A method for detecting fault of lubricating oil cooler of gas turbine

Abstract

The purpose of the present invention is to provide a method for detecting a fault of a gas turbine lubricating oil cooler, comprising the following steps: constructing a return oil temperature change rate model of a gas turbine lubricating oil cooler under different working conditions, wherein the return oil temperature change rate model includes an upper boundary and a control center line; collecting the return oil temperature of the lubricating oil cooler during the operation of the gas turbine, and calculating the inlet and outlet temperature difference and change rate of the lubricating oil cooler; detecting the return oil temperature change rate of the lubricating oil cooler during the actual operation of the unit, and marking the lubricating oil cooler state as abnormal state one when the return oil temperature change rate is greater than a threshold; detecting the inlet and outlet lubricating oil temperature difference of the lubricating oil cooler, and marking it as abnormal state two when the temperature difference exceeds the threshold; comprehensively evaluating the abnormal state one and abnormal state two of the lubricating oil cooler, and when the abnormal state one and abnormal state two occur at the same time, marking the lubricating oil cooler abnormal, and the detection is completed. The present invention can use process information to detect whether there is a fault in the gas turbine lubricating oil cooler.

Description

Technical Field

[0001] The present invention relates to a gas turbine detection method, in particular to a gas turbine lubricating oil cooler detection method. Background Art

[0002] The lubricating oil system of a gas turbine is used to provide the gas turbine with suitable lubricating oil to meet the lubrication, cooling and regulation needs of the gas turbine during normal operation, and to take away the heat from friction and high-temperature parts to maintain the parts within the normal operating temperature range and increase the service life of the parts. The lubricating oil takes away the heat from the gas turbine, and the lubricating oil cooler is used to cool the lubricating oil. When the lubricating oil cooler fails, it cannot meet the unit's requirements for lubricating oil.

[0003] In engineering applications, the difference in lubricating oil temperature between the inlet and outlet of the gas turbine lubricating oil cooler is generally used to detect the failure of the lubricating oil cooler. When the difference in lubricating oil temperature between the inlet and outlet of the lubricating oil cooler is detected to exceed the set threshold, a fault alarm is given. The failure of the lubricating oil cooler is a gradual process. The existing technologies fail to utilize the changes in information during the monitoring of the lubricating oil cooler failure process, and can only alarm after the lubricating oil cooler fails. Summary of the invention

[0004] The object of the present invention is to provide a method for detecting a fault of a gas turbine lubricating oil cooler by utilizing process information to detect the fault of the gas turbine lubricating oil cooler.

[0005] The object of the present invention is achieved in that:

[0006] The present invention provides a method for detecting a fault of a gas turbine lubricating oil cooler, which is characterized by:

[0007] (1) Based on the historical operation data of the unit, a return oil temperature change rate model of the gas turbine lubricating oil cooler under different operating conditions is constructed. The return oil temperature change rate model includes an upper boundary and a control center line;

[0008] (2) Collect the oil return temperature of the lubricating oil cooler during the operation of the gas turbine and calculate the temperature difference and change rate of the lubricating oil cooler inlet and outlet;

[0009] (3) Detecting the oil return temperature change rate of the lubricating oil cooler during actual operation of the unit, and marking the oil return temperature change rate as abnormal state 1 when it is greater than a threshold value;

[0010] (4) Detecting the temperature difference of the lubricating oil at the inlet and outlet of the lubricating oil cooler, and marking it as abnormal state 2 when the temperature difference exceeds a threshold;

[0011] (5) Comprehensively evaluate the abnormal state 1 and the abnormal state 2 of the lubricating oil cooler. When the abnormal state 1 and the abnormal state 2 occur at the same time, the lubricating oil cooler is marked as abnormal and the detection is completed.

[0012] The present invention may also include:

[0013] 1. The establishment process of the oil return temperature change rate model: The horizontal axis parameter uses the engine speed n, and the oil return temperature change rate is calculated as (ΔT 02 -ΔT 01 ) / t, t is the calculation period, ΔT 01 , ΔT 02 Represents the oil cooler temperature difference between the two computing nodes, ΔT 01 , ΔT 02 The arithmetic mean or median value of all temperature difference monitoring values ​​in the neighborhood of the calculation node is selected, and the neighborhood of the calculation node includes the monitoring values ​​before or after the calculation node.

[0014] 2. The control center line is the average value of the temperature difference change rate curve of the lubricating oil cooler under each operating condition during historical normal operation, and the upper boundary is the control center line plus m times the standard deviation calculated based on historical data.

