Alarm method

Abstract

The invention provides an alarm method comprising the following steps: filtering a vibration signal to obtain a first noise signal; judging whether the maximum amplitude of the first noise signal is greater than a preset amplitude threshold, if not, restarting to execute the step a, and if so, recording the first noise signal and acquiring a vibration signal near the first noise signal as a second noise signal; dividing the second noise signal into a plurality of frames of noise signals, performing Fourier transform on each frame of the plurality of frames of noise signals to obtain a plurality of energy spectrums, and performing frequency band division correction on the plurality of energy spectrums to obtain a plurality of corrected energy spectrums; calculating the proportion of the energy of the plurality of frequency components in each corrected energy spectrum in the whole frequency spectrum to obtain a plurality of probability values of the plurality of frequency components, and calculating the spectrum entropy corresponding to each frame through the plurality of probability values; and calculating the average Euclidean distance between the suspected spectral entropy curve and the plurality of standard spectral entropy curves, and if the maximum value of the average Euclidean distance is smaller than a preset distance threshold value, giving an alarm.

Description

Technical Field

[0001] The invention mainly relates to a loose component monitoring system, and in particular to an alarm method. Background Art

[0002] The nuclear power plant system is a highly complex structural system with important safety requirements. The safety of nuclear power plant operation is one of the primary issues in the field of nuclear engineering. Loose parts are often found in the loop coolant system of nuclear power plants. The source is the loosening or falling of parts caused by operation, or it may be parts, tools and other objects left in the system during construction, material replacement or maintenance. There are a large number of parts in the loop system of nuclear power plants that can cause hidden faults. Some parts can be loose or detached from the body. Once this happens, serious faults and accidents in the operation of nuclear power plants may occur, and may cause the shutdown of nuclear power plant reactors. Therefore, the monitoring of loose parts in nuclear power plants is particularly important. The loose parts monitoring system (LPMS) is one of the basic diagnostic tools for nuclear power plant fault detection. It is mainly used for monitoring loose parts in the loop coolant system of nuclear power plants. Its purpose is to detect and locate loose parts or falling parts that may exist in the loop, and issue alarms in time to provide information for possible subsequent operations, thereby ensuring the safety and reliability of nuclear power plant operation.

[0003] The most important function of the loose part monitoring system (LPMS) is to quickly and accurately detect the falling of loose parts and issue an alarm. The design of the alarm algorithm should minimize missed alarms and also reduce the probability of false alarms. There are generally two reasons for missed alarms and false alarms. One is that the energy of the impact signal is too low and is submerged in the background noise. According to a large number of experiments, the spectrum of the loose part impact signal is mainly distributed in the range of 1kHz to 25kHz. Therefore, bandpass filtering (1kHz~25kHz) of the collected impact signal can effectively eliminate most of the background noise. The frequency of the background noise in a nuclear power plant generally does not exceed 6kHz, so the spectrum of the background noise and the collision signal has a large overlap, especially between 3 and 5KHz. The currently commonly used spectrum filtering method is difficult to separate the background noise and the collision signal. When the amplitude of the noise exceeds the preset value, a false alarm is generated, or the effective loose part monitoring signal is submerged in the noise, resulting in missed alarms. The second reason is the interference signals generated by non-falling parts, such as mechanical and electrical noise, large-amplitude electrical transient signals, thermal shock vibration signals and friction vibration signals. These interference signals have the characteristics of large amplitude and short duration like the impact signals of loose parts. In particular, the acceleration time history curves of thermal shock vibration signals and friction vibration signals are very similar to those of loose parts impact signals. Therefore, these interference signals often trigger false alarms. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an alarm method for reducing missed alarms and false alarms in a loose component monitoring system of a nuclear power plant.

