Alerting method

Abstract

The application provides an alarm method, comprising the following steps: a. filtering a 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 value, and if yes, recording the first noise signal and performing step c; c. obtaining the vibration signal before the first noise signal as a background noise signal, establishing an AR model of the background noise signal, and calculating the residual error of the first noise signal and the predicted signal of the AR model; d. calculating the sequential probability ratio test parameter of the residual error, and judging whether the sequential probability ratio is greater than a preset probability ratio threshold value for multiple times, and if yes, recording the time when the sequential probability ratio is greater than the probability ratio threshold value for the first time as a suspected time; e. calculating the amplitude probability density function of the residual error, and calculating the correlation coefficient of the amplitude probability density function of the residual error and the probability density function of a standard signal; and f. obtaining a frequency ratio according to the self-power density spectrum of the first noise signal at the suspected time.

Description

TECHNICAL FIELD

[0001] The present application relates to a loose parts monitoring system, in particular to an alarm method. BACKGROUND

[0002] Nuclear power plant system is a highly complex and important safety requirements of the structure system, the safety of nuclear power plant operation is one of the primary problems in the field of nuclear engineering research. In the nuclear power plant loop coolant system, loose parts are often found, which are caused by the loosening or falling of parts during operation, or may be parts, tools and other objects left in the system during construction, refueling or maintenance. There are a large number of components in the nuclear power plant loop system that can cause hidden faults, some of which can loosen or fall off the body. Once this happens, it may cause serious failure and accident of nuclear power plant operation, and may cause nuclear power plant reactor to shut down. Therefore, the monitoring of loose parts in nuclear power plants is particularly important. Loose parts monitoring system (Loose Parts Monitoring System, LPMS) is one of the basic diagnostic tools for nuclear power plant fault detection, mainly used for monitoring loose parts in the nuclear power plant loop coolant system. Its purpose is to detect and locate loose or falling parts that may exist in the loop, and to issue an alarm in a timely manner to provide information for possible subsequent operations, thereby ensuring the safety and reliability of nuclear power plant operation.

[0003] Quick and accurate monitoring of loose parts falling and issuing an alarm is the most important function of the loose parts monitoring system (Loose Part Monitoring System, LPMS). The design of the alarm algorithm should minimize false alarms, while also reducing the probability of false alarms. There are generally two reasons for false alarms and false alarms. One is that the impact signal energy is too low and is submerged in the background noise. According to a large number of experiments, the frequency spectrum of the loose part impact signal is mainly distributed in 1kHz to 25kHz, so band-pass filtering (1kHz~25kHz) of the collected impact signal can effectively eliminate most of the background noise. The frequency of the background noise in the nuclear power plant is generally not more than 6kHz, so the frequency spectrum of the background noise and the impact signal has a large overlap, especially between 3~5KHz. The current commonly used spectral filtering method cannot separate the background noise and the impact signal. When the amplitude of the noise exceeds the preset value, a false alarm occurs, or the effective loose part monitoring signal is submerged in the noise, resulting in a false alarm. The second reason is the interference signal generated by non-falling parts, such as mechanical, electrical noise, large amplitude electrical transient signal, thermal shock vibration signal and friction vibration signal. Such interference signals have the characteristics of large amplitude and short duration, especially the acceleration time history curve of the thermal shock vibration signal and the friction vibration signal is very similar to that of the loose part impact signal, so these interference signals often trigger false alarms. SUMMARY

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

[0005] To solve the above technical problem, the present application provides an alarm method suitable for monitoring the loosening of internal components of a device, the device having a plurality of acceleration sensors installed therein to obtain vibration signals in real time, the alarm method comprising the following steps: a. filtering the vibration signals 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, starting to execute step a again, if yes, recording the first noise signal and executing step c; c. obtaining the vibration signal before the first noise signal as a background noise signal, establishing an AR model of the background noise signal, and calculating the residual error between the first noise signal and the predicted signal of the AR model; d. calculating the sequential probability ratio test parameter of the residual error, and determining whether the sequential probability ratio is greater than a preset probability ratio threshold multiple times, if not, starting to execute step a again, if yes, recording the time when the sequential probability ratio is greater than the probability ratio threshold for the first time as a suspected time; e. calculating the amplitude probability density function of the residual error, and calculating the correlation coefficient between the amplitude probability density function of the residual error and the probability density function of the standard signal; f. obtaining a frequency ratio from the self-power density spectrum of the first noise signal at the suspected time; g. comparing the correlation coefficient and the frequency ratio with a preset correlation coefficient threshold and a preset frequency ratio threshold respectively, and if there is a signal that makes the correlation coefficient greater than the preset correlation coefficient threshold and the frequency ratio greater than the preset frequency ratio threshold, an alarm is triggered.

