A dual accelerometer strong vibration identification method and monitoring instrument

Abstract

The present invention relates to the field of vibration detection technology, and specifically to a dual-accelerometer strong vibration identification method and monitor. The method comprises: S1 collecting signals from a quartz accelerometer and a MEMS accelerometer; S2 processing the collected vibration signals; S3 identifying strong vibration signals; S4 a strong vibration monitor marking data identified as strong vibration signals, wherein the marking content includes the number of strong vibrations and the maximum amplitude, and storing the strong vibration signal data exceeding a specific threshold in a separate file; S5 integrating and correcting the acceleration data in the marked strong vibration signal data file to obtain parameters generated by the vibration signal as indicators for monitoring the structural health of a building. The present invention has the advantages of accurately distinguishing strong vibration signals from environmental noise signals, effectively reducing the false alarm rate in the vibration signal identification process, enhancing the reliability of the system, and improving the accuracy of strong vibration monitoring of buildings.

Description

Technical Field

[0001] The present invention relates to the technical field of vibration detection, in particular to a dual-accelerometer strong vibration identification method and a monitoring instrument. Background Art

[0002] Building vibration monitoring plays a key role in ensuring public safety, extending building lifespans, optimizing maintenance strategies, improving the quality of living and working environments, and supporting scientific research. Existing technologies primarily employ two approaches to monitoring building vibration: real-time monitoring using vibration sensors, and indirect capture and analysis using image processing techniques. Vibration sensor monitoring is a common approach that directly collects vibration data by installing vibration sensors within the building structure.

[0003] In existing technologies, commonly used accelerometers include quartz accelerometers and MEMS accelerometers (also known as microelectromechanical system accelerometers). Quartz accelerometers are widely used in high-precision vibration monitoring due to their excellent linearity, low noise, and long-term stability. However, they have a long response time, which can lead to some lag in high-frequency dynamic detection. In contrast, MEMS accelerometers are widely used in strong vibration monitoring due to their miniaturization, low cost, and fast response. They can quickly capture strong vibration signals, but their measurement accuracy is low for low-frequency vibration detection and they are easily affected by environmental noise, which can reduce the accuracy and reliability of the monitoring system. In addition, the monitoring results of a single accelerometer in a building are easily affected by the environment near the installation point, for example, human movement can be mistaken for strong vibration. Distributed accelerometers installed at multiple locations face challenges in time synchronization and rapid data integration and analysis.

[0004] Therefore, there is an urgent need to design a monitoring method and detection device that fully utilizes the respective advantages of multiple accelerometers to achieve more comprehensive vibration monitoring, and solve the problems of how to effectively integrate data in the detection system of dual accelerometers, judge the authenticity of vibration in real time, and reduce false alarms. Summary of the Invention

[0005] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a dual-accelerometer strong vibration discrimination method and monitor to combine a quartz accelerometer and a MEMS accelerometer, give full play to the high precision of the quartz accelerometer and the fast response of the MEMS accelerometer, and overcome the interference and error problems of the two accelerometers when synchronously collecting data, improve the accuracy of strong vibration discrimination through consistency analysis, and reduce false alarms.

[0006] In order to achieve the above-mentioned purpose, a dual-accelerometer strong vibration discrimination method is designed, comprising the following steps: step S1. collecting vibration signals, including two accelerometer signals of a quartz accelerometer and a MEMS accelerometer; step S2. performing acceleration restoration, filtering, and de-meaning processing on the collected vibration signals; step S3. performing strong vibration signal discrimination, comprising: step S3.1. processing the vibration signal through a vibration wave picking algorithm; if both accelerometer signals in the vibration signal exceed the threshold, preliminarily judging the vibration signal as a strong vibration signal, and proceeding to step S4; if both do not exceed the threshold, proceeding to step S3.2; step S3.2. comparing the waveform consistency index of the vertical axis of the two accelerometer signals with the threshold: if both consistency indexes exceed the threshold, it is judged to be a strong vibration signal, and proceeding to step S4; if both consistency indexes do not exceed the threshold, it is judged to be non-strong vibration. signal, and enter step S4; if the two consistency indicators only partially exceed the threshold, the three-axis data of the two accelerometer signals are synthesized, and the process goes to step S3.3; step S3.3. The consistency indicator of the waveform of the vertical axis of the synthesized data is judged again. If it exceeds the threshold, it is judged as a strong vibration signal. If it does not exceed the threshold, it is judged as a non-strong vibration signal, and the process goes to step S4; step S4. The strong vibration monitor marks the data judged as strong vibration signals, and the content of the mark includes the number of strong vibrations and the maximum amplitude, and the strong vibration signal data exceeding a specific threshold is formed into a separate file for storage; step S5. The acceleration data in the marked strong vibration signal data file is synthesized, de-averaged, de-linearized, and then integrated and corrected to obtain the parameters generated by the vibration signal as the building structure health monitoring index, and the parameters include: acceleration peak, motion speed peak, displacement peak and duration.

