A new self-learning and adaptive vibration sensing device

Abstract

The present invention discloses a novel self-learning and self-adaptive vibration sensing device, comprising: a vibration sensing element for sensing vibration and transmitting the sensed vibration signal to a single-chip processor; a single-chip processor for data processing, wherein the data processing process comprises receiving the vibration signal transmitted by the vibration sensing element and comparing the vibration signal with a signal model, and when the vibration signal matches the signal model, judging the vibration signal as a valid signal, and when the vibration signal does not match the signal model, judging the vibration signal as an invalid signal; a sensitivity adjustment setting switch for receiving a sensitivity adjustment instruction; a signal matching result display circuit for receiving data information sent by the single-chip processor and displaying the data information; and a data storage circuit for storing data. Compared with the prior art, the present invention has stronger reliability and adaptability.

Description

Technical Field

[0001] The invention relates to the field of vibration sensing technology, and in particular to a novel self-learning and self-adaptive vibration sensing device. Background Art

[0002] With the improvement of modern industrial automation, equipment systems with special and strict requirements for vibration sensing detection are becoming more and more important. In these applications, the reliability and adaptability of vibration sensing devices are crucial. They must be able to accurately distinguish between effective vibration signals and noise interference to avoid false alarms and missed alarms.

[0003] The prior art still has the following problems:

[0004] 1. False positives

[0005] In actual use environments, vibration factors that may cause false alarms are everywhere, such as wind and rain in outdoor environments, shock waves generated by loud noises (such as fireworks and firecrackers), ground vibrations caused by heavy vehicles passing by, etc. These situations can easily lead to false alarms in existing vibration sensing devices on the market.

[0006] 2. The contradiction between sensitivity and false alarm

[0007] In order to reduce false alarms, the usual practice is to increase the trigger threshold of the vibration sensing device, that is, a larger vibration intensity is required to trigger the sensing element. Although this approach can reduce the false alarm rate, it also sacrifices the sensitivity of the sensing device, making it difficult to detect effective vibration signals, that is, the problem of missed alarms is prominent.

[0008] 3. Electromagnetic interference problem

[0009] In actual application scenarios, vibration sensing devices often require long leads to connect to the signal processing unit. Long wires are susceptible to electromagnetic interference, which can lead to misjudgment of the device. Most products on the market currently do not adequately handle electromagnetic interference, which is another important factor affecting the reliability and applicability of vibration sensing devices.

[0010] 4. Technical limitations of existing vibration sensors

[0011] Selection of vibration sensing element: In existing vibration sensors, the sensitivity of the sensing element is differentiated by the trigger force. The sensitivity adjustment is generally achieved by adjusting the elastic force of the vibration element spring and the distance of the trigger contact point.

[0012] Fixed sensitivity problem: Once a certain sensing element is selected, its sensitivity is fixed, which makes the reliability and applicability of the equipment completely dependent on whether there is a clear difference between the vibration generated by the interference source and the effective vibration. Once the strength of the two is similar, false triggering will be inevitable.

[0013] Consistency issues of sensing elements: The difference in spring force is highly related to the material and manufacturing process. Ensuring the consistency of the sensing element itself is a difficulty in the existing technology.

[0014] Influence of installation conditions: In actual use environments, it is difficult to keep the installation position and installation method of the vibration sensor completely consistent, which will cause the actual effective vibration intensity to change. The sensitivity of the sensing element of the existing vibration sensing device is fixed, so a lot of debugging work is required on site to determine the appropriate vibration sensing device, which consumes a lot of time and resources.

[0015] In summary, the prior art has many limitations in ensuring the reliability and adaptability of vibration sensing devices, especially when dealing with complex and changeable actual environmental factors. In view of this, a new self-learning and self-adaptive vibration sensing device is proposed. Summary of the invention

[0016] The purpose of the present invention is to provide a novel self-learning and self-adaptive vibration sensing device, which has strong reliability and adaptability.