[0006] In an embodiment of the present application, obtaining the vibration signal before the first noise signal as a background 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 a center time; and intercepting signal data within a first time period before the center time and a second time period after the center time, and selecting part of the signal data away from the center time in the first time period as the background noise signal.

[0007] In an embodiment of the present application, the alarm method further comprises: selecting different positions of the inner wall of the device struck by the loose component under different operating conditions of the device to obtain a plurality of standard signals; and obtaining the preset amplitude threshold, the preset frequency ratio threshold and the preset correlation coefficient threshold through the plurality of standard signals.

[0008] In an embodiment of the present application, obtaining the preset amplitude threshold through the standard signal comprises the following steps: filtering the plurality of standard signals to obtain a plurality of first analog signals, extracting the maximum amplitudes of the plurality of first analog signals ; recording multiple background noise signals of the device at different power levels by the acceleration sensor, filtering the multiple background noise signals to obtain multiple second analog signals, and extracting maximum amplitudes of the multiple second analog signals (i=1, 2, 3…n); and obtaining the preset amplitude threshold value through the maximum amplitudes of the multiple first analog signals and the maximum amplitudes of the multiple second analog signals.

[0009] In an embodiment of the present application, a formula for obtaining the preset amplitude threshold value through the maximum amplitudes of the multiple first analog signals and the maximum amplitudes of the multiple second analog signals is: , wherein, is the maximum amplitude of the multiple first analog signals is a lower limit value at a 95% confidence level, is the maximum amplitude of the multiple second analog signals is an upper limit value at a 95% confidence level.

[0010] In an embodiment of the present application, obtaining the preset frequency ratio threshold value through the standard signals comprises the following steps: gradually increasing the temperature inside the device, recording friction vibration signals and thermal shock vibration signals occurring during the temperature increasing process, and filtering the friction vibration signals and the thermal shock vibration signals; and calculating frequency ratios of the multiple standard signals, the friction vibration signals and the thermal shock vibration signals; and taking an average of a minimum value of the frequency ratios of the multiple standard signals and a maximum value of the frequency ratios of the friction vibration signals and the thermal shock vibration signals as the preset frequency ratio threshold value.

[0011] In an embodiment of the present application, obtaining the preset correlation coefficient threshold value through the standard signals comprises the following steps: calculating multiple amplitude density functions of the multiple standard signals, arbitrarily selecting two from the multiple amplitude density functions, calculating a correlation coefficient of the selected two amplitude density functions, and obtaining a total of N(N-1) / 2 correlation coefficients; and drawing a probability distribution graph according to the N(N-1) / 2 correlation coefficients, and finding a corresponding correlation coefficient lower limit value at a 95% confidence level from the probability distribution graph as the preset correlation coefficient threshold value; wherein N is the number of the multiple standard signals.

[0012] In an embodiment of the present application, a formula for calculating the frequency ratio is: , wherein, APSD is a self-power density spectrum, low_ub and low_lb represent upper and lower limits of integral frequencies of a low frequency band, and high_ub and high_lb represent upper and lower limits of integral frequencies of a high frequency band.

[0013] In an embodiment of the present application, a formula for establishing an AR model of the background noise signal is: , wherein, is the AR model, and p is a system order.​​ are coefficients of the AR model, is the first noise signal.

[0014] In an embodiment of the present application, the formula for calculating the residual of the first noise signal and the prediction signal of the AR model is: wherein, is the residual, p is the system order, are coefficients of the AR model, is the first noise signal.

[0015] In an embodiment of the present application, the formula for calculating the correlation coefficient of the amplitude probability density function of the residual and the probability density function of the standard signal is: wherein, p is the correlation coefficient, x is the amplitude probability density function of the residual, y is the amplitude probability density function of the standard signal, Cov represents the covariance, and s represents the standard deviation.

[0016] Compared with the prior art, the alarm method provided by the present application can realize real-time and efficient capture of abnormal signals by judging whether the amplitude of the vibration signal exceeds the amplitude threshold, thereby avoiding false alarms; the AR model can effectively strip the interference of background noise, and the AR model is updated in real time according to the change of the background noise, thereby improving the denoising effect; and whether there is an abnormal component in the signal is identified by judging whether the sequential probability checking parameter of the residual signal after stripping the background noise exceeds the probability ratio threshold; and finally, whether the signal is a loose component signal is determined in combination with the amplitude probability density function and the frequency ratio. The three times of judgment effectively reduce the probability of false alarms and missed alarms, and the method still has good stability under the condition that the noise and the impact signal amplitude difference is not obvious, and there are complex background noises such as impulse noise, friction and thermal impact signal similar in shape to the impact signal. In addition, the threshold parameters used in the three times of judgment are determined based on the standard signal of the representative loose component in the simulation experiment, thereby improving the accuracy of alarm judgment. BRIEF DESCRIPTION OF DRAWINGS

[0017] The accompanying drawings are included to provide a further understanding of the present application, and are incorporated and constitute a part of this application, illustrate embodiments of the present application, and together with the description serve to explain the principles of the present application. In the drawings:

[0018] Figure 1 is a flowchart of an alarm method of the present application.

