JP2014092521A - Time synchronization method and measuring system using the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a time synchronization method capable of relatively synchronizing a local time between individual sensors without being affected by delay even if the line speed is low.SOLUTION: The time synchronization method is a time synchronization method in a measuring system which includes: a first measuring device provided with a CPU, storage means, a first sensor, and a first clock; and a second measuring device provided with a CPU, storage means, a second sensor, and a second clock, the first and second measuring devices being connected via a network, and includes the step S1 of identifying the waveform showing time transition of the physical quantity in each sensor, respectively from the measured value of the first sensor and the measured value of the second sensor; the step S2 of comparing the feature part of each identified waveform and calculating a local time lag of the second clock for the local time of the first clock; and the step 3 of subtracting the calculated local time lag from the local time of the second clock and synchronizing the local time of the second clock with the local time of the first clock.

Description

  The present invention relates to a time synchronization method for synchronizing the time of each measurement device in a measurement system including measurement devices installed at a plurality of points and connected to a network.

  Install multiple measuring devices that can measure acceleration, etc. in civil engineering structures such as bridges and building structures (hereinafter referred to as “buildings” together with civil engineering structures and building structures) A method for measuring the vibration of each part such as a building when it occurs and evaluating the residual seismic performance is known (Patent Document 1). In such a system that measures one physical phenomenon with a plurality of measuring devices, it is necessary that the times of all the measuring devices are appropriately synchronized.

  Known time synchronization methods include a method of synchronizing the absolute time indicated by an NTP (Network Time Protocol) server and the local time of the client terminal, or a method of synchronizing the time of a satellite using GPS (Global Positioning System). It has been. Furthermore, Patent Document 2 discloses a multi-point measurement method that can accurately synchronize the time at which each measurement device measures a physical quantity when measuring the physical quantity with a plurality of measurement devices connected via a wireless network. .

JP 2011-95237 A JP 2007-206848 A

  However, in the conventional time synchronization method, when the measuring device is connected to a low-speed network, the deviation from the absolute time due to network delay cannot be resolved.

  The present invention has been made in view of the above, and provides a time synchronization method capable of relatively synchronizing the local time between sensors without being affected by delay even in a low-speed line. Is the main technical issue.

  Even if the times of a plurality of measuring devices constituting one measuring system need to be synchronized with each other with the accuracy required for analysis, it is not necessary that the times of all measuring devices be the same as the absolute time. In other words, even if the absolute time indicated by the NTP server or the like deviates from the local time of the plurality of measuring devices, there is no problem in analysis as long as the times are accurately synchronized between the plurality of measuring devices.

  A time synchronization method according to the present invention includes a CPU, a storage unit that stores a program and data, a first sensor that measures a physical quantity at a set point, and a first timepiece that includes a first clock, and a CPU. And a storage means for storing the program and data, a second measuring device comprising a second sensor for measuring a physical quantity at the installed point, and a second clock, and the first and second measuring devices are connected to a network. A method of synchronizing time in a measurement system connected via a step of identifying a waveform indicating a temporal transition of a physical quantity in each sensor from the measurement value of the first sensor and the measurement value of the second sensor, respectively. S1 is compared with the characteristic portion of each identified waveform to calculate the local time offset of the second clock relative to the local time of the first clock. Step S2 and step S3 of subtracting the calculated local time difference from the local time of the second clock to synchronize the local time of the second clock with the local time of the first clock. And

  With this configuration, the waveform indicating the temporal transition of the physical quantity in each sensor includes a time lag due to network delay, and the characteristic portions of the waveform are respectively compared to calculate the relative time lag between the clocks. The local time of the second clock can be synchronized with the local time of the first clock while suppressing the influence on the time difference due to the network delay between the sensors. Therefore, even if the network is a low-speed line, the local time between the sensors can be relatively synchronized without being affected by the delay.

