RU2813716C1 - Method for diagnosing technical condition of gas turbine engine - Google Patents

Abstract

FIELD: non-destructive testing; gas turbine engines.

SUBSTANCE: invention relates to non-destructive testing of the technical condition of gas turbine engines (GTE). The method for diagnosing the technical condition of a gas turbine engine is to select parameters subject to diagnostic control, the current value of which is measured on the gas turbine engine being diagnosed (P1, P2 … Pn). During engine operation, each of the measured parameters is processed by two channels with different frequency characteristics, channel outputs (P11, P12, P21, P22, …, Pn1, Pn2) are compared with each other and the conclusion that the engine being diagnosed is faulty is made by the magnitude of the difference (ΔP1=P11-P12, ΔP2=P21-P22, …, ΔPn=Pn1-Pn2,). Additionally, the current (nER) and average (nERav) values of the engine rotor speed are determined, as well as the magnitude of their difference (Δ nER=| nER-nERav |), a threshold based on the rotor speed (nERmax) is selected, a time interval τ (τ1, τ2, …, τn) for each of the monitored parameters is selected. If ΔnER exceeds the selected threshold nERmax, then single values of logical signals L (L1, L2, …, Ln), each of which is stored during the corresponding time interval selected for the controlled parameter τ as Δ nER decreases below the threshold nERmax, and with single values of logical signals L, the previously determined values Р11, Р12, Р21, Р22, …, Pn1, Pn2 are assigned equally to the current values of parameters P1, P2 … Pn accordingly.

EFFECT: eliminating the generation of false diagnostic messages when the engine operating mode changes and increasing the accuracy and reliability of the gas turbine engine diagnostic system.

1 cl, 3 dwg

Description

The invention relates to non-destructive testing of the technical condition of gas turbine engines (GTE), namely to methods for technical diagnostics of defects that affect the performance of gas turbine engine components during testing and operation, and can find application in engine building to identify the presence of defects.

The closest to the claimed invention in technical essence and the achieved technical result is a method for diagnosing the technical condition of a gas turbine engine, which consists in selecting parameters that are subject to diagnostic control, the current value of which is recorded on the gas turbine engine being diagnosed, for diagnosing the gas turbine engine according to any of its components for of the latter, at least two parameters are selected that characterize its performance and their maximum permissible deviation values for a given type of engine are experimentally determined, after which, during the operation of the engine at the current time, the average value for each selected parameter is calculated for the previous short and long time periods, with the ratio of a short time period to a long one in the range of 0.002-0.1, and determine their difference, then calculate the ratio of the resulting differences to the corresponding maximum permissible values of parameter deviations, and then sum them up, and if the resulting sum of ratios exceeds one, then a conclusion is drawn about a malfunction engine being diagnosed (RF patent 2745820 C1, 04/01/2021).

As a result of the analysis of this method, it should be noted that in order to diagnose the engine condition, the monitored parameters are processed by two channels with different frequency characteristics, from which the low-frequency channel selects the average value, and the decision about the malfunction is made based on the mismatch of the output signals of the channels. This allows you to obtain reliable results at steady-state engine operating conditions. However, the values of a number of monitored parameters, for example, the temperatures of engine components, significantly depend on the operating mode, therefore, during the transient process, the difference in the output signals of the measuring channels can lead to the formation of false diagnostic signals when the engine and its systems are in good condition.

The technical result of the proposed method for diagnosing a technical condition is to eliminate the generation of false diagnostic messages when the engine operating mode changes and to increase the accuracy and reliability of the gas turbine engine diagnostic system.

The specified technical result is achieved by the fact that in the known method for diagnosing the technical condition of a gas turbine engine, which consists in selecting parameters subject to diagnostic control, the current value of which is measured on the gas turbine engine being diagnosed (P1, P2 ... Pn), during engine operation, each of the measured parameters are processed by two channels with different frequency characteristics, the outputs of the channels (P11, P12, P21, P22, ..., Pn1, Pn2) are compared with each other and by the magnitude of the difference (ΔP1=P11-P12, ΔP2=P21-P22, ΔPn=Pn1 -Pn2,), make a conclusion about the malfunction of the diagnosed engine, according to the proposal, additionally determine the current (nРД) and average (nРДср) values of the engine rotor speed, as well as their module of their difference (ΔnРД=| nРД-nРДср |), select a threshold by rotor rotation frequency (nРДmax), for each of the monitored parameters the time interval τ (τ1, τ2, …, τn) is selected, if ΔnРД exceeds the selected threshold nРДmax, then single values of logical signals L (L1, L2, …, Ln) are generated , each of which is maintained during the corresponding time interval τ selected for the monitored parameter after ΔnРД decreases below the threshold nРДmax, and with single values of logical signals L, the previously determined values P11, P12, P21, P22,..., Pn1, Pn2 are assigned equal current values of parameters P1, P2 ... Pn, respectively.