[0019] Figure 2 is a time course curve diagram of the standard signal of the present application.

[0020] Figures 3-6 is the sequential probability ratio obtained by different preset standard deviations s of the present application. DETAILED DESCRIPTION

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some examples or embodiments of the present application, and for those skilled in the art, the present application can also be applied to other similar scenarios without creative labor on the basis of these drawings. Unless it is clear from the language context or otherwise indicated, the same reference numbers in the drawings represent the same structure or operation.

[0022] As shown in the present application and claims, unless the context clearly indicates otherwise, the words "one", "an", "a", and / or "the" do not mean to specify a single number, but can also include a plurality. Generally, the terms "comprising" and "including" only indicate the inclusion of the steps and elements explicitly identified, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements.

[0023] Unless otherwise specifically indicated, the relative arrangement of the components and steps, numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present application. At the same time, it should be understood that the sizes of the various parts shown in the drawings are not drawn in accordance with the actual proportional relationship. The technology, methods and devices known to those skilled in the relevant art can not be discussed in detail, but under appropriate circumstances, the technology, methods and devices should be considered as part of the authorized description. In all examples shown and discussed here, any specific value should be interpreted as merely exemplary, and not as a limitation. Therefore, other examples of exemplary embodiments can have different values. It should be noted that similar reference numbers and letters represent similar items in the following drawings, so once an item is defined in one 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 orientation words such as "front, back, up, down, left, right", "horizontal, vertical, perpendicular, horizontal" and "top, bottom" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and in the absence of contrary indications, these orientation words do not indicate and imply that the indicated device or element must have a particular orientation or be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the scope of protection of the present application; the orientation words "inner, outer" refer to the inner and outer relative to the contour of the parts themselves.

[0025] For purposes of the description hereinafter, spatially relative terms— such as "above", "below", "up", "down", "between" and the like— can be used to describe an orientation of one element or feature to another element or feature as illustrated in the figures. It will be understood that the spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. For example, if a device is inverted in the figures, a spatially relative term such as "above" or "below" can indicate a different orientation (or a reversed orientation) of the device to that shown in the figures. The device can be otherwise oriented (rotated 90 degrees or at other orientations) and the spatially relative descriptors used herein interpreted accordingly. The terms "first", "second", "third", etc., as used herein do not denote any order, quantity, or importance, but rather are used to distinguish one element from another, and are more especially used to distinguish an element having a certain property from another element lacking such property (e.g., a first element is a "first" element in that it is the first element to have a certain property, while a second element is a "second" element in that it is the second element to have the same property).

[0026] In addition, it should be noted that the use of "first", "second", etc., words to describe a component does not limit the scope of the claims to the corresponding component. The words are only used to distinguish one component from another. Unless otherwise stated, the words are not meant to imply a special ranking of importance of one component over another. Furthermore, although the terms commonly used in the art are selected from the commonly used terms, some terms mentioned in the specification can be selected by the applicant according to his or her judgment, and the detailed meanings thereof are described in the relevant part of the description. In addition, the present application is required to be understood not only by the actual terms used, but also by the meaning implied by each term.

[0027] It will 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 can be present. In contrast, when a component is referred to as being "directly on", "directly connected to", "directly coupled to", or "directly contacting" another component, there are no intervening components present.

[0028] Flowcharts representative of the systems managed according to embodiments of the application are used herein to illustrate the operations according to embodiments of the application. It will be understood that the operations of the embodiments can be carried out in sequence or in parallel, or in a different order than illustrated. Additionally, other operations can be added or removed from the processes, or one or more operations can be added to or removed from the processes.

[0029] Figure 1 is a flowchart of a method of alerting according to an embodiment of the application. Reference is made to Figure 1As shown, the present application provides an alarm method suitable for monitoring the loosening of internal components of a device, the device is internally installed with a plurality of acceleration sensors, real-time vibration signals of corresponding positions are obtained, the alarm method 10 comprises the following steps:

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

[0031] S12: determining whether the maximum amplitude of the first noise signal is greater than a preset amplitude threshold, if not, starting to execute step S11 again, if yes, executing step S13.

