Method and system for monitoring net heat value of fuel gas of combined cycle unit of gas turbine

Abstract

The invention discloses a gas turbine combined cycle unit fuel gas net heat value monitoring method and system, and belongs to the technical field of power system real-time monitoring. Aiming at the problem that the real-time performance and the high precision of fuel gas component detection cannot be considered in the prior art, the method comprises the following steps: acquiring detection data of a laser Raman spectrum on pipeline fuel gas components; performing anomaly detection on the detection data, and replacing the anomaly data with the detection data of the gas chromatograph in a timestamp alignment mode to obtain calibration data; using the calibration data as a training set to fit and train a Raman spectrum detection calibration model; fuel gas components are detected based on the trained model, and the net heat value is calculated. The system comprises an acquisition module, a calibration module, a training module and a net heat value module. By fusing the second-level real-time detection advantage of the laser Raman spectrum and the high-precision characteristic of the gas chromatograph and through dynamic calibration and model training, efficient and accurate monitoring of the net heat value of the fuel gas is achieved, technical support is provided for dynamic regulation and control of the combustion state of the gas turbine, and efficient and low-carbon operation of a unit is promoted.

Description

Technical Field

[0001] This invention relates to the field of real-time monitoring technology for power systems, specifically to a method and system for monitoring the net calorific value of fuel gas in a gas turbine combined cycle unit. Background Technology

[0002] Currently, the fuel gas for combined cycle gas turbine units in my country mainly includes pipeline natural gas from the West-East Gas Pipeline, by-product coal gas from the steel and chemical industries, and LNG from overseas. Future development will expand to include low-carbon / zero-carbon fuel gases such as hydrogen and ammonia. Due to the complex and diverse gas sources, the composition of the fuel gas used in actual unit operation differs significantly from the design parameters. Changes in fuel gas composition directly affect the calorific value of the fuel gas, posing a significant challenge to controlling the combustion stability of the gas turbine. Against this backdrop, gas-fired power plants typically install online fuel gas composition monitoring systems at the plant's pressure regulating station to monitor the calorific value of the fuel gas and more effectively monitor the combustion status of the combined cycle gas turbine unit and make timely combustion adjustments.

[0003] Traditional fuel gas component detection systems primarily rely on gas chromatography (GC). While offering high accuracy (±0.1% calorific value error) and broad component coverage, their long analysis cycles (5-15 minutes) make them unsuitable for real-time combustion state control in combined cycle gas turbine units. Besides GC, some fuel gas component detection systems also rely on laser Raman spectroscopy. Although capable of second-level detection and requiring no consumables, their accuracy for low-concentration components and complex hydrocarbons is insufficient (±1-2% calorific value error), failing to meet the precise combustion state control requirements of combined cycle gas turbine units. In summary, existing technologies cannot simultaneously achieve both real-time performance and high accuracy in fuel gas component detection. This dual deficiency means that current systems cannot provide timely warnings of unstable combustion conditions in gas turbines, nor can they offer precise component feedback for dynamic combustion adjustments, severely hindering the efficient and low-carbon operation of combined cycle gas turbine units. Summary of the Invention

[0004] To address the need for real-time and high-precision fuel gas component detection in the dynamic control of combustion state of gas turbines in combined cycle units, this invention proposes a method for monitoring the net calorific value of fuel gas in combined cycle gas turbine units by integrating multiple detection technologies. Through multimodal data collaboration and dynamic calibration, it provides technical support for the dynamic control of combustion state of gas turbines in combined cycle units.

[0005] This invention is achieved through the following technical solution: A method for monitoring the net calorific value of fuel gas in a gas turbine combined cycle unit includes the following steps: Step 1: Obtain detection data of pipeline fuel gas group using laser Raman spectroscopy; Step 2: Perform anomaly detection on the laser Raman spectroscopy detection data, obtain the anomaly detection data, and use the gas chromatograph detection data to replace the anomaly detection data by timestamp alignment to obtain calibrated detection data; Step 3: Use the calibrated detection data as a training dataset to fit and train the Raman spectroscopy detection calibration model; Step 4: Detect the fuel gas group according to the trained Raman spectroscopy detection calibration model, and calculate the net calorific value of the fuel gas based on the detection data.