[0017] The above technical objectives of the present invention are achieved through the following technical solutions:

[0018] A novel self-learning and self-adaptive vibration sensing device, comprising:

[0019] A vibration sensing element, used for sensing vibration and transmitting the sensed vibration signal to the single chip processor;

[0020] A single-chip processor is used for data processing, wherein the data processing process includes receiving a vibration signal transmitted by a vibration sensing element, and comparing the vibration signal with a signal model. When the vibration signal matches the signal model, the vibration signal is judged to be a valid signal, and when the vibration signal does not match the signal model, the vibration signal is judged to be an invalid signal;

[0021] The sensitivity adjustment setting switch is used to receive a sensitivity adjustment instruction and transmit a sensitivity adjustment signal to the single-chip processor according to the instruction, so that the single-chip processor can adjust the sensitivity;

[0022] The signal matching result display circuit is used to receive the data information sent by the single-chip processor and display the data information;

[0023] A data storage circuit, used for storing data;

[0024] Among them, the vibration sensing element, the sensitivity adjustment setting switch, the signal matching result display circuit and the data storage circuit are electrically connected to the single chip processor respectively.

[0025] In a preferred embodiment, it also includes a lightning protection and electrostatic protection circuit for implementing lightning protection and electrostatic protection, and the lightning protection and electrostatic protection circuit is arranged between the single-chip processor and the vibration sensing element.

[0026] In a preferred embodiment, it also includes a signal filtering circuit for filtering the vibration signal, and the signal filtering circuit is arranged between the single-chip processor and the vibration sensing element.

[0027] In a preferred embodiment, it also includes a waveform extension circuit for proportionally extending the vibration signal waveform so that a narrow signal becomes a wide signal. The waveform extension circuit is arranged between the single-chip processor and the vibration sensing element.

[0028] In a preferred embodiment, it further comprises a self-learning start switch and an adaptive start switch, wherein the self-learning start switch and the adaptive start switch are connected to the single-chip processor, and the self-learning start switch is used to control the start of the self-learning mode;

[0029] After starting the self-learning mode, receiving information of multiple signal waveforms, performing data statistics on the information of the multiple signal waveforms, and establishing a signal model;

[0030] After starting the adaptive start switch, the matching times of different signal models are counted, and the signal model that reaches the preset matching times or the preset matching times ratio is found out, which is defined as the priority model. In the subsequent matching process, the priority model is given priority for matching comparison.

[0031] In a preferred embodiment, it also includes an output level switch for configuring the vibration sensing device to be in a normally open mode and a normally closed mode, and the output level switch is connected to the single-chip processor.

[0032] In a preferred embodiment, it also includes a working mode configuration switch for switching working modes in different working scenarios, and the working mode configuration switch is connected to the single-chip processor.

[0033] In a preferred embodiment, it also includes a signal voltage comparison circuit for converting the peak analog signal of the vibration signal into a square wave signal, and the signal voltage comparison circuit is arranged between the vibration sensing element and the single-chip processor.

[0034] In a preferred embodiment, the process of comparing the vibration signal with the signal model by the single chip processor includes:

[0035] Record the time width of each pulse of the square wave signal converted from the peak analog signal of the vibration signal, the time interval between adjacent pulses and the number of pulses;

[0036] A regression calculation is performed on the time width of each pulse of the square wave signal and the time interval between adjacent pulses. When the time width of each pulse of the square wave signal and the time interval between adjacent pulses are within a preset range, they are adjusted to preset values, and a shaped signal is generated based on the adjusted preset values. The shaped signal is compared with the signal model. When the shaped signal matches the signal model, the vibration signal is judged to be a valid signal. When the shaped signal does not match the signal model, the vibration signal is judged to be an invalid signal.

[0037] In a preferred embodiment, before comparing the vibration signal with the signal model, the single-chip processor also performs the following steps: identifying segments in the square wave signal where the pulse intensity is too high or too low, defining them as suspected superposition segments, and cutting the square wave signal based on the two end points of the suspected superposition segments to form partitions, extending the waveform of the partitions, with the extension time being equal to the duration of the suspected superposition segments, to form a simulated signal, and comparing the simulated signal with the signal model.

[0038] Compared with the prior art, the present invention realizes vibration sensing through a vibration sensing element, uses a single-chip processor to process data, and uses a sensitivity adjustment setting switch to adjust the sensitivity of the processor. During the working process, the vibration signal is first sensed by the vibration sensing element, and the vibration signal is sent to the single-chip processor. The single-chip processor stores a trained signal model, that is, a signal model that meets the characteristics of an effective signal. The vibration signal is compared with the signal model. When the vibration signal matches the signal model, the vibration signal is judged to be a valid signal. When the vibration signal does not match the signal model, the vibration signal is judged to be an invalid signal, thereby achieving stable vibration signal recognition, avoiding the influence of irrelevant vibration signals, and making it have good stability. At the same time, the sensitivity adjustment is achieved through the sensitivity adjustment setting switch. The sensitivity adjustment is mainly to adjust the similarity between the vibration signal and the signal model when matching. When the sensitivity requirement is high, the similarity between the vibration signal and the signal model is set to be high. When the sensitivity requirement is low, the similarity between the vibration signal and the signal model is set to be low, so that it has excellent adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 The invention relates to a structural schematic diagram of a novel self-learning and self-adaptive vibration sensing device.