[0032] S13: recording the first noise signal.

[0033] S14: obtaining the vibration signal before the first noise signal as a background noise signal, establishing an AR model of the background noise signal, and calculating the residual error between the first noise signal and the predicted signal of the AR model.

[0034] S15: calculating the sequential probability ratio test parameter of the residual error.

[0035] S16: determining whether the sequential probability ratio is greater than a preset probability ratio threshold for multiple times, if not, starting to execute step S11 again, if yes, executing step S17.

[0036] S17: recording the time when the sequential probability ratio is greater than the probability ratio threshold for the first time as a suspected time.

[0037] S18: calculating the amplitude probability density function of the residual error, and calculating the correlation coefficient between the amplitude probability density function of the residual error and the probability density function of the standard signal.

[0038] S19: obtaining a frequency ratio from the self-power density spectrum of the first noise signal at the suspected time.

[0039] S20: comparing the correlation coefficient and the frequency ratio with a preset correlation coefficient threshold and a preset frequency ratio threshold respectively, and if there is a signal that makes the correlation coefficient greater than the preset correlation coefficient threshold and the frequency ratio greater than the preset frequency ratio threshold, an alarm is given.

[0040] Specifically, before the alarm method 10 is executed, a plurality of acceleration sensors need to be installed inside the device to obtain vibration signals of corresponding positions in real time. In step S11, the signals obtained by the plurality of acceleration sensors in real time are filtered to eliminate noise in some frequency bands, thereby preliminarily screening background noise and avoiding false alarms. Specifically, according to a large number of experiments, when a part inside the device is loose, the impact signal (hereinafter referred to as a standard signal) of the loose part is mainly distributed in the frequency spectrum of 1 kHz to 25 kHz. Therefore, in the specific implementation, band-pass filtering of 1 kHz to 25 kHz is performed on the collected impact signal to effectively eliminate most of the background noise. However, the background noise will not exceed 6 kHz, and the frequency spectrum of the background noise and the standard signal has a large overlap, especially between 3 kHz and 5 kHz, so it is difficult to separate the background noise and the standard signal by using a general spectrum filtering method, and therefore a subsequent algorithm is needed to further extract the signal.

[0041] In step S12, it is necessary to determine whether the maximum amplitude of the first noise signal after filtering is greater than a preset amplitude threshold. If the maximum amplitude does not exceed the preset amplitude threshold, it means that the standard signal does not appear at this time, that is, the parts inside the device are not loose, and no alarm is needed, and then the method returns 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 it is possible that the parts are loose. The first noise signal is recorded for further processing.

[0042] It can be understood that the background noise of the device at different time is not consistent, and the background noise signal needs to be recorded when the first noise signal which may exist abnormal is further processed. The vibration signal before the first noise signal is obtained in step S14 is the background noise signal, the AR model of the background noise signal is established, and the residual error of the first noise signal and the prediction signal of the AR model is calculated. In an embodiment of the present application, the vibration signal before the first noise signal is obtained as the background noise signal further comprises the following steps: recording the time point with the maximum amplitude of the first noise signal greater than the preset amplitude threshold as the center time point, intercepting the signal data in the first time period before the center time point and the second time period after the center time point, and selecting part of the signal data away from the center time point in the first time period as the background noise signal. It can be understood that for the standard signal, the waveform is usually suddenly appeared at the impact time and maintained for a period of time, so the part of the time range before the center time point is less than the part of the time range after the center time point, and the background noise signal should be before the impact time, that is, in the first time period, otherwise it will be disturbed by the standard signal, and preferably, the selection of the background noise signal should be as far away from the center time point as possible. In a preferred embodiment, 100 ms of signal data is intercepted around the center time point with the super preset amplitude threshold, the time range is 30 ms before the center time point to 70 ms after the center time point, and then 20 ms of background noise data is selected to establish the AR model.

[0043] In an embodiment of the present application, the formula for establishing the AR model of the background noise signal is: , wherein, is the AR model, p is the system order, is the coefficient of the AR model, is the first noise signal. Further, the formula for calculating the residual error of the first noise signal and the prediction signal of the AR model is: , wherein, is the residual error, p is the system order used for establishing the AR model of the background noise signal, is the coefficient of the AR model, is the first noise signal. If it is noise, the prediction residual error becomes white noise, and if the collision signal appears, the residual error is relatively large because it cannot be accurately predicted by noise. The AR model can effectively strip the interference of the background noise, thereby improving the signal-to-noise ratio of the standard signal and the background noise signal. In addition, the AR model established by selecting the background noise signal close to the abnormal signal can be updated in real time according to the change of the background noise, thereby improving the denoising effect.

