Combustion pulsating pressure trend monitoring method and early warning system

Abstract

The invention relates to the technical field of combustion process optimization, and discloses a combustion pulsating pressure trend monitoring method and early warning system, and the method comprises the steps: obtaining combustion pulsating pressure data in a combustion chamber; according to the combustion pulsating pressure data, pressure change trends of different frequency bands in a combustion chamber are predicted, and a multi-frequency-band pressure prediction trend is obtained; according to the multi-band pressure prediction trend, performing early warning on a pressure abnormal condition; according to the invention, based on the pulsating pressure data of different frequency bands in the combustion chamber, the pressure change trend of each frequency band is predicted, tiny fluctuation and abnormal change before a fault can be captured, a potential pressure fluctuation problem can be found in advance, pressure abnormal conditions in different frequency bands are dynamically detected, and the detection accuracy is improved. According to the method, the abnormal pressure position and the abnormal type can be accurately identified, potential abnormity can be early warned in advance, the combustion process is optimized, the source and the property of the problem are accurately positioned, the maintenance efficiency of workers is improved, and therefore the probability of accidents is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of combustion process optimization, and particularly to a method for monitoring the trend of combustion pulsation pressure and an early warning system. Background Art

[0002] Combustion pulsation pressure is an important physical phenomenon generated during the combustion process. It reflects the interaction of various factors such as flame propagation in the combustion chamber, mixing of fuel and air, and dynamic response of the combustion chamber structure. The magnitude and frequency of combustion pulsation pressure have important impacts on the operating stability, safety, and efficiency of equipment. Traditional methods for monitoring combustion pulsation pressure mainly rely on single or multiple pressure sensors to measure the pressure at specific positions in the combustion chamber, making it difficult to comprehensively reflect the pressure distribution throughout the combustion chamber; it is impossible to deeply analyze the frequency characteristics of combustion pulsation pressure, ignoring the differences in abnormal pressure thresholds in different frequency bands; only analyzing real-time pressure anomalies, it is impossible to prevent potential anomalies in advance.

[0003] For example, the Chinese patent with the authorization announcement number CN110966100B discloses a monitoring device and method for combustion oscillation, including: collecting a plurality of dynamic signals in and around the combustion chamber, and the plurality of dynamic signals at least include the pressure of the combustion chamber; analyzing each dynamic signal in the time domain, comparing the time-domain analysis result of each dynamic signal with the corresponding first threshold, and determining whether the time-domain analysis results of all dynamic signals are lower than the corresponding first threshold. If so, enter the early combustion oscillation diagnosis step, otherwise enter the next step; analyzing each dynamic signal in the frequency domain to obtain the indication value of each dynamic signal; performing fusion processing on the indication values to obtain the total indication value; determining whether the total indication value is lower than the second threshold. If so, it is determined that no combustion oscillation has occurred, otherwise it is determined that combustion oscillation has occurred.

[0004] The above prior art has the problems proposed in this background art: due to the complex and variable combustion process in the combustion chamber, it is difficult for a single sensor to comprehensively reflect the pressure distribution throughout the combustion chamber; it is impossible to deeply analyze the frequency characteristics of combustion pulsation pressure, ignoring the differences in abnormal pressure thresholds in different frequency bands; only analyzing real-time pressure anomalies, it is impossible to prevent potential anomalies in advance; to solve at least one of the above problems, the present invention proposes a method for monitoring the trend of combustion pulsation pressure and an early warning system. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the main object of the present invention is to provide a method for monitoring the trend of combustion pulsation pressure and an early warning system, which can effectively solve the problems in the background art. The specific technical solutions of the present invention are as follows:

[0006] A method for monitoring the trend of combustion pulsation pressure, including:

[0007] Obtain the combustion pulsation pressure data in the combustion chamber;

[0008] According to the combustion pulsation pressure data, predict the pressure change trends in different frequency bands in the combustion chamber respectively to obtain multi-frequency band pressure prediction trends;

[0009] Warn of abnormal pressure conditions according to the multi-frequency band pressure prediction trends.

[0010] Specifically, the step of predicting the pressure change trends in different frequency bands in the combustion chamber respectively according to the combustion pulsation pressure data to obtain multi-frequency band pressure prediction trends includes:

[0011] Perform frequency domain conversion on the combustion pulsation pressure data to obtain frequency domain pressure data;

[0012] According to a preset frequency band range, divide the frequency domain pressure data into multiple frequency bands, where the multiple frequency bands include a high frequency band and a non-high frequency band;

[0013] Predict the pressure change trends of each frequency band respectively to obtain multi-frequency band pressure prediction trends.

[0014] Specifically, the step of predicting the pressure change trends of each frequency band respectively to obtain multi-frequency band pressure prediction trends includes:

[0015] For the high frequency band pressure data, use a preset first pressure prediction model to predict the pressure change trend to obtain a first pressure change trend;

[0016] For the non-high frequency band pressure data, predict the pressure change trend according to a preset second pressure prediction model to obtain a second pressure change trend.

[0017] Specifically, the step of using a preset first pressure prediction model to predict the pressure change trend for the high frequency band pressure data to obtain a first pressure change trend includes:

[0018] According to preset high frequency fault types, obtain multiple target frequency spectra, where the high frequency fault types include incomplete combustion, excessive thermoacoustic oscillation, local cracks in the flame tube, and abnormal turbulence characteristics;

[0019] According to the multiple target frequency spectra, divide the high frequency band pressure data to obtain multi-spectrum pressure data;

[0020] Based on historical high frequency band pressure data, extract features from each spectrum pressure data to obtain a multi-spectrum pressure feature set;

[0021] According to the multi-spectrum pressure feature set, use the preset first pressure prediction model to predict each spectrum pressure respectively to obtain a first pressure change trend.

