Frequency spectrum substrate lifting detection method and device, electronic equipment and storage medium

Abstract

The invention provides a spectrum substrate lifting detection method and device, electronic equipment and a storage medium, and relates to the technical field of spectrum analysis. The method comprises the following steps: acquiring to-be-detected vibration time sequence data of a preset duration of machine equipment, and converting the to-be-detected vibration time sequence data into a to-be-detected frequency spectrum containing a frequency sequence and an amplitude sequence; based on a preset rolling window amplitude difference algorithm, calculating to obtain an amplitude difference sequence corresponding to the spectrum to be detected; matching a predetermined reference value sequence by using the amplitude difference sequence, matching an amplitude difference interval which is continuously greater than a reference value and has the maximum lifting energy, and marking a frequency interval corresponding to the amplitude difference interval as a suspected interval; and if the lifting energy corresponding to the suspected interval is greater than an alarm threshold value, determining the suspected interval as a substrate lifting interval. Compared with the prior art, the method can automatically judge the substrate lifting abnormity in the spectrum, and reduces the labor cost in spectrum analysis.

Description

Technical Field

[0001] The present application relates to the technical field of spectrum analysis, and in particular to a spectrum floor rise detection method, device, electronic device and storage medium. Background Art

[0002] Spectrum analysis plays a vital role in modern signal processing and communication systems. By analyzing the signal spectrum, the frequency components, amplitude information and other characteristics of the signal can be obtained, which can then be applied to various fields such as communication, radar, sonar, biomedical imaging, etc. However, in practical applications, the phenomenon of spectrum floor lifting often has a significant impact on the accuracy and reliability of spectrum analysis.

[0003] Spectral floor lift is a common spectral distortion phenomenon, usually caused by external interference, instrument noise, environmental changes or changes in the characteristics of the object being measured. It causes the spectrum baseline to deviate from zero, thus affecting the accurate analysis and detection of the target signal. Especially in weak signal detection and precision measurement situations, the presence of floor lift may cause valuable information to be submerged in baseline drift, affecting subsequent data processing and decision-making.

[0004] Traditional spectrum floor lift detection methods mainly rely on manual visual inspection of spectrum graphs, which is not only time-consuming and laborious, but also difficult to ensure consistency and accuracy. In addition, with the continuous development of spectrum analysis technology and the increase in application demand, automated and intelligent spectrum floor lift detection is particularly important. Summary of the invention

[0005] The purpose of this application is to provide a spectrum floor lift detection method, device, electronic device and storage medium.

[0006] In a first aspect, an embodiment of the present application provides a spectrum floor rise detection method, comprising:

[0007] Acquire the vibration time series data to be detected of the machine equipment with a preset time length, and convert the vibration time series data to be detected into a frequency spectrum to be detected including a frequency sequence and an amplitude sequence;

[0008] Based on a preset rolling window amplitude difference algorithm, an amplitude difference sequence corresponding to the spectrum to be detected is calculated;

[0009] Using the amplitude difference sequence to match a predetermined reference value sequence, matching an amplitude difference interval that is continuously greater than the reference value and has the largest lift energy, and recording the frequency interval corresponding to the amplitude difference interval as a suspected interval;

[0010] If the lifting energy corresponding to the suspected interval is greater than the alarm threshold, the suspected interval is determined as a base lifting interval.

[0011] In a possible implementation, the rolling window amplitude difference algorithm includes:

[0012] For each spectrum, a rolling window is determined based on a preset frequency accuracy and a frequency sequence difference mean of the spectrum, so that the amplitude sequence of the spectrum executes the rolling window, and the amplitude difference corresponding to the intermediate frequency in each rolling window is calculated to obtain an amplitude difference sequence corresponding to the spectrum;

[0013] The frequency sequence difference mean refers to the mean obtained by subtracting the previous value from the next value of the frequency sequence and then averaging the whole value;

[0014] The amplitude difference refers to the amplitude difference obtained by subtracting the second quantile from the first quantile in the rolling window, and the first quantile is greater than the second quantile.

