Method for picking up impact vibration sensor signal

Abstract

The present invention relates to a method for picking up an impact vibration sensor signal, which includes sampling an impact vibration signal, finding its average value, finding the maximum value, determining whether the maximum value is in the middle position of the data, and if so, determining whether the data on both sides of the maximum value are monotonically increasing or decreasing with basically the same increasing and decreasing values, and collecting suitable data. If not, adjust the data and repeat until the requirements are met. The signal collected by the impact vibration sensor can be processed using a time domain method to obtain an impact vibration signal that meets the requirements, and the signal can be collected in real time with high accuracy. At the same time, the result of the time domain processing can be further reinforced by a frequency domain method to improve the reliability and accuracy of the collected signal.

Description

Technical Field

[0001] The present invention relates to the technical field of vibration testing, and particularly to a method for picking up shock vibration sensor signals.

Background Art

[0002] As China becomes a major manufacturing country, the reliability of products has attracted global attention. Before excellent products are put on the market, they need to undergo a series of strict tests such as dust tests, salt spray tests, vibration tests, shock tests, and temperature tests. This is to verify whether the product design meets the requirements.

[0003] In shock testing equipment, the most important parameter is the intensity of the shock. To accurately set the shock intensity, it is necessary to accurately pick up the signal from a shock vibration sensor (which means the general term for a shock sensor and a vibration sensor). However, the current picking up of shock vibration signals cannot accurately pick up the shock vibration signals due to interference such as internal disturbances and external environmental noises. Therefore, multiple pickups and judgments of the pickup results are required, which has a great impact on the quality and efficiency of the test.

[0004] Therefore, there is a need for a technology that can accurately pick up shock vibration signals even under external interference.

Summary of the Invention

[0005] To solve the above problems, the present invention provides a method for picking up shock vibration sensor signals. Thereby, under self and external interferences, the shock vibration signals can be accurately and real-time picked up.

[0006] To solve this technical problem, the technical approach adopted by the present invention is as follows. That is, a method for picking up shock vibration sensor signals, the picking up method including the following steps. S1: Use the impact vibration sensor to sample the vibration frequency, obtain a plurality of sample data, distribute the sample data to a plurality of response cycles according to the determined response cycle, select one response cycle and proceed to step S2. S2: Calculate the average value of the sample data within that response cycle and find the maximum value. S3: Determine whether the maximum value is at the central position of the sampling time Forward If so, proceed to step S6; otherwise, proceed to step S4. S4: Determine whether the sample data meets the requirements of quantity and trend. If it meets the requirements, proceed to step S5; otherwise, select the sample data of the next response cycle and return to step S2. S5: Starting from the effective value (rms value) of the current sample data, move the position of the response cycle, newly obtain a complete response cycle, and return to step S2. S6: Determine whether the left side of the maximum value is monotonically increasing, the right side is monotonically decreasing, and whether the increasing value and the decreasing value are the same. If all are met, it is determined that the sample data within that response cycle meets the requirements; otherwise, select the sample data of the next response cycle and return to step S2.

[0007] In a more specific technical approach, determining whether the sample data meets the requirements of quantity and trend in step S4 includes the following steps. S41: Search for the number of data exceeding the average value on the left side of the maximum value. If the number is less than N, proceed to step S42; otherwise, proceed to step S43. S42: Search for the number of data exceeding the average value on the right side of the maximum value. If the number is less than N, select the sample data of the next response cycle and return to step S2; otherwise, proceed to step S44. S43: Divide the sample data on the left side of the maximum value into k equal parts, add the sample data of each part to obtain k total values, and determine whether they are monotonically increasing. If so, proceed to step S5; otherwise, select the sample data of the next response cycle and return to step S2. S44: Divide the sample data on the right side of the maximum value into k equal parts, add the sample data of each part to obtain k total values, determine whether they are monotonically decreasing. If so, proceed to step S5; otherwise, select the sample data of the next response period and return to step S2.

[0008] In a more specific technical approach, the value of N in steps S41 and S42 is 100.

[0009] In a more specific technical approach, for the determination of monotonic increase on the left side of the maximum value in step S6, the sample data within the response period is divided into q equal parts, the sample data of each part is added to obtain q total values, and a plurality of total values on the left side of the maximum value are selected for determination.

