A method for collecting seismic signals based on MEMS sensors

Abstract

The present invention discloses a method for collecting earthquake signals based on MEMS sensors, including multiple groups of intensity sensors and a single-chip microcomputer, wherein the output ends of the multiple groups of intensity sensors are connected to the single-chip microcomputer, and multiple intensity sensors are used for parallel averaging to improve the signal-to-noise ratio. The present invention proposes a method for parallel averaging of multiple low-cost, low-resolution digital MEMS accelerometers to improve the resolution of small signals, and a low-cost intensity meter based on MEMS acceleration sensors with high resolution has the characteristics of small size, low cost, low power consumption, integration, and intelligence; it has a built-in real-time processing algorithm for earthquake information, can automatically identify earthquake events and calculate seismic parameters, and is easy to install.

Description

Technical Field

[0001] The present invention relates to a seismic signal acquisition technology, and in particular to a seismic signal acquisition method based on a MEMS sensor. Background Art

[0002] The acquisition of seismic signals is an important part of earthquake detection and seismic research. Traditional seismic signal acquisition equipment mainly relies on large, high-precision mechanical and electronic equipment. However, these devices are large, heavy, and expensive, which is not conducive to portability and field operations. In addition, the traditional seismic signal acquisition equipment's signal processing and storage methods are relatively backward and cannot meet the real-time and efficient data processing and storage requirements.

[0003] Intensive observation is a prominent feature of the earthquake intensity rapid reporting network, so the observation cost of each monitoring point should be reduced as much as possible. Therefore, unlike the observation method of traditional seismic stations, the intensity rapid reporting instrument often does not have a dedicated station observation environment, but installs the instrument on the wall of ordinary residential buildings to reduce the observation cost. In the projects of earthquake intensity rapid reporting and earthquake early warning, it is necessary to deploy earthquake sensors with high density, which are easy to install, compact, low cost and low power consumption, so that the earthquake intensity rapid reporting or early warning can be carried out quickly and accurately when a medium-intensity earthquake comes. On the other hand, in order to improve the utilization efficiency of the instrument, and also for the accuracy and reliability of earthquake early warning, it is necessary to be able to record the arrival time of earthquake P waves with small amplitude with high precision, and to improve the signal-to-noise ratio of small earthquakes and distant earthquakes. In order to improve the resolution requirement as much as possible at a low cost, a method for seismic signal acquisition based on MEMS sensors is needed. Summary of the invention

[0004] The object of the present invention is to provide a method for collecting seismic signals based on MEMS sensors.

[0005] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0006] The invention comprises a plurality of groups of intensity sensors and a single chip microcomputer. The output ends of the plurality of groups of intensity sensors are connected to the single chip microcomputer. The single chip microcomputer collects data by parallel connection of a plurality of intensity sensors for average collection so as to improve the signal-to-noise ratio.

[0007] Furthermore, the calculation formula of the intensity sensor SNR is:

[0008] For a noise-free signal x and N sensors in parallel, assume that each sensor has an uncorrelated noise source n i , N noisy sensor outputs X i for:

[0009] X1=x+n1,X2=x+n2,…XN =x+n N

[0010] Each additional noise source n i All have zero mean value, and the noise power is If K 2 is the signal power, then

[0011]

[0012] SNR dB =10logSNR

[0013] The average value of N sensor outputs is:

[0014]

[0015] The sensor noise power after averaging is:

[0016]

[0017] The average SNR is:

[0018]

[0019] If N sensors have the same SNR dB , then the total SNR is:

[0020]

[0021] That is, after N sensors are averaged in parallel, their dynamic range can be increased by 10logN (dB), or the self-noise can be reduced to that of a single sensor.

[0022] Furthermore, the intensity sensor is a low-cost, low-resolution digital MEMS accelerometer.

[0023] Furthermore, an interrupt signal line is provided on the intensity sensor, and the interrupt signal line INT and the chip select line CS of the intensity sensor are respectively connected to the I / O port of the single-chip microcomputer, so that the single-chip microcomputer can independently respond to and read and write each sensor;

[0024] When an interrupt occurs, the microcontroller responds to each interrupt signal in turn, reads real-time data through the SPI data interface and stores it in a common data buffer pool in the order of reading. After reaching a sampling cycle time, the read data is accumulated and the average value is calculated as the collected data for this time.