[0007] Preferably, the method of the present invention further comprises: the specific method of step S3.1 is as follows: by calculating the waveform value R of the vertical axis of the quartz accelerometer signal and the MEMS accelerometer signal respectively a 、R m ; Set the trigger threshold R th and the ending threshold R e ; The threshold exceeding flag bits corresponding to the quartz accelerometer and MEMS accelerometer are F as 、F ms , the exceeding threshold flag F as 、F ms The waveform value R perpendicular to the axis a 、R m and trigger threshold R th The logical relationship is as follows:

[0008]

[0009] There is a common exceeding threshold flag Fs , the common exceeds the threshold flag F s and the threshold flag F as 、F ms The logical relationship is as follows:

[0010] F s =F as and F ms ;

[0011] and means F as and F ms Perform AND logic operation, when F s When the value of is 1 for k consecutive times, it is determined that both accelerometer signals exceed the trigger threshold, k is any set positive integer, and the time t obtained by the global navigation satellite system module is recorded at the same time s ; Also includes setting the end threshold flag F of the quartz accelerometer signal and the MEMS accelerometer signal ae 、F me , end threshold flag F ae 、F me The waveform value R perpendicular to the axis a 、R m and the ending threshold R e The logical relationship is as follows:

[0012]

[0013] There is a common end threshold flag F e , the common end threshold flag F e and the threshold flag F ae 、F me The logical relationship is as follows:

[0014] F e =F ae and F me ;

[0015] and means F ae and F me Perform AND logic operation at t s Monitor F after the start of the time e The value of F e When the values ​​of are all 1, the trigger is judged to be over, and the time t obtained by the global navigation satellite system module is recorded. e , k is any set positive integer.

[0016] Preferably, the method of the present invention further comprises: the specific method of step S3.2 is as follows: step S3.2.1. intercepting two accelerometer signals at time t s to te The waveform data of the vertical axis between the two is smoothed; step S3.2.2. All peak values ​​are calculated; step S3.2.3. The zero crossing point between the first peak value and the second peak value is calculated, and this zero crossing point is used as the first zero crossing point, and multiple zero crossing points and zero crossing point sequences are calculated in sequence; step S3.2.4. The cosine similarity of the corresponding data of the two accelerometer signals is calculated.

[0017] Preferably, the method of the present invention also includes: the method for calculating all peaks described in step S3.2.2 is as follows: after calculating all peaks, the number of peaks N1 is obtained; the waveform data is smoothed again to calculate the number of peaks N2. If N1≠N2, the latter peak calculation is considered valid, otherwise the previous one is valid, and the waveform data is continued to be smoothed and the number of peaks is calculated until the two peak numbers are equal, at which time the previous calculation is considered valid; the peak calculation and identification are completed, and the peak data and the peak corresponding time data are obtained.

[0018] Preferably, the method of the present invention further comprises: the zero-crossing point calculation method in step S3.2.3 is as follows, comprising: step S3.2.3.1. calculating all zero-crossing point time sequences; step S3.2.3.2. obtaining an average time difference Δt based on the peak corresponding time data obtained in all peak calculations, wherein the time data is a peak time sequence [t1, t2, ... t i ], i∈[1, N], where N is the number of peaks obtained by calculation, and the average time difference Δt is calculated as follows:

[0019]

[0020] The average time difference Δt is approximately equal to the peak time difference between adjacent peaks of a vibration waveform, that is, approximately equal to half of a vibration waveform period; Step S3.2.3.3. Set the threshold ε z , the calculation formula is:

[0021]

[0022] Where k is an arbitrary positive integer. This threshold is used to correct the zero-crossing time sequence of the two accelerometers. The correction calculation method is as follows:

[0023]

[0024] Through the above steps, the peak data, peak time data and zero-crossing time series data of the quartz accelerometer and MEMS accelerometer are obtained. xzi, where x can be substituted into a or m, a represents the quartz accelerometer, m represents the MEMS accelerometer, i∈[1,m], and m represents the number of zero crossings of the accelerometer.

[0025] Preferably, the method of the present invention further comprises: the cosine similarity calculation method of step S3.2.4 is as follows:

[0026]

[0027] Among them, S z The expression of n is n=min(M,N), which means taking the smaller value of M and N. N represents the number of zero crossings of the quartz accelerometer and M represents the number of zero crossings of the MEMS accelerometer. azi Represents the zero-crossing sequence of the quartz accelerometer, denoted as [t az1 ,t az2 ,...t azi ],i∈[1,N],t mzi Represents the zero-crossing sequence of the MEMS accelerometer, denoted as [t mz1 ,t mz2 ,...t mzi ]i∈[1, M],; the peak data, peak corresponding time data and zero-crossing time sequence data of the two accelerometers, and then the cosine similarity s is calculated by the peak data of the two accelerometers a , the cosine similarity s is calculated by the peak moments of the two accelerometers az , the cosine similarity s is calculated by the zero-crossing moment of the two accelerometers z , and finally use the average value to integrate and measure, the calculation method is as follows:

[0028]

[0029] Among them, s represents the waveform consistency judgment value, s az represents the cosine similarity of peak time, s z It represents the peak cosine similarity. The closer s is to 1, the more consistent the waveforms of the quartz accelerometer and the MEMS accelerometer are. When s is greater than the set threshold, it means that the waveforms of the two accelerometers in the vertical axis are consistent. The waveform similarity in other directions is determined. The data of the north-south axis and the east-west axis are combined and calculated. The combination formula is as follows:

[0030]

[0031] where a z Indicates vertical axis data, a x Indicates the north-south axis data, a yRepresents the east-west axis data; the waveform consistency judgment of the three direction axes refers to the judgment process of the vertical axis mentioned above. If it exceeds the threshold, the vibration is judged as a strong vibration.

