A method and equipment for optimizing and improving the power generation capacity of gas turbines

Abstract

This invention discloses a method and device for optimizing and improving the power generation of a gas turbine. A three-stage air filtration system (primary bag filter, medium-efficiency V-filter, and high-efficiency fine filter) is connected to the gas turbine's intake duct. A cooling system and a dehumidification system are installed between the medium-efficiency V-filter and the high-efficiency fine filter. Using the pressure difference between each stage and the gas turbine inlet air temperature as input, and the gas turbine output power as output, training data drives an algorithm to obtain a gas turbine output power prediction model. A power generation relationship model is constructed using ambient temperature and humidity, temperature and humidity before and after entering and exiting the cooling and dehumidification system, chilled water system supply and return water temperature and flow rate, pressure difference before and after the primary bag filter, medium-efficiency V-filter, and high-efficiency fine filter, and power generation. An energy consumption relationship model is constructed using chilled water system supply and return water temperature and flow rate, and energy consumption. Multi-objective optimization is performed to maximize power generation, minimize the difference between power generation and predicted power, and minimize energy consumption to determine the optimal inlet temperature point of the gas turbine, thereby improving the gas turbine's power generation.

Description

Technical Field

[0001] This invention relates to the field of gas turbine power generation technology, and more specifically, to a method and device for optimizing and improving the power generation capacity of a gas turbine. Background Technology

[0002] Gas turbines play a vital role in ensuring power and heat supply. With the national dual-carbon goals being proposed, it is necessary to further improve the power generation efficiency of gas turbines and reduce carbon emissions. Existing technologies, such as ZL202011458763.2, disclose a system and method for improving the power output of a gas turbine generator set. This system includes a gas storage tank and an air compressor. The air compressor provides compressed air to the storage tank at high altitudes, ensuring the oxygen content in the compressed air is greater than or equal to 21%. The system monitors the pressure and temperature of the storage tank, calculates the ambient air pressure based on geographical information, and then sets the motor's speed range. The motor speed is then adjusted specifically based on the pressure and temperature of the storage tank. This method, which uses a storage tank to ensure the compressed air meets standard atmospheric pressure, improves the generator set's output power and is mainly applied in areas above 1500m altitude. For example, ZL2021103013732 discloses a gas turbine intake air filtration system. In response to the problem that the high humidity and salt spray in coastal areas cause heavy operating loads on the air filtration system, liquid particles tend to frost on the filter element, causing blockage and filter failure, the system uses a droplet separation unit and a filtration unit to reduce liquid particles in the gas before it enters the air filtration system, thus ensuring the safe and stable operation of the gas turbine.

[0003] Since my country's population is mainly located in the central and eastern regions, and numerous gas turbine power plants have been built outside of high-altitude and coastal areas, long-standing problems exist such as reduced gas turbine power generation due to large pressure differentials and high intake air temperatures in the inlet filtration system. The methods mentioned above are ineffective in the central and eastern regions. The first technology mainly involves a gas storage tank that generates a standard atmospheric pressure, but the plains of the central and eastern regions are at standard atmospheric pressure, making this device unnecessary. The second technology aims to reduce the entry of liquid particles into the filtration system, but it first incorporates a droplet generating unit to produce liquid particles of varying sizes before removal, which is also ineffective in the central and eastern regions.

[0004] Therefore, this application proposes a new solution for increasing the power generation capacity of gas turbines in the central and eastern regions. Summary of the Invention

[0005] To address the aforementioned problems, the purpose of this invention is to propose a method and apparatus for optimizing and enhancing the power generation capacity of a gas turbine. This method increases the power generation capacity of the gas turbine by controlling its inlet temperature. The apparatus not only reduces particulate matter entering the gas turbine inlet but also lowers the air temperature and humidity at the inlet during summer.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a method for optimizing and increasing the power generation capacity of a gas turbine. A three-stage air filtration system consisting of a primary bag filter, a medium-efficiency V-filter, and a high-efficiency fine filter is connected to the gas turbine's intake duct. A cooling system and a dehumidification system are installed between the medium-efficiency V-filter and the high-efficiency fine filter. The air temperature is first cooled, and then the air is dehumidified, with energy provided by a chilled water system. The optimization method includes the following:

[0008] Real-time monitoring data are obtained for the pressure difference at each stage after primary, medium and high efficiency treatment, ambient temperature and humidity, temperature and humidity before and after entering and exiting the cooling and dehumidification system, chilled water supply and return water temperature and flow rate, and gas turbine inlet temperature. Real-time monitoring data of gas turbine output power are also obtained.