[0040] Figure 2 It is a waveform extension circuit.

[0041] Figure 3 It is a schematic diagram of waveform comparison at each stage of waveform extension.

[0042] Figure 4 It is a schematic diagram of the signal model.

[0043] Figure 5 It is a schematic diagram of a square wave signal, a simulated signal 1, and a simulated signal 2 that may be suspected of being superimposed.

[0044] Vibration sensing element 1; single-chip processor 2; sensitivity adjustment setting switch 3; signal matching result display circuit 4; data storage circuit 5; lightning protection and electrostatic protection circuit 6; signal filtering circuit 7; waveform extension circuit 8; self-learning start switch 9; adaptive start switch 10; output level switch 11; working mode configuration switch 12; signal voltage comparison circuit 13; suspected superposition section 14; partition section one 15; partition section two 16; simulated signal one 17; simulated signal two 18. DETAILED DESCRIPTION

[0045] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0046] This specific embodiment is merely an explanation of the present invention and is not a limitation of the present invention. After reading this specification, those skilled in the art may make non-creative modifications to the present embodiment as needed. However, as long as they are within the scope of the claims of the present invention, they are protected by the patent law.

[0047] like Figure 1 As shown, a novel self-learning and self-adaptive vibration sensing device comprises:

[0048] A vibration sensing element 1 is used to sense vibration and transmit the sensed vibration signal to a single chip processor 2;

[0049] The single-chip processor 2 is implemented by a domestic single-chip STC8H8K64U and is used for data processing. The data processing process includes receiving the vibration signal transmitted by the vibration sensing element 1 and comparing the vibration signal with the signal model. When the vibration signal matches the signal model, the vibration signal is judged to be a valid signal. When the vibration signal does not match the signal model, the vibration signal is judged to be an invalid signal.

[0050] The sensitivity adjustment setting switch 3 is used to receive a sensitivity adjustment instruction and transmit a sensitivity adjustment signal to the single-chip processor 2 according to the instruction, so that the single-chip processor 2 can adjust the sensitivity;

[0051] The signal matching result display circuit 4 is used to receive the data information sent by the single-chip processor 2 and display the data information, the data information includes parameters such as the pulse width, number of pulses and pulse interval time of the signal, and the matched preset signal model code;

[0052] Data storage circuit 5, implemented by W25Q32JVSSIQ SPI NOR Flash, is used to store data, specifically, to store signal model data established after self-learning, measurement waveform data, statistical data and configuration information, etc.;

[0053] The vibration sensing element 1 , the sensitivity adjustment setting switch 3 , the signal matching result display circuit 4 and the data storage circuit 5 are electrically connected to the single chip processor 2 respectively.

[0054] A self-learning and self-adaptive novel vibration sensing device of this embodiment realizes vibration sensing through a vibration sensing element 1, uses a single-chip processor to measure, count and analyze signals, and uses a sensitivity adjustment setting switch 3 to adjust the sensitivity of the processor. During the working process, the vibration signal is first sensed by the vibration sensing element 1, and the vibration signal is sent to the single-chip processor 2. The single-chip processor 2 stores a trained signal model, that is, a signal model that meets the characteristics of an effective signal. The vibration signal is compared with the signal model. When the vibration signal matches the signal model, the vibration signal is judged to be a valid signal. When the vibration signal does not match the signal model, the vibration signal is judged to be an invalid signal, thereby achieving stable vibration signal recognition, avoiding the influence of irrelevant vibration signals, and making it have good stability. At the same time, the sensitivity adjustment is achieved by setting the sensitivity adjustment switch 3. The sensitivity adjustment is mainly to adjust the similarity between the vibration signal and the signal model when matching. When the sensitivity requirement is high, the similarity between the vibration signal and the signal model is set to be high. When the sensitivity requirement is low, the similarity between the vibration signal and the signal model is set to be low, so that it has excellent adaptability.

