CN-121976885-A - Gas turbine natural gas hydrogen-loading dynamic combustion control optimization method and system

CN121976885ACN 121976885 ACN121976885 ACN 121976885ACN-121976885-A

Abstract

The invention discloses a method and a system for optimizing natural gas loading dynamic combustion control of a gas turbine, relates to the technical field of combustion control of the gas turbine, and aims to solve the problems of poor stability, exceeding NOx emission and insufficient working condition suitability in loading combustion of the gas turbine. The method comprises the steps of 1, obtaining multidimensional operation parameters such as temperature distribution, pressure pulsation, tail gas components and the like of a combustion chamber, constructing a combustion state matrix, 2, extracting spatial characteristics of a temperature field, pressure and tail gas time sequence characteristics based on the matrix, fusing the spatial characteristics and the pressure and the tail gas time sequence characteristics to obtain a combustion stability index and a NOx emission predicted value, 3, dynamically adjusting the hydrogen loading ratio according to the index and the predicted value, optimizing fuel distribution through a multi-stage swirl injector, inhibiting a local high-temperature area, and 4, determining an optimal combustion control strategy of premixing-dilution, graded diffusion or self-adaptive mixing by combining the real-time combustion state characteristics and load requirements. The system correspondingly comprises a combustion state, prediction, tail gas optimization and strategy control module. The invention realizes double optimization of combustion efficiency and pollutant discharge, adapts to different load working conditions, and improves the stability and economy of hydrogen-doped combustion.

Inventors

XIAO JUNFENG
GUO HAN
ZHENG JUNHUI
ZHANG XIAOXU
WANG FENG
GAO SONG
WANG WEI
LI XIAOFENG
YU QIANQIAN
LI LE
LI DAN
XIA JIAXING

Assignees

西安热工研究院有限公司

Dates

Publication Date: 20260505
Application Date: 20260323

Claims (10)