  A measurement system according to the present invention is a measurement in which a first measurement device having a first timepiece inside, a second measurement device having a second timepiece inside, and a server are all connected via a network. In the system, the server includes a CPU, a storage unit that stores a program and data, and a time synchronization unit that is executed by the CPU, and the first measurement device stores the CPU, the program, and data. Storage means for storing, and a first sensor for measuring a physical quantity at a point where the first measuring device is installed, wherein the second measuring device is a CPU, and storage means for storing a program and data A second sensor for measuring a physical quantity at a point where the second measuring device is installed, and the time synchronization means includes the measurement value of the first sensor and the second sensor. A waveform at each sensor is identified from the measured value of the sir, and a characteristic portion of each identified waveform is compared to calculate a local time shift of the second clock relative to a local time of the first clock. And a time synchronization function for correcting the local time of the second clock by subtracting the local time difference from the local time of the second clock.

  According to the time synchronization method of the present invention, even when the measuring device is connected to a low-speed line, the local time of one measuring device and the time of another measuring device using the time in the feature portion of the physical quantity captured by each measuring device. In order to synchronize with the local time, correction including the network delay is made to each measuring device, so the local time between each sensor is relatively synchronized with extremely high accuracy without being affected by the network delay. be able to.

The figure which shows the measurement system of 1st Embodiment. The figure which shows the flowchart of the time synchronization method which concerns on the measurement system of 1st Embodiment. It is a figure which shows the data of an Example, (a) is a figure which shows the transfer characteristic immediately after time synchronization, (b) is a figure which shows the transfer characteristic after 20 minutes have passed in time synchronization. The figure which shows the measurement system of 2nd Embodiment. The figure which shows the measurement system of 3rd Embodiment

  Hereinafter, each embodiment of the present invention will be described with reference to the drawings. The same reference numerals are used for the same or similar members, or only the subscripts are displayed differently, and duplicated descriptions are omitted. However, the description of each embodiment is for understanding the technical idea of the present invention. Should not be construed as limited to the description of the embodiments.

(First embodiment)
FIG. 1 is a diagram illustrating a measurement system according to the first embodiment. The measurement system 100 includes a first measurement device 10, a second measurement device 20, and a management server 40, which are connected via a network 30. The measurement system 100 is mainly constructed in a building 1 that is a measurement target. The building 1 may be a civil structure such as a bridge or any other architectural structure. The network 30 includes transmission lines with various communication speeds and transfer rates such as optical lines, satellite lines, Asymmetric Digital Subscriber Lines (ADSL), 3G (3rd Generation) lines, LANs (Local Area Networks), and mixed transmission lines. May be used.

  The first measuring apparatus 10 includes a central processing unit (CPU) 11 that controls the entire apparatus in a housing, storage means 12 such as a flash memory or a hard disk, an A / D converter 13, 3 A first acceleration sensor 14 for detecting a temporal change in dimensional acceleration, a temporary storage means 15 such as a RAM, a first clock 16, a wireless antenna 17, and the like are provided. The first measuring device 10 is installed on the floor of the first floor (ground floor) in the building 1. The first measuring device 10 may be installed on the foundation of the building 1. The wireless antenna 17 is connected to the network 30 wirelessly. Instead of the wireless antenna 17, a network interface may be used to connect to the network 30 via an Internet cable.

  The second measuring apparatus 20 includes a CPU 21 that controls the entire apparatus in a housing, a storage unit 22 such as a flash memory or a hard disk, an A / D converter 23, and a first detector that detects temporal changes in three-dimensional acceleration. 2 acceleration sensors 24, temporary storage means 25 such as RAM, a second clock 26, a wireless antenna 27, and the like. The second measuring device 20 is installed on the second floor in the building 1. The wireless antenna 27 is connected to the network 30 wirelessly. Instead of the wireless antenna 27, a network interface may be used to connect to the network 30 via an Internet cable.

  The CPU 11 (21) is a control device that controls the operation of the entire apparatus. The storage means 12 (22) stores the program and stores the calculation result by the CPU as data. The acceleration sensor 14 (24) is connected to the CPU via the A / D converter 13 (23), and transmits the acceleration measurement value as input data to the CPU while sampling.