[0009] A gas turbine output power prediction model is constructed. The algorithm is driven by training data with the pressure difference at each stage and the gas turbine inlet air temperature as inputs and the gas turbine output power as output, so as to obtain the gas turbine output power prediction model.

[0010] Using ambient temperature and humidity, temperature and humidity before and after entering and exiting the cooling and dehumidification system, chilled water supply and return water temperature and flow rate, and pressure difference before and after the primary bag filter, medium-efficiency V filter and high-efficiency fine filter as independent variables, and power generation as dependent variable, a power generation relationship model is constructed.

[0011] Using the supply and return water temperatures and flow rates of the chilled water system as inputs, the energy consumption under different power generation capacities is calculated to form an energy consumption relationship model.

[0012] Based on the power generation relationship model and energy consumption relationship model, a multi-objective optimization is performed to maximize power generation, minimize the difference between power generation and predicted power, and minimize energy consumption. The optimal temperature point at the gas turbine inlet is determined by the Pareto front, and the temperature and flow rate of the chilled water system at this time are determined. The gas turbine power generation is controlled by this temperature and flow rate.

[0013] Furthermore, the K-means clustering algorithm is used to identify abnormal data, including missing and mutated data, for the acquired real-time monitoring data; and a machine learning algorithm is used to repair the abnormal data to obtain preprocessed data for subsequent model building; the machine learning algorithm is a BP neural network or K nearest neighbor classification.

[0014] Furthermore, the data-driven algorithms include artificial neural networks (ANN), support vector regression (SVR), random forests, or extreme gradient boosting algorithms, etc.

[0015] Secondly, the present invention provides a device for optimizing and enhancing the power generation of a gas turbine, wherein the device performs the method described herein.

[0016] Thirdly, the present invention provides a device for optimizing and enhancing the power generation capacity of a gas turbine, the device comprising:

[0017] The air intake filtration system is equipped with a three-stage filtration system consisting of a pre-filter bag filter, a medium-efficiency V-filter, and a high-efficiency fine filter.

[0018] The cooling and dehumidification system consists of a cooling system and a dehumidification system, located between the medium-efficiency V filter and the high-efficiency fine filter. It first cools the air temperature and then dehumidifies the air. The cooling system of the cooling and dehumidification system is powered by a chilled water system.

[0019] Pressure sensors are used to collect the pressure difference before and after the primary bag filter, medium-efficiency V filter, and high-efficiency fine filter;

[0020] Temperature sensors are used to monitor the supply and return water temperatures of chilled water systems;

[0021] Flow sensors monitor the flow rate of chilled water systems;

[0022] Temperature and humidity sensor to collect ambient temperature and humidity, as well as temperature and humidity before and after entering and exiting the cooling and dehumidification system;

[0023] The intelligent control system for intake air filtration acquires data from the pressure sensor, temperature sensor, flow sensor, and temperature and humidity sensor mentioned above, and monitors the pressure difference of the three-stage filtration, the temperature and flow of chilled water supply and return water, the ambient temperature and humidity, and the temperature and humidity before and after entering the cooling and dehumidification system, and communicates with the gas turbine DCS system in real time.

[0024] Finally, the clean, dry, and low-temperature air, after being treated by high-efficiency fine filtration, enters the gas turbine combustion chamber through the gas turbine inlet under the action of the gas turbine air compressor.