[0032] The present invention also provides a dual-accelerometer strong vibration monitor, comprising: at least two monitoring points, the monitoring points are set in the building structure, so a distance is retained between the monitoring points, the distance is at least 2 meters, and a height difference is retained between the monitoring points; a monitoring host, set at the monitoring point, the monitoring host includes: a data processing module 8 and a storage and transmission module, the data processing module 8 includes the strong vibration discrimination method as described above; a plurality of quartz accelerometers, the plurality of quartz accelerometers are fixed on another monitoring point, respectively installed along the vertical axis, the north-south axis and the east-west axis of the ground, and connected to the monitoring host signal; a plurality of MEMS accelerometers, set in the monitoring host and connected to the monitoring host signal.

[0033] Preferably, the monitor of the present invention further includes: a plurality of 32-bit analog-to-digital converter modules, which are connected to the signals of the plurality of quartz accelerometers; a plurality of 24-bit analog-to-digital converter modules, which are connected to the signals of the MEMS accelerometers; the 32-bit analog-to-digital converter modules and the 24-bit analog-to-digital converter modules are connected to the data processing module via a digital isolation module.

[0034] Preferably, the monitor of the present invention further includes: the data processing module includes: a microprocessor and a global navigation satellite system module, and the microprocessor and the global navigation satellite system module are signal-connected to transmit the time signal and geographic coordinate signal acquired by the global navigation satellite system module to the microprocessor.

[0035] Preferably, the monitor of the present invention also includes: the storage and transmission module includes: a storage device, a serial communication interface, a network interface and a host computer, the serial communication interface and the network interface are respectively connected to the host computer and the microprocessor signal, so that the signal between the host computer and the microprocessor is conductive; the storage device is connected to the microprocessor signal, so that the data generated by the microprocessor is stored in the storage device; the storage device includes: at least one of a flash memory, an electrically erasable programmable read-only memory and a secure digital card memory.

[0036] Compared with the prior art, the present invention has the following advantages:

[0037] By integrating quartz accelerometers and MEMS accelerometers, the present invention can accurately distinguish strong vibration signals from environmental noise signals, effectively reducing the false alarm rate in the vibration signal recognition process, enhancing the reliability of the system, and improving the accuracy of strong vibration monitoring of buildings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 , is a flow chart of embodiment 1 of the present invention;

[0039] Figure 2 , is a schematic diagram of the structure of the monitoring instrument of the present invention;

[0040] Figure 3 , is a schematic diagram of the connection relationship of the components of the monitoring instrument of the present invention;

[0041] In the figure: 1 quartz accelerometer, 2 MEMS accelerometer, 3 monitoring host, 4 communication interface, 5 32-bit analog-to-digital converter module, 6 24-bit analog-to-digital converter module, 7 microprocessor, 8 data processing module, 9 serial communication interface, 10 network interface, 11 host computer, 12 flash memory, 13 electrically erasable programmable read-only memory, 14 secure digital card memory, 15 global navigation satellite system module, 16 real-time clock module, 17 digital isolation module. DETAILED DESCRIPTION

[0042] In order to make the purpose, principle and structure of the present invention more clear, it is further described below with reference to the accompanying drawings and specific embodiments.

[0043] Example 1:

[0044] This embodiment provides a dual accelerometer strong vibration discrimination method, which has low computational difficulty and is stable and reliable, and is suitable for running in the microprocessor 7. Figure 1 .

[0045] Step S1: Collect vibration signals through a dual accelerometer monitor.

[0046] Specifically, the microprocessor synchronously acquires data from the quartz accelerometer and the MEMS accelerometer via the ADC module (also known as the analog-to-digital converter module). The raw data from the six channels of the two accelerometers read from the ADC module is first converted based on sensitivity to restore the actual acceleration data for further processing.

[0047] Step S2: The collected vibration signal is subjected to acceleration value restoration, filtering, and de-averaging, and then enters the strong vibration identification stage.

[0048] Specifically, the data of each channel is low-pass filtered to reduce the interference of high-frequency noise on the vibration data. After filtering, the mean of the waveform data of each channel is removed.

[0049] Step S3. Strong vibration judgment:

[0050] Step S3.1. Process the two accelerometer signals using a vibration wave picking algorithm. If both data exceed the threshold within a short period of time, it is preliminarily determined to be a strong vibration signal.

[0051] Specifically, the specific method of step S3.1 is as follows:

[0052] By using the STA / LTA algorithm (also known as the "short-time average / long-time average" algorithm) to monitor and calculate, the waveform values ​​R of the vertical axis of the quartz accelerometer signal and the MEMS accelerometer signal are obtained respectively. a 、R m To reduce the amount of calculation, the parameters of the STA / LTA algorithm used by the two accelerometers are consistent; the trigger threshold R is set th and the ending threshold R e The trigger threshold and the end threshold are also the same. The threshold flags corresponding to the quartz accelerometer and MEMS accelerometer are F as 、F ms .