[0015] In a possible implementation, the calculation formulas for the window length width and step length step of the rolling window are as follows:

[0016] width=ceil(3*freq_precision / m);

[0017] step=ceil(freq_precision / m);

[0018] Among them, ceil is rounded up, freq_precision is the mean of frequency sequence differences, and m is the frequency precision.

[0019] In a possible implementation, the process of determining the reference value sequence is as follows:

[0020] Acquire vibration data of the machine equipment in a historical time period, and divide the vibration data in the historical time period into multiple vibration time series data according to a preset time length;

[0021] Convert each vibration time series data into a spectrum including a frequency sequence and an amplitude sequence;

[0022] For each of the frequency spectrums, based on a preset rolling window amplitude difference algorithm, an amplitude difference sequence corresponding to the frequency spectrum is calculated;

[0023] generating a standard frequency sequence corresponding to the spectrum based on a preset frequency accuracy and a maximum frequency of the spectrum;

[0024] The amplitude difference sequence is corrected by using the nearest neighbor method so that each standard frequency in the standard frequency sequence has a corresponding amplitude difference, thereby obtaining a standard amplitude difference sequence corresponding to the frequency spectrum;

[0025] For each standard frequency, multiple amplitude differences corresponding to the standard frequency are counted as a histogram based on a standard amplitude difference sequence to obtain a template consisting of the standard frequency and its corresponding histogram;

[0026] Traverse the standard frequency sequence, and calculate the reference value of the histogram corresponding to each standard frequency in the template. The reference value is the amplitude difference corresponding to the center of the largest interval in the histogram. The reference value corresponding to each standard frequency constitutes a reference value sequence.

[0027] In a possible implementation, the calculation formula of the lifting energy E is as follows:

[0028]

[0029]

[0030] Among them, x n is the amplitude difference with sequence number n in the amplitude difference sequence, T n is the warning value of the template with sequence number n, e n is the power of number n in the interval of numbers i to j, and the sum of the powers is the lifting energy;

[0031] The warning value is the amplitude difference corresponding to the interval whose cumulative sum from small to large in the histogram reaches 99.5%, and the warning value is ≥ the reference value.

[0032] In a possible implementation, the method further includes:

[0033] If the lift energy corresponding to the suspected interval is less than the learning threshold, the amplitude difference sequence corresponding to the spectrum to be detected is combined with the template to form a new template according to the set learning rate.

[0034] In a possible implementation, merging the amplitude difference sequence corresponding to the spectrum to be detected with the template to form a new template according to a set learning rate includes:

[0035] Traverse the standard frequency sequence, if the current amplitude difference corresponding to a standard frequency falls within a certain interval of the histogram of the standard frequency, execute the histogram value update formula;

[0036] The value update formula is: new value = old value * (1-learning rate) + current amplitude difference * learning rate;

[0037] If the amplitude difference falls outside the histogram when the template is updated, the histogram is updated again with a new interval with a value of 0.

[0038] In a second aspect, an embodiment of the present application provides a spectrum floor rise detection device, comprising:

[0039] A data processing module is used to obtain the vibration time series data to be detected of the machine equipment with a preset time length, convert the vibration time series data to be detected into a frequency spectrum to be detected including a frequency sequence and an amplitude sequence; and calculate the amplitude difference sequence corresponding to the frequency spectrum to be detected based on a preset rolling window amplitude difference algorithm;

[0040] A sequence matching module is used to match a predetermined reference value sequence using the amplitude difference sequence, match an amplitude difference interval that is continuously greater than the reference value and has the largest lift energy, and record the frequency interval corresponding to the amplitude difference interval as a suspected interval;

[0041] The lifting judgment module is used to determine the suspected interval as a base lifting interval if the lifting energy corresponding to the suspected interval is greater than an alarm threshold.

[0042] In a third aspect, the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0043] In a fourth aspect, the present application provides a computer-readable storage medium having computer-readable instructions stored thereon, wherein the computer-readable instructions can be executed by a processor to implement the method described in the first aspect.