[0044] The sequential probability ratio test parameter for calculating the residual error in S15 uses the formula: , which can be simplified as: If yes, the first noise signal obtained in step S13 is defined as a suspected signal, and the time when the sequential probability ratio first exceeds the probability ratio threshold is recorded as a suspected time. If no, the method returns to step S11 to continue monitoring.

[0045] In step S18, the amplitude probability density function of the residual obtained in step S14 is extracted, and a correlation coefficient between the amplitude probability density function and the probability density function of the standard signal is calculated. In an embodiment of the present application, the formula for calculating the correlation coefficient between the amplitude probability density function of the residual and the probability density function of the standard signal is: wherein p is the correlation coefficient, x is the amplitude probability density function of the residual, y is the amplitude probability density function of the standard signal, Cov represents the covariance, and s represents the standard deviation.

[0046] In step S19, the autocorrelation power spectrum of the suspected signal defined in step S13 is extracted, the ratio of the integrals of the autocorrelation power spectrum in the low frequency band and the high frequency band is calculated, and is defined as a frequency ratio. In an embodiment of the present application, the frequency ratio is the ratio of the integral of the autocorrelation power spectrum in the low frequency band to the integral of the autocorrelation power spectrum in the high frequency band. Specifically, the frequency ratio can be calculated by using the following formula: wherein APSD is the autocorrelation power spectrum of the signal for which the power ratio needs to be calculated (at this time, the signal is the suspected signal), low ub and low lb represent the upper and lower limits of the integral frequency of the low frequency band, and high ub and high lb represent the upper and lower limits of the integral frequency of the high frequency band.

[0047] In a specific implementation, the initial values of low ub, low lb, high ub and high lb are respectively set as 15 kHz, 4 kHz, 45 kHz and 17 kHz. On this basis, through a large amount of test data, it is calculated that the frequency ratio range of the standard signal is 2-40, the frequency ratio range of the thermal shock vibration signal caused by thermal expansion is 0.01-0.02, and the frequency ratio range of the friction vibration signal is 0.03-0.08. Therefore, the preset frequency ratio threshold in step S20 is taken as 1.

[0048] In step S20, the correlation coefficient (time domain index) obtained in step S18 and the frequency ratio (frequency domain index) obtained in step S19 are compared with the preset correlation coefficient threshold and the preset frequency ratio threshold respectively. If there is a signal such that the correlation coefficient is greater than the preset correlation coefficient threshold and the frequency ratio is greater than the preset frequency ratio threshold, an alarm is given. Otherwise, the method returns to step S11 to continue monitoring.

[0049] Further, the determination method of the sequential probability ratio test parameter in step S15 and the preset probability ratio threshold in step S20 includes the following steps:

[0050] (1) The residual obtained in step S14 is normalized.

[0051]

[0052] Sequence The sequential probability ratio test is performed on the data according to the above description.

[0053] (2) The standard deviation of the preset noise signal is 1, the standard deviation of the preset vibration signal is 1.1, which is slightly larger than the standard deviation of the noise signal , the preset false alarm rate is , and the preset missed alarm rate is 10%

[0054] (3) Based on the standard deviation, false alarm rate and missed alarm rate preset in step (2), the upper and lower limits of the sequential probability ratio are calculated as follows:

[0055]

[0056]

[0057] (4) The initial value of the sequential probability ratio is given as .

[0058] (5) The sequential probability ratio at the next time point is calculated based on the sequential probability ratio at the previous time point (n = 1, 2, 3, …), if >b and n+1>100, then .

[0059] Figure 2 is the time course curve of the standard signal of the present application, Figures 3-6 is the sequential probability ratio obtained by different preset standard deviations σ of the present application. The execution of the logical judgment "if >b and n+1>100, then " makes the sequential probability ratio reach the upper limit directly and return to zero, so when the background appears an impact signal, the test parameter will jump between 0 and the upper boundary, and reach the upper boundary value many times, as shown in Figure 2 Therefore, step S16 determines whether there is an abnormal component in the signal by judging whether the sequential probability ratio exceeds the preset probability ratio threshold many times.

[0060] To achieve accurate alarm using the sequential probability ratio method, the key is to set appropriate standard deviations and ​However, the mean and variance of the data collected by the loose component monitoring system in real time are changing all the time, and thus cannot be set in advance, and thus the conversion of the sensor signal sequence into a fixed (0, 1) normal distribution can not only simplify the calculation, but more importantly, a unified sequential probability ratio calculation formula can be obtained, which is conducive to the realization of real-time automatic monitoring. Therefore, the normalized signal sequence is used as the input sequence of the sequential probability ratio test, which has many advantages.