[0022] Specifically, for the non-high-frequency band pressure data, the pressure change trend is predicted according to a preset second pressure prediction model to obtain a second pressure change trend, including:

[0023] Feature extraction is performed on historical non-high-frequency band pressure data to obtain a non-high-frequency band pressure feature set;

[0024] According to the non-high-frequency band pressure feature set, the pressure is predicted by using a preset second pressure prediction model to obtain a second pressure change trend.

[0025] Specifically, the early warning of pressure anomalies according to the multi-band pressure prediction trend includes:

[0026] For the first pressure change trend in the high-frequency band, an abnormal warning model is used to give an early warning of pressure anomalies;

[0027] For the second pressure change trend in the non-high-frequency band, an early warning of pressure anomalies is given according to a preset pre-alarm value and high alarm value.

[0028] Specifically, for the first pressure change trend in the high-frequency band, using a preset abnormal warning model to give an early warning of pressure anomalies includes:

[0029] According to the first pressure change trend, a change trend curve of the pressure of each target frequency spectrum is drawn;

[0030] According to the change trend curve, the slope and amplitude increment of the curve are calculated to obtain the trend characteristics of the pressure of each target frequency spectrum;

[0031] According to the preset high-frequency fault types and historical fault pressure data, the alarm threshold is updated by using a preset fault frequency spectrum model to obtain an updated alarm threshold for the pressure of each frequency spectrum;

[0032] The trend characteristics are compared with the updated alarm threshold of the corresponding frequency spectrum to obtain the pressure anomaly position and anomaly type;

[0033] According to the pressure anomaly position and anomaly type, an early warning is issued through a preset abnormal warning model.

[0034] Specifically, obtaining the combustion pulsation pressure data in the combustion chamber includes:

[0035] Windowing processing is performed on the combustion pulsation pressure data to obtain windowed pressure data;

[0036] Resampling processing is performed on the windowed pressure data to obtain resampled data;

[0037] Combine the resampled data to obtain updated combustion pulsation pressure data.

[0038] Specifically, the resampling process of the windowed pressure data to obtain resampled data includes:

[0039] Perform decimation processing on the windowed pressure data to obtain decimated data;

[0040] Perform interpolation processing on the decimated data to obtain resampled data.

[0041] A combustion pulsation pressure warning system for implementing the combustion pulsation pressure trend monitoring method described above includes:

[0042] A data acquisition module that acquires combustion pulsation pressure data in the combustion chamber;

[0043] A pressure change trend prediction module that predicts the pressure change trends in different frequency bands in the combustion chamber based on the combustion pulsation pressure data to obtain multi-frequency band pressure prediction trends;

[0044] A warning module that warns of abnormal pressure conditions based on the multi-frequency band pressure prediction trends.

[0045] Compared with the prior art, the present application has the following beneficial effects:

[0046] Based on the pulsation pressure data in different frequency bands in the combustion chamber, the present application predicts the pressure change trends in each frequency band respectively, can detect potential pressure fluctuation problems in advance, dynamically detect abnormal pressure conditions in different frequency bands, can accurately identify the abnormal pressure position and abnormal type, can give early warnings of potential abnormalities, optimize the combustion process, achieve full combustion of fuel and a stable combustion state, accurately locate the source and nature of the problem, avoid affecting the smooth progress of the combustion process due to abnormal conditions in the combustion chamber, improve the maintenance efficiency and abnormal response speed of the staff, enhance the supervision of the combustion system, and thus reduce the probability of accidents. Description of the Drawings

[0047] Figure 1 It is the working flowchart of the combustion pulsation pressure trend monitoring method in Embodiment 1 of the present invention;

[0048] Figure 2 It is the working flowchart of the high-frequency band pressure abnormal condition detection in Embodiment 1 of the present invention;

[0049] Figure 3 It is the working flowchart of the high-frequency band pressure abnormal warning in Embodiment 1 of the present invention;

[0050] Figure 4This is a schematic structural diagram of the combustion pulsation pressure warning system in Embodiment 3 of the present invention. Detailed implementation manners

[0051] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following detailed description of the specific implementation manners of the present invention will be given in conjunction with the accompanying drawings of the specification.

[0052] In the following description, many specific details are set forth to facilitate a thorough understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0053] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other from other embodiments.

[0054] Embodiment 1:

[0055] This embodiment provides a method for monitoring the combustion pulsation pressure trend. As Figure 1 shown, the method for monitoring the combustion pulsation pressure trend includes:

[0056] S101. Obtain the combustion pulsation pressure data in the combustion chamber;

[0057] S102. According to the combustion pulsation pressure data, respectively predict the pressure change trends in different frequency bands in the combustion chamber to obtain the multi-frequency band pressure prediction trends;

[0058] S103. According to the multi-frequency band pressure prediction trends, give a warning about abnormal pressure conditions.

[0059] The traditional method of identifying and warning abnormal pressure conditions in the combustion chamber based on real-time pulsation pressure data and static thresholds in the combustion chamber cannot meet the requirements of early warning and complex working conditions. The embodiment of this application predicts the long-term pressure change trend in the combustion chamber according to the pulsation pressure data, can detect potential pressure abnormal conditions in the combustion chamber in advance, realizes early warning, and is convenient for formulating maintenance plans in time.

[0060] Specifically, as Figure 1Obtain the pulsating pressure data of the gas turbine during continuous operation, perform data preprocessing on it to prepare for subsequent analysis, conduct spectral analysis on the preprocessed data, analyze the trends of each frequency band separately, input the analysis results into the trained LSTM model, and at the same time set the fault judgment threshold by combining the real fault spectral data and the spectral data based on expert experience, reference documents, and similar equipment, dynamically adjust the threshold, and realize fault judgment; if the judgment result is correct, end the process; when the judgment result is inaccurate, retrain the LSTM model, and then judge again until it is accurate. When a fault situation is predicted, issue a fault warning.

