TW462900B - Vibration diagnosis method for abnormality or malfunctions of roller of hot rolling machine for steel strip - Google Patents

Abstract

This invention relates to a vibration diagnosis method for abnormality or malfunctions of roller of hot rolling machine for steel strip, mainly comprising in sequence vibration signals measuring step; signal calculating step; determining step for effective value and peak-to-peak value of amplitude; effective frequency center, comprehensive determining step for peak frequency, equivalent frequency center, amplitude modulation frequency, and amplitude modulation variation; determining step for primary peak sharpness of secondary frequency spectrum; and comprehensive determining step for equivalent frequency center and frequency concentration, so as to diagnose the characteristic parameters of time domain, frequency domain, and amplitude modulation of each vibration signal. Then, depending upon the combination of the multiple-parameter discriminant, the abnormal condition the working roller and backing roller being either working roller vibration phenomenon, poor lubrication of working roller bearing, abrasion of main shaft coupler, scratches of backing roller surface, or abnormal loading of roller bearing can be determined.

Description

462900

V. Description of the invention (1) The present invention relates to a kind of equipment, especially to a steel strip hot rolling mill roll which can effectively judge the location and cause. = Frequent or faulty vibration diagnosis method ^ Break out the abnormal or faulty vibration diagnosis method of the hot rolling mill Press the 'hot stick and back stick are the most important equipment in the hot rolling production line' and the reason Because the steel strip is cut here for the last time, most of the steel strip size and surface quality are determined here; @ 日 寺, because the hot rolling mill is the lifeblood of the steel plant's output and profit, once the condition of the equipment needs Stopping and overhauling will delay production and cause great losses. Therefore, the condition of the work rolls and back rolls is not only directly related to the rolling output ', but also has a decisive impact on the surface quality of the steel strip; please refer to the first picture and the second Figure, which is a schematic diagram of the structure of the conventional hot rolling mill and its drive system, as shown in the figure, wherein the work roll group 10 can be divided into upper work roll 100 and lower work roll 101, and the work The function of the roller group 10 is to thin the steel strip, and the back roller group 20 can also be divided into the upper back roller 20 and the lower report 202. As for the other functions, it is to transmit and support the rolling force. Due to the surface quality of the steel belt, it is affected by the work roll group 10 and the back roll group 2 The surface condition, the roundness of such grinding, the condition of the bearings of the work roll set 10, the gap between the work roll 『10 and the hot-roller base, and the stability of the back light set 〇 fixing stability and many other factors' It is not easy to grasp and control. In order to reduce the above-mentioned steel belt quality defects caused by equipment abnormalities, and to avoid the loss of production stoppage due to sudden equipment failure, a set of "hot rolling mill work rolls and back roll equipment vibration is precious The "rule" automatically determines the condition of the equipment by measuring the vibration signal, so that when the site operator finds that the equipment is abnormal, it can be immediately eliminated.

Except for 'to avoid continuous rolling out of defective products that are gradually degraded and are about to be damaged, or replace them'. To avoid different descriptions of equipment such as poor vibration diagnosis methods and bearing defects during rolling, rather than small tests within quantitative criteria. The equipment is more complicated, and the defects of the actual equipment are much more complicated. Therefore, the abnormal position and cause of the work roll set 10 and the back roll of the heat engine must be used. However, a large number of rejections were scheduled in advance for scheduled downtime. Focusing on abnormal quality conditions, the diagnosis is that these methods mainly generate artificial vibration diagnosis groups 20 of abnormal types and conditions manually, often when there is no equipment found and repaired on a time-by-time basis. The rules of mind are mostly qualitative, and the defects obtained from the laboratory are used to perform the modeling. The simple manual method is used to accurately diagnose the on-site method. In view of this, in order to break through this difficulty and bottle collar, the inventor of this case relied on his many years. The experience of the hot rolling mill and the two aspects related to vibration detection and analysis have been carefully researched and designed. After continuous improvement, the invention has finally been developed. Therefore, the main purpose of the present invention is to provide a method for diagnosing abnormal or faulty L rolls of a steel strip hot rolling mill, so that it can effectively judge the equipment status of work rolls and back rolls, and diagnose the parts where the abnormality occurs. And the cause. π _According to the above purpose, the main feature of the present invention is to measure the vibration signal of μ at a timing measurement, and analyze the two signs of the time domain, frequency domain and amplitude modulation signals, and find various abnormalities through long-term statistics and tracking. Correspondence between conditions and characteristics of vibration signals; then develop a number of quantitative parameters that are representative of equipment abnormality and fault conditions, and based on the results of statistical analysis, determine the critical values of each parameter, and summarize the quantitative diagnostic rules for various abnormal causes

4629 0

3. Description of the invention (3). Finally, these diagnostic rules are integrated into a complete Ling and computer program. At the same time, the lightning brain 9 breaks the process, and then develops the measurement signal, which can be attributed to: “Cause. Automatically analyze the characteristics of the vibration signal and further diagnose the abnormality of the roller. The method of the two-break method of vibrating gas in the roll of the steel strip hot rolling mill according to the present invention mainly includes the vibration signal amplitude and vibration diagnosis sequence in sequence. Effective value and peak-to-peak judgment ': Frequency :: Judgment step' Judgment of peak frequency, equivalent frequency center concentration judgment '' Secondary spectrum ... sharpness; tuning; time domain characteristic parameters of variational dynamic signals *, frequency domain Special: The parameter letter "diagnostics of each back vibration is strong." The combination of parameter value discriminant formulas, the vehicle is determined by "spindle coupling wear", "poor back bearing lubrication abnormal load" and other abnormal conditions. "," And "Working rollers, for a month of the present, I have a firm understanding, and here is the second part of the present invention: the characteristics are more detailed and clear: The preferred embodiment and the description of the drawings are as follows. First, please refer to the first Three pictures