[0025] Furthermore, the single chip microcomputer filters the real-time data collected by the intensity sensor every second, performs earthquake event triggering threshold determination and calculates earthquake parameters, wherein the earthquake parameters include PGA parameters and PGV parameters.

[0026] Furthermore, the single chip microcomputer is provided with a GPS timing system for responding to GPS second interrupt signals and a watchdog.

[0027] Furthermore, the single chip microcomputer is provided with a watchdog to improve the robustness of the system.

[0028] The beneficial effects of the present invention are:

[0029] The present invention proposes a method of parallel averaging of multiple low-cost, low-resolution digital MEMS accelerometers to improve the resolution of small signals. The low-cost intensity meter based on MEMS acceleration sensors with relatively high resolution has the characteristics of small size, low cost, low power consumption, integration and intelligence. It has a built-in real-time processing algorithm for earthquake information, can automatically identify earthquake events and calculate seismic parameters, and is easy to install. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A schematic diagram of a flow chart of a seismic signal acquisition system based on a MEMS sensor according to the present invention;

[0031] Figure 2 A schematic diagram of parallel connection of sensors in a seismic signal acquisition system based on MEMS sensors according to the present invention;

[0032] Figure 3 A schematic diagram of data reading of a seismic signal acquisition system based on a MEMS sensor according to the present invention;

[0033] Figure 4 A functional structure diagram of an intensity rapid reporting instrument of a seismic signal acquisition system based on a MEMS sensor according to the present invention;

[0034] Figure 5 A schematic diagram of the flow of earthquake event identification and ground motion parameter calculation of a seismic signal acquisition system based on a MEMS sensor of the present invention;

[0035] Figure 6 This is an overall block diagram of acquisition software for a seismic signal acquisition system based on a MEMS sensor according to the present invention;

[0036] Figure 7 A schematic diagram of sensor noise recording comparison of a seismic signal acquisition system based on a MEMS sensor according to the present invention;

[0037] A is the noise record and spectrum density of a single sensor instrument; B is the noise record and spectrum density of multiple sensors in parallel instrument;

[0038] Figure 8 A schematic diagram of the amplitude-frequency characteristics of a seismic signal acquisition system based on a MEMS sensor according to the present invention;

[0039] Fig. 9 A schematic diagram of input and output linear characteristics of a seismic signal acquisition system based on a MEMS sensor according to the present invention; DETAILED DESCRIPTION

[0040] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0041] like Figure 1 As shown, the resolution of low-cost MEMS digital accelerometers is not high. Although it can record the maximum amplitude of larger earthquakes and calculate the corresponding seismic parameters such as PGA and PGV, the present invention adopts a method of parallel averaging of multiple low-cost and low-resolution digital MEMS accelerometers to improve the resolution of small signals, thereby providing a low-cost intensity meter based on MEMS acceleration sensors with higher resolution.

[0042] In this embodiment, multiple identical sensors are used in parallel to collect and average data, which can improve the signal-to-noise ratio SNR. For a noise-free signal x and N sensors in parallel, assume that each sensor has an uncorrelated noise source n i , N noisy sensor outputs X i for:

[0043] X1=x+n1,X2=x+n2,…X N =x+n N

[0044] Each additional noise source n i All have zero mean value, and the noise power is If K 2 is the signal power, then

[0045]

[0046] SNR dB =10logSNR

[0047] The average value of N sensor outputs is:

[0048]

[0049] The sensor noise power after averaging is:

[0050]

[0051] The average SNR is:

[0052]

[0053] If N sensors have the same SNR dB , then the total SNR is:

[0054]

[0055] That is, after N sensors are averaged in parallel, their dynamic range can be increased by 10logN(dB, or the self-noise can be reduced (attenuated) to that of a single sensor.

[0056] The multi-sensor parallel acquisition method uses the sensor connection form of the SPI data transmission interface. Figure 2 As shown. The data input interface MOSI, data output interface MISO, and serial clock SCLK of each sensor are common and connected to the SPI data interface of the control processor microcontroller to reduce the number of pins occupied by the microcontroller. The interrupt signal line INT and chip select line CS of each sensor are respectively connected to the I / O port of the microcontroller so that the microcontroller can independently respond to and read and write each sensor.