[0053] The exceeding threshold flag F as 、F ms The waveform value R perpendicular to the axis a 、R m and trigger threshold R th The logical relationship is as follows:

[0054]

[0055] There is a common exceeding threshold flag F s , the common exceeds the threshold flag F s and the threshold flag F as 、F ms The logical relationship is as follows:

[0056] F s =F as and F ms ;

[0057] and means F as and F ms Perform AND logic operation, when both accelerometers are triggered, F s is 1, otherwise it is 0. In order to reduce the interference caused by accidental factors, when F s When the value of is 1 for k consecutive times (e.g., k=3), it is determined that both accelerometer signals exceed the trigger threshold, k is any set positive integer, and the time t obtained by the global navigation satellite system module is recorded at the same time. s .

[0058] It also includes setting the end threshold flag F of the quartz accelerometer signal and the MEMS accelerometer signal ae 、F me End threshold flag F ae 、F me The waveform value R perpendicular to the axisa 、R m and the ending threshold R e The logical relationship is as follows:

[0059]

[0060] There is a common end threshold flag F e The common end threshold flag F e and the threshold flag F as 、F ms The logical relationship is as follows:

[0061] F e =F ae and F me ;

[0062] and means F ae and F me Perform AND logic operation at t s Monitor F after the start of the time e The value of F e When the values ​​of are all 1, the trigger is judged to be over, and the time t obtained by the global navigation satellite system module is recorded. e , k is any set positive integer.

[0063] Step S3.2. Perform consistency judgment on the vertical waveforms of the two accelerometer signals. If the vertical waveform consistency index exceeds the threshold, it indicates that the two sensors at different positions and different models have consistent responses, and it can be determined as a strong vibration signal of the entire real building; if the vertical waveform consistency index does not exceed the threshold, it is identified as vibration caused by environmental noise, etc. and ignored, thereby reducing false alarms and improving system reliability; if the vertical waveform consistency index only partially exceeds the threshold, the three-axis data of the two accelerometer signals are synthesized and enter step S3.3.

[0064] Specifically, the specific method of step S3.2 is as follows:

[0065] Step S3.2.1. Intercept the two accelerometer signals at time t s to t e The waveform data on the vertical axis between the two is analyzed, and waveform consistency is judged from the perspectives of similarity and amplitude. To reduce computational complexity, waveform similarity is expressed in terms of phase; the more consistent the phase, the more similar the waveforms are. Phase consistency is expressed in terms of zero-crossing consistency. The waveform data is smoothed to reduce interference caused by multiple spikes and jitter near the zero-crossing point.

[0066] Step S3.2.2. Calculate all peak values; calculate the peak value by the improved slope sign change formula. The improvement is that the threshold value in the slope sign change formula is not set. All peak values ​​are calculated first and then the number of peak values ​​N1 is obtained. Smooth the waveform data again and calculate the number of peak values ​​N2. If N1≠N2, the latter peak value calculation is considered valid. Otherwise, the former is considered valid. Continue to smooth the waveform data and calculate the number of peak values ​​until the former and latter peak values ​​are equal. At this time, the former calculation is considered valid. Peak value calculation and identification are completed, and peak value data and peak value corresponding time data are obtained. The accelerometer peak value is [A x1 ,A x2 ,...A xi ], the peak corresponding time is represented by [t x1 ,t x2 ,...t xi ], i∈[1, N]. Where x can be substituted by a or m, a represents a quartz accelerometer, m represents a MEMS accelerometer, and N represents the number of accelerometer waveform peaks.

[0067] Specifically, the calculation method for all peak values ​​in step S3.2.2 is as follows:

[0068] Step S3.2.3. Calculate the zero-crossing point between the first peak and the second peak, take this zero-crossing point as the first zero-crossing point, and calculate multiple zero-crossing points in sequence.

[0069] Specifically, the zero-crossing point calculation method in step S3.2.3 is as follows, including:

[0070] Step S3.2.3.1. Calculate and obtain all zero-crossing time sequences.

[0071] In the prior art, the zero-crossing point calculation logic is as follows:

[0072] {X(t)<0and X(t+1)>0}or{X(t)>0and X(t+1)<0}:

[0073] Where X(t) represents the value of a common time series (such as vibration waveform, temperature waveform, etc.) at time t.

[0074] To avoid interference caused by zero-crossing noise, a threshold ε≤|X(t)-X(t+1)| is generally required. However, this threshold setting method has two problems: first, if the threshold is too small, it will lose its effect; if it is too large, the zero-crossing point will be missed; second, only one position in a section of zero-crossing fluctuation data will be selected as the zero-crossing point, which reduces accuracy.

[0075] The improved solution of the present invention is: calculate all zero-crossing time sequences, recorded as [t xz1 ,txz2 ,...t xzi ], i∈[1,m], where m is the number of zero-crossing points calculated before correction, and x can be substituted into a or m, where a represents a quartz accelerometer and m represents a MEMS accelerometer. i ], i∈[1, N], where N is the number of peaks calculated in S3.2.2.