[0061] Specifically, install high-temperature pulsating pressure sensors in the combustion chamber. These sensors can withstand the high-temperature environment in the combustion chamber and capture the pulsating pressure signals generated during the combustion process in real time. Collect these pulsating pressure data during the combustion process to reflect the state of the combustion process. According to the combustion pulsating pressure data, predict the pressure change trends in different frequency bands in the combustion chamber to obtain the long-term pressure change trends of each frequency band. Each frequency band reflects different pressure characteristics. By predicting the pressure change trends in different frequency bands, the dynamic characteristics of the pressure in the combustion chamber can be comprehensively understood, potential pressure fluctuation problems can be discovered in advance, and more accurate basis can be provided for combustion control and optimization.

[0062] Specifically, use different pressure anomaly analysis methods to perform anomaly analysis on the pressure change trends in different frequency bands, which can specifically detect the specific pressure anomaly positions and anomaly types, realize dynamic anomaly detection, dynamically update the alarm threshold for each frequency band according to historical pressure data and fault types. When the predicted pressure trend exceeds the alarm threshold, issue a warning and send a warning signal to relevant personnel and systems to remind them to take measures for processing in a timely manner. Through dynamic threshold anomaly warning, the alarm threshold can be updated in real time according to the combustion situation in the combustion chamber, improving the system's fault recognition ability and the accuracy of anomaly detection, more accurately positioning the source and nature of the problem, and improving the maintenance efficiency and anomaly response speed of the staff.

[0063] When abnormal pressure conditions are detected in the combustion chamber, the system issues an alarm to the operator through a preset alarm mechanism. The alarm mechanism can issue warnings to the operator through sound, light, or other means, reminding them to pay attention to the abnormal conditions in the combustion chamber. At the same time, the system can also record the abnormal data and relevant information for subsequent analysis and processing. By promptly reminding the operator to pay attention to the abnormal conditions in the combustion chamber, potential safety risks can be avoided, the supervision of the combustion system can be enhanced, and thus the probability of accidents can be reduced. The staff adjusts the combustion process by controlling the combustion speed and temperature according to the predicted state of the combustion chamber. From the fine adjustment of combustion conditions to the stable control of the overall combustion process, the comprehensive optimization of the combustion process ensures the stability of the production process and provides a solid guarantee for the consistency and reliability of product quality.

[0064] Based on the pulsating pressure data in different frequency bands in the combustion chamber, this application predicts the pressure change trends in each frequency band respectively, can detect potential pressure fluctuation problems in advance, dynamically detects abnormal pressure conditions in different frequency bands, can accurately identify the abnormal pressure position and abnormal type, can give early warnings for potential abnormalities, optimize the combustion process, achieve full combustion of fuel and a stable combustion state, accurately locate the source and nature of the problem, avoid the abnormal conditions in the combustion chamber from affecting the smooth progress of the combustion process, improve the maintenance efficiency and abnormal response speed of the staff, enhance the supervision of the combustion system, and thus reduce the probability of accidents.

[0065] Further, predicting the pressure change trends in different frequency bands in the combustion chamber respectively according to the combustion pulsating pressure data to obtain the multi-frequency band pressure prediction trends includes:

[0066] S201: Perform frequency domain conversion on the combustion pulsating pressure data to obtain frequency domain pressure data;

[0067] S202: Divide the frequency domain pressure data into multiple frequency bands according to a preset frequency band range, where the multiple frequency bands include a high frequency band and a non-high frequency band;

[0068] S203: Predict the pressure change trends in each frequency band respectively to obtain the multi-frequency band pressure prediction trends.

[0069] In this embodiment, the resampled combustion pulsation pressure data is converted into frequency-domain data by using Fourier transform. In the analysis of combustion pulsation pressure, the pressure signal varying with time is decomposed into the sum of sine waves of different frequencies by using Fourier transform, so as to reveal the frequency components in the signal. Through frequency-domain analysis, the frequency components in the signal can be visually displayed. The frequency band division is based on the frequency components of the signal. In the analysis of combustion pulsation pressure, the signal can be divided into a high-frequency band and a non-high-frequency band according to the frequency range. This division helps to detect pressure anomalies in different frequency ranges. Specifically, the non-high-frequency band usually refers to the frequency band below a certain threshold. For example, the frequency band below 2 kHz is regarded as the non-high-frequency band, and the high-frequency band usually refers to the frequency band above a certain threshold. For example, the frequency band above 2 kHz is regarded as the high-frequency band. The frequency band division helps to detect pressure anomalies in different frequency ranges in a targeted manner, and can improve the accuracy and efficiency of pressure anomaly detection.

[0070] Specifically, the pressure data in each frequency band has its unique variation law and characteristics. The pressure change trends in different frequency bands are predicted respectively, and the trends and patterns in the historical pressure data are used to predict the pressure change trend in the future time period, so as to obtain the pressure change conditions in each frequency band. This can provide richer and more accurate information for the operation state evaluation, fault diagnosis and optimal control of the combustion system, and can understand the dynamic change of the pressure in the combustion chamber more comprehensively and meticulously. For example, predicting the pressure change in the high-frequency band can detect abnormal fluctuations in the combustion chemical reaction in advance; predicting the pressure trend in the non-high-frequency band can understand the stability of the overall flow and fuel supply in the combustion chamber, providing a strong guarantee for the safe and stable operation of the combustion system.

[0071] Further, the step of predicting the pressure change trends in each frequency band respectively to obtain the multi-frequency band pressure prediction trends includes:

[0072] S301. For the high-frequency band pressure data, use the preset first pressure prediction model to predict the pressure change trend to obtain the first pressure change trend;

[0073] S302. For the non-high-frequency band pressure data, predict the pressure change trend according to the preset second pressure prediction model to obtain the second pressure change trend.