= All: 'Vibration diagnosis shown in this example =; = = Peak motion signal measurement step 1, signal calculation step 2, break 1 = judgment step 3' frequency concentration judgment step 4 ISC frequency, equivalent frequency center 、 Step of adjusting center 42 Step of comprehensive judgment of frequency and amplitude variation of sharp peak 碉 %% 6, Step of judgment of wave height ratio 51 1. The step of determining the sharpness of the main peak of the secondary frequency spectrum The step of comprehensive judgment of equivalent frequency center and frequency concentration 7, and the inverse frequency 4 main peak sharpness judgment step 7i and other steps; wherein, the vibration signal measurement step] is to periodically measure the vibration signal of the work roll bearing of the hot rolling mill; and then the signal calculation step 2 is used to calculate the foregoing. The time-domain characteristics of the vibration signal, the number of frequency-domain characteristic parameters, and the amplitude-modulation characteristic parameter values, the two types of parameters are described separately as follows: 1. Each time-domain characteristic parameter is described. -Wave Temple :: Φ ref: IV: Time of vibration signal received by workers and relatives. Amazing amplitude effective value ums) 'Amplitude peak pair, and wave height rate (Crfct) three tables ^ C Pk). The value of the sub-item parameter is used as the equipment diagnosis. If Y = {y 丨, y2}, the array 'then each time domain characteristic parameter n number c momentum measurement theory meaning is as follows: < 疋 义 and the thing it represents (1) Amplitude effective value (Rms): instead of the standard deviation, which is defined as the vibration signal

Rms

N ′ is less. ) ...... rs = 1 ...... where y〇 represents the average value of the vibration signal

I N less 0 Han U brother ........... 4629

V. Description of the invention (5) (2) Amplitude peak-to-peak value (Pkpk): represents the change range of the vibration signal from maximum to maximum. If max (Y) is used to represent the vibration signal γ = {l '',, ..., Υν Maximum value, mi η (Υ) represents the minimum value, then & 'I

Pkpk = max (F)-min (F) (3) Wave height ratio (Crfct): shows whether the vibration signal has a shock wave and the intensity of the wave, which is defined as engine

Crfct = Pkpk / 2 / Rms 2 · Frequency-domain characteristic parameters Frequency-domain characteristic parameters are based on the frequency spectrum of the vibration signal, and calculate three parameters: the sharp frequency (Fpeak), the equivalent frequency center (FreqCntr), and the frequency = moderate (Fcon). Value as an indicator for device diagnostics. 2 ^ 1 'Sl)' (f 2, S2), ... '(f M,%) 丨 represents the spectral line in the vibration frequency error, that is, (f 丄, \) represents the frequency f; the amplitude at 3 The definition of each frequency domain characteristic parameter and the physical meaning it represents, such as (1) Peak frequency (FPeak): the largest point in the frequency spectrum, if s ·

Ui 丨 u a X s], s2 ’…’ sM}) then Fpeak = f 丨 (2) Equivalent frequency center (FreqCntr): represents the effective center of gravity position of the spectral line distribution in the frequency spectrum. The method is as follows:

5. In the description of the invention (6), the threshold value is set to 10% of the peak frequency Fpeak), and the height of the spectral line is lower than this gate root value, it is regarded as noise and ignored. Secondly, find the peaks in the frequency spectrum, which are defined as follows: Right 5b 2 <5b 丨 <\ and \ &gt;5; + 1 &gt; 3 丨 +2 then (1 ^ '5 丨) is a peak, and each peak is The composed array is represented by {(pf], pS]), (pf2, pS2), ..., (P f η, P Sn) 丨, then the equivalent frequency center is defined as η Σ㈣ χΜ)

FreqCntr =-^ ―............................. ......... (5)

YuPh • ^ 9 ^. Ϊ '^ = Ι (1) Frequency concentration (Fcon) · Represents the degree of frequency concentration in the frequency spectrum, that is, to determine whether the spectral line is concentrated or dispersed, as a diagnosis of flutter The index of the phenomenon is calculated as follows: As in the preceding paragraph, {(pf〗, pSi), (pf2, ps2), ..., (Pfn 'psn) 丨 represents each peak in the spectrum; first find the peak frequency Fpeak in the spectrum , And set a frequency bandwidth BWF (in the preferred embodiment of the present invention, it is set to 20Hz), and then find the peak and psum in the frequency range centered on Fpeak's BWF and the overall peak • value and T sum Ratio, that is,

Psum = Yjps ...... f〇r (Fpeak-BWF) &lt; Pfi ^ (Fpeak + BWF) ................ .... (6) η

Tsum = Y ^ psi ............................. ........................................ .......... (7) i = i

Fcon = Psum / Tsum x 100% ......................... ..........................................(8)

Page 9 1. Characteristic parameters of amplitude modulation 46 V. Description of the invention (7) Some equipment abnormal conditions will cause vibration signal waves or cause periodic I degree modulation; as shown in the figure, they are the invention A schematic diagram of the vibration signal of a better real-time shock wave and a schematic diagram of the variable vibration signal; while the amplitude modulation is special, the amplitude modulation frequency (or shock wave repetition frequency) of the sharpness of the main peak of the inverse spectrum and the main peak of the secondary spectrum are It means that the amplitude of the vibration signal is not a fixed change (usually a periodic change); analyzing the repetition frequency of the shock and the variation of the amplitude modulation can provide considerable help to the cause. In a broad sense, the periodic shock wave also belongs to the amplitude and therefore the amplitude modulation phenomenon of the amplitude modulation frequency and the outgoing frequency of the shock wave repetition frequency can be reflected by the phenomenon that there are multiple high-power various waves in the anti-spectrum ^ Ν ″; Regarding the amplitude modulation frequency (the degree of ΪΓΤ sharpness is as follows: meaning, calculation method and property

