JPH076831B2 - Equipment diagnosis method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention relates to a method for diagnosing equipment applied to rotary machines such as pumps and blowers installed in highly public LNG plants, thermal power plants, and other equipment. It is about.

[Prior Art] The ratio of rotary machines to large plants or equipment such as LNG plants is very high, and their role is also important. When an abnormality or a failure occurs in these rotating machines, economic loss naturally occurs, but when the accident spreads, it becomes a big social problem.

For example, an LNG plant is used as a supply source for city gas, power generation equipment, etc.
If the NG pump fails, it will cause a large loss to the user side and damage the general consumers. Therefore, in order to solve such a problem, a vibration sensor is attached to a typical part of the pump casing, vibration is measured at regular time intervals, Fourier analysis is performed on the measured vibration data, and the data is transferred to data on the frequency axis. Then, processing is performed to obtain a diagnostic index, and the pump is diagnosed for abnormality and prediction of deterioration is performed from the diagnostic index.

Applying the conventional equipment diagnosis method applied to rotating machinery to LNG pumps Fig. 4 to Fig. 6 (a) (b)
Explaining with (C), 1 in FIG. 4 is an LNG pump, 2
Is the casing, 3 is the upper and lower bearings 4, 4'in the casing 2.
A vertical shaft 5 rotatably supported via a vertical shaft 3, an impeller fixed to the vertical shaft 3 in a plurality of stages, a driving device 6 for driving the vertical shaft 3 and the impeller 5, and a lower bearing 4'on the outer periphery of the casing 2. An acceleration detector or a displacement detector, which is arranged at a close position, 8 is an LNG.

Acceleration or displacement is detected by the acceleration detector or displacement detector 7 during the operation of the LNG pump 1, and the frequency analysis of this data is performed to obtain a vibration spectrum distribution such as acceleration or amplitude as shown in FIG.

Generally, vibration of a rotary machine such as a pump is generated by a shaft system (vertical shaft 3
And the impeller 5), bearing system (bearings 4, 4 '), and internal fluid system (LNG8) are generated as vibration sources. The vibration frequency generated by the shaft system, bearing system, and internal fluid system is mainly 0 to
It is distributed in the frequency band of 500Hz, 500Hz-10KHz, 10KHz-20KHz. Therefore, the vibration spectrum distribution shown in FIG.
~ 500Hz, 500Hz ~ 10KHz, 10KHz ~ 20KHz divided into a frequency band A by the shaft system, a frequency band B by the bearing system, a frequency band C by the internal fluid system, each frequency band A, B measured at a certain point , C, the average values of the vibration levels s, b, and c are obtained, and the vibration levels are displayed in time series as shown in FIGS. 6 (a), (b), and (c). FIG. 6 (a) shows the shaft system, FIG. 6 (b) shows the bearing system, and FIG. 6 (c) shows the diagnostic index of the internal fluid system.

Thus, the diagnostic index of FIG. 6 (a) (b) (c) is the axis system,
Since the state of deterioration of the bearing system and internal fluid system is shown over time,
Based on the data measured during normal operation of the LNG pump,
The warning and abnormality diagnostic levels L _C and L _D are set in advance, and the diagnostic index and caution and the abnormality diagnostic levels L _C and L _D obtained in the above procedure in the actual operation of the LNG pump are compared with the LNG pump. It is diagnosed whether or not an abnormality has occurred.

[Problems to be Solved by the Invention] However, in the above-mentioned conventional means, when the diagnostic index is obtained, the entire average of accelerations or amplitudes in a predetermined frequency band is used. If a portion with a high average value diagnosis result occurs, there is a problem that the error in the diagnostic index becomes large, and it may not be possible to perform accurate and accurate diagnosis.

The present invention has been made in view of the above circumstances and has an object to enable accurate equipment diagnosis.

[Means for Solving the Problems] The present invention finds a vibration spectrum based on acceleration or displacement generated by rotation of a rotating body, sets a spectrum pattern based on the basic rotation speed of the rotating body, and sets an integer of the spectrum pattern. A Gaussian distribution in a predetermined range corresponding to a double component multiplied by the vibration spectrum at the frequency for the integer multiple component is integrated and averaged over the predetermined range to obtain a diagnostic index. The diagnostic index is obtained in time series to predict the scheduled inspection time.