[0074] In this embodiment, during the combustion process in the combustion chamber, the high-frequency pressure data is related to some rapidly changing physical phenomena and chemical reactions, such as local turbulence during combustion, micro-scale oscillations of the flame, and rapid changes in the chemical reaction rate. The non-high-frequency pressure data is related to some macroscopic phenomena and slow-changing processes in the combustion system, such as the overall gas flow movement in the combustion chamber, low-frequency fluctuations in the fuel supply system, and the stability of the burner. The pressure change trends in the high-frequency band and the non-high-frequency band are predicted respectively, and the pressure prediction model is trained according to the historical pressure data in this frequency band to obtain a pre-trained pressure prediction model, and the pressure prediction model is used to predict the pressure change trend.

[0075] Further, for the high-frequency pressure data, using a preset first pressure prediction model to predict the pressure change trend to obtain a first pressure change trend, including:

[0076] S401. According to the preset high-frequency fault types, obtain a plurality of target spectra, where the high-frequency fault types include incomplete combustion, excessive thermoacoustic oscillation, local cracks in the flame tube, and abnormal turbulence characteristics;

[0077] S402. According to the plurality of target spectra, divide the high-frequency pressure data to obtain multi-spectrum pressure data;

[0078] S403. Based on the historical high-frequency pressure data, extract features from each spectrum pressure data to obtain a multi-spectrum pressure feature set;

[0079] S404. According to the multi-spectrum pressure feature set, use the preset first pressure prediction model to predict each spectrum pressure respectively to obtain a first pressure change trend.

[0080] In this embodiment, different high-frequency fault types will have specific manifestations on the spectrum of the combustion pulsation pressure. The physical processes and phenomena corresponding to each fault will cause pressure fluctuations in different frequency ranges. For example, for the fault of excessive thermoacoustic oscillation, the pressure data shows that the amplitudes at the 50th harmonic and the 75th harmonic increase rapidly. As Figure 2 , the initial fault alarm threshold is set according to references, experience of similar equipment, expert knowledge, etc. Use a large amount of high-frequency fault data for FFT spectrum analysis, with a time interval of 1 second or 10 minutes, to obtain the data of the high-frequency band spectrum at different time points (T1, T2,..., TN), such as the data in frequency bands such as the 40th harmonic and the 50th harmonic, and analyze the spectrum trends of each frequency band. Input the spectrum trend data into a pre-trained LSTM recurrent neural network, and the network performs operations through the input gate, forget gate, and output gate, and uses the loss function for optimization. Adjust the configuration according to the training results to obtain a fault prediction model to achieve the monitoring of high-frequency faults.

[0081] Specifically, dynamic analysis is performed on the pressure data in the high-frequency band. According to the preset high-frequency fault types, multiple target spectra are obtained. Further dividing the high-frequency band data can obtain more subtle dynamic information. By performing spectral analysis on the high-frequency band signals and combining with the trend analysis of the time series, the change trend of the high-frequency signals can be monitored in real time, so as to predict the following faults in the combustion chamber: incomplete combustion (abnormal fuel-air mixture ratio, uneven fuel distribution), excessive thermoacoustic oscillation (abnormal increase in the amplitude of the thermoacoustic oscillation mode in the combustion chamber), aging of burner components (such as local cracks in the flame tube, blockage or local damage of the combustion nozzle), and abnormal turbulence characteristics (increase in turbulence intensity or abnormal coupling of turbulent combustion).

[0082] Specifically, the pressure characteristic data of each target spectrum (such as 40 times frequency, 50 times frequency up to 200 times frequency) in the high-frequency range are extracted, including characteristics such as the maximum amplitude value and the pressure change rate, to obtain the characteristic set of the pressure of each spectrum. By performing feature extraction on the pressure data of each target spectrum to obtain the multi-spectrum pressure feature set, a large amount of original pressure data can be compressed into representative feature vectors, reducing the amount of data while retaining the key information related to the faults, and more accurately capturing the detailed information and potential fault characteristics in the pressure signal.

[0083] Specifically, the first pressure prediction model is trained using historical pressure data to obtain a pre-trained first pressure prediction model. Specifically, the first pressure prediction model is an LSTM model. According to the pressure feature set, the pressure data of different spectra are predicted respectively to obtain the predicted pressure data in the spectral ranges such as 40 times frequency, 50 times frequency, 75 times frequency, etc. By predicting the future pressure data, the abnormal change trend that may occur in the pressure can be discovered in advance, and a warning can be issued before the abnormality actually occurs, which helps to take timely measures to prevent the occurrence or expansion of the faults.

[0084] Furthermore, for the non-high-frequency band pressure data, the pressure change trend is predicted according to the preset second pressure prediction model to obtain the second pressure change trend, including:

[0085] S501. Feature extraction is performed based on the historical non-high-frequency band pressure data to obtain the non-high-frequency band pressure feature set;

[0086] S502. According to the non-high-frequency band pressure feature set, the second pressure prediction model is used to predict the pressure to obtain the second pressure change trend.

[0087] In this embodiment, the non-high-frequency band pressure data includes various information related to the combustion process. Through feature extraction, a non-high-frequency band pressure feature set is obtained, and features that are representative and indicative of the pressure change trend are excavated, including features such as mean, variance, skewness, and peak value, which can reflect some basic states and laws of the combustion process, such as the stability of combustion and the change of the average pressure level.

[0088] Specifically, the second pressure trend prediction model is an LSTM model. A large amount of historical frequency domain pressure feature data in the non-high-frequency band is used to train the LSTM model to obtain a preset second pressure trend prediction model. The preset second pressure trend prediction model is used to predict the pressure change trend in the non-high-frequency band to obtain the second pressure change trend, which can help the operator understand the change of the pressure in the non-high-frequency band in the combustion chamber in advance and provide a basis for taking measures in advance.

[0089] Further, warning of pressure anomalies according to the predicted trends of multi-band pressures includes:

[0090] S601. For the first pressure change trend in the high-frequency band, a preset anomaly warning model is used to warn of pressure anomalies;

[0091] S602. For the second pressure change trend in the non-high-frequency band, warning of pressure anomalies is carried out according to the preset pre-warning value and high-warning value.