[0055] Furthermore, the vibration sensing element 1 uses the existing mature vibration sensing element 1 as the basic trigger source to establish a whitelist and a blacklist of signal waveforms. During use, for signals that meet the whitelist, the vibration sensing device outputs the correct electrical signal. For signals that meet the blacklist, the system performs statistics and gives an assessment of the severity of the use environment. For situations that are neither on the whitelist nor on the blacklist, the system establishes statistical data for further model research and analysis to guide on-site commissioning personnel to make correct configuration settings.

[0056] Furthermore, for signals that are neither in the whitelist nor in the blacklist, the following process is performed:

[0057] The data characteristics of the collected signal include the number of pulses, pulse width, pulse interval and occurrence time, but are not limited to these data characteristics;

[0058] Statistical data characteristics time law, and according to the time law, eliminate interference sources;

[0059] Statistically analyze the number of pulses, pulse width, pulse interval and occurrence time generated by each trigger, and use a similarity comparison algorithm to infer possible underreporting; combine the underreporting feedback from actual use to verify the authenticity of the underreporting;

[0060] Combined with the judgment result of the authenticity of the underreporting, the whitelist parameters are updated according to the similarity comparison algorithm for confirmation by the on-site debugging personnel.

[0061] The above process replaces the tedious manual debugging process, saves a lot of labor costs, and makes the data results more comprehensive and reliable, that is, by running data to calculate more accurate parameter correction data, the purpose of making the equipment more useful the more it is used is achieved.

[0062] The self-learning and self-adaptive novel vibration sensing device of this embodiment further includes a lightning protection and electrostatic protection circuit 6 for realizing lightning protection and electrostatic protection. The lightning protection and electrostatic protection circuit 6 is arranged between the single-chip processor 2 and the vibration sensing element 1, so as to adapt to the open field environment. The lightning protection and electrostatic protection circuit is implemented by a combination of a gas discharge tube and a TVS tube.

[0063] Furthermore, it also includes a signal filter circuit 7 for filtering the vibration signal. The signal filter circuit 7 is arranged between the single-chip processor 2 and the vibration sensing element 1. The electromagnetic interference is filtered by the signal filter circuit 7, thereby improving the stability of the device. The signal filter circuit 7 adopts a capacitor and inductor filter network, and the filter frequency is calculated and the appropriate parameters are selected to achieve it.

[0064] Furthermore, when the vibration sensing element 1 with high sensitivity vibrates, the pulse width of the oscillation waveform is only at the microsecond level, so a waveform extension circuit 8 is also provided to extend the vibration signal waveform in proportion so that the narrow signal becomes a wide signal. The waveform extension circuit 8 is provided between the single-chip processor 2 and the vibration sensing element 1, thereby improving the measurement accuracy of the single-chip processor 2. The signal filtering circuit 7 adopts a capacitor-inductor filtering network, and the filtering frequency is calculated and the appropriate parameters are selected to achieve it.

[0065] Furthermore, the purpose of waveform stretching is to proportionally amplify the pulse width of a sharp narrow pulse that is difficult to accurately measure in a low-cost manner, and to achieve measurement of a sharp narrow pulse by measuring and calculating a wide pulse.

[0066] like Figure 2As shown, the waveform extension circuit is realized by diode D1, energy storage inductor L1, charging capacitor C1, and discharge resistor R1. Diode D1 uses the unidirectional conduction characteristics of the diode. The peak narrow pulse input from terminal JP1 can charge L1 and C1 through diode D1, preventing the charge on L1 and C1 from discharging in the opposite direction through the input terminal. The A- pin of voltage comparator U1 is high-impedance input hardware with a very large internal resistance, so the discharge mainly depends on resistor R1. The size of resistor R1 determines the speed of discharge.

[0067] The key point of the waveform extension circuit is to make the discharge time mainly depend on the resistance value of resistor R1. The reverse high resistance of D1 and the high resistance design of U1 input pin strengthen the main discharge function of resistor R1.

[0068] The resistance value of resistor R1 can adjust the pulse width amplification degree.

[0069] pass Figure 3 Waveform comparison, waveform extension is actually the extension of the discharge waveform width. The size of the comparison voltage can also affect the width of the square wave. Figure 3 In the figure, (1) is a schematic diagram of a narrow pulse, and (2) is a schematic diagram of a Figure 2 The waveform after the charge and discharge network of D1, L1, C1, and R1, (3) is the square wave after voltage comparison and shaping by the comparator.