(1) Amplitude modulation frequency 皋 r + U or shock wave repetition frequency): The so-called amplitude_g is time-dependent (the amplitude of the vibration signal refers to the periodic variation of the vibration signal through the Λ) shock wave repetition frequency ^ The degree of complex frequency in unit time refers to the period of impulse generation per unit time. As shown in the example in the fourth figure, there is a periodic characteristic parameter system, and the amplitude of the amplitude modulation becomes sharp; There is a method to find the phenomenon of abnormal modulation with so-called amplitude modulation over time, which includes the degree of variation. The significant peak response can be determined by the secondary frequency spectrum or the shock wave repetition frequency and the main frequency of the secondary frequency spectrum. It is said that the degree is not a fixed value, but the number of changes in amplitude modulation and frequency, and the number of occurrences of shock waves. 5. Description of the invention (8) • Analysis of the above frequency method is to first filter the carrier frequency of the vibration, and then This spectrum or shock wave repeats (2) AM modulation variability: The analysis of shock variability is helpful for "quotation", and the amplitude modulation of AM amplitude variation is the method of its hill rate Using the envelope spectrum analysis method, the dynamic signal is rectified and low-pass filtered, and the high frequency ^ 'leaves the low-frequency amplitude-modulated signal', which is the so-called envelope envelope, is used for spectrum analysis to obtain the envelope spectrum 'at the maximum peak' Learn the AM ^ frequency of the vibration signal. "The repetition frequency of the wave or amplitude modulation and the amplitude can provide a considerable amount of ^ anomalies in the diagnosis of the cause of the abnormality of the device. First, the Hilbert transformation is used to find the waveform of the vibration signal: if y (t) represents the time wave of the vibration signal Burt conversion% (1 :) is defined as y ^ t) -dr, t- ·. (9) Using y (t) and yh (t), the vibration signal t amplitude modulation waveform A ( t) ............ Xiong Xiu uses the sixth chart as an example to explain the meaning of the amplitude modulation wave form, then the sixth chart is a vibration signal whose amplitude is modulated; using formula (9) And (10), the amplitude modulation waveform can be obtained as shown in Fig. 6 (b).

4 62 V. Description of the invention (9) A (t); the curve indicated by the arrow in the sixth figure (a) is the curve of &gt; Figure ^); this curve is the vibration signal in the sixth figure U) 'Outline' describes the signal amplitude modulation. Because the magnitude of the amplitude of the vibration signal can obviously reflect the equipment shape, so a quantitative index can represent the degree of amplitude modulation. This indicator is called the amplitude modulation variability Amdev 'and its definition is as follows: The maximum and minimum values of the amplitude modulation waveform A (t) in Figure 6 (b), then

Amdev = (max- min) / (max + min) x 100% ... The above formula defines that for a vibration signal whose amplitude does not change with time, its amplitude modulation variability Amdev is close to 0; otherwise, as in the waveform in Figure 6, the amplitude variability Amdev is close to 丄 〇 (3) inverse Spectral peak sharpness (CepsPeakV): This parameter is used to determine whether there are significant spikes in the inverse spectrum of the vibration signal. At the same time, the amplitude modulation frequency of the frequency can also be reflected by the significant spikes in the inverse frequency Ip. The sharpness of the main peak of the frequency spectrum is used as a quantitative indicator for diagnosing whether "slipping between the inner ring of the work roll bearing and the roll neck" has occurred. Inverse spectrum is defined as taking the logarithm of the spectrum and then performing a Fourier transform. That is, the spectrum of the spectrum; first find the main spikes in the inverse spectrum; if Ceps [i] represents the spectral lines in the inverse spectrum, and

Page 12 5. Description of the invention (10) The maximum inverse spectrum value is Ceps [m], which is the main peak. For a set number of calculation points N, the sharpness of the inverse spectrum main peak can be calculated according to the formula CepsPeakV:

CepsPeakV = | — X [Ceps [m \-Ceps [m + 2 + /]] 41 ......... (12) (4) Quadratic Spectrum Master Peak Spectral Sharpness (DspecPeakV): This parameter is used to determine whether the frequency spectrum of the vibration signal has bearing Λ frequency and its multiple high harmonics, as a quantitative indicator for diagnosing whether the abnormal load of the roller bearing occurs. The frequency range of 0 ~ 1 250Hz is taken out, and then two consecutive spectrum conversions are performed for this spectrum. The obtained is called the secondary spectrum. The main purpose of this signal processing is: when there are multiple harmonic peak frequencies in the spectrum After the above conversion, a significant peak will appear in the secondary spectrum. If Dspec [i] is used to represent each spectral line in the secondary spectrum, and the maximum value of the secondary spectrum is when i = π, then Dspec [m] is Is the main peak. For a set number of different points N ', the secondary spectrum main peak sharpness DspecPeakV can be calculated according to the following formula:

Page 13 462

The peak judgment step 3 is to judge whether the above data is greater than UG (G is the acceleration of gravity, == = == the effect value is whether the peak-to-peak value is greater than 3G, if it is, then ^ or Zhenming μ,-τ is ordered Frequency concentration judgment step 4 'No right, it is judged that the equipment condition is normal and the diagnosis is ended. 隼 中 = Concentration judgment step 4 is to judge that the frequency concentration of the above-mentioned vibration signal J is greater than 40%, then another peak frequency is sequentially performed. Judgment step 41, = and a comprehensive judgment step 42 of amplitude modulation frequency and amplitude modulation variation, if no =, the peak frequency, the equivalent frequency center, the modulation rate and the change are changed. Step 5: and the main peak of the secondary spectrum is sharp Step 6 and 4 comprehensive frequency center and frequency concentration degree comprehensive judgment step 7; the peak frequency judgment step 41 and the amplitude modulation frequency and amplitude modulation variation comprehensive determination step 42 are as follows: ”1. If the vibration signal The peak frequency is less than 40 0Ηζ, and the condition that the amplitude modulation frequency 2 is between 1 and 8 Hz, and the amplitude variation is greater than 40%, then it is judged that the 5 is also a low-frequency chatter and has obvious amplitude modulation. 2. Peak frequency If the rate is less than 400Ηζ, but does not meet the conditions that the AM frequency is between “redundant” and “AM” variability is greater than 40%, the device is judged to be a low-frequency chattering phenomenon. 3. If the peak frequency of the vibration signal is between 4〇 {^ and 80〇1 ^, and satisfy the condition that the amplitude modulation frequency is between 1 and 8ΗZ, and the amplitude variation is greater than 40%, then the device is judged to be high frequency flutter and has obvious amplitude modulation. Phenomenon 4. If the peak frequency is between 400Hz and 800Hz, but does not meet the conditions that the amplitude modulation frequency is between 1 and 8Hz, and the amplitude variation is greater than 40%, then judge

Page 14 V. Description of the invention (12) The device is defined as a high-frequency chattering phenomenon. 5. If the peak frequency of the vibration signal is greater than 12,000 Hz, and the conditions that the amplitude modulation frequency is between 1 and 8 Hz, and the amplitude modulation variation is greater than 40%, the device is judged to have extremely high frequency chattering and obvious Amplitude modulation phenomenon. 6‘If the peak frequency is greater than 1200Hz, but the AM frequency is between 1 and 8Hz, and the AM variation is greater than 40%, then the device is judged to be a south-south frequency vibration phenomenon. 7. If the peak frequency of the vibration signal is between gooHz and 1200Hz, then it is determined that the condition of the equipment cannot be determined and the diagnosis is ended. … The frequency concentration determination step 4 determines that if the frequency concentration of the above-mentioned vibration signal is less than 40%, the comprehensive determination step of the sharp frequency, equivalent frequency center, amplitude modulation frequency, and amplitude modulation variation step 5 is to determine the above-mentioned vibration signal Whether it meets (a) the peak frequency is less than 70 0Hz, (b) the center of the equivalent frequency is less than 1300Hz, (c) the amplitude modulation frequency is between iHz and 16Hz, (d) the amplitude modulation variation is greater than 40%, if not, no judgment is made If yes, then perform another wave height rate determination step 5 1 'It is to determine whether the wave height rate of the above-mentioned vibration signal is greater than 5; if yes, then judge the condition of the equipment as the surface of the back roller is scratched; if not, judge the equipment condition is the spindle coupling Shaft wear. _ In addition, if the frequency concentration of the vibration signal is less than 40%, the secondary spectrum main peak sharpness judgment step 6 is to determine whether the secondary spectrum main peak sharpness of the vibration signal is greater than 68%, and if so, determine the device The condition is that the load of the work roll bearing is abnormal. If not, no judgment is made; at the same time, 'the equivalent frequency center and frequency concentration comprehensive judgment step 7 also judges whether the above-mentioned vibration signal satisfies the equivalent frequency center greater than 1 100 Hz and the frequency is concentrated. Degree small

Page 15 4 62 9 v. V. Description of the invention (13) At 10%, if yes, then perform the inverse spectrum main peak sharpness judgment step 71; if not, no judgment is made; the inverse spectrum main peak sharpness judgment step 71 is to judge the sharpness of the main peak of the inverse spectrum of the vibration signal, as follows: (1) If the sharpness of the main peak of the inverse spectrum of the vibration signal is less than 80%, then the condition of the equipment is caused by bearing oil loss or water ingress. Poor lubrication. (2) If the sharpness of the main peak of the inverse spectrum of the vibration signal is between 80% and 85% ’, it is judged that the condition of the equipment is poor bearing lubrication, or scratches between the roller neck and the inner ring of the bearing. (3) If the sharpness of the main peak of the anti-spectrum of the vibration signal is greater than 85%, it is judged that the equipment condition is dry friction and scratch caused by oil loss between the roller neck and the inner ring of the bearing. Therefore, the diagnosis method of the present invention is used to diagnose The time domain characteristic parameter value, frequency domain characteristic parameter value and amplitude modulation characteristic parameter value of each vibration signal, and then according to a combination of multiple parameter value discriminants, determine whether the status of the work rolls and rolls of the hot rolling mill is "work roll chatter Phenomenon ”,“ Working roller bearing lubrication is not good, “Spindle coupling wear”, “Back roller surface is damaged”, and “Legative”, etc.%; field operators can take precautions in advance based on the diagnosis results.辄 Measures, once the equipment is found to be different, it can be eliminated to avoid continuous rejection of defective products and being rejected in large quantities. When the existing equipment gradually deteriorates and is about to be damaged, it can also be arranged in advance: Repair or replace Accidents during rolling are provided by “low production and maintenance costs”;