[0092] In this embodiment, different methods are used to detect pressure anomalies in different frequency bands, which can perform targeted anomaly detection according to the pressure characteristics within the frequency band and meet the requirements of different frequency band ranges. For the pressure data in the high-frequency band, the pressure signal in the high-frequency band will contain some complex dynamic information and potential fault characteristics. In this embodiment, a preset anomaly warning model is used to identify and classify anomalies based on machine learning, automatically learn the characteristic patterns of the pressure signal in the high-frequency band, capture the characteristic distribution law of the normal pressure signal in the high-frequency band, and the difference characteristics between the abnormal pressure signal and the normal signal, detect the pressure anomalies in the high-frequency band, obtain the pressure anomaly position and anomaly type, and identify the pressure anomalies in the high-frequency band through the anomaly warning model, which can quickly detect the real-time input high-frequency band pressure signal and timely discover anomalies.

[0093] Specifically, for the pressure data in non-high-frequency bands, the abnormal pressure conditions are identified by setting abnormal alarm thresholds. The pressure signals in non-high-frequency bands are relatively stable, and their pressure values usually fluctuate within a certain range. By setting pre-alarm values and high alarm values, the value range of the pressure signal can be divided into different intervals. When the pressure value exceeds the pre-alarm value, it indicates that there may be potential abnormal conditions that need attention; when the pressure value exceeds the high alarm value, it means that the pressure abnormal condition is relatively serious and immediate measures need to be taken.

[0094] Specifically, detecting abnormal pressure conditions based on the pre-alarm value and high alarm value, the pre-alarm value and high alarm value can be set based on the understanding of the normal pressure fluctuation range and the expectation of abnormal conditions. They are set according to the mean and standard deviation of the normal pressure fluctuation. The formula is:

[0095]

[0096] In the formula, μ is the mean pressure within this frequency band, P i is the i-th pressure data value within this frequency band, and N is the total number of data points. The standard deviation calculation formula is as follows:

[0097]

[0098] In the formula, σ is the standard deviation of the pressure within this frequency band. The formulas for setting the pre-alarm value and high alarm value based on the mean and standard deviation are as follows:

[0099] T 预 = μ + k1×σ;

[0100] T 高 = μ + k2×σ;

[0101] In the formula, T 预 is the pre-alarm value, T 高 is the high alarm value, and k1 and k2 are set multiples. The values of k1 and k2 can be determined according to actual needs and safety requirements. Usually, the value of k1 should be less than the value of k2 to ensure that the pre-alarm value is triggered before the high alarm value. When setting the abnormal threshold, the characteristics of the pressure data and the actual application scenario need to be considered. For example, if the data fluctuates greatly, the values of k1 and k2 need to be appropriately increased; if the application scenario has high safety requirements, more stringent abnormal thresholds need to be set. The setting of the abnormal threshold is a dynamic process and needs to be adjusted and optimized according to the actual situation. In actual applications, the abnormal threshold can be adjusted by observing the alarm situation, analyzing data changes, etc. to improve the accuracy and reliability of monitoring.

[0102] When the pulsating pressure value exceeds the pre-alarm value but does not reach the high-alarm value, the system is considered to be in a pre-alarm state and needs to be closely monitored. At this time, by locating the pre-alarm points in the pulsating pressure change curve, the specific time and frequency band of the pre-alarm can be determined. By setting the pre-alarm value and the high-alarm value, hierarchical pre-alarms for pulsating pressure anomalies can be achieved. When the pulsating pressure value is within the pre-alarm range, the system can send a pre-alarm signal in advance to remind the operator to pay attention to the reaction situation in the combustion chamber and take necessary measures to avoid potential risks.

[0103] When the pulsating pressure value exceeds the high-alarm value, the system is considered to be in a serious abnormal state, and immediate measures need to be taken to avoid failures or accidents. At this time, by locating the alarm points in the pulsating pressure change curve, the specific time and frequency band of the alarm can be determined, and an alarm signal can be sent immediately. By setting the high-alarm value, the severity of the pulsating pressure anomaly can be judged. When the pulsating pressure value exceeds the high-alarm value, the system can immediately send an alarm signal to remind the operator to take emergency measures to avoid more serious consequences.

[0104] Furthermore, as Figure 3 described above, for the first pressure change trend in the high-frequency band, a preset abnormal pre-alarm model is used to pre-alarm pressure anomalies, including:

[0105] S701. According to the first pressure change trend, draw the change trend curve of each target spectral pressure;

[0106] S702. According to the change trend curve, calculate the slope and amplitude increment of the curve to obtain the trend characteristics of each target spectral pressure;

[0107] S703. According to the preset high-frequency fault types and historical fault pressure data, use the preset fault spectral model to update the alarm threshold to obtain the updated alarm threshold for each spectral pressure;

[0108] S704. Compare the trend characteristics with the updated alarm threshold of the corresponding spectrum to obtain the pressure anomaly position and anomaly type;

[0109] S705. According to the pressure anomaly position and anomaly type, send a pre-alarm through the preset abnormal pre-alarm model.

[0110] In this embodiment, according to the predicted pressure change trend of each frequency band, a change trend curve of each spectral pressure is drawn to display the pressure change trend and fluctuation. In the pulsating pressure change curve, different spectral positions reflect the change of the pressure signal under different frequency components. The slope represents the change rate of the curve at a certain point, reflecting the change speed of the pressure at that moment. The amplitude increment represents the change amount of the pressure amplitude between adjacent time points or within a specific time period. The slope and amplitude increment of the change trend curve at different positions are calculated to obtain the trend characteristics of each spectral pressure. Different fault types will cause abnormal changes in the slope and amplitude increment of the pressure curve. For example, when a loosening fault occurs in the equipment, there may be a phenomenon of increased slope and abnormal amplitude increment at the spectral positions of certain specific frequencies. By calculating the slope and amplitude increment at the preset spectral positions, abnormal situations can be identified and the types of abnormalities can be judged.