[0006] Preferably, the detection data includes the mole fraction of each component in the fuel gas; The fuel gas comprises methane (CH4), ethane (C2H6), ethylene (C2H4), propane (C3H8), propylene (C3H6), and isobutane (Iso-C4H). 10 n-Butane N-C4H 10 Butene (C4H8), isopentane (Iso-C5H) 12 n-Pentane N-C5H 12 Pentene C5H 10 Hexane C6H 14 Nitrogen (N2), carbon monoxide (CO), carbon dioxide (CO2), water (H2O), hydrogen sulfide (H2S), hydrogen (H2), helium (He), oxygen (O2) and / or argon (Ar).

[0007] Preferably, the step of performing anomaly detection on the laser Raman spectroscopy detection data and obtaining anomaly detection data includes: Adopting improvements The criterion algorithm performs anomaly detection on the detection data of laser Raman spectroscopy and obtains anomaly detection data.

[0008] Preferably, the improved The criterion algorithm performs anomaly detection on laser Raman spectroscopy detection data, including: Obtain key components from the detection data and determine the change in mole fraction of key components between two adjacent time points; The average change in mole fraction at adjacent moments within the sliding window period is determined based on the change in mole fraction. The root mean square error of the change in signal data at adjacent time points within the window period is determined based on the mean of the change. The root mean square error is used to determine whether the detection data of key components are abnormal.

[0009] Preferably, the acquisition of key components in the detection data includes: The components in the test data are sorted according to their mole fractions, and the first few components in the sequence are considered the most critical components. If any critical component data is abnormal, the test data is considered abnormal.

[0010] Preferably, the method of using timestamp alignment to replace abnormal detection data with gas chromatograph detection data to obtain calibrated detection data includes: When abnormal detection data is detected, gas chromatography is used to detect the pipeline fuel gas simultaneously, and the abnormal detection data is replaced with the gas chromatography data according to the timestamp.

[0011] Preferably, the functional expression of the Raman spectroscopy detection calibration model is as follows:

[0012] In the formula, Fuel gas components for Raman spectroscopy detection The actual value of the mole fraction in the training set, Fuel gas components for gas chromatography detection The actual value of the mole fraction in the training set.

[0013] Preferably, the step of detecting the fuel gas batch based on the trained Raman spectroscopy detection calibration model and calculating the net calorific value of the fuel gas based on the detection data includes: Based on the molar mass of each component of the fuel gas Calculate the average relative molecular mass of the fuel gas. and the mass fraction of each component of the fuel gas Based on the theoretical calorific value of each component of the fuel gas Weighted calculation of the theoretical calorific value components of the fuel gas and the total theoretical calorific value of fuel gas Combined with the sensible heat of fuel gas Calculate the real-time net calorific value of fuel gas .

[0014] A monitoring system for the net calorific value of fuel gas in a gas turbine combined cycle unit includes: The acquisition module is used to acquire detection data of pipeline fuel gas groups using laser Raman spectroscopy. The calibration module is used to detect anomalies in the detection data of laser Raman spectroscopy, acquire the abnormal detection data, and replace the abnormal detection data with the detection data of gas chromatograph by timestamp alignment to obtain calibrated detection data; The training module is used to fit and train the Raman spectroscopy detection calibration model using the calibrated detection data as a training dataset. The net calorific value module is used to test the fuel gas batch based on the trained Raman spectroscopy detection calibration model and calculate the net calorific value of the fuel gas based on the detection data.

[0015] An electronic device, comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement the steps of the method for monitoring the net calorific value of fuel gas in the gas turbine combined cycle unit.

