CN114323664A - Method for detecting abnormal gas vibration of gas turbine - Google Patents

Abstract

The object of the present invention is to provide a method for detecting abnormal gas vibration of a gas turbine, comprising the following steps: selecting unit power and vibration amplitude data when the gas turbine is running, using historical data to construct a normal vibration amplitude model, including the upper limit of the vibration amplitude Boundary and control center line; collect the power parameters during the actual operation of the gas turbine to construct the sequence a to be detected, and the different vibration amplitude data of each vibration measurement point to construct the sequence to be detected b, c, d...; select different SPC criteria to combine the logarithmic sequence a and number sequences b, c, d...etc. to perform abnormal detection to determine the vibration system status of the unit. The invention can detect the development trend of abnormal vibration failure, and is a simple, efficient, accurate and real-time detection method for abnormal vibration of gas turbine.

Description

A detection method for abnormal gas vibration of gas turbine

technical field

The invention relates to a gas turbine detection method, in particular to a gas turbine vibration detection method.

Background technique

Because the gas turbine rotor is a high-speed rotating part, it is subjected to high-frequency vibration for a long time, and faults such as rotor eccentricity, blade tipping, and bearing wear of the unit will cause abnormal vibration of the unit. Therefore, the monitoring of abnormal vibration of gas turbine is of great significance for the stable and reliable operation of the unit.

Generally, the vibration of the gas turbine is measured by the acceleration vibration sensor installed on the bearing seat, and some unit vibration sensors are also installed on the casing or casing. The collected vibration signals are mainly responsible for the safety protection of the unit. Once the set threshold is exceeded, an alarm or stop command is given. However, during the actual operation of the unit, the safe operation value and limit value of the unit vibration are actually a change value due to changes in factors such as random unit operating conditions (load) and lubrication system status. Moreover, if anomalies can be found through the change process of these values, the trend of slow-changing faults can be found in advance, and the occurrence of related faults can be prevented in advance.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a detection method for abnormal gas vibration of a gas turbine which can effectively detect the development trend of abnormal vibration failure.

The object of the present invention is achieved in this way:

A method for detecting abnormal gas vibration of a gas turbine of the present invention is characterized in that:

(1) Calculate the vibration amplitude model of the gas turbine under different working conditions through the historical operation data of the unit;

(2) Construct the vibration amplitude series A, B, C..., and calculate the standard deviation D and mean value S of the vibration amplitude under different working conditions;

(3) taking the multiple n·D of the standard deviation D calculated in step (2) to construct the upper boundary of the vibration amplitude model;

(4) take step (2) to calculate and obtain the mean value S as the control center line of the vibration amplitude model;

(5) Collect the power and vibration amplitude of each position during the actual operation of the gas turbine;

(6) Select one or more corresponding detection rules based on the theory of quality control charts to perform abnormal detection on power and vibration amplitude;

(7) If the power detection result is stable and the vibration amplitude detection result is an abnormal change, a corresponding alarm is given, and step (9) is skipped;

(8) If the vibration amplitude detection result is normal, go to step (5) to perform data collection and detection again;

(9) An alarm is given, and the detection is completed.

The present invention can also include:

1. The quality control chart is theoretically based on the following criteria:

The advantage of the present invention is that the present invention can detect the development trend of abnormal vibration failure, and is a simple, efficient, accurate and real-time detection method for abnormal vibration of a gas turbine.

Description of drawings

Figure 1 is an example of the vibration power model when the gas turbine is operating normally;

Fig. 2 is the flow chart of vibration state identification during the operation of the gas turbine;

FIG. 3 is a flowchart of abnormal vibration detection of a gas turbine.

Detailed ways

The present invention will be described in more detail below in conjunction with the accompanying drawings:

1-3, the implementation of the present invention includes the following steps:

The vibration amplitude model of each working condition under the normal state of the unit is established.

Select the historical data of the vibration amplitude u of the unit when the gas turbine is running under different operating conditions, and establish a normal vibration amplitude u-power P relationship model, including the upper boundary of the amplitude and the control center line.

The control center line is the average value of the historical normal operation process curve, and the upper boundary is the control center line plus n times the standard deviation based on historical data statistics. Among them, the value of n is adjusted according to the actual data. This process can be done offline.

Detect the real-time status of gas turbine power and corresponding vibration amplitude.

(1) Collect the power and vibration amplitude time series of each vibration measuring point during the actual operation of the unit;

(2) Input the power data into the historical data model obtained in step 1, and calculate the time series of the theoretical value of vibration amplitude;

(3) According to historical experience, the data state detection rule in the theory of quality control chart (SPC criterion) is selected to monitor the difference state between the vibration amplitude of the unit and the calculated value of the theoretical model. According to historical experience, R1, R2, R3, R5, R11 and R13 are selected for vibration detection of the unit, and the corresponding detailed rules are shown in the SPC criteria list of the present invention.

List of SPC criteria for the present invention:

It is determined that the gas turbine vibration is abnormal.

Based on the vibration state identification of the unit, a comprehensive judgment condition rule for gas turbine vibration abnormality is constructed. If and only when the power of the unit is stable, more than one state of all vibration amplitude time series is marked as abnormal, it is determined that the vibration of the unit is abnormal.

Claims (2)

1. a detection method for abnormal gas vibration of gas turbine, it is characterized in that: (1) Calculate the vibration amplitude model of the gas turbine under different working conditions through the historical operation data of the unit; (2) Construct the vibration amplitude series A, B, C..., and calculate the standard deviation D and mean value S of the vibration amplitude under different working conditions; (3) taking the multiple n·D of the standard deviation D calculated in step (2) to construct the upper boundary of the vibration amplitude model; (4) take step (2) to calculate and obtain the mean value S as the control center line of the vibration amplitude model; (5) Collect the power and vibration amplitude of each position during the actual operation of the gas turbine; (6) Select one or more corresponding detection rules based on the theory of quality control charts to perform abnormal detection on power and vibration amplitude; (7) If the power detection result is stable and the vibration amplitude detection result is an abnormal change, a corresponding alarm is given, and step (9) is skipped; (8) If the vibration amplitude detection result is normal, go to step (5) to perform data collection and detection again; (9) An alarm is given, and the detection is completed. 2. A detection method for abnormal gas vibration of a gas turbine, characterized in that: the quality control chart is theoretically based on the following criteria:

Images