[0076] Step S3.2.3.2. Obtain the average time difference Δt based on the peak corresponding time data obtained from all peak calculations. The time data is a time series [t1, t2, ... t i ], i∈[1, N], where N is the number of peaks calculated in step S3.2.2, and the average time difference Δt is calculated as follows:

[0077]

[0078] The average time difference Δt is approximately the same as the peak time difference between adjacent peaks of a vibration waveform, that is, approximately half of a vibration waveform period;

[0079] Step S3.2.3.3. Setting the threshold ε z , the calculation formula is:

[0080]

[0081] Where k is an arbitrary positive integer, and this threshold is used to calculate the zero-crossing time sequence [t xz1 ,t xz2 ,...t xzi ], i∈[1, m] is corrected, and the correction calculation method refers to the following formula:

[0082]

[0083] Through the above steps, we can obtain peak data, peak moment data and zero-crossing moment sequence data, and thus obtain the zero-crossing sequence, peak moment and peak value of the quartz accelerometer and MEMS accelerometer. Where x can be substituted into a or m, a represents the quartz accelerometer, m represents the MEMS accelerometer, and the zero-crossing moment sequence of the quartz accelerometer is recorded as [t az1 ,t az2 ,...t azi ], i∈[1, N], where N represents the number of zero crossings of the quartz accelerometer, and the peak time and peak value are recorded as [t a1 ,t a2 ,...t ai ],[A a1 ,A a2 ,...Aai ], the zero-crossing time sequence, peak time and peak value of MEMS accelerometer are [t mz1 ,t mz2 ,...t mzi ],[t m1 ,t m2 ,...t mi ],[A m1 ,A m2 ,...A mi ] represents, where i∈[1,M], M represents the number of zero crossing points of the MEMS accelerometer.

[0084] Step S3.2.4. Calculate the cosine similarity of the corresponding data of the two accelerometer signals.

[0085] Specifically, the cosine similarity calculation method in step S3.2.4 is as follows:

[0086]

[0087] Among them, S z Indicates the similarity of the zero-crossing sequence of the two accelerometers. The expression of n is n=min(M,N), that is, taking the smaller value of M and N, t azi represents the zero-crossing sequence of the quartz accelerometer, t mzi Represents the zero-crossing sequence of the MEMS accelerometer, i∈[1, min(M, N)], min(M, N) indicates the selection of the smaller value between M and N, M represents the number of zero-crossing points of the MEMS accelerometer, and N represents the number of zero-crossing points of the quartz accelerometer; the peak data, peak corresponding time data and zero-crossing time sequence data of the two accelerometers are calculated by the above cosine similarity calculation method, and then the cosine similarity s is calculated by the peak data of the two accelerometers a , the cosine similarity s is calculated by the peak moments of the two accelerometers az , the cosine similarity s is calculated by the zero-crossing moment of the two accelerometers z The above three cosine similarities can all be obtained using the same cosine similarity calculation formula, and finally the average value is used for integration and measurement. The calculation method is as follows:

[0088]

[0089] Among them, s represents the waveform consistency judgment value, s az represents the cosine similarity of peak time, s zIt represents the peak cosine similarity. The closer s is to 1, the more consistent the waveforms of the quartz accelerometer and the MEMS accelerometer are. When s is greater than the set threshold (such as 0.8), it means that the waveforms of the two accelerometers in the vertical axis are consistent, and the waveform similarity in other directions is determined. In order to reduce the number of calculation steps, the north-south and east-west data are no longer calculated separately. Instead, they are synthesized and calculated. The synthesis formula is as follows:

[0090]

[0091] where a z Indicates vertical axis data, a x Indicates the north-south axis data, a y Represents the east-west axis data; the waveform consistency judgment of the three direction axes refers to the judgment process of the vertical axis mentioned above. If it exceeds the threshold, the vibration is judged as a strong vibration.

[0092] Step S3.3. The consistency index of the waveform of the vertical axis of the synthesized data is judged again. If it exceeds the threshold, it is judged as a strong vibration signal. If it does not exceed the threshold, it is judged as a non-strong vibration signal and enters step S4.

[0093] Step S4. The strong vibration monitor marks the stored strong vibration data according to the number of strong vibrations and the maximum amplitude, and in addition forms a separate file for strong vibration data exceeding a specific threshold for storage and subsequent further analysis and processing.

[0094] Step S5. The acceleration data in the selected strong vibration data file is synthesized, de-meaned and de-linearized, and then integrated and corrected to obtain the acceleration peak, motion velocity peak, displacement peak, and duration parameters generated by the vibration, which are used as building health monitoring indicators.

[0095] Example 2:

[0096] This embodiment provides a dual accelerometer strong vibration monitor, see Figure 2 、 Figure 3 .include:

[0097] At least two monitoring points are placed in the structure of the building and a certain distance is maintained between them, with a distance of at least more than 2 meters, and there is also a height difference between the monitoring points.