[0005] In order to solve the above technical problems, the present invention provides an alarm method, which is suitable for monitoring the looseness of internal components of a device. A plurality of acceleration sensors are installed inside the device to obtain vibration signals of corresponding positions in real time. The alarm method comprises the following steps: a. filtering the vibration signal to obtain a first noise signal; b. judging whether the maximum amplitude of the first noise signal is greater than a preset amplitude threshold, if not, restarting step a, and if so, recording the first noise signal and obtaining a vibration signal near the first noise signal as a second noise signal; c. dividing the second noise signal into multiple frames of noise signals, performing Fourier transform on each frame of the multiple frames of noise signals to obtain multiple energy spectra, and performing frequency band correction on the multiple energy spectra to obtain multiple corrected energy spectra; d. calculating the proportion of the energy of multiple frequency components in each corrected energy spectrum to the entire frequency spectrum, obtaining multiple probability values ​​of the multiple frequency components, and calculating the spectral entropy corresponding to each frame through the multiple probability values; e. normalizing the multiple spectral entropies to obtain a suspected spectral entropy curve, calculating the average Euclidean distance between the suspected spectral entropy curve and the multiple standard spectral entropy curves, and alarming if the maximum value of the average Euclidean distance is less than the preset distance threshold.

[0006] In one embodiment of the present invention, the formula for performing frequency band correction on multiple energy spectra to obtain multiple corrected energy spectra is: Among them, Y′ m (f i ) is the corrected energy spectrum after correction, Y m (f i ) is the uncorrected energy spectrum, Y m (f i )’s average value.

[0007] In one embodiment of the present invention, the alarm method further includes optimizing the suspected spectral entropy curve and the multiple standard spectral entropy curves using a dynamic time warping method before calculating the average Euclidean distance between the suspected spectral entropy curve and the multiple standard spectral entropy curves.

[0008] In one embodiment of the present invention, obtaining a vibration signal near a first noise signal as a second noise signal includes the following steps: recording the moment when the maximum amplitude of the first noise signal is greater than a preset amplitude threshold as the center moment; and intercepting signal data in a first time period before the center moment and in a second time period after the center moment as the second noise signal.

[0009] In an embodiment of the present invention, the duration of the second time period is greater than or equal to 50 ms.

[0010] In one embodiment of the present invention, the alarm method also includes: selecting loose parts to knock on different positions of the inner wall of the equipment under different operating conditions of the equipment to obtain multiple standard signals; and obtaining a preset amplitude threshold, a standard spectral entropy curve and a preset distance threshold through multiple standard signals.

[0011] In one embodiment of the present invention, obtaining a preset amplitude threshold through a standard signal includes the following steps: filtering a plurality of standard signals to obtain a plurality of first analog signals, extracting the maximum amplitude of the plurality of first analog signals Record multiple background noise signals of the device at different power levels through an acceleration sensor, filter the multiple background noise signals to obtain multiple second analog signals, and extract the maximum amplitude of the multiple second analog signals And a preset amplitude threshold is obtained through the maximum amplitudes of the plurality of first analog signals and the maximum amplitudes of the plurality of second analog signals.

[0012] In one embodiment of the present invention, obtaining a standard spectral entropy curve through a standard signal includes the following steps: f. filtering the standard signal to obtain a first standard signal; g. determining whether the maximum amplitude of the first standard signal is greater than a preset amplitude threshold, if not, restarting step a; if so, recording the first standard signal and obtaining a vibration signal near the first standard signal as a second standard signal; h. dividing the second standard signal into multiple frames of standard signals, performing Fourier transform on each frame of the multiple frames of standard signals to obtain multiple energy spectra, and performing frequency band correction on the multiple energy spectra to obtain multiple corrected energy spectra; i. calculating the proportion of the energy of multiple frequency components in each corrected energy spectrum to the entire frequency spectrum, obtaining multiple probability values ​​of the multiple frequency components, and calculating the spectral entropy corresponding to each frame through the multiple probability values; j. normalizing the multiple spectral entropies to obtain a standard spectral entropy curve.

[0013] In one embodiment of the present invention, obtaining the preset distance threshold through the standard signal includes: calculating the average Euclidean distance d between the spectral entropy curves of any two frames of the standard signal ij , if there exists 0<α≤0.20 such that P{d ij >d α (n)}=α, then d α (n) is the preset amplitude threshold; where d α (n) Satisfaction

[0014] In one embodiment of the present invention, the formula for calculating the average Euclidean distance is: Among them, x′ i1 , x′ i2 ,…,x′ inThese are multiple spectral entropy curves optimized by dynamic time warping method.