1. The natural gas loading dynamic combustion control optimization method of the gas turbine is characterized by comprising the following steps of: Step 1, acquiring multidimensional operation parameters of a combustion chamber of a gas turbine, and constructing a combustion state matrix according to the multidimensional operation parameters; Step 2, acquiring a temperature field space feature vector of a combustion chamber and a time feature vector of pulsation pressure and tail gas components based on a combustion state feature matrix, and fusing the space feature vector and the time feature vector to obtain a combustion stability index and a NOx emission predicted value; step 3, dynamically adjusting the hydrogen loading ratio of the natural gas-hydrogen mixer according to the current combustion stability index and the NOx emission value, optimizing the fuel space distribution through the rotational flow strength of the multi-stage rotational flow fuel injector, and inhibiting a local high temperature region; And 4, determining combustion state characteristics according to the real-time hydrogen loading proportion, the rotational flow strength and the fuel distribution state, and determining an optimal combustion control strategy according to the combustion state characteristics and the load demand.
2. The method for optimizing natural gas loading dynamic combustion control of a gas turbine according to claim 1, wherein said constructing a combustion state matrix based on multi-dimensional operating parameters comprises: the multidimensional operating parameters comprise temperature distribution, pressure pulsation and tail gas components; Acquiring the position and gradient of a high temperature region of a combustion chamber by adopting an infrared thermal imager, and determining temperature distribution according to the position and gradient of the high temperature region; A high-frequency pressure sensor is adopted to acquire a pressure signal of the combustion chamber, and the pressure signal is subjected to fast Fourier transformation to obtain pressure pulsation; the tail gas components include concentrations of NOx, CO, and O 2 ; And constructing a combustion state matrix according to the temperature distribution, the pressure pulsation and the tail gas components.
3. The method for optimizing natural gas loading dynamic combustion control of a gas turbine according to claim 1, wherein the obtaining a temperature field space feature vector of a combustion chamber and a time feature vector of a pulsation pressure and an exhaust gas component based on a combustion state feature matrix comprises: the combustion state feature matrix is input into a CNN-LSTM fusion model, the CNN-LSTM fusion model extracts temperature field space features through a convolutional neural network, and the CNN-LSTM fusion model analyzes time sequence dependency relationship of pressure pulsation and tail gas components through a long-term and short-term memory network to obtain a time feature vector.
4. The method for optimizing natural gas loading dynamic combustion control of a gas turbine according to claim 3, wherein the fusing of the spatial feature vector and the temporal feature vector to obtain the combustion stability index and the predicted NOx emission value comprises: And according to the influence weights of the space feature vector and the time feature vector on combustion stability evaluation and NOx emission, fusing the space feature vector and the time feature vector, and outputting the current combustion stability index and NOx emission value through an output layer of the CNN-LSTM fusion model.
5. The method for optimizing the dynamic combustion control of natural gas loading of a gas turbine according to claim 1, wherein the method for dynamically adjusting the loading ratio of the natural gas-hydrogen mixer is as follows: Comparing the current combustion stability index with a preset value, and adjusting the hydrogen loading ratio according to the comparison result; if the combustion stability index is lower than the threshold value, the combustion is unstable, and the hydrogen loading proportion of the mixer is reduced; And if the combustion stability index is higher than the threshold value and the NOx emission value exceeds the standard, increasing the hydrogen loading ratio.
6. The method for optimizing natural gas loading dynamic combustion control of a gas turbine according to claim 5, wherein optimizing fuel spatial distribution by swirl strength of a multi-stage swirl fuel injector, suppressing a local high temperature zone, comprises: Controlling the blade angle of the multi-stage swirl fuel injector according to the combustion state to optimize the mixing uniformity of fuel and air, and spatially inhibiting the formation of a local high-temperature zone so as to reduce the generation of NOx; The adjusting process is to adjust parameters according to the change of the combustion stability index and the NOx emission value, so as to ensure that the combustion state is always in an optimal interval.
7. The method for optimizing natural gas loading dynamic combustion control of a gas turbine according to claim 1, wherein determining combustion state characteristics based on real-time loading ratio, swirl strength and fuel distribution state, determining an optimal combustion control strategy based on the combustion state characteristics and load demand, comprises: the combustion control strategy includes a premix-dilution mode, a staged diffusion mode, and an adaptive mixing mode; at low load, a premixing-diluting mode is adopted, and the premixed combustion is combined with tail gas recirculation, so that flame temperature and NOx generation are reduced; In the high load, a graded diffusion mode is adopted, and the first stage injection reduces the hydrogen proportion in the mixed fuel to form a stable ignition source; When the load changes, a self-adaptive mixing mode is adopted to dynamically adjust the premixing and diffusion combustion proportion, and the efficiency and the emission are balanced.
8. A gas turbine natural gas loading dynamic combustion control optimization system, comprising: The combustion state module is used for acquiring multidimensional operation parameters of the combustion chamber of the gas turbine and constructing a combustion state matrix according to the multidimensional operation parameters; the prediction module is used for acquiring a temperature field space feature vector of the combustion chamber and a time feature vector of the pulsation pressure and tail gas components based on the combustion state feature matrix, and fusing the space feature vector and the time feature vector to obtain a combustion stability index and a NOx emission value; The tail gas optimizing module is used for dynamically adjusting the hydrogen loading proportion of the natural gas-hydrogen mixer according to the current combustion stability index and the NOx emission value, optimizing the fuel space distribution through the rotational flow intensity of the multi-stage rotational flow fuel injector and inhibiting a local high temperature region; and the strategy control module is used for determining the combustion state characteristics according to the real-time hydrogen loading proportion, the rotational flow strength and the fuel distribution state, and determining the optimal combustion control strategy according to the combustion state characteristics and the load demand.
9. An electronic device, comprising: A memory for storing a computer program; A processor for performing the steps of the gas turbine natural gas loading dynamic combustion control optimization method of any one of claims 1-8 when executing the computer program.
10. A computer readable storage medium, wherein a computer program is stored in the computer readable storage medium, which computer program, when being executed by a processor, implements the steps of the gas turbine natural gas loading dynamic combustion control optimization method according to any one of claims 1-8.