[0057] When an interrupt occurs, the microcontroller responds to each interrupt signal in turn, reads real-time data through the SPI data interface and stores it in a common data buffer pool in the order of reading. After reaching a sampling cycle, the read data is accumulated and the average value is calculated as the collected data for this time. Figure 3 shown.

[0058] The earthquake intensity rapid reporting instrument uses multiple low-cost MEMS digital accelerometers as the seismic sensing unit, a high-performance, low-power 32-bit ARM embedded CPU as the control processor, and uses the Linux operating system to build a low-power data acquisition system. It has a built-in large-capacity SD card local data storage, integrated network communication remote data transmission and control functions, and an embedded earthquake information real-time processing algorithm. It automatically calculates seismic motion parameters and has the characteristics of low cost, low power consumption, miniaturization, and intelligence. It is a small digital strong earthquake instrument that integrates sensors, data acquisition, and communication transmission. The intensity rapid reporting instrument is mainly composed of control management modules, sensor data acquisition modules, data storage management modules, earthquake information processing, GPS time calibration and time services, data services, watchdogs, and power management functions. Figure 4 As shown:

[0059] The interrupt signal line INT and the chip select line CS of the sensor are respectively connected to the I / O port of the single-chip microcomputer, so that the single-chip microcomputer can independently respond to and read and write each sensor;

[0060] When an interrupt occurs, the microcontroller responds to each interrupt signal in turn, reads real-time data through the SPI data interface and stores it in a common data buffer pool in the order of reading. After reaching a sampling cycle time, the read data is accumulated and the average value is calculated as the collected data for this time.

[0061] The control processing system is used to control and manage the work of other functional parts, such as collecting data from sensors, GPS time calibration and data collection timing, local storage management of collected data, communication and remote data services, real-time earthquake event monitoring and earthquake parameter calculation algorithm implementation. In order to take into account the processing power and low power requirements, the central control processor uses the 32-bit low-power ARM9 structure embedded microprocessor S3C2416. S3C2416 is a RISC architecture 32-bit ARM9 high-performance low-power microcontroller processor launched by Samsung Electronics for low-power embedded applications. It uses the ARM920T core, integrates rich on-chip functions, and has a maximum operating frequency of 533MHz; the program running memory uses 64MB SDRAM to meet the needs of system operation and data processing software; 256MB Nand-Flash main memory is used as system and application storage. In order to realize the multi-task processing needs of the acquisition and control system, the stable 3.1.0 kernel Linux embedded operating system is used to manage the application software of each part, and the kernel is tailored according to the system requirements, the driver design of each part of the hardware and the startup parameter configuration are configured. Use multi-thread technology to design Linux applications to achieve parallel processing of multiple tasks in each functional module.

[0062] The sensor data acquisition part collects seismic acceleration signals in real time. In order to improve the dynamic range and reduce the cost of the instrument, multiple low-cost MEMS digital accelerometers are used for parallel acquisition, and the output of each sensor is accumulated and averaged to reduce self-noise.

[0063] The data storage system stores the collected data in the SD card memory. After a strong earthquake occurs, the quick-reporting instrument not only needs to quickly obtain the seismic parameters and transmit them to the monitoring center to guide emergency rescue, but also the earthquake waveform data and the background noise before the earthquake will become precious earthquake analysis data in the future. Large earthquakes often bring great damage, and the surge in communication volume may even cause temporary network congestion. The large number of sensors on the network need to transmit real-time collected data or seismic waveform data to the monitoring center at the same time, which will aggravate the network congestion. The earthquake intensity quick-reporting instrument is equipped with a large-capacity local database storage system, which can download data as needed after the earthquake. The instrument mainly stores three types of data files locally: real-time collected data, one storage file per hour; the second is the event file; the third is the parameter file, which corresponds to the seismic parameters calculated for each event, which is convenient for quick query of historical events in the future. Considering that the severely damaged earthquake areas after a major earthquake require a long recovery time, such as one or two months to collect or download the recorded data, the intensity rapid reporting instrument stores the collected data on the instrument's built-in 32G large-capacity SD card and sets the data file retention period (up to 4 months at a sampling rate of 200SPS). Files that exceed the storage period are automatically deleted by the program, thus ensuring that the storage space can be recycled.