  The management server 40 includes a time synchronization unit 41, a CPU (not shown), a storage unit, a network interface, and the like. The management server 40 is installed outside the building 1, but may be installed inside the building 1. The management server 40 mainly serves to manage the first measuring device 10 and the second measuring device 20. The management server 40 is connected to the network 30 via the Internet cable. The time synchronization unit 41 is a program stored in the storage unit and is mainly executed by the CPU of the management server 40. The time synchronization means 41 synchronizes the time of the first timepiece 16 of the first measuring device 10 and the time of the second timepiece 26 of the second measuring device 20 by executing the following steps S1 to S4.

  FIG. 2 is a diagram illustrating a flowchart of the time synchronization method according to the measurement system of the first embodiment. Before implementing this time synchronization method, the granularity that is the frequency of time synchronization is determined in advance. The first timepiece 16 of the first measuring device 10 and the second timepiece 26 of the second measuring device 20 have variations in time accuracy (individual differences) due to processing of crystal as a vibrator. . Further, in these watches, a time lag occurs mainly due to a temperature change. Therefore, it is preferable to perform time synchronization periodically. The granularity varies depending on the scale and configuration of the measurement system 100. For example, when time synchronization is performed with an accuracy of millisecond [msec] units, the time of the clocks 16 and 26 of each measurement device is 10 [msec] for 1 minute. In a measurement system that is shifted, it is preferably set once every minute, and once every about 1 to 10 minutes. In this way, the measurement system 100 can always obtain measurement value data in which the time of the first clock 16 and the time of the second clock 26 are accurately synchronized. This granularity may be managed so that the time is newly counted again when the time is synchronized using the time of the internal clock of the server 40 or a timer.

  When the time is synchronized for the first time (at the start of time synchronization), the time of the first clock 16 and the time of the second clock 26 are synchronized so that there is no time lag, and the impulse response or the like A time T0 that is delayed until the signal reaches the second measuring device 20 from the first measuring device 10 is calculated. You may calculate similarly to the time gap of the 1st time mentioned later and the 2nd time.

  1. First, an impulse response signal is simultaneously transmitted and received between the first measuring device 10 (first acceleration sensor 14 system) and the second measuring device 20 (second acceleration sensor 24 system) to obtain the first The signals are measured by the acceleration sensor 14 and the second acceleration sensor 24, respectively. Note that a vibration may be applied to the building 1 from below to input a waveform, and the signals may be measured by the first acceleration sensor 14 and the second acceleration sensor 24, respectively. The time for transmitting and receiving this signal only needs to be a time that can acquire a data amount of measurement values sufficient to synchronize the time with each of the acceleration sensors 14 and 24, and the measurement value data is 10 [msec] per sample. It may take a few seconds. And the waveform which shows the time transition in each acceleration sensor from the measured value of the 1st acceleration sensor 14 and the measured value of the 2nd acceleration sensor 24 is each identified (step S1). Here, identifying a waveform indicating temporal transition in the present embodiment means expressing a waveform that approximates the measured value from the measured value of the acceleration sensor by a mathematical expression or the like. Alternatively, the measurement values of the acceleration sensors 14 and 24 may be arranged in time series and may be represented by interpolating between the measurement values in order with lines, or may be a data string in which they are arranged in time series.

  It is preferable that the waveform indicating the temporal transition indicates a transfer characteristic between the first acceleration sensor 14 and the second acceleration sensor 24. By using the waveform indicating this transfer characteristic, the time of the first clock 16 and the time of the second clock 26 can be synchronized as long as the building 1 to be measured is linearly deformed. In addition, since this waveform includes various delay and transfer rate conditions between the first acceleration sensor 14 and the second acceleration sensor 24, the waveform is between the first acceleration sensor 14 and the second acceleration sensor 24. It is possible to use a 3G line with a low transfer rate regardless of the installation location.