[0015] Compared with the prior art, the alarm method provided by the present invention first determines whether the amplitude of the vibration signal exceeds the amplitude threshold, realizes real-time and efficient capture of abnormal signals, avoids missed alarms, and then uses the method of increasing weights to obtain a modified energy spectrum for the energy spectrum of the 6kHz-15kHz frequency band, which can exclude the interference of information outside the frequency band, and can enhance the ability of the spectrum entropy curve to distinguish vibration signals from background noise signals. In addition, the signal processing method based on signal spectrum entropy can detect collision signals with low signal-to-noise ratio, and can exclude the interference of pulse signals such as electrical interference, and the method for determining the preset distance threshold takes into account the influence of the circuit environment, the type of knocking object, the knocking speed, the operating conditions, etc. on the spectrum entropy curve, which helps to improve the accuracy of identifying the impact signal of loose parts through the signal spectrum entropy curve. The Euclidean distance can quantitatively compare the similarity of two time series, and combined with the classification threshold at a certain confidence level determined based on the average Euclidean distance probability density function, it can accurately identify the spectrum entropy curve corresponding to the impact signal of the loose parts, and improve the accuracy of the alarm judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are included to provide a further understanding of the present application, and they are included and constitute a part of the present application. The accompanying drawings illustrate embodiments of the present application and together with the present specification serve to explain the principles of the present invention. In the accompanying drawings:

[0017] Figure 1 It is a flow chart of an alarm method of the present application.

[0018] Figure 2 It is a waveform diagram of a vibration signal of the present application.

[0019] Figure 3 This application is Figure 2 The spectral entropy curve of the uncorrected energy spectrum of the vibration signal is shown.

[0020] Figure 4-6 This application is Figure 2 The spectral entropy curve of the corrected energy spectrum of the vibration signal under different correction parameters is shown. DETAILED DESCRIPTION

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0022] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0023] Unless otherwise specifically stated, the relative arrangement, numerical expressions and numerical values ​​of the parts and steps set forth in these embodiments do not limit the scope of the present application. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to those of ordinary skill in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be considered as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so that once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0024] In the description of the present application, it should be understood that the directions or positional relationships indicated by directional words such as "front, back, up, down, left, right", "lateral, vertical, perpendicular, horizontal" and "top, bottom" are usually based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present application and simplifying the description. Unless otherwise specified, these directional words do not indicate or imply that the device or element referred to must have a specific direction or be constructed and operated in a specific direction, and therefore cannot be understood as limiting the scope of protection of the present application; the directional words "inside and outside" refer to the inside and outside relative to the contours of each component itself.

[0025] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used here to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatially relative descriptions used here are interpreted accordingly.

[0026] In addition, it should be noted that the use of words such as "first" and "second" to define components is only for the convenience of distinguishing the corresponding components. If not otherwise stated, the above words have no special meaning and cannot be understood as limiting the scope of protection of this application. In addition, although the terms used in this application are selected from well-known and commonly used terms, some terms mentioned in the specification of this application may be selected by the applicant at his or her discretion, and their detailed meanings are explained in the relevant parts of the description of this article. In addition, it is required to understand this application not only by the actual terms used, but also by the meaning implied by each term.

[0027] It should be understood that when a component is referred to as being “on another component,” “connected to another component,” “coupled to another component,” or “contacting another component,” it can be directly on, connected to, coupled to, or contacting the other component, or intervening components may be present. In contrast, when a component is referred to as being “directly on another component,” “directly connected to,” “directly coupled to,” or “directly contacting” another component, there are no intervening components present.

[0028] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously. At the same time, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0029] Figure 1 This is a flowchart of an alarm method of the present application. Figure 1As shown, the present invention provides an alarm method suitable for monitoring the looseness of internal components of a device. A plurality of acceleration sensors are installed inside the device to obtain vibration signals of corresponding positions in real time. The alarm method 10 includes the following steps:

[0030] S11: Filter the vibration signal to obtain a first noise signal.

[0031] S12: Determine whether the maximum amplitude of the first noise signal is greater than a preset amplitude threshold. If not, restart step S11. If yes, execute step S13.

[0032] S13: Record the first noise signal and obtain a vibration signal near the first noise signal as a second noise signal.

[0033] S14: dividing the second noise signal into multiple frames of noise signals, performing Fourier transform on each frame of the multiple frames of noise signals to obtain multiple energy spectra, and performing frequency band correction on the multiple energy spectra to obtain multiple corrected energy spectra.