[0010] In a more specific technical approach, for the determination of monotonic decrease on the right side of the maximum value in step S6, the sample data within the response period is divided into q equal parts, the sample data of each part is added to obtain q total values, and a plurality of total values on the right side of the maximum value are selected for determination.

[0011] In a more specific technical approach, after the acquisition of the impact vibration signal is completed, the sample data is processed using a frequency domain processing method to determine whether the impact vibration signal is detected.

[0012] In a more specific technical approach, the frequency domain processing method includes the following steps. S71: Apply FFT to all sample data to obtain a plurality of frequency data. S72: Apply Square-law detection to the plurality of frequency data, and take four frequency data as a set to generate a plurality of complex number data. S73: Calculate the first threshold value, convert the plurality of complex number data into real number data, sum them to calculate the average value, and calculate the standard deviation C σ based on the real number data and the average value. Here, the first threshold value is C p = k·C σand let k = 2. S74: Compare all the frequency data with the first threshold, select the frequency data smaller than the first threshold to form the second frequency data, and calculate the second threshold along steps S72 to S73. S75: Compare all the frequency data with the second threshold and count the number of frequency data exceeding the second threshold. S76: Determine whether the counted number of frequency data exceeds 10% of the total number of frequency data. If it exceeds, it is determined that an impact vibration signal has been detected; otherwise, it is determined that no impact vibration signal has been detected.

[0013] In a more specific technical approach, the complex number data in step S72 includes a real part and an imaginary part. The real part and the imaginary part are represented by the following formulas. The real part is [Number] and The imaginary part is [Number] is. Here, A (j) is the frequency data, and B (j) is the complex number data.

[0014] In a more specific technical approach, the formula for converting complex number data into real number data is as follows. [Number] [Advantages of the Invention]

[0015] The beneficial effects of the present invention are to process the signals collected from the impact vibration sensor in the time domain method to obtain the impact vibration signals that meet the requirements, thereby achieving high accuracy and realizing the collection of real-time signals. Furthermore, the frequency domain method is used to reinforce the results of the time domain processing to improve the reliability and accuracy of the collected signals.

Brief Description of the Drawings

[0016]

Figure 1

Figure 2

Figure 3

Embodiments for Carrying Out the Invention

[0017] Hereinafter, in order to clearly and completely describe the technical concept of this invention, it will be described with reference to the drawings. Obviously, the described embodiments are some embodiments of this invention, not all embodiments. Based on the embodiments according to this invention, other embodiments obtained by those of ordinary skill in the art without creative labor are all included in the protection scope of this invention.

[0018] In the description of this invention, it should be noted that terms such as "center", "upper", "lower", "left", "right", "vertical", "horizontal", "internal", "external", etc. are based on the illustrated positional and directional relationships. This is for the convenience of explaining this invention and simplifying the explanation, and does not indicate or imply that it is necessary to construct and operate in a specific direction or a specific orientation, so it is not understood as a restriction on this invention. Also, terms such as "first", "second", "third", etc. are only used for the purpose of explanation and do not indicate or imply relative importance.

[0019] In the description of this invention, it should be noted that terms such as "attachment", "connection", "connection", etc. should be interpreted in a broad sense. For example, it may be a fixed connection, a removable connection, or an integral connection. Or it may be a mechanical connection or an electrical connection. Or it may be directly connected or indirectly connected through an intermediate medium. It may also be an internal connection between two elements. A person of ordinary skill in this technical field can understand the specific meaning of the above terms in this invention. Furthermore, the technical features related to different embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0020] The response time to impact vibration is 3 to 50 milliseconds, and the maximum frequency response does not exceed 10k. This varies depending on the impact table body and the change in impact intensity and is mainly at low frequencies. Based on the FFT principle, the higher the sampling frequency, the higher the frequency resolution. For example, when the frequency response is 10k, the sampling frequency must be at least 20k, and below that, it cannot reflect the true signal spectrum information. If the sampling frequency of the A / D chip is 100k, the sampling width is 16 bits, and the sampling time is 10ms, the number of samplings within the impact vibration response time is 1000 times, and the data is 2000 bytes. If the data storage depth is set to three times the response time, the required memory is 6000 bytes. When the time window is set to 10 milliseconds, all this memory data corresponds to three impact response cycles.