[0016] Compared with the prior art, the present invention has the following beneficial technical effects: This invention provides a method for monitoring the net calorific value of fuel gas in a combined cycle gas turbine unit. First, a real-time laser Raman spectroscopy detection stage is established to provide rapid initial data. Second, an anomaly detection stage is established to perform real-time anomaly analysis on the Raman spectroscopy detection data. Then, a gas chromatography precise verification stage is constructed to dynamically calibrate the Raman spectroscopy analysis model while correcting Raman spectroscopy detection deviations in real time. Finally, a fuel gas calorific value calculation stage is constructed to determine the net calorific value of the fuel gas in the combined cycle gas turbine unit. This invention provides effective technical support for the dynamic control of combustion state in combined cycle gas turbine units through multi-modal data collaboration and dynamic calibration.

[0017] This application also proposes a monitoring system for the net calorific value of fuel gas in a gas turbine combined cycle unit, an electronic device, and a computer storage medium, which possess all the advantages of the aforementioned monitoring methods for the net calorific value of fuel gas in gas turbine combined cycle units. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of the method for monitoring the net calorific value of fuel gas in a combined cycle gas turbine unit according to the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0021] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] A method for monitoring the net calorific value of fuel gas in a gas turbine combined cycle unit includes the following steps: Step 1: Obtain detection data of pipeline fuel gas group using laser Raman spectroscopy; Step 2: Perform anomaly detection on the laser Raman spectroscopy detection data, obtain the anomaly detection data, and use the gas chromatograph detection data to replace the anomaly detection data by timestamp alignment to obtain calibrated detection data; Step 3: Use the calibrated detection data as a training dataset to fit and train the Raman spectroscopy detection calibration model; Step 4: Detect the fuel gas group according to the trained Raman spectroscopy detection calibration model, and calculate the net calorific value of the fuel gas based on the detection data.

[0023] The method for monitoring the net calorific value of fuel gas in this gas turbine combined cycle unit integrates laser Raman spectroscopy and gas chromatography to construct a complete technical chain of "real-time detection - anomaly calibration - model training - calorific value calculation", which effectively solves the core problem of the difficulty in balancing real-time performance and high accuracy in existing technologies. Its technical solution has significant advantages: First, it uses laser Raman spectroscopy to acquire detection data. Leveraging the technology's second-level response and consumable-free characteristics, it overcomes the limitations of traditional gas chromatography's 5-15 minute analysis cycle, meeting the real-time data timeliness requirements for gas turbine combustion state control and laying the foundation for rapid response to changes in fuel gas composition. Second, it introduces an anomaly detection mechanism and combines it with gas chromatography data for calibration. Through timestamp alignment, it replaces abnormal data from laser Raman spectroscopy with high-precision detection data from the gas chromatograph (±0.1% calorific value error). This retains the real-time advantage of laser Raman spectroscopy while compensating for its insufficient accuracy (±1~2% calorific value error) in analyzing low-concentration components and complex hydrocarbons, ensuring the reliability of the detection data. Third, it trains the Raman spectroscopy detection calibration model with the calibrated detection data, achieving dynamic optimization of the model. With data accumulation, the accuracy and stability of laser Raman spectroscopy detection are continuously improved, enabling the system to continuously adapt to complex and diverse gas sources (such as pipeline natural gas, by-product coal gas, and LNG). The actual operating conditions of the gas turbine (and future low-carbon fuels) are considered. Finally, the net calorific value is calculated based on the test data of the trained model. The accuracy and real-time performance of the calibrated data are integrated, providing accurate calorific value feedback for the dynamic control of the combustion state of the gas turbine. It can not only provide timely warning of unstable combustion states, but also support dynamic combustion adjustment, effectively promoting the efficient and low-carbon operation of the gas turbine combined cycle unit, which is significantly better than the monitoring effect of single technology.

[0024] In some embodiments, anomaly detection is performed on the laser Raman spectroscopy detection data to obtain anomaly detection data. This anomaly detection data is then timestamped and replaced with gas chromatograph detection data to obtain calibrated detection data, including: The improved The criterion algorithm performs anomaly detection on laser Raman spectroscopy detection data, including: Obtain key components from the detection data and determine the change in mole fraction of key components between two adjacent time points; The average change in mole fraction at adjacent moments within the sliding window period is determined based on the change in mole fraction. The root mean square error of the change in signal data at adjacent time points within the window period is determined based on the mean of the change. The root mean square error is used to determine whether the detection data of key components are abnormal.