[0025] Furthermore, the intelligent air intake filtration control system includes:

[0026] Local storage units and cloud platforms are used to store normal data during device operation;

[0027] The analysis unit obtains the total output power and total heat consumption of the gas turbine under factory conditions, and records them as the standard reference value for output power. and heat consumption standard reference value The total output power of the gas turbine under actual operating conditions is... It is obtained from the following formula:

[0028]

[0029] In the formula: —Correction factor for output power based on ambient temperature;

[0030] —The correction factor for the output power due to pressure difference;

[0031] —Correction factor for relative humidity on output power;

[0032] —Correction factor for the low calorific value of fuel on output power;

[0033] —The correction factor for the output power based on the exhaust pressure;

[0034] Under actual operating conditions, the total heat consumption of the gas turbine is calculated using the following formula:

[0035]

[0036] In the formula: —Correction factor for heat loss due to ambient temperature;

[0037] —The correction factor for heat loss due to atmospheric pressure;

[0038] —The correction factor for heat loss due to relative humidity;

[0039] —Correction factor for low calorific value of fuel on heat consumption;

[0040] —The correction factor for heat loss due to exhaust pressure;

[0041] Based on the ambient temperature, the pressure difference of the high-efficiency fine filter, and the humidity before and after entering and exiting the cooling and dehumidification system, establish the relationship curves between output power and heat consumption, and obtain the correction coefficients for output power and heat consumption under their respective conditions.

[0042] Optimize the gas turbine inlet temperature to maximize output power and minimize heat consumption;

[0043] The display unit is used to display the relationship curves between various indicators and output power, the relationship curves between indicators and heat consumption, and the optimization results.

[0044] Furthermore, the intelligent air intake filtration control system is equipped with a large-scale language model based on Transformers. This large-scale language model takes as input the pressure difference of each stage after primary, medium, and high-efficiency treatment, the ambient temperature and humidity, the temperature and humidity before and after entering and exiting the cooling and dehumidification system, the supply and return water temperature and flow rate of the chilled water system, and the gas turbine inlet temperature. It takes as output power and heat consumption as output. The output of the large-scale language model based on Transformers is compared with the set normal range of output power and heat consumption to determine whether there is a fault in the gas turbine, which is used for fault diagnosis of the gas turbine intake system.

[0045] Furthermore, the intake filtration system, while ensuring intake pressure differential, optimizes the structure and effective filtration area of ​​each filter element, setting up three filter elements in sequence: a pre-filter bag filter, a medium-efficiency V filter, and a high-efficiency fine filter. The pre-filter bag filter adopts a bag structure with at least 10 bags, and the filter material is M6 grade high-efficiency glass fiber. The medium-efficiency V filter adopts a 5V structure, and the filter material is F8 grade high-efficiency glass fiber. The high-efficiency fine filter adopts a cylindrical structure, and the diameter and length are adjusted according to the intake space. The filter material is F9 grade fully synthetic composite fiber.

[0046] Furthermore, the cooling system adopts a layered and grouped configuration, grouped by top and bottom and layered by inside and outside. The cooling coils within a group are connected in series, and the cooling coils between different layers are independent of each other. Each cooling coil includes a heat dissipation channel, and two heat dissipation fins 2 are provided in each heat dissipation channel. The two ends of the heat dissipation pipes of each layer are connected to the chilled water system through a chilled water supply pipe 3 and a chilled water return pipe 4, respectively.

[0047] At least four separate cooling coils, each with at least three layers, with the angle between the two heat dissipation fins being 60°~120°, and the heat dissipation fins being at 30°~60° to the airflow direction. The two heat dissipation fins need to overlap 1-2cm horizontally. The thickness of each cooling coil is 10-30cm. The heat dissipation pipes are made of copper or stainless steel, and the heat dissipation fins are made of corrosion-resistant aluminum foil or stainless steel sheets.

[0048] The dehumidification system uses a water mist filter, a solid adsorbent, or a lithium chloride rotary dehumidification system with a thickness of 5-10 cm and a cross-sectional area adjusted according to the air intake space.

[0049] Furthermore, in summer, outdoor air that is hot, humid, and dusty enters the air intake filtration system and passes through a pre-filter bag filter with 10 bags of M6 grade filter media, which removes about 60% of dust particles with a diameter of ≥0.4μm; then it passes through a medium-efficiency V filter with 5V structure F8 filter media, which removes about 90% of dust particles with a diameter of ≥0.4μm, resulting in cleaner air.