[0061] When the normalized signal sequence is used as the sequential probability ratio test, the noise and vibration signal standard deviation and in the above formula have a great influence on the test results. After a large amount of data calculation, the standard deviation of the preset noise signal is 1, and the standard deviation of the preset vibration signal is 1.1~1.9, slightly larger than the standard deviation of the noise signal . Setting the parameters in this way will obtain good monitoring effect, as shown in Figures 2-6 . Figure 2 The original signal time history is given in Figures 3-6 , and the sequential probability ratios obtained by the values of the standard deviation of the different preset vibration signals are given in . It can be seen that when is greater than located in the interval 1.1~1.9, the sequential probability ratio method can better test the abnormal components in the signal. However, only this method cannot completely determine whether it is a loose component signal, because pulse noise, friction vibration signal, thermal shock vibration signal, etc. will be detected as suspected signals by this method, and further analysis is still needed to determine whether it is a real loose part alarm.

[0062] In an embodiment of the present application, the alarm method further comprises: selecting different positions of the loose components knocking the inner wall of the device under different operating conditions of the device to obtain a plurality of standard signals; and obtaining a preset amplitude threshold, a preset frequency ratio threshold and a preset correlation coefficient threshold through the plurality of standard signals. Specifically, the plurality of standard signals are obtained through simulation experiments: deploying the loose component monitoring system on a simulated loop test bench or an actual reactor loop, selecting a certain number of steel balls, bolts, pins and other representative loose components in the primary loop, under different operating conditions of the loop test bench, knocking the different positions of the primary loop main equipment container wall at a speed not less than 0.68J impact energy through free fall, pendulum and other ways, recording the impact signal of the loose component as the standard signal, and further processing the standard signal to obtain the preset amplitude threshold, the preset frequency ratio threshold and the preset correlation coefficient threshold and other preset data.

[0063] In an embodiment of the present application, the preset amplitude threshold value is obtained from the standard signals by the following steps: filtering the standard signals to obtain a plurality of first analog signals, extracting the maximum amplitudes of the first analog signals ; recording a plurality of background noise signals of the device at different power levels by the acceleration sensor, filtering the background noise signals to obtain a plurality of second analog signals, extracting the maximum amplitudes of the second analog signals (i = 1, 2, 3…n); and obtaining the preset amplitude threshold value from the maximum amplitudes of the first analog signals and the maximum amplitudes of the second analog signals.

[0064] In an embodiment of the present application, the formula for obtaining the preset amplitude threshold value from the maximum amplitudes of the first analog signals and the maximum amplitudes of the second analog signals is: , wherein, is the maximum amplitude of the first analog signals is the lower limit value at the 95% confidence level, is the maximum amplitude of the second analog signals is the upper limit value at the 95% confidence level.

[0065] In an embodiment of the present application, the preset frequency ratio threshold value is obtained from the standard signals by the following steps: gradually increasing the temperature inside the device, recording the friction vibration signals and thermal shock vibration signals occurring during the temperature rising process, and filtering the friction vibration signals and thermal shock vibration signals; and calculating the frequency ratios of the standard signals, the friction vibration signals and the thermal shock vibration signals; and taking the average of the minimum value of the frequency ratios of the standard signals and the maximum value of the frequency ratios of the friction vibration signals and the thermal shock vibration signals as the preset frequency ratio threshold value. It can be understood that the formula is also used when calculating the preset frequency ratio threshold value of the standard signals. In the formula, APSD is the auto-power spectral density of the signal for which the power ratio needs to be calculated (at this time, it is the standard signal), low_ub and low_lb represent the upper and lower limits of the integral frequency of the low frequency band, and high_ub and high_lb represent the upper and lower limits of the integral frequency of the high frequency band.

[0066] In an embodiment of the present application, the preset correlation coefficient threshold value is obtained from the standard signals by the following steps: calculating a plurality of amplitude density functions of the standard signals, arbitrarily selecting two from the plurality of amplitude density functions, calculating the correlation coefficient of the selected two amplitude density functions, and obtaining a total of correlation coefficients; and drawing a probability distribution graph according to the correlation coefficients, and finding the corresponding lower limit value of the correlation coefficient at the 95% confidence level from the probability distribution graph as the preset correlation coefficient threshold value; wherein N is the number of the standard signals.