[0025] The components in the test data are sorted according to their mole fractions, and the first few components in the sequence are considered the most critical components. If any critical component data is abnormal, the test data is considered abnormal.

[0026] When abnormal detection data is detected, gas chromatography is used to detect the pipeline fuel gas simultaneously, and the abnormal detection data is replaced with the gas chromatography data according to the timestamp.

[0027] It should be noted that, due to the high detection efficiency of laser Raman spectroscopy, after abnormal detection data is detected, an online gas chromatograph is immediately used for detection. During the data replacement process, the first data from the gas chromatograph is spliced ​​with the first abnormal detection data to form the calibrated detection data.

[0028] In another embodiment, historical detection data from laser Raman spectroscopy and gas chromatography over the same time period are acquired; The detection data of the laser Raman spectroscopy was compared with the detection data of the gas chromatograph as a standard to identify abnormal detection data in the laser Raman spectroscopy detection data. The abnormal detection data was then replaced with the detection data of the gas chromatograph to obtain corrected detection data.

[0029] The replacement data needs to be consistent with the timestamp of the anomaly detection data.

[0030] like Figure 1 As shown, a monitoring method for real-time determination of the net calorific value of fuel gas in a gas turbine combined cycle unit includes the following steps: Step 1: The plant's pressure regulating station is equipped with a laser Raman spectrometer and its supporting equipment to establish a real-time laser Raman spectroscopy detection system. This system enables online detection of the pipeline fuel gas and real-time acquisition of the mole fraction of each component in the fuel gas. It achieves sub-second response times and provides rapid initial data detection. Details are as follows:

[0031] In the formula, For rapid detection of the mole fractions of each component in fuel gas using a laser Raman spectrometer. , for fuel gas Components, For each set of data, the current timestamp is... This refers to the fuel gas components detected by Raman spectroscopy. At the current timestamp The mole fraction.

[0032] Considering the complexity and diversity of fuel gas in combined cycle gas turbine units, and referring to relevant standards and specifications for fuel gas component testing, the main fuel gas components tested include: methane (CH4), ethane (C2H6), ethylene (C2H4), propane (C3H8), propylene (C3H6), and isobutane (Iso-C4H). 10 n-Butane N-C4H 10 Butene (C4H8), isopentane (Iso-C5H) 12 n-Pentane N-C5H 12 Pentene C5H 10 Hexane C6H 14 Nitrogen (N2), carbon monoxide (CO), carbon dioxide (CO2), water (H2O), hydrogen sulfide (H2S), hydrogen (H2), helium (He), oxygen (O2), argon (Ar), etc. Components.

[0033] Step 2, utilizing improvements The algorithm establishes a data signal anomaly detection step, performing real-time anomaly analysis and judgment on Raman spectroscopy detection data. If the data signal is normal, it is directly transmitted to the calorific value calculation step; if the data signal is abnormal, gas chromatography verification is immediately triggered. Details are as follows: With a certain component of fuel gas For example, suppose At the current stable operating time, the known components At any moment The mole fraction is , Then at time Compared to the previous moment change in mole fraction ,for:

[0034] The average change in mole fraction between adjacent time points within the anomaly detection sliding window period is determined based on the change in mole fraction. ,for:

[0035] In the formula, The sliding window period for detecting mole fraction anomalies. It can adaptively adjust the window based on historical data.

[0036] The root mean square error of the signal data change at adjacent time points within the window period is determined based on the mean of the change. ,for:

[0037] Determine the absolute value of the change in each signal data within the window period. and Size, if Then the composition within the window period can be confirmed. mole fraction If it is a normal value, then it is an abnormal value.

[0038] Similarly, a sliding detection window enables dynamic detection of abnormal mole fractions.

[0039] Similarly, anomaly analysis of the mole fraction of each component in the fuel gas is performed to detect in real time whether there are any abnormalities in the Raman spectroscopy detection. If the data signal is abnormal, a gas chromatographic verification is triggered.