[0044] Compared with the prior art, the spectrum base lift detection method provided by the present application obtains the vibration time series data to be detected of the preset duration of the machine equipment, and converts the vibration time series data to be detected into a spectrum to be detected containing a frequency sequence and an amplitude sequence; based on the preset rolling window amplitude difference algorithm, the amplitude difference sequence corresponding to the spectrum to be detected is calculated; the amplitude difference sequence is used to match the predetermined reference value sequence, and the amplitude difference interval that is continuously greater than the reference value and has the largest lift energy is matched, and the frequency interval corresponding to the amplitude difference interval is recorded as a suspected interval; if the lift energy corresponding to the suspected interval is greater than the alarm threshold, the suspected interval is determined as a base lift interval. Compared with the prior art, the present application can automatically determine the base lift anomaly in the spectrum and reduce the manpower cost in spectrum analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0046] Figure 1 A flow chart of a spectrum floor lift detection method provided in an embodiment of the present application is shown;

[0047] Figure 2 A schematic diagram of the structure of a spectrum floor lift detection device provided in an embodiment of the present application is shown;

[0048] Figure 3 A schematic diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0049] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0050] It should be noted that, unless otherwise specified, other technical terms or scientific terms used in this application should have the common meanings understood by technicians in the field to which this application belongs.

[0051] In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0052] The present application provides a spectrum floor lift detection method, device, electronic device and storage medium, which are described in detail below with reference to the accompanying drawings.

[0053] Please refer to Figure 1 , Figure 1 : is a flow chart of a spectrum floor lift detection method provided in an embodiment of the present application, comprising the following steps S101 to S106:

[0054] S101, obtaining vibration time series data to be detected of a preset duration of a machine device, and converting the vibration time series data to be detected into a frequency spectrum to be detected including a frequency sequence and an amplitude sequence;

[0055] S102, calculating and obtaining an amplitude difference sequence corresponding to the spectrum to be detected based on a preset rolling window amplitude difference algorithm;

[0056] S103, using the amplitude difference sequence to match a predetermined reference value sequence, matching an amplitude difference interval that is continuously greater than the reference value and has the largest lift energy, and recording a frequency interval corresponding to the amplitude difference interval as a suspected interval;

[0057] S104: If the lifting energy corresponding to the suspected interval is greater than an alarm threshold, the suspected interval is determined as a base lifting interval.

[0058] For ease of understanding, the process of determining the reference value sequence in step S103 is first introduced.

[0059] The process of determining the reference value sequence is as follows:

[0060] 1. Obtain vibration data of the machine equipment in a historical time period, and divide the vibration data in the historical time period into multiple vibration time series data according to a preset time length;

[0061] Specifically, obtain the signal data of a period of time of the vibration sensor installed on the machine equipment, and at the same time obtain the data collection frequency (also called sampling rate) of this vibration sensor; split N vibration time series data according to the conditions of starting the machine equipment and the maximum length of each sequence.

[0062] The maximum length of the sequence can be determined according to the data acquisition frequency and acquisition duration of the vibration sensor.

[0063] 2. Convert each vibration time series data into a spectrum containing a frequency sequence and an amplitude sequence;

[0064] Specifically, each vibration time series data is subjected to Butterworth filtering according to the set filter order and frequency cutoff range, thereby filtering out frequency intervals of no interest; the vibration time series data is converted into a spectrum using fast Fourier transform (FFT), taking only the first half of the spectrum, and the amplitude is converted from a complex number to a real number by taking the absolute value. Each spectrum contains a frequency sequence and an amplitude sequence, and one sequence number corresponds to one frequency and one amplitude.

[0065] For example, the vibration sensor data acquisition frequency is 50000Hz, 100 hours of vibration signal data, the data unit is m / s 2 , the equipment was turned on for 100 hours, and each vibration time series data is 60 seconds long. Then, 6000 60-second data can be obtained by dividing the 100-hour data, and the length of each vibration time series data is 3 million. After filtering, the data length remains unchanged. Take the first half of the spectrum, and finally the length of the frequency sequence and amplitude sequence in each spectrum is 1.5 million, and the range of the frequency sequence is 0-25000Hz.