[0034] S15: Calculate the proportion of the energy of multiple frequency components in each corrected energy spectrum to the entire spectrum, obtain multiple probability values ​​of the multiple frequency components, and obtain the spectrum entropy corresponding to each frame through the calculation of the multiple probability values.

[0035] S16: performing normalization processing on the multiple spectral entropies to obtain a suspected spectral entropy curve, and calculating an average Euclidean distance between the suspected spectral entropy curve and the multiple standard spectral entropy curves.

[0036] S17: Determine whether the maximum value of the average Euclidean distance is less than a preset distance threshold. If not, restart step S11. If yes, generate an alarm.

[0037] Specifically, before executing the alarm method 10, it is necessary to install multiple acceleration sensors inside the device to obtain the vibration signals of the corresponding positions in real time. In step S11, the signals obtained by multiple acceleration sensors in real time are filtered to eliminate the noise of some frequency bands, realize the initial screening of background noise, and avoid false alarms. Specifically, according to a large number of experiments, when there are parts loose inside the device, the frequency spectrum of the loose parts impact signal (hereinafter referred to as the standard signal) is mainly distributed in the range of 1kHz to 25kHz. Therefore, in the specific implementation, the collected impact signal is band-pass filtered at 1kHz to 25kHz to effectively eliminate most of the background noise. However, the background noise will not exceed 6kHz, and the spectrum of the background noise and the standard signal has a large overlap, especially between 3 and 5KHz. It is difficult to separate the background noise and the standard signal using a general spectrum filtering method, so a subsequent algorithm is required to further extract the signal.

[0038] In step S12, it is necessary to determine whether the maximum amplitude of the first noise signal after filtering is greater than the preset amplitude threshold. If the maximum amplitude does not exceed the preset amplitude threshold, it means that no standard signal appears at this time, that is, no parts are loose inside the equipment, and no alarm is required. Then return to step S11 to continue monitoring. If the maximum amplitude of the first noise signal is greater than the preset amplitude threshold, it is determined that the first noise signal is abnormal, and parts may be loose. The first noise signal is recorded for further processing.

[0039] In one embodiment of the present invention, obtaining the vibration signal near the first noise signal as the second noise signal in step S13 includes: recording the moment when the maximum amplitude of the first noise signal is greater than the preset amplitude threshold as the center moment, and intercepting the signal data in the first time period before the center moment and the second time period after the center moment as the second noise signal. It can be understood that when there is a loose component inside the device and an impact occurs, for the standard signal, its waveform usually appears suddenly at the impact moment and lasts for a period of time, so the intercepted time range before the center moment is less than the part after the center moment, that is, the first time period before the center moment is basically a background noise signal, and the waveform of the impact signal is generally located in the second time period after the center moment, so in some common embodiments of the present invention, the duration of the first time period is usually less than the duration of the second time period, and the signal interception method can completely preserve the waveform curve of the impact signal, and can make the overall scale and recording time of all intercepted signal segments consistent, which is convenient for the subsequent calculation of the Euclidean distance based on the dynamic time warping method, and is convenient for calculating the arrival time difference of the recorded signals of different channels when performing the loose component positioning algorithm. In order to further ensure that the waveform curve of the impact signal can be completely preserved, in a preferred embodiment, the duration of the second time period is greater than or equal to 50 ms.

[0040] In a preferred embodiment, 100ms of signal data is intercepted around the central moment exceeding the preset amplitude threshold, and the interception time range is 30ms before the central moment to 70ms after the central moment, that is, the first time period is 30ms, the second time period is 70ms, and the second noise signal is 100ms in length.

[0041] In step S14, the intercepted second noise signal is framed, each frame signal is Fourier transformed to obtain multiple energy spectra, and the multiple energy spectra are corrected by frequency bands to obtain multiple corrected energy spectra. In a specific embodiment, the total number of frames is M, and each frame signal is Fourier transformed to obtain an energy spectrum Y m (f i), where m = 1, 2, 3 ... M and i = 1, 2, 3 ... N, N is the total number of frequency band components. That is, in a preferred embodiment, the second noise signal is divided into M frames, and the M frame signals are respectively subjected to Fourier transform to obtain M energy spectra, and then the M energy spectra are corrected into N frequency bands to obtain multiple corrected energy spectra.