[0021] The method for picking up the impact vibration sensor signal shown in Figure 1 includes the following steps.

[0022] S1: Use an impact vibration sensor to sample the vibration frequency, obtain a plurality of sample data, and distribute the plurality of sample data to a plurality of response cycles according to a determined response cycle. Select one response cycle. When performing software processing, assign the plurality of sample data within the response cycle to the D[i] array respectively, and proceed to step S2.

[0023] S2: Obtain the average value Dave of a plurality of sample data within the response period, that is, the plurality of sample data in the D[i] array, and at the same time find the maximum value Dmax in the D[i] array.

[0024] S3: Determine whether the maximum value is at the central position of the sampling time of the plurality of sample data Forward That is, determine whether the number of sample data on the left side of Dmax is equal to the number of sample data on the right side. If they are equal, it is considered that the sample data meets the basic requirements and proceed to step S6 for further processing. If they are not equal, it is determined that the sample data does not meet the requirements and adjustment is necessary, so proceed to step S4 for further processing.

[0025] S4: Determine whether the sample data meets the requirements of quantity and trend. As shown in FIG. 2, the specific judgment process regarding the quantity and trend includes the following steps.

[0026] S41: Search for the number of data exceeding the average value on the left side of the maximum value, and determine whether the number of sample data meeting this condition is less than N. Here, N is set to 100. If it is less than N, it is determined that the sample data on the left side of the maximum value Dmax does not meet the requirements and proceed to step S42. Otherwise, it is determined that the sample data on the left side of the maximum value Dmax meets the requirements and proceed to step S43, where the sample data on the left side of the maximum value Dmax is processed in step S43.

[0027] S42: Search for the number of data exceeding the average value on the right side of the maximum value, and determine whether the number of sample data meeting this condition is less than N. Here, N is set to 100. If it is less than N, it is determined that the sample data on the right side of the maximum value Dmax does not meet the requirements, reassign the sample data of the next response period to the D[i] array and return to step S2. Otherwise, it is determined that the sample data on the right side of the maximum value Dmax meets the requirements and proceed to step S44, where the sample data on the right side of the maximum value Dmax is processed.

[0028] S43: At this stage, a plurality of sample data on the left side of the maximum value Dmax are processed. That is, the sample data on the left side of the maximum value Dmax are evenly divided into k parts, and the sample data of each part are added to obtain k total values. Then, it is determined whether the k total values are monotonically increasing based on the time sequence of sampling. If they are monotonically increasing, it is determined that the requirements are met, and if so, proceed to step S5. Otherwise, it is determined that the requirements are not met, the sample data of the next response cycle are selected, reallocated to the D[i] array, and return to step S2.

[0029] S44: At this stage, a plurality of sample data on the right side of the maximum value Dmax are processed. That is, the sample data on the right side of the maximum value Dmax are evenly divided into k parts, and the sample data of each part are added to obtain k total values. Then, it is determined whether the k total values are monotonically decreasing based on the time sequence of sampling. If they are monotonically decreasing, it is determined that the requirements are met, and if so, proceed to step S5. Otherwise, it is determined that the requirements are not met, the sample data of the next response cycle are selected, reallocated to the D[i] array, and return to step S2.

[0030] S5: At this stage, the position of the response cycle is adjusted. That is, starting from the valid value (actual effective value) of the current sample data (the sample data on the left side of the maximum value Dmax or the sample data on the right side of the maximum value Dmax), the position of the response cycle is moved to newly obtain a plurality of response cycles. After selecting the sample data of the first response cycle after the movement and reallocating it to the D[i] array, return to step S2 to repeat the operation.

[0031] S6: Determine whether the left side of the maximum value Dmax is monotonically increasing, the right side of the maximum value Dmax is monotonically decreasing, and whether the monotonically increasing value and the monotonically decreasing value are the same.