[0066] 3. For each of the frequency spectra, based on a preset rolling window amplitude difference algorithm, calculate an amplitude difference sequence corresponding to the frequency spectrum;

[0067] Specifically, the rolling window amplitude difference algorithm includes:

[0068] For each spectrum, a rolling window is determined based on a preset frequency accuracy and a frequency sequence difference mean of the spectrum, so that the amplitude sequence of the spectrum executes the rolling window, and the amplitude difference corresponding to the intermediate frequency in each rolling window is calculated to obtain an amplitude difference sequence corresponding to the spectrum;

[0069] The frequency sequence difference mean refers to the mean obtained by subtracting the previous value from the next value of the frequency sequence and then averaging the whole value;

[0070] The amplitude difference refers to the amplitude difference obtained by subtracting the second quantile from the first quantile in the rolling window, and the first quantile is greater than the second quantile. For example, the first quantile is 50% and the second quantile is 1%.

[0071] Specifically, the calculation formulas for the window length width and step length step of the rolling window are as follows:

[0072] width=ceil(3*freq_precision / m);

[0073] step=ceil(freq_precision / m);

[0074] Among them, ceil is rounded up, freq_precision is the mean of frequency sequence differences, and m is the frequency precision.

[0075] In the present application, the amplitude in each rolling window obtains a percentile value according to the set first quantile and second quantile, and the difference between the two percentile values ​​is the rolling window amplitude difference; all rolling window amplitude differences of each spectrum constitute their own amplitude difference sequence.

[0076] 4. Generate a standard frequency sequence corresponding to the spectrum based on a preset frequency accuracy and a maximum frequency of the spectrum;

[0077] For example, if the frequency accuracy is 0.1 Hz and the maximum frequency is 5000 Hz, the standard frequency sequence is a sequence starting with 0.1, with a step size of 0.1, ending at 5000, and a length of 50000. The length of each piece of data may be different, and the frequency sequence will also be different after FFT. The purpose of the standard frequency sequence is to align each piece of data to facilitate subsequent template training, matching, and merging.

[0078] 5. Using the nearest neighbor method to correct the amplitude difference sequence, so that each standard frequency in the standard frequency sequence has a corresponding amplitude difference, and obtaining a standard amplitude difference sequence corresponding to the spectrum;

[0079] Specifically, the amplitude difference sequence of each spectrum is corrected using the nearest neighbor method according to the standard frequency sequence, that is, the standard frequency sequence is traversed, and if a frequency has no corresponding amplitude difference, it is replaced by the amplitude difference of the nearest frequency, thereby generating a new amplitude difference sequence for each spectrum, that is, the standard amplitude difference sequence.

[0080] Assuming that the frequency sequence difference mean of a spectrum is 0.01, and the upper and lower percentiles are 50% and 1%, the amplitude sequence of this spectrum executes a rolling window with a window length of 30 and a step length of 10. For each window, the 50% quantile-1% quantile is calculated as the amplitude difference corresponding to the middle frequency of this window. Due to the rolling window, the length of the newly generated frequency sequence and amplitude difference sequence is 499998, that is, (5000000-30) / 10+1. Then, the nearest neighbor method is used to correct each amplitude difference sequence, and the length of the obtained standard amplitude difference sequence is 50000.

[0081] 6. For each standard frequency, multiple amplitude differences corresponding to the standard frequency are counted as a histogram based on the standard amplitude difference sequence to obtain a template consisting of the standard frequency and its corresponding histogram;

[0082] Specifically, the histogram distribution of multiple amplitude differences of the same frequency is used as a template. Traverse the standard frequency sequence, all amplitude difference sequences form a sequence at the frequency, and calculate the histogram of this sequence as the template value of the frequency. The standard frequency sequence and its corresponding histogram sequence are the template.

[0083] Exemplarily, the template includes a frequency sequence and a histogram sequence with a length of 50,000. The number of histogram intervals is initially set to 50. The frequency sequence is a standard frequency sequence, and the histogram sequence is the normalized result of grouping amplitude difference sequences with a length of 6,000 each. Normalization means that the sum of the values ​​of each histogram is 1.