[0070] Furthermore, it also includes a self-learning start switch 9 and an adaptive start switch 10, which are used to set the self-learning mode, the adaptive working mode and the normal working mode. The self-learning start switch 9 and the adaptive start switch 10 are connected to the single-chip processor 2, and the self-learning start switch 9 is used to control the start of the self-learning mode.

[0071] After starting the self-learning mode, the information of multiple signal waveforms is received, and the information of multiple signal waveforms is statistically analyzed to establish a signal model. In the self-learning mode, the single-chip processor 2 measures and records the time width of each pulse of a series of signal waveforms (pulse time 1, pulse time 2, etc.), the time interval between adjacent pulses (pulse interval time 1, pulse interval time 2, etc.), and the number of pulses. The data is statistically analyzed multiple times to establish a basic model of the signal. Based on the basic model, the comparison signal model of the corresponding pulse time width, pulse interval time and pulse number with different similarities is calculated. In the working mode, the measured signal waveform is compared with the comparison signal model to determine whether the signal waveform is a valid waveform.

[0072] After starting the adaptive start switch 10, the number of matches of different signal models is counted, and the signal model that reaches the preset number of matches or the preset proportion of the number of matches is found, which is defined as the priority model. In the subsequent matching process, the priority model is given priority for matching and comparison. The main purpose of this is: in specific equipment products, the similarity of application scenarios is generally high, and a certain number of basic waveform models can be preset. In the adaptive mode, the most matching signal comparison waveform model is found by counting the number of matches of different basic waveform models. In the working mode, the system dynamically adjusts the sensitivity. The longer it runs, the more matching the model is, and the compatibility is sensitive and reliable. This makes the applicability more and more extensive.

[0073] In the process of finding the priority model, it can be calculated based on the number of matches. When a signal model is matched more than a predetermined number of times, it is defined as the priority model. It can also be calculated based on the proportion of matching times. When the proportion of the number of times a signal model is matched reaches a predetermined proportion of the total number of times, it is defined as the priority model.

[0074] Furthermore, it also includes an output level switch 11 for configuring the normally open mode and the normally closed mode of the vibration sensing device, which is compatible with the two level output modes currently on the market. The output level switch 11 is connected to the single-chip processor 2.

[0075] Furthermore, it also includes a working mode configuration switch 12, which is used to switch the working modes of different working scenarios. The working mode configuration switch 12 is connected to the single-chip processor 2. The switching of the working mode is achieved by adjusting the parameter configuration settings. Different parameter configurations are adapted to different application scenarios. The working mode configuration switch 12 can adopt a toggle switch.

[0076] Furthermore, it also includes a signal voltage comparison circuit 13, which is used to convert the peak analog signal of the vibration signal into a square wave signal for digital processing. The signal voltage comparison circuit 13 is arranged between the vibration sensing element 1 and the single-chip processor 2. The signal voltage comparison circuit 13 is designed based on the LM393 chip to form a signal voltage comparator.

[0077] Furthermore, the reference voltage of the comparator is controlled and adjusted by the single chip microcomputer, and the purpose of positively adjusting the sensitivity is achieved by dynamically controlling the reference voltage.

[0078] Furthermore, the signal voltage comparison circuit is used to eliminate the influence of the load factor on the output end of the charge and discharge circuit. The reason is that the output end of the charge and discharge circuit needs to meet the requirements of connecting a higher load resistor to eliminate the discharge influence caused by too small a load. The combined use of the voltage comparator circuit solves the multiple purposes of the load requirement on the output end of the charge and discharge circuit and converting the peak analog signal of the vibration signal into a square wave signal.

[0079] The process of comparing the vibration signal with the signal model by the single chip processor 2 includes:

[0080] S1, recording the time width of each pulse of the square wave signal converted from the peak analog signal of the vibration signal, the time interval between adjacent pulses and the number of pulses;

[0081] S2, performing regression calculation on the time width of each pulse of the square wave signal and the time interval between adjacent pulses, the process of regression calculation is: when the time width of each pulse of the square wave signal and the time interval between adjacent pulses are within a preset range, adjusting them to a preset value, and generating a shaped signal according to the adjusted preset value,

[0082] The shaped signal is compared with the signal model. When the shaped signal matches the signal model, the vibration signal is judged to be a valid signal. When the shaped signal does not match the signal model, the vibration signal is judged to be an invalid signal.