[0111] Specifically, the alarm thresholds in multiple preset fault type standards can be obtained based on a large amount of experimental data, actual operation experience, and theoretical analysis. The preset fault spectrum model is trained using historical fault pressure data to obtain a pre-trained fault spectrum model. According to the fault type and fault pressure data, the alarm thresholds for each frequency band are dynamically updated to obtain updated alarm thresholds. The updated alarm thresholds can better adapt to the changes in different working conditions and fault characteristics, reducing false alarms and missed alarms.

[0112] Specifically, according to the updated alarm thresholds, the trend characteristics are compared with the corresponding alarm thresholds. When the trend characteristics exceed the threshold range, it indicates that the spectral pressure is abnormal. According to the corresponding relationship between different high-frequency fault types and spectral pressure change characteristics, the type of abnormality can be further determined. The time and position information corresponding to the abnormal spectral pressure are recorded to determine the specific position and type of the pressure abnormality. The pressure abnormality position and type are promptly conveyed to relevant personnel. According to the pressure abnormality position and type, through the preset abnormal warning model, appropriate warning methods and contents can be selected, enabling operators to quickly locate and handle abnormalities, reducing the fault handling time, and improving the operation efficiency of the combustion system.

[0113] Further, the obtaining of the combustion pulsating pressure data in the combustion chamber includes:

[0114] S801. Perform windowing processing on the combustion pulsating pressure data to obtain windowed pressure data;

[0115] S802. Perform resampling processing on the windowed pressure data to obtain resampled data;

[0116] S803. Combine the resampled data to obtain updated combustion pulsating pressure data.

[0117] In this embodiment, windowing processing is performed on the combustion pulsation pressure data to obtain windowed pressure data. Windowing processing is a method of windowing continuous signals. In time series analysis, data is divided into multiple small pieces by applying a sliding window to the data. Common windows include rectangular windows, triangular windows, Hanning windows, Hamming windows, Gaussian windows, etc. An appropriate window can be selected according to the data characteristics and actual requirements. In the processing of real-time combustion pulsation pressure data, windowing processing can help reduce the boundary effect of the data and improve the ability to analyze the local characteristics of the data. By windowing processing, the boundary of the data can be smoothed, and the boundary effect caused by data truncation can be reduced. The windowed data can be more focused on a local time period or spatial region, thereby improving the ability to analyze the local characteristics of the combustion pulsation pressure data.

[0118] Specifically, resampling processing is performed on the windowed pressure data to obtain resampled data. The sampling rate is adaptively adjusted according to the data characteristics to obtain sampling data that better meets the requirements. By resampling processing, redundant information in the original data can be removed, and the quality and usability of the data can be improved. The resampled data is combined to form complete resampled combustion pulsation pressure data, which can maintain the integrity of the data and facilitate subsequent analysis of the combustion pulsation pressure.

[0119] Further, the resampling processing of the windowed pressure data to obtain resampled data includes:

[0120] S901. Decimation processing is performed on the windowed pressure data to obtain decimated data;

[0121] S902. Interpolation processing is performed on the decimated data to obtain resampled data.

[0122] In this embodiment, integer decimation is used to perform decimation processing on the windowed pressure data to obtain decimated data. Integer decimation means that for the original sampling sequence x(p), where p is the sampling point order of the original sampling sequence, one sampling point is extracted every D - 1 data to form a new sequence xD(m), where:

[0123] xD(m) = x(mD);

[0124] Where D is the decimation factor, and D is a positive integer. This decimation method is equivalent to reducing the data sampling frequency, that is, the spectral period of the signal after decimation is reduced to 1 / D of the original. By adjusting the decimation factor D, the sampling rate of the data can be adjusted to adapt to different application scenarios. Suppose there is an original sampling sequence x(n) = {1, 2, 3, 4, 5, 6, 7, 8}, and the decimation factor D = 2, then the new sequence xD(m) after decimation is {1, 3, 5, 7}; by processing the original sequence using the integer multiple decimation method, data is taken out at equal intervals and reordered, which can significantly reduce the amount of data, while retaining the main features of the signal, reducing the sampling rate to a suitable processing range, and at the same time eliminating the interference of out-of-band signal spectra and noise, reducing storage and calculation costs, and facilitating subsequent signal processing and analysis.

[0125] Specifically, the data after decimation is interpolated using the fractional multiple interpolation method to obtain the resampled data. After integer multiple decimation, the sampling rate of the data is reduced. To increase the sampling rate of the data, the fractional multiple interpolation method can be used. Through interpolation processing, denser data points can be obtained, so as to more accurately describe the change of the signal. In the fractional multiple interpolation method, it is necessary to first convert the ratio of the resampling frequency to the initial sampling frequency (i.e., the resampling multiple) into the form of a ratio of relatively prime integers, that is:

[0126] L = I / f;

[0127] Where L is the resampling multiple, I is the interpolation multiple, and f is the decimation multiple. First, perform I-fold interpolation on xD(m), and then perform f-fold decimation to obtain the resampled sequence. For example, assume that the resampling multiple L is 0.75, and the sampling frequency of the initial sampling sequence is 2×10 4 Hz, then the resampling frequency is 1.5×10 4 Hz. During the resampling process, the resampling multiple 0.75 needs to be regularized to 3 / 4 first, and at this time, the interpolation and decimation multiples I and f are obtained. Then, first perform 3-fold interpolation on the initial sampling sequence to obtain a discrete sequence with a sampling frequency of 6×10 4 Hz, and then perform 4-fold decimation to finally obtain a resampled sequence with a sampling frequency of 1.5×10 4 Hz.

[0128] Specifically, the number of sampling points determines the sampling time, and the sampling time is also the insertion position of the sampling value. The number of resampling points is related to the number of resampled values. After the number of resampled values is determined, the sampling time series is also determined. When 0 < L < 1, the change in the number of sampling points before and after resampling is described in two cases:

[0129] When the last point of the resampled sequence coincides with the last point of the initial sampling sequence, the number of sampling points of the resampled sequence can be calculated by the following formula:

[0130] Nr = L(N0 - 1) + 1;

[0131] Where N0 is the initial number of sampling points; N r is the number of resampling points.