Page 16

Here are six practical facts. 5. Description of the invention (14) In order to verify the effectiveness of this diagnostic method, the illustrations are shown with drawings and photos as follows: Example 1: Low frequency chatter phenomenon of hot rolling mill rolls Chatter phenomenon refers to equipment Harmonic vibration near a single frequency; because the vibration frequency is limited to a very narrow frequency band, so that kinetic energy is concentrated at a single frequency, it is easy to generate vibration rollers on the work roll, back roll surface and steel belt. Please refer to Figure 7 and Figure 8 at the same time, which are examples of vibration signals of the present invention-preferred embodiment (1) The vibration signal of the hot rolling mill roll bearing is not intended and the frequency spectrum of the vibration signal of Figure 7 (1) A schematic diagram, as shown in the figure; the seventh diagram shows that the vibration pattern is a simple harmonic close to a single frequency, and the eighth diagram shows that the vibration spectrum is a peak of energy concentration. The vibration signal data of this example 1 is as follows: the amplitude effective value is 1, 36G, the amplitude peak-to-peak value is 516G, the frequency concentration is 100%, the peak frequency is 299Hz, the amplitude modulation frequency is 73 Hz, and the amplitude modulation variation is 18.2%; At the same time, please review the second picture, the result of using the diagnostic method of the present invention is "low frequency appearance vibration phenomenon"; please also refer to the photo-the vibration roller marks found on the work roll. The main box with low-frequency chattering phenomenon is not stable in the hot rolling mill, and the gap is too large, so that the rolling frequency is mainly the bearing inner ring or wave. The main cause of the vibration stick marks is the work roll and back roll bearing or work concept. Caused by sloshing between the bearing housing and the hot rolling mill; the passing frequency of the vibration outer ring or its 2 times the harmonic distance is between 50 and 65 mm.

Page 17

"Column 2 'Poor lubrication of the work roll bearings of the hot rolling mill (oil loss or water ingress) 2 The bearing's poor lubrication caused by oil loss or water ingress will rise sharply to the south, and the vibration frequency is characterized by 5 disturbances in the high frequency range The spectrum of the distribution 'This is due to the metal-to-metal dry friction. Please refer to the schematic diagram of the vibration spectrum of the work roll bearing of the ninth circle of a preferred embodiment of the present invention. Compared with the tenth diagram, which is a schematic diagram of the vibration spectrum of a normal bearing of a preferred embodiment of the present invention, it can be found that the two are different in the high-frequency region. The vibration signal data of this example 2 is as follows: The amplitude effective value is 0, 46G, the amplitude peak-to-peak value is 3.47 (;, the frequency concentration is 6.7%, the equivalent frequency center is 1589Hz, and the sharpness of the main peak of the inverse spectrum is 76.3%. At the same time, please review the second figure and use the The result of the diagnostic method was "Poor lubrication caused by oil loss or water in the bearing." After disassembling the equipment on site, it was found that there was oil loss and water ingress. Please also refer to the photos of the second series of work roll thrust bearings for water ingress. , Cause poor lubrication and burn If the situation of poor bearing lubrication is not detected early and handled in a timely manner, it may be burned during rolling, and even cause major damage to the rollers. Example 3: Hot Fortunately, the slipping and scratching between the inner ring of the roller bearing and the roll Please refer to the eleventh diagram for the vibration signal example (c) of a preferred embodiment of the present invention. The vibration spectrum of a work roll bearing. It can be seen that its vibration characteristics are the same as the bearings in the previous example.

Run five. Description of the invention (16)-scattered form 'and many large vibrations in the high frequency region. This is a phenomenon of metal-to-metal dry friction. One picture and the ninth picture, we can find _; ^ There is still an obvious difference, that is, the red ^ netting sleeve bearing poorly lubricated spectrum in the ninth picture is messy in the high frequency range. _ ”Valence, and in the eleventh figure, the frequency spectrum between the inner ring of the bearing and the roller neck slips to complex α &amp; N 炀. In the frequency domain, the spectrum is a 愔 π moon shape arranged at equal intervals.

The side-wave bands of the medium-interval distribution of the frequency spectrum show that the bearing ^ I door rotates regularly during the rotation. — The difference between the above-mentioned bearing poor lubrication and roll neck slipping to injury 'is the inverse spectrum analysis using vibration signals; please refer to the twelfth figure which is a schematic diagram of the inverse frequency of the eleventh figure of a preferred embodiment of the present invention. In the case where the eleventh spectral line is evenly spaced, ^ 'the inverse spectrum will show significant spikes. On the other hand, if the ninth spectral line is in a random distribution, there will be no significant spikes in the inverse spectrum; The signal of the eleven figure, that is, the vibration signal data of this example 3 is as follows: the amplitude effective value is 0.5G, the amplitude peak-to-peak value is 4.8G, the frequency concentration is 6, 3%, and the equivalent frequency center is 1707 · 6 Hz. 'Anti-spectrum main drop sharpness 93.6%' Please also cooperate with the second picture shown below; the result of using the diagnostic method of the present invention is "dry friction and brake injury caused by oil loss between the roller neck and the inner ring of the bearing"; After the device, it was found that there was a scratch between the roller neck and the inner ring of the bearing 4; please also refer to the photo. After removing the bearing of the three series of work rolls, it was found that the roll neck was slipping with the inner ring of the bearing and