[0032] First, divide the multiple sample data within the response period into q equal parts, add the sample data of each part to obtain q total values. Since the maximum value Dmax is at the central position of the D[i] array, the quantities of the total values on the left and right sides of the maximum value Dmax are the same.

[0033] After that, select all the total values on the left side of the maximum value Dmax, and determine whether all the total values are monotonically increasing based on the order of the collection time.

[0034] After that, select all the total values on the right side of the maximum value Dmax, and determine whether all the total values are monotonically decreasing based on the order of the collection time.

[0035] Finally, obtain the increasing values on the left side of the maximum value Dmax and the decreasing values on the right side of the maximum value Dmax, and compare them. If the difference between the two is within 5%, it is recognized that the requirements are met.

[0036] If all of the above three states meet the requirements, the sample data is accurate and can be used. If any one of the above three states does not meet the requirements, the sample data is inaccurate, select the sample data of the next response period, reassign it to the D[i] array, and return to step S2.

[0037] The impact vibration signal obtained by the above method is obtained by time-domain processing. In order to ensure a more accurate impact vibration signal, after the above steps are completed, perform frequency-domain processing on the sample data to determine whether an impact vibration signal is detected. Thereby, the accuracy of the signal is enhanced.

[0038] As shown in Figure 3, the method of this frequency-domain processing includes the following steps.

[0039] S71: Apply the FFT (Fast Fourier Transform) operation to all the sample data in the D[i] array within this cycle that has been collected to obtain a plurality of frequency data, and configure it into the A[i] array. Here, the number of frequency data in the A[i] array is the same as the number of sample data in the D[i] array.

[0040] S72: Apply Square-law detection to all the frequency data in the A[i] array, and take four frequency data as a set to generate a plurality of complex number data. Here, the complex number data includes a real part and an imaginary part.

[0041] The real part is the difference between the first frequency data and the third frequency data among the four frequency data divided by 2.

[0042]

Number

[0043] The imaginary part is the difference between the second frequency data and the fourth frequency data among the four frequency data divided by 2.

[0044]

Number

[0045] Here, A (j) is frequency data, and B (j) is complex number data.

[0046] S73: As described below, calculate the first threshold value.

[0047] First, convert a plurality of complex number data into real number data (detection).

[0048] The real number data is calculated based on the following formula.

Number

[0049] Then, the real data is summed up and the average value is calculated. The average value is

Number

[0050] After that, based on the real data and the average value, the standard deviation C σ is calculated.

[0051] The standard deviation is as follows.

Number

[0052] The first threshold is C p = k·C σ where k = 2.

[0053] S74: Compare all the frequency data in the A[i] array with the first threshold, select the frequency data smaller than the first threshold, form new frequency data, and calculate the second threshold along steps S72 to S73.

[0054] S75: Compare all the frequency data in the A[i] array with the second threshold and count the number of frequency data exceeding the second threshold.

[0055] S76: Determine whether the number of the counted frequency data exceeds 10% of the number of all the frequency data. If it exceeds, it is determined that an impact vibration signal is detected; otherwise, it is determined that no impact vibration signal is detected.

[0056] If no impact vibration signal is detected after the frequency domain processing, select the sample data of the next response period, reassign it to the D[i] array, and return to step S2.

[0057] Generally speaking, first, a signal is acquired by an impact vibration sensor, the signal is selected by time domain processing, and the effective signal is adjusted and output. As a result, the acquired signal has high accuracy and real-time performance. Next, the signal acquired by frequency domain processing is further judged, and the accuracy and reliability of the signal can be further improved.

[0058] It should be emphasized that the above is merely an example of the present invention and is not any form of restriction on the present invention. Based on the technical essence of the present invention, any simple changes, equivalent changes, and modifications to the above embodiments are still included in the scope of the technical concept of the present invention.