[0050] When the outdoor temperature exceeds 22°C, the air passes through a medium-efficiency V filter and enters the cooling system connected to the chilled water system. The cooling system consists of at least four sets of cooling coils. When the chilled water system is turned on, the chilled water enters the cooling coils in the cooling system to cool the high-temperature air to 22°C.

[0051] The cooled air enters the dehumidification system after being processed by the cooling system. The dehumidification system removes condensate and excess moisture from the air, turning it into cleaner, cooler, and drier air, which then enters the high-efficiency filter.

[0052] The high-efficiency fine filter uses a cylindrical structure and 1m long F9 grade fully synthetic composite fiber to filter out about 95% of dust particles with a diameter ≥0.4μm, turning them into clean, low-temperature, dry air. This clean air then enters the gas turbine combustion chamber through the gas turbine inlet under the action of the air compressor.

[0053] Compared with the prior art, the present invention provides a method and device for optimizing and improving the power generation capacity of a gas turbine, which has at least the following advantages:

[0054] 1. The present invention is equipped with an intelligent control system for intake air filtration, which can intelligently regulate the relevant parameters of the equipment before the gas turbine intake, such as the three-stage filtration pressure difference and the temperature and flow rate of chilled water supply and return water, thereby improving the power generation of the gas turbine.

[0055] 2. The improvement method of this invention monitors parameters such as gas turbine power generation, inlet temperature, three-stage filter pressure difference, chilled water supply and return water temperatures, and temperature and humidity before and after entering and exiting the cooling and dehumidification system in real time. It performs multi-objective optimization to maximize power generation, minimize the difference between the generated power and the predicted power, and minimize energy consumption. The optimal gas turbine inlet temperature is determined using the Pareto front, and the temperature and flow rate of the chilled water system at this point are also determined. These temperature and flow rates are then used to control gas turbine power generation, maintaining a high power output. Simultaneously, it effectively ensures the intake air filtration and cooling / dehumidification effects.

[0056] 3. This invention regulates and controls the inlet air temperature of the gas turbine, solving the problem of reduced power generation caused by high inlet air temperature in summer. It achieves intelligent control with high precision and fast response.

[0057] 4. This invention features a three-stage filter structure, enhancing the filtration efficiency of each stage. The intermediate filter employs a special structure, such as a 5V structure, to increase the flow area and reduce pressure drop. This ensures the quality of the intake air filtration while extending the filter lifespan and reducing the cost of frequent filter replacements. The service life of the primary bag filter and the intermediate V filter exceeds 4000 hours or one year, while the service life of the high-efficiency fine filter is no less than 8000 hours. This solves the problems of short service life and easy filter clogging in existing single-stage or two-stage intake filtration systems. Under the conditions of limited intake space and a high flow rate of 108.5 m³ / s, this invention adds a high-efficiency fine filter, achieving increased filtration efficiency within a limited space and under high flow conditions, and reducing the frequency of filter replacement (e.g., existing technologies require filter replacement at least 3-5 times per year, while this application requires replacement approximately once a year). With the three-stage filtration system, the pressure difference can be reduced from 60 mmH₂O to 40-42 mmH₂O, a reduction of 33%.

[0058] 5. This invention can effectively control the gas turbine inlet temperature in summer and winter through an intelligent control system, accurately control the gas turbine inlet air temperature in winter and summer, ensure the gas turbine's high-efficiency power generation, and at the same time, the dehumidification system can effectively reduce the humidity of the inlet air, prevent humid air from entering the high-efficiency fine filter system, and avoid the risk of fine filter failure. This can increase the maximum power generation of the gas turbine by 2WM / h, an increase of 5.3%. Attached Figure Description

[0059] Figure 1 A schematic diagram of the overall structure of an embodiment of the power generation optimization and enhancement device for gas turbines according to the present invention;

[0060] Figure 2 A top view of the cooling system of an embodiment of the power generation optimization and enhancement device for gas turbines according to the present invention;