[Operation] From the vibration spectrum obtained from the acceleration or displacement generated by the rotation of the rotating body, a spectrum pattern based on the basic rotation speed of the rotating body is set, and a predetermined range determined in advance corresponding to an integer multiple component of the spectrum pattern. The Gaussian distribution of is multiplied by the vibration spectrum at the frequency for the integral multiple component is integrated over a predetermined range and averaged, the diagnostic index is determined, and the diagnostic index is determined in time series, Since the scheduled inspection time is predicted from the diagnostic index required for, it is possible to accurately determine the scheduled inspection time and improve the reliability of the equipment.

Embodiments Embodiments of the present invention will be described below with reference to the accompanying drawings.

1 to 3 show an embodiment of the present invention.

Similarly to the conventional case, the acceleration detector or the displacement detector 7 shown in FIG. 4 detects the acceleration or the displacement during the operation of the rotating body such as the LNG pump, and performs the frequency analysis to perform the first analysis.
As shown in the figure, a vibration spectrum distribution such as acceleration or amplitude is obtained.

The vibration generated by the rotating body has the largest component due to the basic rotational speed, and often contains information due to abnormalities in the shaft system and bearing system (particularly bush bearings).
The derived component is also generated in a part of the basic number of revolutions that is an integral multiple or a fraction of an integer, and this derived component often becomes “fluctuation” of the number of revolutions due to load fluctuation.

However, by setting a spectrum pattern based on the basic rotation speed and filtering a Gaussian distribution in the range of nf ₀ ± Δf with respect to this integral multiple component, spectrum “fluctuation” is prevented. For each component, the average value within nf ₀ ± Δf is used as the diagnostic index _n. Evaluate by.

Here, Sp (f); Vibration spectrum f ₀ ; Fundamental frequency Δf; Width f for obtaining the spectrum pattern around the fundamental frequency or a frequency that is an integral multiple of that frequency f; Frequency W; Coefficient K; Constant e; Natural logarithm When the diagnostic index _n (n = 1,2,3, ...) Obtained by the equation (i) is displayed in time series, it becomes as shown in FIG. 3, and the time T ₀ of regular inspection can be predicted from this graph. Therefore, since the error of the diagnostic index is small, it is possible to accurately diagnose the regular inspection time, and it is possible to know the stability of the equipment after the regular inspection.

It should be noted that the present invention is not limited to the above-described embodiments, and it goes without saying that various changes can be made without departing from the gist of the present invention.

[Effects of the Invention] According to the equipment diagnosis method of the present invention, since it is possible to accurately diagnose the time of regular inspection of equipment, there is an excellent effect that the reliability of equipment is improved.

[Brief description of drawings]

FIG. 1 is a graph showing the relationship between the frequency and the vibration spectrum obtained when the method of the present invention is performed, FIG. 2 is a graph showing the relationship between the frequency and the Gaussian distribution, and FIG. 3 is a graph showing the diagnostic index in time series. A graph, FIG. 4 is a general explanatory view of the LNG pump, FIG. 5 is a graph showing the relationship between the frequency and the vibration spectrum obtained when performing the conventional method, and FIG. 6 (a).
(B) and (c) are graphs in which the diagnostic indexes in the conventional method are displayed in time series. In the figure, 1 is an LNG pump, and 7 is an acceleration detector or a displacement detector.

Claims (1)

[Claims] 1. A vibration spectrum is obtained based on acceleration or displacement generated by the rotation of a rotating body, and a spectrum pattern is set by the basic rotation speed of the rotating body, which is predetermined in correspondence with an integral multiple component of the spectrum pattern. A product obtained by multiplying the vibration spectrum at the frequency for the integer multiple component by a Gaussian distribution in a predetermined range is integrated and averaged over the predetermined range determined in advance to obtain a diagnostic index, and the diagnostic index is time-series. A method for diagnosing equipment, characterized by predicting the scheduled inspection time.