Claims

1. Step S1 of sampling the vibration frequency using a shock vibration sensor to obtain a plurality of sample data, distributing the plurality of sample data into a plurality of response cycles according to the determined response cycle, selecting one response cycle, and proceeding to step S2; Step S2 of obtaining the average value of the plurality of sample data within the selected response cycle and finding the maximum value; Step S3 of determining whether the maximum value is at the central position in the order of sampling time. If so, proceed to step S6; otherwise, proceed to step S4; Step S4 of determining whether the sample data meets the requirements of quantity and trend. If it meets the requirements, proceed to step S5; otherwise, select the sample data of the next response cycle and return to step S2; Step S5 of starting from the effective value of the current sample data, moving the position of the response cycle, newly obtaining a complete response cycle, and returning to step S2; Step S6 of determining whether the left side of the maximum value is monotonically increasing and the right side is monotonically decreasing, and whether the monotonically increasing value and the monotonically decreasing value are the same. If all are satisfied, it is determined that the sample data within that response cycle meets the requirements; otherwise, select the sample data of the next response cycle and return to step S2. A method for picking up shock vibration sensor signals, characterized by including the above steps.

2. The determination in step S4 of whether the sample data meets the requirements of quantity and trend is as follows: Step S41 of searching for the number of data exceeding the average value on the left side of the maximum value. If the number is less than N, proceed to step S42; otherwise, proceed to step S43; Step S42 of searching for the number of data exceeding the average value on the right side of the maximum value. If the number is less than N, select the sample data of the next response cycle and return to step S2; otherwise, proceed to step S44; Step S43 of equally dividing the plurality of sample data on the left side of the maximum value into k parts, adding the sample data of each part to obtain k total values, and determining whether they are monotonically increasing. If so, proceed to step S5; otherwise, select the sample data of the next response cycle and return to step S2; Divide a plurality of sample data on the right side of the maximum value into k equal parts, add the sample data of each part to obtain k total values, determine whether they are monotonically decreasing, if so, proceed to step S5, otherwise select the sample data of the next response period and return to step S2, step S44, which is characterized in that the method for picking up the impact vibration sensor signal according to claim 1.

3. The value of N in steps S41 and S42 is 100, which is characterized in that the method for picking up the impact vibration sensor signal according to claim 2.

4. The determination that the left side of the maximum value in step S6 is monotonically increasing includes dividing a plurality of sample data within the response period into q equal parts, adding the sample data of each part to obtain q total values, and selecting and determining a plurality of total values on the left side of the maximum value, which is characterized in that the method for picking up the impact vibration sensor signal according to claim 1.

5. The determination that the right side of the maximum value in step S6 is monotonically decreasing includes dividing a plurality of sample data within the response period into q equal parts, adding the sample data of each part to obtain q total values, and selecting and determining a plurality of total values on the right side of the maximum value, which is characterized in that the method for picking up the impact vibration sensor signal according to claim 1.

6. After the acquisition of the impact vibration signal is completed, it includes processing the sample data using a frequency domain processing method and determining whether the impact vibration signal is detected, which is characterized in that the method for picking up the impact vibration sensor signal according to claim 1.

7. The method of frequency domain processing is Applying a fast Fourier transform to all sample data to obtain a plurality of frequency data, step S71; Applying square detection to the plurality of frequency data, taking four frequency data as a set, and generating a plurality of complex number data, step S72; Convert a plurality of complex number data into real number data, sum them up to calculate an average value, and calculate a standard deviation C based on the real number data and the average value σ By calculating, set the first threshold C with k = 2 ｐ = k · C σ Step S73 of calculating; Comparing all frequency data with a first threshold value, selecting frequency data smaller than the first threshold value to form second frequency data, and calculating a second threshold value along steps S72 to S73, step S74; Comparing all frequency data with the second threshold value and counting the number of frequency data exceeding the second threshold value, step S75; Step S76 of determining whether the number of counted frequency data exceeds 10% of the total number of frequency data, and if it exceeds, determining that an impact vibration signal has been detected, and if not, determining that an impact vibration signal has not been detected, is included. The method for picking up an impact vibration sensor signal according to claim 6 is characterized by this. [

8. ] The complex number data in step S72 includes a real part and an imaginary part, and A （ｊ） is frequency data, and B （ｊ） is complex number data, and the real part is 【Number 1】 represented as, and the imaginary part is 【Number 2】 The method for picking up an impact vibration sensor signal according to claim 7 is characterized by including being represented as this.

9. 【Fig. 3】 The method for picking up an impact vibration sensor signal according to claim 8 is characterized by including converting complex number data into real number data according to the formula.

Images