[0098] The monitoring host 3 is installed at one of the monitoring points and includes a data processing module 8 and a storage and transmission module. The data processing module 8 integrates the strong vibration identification method described in the first embodiment.

[0099] The data processing module 8 further includes a microprocessor 7 and a global navigation satellite system module 15 . The microprocessor 7 is signal-connected to the global navigation satellite system module 15 for transmitting the time signal and geographic coordinate signal acquired by the global navigation satellite system module 15 to the microprocessor 7 .

[0100] The storage and transmission module includes a storage device, a serial communication interface 9, a network interface 10, and a host computer 11. The serial communication interface 9 and the network interface 10 form the communication interface 11. The serial communication interface 9 and the network interface 10 respectively establish signal connections with the host computer 11 and the microprocessor 7, enabling signal transmission between the host computer and the microprocessor 7. Through the serial communication interface 9 and the network interface 10, the system can communicate with external computers or other devices, facilitating data transmission and monitoring. The serial communication interface 9 is primarily used for debugging with the host computer 11. Through the RJ45 network interface 10, the system can be connected to the network to achieve remote data transmission and monitoring, which is particularly important for real-time monitoring.

[0101] A storage device is connected to the microprocessor 7 to store data generated by the microprocessor 7. The storage device includes at least one of a flash memory 12, an electrically erasable programmable read-only memory 13, and a secure digital card memory 14. The flash memory 12 is a W25Q256 memory and stores firmware for program updates. The electrically erasable programmable read-only memory 13 is a 24LC08BT-I / OT memory and is used to store parameters such as trigger thresholds. The secure digital card memory 14 is used to store large amounts of vibration data and historical records. This design ensures long-term data storage and accessibility, facilitating subsequent earthquake analysis and research.

[0102] The system also includes multiple quartz accelerometers 1, which are fixed at another monitoring point and installed along the vertical, north-south, and east-west axes of the ground. They are also connected to the monitoring host 3 for signals. These quartz accelerometers 1 are enclosed in a housing and connected to the monitoring host 3 via cables. The quartz accelerometers 1 are FH2000D quartz flexible accelerometers, which offer high precision and stability.

[0103] Similarly, the system is equipped with multiple MEMS accelerometers 2. Due to their small size, these are directly integrated onto the circuit board of the monitoring host 3 and connected to it for signal communication. The MEMS accelerometers 2 are ADXL356 3-axis analog output accelerometers with low noise, low drift, and low power consumption.

[0104] In order to reduce the interference caused by human factors to this device, the monitoring point where the quartz accelerometer 1 is located should be installed as far away as possible from the monitoring point where the monitoring host 3 containing the MEMS accelerometer 2 is installed. For example, the connecting cable between the quartz accelerometer 1 and the monitoring host 3 is designed to be more than 2 meters long, the quartz accelerometer 1 is installed vertically on the wall, and the monitoring host 3 is installed horizontally on the ground. In this way, the response signals of the two sensors for the same signal generated by human activities are likely to be different, so that they can be identified and filtered out through consistency. As for the overall vibration of the building caused by strong vibration, two accelerometers of different models at different locations can produce more consistent responses, thereby improving the accuracy of the judgment.

[0105] The system also includes multiple 32-bit analog-to-digital converter modules 5, which are connected to the quartz accelerometer 1 for signal transmission. And multiple 24-bit analog-to-digital converter modules 6, which are connected to the MEMS accelerometer 2 for signal transmission. These analog-to-digital converter modules are connected to the data processing module 8 via the digital isolation module 17. The quartz accelerometer 1 and the MEMS accelerometer 2 are both analog accelerometers, which facilitates synchronous data acquisition using analog-to-digital converter modules. For example, three quartz accelerometers 1 and three-axis MEMS accelerometers 2 have a total of six channels of data. In order to ensure high consistency in data sampling time, six analog-to-digital converter modules are required for synchronous sampling. In order to give full play to the high precision and low noise characteristics of the quartz accelerometer 1, three 32-bit analog-to-digital converter modules 5 are selected to synchronously acquire the data from the three quartz accelerometers 1. For the MEMS accelerometer 2, three 24-bit analog-to-digital converter module 6 chips can meet the requirements. The 32-bit analog-to-digital converter module 5 uses the LTC2500-32 series chip, while the 24-bit analog-to-digital converter module 6 uses the LTC2512-24 series chip. To improve vibration data acquisition accuracy, the analog-to-digital converter chip in the monitoring host 3 is connected to the data processing module 8 via a digital isolation module 17. This ensures stable and accurate data transmission and reduces the impact of noise and external interference on data acquisition. The isolation chip signal is ADUM262N0BRIZ.

[0106] The present invention uses a microprocessor 7 as the main control unit and data processing unit, model STM32F429ZGT6, which controls six analog-to-digital converter chips through a digital isolation module 17 to collect data from a quartz accelerometer 1 and a MEMS accelerometer 2. The microprocessor 7 is connected to the global navigation satellite system module 15 and the real-time clock module 16 for signal conduction, and obtains precise time to ensure the time accuracy of the vibration data. The monitor integrates a secure digital card memory 14 to store accelerometer data. The hourly acceleration sensor data is synthesized into a file and stored in the secure digital card memory 14. In order to reduce the overall data storage volume, each hour's data file is marked according to the number of strong vibration signals and the maximum vibration amplitude. Files with a small number or low vibration amplitude are deleted or overwritten first when the storage space is full.