[0042] Specifically, in one embodiment, the formula for performing frequency band correction on multiple energy spectra to obtain multiple corrected energy spectra is: Among them, Y′ m (f i ) is the corrected energy spectrum after correction, Y m (f i ) is the uncorrected energy spectrum, Y m (f i ). In simple terms, this process uses the method of increasing weights to obtain the corrected energy spectrum Y′ for the energy spectrum outside the 6kHz-15kHz frequency band. m (f i ), the original energy spectrum is kept unchanged within the frequency band, and a larger number is assigned to the energy spectrum outside the 6kHz-15kHz frequency band. In specific implementation, the energy spectrum outside the 6kHz-15kHz frequency band is assigned a value of in is the energy spectrum Y m (f i )’s average value.

[0043] Figure 2 is a waveform diagram of a vibration signal of the present application, Figure 3 This application is Figure 2 The spectral entropy curve of the uncorrected energy spectrum of the vibration signal is shown. Figure 4-6 This application is Figure 2 The spectral entropy curve of the corrected energy spectrum of the vibration signal under different correction parameters is shown. Figure 2-6 As shown, it is obvious that the spectral entropy curve calculated by using the modified energy spectrum can effectively eliminate the interference of information outside the frequency band, and can significantly enhance the ability of the spectral entropy curve to distinguish the second noise signal (i.e., the suspected impact signal) from the background noise signal.

[0044] In step S15, for each frame of the modified energy spectrum, the proportion of the energy of each frequency component in the entire spectrum is calculated to obtain the probability value of each frequency component. Then substitute the probability value into the spectral entropy calculation formula Get the spectral entropy corresponding to each frame. Generally speaking, the spectrum of the impact signal is much wider than the spectrum of the background noise signal. The spectrum of the background noise signal is generally not higher than 6kHz, while the frequency band of the impact signal is very wide at the beginning, and the frequency band gradually narrows as the signal attenuates. Therefore, when the signal-to-noise ratio is very low, the signal-to-noise ratio is still very high in the high frequency band. According to this principle, it is possible to determine whether a collision signal appears by calculating the spectral entropy of the signal in the high frequency band. Specifically, the spectral entropy of the impact signal has the following characteristics: the spectral entropy of the vibration signal is different from the spectral entropy of the background noise. As long as the spectrum remains unchanged, the spectral entropy will not change. Generally, the spectral entropy of a flat spectrum is relatively low, and when an impact signal appears, the spectral entropy will decrease.

[0045] In step S16, the spectral entropy curve obtained in step S15 is normalized to obtain a suspected spectral entropy curve, which can be specifically expressed by the formula: Among them, H normal is the suspected spectral entropy, H is the spectral entropy without normalization, min(H) is the minimum value of the spectral entropy curve, and max(H) is the maximum value of the spectral entropy curve

[0046] In one embodiment of the present invention, the alarm method further includes optimizing the suspected spectral entropy curve and the multiple standard spectral entropy curves using a dynamic time warping method before calculating the average Euclidean distance between the suspected spectral entropy curve and the multiple standard spectral entropy curves, and then calculating the average Euclidean distance between the suspected spectral entropy curve and the multiple standard spectral entropy curves.

[0047] Specifically, the dynamic time warping algorithm is an algorithm that calculates the similarity between two sets of time series data. The core idea of ​​the algorithm is to establish a nonlinear correspondence between the time axes of the two sets of data. Since the length of the sequence fragments representing the same meaning may be different each time the same event occurs, the two sequences will not completely match. In this case, directly using the Euclidean distance to calculate the similarity is prone to incorrect recognition results. The dynamic time adjustment algorithm stretches or compresses the time series and uses dynamic programming to make it as similar as possible, and on this basis calculates the similarity between the two time series.

[0048] Since there is a time difference between the impact signals of loose parts received by acceleration sensors at different positions, the suspected spectral entropy curve and the standard spectral entropy curve obtained by processing the vibration signals intercepted by different channels in steps S11-S16 are similar in overall shape, but are not aligned on the time axis. The dynamic time warping algorithm extends, shortens, and translates the time series to calculate the similarity between the two time series.