[0015] The advantage of the present invention is that the present invention can use process information to detect whether there is a fault in the gas turbine lubricating oil cooler. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of the present invention;

[0017] Figure 2 This is the temperature difference change rate curve of the lubricating oil cooler under various operating conditions during normal operation of the gas turbine. DETAILED DESCRIPTION

[0018] The present invention is described in more detail below with reference to the accompanying drawings:

[0019] Combination Figure 1-2 , the implementation of the present invention comprises the following steps:

[0020] A return oil temperature change rate model of the oil cooler under various operating conditions during normal operation of the unit is established.

[0021] The historical data of the normal operation of the gas turbine is selected to calculate the return oil temperature change rate of the oil cooler under different working conditions, and establish a normal return oil temperature change rate curve, including the upper boundary and the control center line.

[0022] 1) The horizontal axis parameters are the engine speed n, the gas turbine exhaust temperature T5 or the engine power N;

[0023] 2) The calculation of the return oil temperature change rate is:

[0024] (ΔT 02 -ΔT 01 ) / t

[0025] Where t is the calculation period; ΔT 01 , ΔT 02 Respectively represent the temperature difference of the oil cooler at the two computing nodes. Considering the instability of the actual monitoring value, ΔT 01 , ΔT 02 The arithmetic mean or median value of all temperature difference monitoring values ​​in the neighborhood of the calculation node is selected, and the neighborhood of the calculation node includes several monitoring values ​​before or after the calculation node.

[0026] 3) The return oil temperature change rate curve includes the upper boundary and the control center line, which is a trend curve that changes with the working conditions. The control center line is the average value (geometric mean, arithmetic mean or other weighted mean) of the temperature difference change rate curve of the lubricating oil cooler under each working condition during historical normal operation, and the upper boundary is the control center line plus m times the standard deviation based on historical data statistics. The value of m is adjusted according to the actual data. This process can be performed offline.

[0027] The oil return temperature of the lubricating oil cooler during the operation of the gas turbine is collected, and the temperature difference and change rate of the lubricating oil cooler inlet and outlet are calculated.

[0028] 1) Collect the speed (n), gas turbine exhaust temperature (T5), power (N) and oil cooler temperature difference (ΔT) during the operation of the gas turbine;

[0029] 2) Preprocessing the collected gas turbine operation data, including filtering, conversion, standardization, etc., and calculating the return oil temperature change rate of the lubricating oil cooler for a given calculation period t;

[0030] Compare the current oil return temperature change rate of the gas turbine with the oil return temperature difference change rate curve of the oil return cooler under normal operation. When the difference between the oil return temperature change rate of the gas turbine and the oil return temperature difference change rate of the oil return cooler under normal operation is greater than the threshold, abnormal state 1:

[0031] Step 4: If the distance between the calculated value of the temperature difference change rate and the control center line is greater than the distance between the upper boundary and the control center line, then it is marked as abnormal state 2;

[0032] Step 5: Evaluate abnormal state 1 and abnormal state 2 of the lubricating oil cooler. When abnormal state 1 and abnormal state 2 occur at the same time, mark the lubricating oil cooler as abnormal and the detection is completed.

Claims

1. A method for detecting a fault in a gas turbine oil cooler, characterized in that: (1) Based on the historical operation data of the unit, a return oil temperature change rate model of the gas turbine lubricating oil cooler under different operating conditions is constructed. The return oil temperature change rate model includes an upper boundary and a control center line; (2) Collect the oil return temperature of the lubricating oil cooler during the operation of the gas turbine and calculate the temperature difference and change rate of the lubricating oil cooler inlet and outlet; (3) Detecting the oil return temperature change rate of the lubricating oil cooler during actual operation of the unit, and marking the oil return temperature change rate as abnormal state 1 when it is greater than a threshold value; (4) Detecting the temperature difference of the lubricating oil at the inlet and outlet of the lubricating oil cooler, and marking it as abnormal state 2 when the temperature difference exceeds a threshold; (5) Comprehensively evaluate the abnormal state 1 and the abnormal state 2 of the lubricating oil cooler. When the abnormal state 1 and the abnormal state 2 occur at the same time, the lubricating oil cooler is marked as abnormal and the detection is completed; The establishment process of the oil return temperature change rate model: The horizontal axis parameter uses the engine speed n, and the calculation of the oil return temperature change rate is (ΔT 02 -ΔT 01 ) / t, t is the calculation period, ΔT 01 , ΔT 02 Represents the oil cooler temperature difference between the two computing nodes, ΔT 01 , ΔT 02 The arithmetic mean or median value of all temperature difference monitoring values ​​in the neighborhood of the calculation node is selected, and the neighborhood of the calculation node includes the monitoring values ​​before or after the calculation node.

2. A method for detecting a fault of a gas turbine lubricating oil cooler according to claim 1, characterized in that: The control center line is the average value of the temperature difference change rate curves of the lubricating oil cooler under various operating conditions during historical normal operation, and the upper boundary is the control center line plus m times the standard deviation calculated based on historical data.

Images