JP7386490B1 - pressure measuring device - Google Patents

Abstract

An object of the present invention is to ensure high accuracy by configuring a plurality of inexpensive pressure sensors, and also to enable pressure measurement with guaranteed reliability even when measuring at multiple points. A plurality of pressure sensors 4-1, 4-2, 4-3, ... 4-n are arranged in the device, and measurement data collected from the plurality of pressure sensors at predetermined synchronized timing is In a pressure measuring device that obtains measured values through averaging processing 7, the range in which the plurality of pressure sensors are arranged is determined according to the pressure resolution of the waves to be measured. [Selection diagram] Figure 3

Description

The present invention relates to a pressure measuring device composed of a plurality of pressure sensors.

By measuring infrasound (infrasound), which is sound with a frequency lower than the human audible frequency (approximately 20Hz to 20KHz), it can be used to predict natural disasters such as earthquakes and volcanic activity, and to evaluate the structure of buildings. Research such as resource exploration is underway.
As a pressure gauge capable of measuring these infrasounds, for example, a high-precision pressure gauge using a crystal resonator manufactured by Paro Scientific is known. Although this pressure gauge can achieve a resolution of 1/1 billion and an accuracy of 0.01% FS, it is an expensive measuring instrument.

By the way, capacitance sensors manufactured using MEMS (Micro Electro Mechanical Systems) technology are inexpensive, compact, power-saving devices that are used in a variety of fields. However, it is not practical to use it alone to measure infrasound as described above because the SN ratio and dynamic range are insufficient.
As a means to solve such problems, for example, a technique using a plurality of MEMS acceleration sensors in a seismometer has been disclosed (for example, Patent Document 1).

Patent Document 1 aims to reduce noise caused by thermal noise included in the output of a MEMS acceleration sensor and reduce quantization errors caused by an A/D converter. To reduce noise due to thermal noise, each analog output signal from the sensor is subjected to addition processing to reduce the noise component, and noise in an unnecessary frequency range is removed through a bandpass filter. To reduce quantization errors, after oversampling, an IIR (infinite impulse response) low-pass filter is used to remove noise in real time.
In this way, in order to reduce noise, it is preferable to increase the number of sensors (N) because the noise component is proportional to the reciprocal of the square root of the number of sensors (N) through addition processing. When the number of sensors increases, the mounting area increases, and even if the measurement timing is the same, the waves detected by sensors separated by the same distance will differ, causing a problem of lowering the resolution of the measuring instrument.

Infrasound is also used in scientific research and applied fields, taking advantage of its ability to propagate long distances through the atmosphere, and is used to monitor earthquakes and volcanic eruptions, observe meteorological phenomena, and observe tsunamis. There is. Because the frequency of sound waves is low, it is also used for structural evaluation of buildings and underground facilities, and exploration of underground resources.
In such a case, when it is necessary to conduct observations at multiple points, it is required to perform observations at accurate and synchronized times at each observation point.

Japanese Patent Application Publication No. 2009-31032

The problem we are trying to solve is that although measuring infrasound is important for observing natural phenomena, high-performance pressure measurement equipment is expensive, especially when observing at multiple points. The point is that if it is necessary, a large amount of cost will be involved.

The present invention uses a plurality of inexpensive, miniaturized, and power-saving capacitance sensors manufactured using MEMS (Micro Electro Mechanical Systems) technology as pressure sensors, and also uses these sensors in accordance with the required resolution. The most important feature is that the area to be placed on the board can be determined. Furthermore, the system is configured to ensure the reliability of synchronized measurement data even when used at multiple locations.

The pressure measuring device of the present invention is configured to use a plurality of inexpensive capacitance sensors to obtain the required resolution. Additionally, by using GPS (Global Positioning System) to synchronize multiple sensors and adding time stamps to the measurement data, the reliability of measurement data synchronized at multiple points can be guaranteed.

FIG. 3 is a diagram illustrating the arrangement of pressure sensors on a substrate of a pressure measuring device according to the present invention. FIG. 3 is a diagram illustrating the distance between pressure sensors corresponding to the pressure resolution of waves to be measured according to the present invention. 1 is a conceptual diagram showing the overall configuration of a pressure measuring device according to the present invention. It is a figure explaining the flow of processing of a pressure measurement device concerning the present invention.

We have achieved a pressure measurement device that can obtain the required resolution with an inexpensive configuration and can acquire measurement data synchronized at multiple points.
Embodiments according to the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram illustrating the arrangement of pressure sensors on a substrate of a pressure measuring device according to the present invention, in which 1 is a substrate on which a circuit pattern is printed, 2-1, . . . 2-8 are substrates. This is a capacitance sensor as a pressure sensor manufactured using MEMS technology located at 1.
Each capacitance sensor has the same performance, and is arranged in a horizontal row of four at equal intervals and in two vertical rows of eight in total. The distance X between capacitance sensors is the largest between sensor 2-1 and sensor 2-8 or between sensor 2-4 and sensor 2-5.
In this state, when each capacitance sensor measures a constant frequency, the measured values will differ depending on the distance between the sensors even if each sensor starts measurement at the same timing.
That is, originally, the values measured by each sensor should be the same, but the amount of deviation is reflected in the resolution. The largest deviation occurs between the sensor outputs of the sensor 2-1 and the sensor 2-8, and the sensor output of the sensor 2-4 and the sensor 2-5, which is the limit value of the resolution.