[0049] Taking the suspected spectral entropy Q and the standard spectral entropy C as examples, the specific explanation is as follows:

[0050] The lengths of the suspected spectral entropy Q and the standard spectral entropy C are n and m respectively, Q = [q1, q2, ..., q i ,…,q n ],C=[c1,c2,…,c j ,…,c m ], in order to align the two sequences, we need to construct an n×m matrix grid and use the Euclidean distance d(q i ,c j )=(q i -c j ) 2 Indicates q i and c j The distance between two points, dynamic time warping can be reduced to finding a path through several grid points in this grid. The grid points passed by the path are the alignment points of the two sequences. The best path is the path that minimizes the accumulated distance along the path. This path can be obtained through a dynamic programming algorithm. The best aligned sequence [x′] after stretching, compression and translation by the above dynamic time warping method is i1 , x′ i2 ,…,x′ in ] and [x′ j1 , x′ j2 ,…,x′ jn The average Euclidean distance of ] is:

[0051]

[0052] In one embodiment of the present invention, the alarm method further includes: selecting loose parts to knock different positions of the inner wall of the equipment under different operating conditions of the equipment to obtain multiple standard signals; and obtaining a preset amplitude threshold, a preset probability ratio threshold and a preset correlation coefficient threshold through multiple standard signals. Specifically, obtaining multiple standard signals is achieved through simulation experiments: deploying a loose parts monitoring system on a simulated loop test bench or an actual reactor loop, selecting a number of steel balls, bolts, pins, etc. within a certain mass range to represent loose parts in a loop, and knocking different positions of the container wall of the main equipment of the loop at a speed of not less than 0.68J of impact energy by free fall, simple pendulum, etc. under different operating conditions of the loop test bench, recording the impact signal of the loose parts as a standard signal, and further processing based on the standard signal to obtain preset data such as a preset amplitude threshold, a standard spectral entropy curve and a preset distance threshold.

[0053] In one embodiment of the present invention, obtaining a preset amplitude threshold through a standard signal includes the following steps: filtering a plurality of standard signals to obtain a plurality of first analog signals, extracting the maximum amplitude of the plurality of first analog signals Record multiple background noise signals of the device at different power levels through an acceleration sensor, filter the multiple background noise signals to obtain multiple second analog signals, and extract the maximum amplitude of the multiple second analog signals And a preset amplitude threshold is obtained through the maximum amplitudes of the plurality of first analog signals and the maximum amplitudes of the plurality of second analog signals.

[0054] The standard spectral entropy curve takes into account the influence of the circuit environment, the type of knocking object, the knocking speed, the operating conditions, etc. on the spectral entropy curve, which helps to improve the accuracy of identifying the impact signal of the loose component through the signal spectral entropy curve. In one embodiment of the present invention, the standard spectral entropy curve is obtained by using the standard signal, that is, the standard signal is used as the vibration signal to perform the following Figure 1 The steps S11-S16 shown specifically include the following steps:

[0055] (1) filtering the standard signal to obtain a first standard signal;

[0056] (2) determining whether the maximum amplitude of the first standard signal is greater than a preset amplitude threshold, if not, restarting step (1), and if so, recording the first standard signal and obtaining a vibration signal near the first standard signal as a second standard signal;

[0057] (3) dividing the second standard signal into multiple frames of standard signals, performing Fourier transform on each frame of the multiple frames of standard signals to obtain multiple energy spectra, and performing frequency band correction on the multiple energy spectra to obtain multiple corrected energy spectra;

[0058] (4) calculating the proportion of the energy of multiple frequency components in each corrected energy spectrum to the entire spectrum, obtaining multiple probability values ​​of the multiple frequency components, and obtaining the spectral entropy corresponding to each frame through the calculation of the multiple probability values;

[0059] (5) Multiple spectral entropies are normalized to obtain a standard spectral entropy curve.

[0060] Furthermore, in this embodiment, obtaining the preset distance threshold through the standard signal is also performed based on the standard spectral entropy curve of the standard signal. Specifically, according to the above steps (1)-(5), a total of m groups of signals [x i1 ,x i2 ,…,x in ], i = 1, 2, ... m, based on the dynamic time warping method, the spectral entropy curve of any two groups of signals is calculated, that is, the average Euclidean distance d between the i-th group of data and the j-th group of data ij , the formula can be preferably used: Among them, x′ i1 , x′ i2 ,…,x′ in These are multiple spectral entropy curves optimized by dynamic time warping method.

