Information processing device

Abstract

To provide an information processing device with which it is possible to identify, and record as a log, a detailed classification of received vibration and mechanical shock.SOLUTION: An information processing device comprises an acceleration sensor and a control unit. The acceleration sensor preserves, in a first storage unit, an acceleration data group for a prescribed number of measurements and having been measured after acceleration, when acceleration exceeds at least a first threshold. The control unit calculates an inclination of jerk norm, acceleration amplitude and a frequency of the acceleration data group. Meanwhile, the control unit classifies the acceleration data group for the prescribed number of measurements into one of high-frequency vibration, low-frequency vibration and mechanical shock on the basis of the inclination of jerk norm, the acceleration amplitude and the frequency and determines a level of vibration or mechanical shock. The control unit preserves, in a second storage unit, a log that indicates the classification result of the acceleration data group for the prescribed number of measurements and the determination result of the level of vibration or mechanical shock.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] FIELD An embodiment of the present invention relates to an information processing device. [Background technology]

[0002] HDDs (Hard Disk Drives), which are auxiliary storage devices built into information processing devices, have moving parts and a complex structure. Therefore, external shocks and vibrations can damage the magnetic head or magnetic disk, corrupt data, or cause problems such as OS startup failure. For this reason, a technology has been developed that detects vibrations and shocks received by information processing devices based on acceleration measured by an acceleration sensor. The detected vibrations and shocks are used by administrators for maintenance of information processing devices and for analyzing the causes of malfunctions. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2020-155169 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the conventional techniques do not specify detailed classifications of the vibrations and shocks that the information processing device receives. [Means for solving the problem]

[0005] An information processing device according to an embodiment includes an acceleration sensor and a control unit. The acceleration sensor measures acceleration at regular intervals, and when the acceleration exceeds at least a first threshold, stores a group of acceleration data measured a specified number of times after the acceleration in a first storage unit. The control unit calculates the slope of a jerk norm of the acceleration data group, the maximum or average value of the acceleration amplitude, which is the amplitude of the acceleration data group, and the frequency of the acceleration data group. The control unit also stores a log indicating a first level of impact in a second storage unit when the slope of the jerk norm is less than a second threshold and the maximum or average value of the acceleration amplitude is greater than a third threshold. The control unit also stores a log indicating a second level of impact in a second storage unit when the slope of the jerk norm is less than the second threshold and the maximum or average value of the acceleration amplitude is less than the third threshold. The control unit also stores a log indicating a first level of high-frequency vibration in a second storage unit when the slope of the jerk norm exceeds at least the second threshold and the frequency exceeds at least a fourth threshold. The control unit stores in the second storage unit a log indicating a first level of low-frequency vibration when the slope of the jerk norm is less than the second threshold, the frequency is less than the fourth threshold, and the acceleration amplitude is greater than the fifth threshold. The control unit stores in the second storage unit a log indicating a second level of low-frequency vibration when the slope of the jerk norm exceeds at least the second threshold, the frequency is less than the fourth threshold, and the acceleration amplitude is less than the fifth threshold. [Brief explanation of the drawings]

[0006] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of an information processing apparatus according to an embodiment. [Figure 2] FIG. 2 is a main flowchart showing an example of the overall flow of a vibration pattern and abnormality determination process executed by the information processing device according to the embodiment. [Figure 3] FIG. 3 is a flowchart showing an example of the flow of the acceleration data acquisition process performed by the microcontroller according to the embodiment. [Figure 4]FIG. 4 is a flowchart showing an example of a processing flow by the CPU according to the embodiment. [Figure 5] FIG. 5 is a flowchart showing an example of the flow of a process for calculating the gradient of the jerk norm, the acceleration amplitude, and the frequency, which is executed by the CPU according to the embodiment. [Figure 6] FIG. 6 is a flowchart showing an example of the flow of the jerk calculation process executed by the CPU according to the embodiment. [Figure 7] FIG. 7 is a flowchart showing an example of the flow of a jerk norm calculation process executed by the CPU according to the embodiment. [Figure 8] FIG. 8 is a flowchart showing an example of the flow of the jerk norm averaging process executed by the CPU according to the embodiment. [Figure 9] FIG. 9 is a flowchart showing an example of the flow of a vibration pattern and abnormality determination process executed by the CPU according to the embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of a log file according to the embodiment. [Figure 11] FIG. 11 is a graph showing an example of acceleration data measured by the acceleration sensor according to the embodiment. [Figure 12] FIG. 12 is a graph showing another example of acceleration data measured by the acceleration sensor according to the embodiment. [Figure 13] FIG. 13 is a graph showing an example of jerk data calculated from the acceleration data of FIG. [Figure 14] FIG. 14 is a graph showing an example of jerk data calculated from the acceleration data of FIG. [Figure 15] FIG. 15 is a graph showing an example of jerk norm data calculated from the jerk data of FIG. [Figure 16] FIG. 16 is a graph showing an example of jerk norm data calculated from the jerk data of FIG. [Figure 17] FIG. 17 shows an example of a moving average value calculated from the jerk norm data of FIG. [Figure 18] FIG. 18 shows an example of a moving average value calculated from the jerk norm data of FIG. [Figure 19] FIG. 19 is a diagram showing an example of a straight line obtained by linearly approximating the averaged jerk norm data of FIG. [Figure 20] FIG. 20 is a diagram showing an example of a straight line obtained by linearly approximating the averaged jerk norm data of FIG. DETAILED DESCRIPTION OF THE INVENTION

[0007] (Embodiment) 1 is a diagram showing an example of the configuration of an information processing device 1 according to this embodiment. As shown in Fig. 1, the information processing device 1 includes an acceleration sensor 11, a microcontroller 12, a CPU (Central Processing Unit) 13, storage 14, a chipset 15, and a main memory 16.

[0008] The acceleration sensor 11 is, for example, a MEMS (Micro Electro Mechanical System) digital acceleration sensor to which MEMS technology is applied, and measures acceleration in three axial directions of the X-axis, Y-axis, and Z-axis.

[0009] The acceleration sensor 11 includes a control circuit 110 , a buffer memory 111 , and a sensor element 112 .

[0010] The sensor element 112 is an element that detects the movement of an object. Specifically, a piezoelectric element, a piezo-resistive element, or the like is used as the sensor element 112 depending on the type of the acceleration sensor 11. In this embodiment, the type of the acceleration sensor 11 is not particularly limited.