[0061] Figure 3 A cross-sectional view of the cooling system of one embodiment of the power generation optimization and enhancement device for gas turbines according to the present invention. Detailed Implementation

[0062] To more clearly illustrate the technical solution and advantages of the present invention, the working principle of the technical solution of the present invention in summer will be fully described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0063] Example 1

[0064] Figure 1 This is a schematic diagram of a gas turbine power generation optimization and enhancement method and device provided in an embodiment of this application. The device for optimizing and enhancing the power generation of a gas turbine includes:

[0065] The air intake filtration system is equipped with a three-stage system: a primary bag filter, a medium-efficiency V filter, and a high-efficiency fine filter.

[0066] The cooling and dehumidification system is located between the medium-efficiency V filter and the high-efficiency fine filter. It first cools the air temperature and then dehumidifies the air.

[0067] The intelligent control system for intake air filtration can monitor the pressure difference of the three-stage filtration system, monitor the ambient temperature and humidity and the temperature and humidity before and after entering the cooling and dehumidification system, monitor the supply and return temperature and flow of chilled water, and communicate with the gas turbine DCS system in real time.

[0068] Finally, the clean, dry, and low-temperature air, after being treated by high-efficiency fine filtration, enters the gas turbine combustion chamber through the gas turbine inlet under the action of the gas turbine air compressor.

[0069] Preferably, the intake filtration system includes: the more filter levels, the greater the pressure difference; under the premise of ensuring the intake pressure difference, by optimizing the structure and effective filtration area of ​​each filter level, a three-stage filter system is set, namely, a pre-filter bag filter, a medium-efficiency V filter, and a high-efficiency fine filter. The pre-filter bag filter adopts a bag structure with at least 10 bags, and the filter material is M6 grade high-efficiency glass fiber; the medium-efficiency V filter adopts a 5V structure, and the filter material is F8 grade high-efficiency glass fiber; the high-efficiency fine filter adopts a cylindrical structure, and the diameter and length are appropriately adjusted according to the intake space, and the filter material is F9 grade fully synthetic composite fiber.

[0070] Preferably, the cooling and dehumidification system includes a cooling system and a dehumidification system. The cooling system is connected to a chilled water system, and the chilled water system includes a chilled water supply pipe 3 and a chilled water return pipe 4, such as... Figure 2 As shown. The cooling system adopts a layered and grouped configuration, grouped by top and bottom, and layered by inside and outside. The cooling coils within a group are connected in series, and the cooling coils between different layers are independent of each other. Each cooling coil includes a heat dissipation channel, and two heat dissipation fins 2 are provided in each heat dissipation channel. The two ends of the heat dissipation pipes of each layer are connected to the chilled water system through a chilled water supply pipe 3 and a chilled water return pipe 4, respectively.

[0071] At least four separate cooling coils, each with at least three layers, with the angle between the two heat dissipation fins being 60°~120°, and the heat dissipation fins being at 30°~60° to the airflow direction. The two heat dissipation fins need to overlap 1-2cm horizontally. The thickness of each cooling coil (i.e., the width in the top view) is 10-30cm. The heat dissipation pipes are made of copper or stainless steel, and the heat dissipation fins are made of corrosion-resistant aluminum foil or stainless steel sheets.

[0072] Existing cooling systems are all single units with a single inlet and single outlet, resulting in long cooling coils and the longest circulation loop exceeding 20m, leading to high energy consumption of the circulation pump. This embodiment adopts a grouped and layered structure, dividing the existing unit into four separate groups of cooling coils, each with at least three layers of copper tubing, increasing heat exchange efficiency and reducing the longest circulation loop by more than 50%.

[0073] The dehumidification system can use a water mist filter, a solid adsorbent such as silica gel, or a lithium chloride rotary dehumidification system with a thickness of 5~10cm and a cross-sectional area that is adjusted according to the air intake space.