[0083] In fact, when adjusting the time width of each pulse of the square wave signal and the time interval between adjacent times, multiple adjustment processes may be involved. Specifically, corresponding to each signal model, a targeted adjustment value is set. For example, if there are signal model A, signal model B, signal model C, signal model D, signal model E and signal model F, then there are adjustment values ​​corresponding to signal model A, signal model B, signal model C, signal model D, signal model E and signal model F respectively.

[0084] In this embodiment, a signal model is used as an example for explanation. Figure 4, which has pulse width time 1, pulse width time 2 and pulse width time 3, has pulse interval time 1 and pulse interval time 2, pulse width time 1, pulse width time 2 and pulse width time 3 are 200ms, 100ms and 130ms respectively, pulse interval time 1 and pulse interval time 2 are 230ms and 390ms respectively. When processing the square wave signal, for this signal model, when the first pulse width time of the square wave signal is in the range of 190-210ms, its pulse width time is adjusted to 200ms, when the square wave When the second pulse width of the signal is within the range of 95-105ms, the pulse width is adjusted to 100ms. When the third pulse width of the square wave signal is within the range of 125-135ms, the pulse width is adjusted to 130ms. When the first pulse interval of the square wave signal is within the range of 220-240ms, the first pulse interval is adjusted to 230ms. When the second pulse interval of the square wave signal is within the range of 380-400ms, it is adjusted to 390ms. Through the above adjustment process, the square wave signal can be directly processed into a state that can be directly compared with the signal model. When the processed square wave signal is consistent with the signal model, it indicates that the signal is a valid signal for the signal model. When the processed square wave signal is inconsistent with the signal model, it indicates that the signal is an invalid signal for the signal model.

[0085] Through the above process, the difficulty of similarity judgment is simplified, and the square wave signal can be directly compared with the signal model after processing. At the same time, it makes it possible to realize direct parameter control similarity judgment, that is, by adjusting the inclusion range of pulse width time and pulse interval time, the sensitivity of the sensing device can be adjusted. When the sensitivity requirement is high, the range interval is set narrower, and when the sensitivity requirement is low, the range interval is set wider.

[0086] Furthermore, before comparing the vibration signal with the signal model, the single-chip processor 2 also performs the following steps: identifying the sections of the square wave signal with too high or too low pulse intensity, defining them as suspected superposition sections, and cutting the square wave signal based on the two end points of the suspected superposition sections to form partitions, extending the waveform of the partitions, and the extension time is equal to the duration of the suspected superposition sections to form simulated signals, and comparing the simulated signals with the signal model.

[0087] The presence of a section with too high or too low pulse intensity in the square wave signal indicates that waves may have been superimposed in this section, causing the pulse intensity to be enhanced or reduced. Too high or too low means that the pulse intensity of a certain section is higher or lower than the pulse intensity of other sections and exceeds a certain range.

[0088] In order to realize the splitting and identification of superposition situations, the above processing steps are specially set up, refer to Figure 5According to the above method, the square wave signal is cut and separated at both ends of the suspected superposition segment 14 to form two partitions, which are defined as partition one 15 and partition two 16. The waveform of the two partitions is extended, and the extension time is equal to the duration of a superposition segment to form a simulated signal one 17 and a simulated signal two 18. The dotted part in the figure is the extended part, and the simulated signal one 17 and the simulated signal two 18 are inferred waveform signals. When identifying the vibration signal, the square wave signal that has not been cut and the simulated signal one 17 and the simulated signal two 18 are matched with the signal model respectively, so that while splitting and identifying the possible superposition signal, the square wave signal that may be a valid signal will not be excluded, making the signal detection more complete and accurate.

[0089] In this embodiment, the stretching of the partition segments and the stretching process of the waveform stretching circuit are significantly different and should not be confused with each other.

[0090] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "include..." or "comprise..." do not exclude the existence of other elements in the process, method, article or terminal device including the elements. In addition, in this article, "greater than", "less than", "exceed" and the like are understood to exclude the number itself; "above", "below", "within" and the like are understood to include the number itself.

[0091] The above description of the embodiments is to facilitate the understanding and use of the present invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to the embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above embodiments. Improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope of the present invention should be within the protection scope of the present invention.