[0040] Step 3: The plant's pressure regulating station is equipped with an online gas chromatograph and its supporting equipment to establish a precise gas chromatographic verification process. Once the gas chromatographic verification is triggered, the pipeline fuel gas is monitored in real time, high-precision component analysis is performed, and the mole fraction of each component in the fuel gas is output. It is used for two purposes: firstly, to dynamically calibrate the Raman spectroscopy analysis model, and secondly, to correct Raman spectroscopy detection biases in real time. Specifically:

[0041] In the formula, This refers to the mole fractions of each component in fuel gas that are detected in real time by an online gas chromatograph. , for fuel gas Components, For each set of data, the current timestamp is... This refers to the fuel gas components detected by gas chromatography. At the current timestamp The mole fraction.

[0042] First, the mole fraction of each component of the fuel gas was determined by gas chromatography. The timestamp will be used to detect the mole fraction of each component of the fuel gas using Raman spectroscopy. Preprocessing is performed to retain timestamp-aligned historical data, which is then used to form the calibration model fitting dataset. This dataset is then divided into two parts, with 80% serving as the training set for the calibration model fitting, and the timestamps being... 20% was used as the test set for calibrating the model fit, with timestamps of 1000 and 1000. .

[0043] Secondly, regarding the components of the fuel gas Using the training set data, Raman spectroscopy detection calibration models were fitted separately, while the mean square error was comprehensively calculated on the test set data. Coefficient of determination The model calibration effect was analyzed using indicators such as [list of indicators].

[0044] Constructor function ,make

[0045] In the formula, Fuel gas components for Raman spectroscopy detection In the training set The actual value of the mole fraction. Fuel gas components for gas chromatography detection In the training set The actual value of the mole fraction. .

[0046] At this time, the function This refers to the use of gas chromatography to detect fuel gas components using Raman spectroscopy. The calibration model for mole fraction can be in the form of linear regression, polynomial, exponential function, logarithmic function, power function, etc.

[0047] Meanwhile, in the test set, there are

[0048] In the formula, Fuel gas components for Raman spectroscopy detection In the test set The actual value of the mole fraction. Fuel gas components for Raman spectroscopy detection In the test set The mole fraction calibration value, .

[0049] Calculate the mean square error :

[0050] In the formula, Fuel gas components for gas chromatography detection In time The actual value of the mole fraction. Fuel gas components for Raman spectroscopy detection In time The mole fraction calibration value.

[0051] Mean square error The smaller the value, the better the calibration model will perform.

[0052] Calculate the coefficient of determination :

[0053] In the formula, Fuel gas components for Raman spectroscopy detection The mean of the actual values ​​of the mole fraction in the test set.

[0054] Coefficient of determination The closer the value is to 1, the better the calibration model performs.

[0055] By using a well-fitted Raman spectroscopy detection calibration model, the Raman spectroscopy detection deviation can be corrected in real time. At the same time, the Raman spectroscopy detection calibration model is dynamically updated as historical datasets are accumulated.

[0056] Step 4: Detect the fuel gas batch according to the trained Raman spectroscopy detection calibration model, and calculate the net calorific value of the fuel gas based on the detection data.

[0057] Mole fractions of fuel gas components based on fusion Raman spectroscopy detection and gas chromatography correction This involves constructing a step for calculating the calorific value of fuel gas. This is done by considering the molar mass of each component of the fuel gas. Calculate the average relative molecular mass of the fuel gas. and the mass fraction of each component of the fuel gas Combining the theoretical calorific value of each component of the fuel gas... Weighted calculation of the theoretical calorific value components of the fuel gas and the total theoretical calorific value of fuel gas Furthermore, the sensible heat of the fuel gas should be considered. Add them together to calculate the real-time net calorific value of the fuel gas. This information is then fed back to the gas turbine control system, thereby supporting the dynamic regulation of the combustion state of the gas turbine in the combined cycle unit. Specifically: First, the mole fractions of each component of the fuel gas were determined based on a combination of Raman spectroscopy detection and gas chromatography correction. ,have:

[0058] Secondly, considering the molar mass of each component of the fuel gas Calculate the average relative molecular mass of the fuel gas. and the mass fraction of each component of the fuel gas ,Right now:

[0059]

[0060] Secondly, considering the theoretical calorific value of each component of the fuel gas... Weighted calculation of the theoretical calorific value components of the fuel gas and the total theoretical calorific value of fuel gas ,Right now:

[0061]

[0062] Furthermore, considering the sensible heat of the fuel gas The real-time net calorific value of the fuel gas is calculated by summing the results. ,Right now:

[0063] Finally, the above calculation results are fed back to the gas turbine control system, thereby providing support for the dynamic regulation of the combustion state of the gas turbine in the combined cycle unit.

[0064] The above-described specific embodiments demonstrate that the monitoring method for determining the net calorific value of fuel gas in a combined cycle gas turbine unit according to the present invention, through multimodal data collaboration and dynamic calibration, can provide effective technical support for the dynamic control of the combustion state of the gas turbine in a combined cycle unit.

[0065] Example 2 A monitoring system for the net calorific value of fuel gas in a gas turbine combined cycle unit includes: The acquisition module is used to acquire detection data of pipeline fuel gas groups using laser Raman spectroscopy. The calibration module is used to detect anomalies in the detection data of laser Raman spectroscopy, acquire the abnormal detection data, and replace the abnormal detection data with the detection data of gas chromatograph by timestamp alignment to obtain calibrated detection data; The training module is used to fit and train the Raman spectroscopy detection calibration model using the calibrated detection data as a training dataset. The net calorific value module is used to test the fuel gas batch based on the trained Raman spectroscopy detection calibration model and calculate the net calorific value of the fuel gas based on the detection data.

[0066] It should be noted that, in the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another device, or some features may be ignored or not executed. The modules described as separate components may or may not be physically separated. The components shown as modules may be one or more physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs.

[0067] Furthermore, in the various embodiments of the present invention, the modules can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0068] An electronic device provided in this application includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the method for monitoring the net calorific value of fuel gas in a gas turbine combined cycle unit as described in any of the above embodiments.

[0069] Another electronic device provided in this application embodiment may further include: an input port connected to a processor for transmitting multimodal data collected by an external acquisition device to the processor; a display unit connected to the processor for displaying the processor's processing results to the outside world; and a communication module connected to the processor for enabling communication between the electronic device and the outside world. The display unit may be a display panel, a laser scanning display, etc.; the communication method adopted by the communication module includes, but is not limited to, Mobile High Definition Link (HML), Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), and wireless connection (including Wi-Fi, Bluetooth, Bluetooth Low Energy, and IEEE 802.11s-based communication technology).

[0070] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the method for monitoring the net calorific value of fuel gas in a gas turbine combined cycle unit as described in any of the above embodiments.

[0071] For descriptions of relevant parts of the monitoring system, electronic equipment, and computer-readable storage medium for the net calorific value of fuel gas in a gas turbine combined cycle unit provided in this application, please refer to the detailed description of the corresponding parts in the monitoring method for the net calorific value of fuel gas in a gas turbine combined cycle unit provided in this application, which will not be repeated here. Furthermore, parts of the technical solutions provided in this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0072] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method of monitoring the net heating value of a fuel gas for a gas turbine combined cycle unit, characterized by, The method comprises the following steps: Step 1, obtaining detection data of pipeline fuel gas group by laser Raman spectroscopy; Step 2, performing abnormality detection on the detection data of laser Raman spectroscopy to obtain abnormal detection data, using the detection data of a gas chromatograph to replace the abnormal detection data in a time stamp manner to obtain calibrated detection data; Step 3, using the calibrated detection data as a training data set to perform fitting training on a Raman spectroscopy detection calibration model; Step 4, detecting the fuel gas group according to the trained Raman spectroscopy detection calibration model, and calculating the net heat value of the fuel gas according to the detection data.