[0084] 7. Traverse the standard frequency sequence, and calculate the reference value of the histogram corresponding to each standard frequency in the template. The reference value is the amplitude difference corresponding to the center of the largest interval in the histogram. The reference value corresponding to each standard frequency constitutes a reference value sequence.

[0085] Traverse the standard frequency sequence, and calculate the reference value and warning value of the histogram corresponding to each frequency in the template. The reference value is the amplitude corresponding to the center of the interval with the largest histogram value, and the warning value is the amplitude corresponding to the interval where the cumulative sum of the histogram amplitude intervals from small to large reaches 99.5%. If the warning value is ≥ the reference value, the reference value sequence and warning value sequence of the template can be calculated.

[0086] In step S101, the signal data of a preset time length of the vibration sensor installed on the machine equipment is obtained as the vibration time series data to be detected. The vibration time series data to be detected is converted into a spectrum using Fast Fourier Transform (FFT), and only the first half of the spectrum is taken to obtain the spectrum to be detected.

[0087] In step S102, based on the preset rolling window amplitude difference algorithm, the amplitude difference sequence corresponding to the spectrum to be detected is calculated. Exemplarily, in real-time spectrum analysis detection, the latest vibration time series data is obtained and converted into an amplitude difference sequence, and the length of this amplitude difference sequence is 50,000.

[0088] In step S103, the amplitude difference sequence corresponding to the spectrum to be detected is matched with a predetermined reference value sequence to match an amplitude difference interval that is continuously greater than the reference value and has the largest lift energy, and the frequency interval corresponding to the amplitude difference interval is recorded as a suspected interval.

[0089] Specifically, traverse the standard frequency sequence, if the value of the amplitude difference corresponding to the latest part of a frequency is greater than the reference value corresponding to the frequency in the template, mark the frequency number; obtain the interval with continuous marked numbers; calculate the lift energy of any sub-continuous interval in the continuous interval, if there is no continuous interval, add a new interval with a lift energy of 0 and a number interval of 0 to the length of the standard frequency sequence. Output the interval with the maximum lift energy among these (sub) continuous intervals as the suspected interval.

[0090] Specifically, the calculation formula of the lifting energy E is as follows:

[0091]

[0092]

[0093] Among them, x n is the amplitude difference with sequence number n in the amplitude difference sequence, T n is the warning value of the template with sequence number n, e n is the power of number n in the interval of numbers i to j, and the sum of the powers is the lifting energy.

[0094] For example, first calculate the baseline value sequence and warning value sequence of the template, and then calculate the continuous interval that is greater than the baseline value sequence. Assuming that there is only one continuous interval and all values ​​in this interval are greater than the warning value, there is no need to calculate the energy of the sub-continuous interval, and directly calculate the lift energy of this interval. This interval is the interval to be output.

[0095] In step S104, if the lift energy corresponding to the suspected interval is greater than the alarm threshold, the suspected interval is determined as a base lift interval. In practical applications, if the lift energy of the suspected interval is greater than the alarm threshold, the base lift interval and lift energy are output.

[0096] The embodiments of the present application can automatically determine the base lift anomaly in the spectrum, thereby reducing the manpower cost in spectrum analysis.

[0097] In some embodiments, the above method may further include the following steps:

[0098] If the lift energy corresponding to the suspected interval is less than the learning threshold, the amplitude difference sequence corresponding to the spectrum to be detected is combined with the template to form a new template according to the set learning rate.

[0099] The process of updating the template according to the set learning rate includes:

[0100] Traverse the standard frequency sequence, if the current amplitude difference corresponding to a standard frequency falls within a certain interval of the histogram of the standard frequency, execute the histogram value update formula;

[0101] The value update formula is: new value = old value * (1-learning rate) + current amplitude difference * learning rate; the old value refers to the value in the template before updating, and the new value refers to the value in the template after updating;

[0102] If the amplitude difference falls outside the histogram when the template is updated, the histogram is updated again with a new interval with a value of 0.

[0103] After updating, each histogram is normalized, that is, each interval value of each histogram is divided by the cumulative sum of the histogram so that the sum of the histogram values ​​is 1.