The claimed invention is illustrated by the following detailed description of its implementation with reference to the figures, which show diagrams of the gas turbine engine diagnostic system.

In fig. Figure 1 shows the general enlarged structure of the diagnostic system, which shows:

- P1, P2, …, Pn - sensors for measuring parameters subject to diagnostic control;

- each of the parameters is processed by a signal processing block containing two measurement channels with different frequency characteristics, the structure of the block is shown in Figure 2, the output signal of the block is the difference in the output signals of the measuring channels;

- for each of the parameters, a logical signal L (L1, L2, …, Ln) is generated by its own logical timer;

- n _RD - engine rotor speed measurement sensor;

- block for generating the “Transient mode” sign, which controls the operation of timers;

- a decision-making block that generates a fault signal depending on the difference in the output signals of the measuring channels.

A logical timer is a logic element that has one input and one output. The output value of the timer is a logical signal L (L1, L2, ..., Ln).

L is set to "1" instantly if the timer input value is set to "1". L is set to “0” if the timer input value is held at “0” for time τ. The initial state of the timer output is “0”.

After changing the operating mode, the parameter values are stabilized at a new level through the transition process.

Timer times τ are selected for each of the parameters individually (τ1, τ2, ..., τn) and correspond to the times of transient modes for each of the monitored parameters. The transition time consists of two intervals - the time of transition of the engine to a new state and the time of establishment of the controlled parameter. The time to establish controlled parameters individually for each parameter is determined by the design features of the engine and can be determined experimentally.

Figure 2 shows the structure of the signal processing block for parameter P1; other signal processing blocks in the system are designed in a similar way.

The input signals of the block are the value of the measured parameter (P1), and the logical signal L1 corresponding to the parameter. Block output signal: difference in output signals of measuring channels:

ΔP1=P1 ₁ -P1 ₂ .

The signal processing block contains:

- the first measurement channel, consisting of integrator 2, covered by negative feedback through summing amplifier 1, the signal from sensor P1 is connected to the first input of summing amplifier 1, the value of integrator 2 is the measurement signal of the first channel P1 ₁ ;

- the first measurement channel, consisting of a second integrator 4, covered by negative feedback through the second summing amplifier 3, a signal from sensor P1 is connected to the first input of summing amplifier 3, the value of integrator 4 is the measurement signal of the second channel P1 ₂ ;

- adder 5, to the inputs of which the outputs of integrators 2 and 4 are connected, the adder generates the difference in the signals of the measurement channels: ΔPl=P1 ₁ -P1 ₂ ;

- key 6, to the controlled input of which the logical signal L1 is connected, the signal from the sensor P1 is connected to the functional input of the key, and the output of the key is connected to the inputs for setting the values of integrators 2 and 4; when a signal “1” arrives at the controlled input of key 6, the key closes and the signal from sensor P1 is sent to the inputs for setting the values of integrators 2 and 4, the values accumulated by the integrators are replaced with the current values of parameter P1.

The integrator, covered by feedback, is a first-order filter, the time constant of which determines the cutoff frequency of the filter and is numerically equal to the reciprocal of the gain of the summing amplifier.

In the general case, the time constants of the first 2 and second 4 integrators differ several times (10...500). For definiteness, we will assume that integrator 2 has a time constant of the order of T=6 minutes, and integrator 4 - 240⋅T=24 hours.

The output value (P1 ₂ ) of the second integrator 4 (“slow”) reflects the state of the engine according to the controlled parameter P1, taking into account external factors (temperature, pressure) and is the basis for detecting a failure when the value (P1 ₁ ) of the first integrator 2 (“fast”) changes ), which represents the "instantaneous" state of the engine. The appearance of a difference between the output values of the integrators means a change in the state of the node, characteristic of a failure. This difference is calculated on the summing amplifier 5 and analyzed in the decision-making block.

The structure of the “Transient Mode” signal generation block is presented in Figure 3.

The input signal of the block is a signal from the engine rotor speed sensor - n _RD .

The output signal of the block is a logical signal “Transient mode”, which controls the operation of timers in the diagnostic system.