[0074] Example 2

[0075] In this embodiment, the intelligent control system for intake air filtration includes:

[0076] Local storage units and cloud platforms are used to store normal data during device operation;

[0077] The analysis unit obtains the total output power and total heat consumption of the gas turbine under factory conditions, and records them as the standard reference value for output power. and heat consumption standard reference value (Unit: [kJ / kWh]), then the total output power of the gas turbine under actual operating conditions. It is obtained from the following formula:

[0078]

[0079] In the formula: —Correction factor for output power based on ambient temperature;

[0080] —The correction factor for the output power due to pressure difference;

[0081] —Correction factor for relative humidity on output power;

[0082] —Correction factor for the low calorific value of fuel on output power;

[0083] —The correction factor for the output power based on the exhaust pressure;

[0084] Under actual operating conditions, the total heat consumption of the gas turbine is calculated using the following formula:

[0085]

[0086] In the formula: —Correction factor for heat loss due to ambient temperature;

[0087] —The correction factor for heat loss due to atmospheric pressure;

[0088] —The correction factor for heat loss due to relative humidity;

[0089] —Correction factor for low calorific value of fuel on heat consumption;

[0090] —The correction factor for heat loss due to exhaust pressure;

[0091] Based on the ambient temperature, the pressure difference of the high-efficiency fine filter, and the humidity before and after entering and exiting the cooling and dehumidification system, establish the relationship curves between output power and heat consumption, and obtain the correction coefficients for output power and heat consumption under their respective conditions.

[0092] Optimize the gas turbine inlet temperature to maximize output power and minimize heat consumption;

[0093] The display unit is used to display the relationship curves between various indicators and output power, the relationship curves between indicators and heat consumption, and the optimization results.

[0094] The correction factors for the fuel's lower calorific value, atmospheric pressure, and exhaust pressure mentioned above can be set as constants based on experience.

[0095] Monitor the pressure difference before and after the primary bag filter, medium-efficiency V filter, and high-efficiency fine filter. When the pressure difference exceeds the limit, issue an early warning and replace the filter in time. Monitor the ambient temperature and humidity and the temperature and humidity before and after the cooling and dehumidification system to provide the control system with cooling system temperature and humidity monitoring data. Monitor the supply and return water temperature and flow rate of the chilled water system to provide the control system with chilled water system temperature and flow rate monitoring data.

[0096] The acquired real-time monitoring data is processed using the K-means clustering algorithm to identify anomalous data, including missing and mutated data. Machine learning algorithms are then used to repair the anomalous data, resulting in preprocessed data for subsequent model building. The machine learning algorithms used are either backpropagation neural networks or K-nearest neighbor classification. The preprocessed data is stored on both the local system and the cloud platform.

[0097] Output power and heat consumption are important indicators of gas turbine power generation. Through the intelligent control system of the intake system, the gas turbine inlet temperature is optimized to maximize output power and minimize heat consumption, so as to keep these two indicators at the optimal operating level.

[0098] The display unit acquires data from the analysis unit and displays parameters such as ambient temperature and humidity, pressure differential of the three-stage filtration system, supply and return water temperature and flow rate of the chilled water system, inlet and outlet temperature and humidity of the cooling and dehumidification system, and output power and heat consumption of the gas turbine, as well as related relationship curves.

[0099] The control system is connected to the gas turbine DCS system. The intelligent control system for intake air filtration incorporates a large-scale language model based on Transformers. This model takes as input the pressure differentials after primary, secondary, and high-efficiency treatments, ambient temperature and humidity, temperature and humidity before and after entering and exiting the cooling and dehumidification system, chilled water supply and return water temperatures and flow rates, and the gas turbine inlet temperature. It outputs power and heat consumption. The outputs of the Transformers-based language model are compared with the set normal ranges for power and heat consumption to determine if the gas turbine has a fault, thus performing fault diagnosis of the gas turbine intake system. The DCS system implements real-time closed-loop control of the gas turbine intake cooling system's intake temperature. The Transformers-based language model is a generative pre-trained model, demonstrating powerful capabilities in human interaction, code generation, and reasoning using common sense and domain knowledge.