Page 19 46290 V. Description of the invention (17) Example. Dry rubbing and scratching. Please refer to Figures 13 and 14 at the same time when the surface of the hot rolling mill is damaged. Please refer to Figures 13 and 14 at the same time. This is a relatively good sound of the present invention. Schematic diagram of the vibration signal of the roller bearing and the frequency spectrum of the first work (4) Dry sound, the vibration signal of Zhongding Yiyuan 乂 clearly see the periodic shock wave, which repeats: ,, rotation frequency. The vibration signal data of this example 4 For example, under the car, the effective value of the wide amplitude is 0 '22G, the peak amplitude is 3 57G, and the concentration ratio is 14.9%' peak frequency 58δ 4Η2 'equivalent = center 543Hz, the amplitude modulation frequency is 3. 7Ηζ, the amplitude modulation variation%, The wave height rate is 8.2. Please also review the results shown in the second figure. 7 The results of using the diagnostic method of the present invention are "the surface of the back roller is scratched: when disassembling the device, it is found that the surface of the back roller is indeed scratched. , Please also refer to the photo of the back roller of the four series; when there is a scratch on the back roller, when the mark turns to contact with the wheel during rolling, the stress state of the contact surface changes instantly, As a result, the "view is subjected to a strong pulse" so the vibration signal will appear ^ Shock wave.… Example 5: Please refer to Figure 15 and Figure 16 for the abrasion of the main shaft coupling of the hot rolling mill, which are the work roll bearings of the vibration signal example (five) of a preferred embodiment of the present invention, respectively. The schematic diagram of the vibration signal and the spectrum (five) of the vibration signal I in Figure 15 are shown in the figure.

Page 20 V. Description of the invention (18) The periodic shock wave is clearly seen, and its repetition frequency is twice the rotation speed of the work rolls; the spectrum of the sixteenth figure shows that the structure of the hot rolling mill is excited by pulsed knocking The case of natural frequencies. The vibration signal data of this example 5 are as follows: the amplitude effective value is 0.57G, the amplitude peak-to-peak value is 5.27G, the frequency concentration is 24%, the peak frequency is 145.3112 ', the effective frequency center is 411.4112, and the amplitude modulation frequency is 12.2Hz'. The degree is 85 丨%, the wave height rate is shown in the second figure in conjunction with the review; the result of using the diagnostic method of the present invention is "spindle coupling wear"; when disassembling the device, it was found that the "spindle coupling is indeed excessive." In the case of abrasion, the reason for the periodic shock wave shown in Figure 15 is that the gap between the coupling and the work head is too large due to excessive wear, which causes the coupling to lose support every half turn of the work roll and downward Drop hits the work roll head; therefore, there are two knocks per revolution, that is, the phenomenon that the repetition frequency of the shock wave is twice the speed of the work roll. Β: The load of the work roll bearing of the hot rolling mill is abnormal. Example of a vibration signal of a preferred embodiment of the invention (six) A schematic diagram of the vibration signal spectrum of a work roll bearing. BPF0 in the figure represents the bearing outer ring passing frequency, which is a certain point of the bearing outer ring and unit time. The number of rollers passed; 2 χ ’3 X ′ 4 X… · # represents 2 times, 3 times, 4 times, etc. of B P F 0; this phenomenon indicates that there is an abnormal condition in the work roll bearing. The vibration signal data of this example 6 is as follows: the amplitude effective value is 0.76G ′, the amplitude peak-to-peak value is 4.6G, and the frequency concentration is 312%.

4 6 2 9 J ： V. Description of the invention (19) Times 80_ Figures are not, use the diagnosis of the present invention to prepare for the private .. / The result of the meeting is at the same time please review the abnormal bearing load of the roller; The result of inspection &gt; was the most ice-cold s, fen # * # bearing inspection results, found that the negative :: ί Ϊ has abnormal wear marks, showing that there are deep scratches in the circle, showing that this bearing has- Road alignment or angle deviation, contact with the work rolls may cause excessive load due to position, on the other hand, from 3: bearing, 'cause shaft black, and local scorch :::: f often Please also refer to the photos of the five series of abnormal conditions that cause excessive wear. To sum up, the method for diagnosing the hot rolling of steel strip according to the present invention can indeed measure the vibration 4; the characteristic value of the characteristic parameters of the vibration, the characteristic value of the frequency domain and the amplitude of the characteristic parameters; The location and cause of the abnormality is w. The present invention is sincere: progress: one, but no similar or similar items have been disclosed in the city, and the patent requirements of Cheng == have been filed in accordance with the provisions of the Patent Law. Application. Ying of d kg but, those described above, the present invention is only one preferred system = specification and the scope of the patent application of the present invention where m = whom drug change, should be included within the scope of the present invention. Seeking government ... Structure Page 22 Brief description of the diagram The first diagram is a schematic diagram of the structure of a conventional hot rolling mill. The second figure is a schematic structural diagram of a driving system of a conventional hot rolling mill. The third diagram is a flowchart of steps in a preferred embodiment of the present invention. The fourth diagram is a schematic diagram of a vibration signal with a periodic shock wave according to one of the preferred embodiments of the present invention. The fifth diagram is a schematic diagram of a vibration signal with periodic amplitude modulation in a preferred embodiment of the present invention. The sixth diagram is a schematic diagram of the amplitude modulation waveform of a preferred embodiment of the present invention. The seventh diagram is a schematic diagram of the vibration signal example (a) of a preferred embodiment of the present invention. The eighth diagram is a spectrum (a) diagram of the vibration signal of the seventh diagram of a preferred embodiment of the present invention. The ninth figure is a schematic diagram of the vibration spectrum of a work roll bearing in the vibration signal example (2) of a preferred embodiment of the present invention. The tenth figure is a schematic diagram of a vibration spectrum of a normal bearing according to a preferred embodiment of the present invention. The eleventh figure is a schematic diagram of the vibration spectrum of an i-roller bearing in the vibration signal example (c) of a preferred embodiment of the present invention. The twelfth line is the inverse spectrum of the eleventh figure of a preferred embodiment of the present invention. The thirteenth figure is a schematic diagram of a vibration signal of a work roll bearing in the example (4) of a vibration signal of a preferred embodiment of the present invention. The fourteenth diagram is a vibration message of the thirteenth diagram of a preferred embodiment of the present invention.