[0011] Control circuit 110 is an electronic circuit such as an ASIC (Application Specific Integrated Circuit), and controls acceleration sensor 11. More specifically, control circuit 110 measures acceleration in three axial directions, i.e., the X-axis, Y-axis, and Z-axis, based on the detection result of the movement of an object by sensor element 112.

[0012] The control circuit 110 measures acceleration at regular intervals, and when the measured acceleration exceeds at least a first threshold, stores in the buffer memory 111 acceleration data for a regular number of measurements taken after the acceleration exceeded the first threshold. The value of the first threshold is not particularly limited and may be determined based on the results of a preliminary evaluation or field test by the developer of the information processing device 1. The regular time is the acceleration measurement cycle, but the length of the regular time is not particularly limited. When the measured acceleration is equal to the first threshold, it may be included in either "exceeding the first threshold" or "smaller than (below) the first threshold." Similarly, when comparing with other thresholds in this embodiment, a value equal to the threshold may be included in either "exceeding the threshold" or "smaller than the threshold." Whether a value equal to the threshold is included in "exceeding the threshold" or "smaller than the threshold" is not particularly limited. In this embodiment, the term "greater than" may be interpreted as "exceeding." In this embodiment, the term "equal to or less than" may be interpreted as "less than."

[0013] In this embodiment, the specified number of measurements is, for example, 32. That is, when the control circuit 110 of the acceleration sensor 11 detects acceleration exceeding at least the first threshold, it stores the measurement results of 32 accelerations, including the detected acceleration, in the buffer memory 111. Note that the specified number of measurements is not limited to 32. The acceleration data for the specified number of measurements is an example of an acceleration data group in this embodiment.

[0014] The buffer memory 111 is a storage device that temporarily stores acceleration data, and is an example of a first storage unit in this embodiment. Specifically, the buffer memory 111 is a FIFO (First in First out) buffer. The buffer memory 111 becomes full when it has stored acceleration data for a specified number of measurements. When the buffer memory 111 becomes full, the acceleration data for the specified number of measurements stored in the buffer memory 111 is read by the microcontroller 12.

[0015] The microcontroller 12 is a control device that controls the acceleration sensor 11. More specifically, the microcontroller 12 periodically monitors the buffer memory 111, and when the buffer memory 111 becomes full, that is, when 32 sets of acceleration data have been stored in the buffer memory 111 in this embodiment, the microcontroller 12 acquires the 32 sets of acceleration data from the buffer memory 111.

[0016] The microcontroller 12 sends the 32 acceleration measurement results obtained from the buffer memory 111 of the acceleration sensor 11 to the CPU 13 via the chipset 15. The microcontroller 12 may also store the 32 acceleration measurement results obtained from the buffer memory 111 of the acceleration sensor 11 in the storage 14 via the chipset 15.

[0017] Furthermore, when the acceleration sensor 11 is started up, the microcontroller 12 executes an initialization process for the acceleration sensor 11. In the initialization process for the acceleration sensor 11, the microcontroller 12 sets various parameters and a first threshold value stored in a storage device (not shown) in the microcontroller 12 in the acceleration sensor 11.

[0018] The chipset 15 controls the exchange of data between the CPU 13 , the storage 14 , and the microcontroller 12 .

[0019] The CPU 13 is the main CPU of the information processing device 1. The CPU 13 acquires acceleration data for a specified number of measurements stored in the buffer memory 111 of the acceleration sensor 11 from the microcontroller 12 via the chipset 15, and calculates the slope of the jerk norm, the acceleration amplitude, and the frequency of the shock or vibration from the acquired acceleration data for the specified number of measurements. Hereinafter, in this embodiment, the frequency of the shock or vibration will be simply referred to as the frequency.

[0020] The slope of the jerk norm is the slope of a straight line obtained by linearly approximating the jerk norm data that has been averaged. More specifically, the CPU 13 calculates jerk data from the acquired acceleration data group, and calculates jerk norm data from the calculated jerk data. The CPU 13 then calculates a moving average of the calculated jerk norm data. The CPU 13 linearly approximates the averaged jerk norm data to obtain a straight line. The CPU 13 obtains the angle between the straight line and the initial line indicating 0°, thereby obtaining the slope of the jerk norm. Specific examples of the slope of the jerk norm will be described later with reference to FIGS. 19 and 20.

[0021] The CPU 13 determines the vibration pattern of the measured acceleration data and whether or not there is an abnormality based on the calculated gradient of the jerk norm, the acceleration amplitude, and the frequency. The CPU 13 is an example of a control unit in this embodiment.

[0022] The vibration pattern is a category that classifies the acceleration data group into one of "impact," "low-frequency vibration," and "high-frequency vibration."

[0023] The presence or absence of an abnormality indicates whether the shock or vibration is at an abnormal level. An abnormal level of shock or vibration is, for example, shock or vibration that exceeds the magnitude expected for normal operation of the information processing device 1. The conditions of the jerk norm slope, acceleration amplitude, and frequency that serve as the criteria for an abnormal level of shock or vibration are determined by second to fifth thresholds, which will be described later. The values ​​of the second to fifth thresholds are not particularly limited, but may be determined based on the results of a prior evaluation or field test by the developer of the information processing device 1. The abnormal level is an example of the first level in this embodiment. A level below the abnormal level is an example of the second level in this embodiment.

[0024] In this embodiment, if acceleration exceeding at least the first threshold is measured and the vibration pattern is a "high-frequency vibration," the CPU 13 classifies the vibration as "abnormal" regardless of the magnitude of the vibration.

[0025] For example, in this embodiment, the CPU 13 classifies a vibration pattern as "high-frequency vibration" if the frequency is 1000 Hz or higher. Vibrations of 1000 Hz or higher are considered to be relatively high-frequency vibrations in the environment in which the information processing device 1 is installed. For example, in recent years, reports have been made of HDD failures due to the operation of gas-type fire extinguishing equipment installed in server rooms and degradation of HDD performance due to HDD chatter. These vibrations are in the frequency range of approximately 1000 Hz to 2000 Hz, and vibrations in this frequency range may not be considered in the vibration / shock resistance specifications of the information processing device 1. In other words, even if the movement is not as sudden as a shock, if the vibration frequency is high, it may have some effect on the information processing device 1. Therefore, if acceleration exceeding the first threshold is measured and the vibration pattern is "high-frequency vibration," the CPU 13 classifies the measured acceleration as an abnormal level of high-frequency vibration not anticipated by the designer of the information processing device 1. Note that the frequency conditions for classifying a vibration pattern as "high-frequency vibration" are not limited to the above values.