[0061] Furthermore, the calculation of the preset distance threshold requires a given α, 0<α≤0.20. If there is P{d ij >d α (n)}=α, then d α (n) is the supremum of the threshold when the longest distance method and the shortest distance method are used for classification at the α level, that is, the preset distance threshold. α The value of (n) can be calculated based on the average Euclidean distance d ij The distribution function is determined by:

[0062] Finally, in step S17, it is determined whether the maximum value of the average Euclidean distance is less than a preset distance threshold. If not, it is considered that no collision has occurred and step S11 is restarted. If so, an alarm is issued.

[0063] The method for determining the preset distance threshold also takes into account the impact of the loop environment, the type of knocking object, the knocking speed, the operating conditions, etc. on the spectral entropy curve, which helps to improve the accuracy of identifying the impact signals of loose parts through the signal spectral entropy curve; in addition, based on the average Euclidean distance probability density distribution function, the Euclidean distance threshold at different confidence levels can be given scientifically and reasonably, thereby improving the accuracy of identifying the impact signals of loose parts.

[0064] Compared with the prior art, the alarm method provided by the present invention first determines whether the amplitude of the vibration signal exceeds the amplitude threshold, realizes real-time and efficient capture of abnormal signals, avoids missed alarms, and then uses the method of increasing weights to obtain a modified energy spectrum for the energy spectrum of the 6kHz-15kHz frequency band, which can exclude the interference of information outside the frequency band, and can enhance the ability of the spectrum entropy curve to distinguish vibration signals from background noise signals. In addition, the signal processing method based on signal spectrum entropy can detect collision signals with low signal-to-noise ratio, and can exclude the interference of pulse signals such as electrical interference, and the method for determining the preset distance threshold takes into account the influence of the circuit environment, the type of knocking object, the knocking speed, the operating conditions, etc. on the spectrum entropy curve, which helps to improve the accuracy of identifying the impact signal of loose parts through the signal spectrum entropy curve. The Euclidean distance can quantitatively compare the similarity of two time series, and combined with the classification threshold at a certain confidence level determined based on the average Euclidean distance probability density function, it can accurately identify the spectrum entropy curve corresponding to the impact signal of the loose parts, and improve the accuracy of the alarm judgment.

[0065] The basic concepts have been described above. Obviously, for those skilled in the art, the above invention disclosure is only used as an example and does not constitute a limitation of the present application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements and amendments to the present application. Such modifications, improvements and amendments are suggested in the present application, so such modifications, improvements and amendments still belong to the spirit and scope of the exemplary embodiments of the present application.

[0066] At the same time, the present application uses specific words to describe the embodiments of the present application. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0067] Some aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". The processor may be one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DAPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, or combinations thereof. In addition, various aspects of the present application may be expressed as computer products located in one or more computer-readable media, which include computer-readable program codes. For example, computer-readable media may include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, tapes ...), optical disks (e.g., compact disks CDs, digital versatile disks DVDs ...), smart cards, and flash memory devices (e.g., cards, sticks, key drives ...).

[0068] A computer-readable medium may include a propagated data signal containing computer program code, such as in baseband or as part of a carrier wave. The propagated signal may have a variety of manifestations, including electromagnetic, optical, etc., or a suitable combination. A computer-readable medium may be any computer-readable medium other than a computer-readable storage medium, which may be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit a program for use. The program code on the computer-readable medium may be propagated via any suitable medium, including radio, cable, fiber optic cable, radio frequency signal, or similar medium, or any combination of the above mediums.

[0069] Similarly, it should be noted that in order to simplify the description of the disclosure of this application and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this application, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the object of this application are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0070] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0071] Although the present application has been described with reference to the current specific embodiments, ordinary technicians in this technical field should recognize that the above embodiments are only used to illustrate the present application, and various equivalent changes or substitutions may be made without departing from the spirit of the present application. Therefore, as long as the changes and modifications to the above embodiments are within the essential spirit of the present application, they will fall within the scope of the claims of the present application.