[0132] When the last point of the resampling sequence does not coincide with that of the initial sampling sequence, the number of resampling points can be calculated by the following formula:

[0133] N r = [L(N0 - 1)] + 1;

[0134] Where [L(N0 - 1)] represents rounding down.

[0135] When L > 1, in the resampling process, the method of adding points is adopted, and the number of resampling points can be calculated by the following formula:

[0136] N r = [N0 × L] + 1;

[0137] After the number of resampling points is determined, the resampled data can be expressed as:

[0138]

[0139] Where t r (n) is the resampled time series data, n is the sampling point order of the resampled time series data, f r is the resampling frequency. First, the sampling rate is reduced to a suitable processing range through integer multiple decimation filtering, while eliminating the interference of out-of-band signal spectrum and noise. Then, fractional multiple interpolation filtering is used to adjust the sampling rate to synchronize with the transmitted symbols, improve the sampling rate of the data, and make the data smoother and more continuous.

[0140] In other embodiments, during the frequency band division process, real-time correction is performed. According to the preset frequency range, the pressure data is divided, and the obtained frequency division result is not accurate enough. There is a situation where data belonging to the high-frequency pressure range is divided into the low-frequency data; the high-frequency band pressure data is related to the rapid changes and high-frequency oscillations during the combustion process. Using these data, the fine characteristics of the combustion process can be captured. When the high-frequency band data is misclassified into the low-frequency band, it will interfere with the judgment of the combustion state; the division process is corrected according to the combustion curve.

[0141] Specifically, the initial high-frequency interval is determined according to the combustion curve, compared with the divided frequency band interval, the interval where the high-frequency band data misclassified into the low-frequency band is located is determined, the misclassified data is compared with the combustion curve, and according to the overall trend and characteristics of the combustion curve, the value and time position of the misclassified data are preliminarily adjusted to make the data trend match the combustion process.

[0142] Analyze the dynamic change rate of the combustion process according to the slope of the combustion curve in adjacent intervals of mis-segmented data. If the slope of the combustion curve of the mis-segmented data increases, it indicates that the combustion process changes faster. At this time, compress the mis-segmented data to make the data more compact on the time axis. If the slope decreases, it shows that the combustion process changes slower, then stretch the data to make the data extend on the time axis to better reflect the combustion state, and obtain the corrected mis-segmented pressure curve. Replace the initial mis-segmented pressure curve with the corrected mis-segmented pressure curve and divide it into the high-frequency range for subsequent analysis. By correcting the frequency band of the pressure data, the mis-segmentation problem of the high-frequency band data can be effectively corrected, and the corrected curve helps to more accurately monitor the dynamic changes during the combustion process and timely detect abnormal combustion conditions.

[0143] Embodiment 2:

[0144] In this embodiment, a cylindrical combustion chamber with a height of 10 meters and a diameter of 5 meters is provided. Pulse pressure sensors are installed in the combustion chamber. When the combustion process is in progress, these sensors will collect the combustion pulse pressure data in the combustion chamber in real time. Apply a Hanning window to the combustion pulse pressure data to obtain the windowed pressure data. Use the dynamic resampling method to resample the windowed pressure data to obtain the resampled pressure data. Combine the resampled pressure data to obtain the complete resampled combustion pulse pressure data.

[0145] According to the resampled combustion pulse pressure data, segment the frequency domain data. Set the frequency range of 300 - 2500 Hz as the non-high-frequency band, and set the frequency range of 2500 - 5000 Hz as the high-frequency band. Set the pre-alarm value and high-alarm value in each frequency band as shown in the following table:

[0146] Frequency band Frequency / Hz Pre-alarm value / kPa High alarm value / kPa Non-high frequency band 300～2500 17.24 24.13 High frequency band 2500～5000 9.07 12.45

[0147] Use the pressure anomaly detection method to detect the pressure anomaly conditions in different frequency bands in the combustion chamber. In each frequency band, use the preset pressure trend prediction model to predict the pressure trend respectively and draw the pressure change curve. When the pressure pulsation value is lower than the pre-alarm value, no alarm is issued. When the pressure pulsation value is higher than the pre-alarm value and lower than the high-alarm value, a warning is issued to remind the staff to pay attention to the reaction situation in the combustion chamber. When the pressure pulsation value is higher than the high-alarm value, an alarm is issued, and the staff takes emergency measures such as directly shutting down the machine.

[0148] Dynamically update the abnormal alarm threshold of the abnormal type according to the historical data of each octave pressure in the high-frequency band to obtain the updated abnormal threshold. For different specific octave positions in the high-frequency band, calculate the slope or amplitude increment of the pressure change trend, and compare it with the updated abnormal threshold related to different faults. When the trend index exceeds the specific threshold, trigger the alarm mechanism. The fault types and judgment criteria are as follows:

[0149] Incomplete combustion: The energy in the high-frequency band continuously increases within a specific frequency range (such as 40 octaves to 60 octaves), the slope exceeds the set threshold (such as >5% / min), and the broadband random noise energy increases significantly;

[0150] Excessive thermoacoustic oscillation: The amplitude of a specific spectrum (such as 50 octaves, 75 octaves) in the high-frequency band increases rapidly, reaching the critical value of the combustion chamber thermoacoustic mode, and the trend slope exceeds 10% / min;

[0151] Local crack in the flame tube: New frequency components or a sharp increase in amplitude appear in the high-frequency spectrum (such as 80 octaves to 100 octaves), and the spectrum characteristics show periodic fluctuations;

[0152] Abnormal turbulence characteristics: The randomness of the overall energy distribution in the high-frequency band increases, the spectral energy concentration decreases (the specific frequency peak disappears or the distribution expands), or the trend slope is continuously positive.