[0026] In other words, based on the calculated slope of the jerk norm, the acceleration amplitude, and the frequency, the CPU 13 classifies the measured acceleration data into one of "abnormal level of impact," "non-abnormal level of impact," "abnormal level of high frequency vibration," "abnormal level of low frequency vibration," and "non-abnormal level of low frequency vibration." Details of the vibration pattern and the process of determining whether or not there is an abnormality will be described later.

[0027] The CPU 13 stores a log indicating the vibration pattern and the determination result of whether or not there is an abnormality in the storage 14. As an example of a method of storing the log in the storage 14, the CPU 13 writes the log indicating the vibration pattern and the determination result of whether or not there is an abnormality to a log file stored in the storage 14. In this embodiment, the log file includes at least the time the log was written, the vibration pattern, and the presence or absence of an abnormality. The slope of the jerk norm, the acceleration amplitude, and the frequency calculated by the CPU 13 may also be written to the log file. The storage destination of the log file in the storage 14 is not particularly limited, but may be, for example, a folder accessible only by a maintenance technician with prescribed authority.

[0028] In this embodiment, whenever acceleration sensor 11 detects acceleration exceeding the first threshold, CPU 13 executes a process for determining the vibration pattern and the presence or absence of an abnormality, and writes the result to a log file. That is, when acceleration sensor 11 detects acceleration exceeding the first threshold, CPU 13 writes at least one of a log indicating an abnormal level of impact, a log indicating a non-abnormal level of impact, a log indicating an abnormal level of high-frequency vibration, a log indicating an abnormal level of low-frequency vibration, and a log indicating a non-abnormal level of low-frequency vibration to a log file stored in storage 14.

[0029] The storage 14 is an auxiliary storage device, such as an HDD (Hard Disk Drive), built into the information processing device 1. The storage 14 stores log files generated by the CPU 13. The storage 14 also stores various programs executed by the CPU 13. The storage 14 is an example of a second storage unit in this embodiment.

[0030] The main memory 16 is a readable and writable storage device such as a RAM (Random Access Memory), and is the primary storage device of the information processing device 1. The information processing device 1 may further include a ROM (Read Only Memory) or the like in addition to the configuration shown in FIG.

[0031] Next, the flow of processing executed by the information processing device 1 configured as above will be described.

[0032] FIG. 2 is a main flowchart showing an example of the overall flow of the vibration pattern and abnormality determination process executed by the information processing device 1 according to this embodiment.

[0033] First, when the information processing device 1 is started up, the microcontroller 12 executes an initialization process for the acceleration sensor 11 (S1). Specifically, the microcontroller 12 sets various parameters and a first threshold value stored in a storage device (not shown) in the microcontroller 12 in the acceleration sensor 11.

[0034] Then, the acceleration sensor 11 measures the acceleration of the information processing device 1 (S2).

[0035] Next, the control circuit 110 of the acceleration sensor 11 compares the measured acceleration with a first threshold. If the measured acceleration is less than the first threshold (S3 "No"), the control circuit 110 does not save the measured acceleration. Then, the acceleration sensor 11 waits until a specified time has passed since the previous measurement (S4 "No"), and when the specified time has passed since the previous measurement (S4 "Yes"), returns to the processing of S2 and measures the acceleration of the information processing device 1.

[0036] Furthermore, if the measured acceleration is greater than the first threshold value (S3 "Yes"), the control circuit 110 of the acceleration sensor 11 stores the first acceleration data, which is the measurement result of the acceleration, in the buffer memory 111 (S5).

[0037] Then, the acceleration sensor 11 waits until a specified time has passed since the previous measurement (S6 "No"), and when the specified time has passed since the previous measurement (S6 "Yes"), the acceleration sensor 11 measures the acceleration of the information processing device 1 (S7).

[0038] Then, the control circuit 110 of the acceleration sensor 11 stores the acceleration data, which is the measurement result of the measured acceleration, in the buffer memory 111 (S8). The processes of S6 to S9 are repeated until the acceleration data stored in the buffer memory 111 reaches the specified number of measurements (32 in this embodiment). Through the processes of S5 to S9, the acceleration sensor 11 stores in the buffer memory 111 a group of acceleration data for the specified number of measurements measured after the acceleration exceeded the first threshold.

[0039] Here, the microcontroller 12 periodically monitors the buffer memory 111, and when the buffer memory 111 becomes full, that is, when 32 acceleration measurement results are stored in the buffer memory 111 in this embodiment (S9 "Yes"), the microcontroller 12 reads acceleration data for a specified number of measurements from the buffer memory 111 of the acceleration sensor 11. Once the microcontroller 12 has read the acceleration data for the specified number of measurements, the control circuit 110 of the acceleration sensor 11 deletes the acceleration data in the buffer memory 111 (S10).

[0040] Then, the process returns to S2, and the processes of S2 to S10 are repeatedly executed while the information processing device 1 is operating. When the power supply of the information processing device 1 is turned off, the process of this flowchart ends.

[0041] FIG. 3 is a flowchart showing an example of the flow of the acceleration data acquisition process performed by the microcontroller 12 according to this embodiment.

[0042] The microcontroller 12 accesses the acceleration sensor 11 (S21), and if the acceleration data for the specified number of measurements is stored in the buffer memory 111 of the acceleration sensor 11 (S22 "Yes"), the microcontroller 12 reads the acceleration data for the specified number of measurements from the buffer memory 111 of the acceleration sensor 11 (S23). The microcontroller 12 sends the acceleration data for the specified number of measurements acquired from the buffer memory 111 of the acceleration sensor 11 to the CPU 13 via the chipset 15. Then, the process returns to S21.

[0043] If the acceleration data for the specified number of measurements has not been stored in the buffer memory 111 of the acceleration sensor 11 (S22 "No"), the process returns to S21. While the information processing device 1 is operating, the processes of S21 to S23 are repeatedly executed. When the power of the information processing device 1 is turned off, the process of this flowchart ends.