[0064] The seismic information processing system filters the real-time data collected every second, determines the threshold of earthquake event triggering, and calculates seismic parameters such as PGA and PGV. The PGA and PGV calculation and processing procedures are as follows: Figure 5 As shown in the figure, the GPS time correction system responds to the GPS second interrupt signal, reads the GPS information through the serial port and parses the time, coordinates, etc. When the GPS signal is valid, the instrument system time is corrected with the read GPS time to ensure the time accuracy of sensor data collection. The remote data service system is connected through the network (wired or wireless) to achieve status monitoring, real-time waveform monitoring, FTP data file download, instrument parameter setting, etc. of the intensity rapid reporting instrument through remote link. The watchdog is used to improve the robustness of the system operation, monitor the program operation in real time, and the program can automatically restart in unexpected situations.

[0065] The earthquake intensity rapid reporting instrument adopts the embedded Linux operating system with kernel trimming, built-in earthquake information processing algorithm, real-time acquisition of ground vibration acceleration signal, automatic calculation of ground vibration parameters, and is an intelligent strong earthquake monitoring instrument. Since the Linux operating system is used as the system software platform, the instrument can also run third-party software, add some professional extension functions and special algorithms, making it suitable for a variety of applications.

[0066] The acquisition software adopts a top-down hierarchical modular design, such as Figure 6As shown in the figure, some functions of each module are relatively independent, and data exchange is realized by providing data interface or sharing data buffer pool. According to the characteristics of signal processing, the entire acquisition software is divided into application layer and driver layer. The application layer uses multi-threading technology to realize multi-task processing, mainly including real-time data acquisition module, local database storage management module, seismic information processing algorithm implementation module and remote data service module. The real-time data acquisition module receives the asynchronous notification signal of the underlying layer and wakes up the work of this module. It calls the standard file function read() to read the real-time acquisition data from the driver storage space and store it in the application data buffer pool. The data buffer pool is a storage area for circular reading and writing in the memory space. The seismic information processing algorithm implementation module performs seismic information processing and program implementation such as seismic event discrimination and seismic motion parameter calculation on the real-time data stored in the data buffer pool. The local database storage management module creates the corresponding file directory to store and manage the real-time data, event waveform data, and parameter data by directory, and deletes the data that exceeds the time limit so that the data storage space can be recycled. As a data remote server, the remote data service module provides the client with real-time continuous data transmission, remote monitoring of instrument status, real-time transmission of seismic motion parameters, query and download management of remote FTP historical data files, etc.

[0067] The driver layer program implements the configuration of sensors, the reading of sensor data, and the implementation of file operation functions for reading and writing kernel space data, etc. It mainly includes the driver of the hardware data read and write bus interface, the hardware interrupt driver, the asynchronous notification used by the kernel space driver to send read and write request signals to the user space application, and the implementation of file operation functions such as read() and write(). The layered design and modularization method improves the stability and reliability of the software system operation, and also facilitates the upgrade and maintenance management of the software.

[0068] In order to compare the effect of multiple sensors in parallel, the instrument with a single sensor and the instrument with 8 sensors in parallel were placed close to each other and in the same direction for simultaneous observation. Figure 7 A and Figure 7 B. It can be seen that the RMS noise of the instrument with 8 sensors connected in parallel in the frequency band of 0.1Hz to 20Hz is about 0.096cm / s 2 Reduced to about 0.034cm / s 2 .

[0069] The frequency response and linearity of the earthquake intensity rapid reporting instrument were tested on the vibration table. During the test, the sensing direction of the earthquake intensity rapid reporting instrument was consistent with the movement direction of the vibration table. The vibration table was input with sinusoidal test signals of different amplitudes and frequencies, and the amplitude-frequency response and linearity were obtained as shown in Tables 1, 2 and Figure 8 , Fig. 9As shown. Since the low-frequency characteristics of the accelerometer require the use of an expensive long-stroke linear vibration table to achieve a good signal-to-noise ratio, the instrument amplitude-frequency characteristics only test frequencies above 0.5Hz. It can be seen that after multiple sensors are averaged in parallel, the amplitude-frequency response of the input and output is still flat ( Figure 8 ), the linear characteristics can also be guaranteed ( Fig. 9 ).