P.23 4 Θ2 91 Simple explanation of the formula (4). The fifteenth figure is a schematic diagram of the vibration signal of the light-duty bearing example (fifth) of a preferred embodiment of the present invention. The sixteenth diagram is a schematic diagram (fifth) of the vibration signal spectrum of the fifteenth diagram of a preferred embodiment of the present invention. The seventeenth figure is a vibration signal spectrum of a vibration signal example (six) of a preferred embodiment of the present invention. The photo shows the vibration marks found on the work roll. Photograph 2 The work roll thrust bearing is exposed to water, which causes poor lubrication and burns out. 9 Photograph 3 rolls of the work roll roll neck and the bearing inner ring slip and dry friction, resulting in injury. Photo of the four series of back rollers scratched. Photo Five series of work roll bearing abnormal load caused excessive wear. Reference number in the figure: 1 vibration signal measurement step 2 signal calculation step 3 amplitude actual value and peak-to-peak determination step 4 frequency concentration determination step 4 L spike frequency determination step 42 amplitude modulation frequency and amplitude modulation variation comprehensive determination step 5 Comprehensive judgment step of peak frequency, equivalent frequency center, amplitude modulation frequency, and amplitude modulation variation step 5 1 wave height ratio determination step 6 secondary peak sharpness judgment step

Page 24

Page 25

Claims (1)