[0044] Although the standby time is not shown in FIG. 5, the buffer memory 111 may be configured to access S21 at regular intervals.

[0045] 4 is a flowchart showing an example of the flow of processing by the CPU 13 according to this embodiment. The processing of this flowchart starts when the microcontroller 12 sends acceleration data from the acceleration sensor 11 for a specified number of measurements to the CPU 13.

[0046] The CPU 13 acquires the acceleration data for a specified number of measurements stored in the buffer memory 111 of the acceleration sensor 11 from the microcontroller 12 via the chipset 15, and executes a process of calculating the slope, acceleration amplitude, and frequency of the jerk norm based on the acquired acceleration data for the specified number of measurements (S11). The details of the process of calculating the slope, acceleration amplitude, and frequency of the jerk norm will be described later with reference to FIGS. 5 to 8.

[0047] Next, the CPU 13 executes a process of determining the vibration pattern and the presence or absence of an abnormality based on the calculated slope of the jerk norm, the acceleration amplitude, and the frequency (S12). The CPU 13 writes the vibration pattern and the determination result of the presence or absence of an abnormality together with time information in a log file in the storage 14.

[0048] Next, the calculation process of the jerk norm gradient, acceleration amplitude, and frequency in S11 in the flowchart of FIG. 4 will be described.

[0049] 5 is a flowchart showing an example of the flow of a process for calculating the gradient of the jerk norm, the acceleration amplitude, and the frequency, which is executed by the CPU 13 according to this embodiment. This flowchart is a subroutine corresponding to S11 in the flowchart of FIG.

[0050] First, the CPU 13 acquires the acceleration data for a predetermined number of measurements stored in the buffer memory 111 of the acceleration sensor 11 from the microcontroller 12 via the chipset 15 (S1110).

[0051] Then, the CPU 13 calculates the jerk (jerk) for each of the x, y and z axes from the acceleration data acquired for the specified number of measurements (S1120). Details of the jerk calculation process will be described with reference to FIG.

[0052] Next, the CPU 13 calculates a jerk norm from the calculated jerk of each of the x, y, and z axes (S1130). Details of the jerk norm calculation process will be described with reference to FIG.

[0053] Then, the CPU 13 executes averaging processing of the calculated jerk norm (S1140). The jerk norm averaging processing is processing for calculating, for example, a moving average value of the jerk norm. Details of the jerk norm averaging processing will be described with reference to FIG. 8.

[0054] Then, the CPU 13 linearly approximates the averaged jerk norm (S1150). By linearly approximating the averaged jerk norm, the rate of change of the jerk norm is expressed as the slope of a straight line.

[0055] Then, the CPU 13 calculates the gradient (rate of change) of the linearly approximated jerk norm, the acceleration amplitude, and the frequency (S1160).

[0056] Specifically, the CPU 13 calculates and averages the norm of the acceleration data acquired for a specified number of measurements in the same manner as for the jerk data, and calculates the intercept of a straight line obtained by linearly approximating the data as the magnitude of the acceleration amplitude. The averaging process for the specified number of measurements is a process of calculating a moving average of the acceleration data for the specified number of measurements. The method for calculating the moving average of the acceleration data for the specified number of measurements is the same as the jerk norm averaging process described in FIG. 8. In this embodiment, the intercept of a straight line obtained by linearly approximating a group of acceleration data that has been subjected to moving averaging and then converted to absolute values ​​is referred to as the acceleration amplitude.

[0057] In the case of acceleration caused by an impact, the linear approximation generally slopes downward to the right, so the intercept value is the maximum value of the amplitude in the acceleration data for the specified number of measurements, or a value close to the maximum value.In the case of acceleration caused by vibration, the linear approximation generally approaches horizontal, so the intercept value is the average value of the amplitude in the acceleration data for the specified number of measurements, or a value close to the average value.

[0058] The method for calculating the acceleration amplitude may be the same as that for calculating the jerk amplitude, and is not limited to this. For example, the CPU 13 may calculate the maximum amplitude value of the acceleration data group as the acceleration amplitude. Furthermore, the CPU 13 may use the maximum amplitude value in the case of an impact, or the average amplitude value in the case of a vibration, to determine whether or not there is an abnormality, as described below.

[0059] Furthermore, the CPU 13 calculates the frequency of the acceleration based on the acceleration amplitude and the jerk amplitude. The acceleration frequency is calculated using the following equations (1) to (3). In equations (1) to (3), acceleration is defined as P, jerk as Q, and frequency as f1. A in equation (1) represents the acceleration amplitude, and 2πAf1 in equation (2) represents the jerk amplitude. Note that, to simplify the calculations, equations (1) to (3) do not take into account phase shifts. Note that the equations for calculating the acceleration frequency are not limited to equations (1) to (3).

[0060]

number

[0061] Generally, acceleration caused by an impact given to the information processing device 1 is characterized by an initial large jerk norm that then rapidly decreases. Therefore, the slope of the linearly approximated jerk norm increases in the negative direction. In contrast, acceleration caused by vibration given to the information processing device 1 shows little change in the magnitude of the linearly approximated jerk norm, and the slope is close to 0°.

[0062] In this embodiment, a horizontal state of the linearly approximated jerk norm is defined as 0°, negative angles are represented by negative values, and positive angles are represented by positive values. When the linearly approximated jerk norm is tilted at a negative angle, the stronger the degree of tilt, the further it deviates from the horizontal state, and therefore the smaller the angle indicating the tilt of the linearly approximated jerk norm becomes.

[0063] Specifically, the line obtained by linearly approximating the jerk norm calculated from the acceleration caused by the impact slopes more sharply downward to the right than the line obtained by linearly approximating the jerk norm calculated from the acceleration caused by vibration, and therefore the slope of the line obtained by linearly approximating the jerk norm calculated from the acceleration caused by the impact is smaller than the slope of the line obtained by linearly approximating the jerk norm calculated from the acceleration caused by vibration. This characteristic of the slope of the linearly approximated jerk norm makes it possible to classify the causes of the acceleration of the information processing device 1 into shock and vibration.

[0064] Here, the processing of this flowchart ends, and the process proceeds to S12 in FIG.