[0067] Compared with the prior art, the alarm method provided by the application first judges whether the amplitude of the vibration signal exceeds the amplitude threshold value, realizes real-time and efficient capture of abnormal signals, and avoids false alarms; the AR model can effectively strip the interference of background noise, and the AR model is updated in real time according to the change of the background noise, improves the denoising effect, and judges whether the sequential probability check parameter of the residual signal after stripping the background noise exceeds the probability ratio threshold value, to identify whether there is an abnormal component in the signal; finally, the amplitude probability density function and the frequency ratio two characteristic parameters are combined to judge whether the signal is a loose component signal. Three times of judgment effectively reduces the probability of false positives and false negatives, and has good stability under the condition that the noise and impact signal amplitude difference is not obvious, and there are complex background noises such as pulse noise, friction, thermal impact signal and the like similar to the shape of the impact signal. In addition, the various threshold parameters used in the three judgment processes are determined based on the standard signals of representative loose components in the simulation experiment, which improves the accuracy of alarm judgment.

[0068] The above has described the basic concepts, and it is obvious that the above-mentioned application disclosure is only used as an example and does not constitute a limitation on the application. Although it is not explicitly stated here, those skilled in the art can make various modifications, improvements and corrections to the application. Such modifications, improvements and corrections are suggested in the application, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the application.

[0069] At the same time, specific words are used in the application to describe the embodiments of the application. As "one embodiment", "an embodiment", and / or "some embodiments" means a certain feature, structure or characteristic related to at least one embodiment of the application. Therefore, it should be emphasized and noted that the "an embodiment" or "one embodiment" or "an alternative embodiment" mentioned in different places in the specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the application can be properly combined.

[0070] Aspects of the application can be implemented in hardware, software (including firmware, resident software, micro-code, etc.), or a combination thereof. The foregoing hardware or software can be referred to as a "data block", "module", "engine", "unit", "component", or "system". A processor can 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, micro-controllers, microprocessors, or combinations thereof. Furthermore, aspects of the application can be implemented as a computer product, which can include a computer-readable medium having stored thereon a computer program coded to perform the method. For example, the computer-readable medium can include, but is not limited to, magnetic storage devices (e.g., hard disk, floppy disk, magnetic strips...), optical disks (e.g., compact disk (CD), digital versatile disk (DVD)...), smart cards, and flash devices (e.g., card, stick, key drive...).

[0071] The computer readable medium can include a propagated data signal with computer program coded therein, for example, in baseband or as a carrier wave. The propagated signal can take on many forms, including but not limited to, electro-magnetic, optical, or any suitable combination thereof. Computer readable medium can be any medium that can be read by a machine (e.g., a computer) and can contain or store a computer program to be used by the machine. A computer program can be propagated in any suitable form, including but not limited to, electro-magnetic, optical, or any suitable combination thereof. Computer readable medium can be any medium that can be read by a machine (e.g., a computer) and can contain or store a computer program to be used by the machine. A computer program can be propagated in any suitable form, including but not limited to, electro-magnetic, optical, or any suitable combination thereof.

[0072] Similarly, it is to be noticed that the term "comprising", used in the description, is not intended to exclude other features but to include other features than the ones described so that also other embodiments than the embodiments specifically described are possible. Furthermore, it is to be noted that the use of the term "example" in the description is not intended to exclude other embodiments than the ones specifically described.

[0073] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates 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 be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of scope in some embodiments of this application are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0074] Although this application has been described with reference to specific embodiments, those skilled in the art should recognize that the above embodiments are only used to illustrate this application, and various equivalent changes or substitutions can be made without departing from the spirit of this application. Therefore, any changes or modifications to the above embodiments within the essential spirit of this application will fall within the scope of the claims of this application.

Claims

1. An alarm method suitable for monitoring the loosening of internal components of equipment, characterized in that, The device is equipped with multiple acceleration sensors to acquire vibration signals at corresponding locations in real time. The alarm method includes the following steps: a. Filter the vibration signal to obtain a first noise signal; b. Determine whether the maximum amplitude of the first noise signal is greater than a preset amplitude threshold. If not, restart step a. If yes, record the first noise signal and execute step c. c. Obtain the vibration signal preceding the first noise signal as the background noise signal, establish an AR model of the background noise signal, and calculate the residual between the first noise signal and the predicted signal of the AR model; d. Calculate the sequential probability ratio test parameter of the residual, and determine whether the sequential probability ratio is greater than the preset probability ratio threshold multiple times. If not, restart step a. If so, record the moment when the sequential probability ratio first exceeds the probability ratio threshold as the suspected moment. e. Calculate the magnitude probability density function of the residual, and calculate the correlation coefficient between the magnitude probability density function of the residual and the probability density function of the standard signal; f. Obtain the frequency ratio based on the self-power density spectrum of the first noise signal at the suspected time, wherein the frequency ratio is the ratio of the integral of the self-power density spectrum in the low-frequency band to the integral of the self-power density spectrum in the high-frequency band. g. Compare the correlation coefficient and the frequency ratio with a preset correlation coefficient threshold and a preset frequency ratio threshold, respectively. If there is a signal that makes the correlation coefficient greater than the preset correlation coefficient threshold and the frequency ratio greater than the preset frequency ratio threshold, then an alarm is triggered.