The signal generation block “Transient mode” contains:

- integrator 8, covered by negative feedback through summing amplifier 7, the signal from sensor n _RD is connected to the first input of summing amplifier 7; the output value of the integrator 8 is the average value of the rotor speed (n _RDsr );

- rectifier 9, to the input of which the output of summing amplifier 7 is connected; at the output of rectifier 9, the module of the difference between the current and average value of the rotation speed Δn _RD is formed: Δn _RD =| n _RD -n _R d _av |;

- comparator 10, to the input of which the output of the rectifier 9 is connected; the comparator generates a logical one signal (“1”) at its output when Δn _RD exceeds the comparator setting - the threshold for rotation frequency (n _R d _max ), and generates a logical zero signal (“0”) when Δn _RD is less than n _R d _max ; signal “1” corresponds to the transient mode of engine operation.

In the transition mode, the initial conditions of the measuring channel integrators are set. After the end of the transition mode for each of the monitored parameters, its measuring channels are put into operation.

If you do not take additional actions, then when the engine operating mode changes, the “slow” integrator 4 in the signal processing block will not have time to change its value during the transition process, and a difference will arise at the adder 5, which can lead to the formation of a failure by the decision-making block.

In the transient operating mode of the gas turbine engine, a mismatch occurs between the actual rotor speed (n _RD ), measured by the sensor n _R d, and its average value (integrator signal 8 - n _R d _av ). When the module of the difference generated by the rectifier 9 of the block for generating the “Transient mode” sign exceeds the response threshold of the comparator 10, the latter generates a logical signal “1” at its output, meaning that the engine is in a transient operating mode. When the engine rotor speed reaches a new set level, the value of integrator 8 will begin to approach it, the magnitude of the difference will decrease, and the reverse operation of comparator 10 will occur - setting its output to “0”, meaning that the engine is in a steady state of operation.

Since some of the operating parameters of the gas turbine engine continue to change for some time after the end of the change in rotation speed (for example, parameters such as the temperature of the gases behind the turbine, the temperature of the oil, fuel, etc.), the signal about the end of the transition process to the diagnostic system should be removed with a delay determined for each parameter individually. This delay is implemented by system timers.

In accordance with the signal of the corresponding timer, key 6 in the measurement block is closed, and the signal from sensor P1 is supplied to the input of integrators 2 and 4, held there throughout the entire transition process, replacing the accumulated value of the integrators with the current value of parameter P1. The difference generated by adder 5 becomes equal to zero.

After the end of the transient process, the signal L1 = “1” at the output of the timer with a delay τ1 is removed, and the system continues to monitor engine operation in the normal manner.

The operation of signal processing blocks and timers corresponding to the P parameters is identical. The selected delays ensure that the signal processing units generate zero mismatches between measurement channels in transient processes.

Thus, in steady-state operating conditions, the system determines the deviation of the monitored parameters from the average value and generates diagnostic messages when the deviations exceed the tolerance. In transient operating modes the control is disabled.

This allows for early detection of failures due to the high sensitivity of the system and at the same time eliminates false generation of diagnostic messages, thereby increasing the accuracy and reliability of the diagnostic system.

Claims (10)

A method for diagnosing the technical condition of a gas turbine engine, which consists in the fact that select parameters subject to diagnostic control, the current value of which is measured on the gas turbine engine being diagnosed (P1, P2 ... Pn), during engine operation, each of the measured parameters is processed by two channels with different frequency characteristics, the channel outputs (P1 ₁ , P1 ₂ , P2 ₁ , P2 ₂ ,..., Pn ₁ , Pn ₂ ) are compared with each other and by the magnitude of the difference (ΔP1=P1 ₁ -P1 ₂ , ΔP2=P2 ₁ -P2 ₂ , ΔPn=Pn ₁ -Pn ₂ ,), conclude that the engine being diagnosed is faulty, characterized in that additionally determine the current (n _R d) and average (n _R d _av ) values of the engine rotor speed, as well as their module of their difference (Δn _R d = | n _R d-n _Rdsr |), select a threshold based on rotor speed (n _Р d _max ), for each of the monitored parameters, select the time interval τ (τ1, τ2, …, τn), if Δn _RD exceeds the selected threshold n _R d _max , then single values of logical signals L (L1, L2, ..., Ln) are formed, each of which is stored during the corresponding time interval τ selected for the controlled parameter after Δn _Р d decreases below the threshold n _Р d _max , and for single values of logical signals L, the previously determined values P1 ₁ , P1 ₂ P2 ₁ , P2 ₂ , ..., Pn ₁ , Pn ₂ are assigned equal to the current values of the parameters P1, P2 ... Pn, respectively.

Images