[0065] Next, the jerk calculation process in S1120 in the flowchart of FIG. 5 will be described.

[0066] 6 is a flowchart showing an example of the flow of the jerk calculation process according to this embodiment, which is executed by the CPU 13. This flowchart is a subroutine corresponding to S1120 in the flowchart of FIG.

[0067] First, the CPU 13 assigns "0" to a variable i for counting the number of times of processing (S1121). The variable i="0" represents the first data in the acceleration data for the specified number of measurements.

[0068] Next, the CPU 13 calculates the difference between the acceleration in the i-th measurement and the acceleration in the (i+1)-th measurement for each of the X-axis, Y-axis, and Z-axis (S1122). The value of the difference is the i-th jerk.

[0069] Specifically, the “j x (i)” indicates the i-th jerk in the X-axis direction. “α x (i)-α x (i+1)” is the formula for calculating the difference between the acceleration in the X-axis direction in the i-th measurement and the acceleration in the X-axis direction in the i+1-th measurement. Similarly, “j y (i)” indicates the i-th jerk in the Y-axis direction, and “j z (i)” indicates the i-th jerk in the Z-axis direction.

[0070] Next, the CPU 13 assigns "i+1" to the variable i (S1123).

[0071] Then, if the value of the variable i after substituting "i+1" is less than "N-1" (S1124 "Yes"), the process returns to S1122, and the CPU 13 calculates the difference between the acceleration corresponding to the next measurement count and the acceleration corresponding to the next measurement count as the next jerk. N indicates the number of times acceleration is measured. In this embodiment, N=32. Note that in this flowchart, the data when i=0 is considered to be the first data, so when N=32, i=31 is the last data.

[0072] If the value of the variable i after substituting "i+1" is equal to or greater than "N-1", that is, if the value of the variable i after substituting "i+1" is equal to or greater than 31 (S1124 "No"), the CPU 13 calculates the i-th X-axis, Y-axis, and Z-axis jerk value "j x (i)”, “j y (i)” and “j z The previously calculated (i-1)-th jerk values ​​for the X-axis, Y-axis, and Z-axis are substituted into each of "(i)" (S1125). Since the jerk is the difference between the acceleration measured the i-th time and the acceleration measured the (i+1)th time, there is no object for calculating the difference in the last measured acceleration. For this reason, in the processing of S1125, the previous jerk value is substituted for the final jerk value.

[0073] Here, the processing of this flowchart ends, and the process proceeds to S1130 in FIG.

[0074] Next, the jerk norm calculation process in S1130 in the flowchart of FIG. 5 will be described.

[0075] 7 is a flowchart showing an example of the flow of the jerk norm calculation process according to this embodiment, which is executed by the CPU 13. This flowchart is a subroutine corresponding to S1130 in the flowchart of FIG.

[0076] First, the CPU 13 assigns "0" to a variable i for counting the number of times of processing (S1131). In this flowchart, the variable i="0" also represents the first data in the acceleration data for the specified number of measurements.

[0077] Next, the CPU 13 obtains the i-th jerk norm "||f(i)||" from the i-th jerk of the X-axis, Y-axis, and Z-axis by the following equation (4) (S1132). x 2 (i)” indicates the square of the i-th jerk in the X-axis direction. Similarly, “j y 2 (i)” indicates the square of the i-th jerk in the Y-axis direction, and “j z 2 (i)” indicates the square of the i-th Z-axis jerk.

[0078]

number

[0079] Next, the CPU 13 assigns "i+1" to the variable i (S1133).

[0080] If the value of the variable i after substituting "i+1" is less than "N-1" (S1134 "Yes"), the process returns to S1132, and the CPU 13 calculates the jerk norm corresponding to the next measurement count. As in FIG. 6, N indicates the number of acceleration measurements. In this embodiment, N=32.

[0081] If the value of the variable i after substituting "i+1" is equal to or greater than "N-1", that is, if the value of the variable i after substituting "i+1" is equal to or greater than 31 (S1134 "No"), the CPU 13 substitutes the previously calculated (i-1)-th jerk norm value "||f(i)||" for the ith X-axis, Y-axis, and Z-axis jerk norm value (S1135).

[0082] Here, the processing of this flowchart ends, and the process proceeds to S1140 in FIG.

[0083] Next, the jerk norm averaging process in S1140 in the flowchart of FIG. 5 will be described.

[0084] 8 is a flowchart showing an example of the flow of the jerk norm averaging process according to this embodiment, which is executed by the CPU 13. This flowchart is a subroutine corresponding to S1140 in the flowchart of FIG.

[0085] In the jerk norm averaging process in this flowchart, the CPU 13 calculates the moving average value of the jerk norm using a sliding window method. In this embodiment, the window size "n" is, for example, "4." That is, the CPU 13 repeats the process of calculating the average value for each of four consecutive jerk norm data. In this embodiment, the average of the jerk norm refers to the moving average.

[0086] First, the CPU 13 assigns "0" to a variable i for counting the number of times of processing (S1141). In this flowchart, the variable i="0" also represents the first data in the acceleration data for the specified number of measurements.

[0087] Next, the CPU 13 assigns "0" to the variable sum indicating the sum value. The CPU 13 also assigns "i" to the variable k for specifying the jerk norm to be added to calculate the moving average (S1142).

[0088] Then, the CPU 13 assigns the sum of sum and the k-th jerk norm "||f(k)||" to the variable sum (S1143).

[0089] Next, the CPU 13 assigns "k+1" to the variable k (S1142).

[0090] Then, if the value of variable k after substituting "k+1" is less than "i+n" (S1145 "Yes"), that is, if the addition process of the jerk norms included in the currently processed window has not been completed, the process returns to S1143, and the next jerk norm included in the currently processed window is added to variable sum. The processes of S1143 to S1145 are repeated until all jerk norms included in the currently processed window have been added. For example, if the window size "n" is "4", in the first loop process, the processes of S1143 to S1145 are repeated until the first through fourth jerk norms have been added.

[0091] Then, if the value of the variable k after substituting "k+1" is equal to or greater than "i+n" (S1145 "No"), the CPU 13 divides the variable sum by n to obtain the moving average value "||j smooth (i) Calculate ||” (S1146).

[0092] Next, the CPU 13 assigns "i+1" to the variable i (S1147).