FIG. 2 is a diagram illustrating the distance between pressure sensors corresponding to the pressure resolution of waves to be measured, according to the present invention. For example, when wave 3 to be measured has amplitude A = 100 Pa, frequency = 5 Hz, and is measured at a pressure resolution (dA) of 0.1 Pa, and the wave propagation speed is 340 m/s, the wavelength λ is 340 ÷ 5 = It is 68m. Here, when the wave between λ/2 is linearly approximated, since there is a relationship of A:dA=λ/2:X, X=34 mm.
That is, the maximum wavelength shift at this pressure resolution of 0.1 Pa is 34 mm, and the distance between the pressure sensors must be within 34 mm.
In this way, the range (distance) in which multiple pressure sensors are placed can be determined, and the range X in which the sensors are placed can be narrowed so as to shorten the section where the wave motion is linearly approximated. By increasing the accuracy, the wave motion to be measured can be more closely approximated, and the accuracy (pressure resolution) can be improved.

FIG. 3 is a conceptual diagram showing the overall configuration of the pressure measuring device according to the present invention, and 4-1, 4-2, 4-3, ... 4-n are n pieces manufactured by MEMS technology. Capacitance sensors 5-1, 5-2, 5-3, . . . 5-n are n A/D converters that convert the output signals of n capacitance sensors into digital signals, 6 -1, 6-2, 6-3, . . . 6-n are n signal processing units that perform oversampling, etc. of digital signals output from n A/D converters to be described later, and 7 is An arithmetic unit that averages the data obtained from n signal processing units; 8 is an IIR low-pass filter (LPF) that cuts the high frequency portion of the averaged signal; 9 is a synchronization circuit 10, which will be described later. A GPS (Global Positioning System) 10 sends signals to the camera and adds time stamps to measured values, and n A/D converters 6-1, 6-2, 6-3, . . . . . 6-n is activated simultaneously to convert the output signal into a time-series digital signal, and 11 is an output section that outputs a measured value at a predetermined timing.

As explained in Fig. 1, the range in which the n capacitance sensors can be arranged is determined so that a predetermined accuracy (pressure resolution) can be obtained, and within that range, there are two-dimensional or three-dimensional It is placed on a substrate that has a similar shape.
Pressure signals measured from these sensors are converted into digital signals by n A/D converters. The timing at which the conversion is started is that the synchronization circuit 10 sends a synchronization signal (every 0.1ms in this embodiment) to each A/D converter based on the pulse 1PPS outputted from the GPS 9 every second. will be started.

The signals sampled at a predetermined timing based on the synchronization signal from the synchronization circuit 10 and output from the n A/D converters are each input to a signal processing section 6-1 or the like. In the signal processing units 6-1, . . . , 6-n, the temperature is measured at the same time and the pressure signal, which is the input signal, is corrected, since capacitance sensors using MEMS have large noise depending on thermal noise. are doing.
That is, the pressure signal and temperature measurement value sampled at a predetermined timing are oversampled, and the average of the pressure signal corrected by the temperature measurement value is calculated, and the unique noise components included in the individual sensor outputs are calculated. It will decrease in proportion to the reciprocal of the square root of the number of oversamplings.
Here, the condition is that each sensor has the same performance, but there is a deviation in the frequency response performance of each sensor, and when sampling is performed at a period within that frequency band, Sensors with long time constants and low response performance tend to always measure low displacement amounts. For this purpose, it is necessary to reduce the individual sensor output to a frequency band that is not affected by the measurement period by oversampling processing.

Then, the signals output from each of the signal processing units 6-1, 6-2, 6-3, ... 6-n are averaged by the calculation unit 7, and the signals are averaged between the n sensor outputs. Noise components are reduced.
In other words, by averaging the total number of capacitive sensors (n), the noise component between each sensor output will be reduced in proportion to the reciprocal of the square root of n, and as a result, the dynamic range of measurement will be reduced. It will be expanded.

The output signal averaged by the arithmetic unit 7 passes through a low-pass filter 8, where unnecessary noise components in a high frequency range are removed. The low-pass filter 8 is an IIR filter and is continuously processed in real time.
The cutoff frequency of the low-pass filter 8 can be set as a parameter in consideration of the measurement band required depending on the intended use.

FIG. 4 is a diagram explaining the processing flow of the pressure measuring device according to the present invention, and shows the capacitance sensors 4-1, 4-2, 4-3, ... 4-n and their built-in components. Temperature measurement (S10) and pressure measurement (S20) are performed using a temperature sensor (not shown), and each measurement data is oversampled (S30). Then, the digital signals output from the n A/D converters 5-1, 5-2, 5-3, ... 5-n are processed by the signal processing sections 6-1, 6-2, 6-3. , . . . 6-n, pressure data is obtained in which the temperature-dependent noise component of each capacitance sensor is corrected in accordance with the measurement data obtained by temperature measurement (S40).