[0107] This monitor supports both adapter and lithium battery power, and integrates voltage and current monitoring with overvoltage and overcurrent protection. The voltage provided by the adapter and lithium battery passes through a switching power supply module and a low-dropout voltage regulator to provide a stable, matched low-voltage power source for each module. The lithium battery serves as a backup power source, providing continuous power in the event of a main power outage, ensuring continuous operation of the monitoring system.

[0108] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent replacement or change made by any technician familiar with the technical field within the technical scope disclosed by the present invention based on the technical solution and novel concept of the present invention should be covered by the scope of protection of the present invention.

Claims

1. A dual accelerometer strong vibration identification method, characterized in that: The steps include: Step S1. collecting vibration signals, including two accelerometer signals of a quartz accelerometer and a MEMS accelerometer; Step S2: performing acceleration restoration, filtering, and de-averaging on the collected vibration signal; Step S3: performing strong vibration signal identification, including: Step S3.

1. Process the vibration signal through the vibration wave picking algorithm. If both accelerometer signals in the vibration signal exceed the threshold, it is preliminarily determined that the vibration signal is a strong vibration signal; Step S3.

2. Compare the waveform consistency index of the vertical axis of the two accelerometer signals with the threshold value to determine: If both consistency indicators exceed the threshold, it is determined to be a strong vibration signal and the process goes to step S4; If both consistency indicators do not exceed the threshold, it is determined to be a non-strong vibration signal and the process goes to step S4; If the two consistency indicators only partially exceed the threshold, the three-axis data of the two accelerometer signals are synthesized and the process proceeds to step S3.3; Step S3.

3. The consistency index of the waveform of the vertical axis of the synthesized data is judged again. If it exceeds the threshold, it is determined to be a strong vibration signal. If it does not exceed the threshold, it is determined to be a weak vibration signal and the process proceeds to step S4. Step S4. The strong vibration monitor marks the data identified as strong vibration signals, including the number of strong vibrations and the maximum amplitude, and stores the strong vibration signal data exceeding a specific threshold in a separate file; Step S5. The acceleration data in the marked strong vibration signal data file is synthesized, de-averaged, de-linearized, and then integrated and corrected to obtain the parameters generated by the vibration signal as building structural health monitoring indicators, including: acceleration peak, motion velocity peak, displacement peak and duration.

2. A dual accelerometer strong vibration discrimination method according to claim 1, characterized in that: The specific method of step S3.1 is as follows: The waveform values ​​R of the vertical axis of the quartz accelerometer signal and the MEMS accelerometer signal are obtained by calculation. a 、R m ; Set the trigger threshold R th and the ending threshold R e ; The threshold exceeding flag bits corresponding to the quartz accelerometer and MEMS accelerometer are F as 、F ms , the exceeding threshold flag F as 、F ms The waveform value R perpendicular to the axis a 、R m and trigger threshold R th The logical relationship is as follows: There is a common exceeding threshold flag F s , the common exceeds the threshold flag F s and the threshold flag F as 、F ms The logical relationship is as follows: F s =F as and F ms ; and means F as and F ms Perform AND logic operation, when F s When the value of is 1 for k consecutive times, it is determined that both accelerometer signals exceed the trigger threshold, k is any set positive integer, and the time t obtained by the global navigation satellite system module is recorded at the same time s ; It also includes setting the end threshold flag F of the quartz accelerometer signal and the MEMS accelerometer signal ae 、F me , end threshold flag F ae 、F me The waveform value R perpendicular to the axis a 、R m and the ending threshold R e The logical relationship is as follows: There is a common end threshold flag F e , the common end threshold flag F e and the threshold flag F as 、F ms The logical relationship is as follows: F e =F ae and F me ; and means F ae and F me Perform AND logic operation at t s Monitor F after the start of the time e The value of F e When the values ​​of are all 1, the trigger is judged to be over, and the time t obtained by the global navigation satellite system module is recorded. e , k is any set positive integer.

3. A dual accelerometer strong vibration identification method as claimed in claim 2, characterized in that: The specific method of step S3.2 is as follows: Step S3.2.

1. Intercept the two accelerometer signals at time t s to t e The waveform data of the vertical axis between the two axes is smoothed; Step S3.2.

2. Calculate all peak values; Step S3.2.

3. Calculate the zero-crossing point between the first peak and the second peak, take this zero-crossing point as the first zero-crossing point, and sequentially calculate multiple zero-crossing points and zero-crossing point time sequences; Step S3.2.

4. Calculate the cosine similarity of the corresponding data of the two accelerometer signals.

4. A dual accelerometer strong vibration identification method as claimed in claim 3, characterized in that: The calculation method for all peak values ​​in step S3.2.2 is as follows: After calculating all the peaks, the number of peaks N1 is obtained; the waveform data is smoothed again to calculate the number of peaks N2. If N1≠N2, the latter peak calculation is considered valid, otherwise the previous one is valid. Continue to smooth the waveform data and calculate the number of peaks until the two peak numbers are equal. At this time, the previous calculation is considered valid; the peak calculation and recognition are completed, and the peak data and the peak corresponding time data are obtained.