[0093] Then, if the value of the variable i after substituting "i+1" is less than "N-(n-1)" (S1148 "Yes"), that is, if the last window has not been reached, the process returns to S1142, and the CPU 13 calculates the moving average value of the jerk norm in the next window.

[0094] If the value of the variable i after substituting "i+1" is equal to or greater than "N-(n-1)" (S1148 "No"), the process of this flowchart ends and proceeds to the process of S1150 in FIG.

[0095] If the window size "n" is "4" and the total number of samples "N" of the jerk norm is "32", and the moving calculation is performed while shifting the window by one sample data, the number of samples of the moving average value of the jerk norm obtained by this flowchart will be 29.

[0096] In this flowchart, the window size is described as "4", but this window size is an example and is not limited to this.

[0097] Next, the vibration pattern and abnormality determination process of S12 in the flowchart of FIG. 4 will be described.

[0098] 9 is a flowchart showing an example of the flow of a vibration pattern and abnormality determination process executed by the CPU 13 according to this embodiment. This flowchart is a subroutine corresponding to S12 in the flowchart of FIG.

[0099] First, the CPU 13 compares the slope of the jerk norm with a second threshold. In this embodiment, the slope of the linearly approximated jerk norm is expressed as 0° when it is horizontal, and is expressed as a positive number when the linearly approximated jerk norm slopes upward to the right, and as a negative number when it slopes downward to the right. That is, in this embodiment, the slope of the linearly approximated jerk norm increases as it moves counterclockwise away from the starting line indicating 0°, and decreases as it moves clockwise away from the starting line indicating 0°. The slope of the jerk norm will be specifically described with reference to FIGS. 19 and 20.

[0100] Because an impact waveform is a single event, it is characterized by an initial large-amplitude acceleration that then decays rapidly. On the other hand, a vibration waveform does not decay as rapidly as an impact and is characterized by a continuous waveform. For this reason, the slope of the line linearly approximating the jerk norm calculated from the acceleration caused by an impact is smaller than the slope of the line linearly approximating the jerk norm calculated from the acceleration caused by vibration.

[0101] If the gradient of the jerk norm is smaller than the second threshold value (S1201 "Yes"), the CPU 13 classifies the vibration pattern of the measured acceleration as "impact" (S1202).

[0102] Next, the CPU 13 compares the acceleration amplitude with a third threshold value. If the acceleration amplitude is greater than the third threshold value (S1203 "Yes"), the CPU 13 classifies the impact as "abnormal" (S1204).

[0103] Then, the CPU 13 writes a log indicating the vibration pattern and the determination result of whether or not there is an abnormality in a log file stored in the storage 14 (S1205). Specifically, in this case, the CPU 13 writes a log indicating an abnormal level of impact in the log file.

[0104] If the acceleration amplitude is equal to or less than the third threshold (S1203 "Yes"), the CPU 13 classifies the impact as "normal" (S1206). In this case, the process proceeds to S1205, and the CPU 13 writes a log indicating that the impact is not at an abnormal level into the log file.

[0105] Furthermore, if the slope of the jerk norm is equal to or greater than the second threshold (S1201 "No"), the CPU 13 identifies the measured acceleration as "vibration." Then, the CPU 13 compares the frequency of the vibration with a fourth threshold. If the frequency of the vibration is equal to or greater than the fourth threshold (S1207 "No"), the CPU 13 classifies the vibration pattern of the measured acceleration as "high-frequency vibration" (S1208). The fourth threshold is, for example, 1000 Hz.

[0106] When the acceleration equal to or greater than the first threshold is measured and the vibration pattern is "high frequency vibration," the CPU 13 classifies the vibration as "abnormal" regardless of the magnitude of the vibration (S1209). In this case, the CPU 13 proceeds to the process of S1205, and writes a log indicating the abnormal level of high frequency vibration to the log file.

[0107] If the frequency of the vibration is lower than the fourth threshold value (S1207 "Yes"), the CPU 13 classifies the vibration pattern of the measured acceleration as "low frequency vibration" (S1210).

[0108] Next, the CPU 13 compares the acceleration amplitude with a fifth threshold. If the acceleration amplitude is greater than the fifth threshold (S1211 "Yes"), the CPU 13 classifies the low-frequency vibration as "abnormal" (S1212). In this case, the process proceeds to S1205, and the CPU 13 writes a log indicating the abnormal level of low-frequency vibration to a log file.

[0109] In this embodiment, the fifth threshold is smaller than the third threshold. The third threshold is a threshold that conforms to the shock resistance specifications of the information processing device 1. The fifth threshold is a threshold that conforms to the vibration resistance specifications of the information processing device 1. The magnitude relationship between the third threshold and the fifth threshold is not limited to this, and the third threshold and the fifth threshold may be the same value.

[0110] On the other hand, if the acceleration amplitude is equal to or less than the fifth threshold (S1211 "No"), the CPU 13 classifies the low-frequency vibration as "normal" (S1213). In this case, the process proceeds to S1205, and the CPU 13 writes a log indicating that the low-frequency vibration is not at an abnormal level into the log file.

[0111] Here, the processing of this flowchart ends, and the process returns to the processing of FIG.

[0112] Next, the log files stored in the storage 14 of this embodiment will be described.

[0113] Fig. 10 is a diagram showing an example of a log file 90 according to this embodiment. As shown in Fig. 10, the log file 90 is, for example, in a comma-separated file format, in which time, vibration pattern, presence or absence of abnormality, jerk norm gradient, acceleration amplitude, and frequency are associated with each other. The time is the time when the CPU 13 wrote the log. The presence or absence of abnormality is an example of the level of impact and vibration in this embodiment.

[0114] The log file 90 stored in the storage 14 is collected and checked by a maintenance technician, for example, at the timing of maintenance. The maintenance technician may use the logs recorded in the log file 90 to make suggestions to the user for environmental improvements, etc.

[0115] The file format and items of the log file are not limited to the example shown in FIG. 10, and general log file specifications may be adopted as appropriate.

[0116] Next, specific examples of acceleration data measured by the acceleration sensor 11 in this embodiment, and jerk data and jerk norm calculated from the acceleration data will be described.