[0104] For example, assuming the learning threshold is 0.4m 2 / s 4 , the alarm threshold is 0.5m 2 / s 4 The suspected interval number is 2000-5000, and the lifting energy is 0.8m 2 / s 4 , so an alarm needs to be output. The final output base lift range is 200.1Hz~500.1Hz and the lift energy is 0.8m 2 / s 4 .

[0105] The embodiments of the present application can not only automatically determine the base lift anomaly in the spectrum and reduce the manpower cost in spectrum analysis, but also update the template in real time to adapt to the slow changes in data caused by equipment aging, etc.

[0106] In the above-mentioned embodiment, a spectrum base lift detection method is provided, and correspondingly, the present application also provides a spectrum base lift detection device. The spectrum base lift detection device provided in the embodiment of the present application can implement the above-mentioned spectrum base lift detection method, and the spectrum base lift detection device can be implemented by software, hardware, or a combination of software and hardware. For example, the spectrum base lift detection device can include integrated or separate functional modules or units to perform the corresponding steps in the above-mentioned methods. Please refer to Figure 2 As shown, the spectrum floor lift detection device 10 includes:

[0107] The data processing module 101 is used to obtain the vibration time series data to be detected of the machine equipment with a preset time length, convert the vibration time series data to be detected into a spectrum to be detected including a frequency sequence and an amplitude sequence; and calculate the amplitude difference sequence corresponding to the spectrum to be detected based on a preset rolling window amplitude difference algorithm;

[0108] A sequence matching module 102 is used to match a predetermined reference value sequence using the amplitude difference sequence, match an amplitude difference interval that is continuously greater than the reference value and has the largest lift energy, and record the frequency interval corresponding to the amplitude difference interval as a suspected interval;

[0109] The lifting judgment module 103 is configured to determine the suspected interval as a base lifting interval if the lifting energy corresponding to the suspected interval is greater than an alarm threshold.

[0110] In a possible implementation, the rolling window amplitude difference algorithm includes:

[0111] For each spectrum, a rolling window is determined based on a preset frequency accuracy and a frequency sequence difference mean of the spectrum, so that the amplitude sequence of the spectrum executes the rolling window, and the amplitude difference corresponding to the intermediate frequency in each rolling window is calculated to obtain an amplitude difference sequence corresponding to the spectrum;

[0112] The frequency sequence difference mean refers to the mean obtained by subtracting the previous value from the next value of the frequency sequence and then averaging the whole value;

[0113] The amplitude difference refers to the amplitude difference obtained by subtracting the second quantile from the first quantile in the rolling window, and the first quantile is greater than the second quantile.

[0114] In a possible implementation, the calculation formulas for the window length width and step length step of the rolling window are as follows:

[0115] width=ceil(3*freq_precision / m);

[0116] step=ceil(freq_precision / m);

[0117] Among them, ceil is rounded up, freq_precision is the mean of frequency sequence differences, and m is the frequency precision.

[0118] In a possible implementation, the process of determining the reference value sequence is as follows:

[0119] Acquire vibration data of the machine equipment in a historical time period, and divide the vibration data in the historical time period into multiple vibration time series data according to a preset time length;

[0120] Convert each vibration time series data into a spectrum including a frequency sequence and an amplitude sequence;

[0121] For each of the frequency spectrums, based on a preset rolling window amplitude difference algorithm, an amplitude difference sequence corresponding to the frequency spectrum is calculated;

[0122] generating a standard frequency sequence corresponding to the spectrum based on a preset frequency accuracy and a maximum frequency of the spectrum;

[0123] The amplitude difference sequence is corrected by using the nearest neighbor method so that each standard frequency in the standard frequency sequence has a corresponding amplitude difference, thereby obtaining a standard amplitude difference sequence corresponding to the frequency spectrum;

[0124] For each standard frequency, multiple amplitude differences corresponding to the standard frequency are counted as a histogram based on a standard amplitude difference sequence to obtain a template consisting of the standard frequency and its corresponding histogram;

[0125] Traverse the standard frequency sequence, and calculate the reference value of the histogram corresponding to each standard frequency in the template. The reference value is the amplitude difference corresponding to the center of the largest interval in the histogram. The reference value corresponding to each standard frequency constitutes a reference value sequence.