As one of the methods for identifying the waveform, the waveform is expressed using a finite impulse response (FIR) as shown in the following formula (1), and the LMS (Least Mean Square) is used. There is a method of identifying a waveform by determining a coefficient hm (n) by an algorithm.
Y (n) = Σhm (n) · X (nm) (1)
However, X (n) is an input signal, Y (n) is an output signal, hm (n) is a weighting factor of X (n), and m = 0, 1,..., N−1. Is an integer.
In identifying the waveform, an algorithm such as FLMS (Fast Least Mean Square), RLS (Recursive Least Square) or steepest descent method may be used in addition to the LMS algorithm.

Specifically, the following processes (a) to (c) are performed.
(A) First, the average value of the measurement values of the first acceleration sensor 14 is calculated, and the average value is subtracted from the measurement value of each first acceleration sensor 14 to remove the offset. Similarly, the average value of the measurement values of the second acceleration sensor 24 is subtracted from the measurement value of each second acceleration sensor 24 to remove the offset.
(B) Next, the measured values of the acceleration sensors 14 and 24 from which the offset has been removed in the process (a) are normalized in the range of ± 1.
(C) Next, the measured values of the acceleration sensors 14 and 24 normalized in the process (b) are substituted into the above equation (1), and the coefficient hm (n) is calculated by the LMS so that the following equation (2) is satisfied. ) Respectively.
Y ′ (n) = Σhm (n) · X ′ (nm) (2)
However, X ′ (n) is a measured value of the acceleration sensor 14 normalized in the above (b), and Y ′ (n) is a measured value of the acceleration sensor 24 normalized in the above (b). The number (number of taps) N of the coefficients hm (n) is about 30 to 100.

  By performing the above processes (a) to (c), transfer characteristics from the first measurement device 10 (first acceleration sensor 14 system) to the second measurement device 20 (second acceleration sensor 24 system) are described. Is represented by the above formula (1). Further, by replacing the measured values of X ′ (n) and Y ′ (n) in the process (c) and determining the coefficient hm (n) by LMS, the second measurement device 20 performs the first measurement. A waveform indicating a transfer characteristic to the device 10 is expressed by the above formula (1).

  2. Next, feature portions of the identified waveform are extracted and compared, and a time lag of the second clock 26 with respect to the time of the first clock 16 is calculated (step S2). For example, the first time point corresponding to the feature portion in the measurement value of the first acceleration sensor 14 and the first point identified in step S1 as the peak point that first protrudes upward in time series. First time corresponding to each peak point is calculated by calculating a second time corresponding to the characteristic part in the waveform (the above formula (1)) indicating the transfer characteristic from the measuring device 10 to the second measuring device 20. And the second time are calculated, and the difference is defined as a time lag.

  In step S1, a waveform indicating the transfer characteristic from the first measurement apparatus 10 to the second measurement apparatus 20 and a waveform indicating the transfer characteristic from the second measurement apparatus 20 to the first measurement apparatus 10 are identified. The time difference of the second clock 26 with respect to the time of the first clock 16 may be calculated. As described above, when a waveform showing transfer characteristics is identified by LMS using FIR, the waveform converges with the passage of time. Therefore, the advance or delay of the time of each clock indicates whether or not each waveform has converged when the waveform is identified by LMS using FIR as described above, or the degree to which the amplitude of each waveform is attenuated, etc. Judge with. For example, when each waveform is identified by LMS as described above, the converged waveform is extracted by comparing each waveform, and corresponds to the peak point first convex upward in time series in this converged waveform The time to be set is the time difference between the time of the first clock 16 and the time of the second clock 26.

  Further, in a vibration environment with a pulse input, the time difference of the second clock 26 relative to the time of the first clock 16 can be calculated using the measurement values as they are without identifying the measurement values of the respective acceleration sensors. Good. This vibration environment with pulse input refers to an environment in which vibration occurs on time, such as mechanical vibration or an iron bridge through which a train passes.