[0117] FIG. 11 is a graph showing an example of acceleration data measured by the acceleration sensor 11 according to this embodiment. The horizontal axis of FIG. 11 represents the number of data samples, and the vertical axis represents acceleration. FIG. 11 shows, in a line graph, 32 points of acceleration sample data measured 32 times at specified time intervals by the acceleration sensor 11. The acceleration waveform shown in FIG. 11 is an example of acceleration on the X-axis, Y-axis, and Z-axis generated by an impact. Since the impact waveform is a one-off event, it is characterized by an initial large-amplitude acceleration being recorded, which then rapidly decays thereafter.

[0118] Fig. 12 is a graph showing another example of acceleration data measured by the acceleration sensor 11 according to this embodiment. As in Fig. 11, the horizontal axis of Fig. 12 represents the number of data samples, and the vertical axis represents acceleration. Note that, because the waveform shown in Fig. 12 has smaller acceleration values ​​than the waveform shown in Fig. 11, for ease of display, the vertical axis of Fig. 12 is scaled in finer units than in Fig. 11.

[0119] The acceleration waveform shown in Figure 12 is an example of acceleration on the X-, Y-, and Z-axes caused by vibration. A vibration waveform is characterized by not decaying as rapidly as an impact, but rather being a continuous waveform. Vibrations / impacts that occur in the real world often have random waveforms that are a combination of multiple components, but generally the above characteristics are maintained.

[0120] Figure 13 is a graph showing an example of jerk data calculated from the acceleration data of Figure 11. The horizontal axis of Figure 13 represents the number of data samples, and the vertical axis represents the jerk. The jerk data shows, more clearly than the acceleration data of Figure 11, that a large amplitude acceleration is recorded at the beginning of the shock waveform, followed by a rapid attenuation.

[0121] Fig. 14 is a graph showing an example of jerk data calculated from the acceleration data of Fig. 12. As with Fig. 13, the horizontal axis of Fig. 14 represents the number of data samples, and the vertical axis represents jerk. The jerk data also shows the characteristics of a vibration waveform with a gradual attenuation and a continuous waveform.

[0122] Fig. 15 is a graph showing an example of jerk norm data calculated from the jerk data of Fig. 13. Fig. 16 is a graph showing an example of jerk norm data calculated from the jerk data of Fig. 14. The horizontal axis of Figs. 15 and 16 represents the number of data samples, and the vertical axis represents the jerk norm. The CPU 13 calculates the jerk norm data from the jerk data on the X-axis, Y-axis, and Z-axis, and as a result, the three graphs in Figs. 13 and 14 are integrated into one graph in Figs. 15 and 16.

[0123] Next, an example of averaging processing of jerk norm data will be described.

[0124] Fig. 17 shows an example of a moving average value calculated from the jerk norm data of Fig. 15. Fig. 18 shows an example of a moving average value calculated from the jerk norm data of Fig. 16. The horizontal axis of Figs. 17 and 18 represents the number of data samples, and the vertical axis represents the averaged jerk norm. By subjecting the jerk norm data to averaging processing by the CPU 13, waveform disturbances are smoothed, facilitating subsequent linear approximation.

[0125] Next, an example of a linear approximation of the averaged jerk norm data will be described.

[0126] Fig. 19 is a diagram showing an example of a straight line L1 obtained by linearly approximating the averaged jerk norm data of Fig. 17. The horizontal axis of Fig. 19 represents the number of data samples, and the vertical axis represents the averaged jerk norm. Fig. 19 also shows an initial line L0 and a straight line L3 that illustrates the second threshold value "-g°" to explain the slope of the straight line L1.

[0127] Generally, the slope of the straight line L1, which is a linear approximation of the jerk norm data calculated from the acceleration caused by the impact, slopes downward to the right. Therefore, as shown in Figure 19, the angle θ1 between the initial line L0 and the straight line L1 is a negative angle "-d°". This negative angle "-d°" is the slope of the straight line L1.

[0128] In the example shown in FIG. 19, the angle θ3 between the initial line L0 and the straight line L3 is "-g°," which is an example of the second threshold value. The angle θ1 between the initial line L0 and the straight line L1, which is a linear approximation of the jerk norm data, is "-d°," which is smaller than the second threshold value "-g°." In this case, as described in FIG. 9, the CPU 13 classifies the vibration pattern of the acceleration from which the straight line L1 is calculated as "impact."

[0129] 20 is a diagram showing an example of a straight line L2 obtained by linearly approximating the averaged jerk norm data of FIG. 18. The initial line L0 and straight line L3 shown in FIG. 20 are the same as those in FIG.

[0130] The angle θ2 between the initial line L0 and the straight line L2 is a negative angle, "-e°." This negative angle, "-e°," is the slope of the straight line L2. Generally, the slope of the straight line L2, which is a linear approximation of the jerk norm data calculated from the acceleration caused by vibration, slopes more gently downward to the right than the slope of the straight line L1, which is a linear approximation of the jerk norm data calculated from the acceleration caused by impact. Therefore, although the angle θ2 between the initial line L0 and the straight line L2, "-e°," is a negative angle, it is greater than the angle θ1, "-d°," between the initial line L0 and the straight line L1. Furthermore, the angle θ1, "-e°," between the initial line L0 and the straight line L2, is greater than the second threshold value, "-g°." In this case, the CPU 13 classifies the vibration pattern of the acceleration from which the straight line L2 is calculated as "high-frequency vibration" or "low-frequency vibration."

[0131] The CPU 13 also calculates the intercept values ​​of the lines L1 and L2 as the amplitude of the jerk data. In Figures 19 and 20, the number of data samples on the horizontal axis starts at "1," so the vertical axis value at "1" on the horizontal axis represents the amplitude of the jerk data. The method for calculating the amplitude of the jerk data may be the same as the method for calculating the acceleration amplitude, and is not limited to this. For example, if the maximum value of the amplitude of the acceleration data group is used as the acceleration amplitude, the CPU 13 calculates the maximum value of the amplitude of the jerk data group as the amplitude of the jerk data. If the average value of the amplitude of the acceleration data group is used as the acceleration amplitude, the CPU 13 calculates the average value of the amplitude of the jerk data group as the amplitude of the jerk data.