[0070] Table 1 Amplitude-frequency characteristic test (sampling rate: 200SPS)

[0071]

[0072] Table 2 Linear characteristics test (test frequency: 5Hz)

[0073]

[0074] The performance indicators of the earthquake intensity rapid reporting instrument adopted in this technical solution and the standard requirements of the China Earthquake Administration are compared in Table 3. It can be seen that the indicators of the earthquake intensity rapid reporting instrument developed by using low-cost MEMS accelerometers through the correlation averaging method of multi-sensor parallel connection can fully meet the use requirements.

[0075] Table 3 Indicators of the intensity rapid reporting instrument with parallel sensors and standards of the China Earthquake Administration

[0076]

[0077]

[0078] The present invention proposes a method of parallel averaging (correlation averaging) of multiple low-cost and low-resolution digital MEMS accelerometers to improve the resolution of small signals, and develops a low-cost MEMS accelerometer with high resolution. The instrument adopts ARM+Linux embedded computer technology, built-in SD card data storage, and can cyclically store about 12 months of collected data (sampling rate 50SPS) without supervision; it has the characteristics of small size, low cost, low power consumption, integration, and intelligence; its built-in real-time processing algorithm for earthquake information can automatically identify earthquake events and calculate seismic parameters; it is easy to install in practical applications. The development results show that by adopting the parallel averaging acquisition method of 8 low-cost sensors, the instrument noise is reduced from 0.096cm / s2 to 0.034cm / s2 in the bandwidth of 0.1Hz~20Hz, and the dynamic range reaches more than 90dB. In actual observations, the Ms2.2 near earthquake and the Ms5.1 200km far earthquake recorded both have a high signal-to-noise ratio. In addition, the instrument has a flat amplitude-frequency response and good linear characteristics, and can meet the requirements for rapid intensity reporting put forward by the China Earthquake Networks Center.

[0079] The technical solution of the present invention is not limited to the above-mentioned specific embodiments. All technical variations made according to the technical solution of the present invention fall within the protection scope of the present invention.

Claims

1. A method for collecting seismic signals based on a MEMS sensor, comprising a single chip microcomputer, wherein the single chip microcomputer is connected to a monitoring center via a wireless communication module, wherein: It also includes a plurality of intensity sensors, the output ends of the plurality of intensity sensors are connected to the single chip microcomputer, and the single chip microcomputer collects data by parallel connection of the plurality of intensity sensors for average collection to improve the signal-to-noise ratio; The calculation formula of the intensity sensor SNR is: For a noise-free signal x and N sensors in parallel, assume that each sensor has an uncorrelated noise source n i , N noisy sensor outputs X i for: X1=x+n1,X2=x+n2,···X N =x+n N Each additional noise source n i All have zero mean value, and the noise power is If K 2 is the signal power, then SNR dB =10logSNR The average value of N sensor outputs is: The sensor noise power after averaging is: The average SNR is: If N sensors have the same SNR dB , the total SNR is: That is, after N sensors are averaged in parallel, their dynamic range can be increased by 10logN (dB), or the self-noise can be reduced to that of a single sensor.

2. A method for collecting seismic signals based on MEMS sensors according to claim 1, characterized in that: The intensity sensor is a low-cost, low-resolution digital MEMS accelerometer.

3. The method for collecting seismic signals based on MEMS sensors according to claim 1, characterized in that: The intensity sensor is provided with an interrupt signal line, and the interrupt signal line INT and the chip select line CS of the intensity sensor are respectively connected to the I / O port of the single-chip microcomputer, so that the single-chip microcomputer can independently respond to and read and write each sensor; When an interrupt occurs, the single chip microcomputer responds to each interrupt signal in turn, reads real-time data through the SPI data interface and stores it in a common data buffer pool in the order of reading. After reaching a sampling cycle time, the read data is accumulated and the average value is calculated as the collected data for this time.

4. The method for collecting seismic signals based on MEMS sensors according to claim 1, characterized in that: The single chip microcomputer filters the real-time data collected by the intensity sensor every second, performs earthquake event triggering threshold judgment and calculates earthquake parameters, and the earthquake parameters include PGA parameters and PGV parameters.

5. The method for collecting seismic signals based on MEMS sensors according to claim 1, characterized in that: The single chip microcomputer is provided with a GPS timing system for responding to GPS second interruption signals and a watchdog.

6. The method for collecting seismic signals based on MEMS sensors according to claim 1, characterized in that: The single chip microcomputer is provided with a watchdog to improve the system robustness.

Images