Claims

1. A new self-learning and self-adaptive vibration sensing device, characterized in that: include: A vibration sensing element, used for sensing vibration and transmitting the sensed vibration signal to the single chip processor; A single-chip processor is used for data processing, wherein the data processing process includes receiving a vibration signal transmitted by a vibration sensing element, and comparing the vibration signal with a signal model. When the vibration signal matches the signal model, the vibration signal is judged to be a valid signal, and when the vibration signal does not match the signal model, the vibration signal is judged to be an invalid signal; The sensitivity adjustment setting switch is used to receive a sensitivity adjustment instruction and transmit a sensitivity adjustment signal to the single-chip processor according to the instruction, so that the single-chip processor can adjust the sensitivity; The signal matching result display circuit is used to receive the data information sent by the single-chip processor and display the data information; A data storage circuit, used for storing data; Wherein, the vibration sensing element, the sensitivity adjustment setting switch, the signal matching result display circuit and the data storage circuit are electrically connected to the single chip processor respectively; It also includes a signal voltage comparison circuit, which is used to convert the peak analog signal of the vibration signal into a square wave signal, and the signal voltage comparison circuit is arranged between the vibration sensing element and the single-chip processor; The process of comparing the vibration signal with the signal model by the microcontroller processor includes: Record the time width of each pulse of the square wave signal converted from the peak analog signal of the vibration signal, the time interval between adjacent pulses and the number of pulses; A regression calculation is performed on the time width of each pulse of the square wave signal and the time interval between adjacent pulses. The process of the regression calculation is: when the time width of each pulse of the square wave signal and the time interval between adjacent pulses are within a preset range, they are adjusted to a preset value, and a shaped signal is generated according to the adjusted preset value; The shaped signal is compared with the signal model. When the shaped signal matches the signal model, the vibration signal is judged to be a valid signal. When the shaped signal does not match the signal model, the vibration signal is judged to be an invalid signal.

2. A self-learning and self-adaptive novel vibration sensing device according to claim 1, characterized in that: It also includes a lightning protection and electrostatic protection circuit for realizing lightning protection and electrostatic protection. The lightning protection and electrostatic protection circuit is arranged between the single-chip processor and the vibration sensing element.

3. A self-learning and self-adaptive novel vibration sensing device according to claim 1, characterized in that: It also includes a signal filtering circuit for filtering the vibration signal, and the signal filtering circuit is arranged between the single-chip processor and the vibration sensing element.

4. A self-learning and self-adaptive novel vibration sensing device according to claim 1, characterized in that: It also includes a waveform extension circuit, which is used to extend the vibration signal waveform in proportion so that a narrow signal becomes a wide signal. The waveform extension circuit is arranged between the single-chip processor and the vibration sensing element.

5. A self-learning and self-adaptive novel vibration sensing device according to claim 1, characterized in that: It also includes a self-learning start switch and an adaptive start switch, wherein the self-learning start switch and the adaptive start switch are connected to the single-chip processor, and the self-learning start switch is used to control the start of the self-learning mode; After starting the self-learning mode, receiving information of multiple signal waveforms, performing data statistics on the information of the multiple signal waveforms, and establishing a signal model; After starting the adaptive start switch, the matching times of different signal models are counted, and the signal model that reaches the preset matching times or the preset matching times ratio is found out, which is defined as the priority model. In the subsequent matching process, the priority model is given priority for matching comparison.

6. A self-learning and self-adaptive novel vibration sensing device according to claim 1, characterized in that: It also includes an output level switch for configuring a normally open mode and a normally closed mode of the vibration sensing device, and the output level switch is connected to the single-chip processor.

7. A novel self-learning and self-adaptive vibration sensing device according to claim 1, characterized in that: It also includes a working mode configuration switch for switching working modes in different working scenarios, and the working mode configuration switch is connected to the single-chip processor.

8. A self-learning and self-adaptive novel vibration sensing device according to claim 1, characterized in that: Before comparing the vibration signal with the signal model, the microcontroller processor also performs the following steps: identifying the sections of the square wave signal with too high or too low pulse intensity, defining them as suspected superposition sections, and cutting the square wave signal based on the two end points of the suspected superposition sections to form partitions, extending the waveform of the partitions, and the extension time is equal to the duration of the suspected superposition sections to form a simulated signal, and comparing the simulated signal with the signal model.

Images