[0132] As described above, the information processing device 1 of this embodiment includes an acceleration sensor 11 and a CPU 13. The acceleration sensor 11 measures acceleration at regular intervals. If the measured acceleration is equal to or greater than a first threshold, the acceleration sensor 11 stores a group of acceleration data in the buffer memory 111, including first acceleration data representing the measurement result of the acceleration equal to or greater than the first threshold. The CPU 13 also acquires the group of acceleration data and calculates the slope of the jerk norm, the acceleration amplitude, and the frequency of the shock or vibration from the group of acceleration data. The CPU 13 classifies the measured acceleration data into one of the vibration patterns of shock, high-frequency vibration, and low-frequency vibration based on the calculated slope and frequency of the jerk norm. If the vibration pattern is classified as shock or low-frequency vibration, the CPU 13 further determines whether an abnormality exists based on the acceleration amplitude. The CPU 13 stores the vibration pattern and the determination result of the abnormality in the storage 14. Therefore, the information processing device 1 of this embodiment can record detailed classifications of the vibrations and shocks received by the information processing device 1 as logs.

[0133] Generally, when an information processing device 1 is used in a field such as a social infrastructure system, it is required to operate stably for a long period of time. By recording the detailed classification of the vibrations and shocks received by the information processing device 1 as a log, as in the information processing device 1 of this embodiment, maintenance personnel can analyze in detail the shocks or vibrations received by the information processing device 1. This can improve the accuracy of maintenance according to the records of the shocks or vibrations and of analysis of the causes of any malfunctions that occur, thereby contributing to the long-term stable operation of the information processing device 1.

[0134] Furthermore, when the acceleration sensor 11 detects acceleration equal to or greater than the first threshold, the information processing device 1 of this embodiment writes at least one of a log indicating an abnormal level of impact, a log indicating a non-abnormal level of impact, a log indicating an abnormal level of high frequency vibration, a log indicating an abnormal level of low frequency vibration, and a log indicating a non-abnormal level of low frequency vibration to a log file stored in the storage 14. According to the information processing device 1 of this embodiment, by leaving a log whenever acceleration equal to or greater than the first threshold is detected, it is possible to leave a detailed record of impacts and vibrations that may affect the information processing device 1, not just when an abnormality is determined.

[0135] Furthermore, the log file 90 of this embodiment includes at least the time the log was written, the vibration pattern, and whether or not an abnormality occurred, so that a maintenance technician who checks the log file 90 can use this information to analyze the cause of a malfunction in the information processing device 1.

[0136] (Variation) The processes described in the above embodiment as processes executed by the CPU 13 may be executed by the microcontroller 12. In this case, the microcontroller 12 is an example of a control unit. According to the information processing device 1 of this modification, it is possible to reduce the processing load on the CPU 13 while achieving the same effects as those of the above embodiment.

[0137] The program executed by the information processing device 1 of this embodiment is provided as a file in an installable or executable format recorded on a computer-readable recording medium such as a CD-ROM, a flexible disk (FD), a CD-R, or a DVD (Digital Versatile Disk).

[0138] The program executed by the information processing device 1 of this embodiment may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. The program executed by the information processing device 1 of this embodiment may be provided or distributed via a network such as the Internet.

[0139] Furthermore, the program executed by the information processing device 1 of this embodiment may be provided by being pre-installed in a ROM (not shown) of the information processing device 1 or the like.

[0140] The program executed by the information processing device 1 of this embodiment has a modular structure including the functions of the CPU 13 described above, and in terms of actual hardware, the CPU 13 reads and executes the program from a storage medium such as the storage 14, thereby loading each of the above-mentioned parts onto a main storage device such as the main memory 16, and generating modules corresponding to the functions of the CPU 13 described above on the main storage device.

[0141] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0142] 1. Information processing equipment 11 Acceleration sensor 12 Microcontrollers 13 CPU 14. Storage 15 chipsets 16 Main Memory 90 Log Files 110 control circuit 111 Buffer Memory 112 Sensor element L0 starting line L1~L3 straight line

Claims

1. an acceleration sensor that measures acceleration at predetermined time intervals, and when the acceleration exceeds at least a first threshold, stores a group of acceleration data measured for a predetermined number of measurements after the acceleration is measured in a first storage unit; a slope of a jerk norm of the acceleration data group, a maximum value or an average value of the acceleration amplitude, which is the amplitude of the acceleration data group, and a frequency of the acceleration data group are calculated; if the slope of the jerk norm is less than a second threshold and the maximum value or the average value of the acceleration amplitude is greater than a third threshold, a log indicating a first level impact is stored in a second storage unit; if the slope of the jerk norm is less than the second threshold and the maximum value or the average value of the acceleration amplitude is less than the third threshold, a log indicating a second level impact is stored in the second storage unit; and if the slope of the jerk norm exceeds at least the second threshold, a control unit that stores in the second storage unit a log indicating a first level of high frequency vibration when the frequency exceeds at least a fourth threshold, stores in the second storage unit a log indicating a first level of low frequency vibration when the slope of the jerk norm is less than the second threshold, the frequency is less than the fourth threshold, and the acceleration amplitude is greater than a fifth threshold, and stores in the second storage unit a log indicating a second level of low frequency vibration when the slope of the jerk norm exceeds at least the second threshold, the frequency is less than the fourth threshold, and the acceleration amplitude is less than the fifth threshold; An information processing device comprising:

2. When the acceleration sensor detects acceleration exceeding the first threshold, the control unit writes at least one of a log indicating the first level impact, a log indicating the second level impact, a log indicating the first level high-frequency vibration, a log indicating the first level low-frequency vibration, and a log indicating the second level low-frequency vibration to a log file stored in the second storage unit. The information processing device according to claim 1 .

3. the log file includes at least a log writing time, a vibration pattern, and a level of shock and vibration; The vibration pattern indicates whether the acceleration data group corresponds to a classification of an impact, a low-frequency vibration, or a high-frequency vibration. The information processing device according to claim 2 .

4. The control unit is a main CPU (Central Processing Unit) of the information processing device. The information processing device according to claim 1 .

5. The control unit is a microcontroller that controls the acceleration sensor. The information processing device according to claim 1 .

6. The control unit Calculating jerk data from the acceleration data group; Calculating jerk norm data from the calculated jerk data; Calculating a moving average of the calculated jerk norm data; A straight line is obtained by linearly approximating the averaged jerk norm data. The angle between the straight line and the initial line is calculated to obtain the gradient of the jerk norm. The information processing device according to claim 1 .

Images