  Specifically, the ratio STA1 / LTA1 between the short time average STA1 and the long time average LTA1 is calculated from the measurement value of the first acceleration sensor 14, and similarly, the short time average is calculated from the measurement value of the second acceleration sensor 24. The ratio STA2 / LTA2 between STA2 and long-term average LTA2 is calculated. Then, a time lag is calculated from a comparison between the ratio STA1 / LTA1 and the ratio STA2 / LTA2. This ratio between short-time average and long-term average indicates the self-similarity of the waveform, and it is easy to extract and compare the time of the moment when the self-similarity is lost by inputting the pulse waveform. Can do. Note that when using the ratio of the short-time average to the long-time average, it is necessary to calculate the long-time average. Therefore, the measurement value is acquired in a longer time than when the waveform is identified.

  3. Next, the calculated time lag is subtracted from the time of the second clock 26 to synchronize the time of the second clock 26 with the time of the first clock 16 (step S3). At this time, the time of the second clock 26 is synchronized with the time of the first clock 16 with reference to the time T0 at the start of the synchronization. This time synchronization may be performed each time, but it may be performed later by recording a time lag.

  4). Next, S1 to S3 are repeated with a preset granularity until the measurement of each acceleration sensor is completed (step S4). Through the processes S1 to S4 described above, the time of the first clock 16 and the time of the second clock 26 can be synchronized with high accuracy. Further, if the granularity is set so that the time is synchronized before the time lag of 10 [msec] or more occurs, the synchronization can be continued with the accuracy of millisecond [msec] units.

(Example)
A time synchronization experiment was performed using a PC table having a height of 1.2 [m] divided into two upper and lower stages as a test body. This table was placed on the floor, the first measurement device (IT strong motion meter) and the projector with the power turned on as the heat source were installed in the lower row, and the second measurement device (IT strong motion meter) was installed in the lower row. . Each of the first measurement device and the second measurement device includes a network interface, and each interface is connected to one LAN via a LAN cable. The management server was installed at a location different from the table and connected to the same LAN as the first and second measurement devices via a LAN cable. A part of the wireless LAN was used between the second measuring device and the LAN.

  In the time synchronization experiment, first, the time of the clock of the first measuring device and the time of the second measuring device are synchronized so that there is no time lag, and free vibration from the floor is given to the PC table as an input waveform. And measured by each acceleration sensor. Further, 20 minutes thereafter, free vibration from the floor was applied to the PC table and measured by each acceleration sensor.

  FIG. 3 is a diagram illustrating an embodiment. FIG. 3A is a diagram illustrating a transfer characteristic immediately after time synchronization, and FIG. 3B is a diagram illustrating a transfer characteristic after 20 minutes have elapsed in time synchronization. A waveform D1 in FIG. 3 indicates a transfer characteristic from the first measurement device to the second measurement device. A waveform D2 indicates a transfer characteristic from the second measuring device to the first measuring device. The waveforms D1 and D2 indicating the respective transfer characteristics are both identified by LMS after performing step S1.

  As shown in FIG. 3A, immediately after time synchronization, the waveform D1 converges at 101 samples (1010 [msec]), and the waveform D2 does not converge at 101 samples. From this, it was found that the waveform D1 is delayed from the waveform D2. Further, as shown in FIG. 3B, it was found that after 20 minutes have passed since time synchronization, the waveform D1 has not converged at 101 samples, and the waveform D2 has converged at 101 samples. From this, it was found that the waveform D2 is delayed from the waveform D1.