Claims

1. An alarm method, suitable for monitoring the looseness of internal components of a device, characterized in that: The device is internally installed with a plurality of acceleration sensors to obtain vibration signals at corresponding positions in real time. The alarm method comprises the following steps: a. filtering the vibration signal to obtain a first noise signal; b. determining whether the maximum amplitude of the first noise signal is greater than a preset amplitude threshold, if not, restarting step a, and if so, recording the first noise signal and obtaining a vibration signal near the first noise signal as a second noise signal; c. dividing the second noise signal into multiple frames of noise signals, performing Fourier transform on each frame of the multiple frames of noise signals to obtain multiple energy spectra, and performing frequency band correction on the multiple energy spectra to obtain multiple corrected energy spectra; d. Calculating the proportion of the energy of the multiple frequency components in each of the modified energy spectrum to the entire spectrum, obtaining multiple probability values ​​of the multiple frequency components, and calculating the spectral entropy corresponding to each frame through the multiple probability values; e. performing normalization processing on the multiple spectral entropies to obtain a suspected spectral entropy curve, calculating the average Euclidean distance between the suspected spectral entropy curve and multiple standard spectral entropy curves, and generating an alarm if the maximum value of the average Euclidean distance is less than a preset distance threshold.

2. The alarm method according to claim 1, characterized in that: The formula for performing frequency band correction on the multiple energy spectra to obtain multiple corrected energy spectra is: Among them, Y ′ m (f i ) is the corrected energy spectrum after correction, Y m (f i ) is the uncorrected energy spectrum, Y m (f i )’s average value.

3. The alarm method according to claim 2, characterized in that: The method further includes optimizing the suspected spectral entropy curve and the multiple standard spectral entropy curves by using a dynamic time warping method before calculating the average Euclidean distance between the suspected spectral entropy curve and the multiple standard spectral entropy curves.

4. The alarm method according to claim 3, characterized in that: Acquiring a vibration signal near the first noise signal as a second noise signal comprises the following steps: Recording the time when the maximum amplitude of the first noise signal is greater than the preset amplitude threshold as the center time; and Signal data in a first time period before the center moment and in a second time period after the center moment are intercepted as the second noise signal.

5. The alarm method according to claim 4, characterized in that: The duration of the second time period is greater than or equal to 50 ms.

6. The alarm method according to claim 4, characterized in that: Also includes: Selecting loose parts to knock different positions of the inner wall of the device under different operating conditions of the device to obtain a plurality of standard signals; as well as The preset amplitude threshold, the standard spectral entropy curve and the preset distance threshold are obtained through the multiple standard signals.

7. The alarm method according to claim 6, characterized in that: Obtaining the preset amplitude threshold through the standard signal comprises the following steps: Filter the plurality of standard signals to obtain a plurality of first analog signals, and extract the maximum amplitude of the plurality of first analog signals. Record multiple background noise signals of the device at different power levels by the acceleration sensor, filter the multiple background noise signals to obtain multiple second analog signals, and extract the maximum amplitude of the multiple second analog signals as well as The preset amplitude threshold is obtained according to the maximum amplitude of the plurality of first analog signals and the maximum amplitude of the plurality of second analog signals.

8. The alarm method according to claim 6, characterized in that: Obtaining the standard spectral entropy curve by using the standard signal comprises the following steps: f. filtering the standard signal to obtain a first standard signal; g. Determine whether the maximum amplitude of the first standard signal is greater than a preset amplitude threshold, if not, restart step a, if so, record the first standard signal and obtain a vibration signal near the first standard signal as a second standard signal; h. dividing the second standard signal into multiple frames of standard signals, performing Fourier transform on each frame of the multiple frames of standard signals to obtain multiple energy spectra, and performing frequency band correction on the multiple energy spectra to obtain multiple corrected energy spectra; i. Calculate the proportion of the energy of the multiple frequency components in each of the modified energy spectrum to the entire spectrum, obtain multiple probability values ​​of the multiple frequency components, and calculate the spectral entropy corresponding to each frame through the multiple probability values; j. Normalizing the plurality of spectral entropies to obtain a standard spectral entropy curve.

9. The alarm method according to claim 8, characterized in that: Obtaining the preset distance threshold by using the standard signal includes: Calculate the average Euclidean distance d between the spectral entropy curves of any two frames of standard signals ij , if there exists 0<α≤0.20 such that P{d ij >d α (n)}=α, then d α (n) is the preset amplitude threshold; wherein, d α (n) Satisfaction 10. The alarm method according to claim 9, characterized in that: The formula for calculating the average Euclidean distance is: Among them, x′ i1 , x′ i2 ,…,x′ in These are multiple spectral entropy curves optimized by dynamic time warping method.

Images