5. A dual accelerometer strong vibration identification method as claimed in claim 4, characterized in that: The zero-crossing point calculation method in step S3.2.3 is as follows, including: Step S3.2.3.

1. Calculate all zero-crossing time sequences; Step S3.2.3.

2. Obtain the average time difference Δt based on the peak corresponding time data obtained from all peak calculations. The time data is the peak time sequence [t1, t2, ... t i ], i∈[1, N], where N is the number of peaks obtained by calculation, and the average time difference Δt is calculated as follows: The average time difference Δt is approximately the same as the peak time difference between adjacent peaks of a vibration waveform, that is, approximately half of a vibration waveform period; Step S3.2.3.

3. Setting the threshold ε z , the calculation formula is: Where k is an arbitrary positive integer. This threshold is used to correct the zero-crossing time sequence of the two accelerometers. The correction calculation method is as follows: Through the above steps, the peak data, peak time data and zero-crossing time series data of the quartz accelerometer and MEMS accelerometer are obtained. xzi , where x can be substituted into a or m, a represents the quartz accelerometer, m represents the MEMS accelerometer, i∈[1,m], and m represents the number of zero crossings of the accelerometer.

6. A dual accelerometer strong vibration identification method as claimed in claim 5, characterized in that: The cosine similarity calculation method of step S3.2.4 is as follows: Among them, S z The expression of n is n=min(M,N), which means taking the smaller value of M and N. N represents the number of zero crossings of the quartz accelerometer and M represents the number of zero crossings of the MEMS accelerometer. azi Represents the zero-crossing sequence of the quartz accelerometer, denoted as [t az1 ,t az2 ,...t azi ],i∈[1,N],t mzi Represents the zero-crossing sequence of the MEMS accelerometer, denoted as [t mz1 ,t mz2 ,...t mzi ]i∈[1, M]; the peak data, peak corresponding time data and zero-crossing time sequence data of the two accelerometers, and then the cosine similarity s is calculated by the peak data of the two accelerometers a , the cosine similarity s is calculated by the peak moments of the two accelerometers az , the cosine similarity s is calculated by the zero-crossing moment of the two accelerometers z , and finally use the average value to integrate and measure, the calculation method is as follows: Among them, s represents the waveform consistency judgment value, s az represents the cosine similarity of peak time, s z It represents the peak cosine similarity. The closer s is to 1, the more consistent the waveforms of the quartz accelerometer and the MEMS accelerometer are. When s is greater than the set threshold, it means that the waveforms of the two accelerometers in the vertical axis are consistent. The waveform similarity in other directions is determined. The data of the north-south axis and the east-west axis are combined and calculated. The combination formula is as follows: where a z Indicates vertical axis data, a x Indicates the north-south axis data, a y Represents the east-west axis data; the waveform consistency judgment of the three direction axes refers to the judgment process of the vertical axis mentioned above. If it exceeds the threshold, the vibration is judged as a strong vibration.

7. A dual accelerometer strong vibration monitor, characterized in that: include: At least two monitoring points, the monitoring points being arranged in the building structure so that there is a spacing between the monitoring points, the spacing being at least 2 meters, and a height difference being maintained between the monitoring points; A monitoring host is provided at a monitoring point, the monitoring host comprising: a data processing module and a storage and transmission module, the data processing module comprising the strong vibration discrimination method according to any one of claims 1 to 6; A plurality of quartz accelerometers, which are fixed at another monitoring point, are respectively installed along the vertical axis, the north-south axis, and the east-west axis of the ground, and are connected to the monitoring host signal; Several MEMS accelerometers are arranged in the monitoring host and are connected to the monitoring host signal.

8. The dual accelerometer strong vibration monitor according to claim 7, characterized in that: Also includes: a plurality of 32-bit analog-to-digital converter modules, connected to the plurality of quartz accelerometer signals; Several 24-bit analog-to-digital converter modules connected to the MEMS accelerometer signals; The 32-bit analog-to-digital converter module and the 24-bit analog-to-digital converter module are connected to the data processing module via a digital isolation module.

9. A dual accelerometer strong vibration monitor as claimed in claim 7, characterized in that: The data processing module includes: A microprocessor and a global navigation satellite system module, wherein the microprocessor and the global navigation satellite system module are signal-connected for transmitting a time signal and a geographic coordinate signal acquired by the global navigation satellite system module to the microprocessor.

10. The dual accelerometer strong vibration monitor according to claim 7, characterized in that: The storage transmission module includes: a storage device, a serial communication interface, a network interface and a host computer, wherein the serial communication interface and the network interface are respectively connected to the host computer and the microprocessor, so that the signals between the host computer and the microprocessor are connected; the storage device is connected to the microprocessor, so that the data generated by the microprocessor is stored in the storage device; The storage device includes at least one of a flash memory, an electrically erasable programmable read-only memory, and a secure digital card memory.

Images