The corrected pressure data (measured data) from each sensor is averaged by the calculation unit 7 (S50), and outputted to the output unit 11 through the low-pass filter 8 (S60).
Here, according to the desired cycle to be output to the outside, temperature measurement (S10) is performed from the measurement data converted into a digital signal based on the sample cycle from the synchronous circuit 10 until that cycle is reached (N). , pressure measurement (S20), and the like are sequentially repeated (S70), and the measured values are stored in the output unit 11 at each timing.

When a predetermined period for outputting to the outside has been reached (Y), the measured values stored in the output unit 11 up to that point are averaged (S80). By doing this, it is possible to obtain an effect equivalent to actually oversampling the desired period to be output and taking an arithmetic average. This allows highly accurate pressure measurements to be obtained. Next, the GPS 9 attaches a time stamp at that time (S90), and outputs the pressure measurement value to the outside from the output unit 11 in a predetermined format (S100).

In this way, attaching a time stamp to the measured value is necessary, for example, when observing pressure at multiple distant points, it is necessary to know the absolute time and synchronization timing of the measured value at each point. This is an effective means of identifying the location of the source, such as the epicenter.
In other words, the pressure measurement device of the present invention uses GPS to output measurement values with time stamps that are synchronized with the world standard time, making it possible to perform time-series synchronized measurements even in a multi-point observation network. be able to. For example, the location and scale of the source can be estimated by obtaining the time of occurrence and amount of attenuation from a large number of infrasound measurements taken several kilometers apart.

In addition, instead of the processing method that performs the arithmetic averaging described above, two sensors are provided as redundant sensors among the N sensors, and the initial default value is set to, for example, 1 atm, and the measured data from each sensor is Find the maximum value with the largest change and the minimum value with the smallest change. Then, the N measurement data excluding the maximum value and minimum value measurement data are added and averaged to obtain the measurement value at this timing. Thereafter, in the same way, the measured value obtained by averaging at the previous timing is compared with the measured data of each (N+2) sensor, the maximum value and the minimum value are determined, and these measured data are The average of the N measurement data is calculated to obtain the measurement value at this timing. Note that if there are multiple same maximum or minimum values, it is considered that the measured data does not contain large noise components (close to the actual measured value), so the average value is calculated using the measured data of N sensors determined in advance. to ask for.

Based on the previous measurement value, the maximum and minimum measurement data are the largest deviations from the current measurement value (average value), and are thought to contain the largest noise component, so the average is calculated in this way. By doing so, the measurement data that includes the largest noise component is removed and averaged, making it much closer to the actual measured value compared to the above processing method that adds and averages the measurement data from all sensors. , the accuracy of the measuring device can be increased.
Note that the present invention is not limited to the embodiments described above, and it goes without saying that various modifications and applications can be made within the scope of the gist of the present invention.

It has become possible to realize an inexpensive and highly accurate pressure measurement device, and it has become possible to obtain useful information in fields where infrasound measurement is essential, such as the natural science field, civil engineering and construction field, disaster prevention field, and resource exploration field. can.

1 Board 2-1 Capacitance sensor 3 Wave 4-1 Capacitance sensor 5-1 A/D converter
6-1 Signal processing section 7 Arithmetic section 8 Low-pass filter 9 GPS
10 Synchronous circuit 11 Output section

Claims (7)

A pressure measuring device in which a plurality of pressure sensors are arranged in the device, and measurement data collected from the plurality of pressure sensors at predetermined synchronized timing is averaged to obtain a measured value,
A pressure measurement device , characterized in that a range in which a plurality of capacitance sensors manufactured by MEMS technology are arranged as the plurality of pressure sensors is determined according to a pressure resolution of waves to be measured . The pressure measuring device according to claim 1, wherein the averaged measurement data is configured to have noise removed through a low-pass filter having a predetermined cutoff frequency. . A synchronization signal is generated at the predetermined synchronization timing based on GPS and sent to the plurality of capacitance sensors, and a time stamp is added to the measurement value and outputted. Item 1. Pressure measuring device according to item 1. Claim characterized in that the averaging process is performed after each of the measurement data obtained from the plurality of capacitance sensors is oversampled and averaged during the predetermined synchronization timing. Item 1. Pressure measuring device according to item 1. The pressure according to claim 1, wherein the measured value is configured such that the measured value is obtained by averaging the measured values obtained until reaching a predetermined period of outputting to the outside. measuring device. The pressure measuring device according to claim 1, wherein the wave to be measured is an infrasound wave . The averaging process is characterized in that the averaging process is performed by excluding measurement data that has the largest and smallest difference from the measurement value obtained by the previous averaging process from the measurement data. Item 1. Pressure measuring device according to item 1.