[0153] Judge the abnormal position and type of pressure according to the pressure change trend, and timely remind the operator to pay attention to the abnormal situation in the combustion chamber, which can avoid potential safety risks, improve the supervision of the combustion system, and thus reduce the probability of accidents.

[0154] Embodiment 3:

[0155] A combustion pulsation pressure warning system, such as Figure 4 , for implementing the combustion pulsation pressure trend monitoring method described above, includes:

[0156] A data acquisition module, which acquires the combustion pulsation pressure data in the combustion chamber;

[0157] A pressure change trend prediction module, which predicts the pressure change trends in different frequency bands in the combustion chamber respectively according to the combustion pulsation pressure data to obtain the multi-frequency band pressure prediction trends;

[0158] An early warning module, which gives an early warning of abnormal pressure conditions according to the multi-frequency band pressure prediction trends.

[0159] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and what is described in the above embodiments and the specification is only to illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. A combustion pulsation pressure trend monitoring method, characterized in that: include: Obtain combustion pulsation pressure data in the combustion chamber; According to the combustion pulsation pressure data, respectively predict the pressure change trends of different frequency bands in the combustion chamber to obtain a multi-band pressure prediction trend; According to the multi-band pressure prediction trend, an early warning is issued for abnormal pressure conditions.

2. The combustion pulsation pressure trend monitoring method according to claim 1, characterized in that: The method of predicting the pressure change trends of different frequency bands in the combustion chamber according to the combustion pulsation pressure data to obtain a multi-band pressure prediction trend includes: The combustion pulsation pressure data is converted into frequency domain to obtain frequency domain pressure data; According to a preset frequency band range, the frequency domain pressure data is divided into a plurality of frequency bands, wherein the plurality of frequency bands include a high frequency band and a non-high frequency band; The pressure variation trend of each frequency band is predicted respectively to obtain the multi-band pressure prediction trend.

3. The combustion pulsation pressure trend monitoring method according to claim 2, characterized in that: The pressure change trend of each frequency band is predicted respectively to obtain a multi-band pressure prediction trend, including: For high-frequency band pressure data, a preset first pressure prediction model is used to predict the pressure change trend to obtain a first pressure change trend; For non-high-frequency band pressure data, a pressure change trend prediction is performed according to a preset second pressure prediction model to obtain a second pressure change trend.

4. The combustion pulsation pressure trend monitoring method according to claim 3, characterized in that: The method of predicting the pressure change trend of the high-frequency band pressure data by using a preset first pressure prediction model to obtain a first pressure change trend includes: According to the preset high-frequency fault types, a plurality of target frequency spectra are obtained, wherein the high-frequency fault types include incomplete combustion, excessive thermoacoustic oscillation, local cracks in the flame tube, and abnormal turbulence characteristics; According to the multiple target spectra, the high-frequency band pressure data is divided to obtain multi-spectrum pressure data; Based on the historical high-frequency pressure data, feature extraction is performed on each spectrum pressure data to obtain a multi-spectrum pressure feature set; According to the multi-spectrum pressure feature set, each spectrum pressure is predicted respectively using a preset first pressure prediction model to obtain a first pressure change trend.

5. The combustion pulsation pressure trend monitoring method according to claim 3, characterized in that: For the non-high-frequency band pressure data, the pressure change trend is predicted according to a preset second pressure prediction model to obtain the second pressure change trend, including: Feature extraction is performed based on historical non-high frequency band pressure data to obtain a non-high frequency band pressure feature set; According to the non-high-frequency band pressure feature set, the pressure is predicted using a preset second pressure prediction model to obtain a second pressure change trend.

6. The combustion pulsation pressure trend monitoring method according to claim 1, characterized in that: The step of issuing an early warning for abnormal pressure conditions according to the multi-band pressure prediction trend includes: For the first pressure change trend in the high frequency band, a preset abnormal warning model is used to warn of pressure anomalies; For the second pressure change trend in the non-high frequency band, an early warning of abnormal pressure situation is issued according to the preset early alarm value and high alarm value.

7. The combustion pulsation pressure trend monitoring method according to claim 6, characterized in that: The first pressure change trend in the high frequency band is used to warn of the pressure anomaly using a preset abnormal warning model, including: According to the first pressure change trend, a change trend curve of each target spectrum pressure is drawn; According to the change trend curve, the slope and amplitude increment of the curve are calculated to obtain the trend characteristics of each target spectrum pressure; According to the preset high-frequency fault type and historical fault pressure data, the alarm threshold is updated using the preset fault spectrum model to obtain the updated alarm threshold for each spectrum pressure; Comparing the trend feature with the updated alarm threshold of the corresponding spectrum to obtain the abnormal pressure position and abnormal type; According to the abnormal pressure location and abnormal type, an early warning is issued through a preset abnormal early warning model.

8. The method for monitoring combustion pulsation pressure trend according to claim 1, characterized in that: The step of obtaining combustion pulsation pressure data in the combustion chamber comprises: Performing windowing processing on the combustion pulsation pressure data to obtain windowed pressure data; Resampling the windowed pressure data to obtain resampled data; The resampled data are combined to obtain updated combustion pulsation pressure data.

9. The combustion pulsation pressure trend monitoring method according to claim 8, characterized in that: The resampling process is performed on the windowed pressure data to obtain resampled data, including: Extracting the windowed pressure data to obtain extracted data; Interpolation processing is performed on the extracted data to obtain resampled data.

10. Combustion pulsation pressure warning system, characterized in that: A method for monitoring combustion pulsation pressure trend according to any one of claims 1 to 9, comprising: A data acquisition module, for acquiring combustion pulsation pressure data in the combustion chamber; A pressure change trend prediction module predicts the pressure change trends of different frequency bands in the combustion chamber according to the combustion pulsation pressure data to obtain a multi-band pressure prediction trend; The early warning module issues an early warning for abnormal pressure conditions according to the multi-band pressure prediction trend.

Images