  Immediately after the time synchronization, the peak point d1 of the waveform D1 is the fifth sample, so the input waveform responds with a delay of 5 samples (50 [msec]) before reaching the second measurement device from the first measurement device. I found out that Since the peak point d2 of the waveform D2 is the 15th sample after 20 minutes have passed since time synchronization, the input waveform is 15 samples (150 [msec] before reaching the first measurement device from the second measurement device. ]) It was found that the response was delayed, and conversely, the time of the first measuring device was found to be advanced by 15 samples (150 [msec]) with respect to the time of the clock of the second measuring device. Therefore, the time synchronization means of the management server calculates 200 [msec] (= 50 [msec] − (− 150 [msec])) from the time of the clock of the second measuring device after 20 minutes have elapsed from the time synchronization. It was found that the time of the clock of the second measuring device is relatively synchronized with the time of the clock of the first measuring device by subtracting.

  In the measurement system 100 according to the first embodiment, even if the first measurement device 10 and the second measurement device 20 are connected to a low-speed line, the measurement system 100 has transfer characteristics from the measurement values measured by the acceleration sensors 14 and 24. Since the waveform is identified and the time in the identified feature portion is used to synchronize the time of the clock 16 of the first measuring device 10 and the time of the clock 26 of the second measuring device 20, the network delay Is corrected to the measurement value measured by each of the acceleration sensors 14 and 24, and the time between the first measurement device 10 and the second measurement device 20 is accurately compared without being affected by the delay of the network. Can be synchronized.

(Second Embodiment)
FIG. 4 is a diagram illustrating a measurement system according to the second embodiment. The main difference between the measurement system of the second embodiment and the first measurement system is that the first measurement device is provided with time synchronization means, and the management server is not required for time synchronization. The measurement system 200 includes a first measurement device 10 a and a second measurement device 20, which are connected via a network 30. The measurement system 200 is mainly constructed in a building 1 that is a measurement target.

  The first measuring device 10a detects a change in the three-dimensional acceleration with time, a CPU 11 for controlling the entire device in a housing, a storage unit 12 such as a flash memory or a hard disk, an A / D converter 13, and the like. 1 acceleration sensor 14, a temporary storage means 15 such as a RAM, a first timepiece 16, a wireless antenna 17, a time synchronization means 18, and the like. The first measuring device 10 is installed on the floor of the first floor (ground floor) in the building 1. The first measuring device 10 may be installed on the foundation of the building 1. The wireless antenna 17 is connected to the network 30 wirelessly. Instead of the wireless antenna 17, a network interface may be used to connect to the network 30 via an Internet cable. The time synchronization unit 18 performs the same processing as the time synchronization unit 41 of the management server 40.

  The measurement system 200 of the second embodiment is not affected by the delay of the network in the same manner as the measurement system 100 of the first embodiment, and the time and the first clock 16 of the first measurement device 10a are the same. The time of the second timepiece 26 of the second measuring device 20 can be relatively synchronized with high accuracy.

(Third embodiment)
FIG. 5 is a diagram illustrating a measurement system according to the third embodiment. The main difference between the measurement system of the third embodiment and the first measurement system is that a plurality of second measurement devices are provided, and each of the second measurement devices is installed on a plurality of floors of a building. The time synchronization means synchronizes the time of the first measurement device and the second measurement device installed on each floor. The measurement system 300 includes a first measurement device 10, a plurality of second measurement devices (20 a, 20 b, 20 c...), And a management server 40 a, which are connected via the network 30. The measurement system 300 is mainly constructed in a multi-storey building 1a that is a measurement target.

  Each of the second measuring devices (20a, 20b, 20c...) Has the same configuration as that of the second measuring device 20, and includes a CPU that controls the entire device in a housing, a flash memory, a hard disk, and the like. A storage unit, an A / D converter, a second acceleration sensor that detects a temporal change in three-dimensional acceleration, a temporary storage unit such as a RAM, a second watch, a wireless antenna, and the like are provided. . The second measuring device is installed on the floor of each floor in the building 1. The wireless antenna is connected to the network 30 wirelessly. You may make it connect to the network 30 via an internet cable using a network interface instead of a wireless antenna.