[0126] In a possible implementation, the calculation formula of the lifting energy E is as follows:

[0127]

[0128]

[0129] Among them, x n is the amplitude difference with sequence number n in the amplitude difference sequence, T n is the warning value of the template with sequence number n, e n is the power of number n in the interval of numbers i to j, and the sum of the powers is the lifting energy;

[0130] The warning value is the amplitude difference corresponding to the interval whose cumulative sum from small to large in the histogram reaches 99.5%, and the warning value is ≥ the reference value.

[0131] In a possible implementation, the lifting judgment module is further used to:

[0132] If the lift energy corresponding to the suspected interval is less than the learning threshold, the amplitude difference sequence corresponding to the spectrum to be detected is combined with the template to form a new template according to the set learning rate.

[0133] In a possible implementation, the lifting judgment module is specifically used to:

[0134] Traverse the standard frequency sequence, if the current amplitude difference corresponding to a standard frequency falls within a certain interval of the histogram of the standard frequency, execute the histogram value update formula;

[0135] The value update formula is: new value = old value * (1-learning rate) + current amplitude difference * learning rate;

[0136] If the amplitude difference falls outside the histogram when the template is updated, the histogram is updated again with a new interval with a value of 0.

[0137] The spectrum floor rise detection device provided in the embodiment of the present application and the spectrum floor rise detection method provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0138] An embodiment of the present application also provides an electronic device corresponding to the method provided in the above embodiment, and the electronic device may be an electronic device for a client, such as a mobile phone, a laptop computer, a tablet computer, a desktop computer, etc., to perform the above-mentioned spectrum floor lift detection method.

[0139] Please refer to Figure 3 , which shows a schematic diagram of an electronic device provided by some embodiments of the present application. Figure 3 As shown, the electronic device 20 includes: a processor 200, a memory 201, a bus 202 and a communication interface 203, and the processor 200, the communication interface 203 and the memory 201 are connected via the bus 202; the memory 201 stores a computer program that can be run on the processor 200, and when the processor 200 runs the computer program, it executes the spectrum floor lift detection method provided in any of the aforementioned embodiments of the present application.

[0140] The memory 201 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 203 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0141] The bus 202 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 201 is used to store a program, and the processor 200 executes the program after receiving an execution instruction. The spectrum floor lift detection method disclosed in any implementation of the embodiment of the present application may be applied to the processor 200, or implemented by the processor 200.

[0142] The processor 200 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 200. The above processor 200 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a readily available programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 201, and the processor 200 reads the information in the memory 201 and completes the steps of the above method in combination with its hardware.

[0143] The electronic device provided in the embodiment of the present application and the spectrum floor lift detection method provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0144] An embodiment of the present application also provides a computer-readable storage medium corresponding to the spectrum floor rise detection method provided in the aforementioned embodiment, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the spectrum floor rise detection method provided in any of the aforementioned embodiments.

[0145] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0146] The computer-readable storage medium provided in the above-mentioned embodiments of the present application and the spectrum floor rise detection method provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and they should all be included in the scope of the claims and specification of the present application.

Claims

1. A spectrum floor lift detection method, characterized in that: include: Acquire the vibration time series data to be detected of the machine equipment with a preset time length, and convert the vibration time series data to be detected into a frequency spectrum to be detected including a frequency sequence and an amplitude sequence; Based on a preset rolling window amplitude difference algorithm, an amplitude difference sequence corresponding to the spectrum to be detected is calculated; Using the amplitude difference sequence to match a predetermined reference value sequence, matching an amplitude difference interval that is continuously greater than the reference value and has the largest lift energy, and recording the frequency interval corresponding to the amplitude difference interval as a suspected interval; If the lifting energy corresponding to the suspected interval is greater than the alarm threshold, the suspected interval is determined as a base lifting interval.