2. The alarm method as described in claim 1, characterized in that, Obtaining the vibration signal prior to the first noise signal as the background noise signal includes the following steps: The moment when the maximum amplitude of the first noise signal exceeds the preset amplitude threshold is recorded as the center moment; and The signal data within the first time period before the center time and the second time period after the center time are extracted, and the portion of the signal data far from the center time in the first time period is selected as the background noise signal.

3. The alarm method as described in claim 1, characterized in that, Also includes: By tapping different locations on the inner wall of the equipment under different operating conditions of the equipment, multiple standard signals are obtained. as well as The preset amplitude threshold, the preset frequency ratio threshold, and the preset correlation coefficient threshold are obtained from the multiple standard signals.

4. The alarm method as described in claim 3, characterized in that, Obtaining the preset amplitude threshold from the standard signal includes the following steps: The multiple standard signals are filtered to obtain multiple first analog signals, and the maximum amplitude of the multiple first analog signals is extracted. ; The device records multiple background noise signals at different power levels using the accelerometer. These background noise signals are then filtered to obtain multiple second analog signals. The maximum amplitude of each of the multiple second analog signals is then extracted. (i=1,2,3…n); as well as The preset amplitude threshold is obtained by using the maximum amplitude of the plurality of first analog signals and the maximum amplitude of the plurality of second analog signals.

5. The alarm method as described in claim 4, characterized in that, The formula for obtaining the preset amplitude threshold by the maximum amplitude of the plurality of first analog signals and the maximum amplitude of the plurality of second analog signals is as follows: ， in, It is the maximum amplitude of the plurality of first analog signals. The lower limit at a 95% confidence level It is the maximum amplitude of the plurality of second analog signals. The upper limit at a 95% confidence level.

6. The alarm method as described in claim 3, characterized in that, Obtaining the preset frequency ratio threshold from the standard signal includes the following steps: The internal temperature of the device is gradually increased, and the friction vibration signal and thermal shock vibration signal that occur during the heating process are recorded. The friction vibration signal and the thermal shock vibration signal are then filtered. as well as Calculate the frequency ratio of the plurality of standard signals, the friction vibration signal, and the thermal shock vibration signal; as well as The average of the minimum frequency ratio of the plurality of standard signals and the maximum frequency ratio of the friction vibration signal and the thermal shock vibration signal is used as the preset frequency ratio threshold.

7. The alarm method as described in claim 3, characterized in that, Obtaining the preset correlation coefficient threshold from the standard signal includes the following steps: Calculate multiple amplitude density functions for the multiple standard signals, arbitrarily select two from the multiple amplitude density functions, and calculate the correlation coefficient between the two selected amplitude density functions. A total of [number] amplitude density functions can be obtained. One correlation coefficient; as well as according to A probability distribution plot is drawn for each correlation coefficient. From the probability distribution plot, the lower limit of the correlation coefficient at a 95% confidence level is identified as the preset correlation coefficient threshold. N is the number of the plurality of standard signals.

8. The alarm method according to any one of claims 1-7, characterized in that, The formula for calculating the frequency ratio is: ， Wherein, APSD is the self-power density spectrum, low_ub and low_lb represent the upper and lower limits of the integral frequency in the low-frequency band, and high_ub and high_lb represent the upper and lower limits of the integral frequency in the high-frequency band.

9. The alarm method as described in claim 8, characterized in that, The formula for establishing the AR model of the background noise signal is as follows: ， in, Let p be the system order of the AR model. The coefficients of the AR model are... This is the first noise signal.

10. The alarm method as described in claim 9, characterized in that, The formula for calculating the residual between the first noise signal and the predicted signal of the AR model is as follows: ， in, Let p be the residual, and p be the order of the system. The coefficients of the AR model are... This is the first noise signal.

11. The alarm method as described in claim 8, characterized in that, The formula for calculating the correlation coefficient between the magnitude probability density function of the residual and the probability density function of the standard signal is as follows: ， Where ρ is the correlation coefficient, x is the amplitude probability density function of the residual, y is the amplitude probability density function of the standard signal, Cov represents the covariance, and σ represents the standard deviation.

Images