  The management server 40a includes a time synchronization unit 41a, a CPU (not shown), a storage unit, a network interface, and the like. The management server 40a is installed outside the building 1, but may be installed inside the building 1. The management server 40a mainly serves to manage the first measurement device 10 and the second measurement devices (20a, 20b, 20c,...). The management server 40a is connected to the network 30 via the Internet cable. The time synchronization means 41a is a program stored in the storage means, and is mainly executed by the CPU of the management server 40a. The time synchronization unit 41a performs the same processing as the time synchronization unit 41 of the management server 40, and the time of the first clock 16 of the first measurement device 10 and the second clock of each second measurement device. Synchronize each time. In this time synchronization, the waveform indicating the transfer characteristic from the first measurement device 10 to the second measurement device 20a, the waveform indicating the transfer characteristic from the first measurement device 10 to the second measurement device 20b, the first Waveforms having transfer characteristics corresponding to the number of second measuring devices are identified, such as a waveform showing transfer characteristics from one measuring device 10 to the second measuring device 20c.

  The measurement system 300 according to the third embodiment identifies waveforms indicating transfer characteristics from the first measurement device 10 to the second measurement devices (20a, 20b, 20c,...) Installed on each floor. Then, the time of the clock 16 of the first measuring device 10 and the time of the clock of the second measuring device (20a, 20b, 20c,...) Are synchronized with each other using the time in the identified characteristic part. Therefore, the correction including the network delay is made to the measurement value measured by each acceleration sensor, and the first measurement device 10 and each second measurement device (20a) with high accuracy without being affected by the network delay. , 20b, 20c,...)) Can be relatively synchronized.

  The measurement system 200 of the second embodiment also includes a plurality of second measurement devices (20a, 20b, 20c,...) And is installed on a plurality of floors of the building. The time of the clock 16 of 10a may be synchronized with the time of the clock of the second measuring device (20a, 20b, 20c,...).

1 Building 10 First Measurement Device 11 CPU (Central Processing Unit)
12 storage means 13 AD converter 14 first acceleration sensor 15 temporary storage means (RAM)
16 First clock 17 Radio antenna 18 Time synchronization means 20 Second measuring device 21 CPU (Central Processing Unit)
22 storage means 23 AD converter 24 second acceleration sensor 25 temporary storage means (RAM)
26 Second clock 27 Wireless antenna 30 Network 40, 40a Management server 41, 41a Time synchronization means 100, 200, 300 Measuring system

A time synchronization method according to the present invention includes a CPU, a storage unit that stores a program and data, a first sensor that measures a physical quantity at a set point, and a first timepiece that includes a first clock, and a CPU. And a storage means for storing the program and data, a second measuring device comprising a second sensor for measuring a physical quantity at the installed point, and a second clock, and the first and second measuring devices are connected to a network. A time synchronization method in a measurement system connected via the at least one of the measured values of the first sensor and the measured value of the second sensor, and at least one of which indicates the temporal transition of the physical quantity in each sensor. a step S1 to identify the waveform indicating the transfer characteristic between the first measuring device and the second measuring device respectively, comparing the features portion of each waveform identified Step S2 of calculating the local time difference of the second clock with respect to the local time of the first clock, and subtracting the calculated local time difference from the local time of the second clock. And a step S3 for synchronizing the local time with the local time of the first clock.

A measurement system according to the present invention is a measurement in which a first measurement device having a first timepiece inside, a second measurement device having a second timepiece inside, and a server are all connected via a network. In the system, the server includes a CPU, a storage unit that stores a program and data, and a time synchronization unit that is executed by the CPU, and the first measurement device stores the CPU, the program, and data. Storage means for storing, and a first sensor for measuring a physical quantity at a point where the first measuring device is installed, wherein the second measuring device is a CPU, and storage means for storing a program and data A second sensor for measuring a physical quantity at a point where the second measuring device is installed, and the time synchronization means includes the measurement value of the first sensor and the second sensor. Comparing the features portion of the at least one identified waveform indicating the transfer characteristic between the second sensor and the first sensor, respectively, the waveforms identified from the measured value Sir a waveform at each sensor Then, the local time difference of the second clock is calculated with respect to the local time of the first clock, and the local time difference of the second clock is subtracted from the local time difference of the second clock. It is characterized by having a time synchronization function for correcting the above.