2. The method according to claim 1, characterized in that: The rolling window amplitude difference algorithm includes: For each spectrum, a rolling window is determined based on a preset frequency accuracy and a frequency sequence difference mean of the spectrum, so that the amplitude sequence of the spectrum executes the rolling window, and the amplitude difference corresponding to the intermediate frequency in each rolling window is calculated to obtain an amplitude difference sequence corresponding to the spectrum; The frequency sequence difference mean refers to the mean obtained by subtracting the previous value from the next value of the frequency sequence and then averaging the whole value; The amplitude difference refers to the amplitude difference obtained by subtracting the second quantile from the first quantile in the rolling window, and the first quantile is greater than the second quantile.

3. The method according to claim 2, characterized in that The calculation formulas for the window length width and step length step of the rolling window are as follows: width=ceil(3*freq_precision / m); step=ceil(freq_precision / m); Among them, ceil is rounded up, freq_precision is the mean of frequency sequence differences, and m is the frequency precision.

4. The method according to claim 2, characterized in that: The process of determining the reference value sequence is as follows: Acquire vibration data of the machine equipment in a historical time period, and divide the vibration data in the historical time period into multiple vibration time series data according to a preset time length; Convert each vibration time series data into a spectrum including a frequency sequence and an amplitude sequence; For each of the frequency spectrums, based on a preset rolling window amplitude difference algorithm, an amplitude difference sequence corresponding to the frequency spectrum is calculated; generating a standard frequency sequence corresponding to the spectrum based on a preset frequency accuracy and a maximum frequency of the spectrum; The amplitude difference sequence is corrected by using the nearest neighbor method so that each standard frequency in the standard frequency sequence has a corresponding amplitude difference, thereby obtaining a standard amplitude difference sequence corresponding to the frequency spectrum; For each standard frequency, multiple amplitude differences corresponding to the standard frequency are counted as a histogram based on a standard amplitude difference sequence to obtain a template consisting of the standard frequency and its corresponding histogram; Traverse the standard frequency sequence, and calculate the reference value of the histogram corresponding to each standard frequency in the template. The reference value is the amplitude difference corresponding to the center of the largest interval in the histogram. The reference value corresponding to each standard frequency constitutes a reference value sequence.

5. The method according to claim 4, characterized in that The calculation formula of the lifting energy E is as follows: Among them, x n is the amplitude difference with sequence number n in the amplitude difference sequence, T n is the warning value of the template with sequence number n, e n is the power of number n in the interval of numbers i to j, and the sum of the powers is the lifting energy; The warning value is the amplitude difference corresponding to the interval whose cumulative sum from small to large in the histogram reaches 99.5%, and the warning value is ≥ the reference value.

6. The method according to claim 4, characterized in that The method further comprises: If the lift energy corresponding to the suspected interval is less than the learning threshold, the amplitude difference sequence corresponding to the spectrum to be detected is combined with the template to form a new template according to the set learning rate.

7. The method according to claim 6, characterized in that The step of merging the amplitude difference sequence corresponding to the spectrum to be detected with the template to form a new template according to the set learning rate includes: Traverse the standard frequency sequence, if the current amplitude difference corresponding to a standard frequency falls within a certain interval of the histogram of the standard frequency, execute the histogram value update formula; The value update formula is: new value = old value * (1-learning rate) + current amplitude difference * learning rate; If the amplitude difference falls outside the histogram when the template is updated, the histogram is updated again with a new interval with a value of 0.

8. A spectrum floor lift detection device, characterized in that: include: A data processing module is used to obtain the vibration time series data to be detected of the machine equipment with a preset time length, convert the vibration time series data to be detected into a frequency spectrum to be detected including a frequency sequence and an amplitude sequence; and calculate the amplitude difference sequence corresponding to the frequency spectrum to be detected based on a preset rolling window amplitude difference algorithm; A sequence matching module is used to match a predetermined reference value sequence using the amplitude difference sequence, match an amplitude difference interval that is continuously greater than the reference value and has the largest lift energy, and record the frequency interval corresponding to the amplitude difference interval as a suspected interval; The lifting judgment module is used to determine the suspected interval as a base lifting interval if the lifting energy corresponding to the suspected interval is greater than an alarm threshold.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: Computer-readable instructions are stored thereon, and the computer-readable instructions can be executed by a processor to implement the method according to any one of claims 1 to 7.

Images