Description

Gas turbine natural gas hydrogen-loading dynamic combustion control optimization method and system Technical Field The invention relates to the technical field of combustion control of gas turbines, in particular to a method and a system for optimizing natural gas hydrogen-loading dynamic combustion control of a gas turbine. Background The gas turbine hydrogen-loading combustion technology is an important way to realize low-carbonization power generation, but the high diffusivity and rapid combustion characteristics of hydrogen lead to the reduction of combustion stability when the hydrogen loading ratio is changed. In the prior art, a control strategy for fixing the hydrogen loading proportion is difficult to adapt to load fluctuation and fuel characteristic change, and the problems of flame flickering, local high-temperature region generation, exceeding of NOx emission and the like are easily caused. In addition, the traditional method lacks real-time monitoring and feedback control on the temperature field, the pressure field and the flame dynamic characteristics in the combustion chamber, so that the combustion efficiency and the emission reduction target cannot be considered. In recent years, a dynamic hydrogen loading regulation technology is proposed, but the combustion state cannot be accurately reflected due to low data fusion degree of a sensor in the dynamic hydrogen loading process, and secondly, a control strategy is single, and a multi-mode combustion mode is not combined to adapt to different working conditions, so that long-term stable operation is difficult to realize. Disclosure of Invention Aiming at the problems in the prior art, the invention provides a method and a system for optimizing natural gas hydrogen-loading dynamic combustion control of a gas turbine, which realize double optimization of combustion efficiency and pollutant emission through multi-source sensor data fusion, dynamic hydrogen-loading proportion regulation and multi-mode combustion mode cooperative switching. The invention is realized by the following technical scheme: In a first aspect, the application provides a gas turbine natural gas loading dynamic combustion control optimization method, comprising the following steps: Step 1, acquiring multidimensional operation parameters of a combustion chamber of a gas turbine, and constructing a combustion state matrix according to the multidimensional operation parameters; Step 2, acquiring a temperature field space feature vector of a combustion chamber and a time feature vector of pulsation pressure and tail gas components based on a combustion state feature matrix, and fusing the space feature vector and the time feature vector to obtain a combustion stability index and a NOx emission predicted value; step 3, dynamically adjusting the hydrogen loading ratio of the natural gas-hydrogen mixer according to the current combustion stability index and the NOx emission value, optimizing the fuel space distribution through the rotational flow strength of the multi-stage rotational flow fuel injector, and inhibiting a local high temperature region; And 4, determining combustion state characteristics according to the real-time hydrogen loading proportion, the rotational flow strength and the fuel distribution state, and determining an optimal combustion control strategy according to the combustion state characteristics and the load demand. Preferably, the constructing a combustion state matrix according to the multi-dimensional operation parameters includes: the multidimensional operating parameters comprise temperature distribution, pressure pulsation and tail gas components; Acquiring the position and gradient of a high temperature region of a combustion chamber by adopting an infrared thermal imager, and determining temperature distribution according to the position and gradient of the high temperature region; A high-frequency pressure sensor is adopted to acquire a pressure signal of the combustion chamber, and the pressure signal is subjected to fast Fourier transformation to obtain pressure pulsation; the tail gas components include concentrations of NOx, CO, and O 2; And constructing a combustion state matrix according to the temperature distribution, the pressure pulsation and the tail gas components. Preferably, the obtaining a temperature field space feature vector of the combustion chamber and a time feature vector of the pulsation pressure and the tail gas component based on the combustion state feature matrix includes: the combustion state feature matrix is input into a CNN-LSTM fusion model, the CNN-LSTM fusion model extracts temperature field space features through a convolutional neural network, and the CNN-LSTM fusion model analyzes time sequence dependency relationship of pressure pulsation and tail gas components through a long-term and short-term memory network to obtain a time feature vector. Preferably, the fusing the spatial feature vector and the temporal