Claims (8)

A first measuring device comprising a CPU, storage means for storing programs and data, a first sensor for measuring a physical quantity at a set point, and a first clock;
A second measuring device comprising a CPU, storage means for storing a program and data, a second sensor for measuring a physical quantity at a set point, and a second clock;
A time synchronization method in a measurement system in which the first and second measurement devices are connected via a network,
Identifying each of the waveforms indicating the temporal transition of the physical quantity in each sensor from the measured value of the first sensor and the measured value of the second sensor;
Calculating a local time offset of the second clock relative to the local time of the first clock by comparing the identified characteristic portions of each waveform;
And subtracting the calculated local time difference from the local time of the second clock to synchronize the local time of the second clock with the local time of the first clock. Method. 2. The time synchronization method according to claim 1, wherein the time synchronization means includes step S4 that periodically repeats steps S1 to S3. 3. The time synchronization method according to claim 2, wherein the time synchronization means includes step Sa <b> 4 that repeats at a cycle of once every 1 to 10 minutes. The step S1 includes a step Sa1 for identifying a conversion equation for converting the measurement value of the first sensor into the measurement value of the second sensor by an algorithm of the least mean square method,
The step S2 includes a first time corresponding to a peak point first convex upward in the measurement value of the first sensor and a peak point first convex upward in the identified conversion formula. A corresponding second time is calculated,
4. The method according to claim 1, further comprising: a step Sa <b> 2 that sets a difference between a first time corresponding to the peak point and a second time corresponding to the peak point as the local time lag. 5. The time synchronization method according to item. For vibration environment with pulse input,
Step S2 includes
A ratio STA1 / LTA1 between the short-term average STA1 and the long-term average LTA1 is calculated from the measurement value of the first sensor,
A ratio STA2 / LTA2 between the short time average STA2 and the long time average LTA2 is calculated from the measurement value of the second sensor,
The time synchronization method according to any one of claims 1 to 3, further comprising a step Sa3 for calculating a time lag from a comparison between the ratio STA1 / LTA1 and the ratio STA2 / LTA2. A measurement system in which a first measurement device having a first clock inside, a second measurement device having a second clock inside, and a server are all connected via a network,
The server includes a CPU, storage means for storing a program and data, and time synchronization means executed by the CPU,
The first measuring device includes a CPU, storage means for storing a program and data, and a first sensor for measuring a physical quantity at a point where the first measuring device is installed,
The second measuring device includes a CPU, storage means for storing a program and data, and a second sensor for measuring a physical quantity at a point where the second measuring device is installed,
The time synchronization means includes
Identifying the waveform of each sensor from the measurement value of the first sensor and the measurement value of the second sensor,
Comparing the identified features of each waveform to calculate the local time offset of the second clock relative to the local time of the first clock;
A measurement system comprising a time synchronization function for correcting the local time of the second clock by subtracting the calculated local time lag from the local time of the second clock. A measurement system in which a first measurement device having a first timepiece inside and a second measurement device having a second timepiece inside are connected via a network,
The first measurement device includes a CPU, storage means for storing a program and data, a first sensor for measuring a physical quantity at a point where the first measurement device is installed, and a time executed by the CPU. Synchronization means,
The second measuring device includes a CPU, storage means for storing a program and data, and a second sensor for measuring a physical quantity at a point where the second measuring device is installed,
The time synchronization means includes
Identifying the waveform of each sensor from the measurement value of the first sensor and the measurement value of the second sensor,
Comparing the identified features of each waveform to calculate the local time offset of the second clock relative to the local time of the first clock;
A measurement system comprising a time synchronization function for correcting the local time of the second clock by subtracting the calculated local time lag from the local time of the second clock. The measurement system according to claim 6 or 7, wherein there are a plurality of the second measurement devices.

Images