462 Patent application scope 1: A method for vibration diagnosis of abnormal or faulty rolls of a steel strip hot rolling mill, which includes the following steps: A vibration signal measurement step, which is a vibration signal of a surveyor; Hard steps, which are calculated separately The flavor value of the aforementioned vibration signal: ί 中 哕 # frequency domain characteristic parameter value and amplitude modulation characteristic parameter wave opening, and calculation: ί characteristic parameter value is for the time and vibration rate of the vibration signal (ch field value (Rms ), Amplitude peak-to-peak (PkPk) 4 two parameter values, as an array of μ prepared diagnostic indicators, then; Υ2 ··, '"represents-pen vibration measurement (1) amplitude effective value (Rms): Where yQ represents the average value of the vibration signal 1 N 7〇 = ^ Σχ · iy Μ (2) amplitude peak-to-peak value (FkPk): Pkpk ^ rugged ―min (r) where max (γ) represents the vibration signal γ Υι 'y2 &quot; · 5 Υμ Page 26 6. The maximum value in the scope of patent application, min (Y) represents the minimum value; (3) Wave height ratio (Crfct): Crfct ^ = Pkpk / 2 / Rms This frequency domain characteristic parameter is for vibration signals. Frequency spectrum. Calculate three parameter values including peak frequency (Fpeak), equivalent frequency center (FreqCntr), and frequency concentration (Fcon) as indicators for equipment diagnosis. If ((H), (f2, s2), '. ,, UM, sM) i represents the spectral lines in the vibration spectrum, that is, (fi, Si) represents the amplitude at the frequency d is si, then: (1) Fpeak: Fpeak = f1 if si = max ({s !, s2,…, Sj [}) (2) Equivalent frequency center (FreqCntr): &quot; First set a threshold for the frequency spectrum (in the present invention In the preferred implementation, 'the threshold value is set to 10% of the peak frequency at ⑽ ^^)' The height of the spectral line is lower than this threshold value 'is considered as noise and ignored, and then the peak in the frequency spectrum is found, Its definition is as follows: and Si &gt; Si + i &gt; Si + 2 then (f ;, Si) is a spike 'then the array formed by each spike is {(pf 1, psl), (pf 2' ps2) '(pfn 'pSn)} means that the equivalent frequency center is defined as FreqCntr = ϋ- ί &gt; · il Page 27 a 62 VI. Patent Application Range (3) Frequency Concentration (Fcon): As in the preceding paragraph, {(pf !, ps,) '(pf2' ps2), ..., (pfn, pSn)} represents the frequency spectrum Each peak; first find the peak frequency Fpeak 'in the spectrum and set a frequency bandwidth BWF and then find the ratio of the peak and psum in the frequency range centered on Fpeak's BWF to the overall peak and Tsum, that is, Psum = J ^ pSi ..... for (Fpeak-BWF) &lt; Pf. &Lt; (Fpeak + BWF) n Tsum ~ Σ 涔 'Fcon = Psum / Tsum x 100%. The amplitude modulation feature parameter includes the amplitude Modulation frequency (or shock wave repetition frequency) and amplitude modulation variability (Amdev) two parameters, where: _ (!) Amplitude modulation frequency (or shock wave repetition frequency): It refers to the amplitude of the vibration signal in unit time , And the repetition frequency of the shock wave refers to the number of times ς occurs per unit time. (2) Amplitude Variation (Amdev) of Qingbo: The time waveform of the amplitude modulation of the dynamic signal, then it uses the Hilbert transform to find the vibration waveform: if y (t) represents the vibration signal, Hilbert Special conversion yh (t) is defined as Page 28 6. Scope of patent application &amp; Using y (t) and yh (t) ', the degree modulation waveform of the vibration signal A (t) A (0 ^ ^ y2 (0 + yzk (t ) Amdev = (max- min) / (max + min) x 100% * where 'max and min represent the maximum and minimum values of the amplitude modulation waveform A (t) in the sixth graph (b), respectively; The peak-to-peak determination step is to determine whether the amplitude effective value of the above-mentioned vibration signal is greater than a set critical value (Q 4G 'G is the acceleration of gravity, which is about 9.8 m / s 2), or whether the amplitude ridge is to the peak. If it is greater than the set critical value (3G), if yes, proceed to the next step; if not, determine that the equipment condition is normal and end the diagnosis; the frequency concentration determination step is to determine if the frequency concentration of the vibration signal is greater than the set Critical value (40%), it is judged that the device has a flutter phenomenon and the diagnosis is ended; if not, the next step is performed; a comprehensive judgment step of the peak frequency, the equivalent frequency center, the amplitude modulation frequency, and the amplitude modulation variation, which is a judgment Whether the above-mentioned vibration signal meets the small peak frequency 700Hz, the center of the equivalent frequency is less than 1300Hz, and the AM frequency is between 1Hz and 16Hz, and the AM variation is greater than 40% 'If yes, the condition of the equipment is determined to be the surface of the back roller to the injury or the spindle Shaft wear, if not, no judgment L; Page 29 462 ------ The sharpness of the main peak of the secondary spectrum is to determine whether the sharpness of the main peak of the secondary spectrum of the above-mentioned vibration I signal is greater than 68%. If it is', then it is judged that the equipment condition is abnormal bearing load. If not, it is not judged that the equivalent frequency center and frequency are judged as the above vibration signal is 1100Hz, and the condition of the frequency concentration degree is that the bearing lubrication is poor. ^ According to the first normal or fault vibration diagnosis method of the scope of patent application, if the step is to determine the vibration signal (40%), then the amplitude modulation frequency and amplitude modulation variability of the vibration signal peak frequency Hz) 'are satisfied in sequence and the amplitude modulation is satisfied. If the frequency is greater than 40%, the amplitude modulation phenomenon will be displayed. If the sharp amplitude modulation frequency is between 1 and 8Hz, the device is judged to have a peak frequency between the set threshold (80 Hz). Time, and the amplitude modulation variability is high frequency flutter with obvious amplitude concentration concentration comprehensive judgment step, whether it satisfies the equivalent frequency center is greater than less than 10%, if it is, then it is judged that the device or the bearing and the neck are scratched. If not, the rolls of the steel strip hot rolling mill described in the item are different. The frequency concentration judgment frequency concentration is greater than a set threshold value, a peak frequency judgment step and a comprehensive judgment step; it is a judgment, If it is less than the set low frequency critical value (4 0 〇 is between 1 and 8 Η z, and the amplitude modulation variation determines that the device is low frequency flutter and the bright peak frequency is less than 400 Hz, but it does not meet the interval, and the amplitude variation is adjusted. Low frequency flutter phenomenon at 40%; if the vibration signal has a low frequency critical value (40 0Hz) and high frequency and meets the condition that the amplitude modulation frequency is between 1 and 8Hz and 40%, then the device ’s degree modulation phenomenon is judged. Frequency between On page 30, the scope of patent application is between 40Hz and 800Hz, but the condition that the amplitude modulation frequency is between 1 and 8Hz and the amplitude variation is greater than 40% is judged as a high-frequency chattering phenomenon; If the peak frequency of the vibration signal is greater than the set extreme high frequency critical value (1200Hz), and the conditions that the amplitude modulation frequency is between 1 and 8 Η z 'and the amplitude variation is greater than 40%, the device is judged to be extremely polar. South frequency flutter and obvious amplitude modulation phenomenon. If the peak frequency is greater than 1200Hz, but does not meet the condition that the amplitude modulation frequency is between 1 and 8Hz and the amplitude variation is greater than 40%, then the device is determined to be connected ~ Frequency flutter phenomenon; if the sharp bee frequency of the vibration signal is between setting 11 and #frequency critical value (800 Hz) and extreme high frequency critical value (Kook), then it is judged that the condition of the device cannot be judged and the judgment is ended By. Time • According to the application of the special scope of the first special scope of the steel strip hot rolling mill vibration diagnosis method 'where, if the central peak frequency, the above-mentioned vibration signal meet the peak frequency is less than 700, and the same frequency Comprehensive judgment step of center, AM frequency and AM variability + 丨 J-break cause ratio 4 If the center is less than 1300HZ, and the AM frequency is between 1Hz * 16Hz and the AM variability is greater than 40%, then another high-break step is performed. , It is to determine whether the wave height rate of the vibration signal is greater than 5 years, then determine the condition of the equipment is the surface of the back roller is scratched, if not, 'if the condition of the equipment is the wearer of the main shaft joint Korean device. According to the method of vibration diagnosis of steel strip hot rolling often or fault described in item i of the patent application scope, where: 'roll different one one *, one ... again, the center of the equivalent frequency and%, then Another Page 31 4 62 90: 、 Scope of patent application—the sharpness step of the main peak of the anti-spectrum, if the sharpness of the main peak of the anti-spectrum of the vibration signal is less than 80%, then the condition of the equipment is that the bearing is out of oil or water is inadequate lubrication. If the sharpness of the main peak of the inverse spectrum of the vibration signal is between 80% and 85%, then the equipment condition is judged to be poor bearing lubrication or scratches between the roller neck and the bearing inner ring. If the inverse spectrum of the vibration signal is If the sharpness of the main peak is greater than 85%, it is judged that the condition of the equipment is dry friction and scratch caused by oil loss between the roller head and the inner ring of the bearing. Page 32