information is equal to or greater than a predetermined value; A circuit life prediction device characterized by:

9. 4. The circuit life prediction device according to claim 3, The monitor module includes: including non-volatile memory, storing the monitor data in the nonvolatile memory; The monitor data is continuously stored even when the electronic system is reused and applied to another system. A circuit life prediction device characterized by:

10. A circuit life prediction system using a circuit life prediction device having one or more processors and one or more memory resources, The processor: acquiring monitor data of power conversion efficiency of a power conversion circuit of an electronic system including the power conversion circuit whose life span is to be predicted; an approximate expression calculation step of predicting a future change in the power conversion efficiency of the power conversion circuit using the monitor data; a remaining life calculation step of calculating a time until the power conversion efficiency reaches a predetermined life threshold as a remaining life; A circuit life prediction system comprising:

11. A circuit life prediction method using a circuit life prediction device having one or more processors and one or more memory resources, comprising: The processor: acquiring monitor data of power conversion efficiency of a power conversion circuit of an electronic system including the power conversion circuit whose life span is to be predicted; an approximate expression calculation step of predicting a future change in the power conversion efficiency of the power conversion circuit using the monitor data; a remaining life calculation step of calculating a time until the power conversion efficiency reaches a predetermined life threshold as a remaining life; A circuit life prediction method comprising:

Images