2. The method of claim 1, wherein the method further comprises: The detection data comprises the mole fraction of each component in the fuel gas; The components of the fuel gas include methane CH4, ethane C2H6, ethylene C2H4, propane C3H8, propylene C3H6, iso-butane Iso-C4H 10 , n-butane N-C4H 10 , butylene C4H8, iso-pentane Iso-C5H 12 , n-pentane N-C5H 12 , pentene C5H 10 , hexane C6H 14 , nitrogen N2, carbon monoxide CO, carbon dioxide CO2, water H2O, hydrogen sulfide H2S, hydrogen H2, helium He, oxygen O2 or / and argon Ar.

3. The method of claim 1, wherein the method further comprises: The abnormality detection on the detection data of laser Raman spectroscopy to obtain abnormal detection data comprises: Adopting improvements The criterion algorithm performs anomaly detection on the detection data of the laser Raman spectrum to obtain anomaly detection data.

4. A method of monitoring the net calorific value of the fuel gas of a gas turbine combined cycle unit according to claim 3, characterized in that, The improved The criterion algorithm performs anomaly detection on detection data of laser Raman spectrum, comprising: obtaining key components in the detection data, and determining the mole fraction change amount of the key components at adjacent two time points; determining the mean value of the mole fraction change amount at adjacent time points within a sliding window period according to the mole fraction change amount; determining the root mean square error of the signal data change amount at adjacent time points within the window period according to the mean value of the change amount; determining whether the detection data of the key components is abnormal according to the root mean square error.

5. A method of monitoring the net calorific value of the fuel gas of a gas turbine combined cycle unit according to claim 4, characterized in that, The obtaining of the key components in the detection data comprises: sorting each component in the detection data according to the mole fraction, and taking the first few components in the sequence as the key components, and when any key component data is abnormal, the detection data is abnormal.

6. The method of claim 1, wherein the method further comprises: The using of the detection data of the gas chromatograph to replace the abnormal detection data in a time stamp manner to obtain the calibrated detection data comprises: when the abnormal detection data is detected, synchronously detecting the pipeline fuel gas by the gas chromatograph, and replacing the abnormal detection data with the detection data of the gas chromatograph according to the time stamp.

7. The method of claim 1, wherein the method further comprises: The function expression of the Raman spectroscopy detection calibration model is as follows: wherein Fuel gas components for Raman spectroscopy detection At the actual value of the mole fraction of the training set, Fuel gas components for gas chromatography detection At the actual value of the mole fraction of the training set.

8. The method of claim 1, wherein the method further comprises: The detection of the fuel gas group according to the trained Raman spectroscopy detection calibration model, and the calculation of the net heat value of the fuel gas according to the detection data comprises: According to the molar mass of each component of the fuel gas , the average relative molecular mass of the fuel gas , and the mass fraction of each component of the fuel gas , combined with the theoretical combustion heat value of each component of the fuel gas , the theoretical heat value component of each component of the fuel gas is weighted and calculated , and the total theoretical heat value of the fuel gas , combined with the sensible heat of the fuel gas , the real-time net heat value of the fuel gas is calculated .

9. A system for monitoring the net heating value of a fuel gas for a gas turbine combined cycle unit, characterized by, comprises: a collection module for obtaining detection data of pipeline fuel gas group by laser Raman spectroscopy; a calibration module for performing abnormality detection on the detection data of laser Raman spectroscopy to obtain abnormal detection data, using the detection data of a gas chromatograph to replace the abnormal detection data in a time stamp manner to obtain calibrated detection data; a training module for using the calibrated detection data as a training data set to perform fitting training on a Raman spectroscopy detection calibration model; a net heat value module for detecting the fuel gas group according to the trained Raman spectroscopy detection calibration model, and calculating the net heat value of the fuel gas according to the detection data.

10. An electronic device, comprising: comprises: a memory for storing a computer program; a processor for executing the computer program to implement the steps of the method for monitoring the net heat value of the fuel gas turbine combined cycle unit according to any one of claims 1-8.

Images