[0100] Example 3

[0101] This embodiment describes a method for optimizing and increasing the power generation capacity of a gas turbine. A three-stage air filtration system—a primary bag filter, a medium-efficiency V-filter, and a high-efficiency fine filter—is connected to the gas turbine's intake duct. A cooling system and a dehumidification system are installed between the medium-efficiency V-filter and the high-efficiency fine filter. The air temperature is cooled first, and then dehumidified, with energy provided by a chilled water system. The optimization method includes the following:

[0102] Real-time monitoring data are obtained for the pressure difference at each stage after primary, medium and high efficiency treatment, ambient temperature and humidity, temperature and humidity before and after entering and exiting the cooling and dehumidification system, chilled water supply and return water temperature and flow rate, and gas turbine inlet temperature. Real-time monitoring data of gas turbine output power are also obtained.

[0103] A gas turbine output power prediction model is constructed. The algorithm is driven by training data with the pressure difference at each stage and the gas turbine inlet air temperature as inputs and the gas turbine output power as output, so as to obtain the gas turbine output power prediction model.

[0104] Using ambient temperature and humidity, temperature and humidity before and after entering and exiting the cooling and dehumidification system, chilled water supply and return water temperature and flow rate, and pressure difference before and after the primary bag filter, medium-efficiency V filter and high-efficiency fine filter as independent variables, and power generation as dependent variable, a power generation relationship model is constructed.

[0105] Using the supply and return water temperatures and flow rates of the chilled water system as inputs, the energy consumption under different power generation capacities is calculated to form an energy consumption relationship model.

[0106] Based on the power generation relationship model and energy consumption relationship model, a multi-objective optimization is performed to maximize power generation, minimize the difference between power generation and predicted power, and minimize energy consumption. The optimal temperature point at the gas turbine inlet is determined by the Pareto front, and the temperature and flow rate of the chilled water system at this time are determined. The gas turbine power generation is controlled by this temperature and flow rate.

[0107] In this example, the data acquisition time was 10 minutes. Each acquisition obtained real-time monitoring data on the pressure difference at each stage after primary, medium, and high-efficiency treatment, ambient temperature and humidity, temperature and humidity before and after entering and exiting the cooling and dehumidification system, chilled water supply and return water temperature and flow rate, and gas turbine inlet temperature. Real-time monitoring data on gas turbine output power was also obtained.

[0108] Based on historical monitoring data, a data-driven algorithm is used to train a gas turbine output power prediction model. In this embodiment, support vector regression is used for training. A set of sample data is formed by the pressure difference at each stage, the gas turbine inlet air temperature, and the gas turbine output power. A large amount of historical monitoring data constitutes the dataset. 80% of the data in the dataset is used for training, and 20% of the data is used for validation. This model can predict the output power value at future time or within a certain period of time.

[0109] The construction process of the gas turbine output power prediction model includes steps such as data preprocessing, feature engineering, model training, model evaluation, and result visualization. Feature engineering uses methods such as correlation analysis, recursive feature elimination, and dimensionality reduction to select appropriate features as model input. Appropriate data-driven algorithms and hyperparameters are selected for model training. Model evaluation employs algorithms such as artificial neural networks (ANN), support vector regression (SVR), random forests, and extreme gradient boosting.

[0110] Example 4

[0111] In this embodiment, during summer, outdoor air with high temperature, high humidity, and dust enters the air intake filtration system and passes through a pre-filter bag filter, which uses 10 bags of M6 grade filter material to filter out about 60% of the dust particles with a diameter ≥0.4μm; then it passes through a medium-efficiency V filter, which uses 5V structure F8 filter material to filter out about 90% of the dust particles with a diameter ≥0.4μm, resulting in cleaner air.

[0112] When the outdoor temperature exceeds 22℃, the relatively clean, high-temperature air passes through a medium-efficiency V filter and then enters the cooling system connected to the chilled water system. The cooling system consists of at least four sets of cooling coils, each using a three-row copper pipe and corrosion-resistant aluminum foil structure with a thickness of 10-30cm. The intelligent control system activates the chilled water system, and chilled water enters the cooling coils in the cooling system to cool the high-temperature air to 22℃.

[0113] After passing through the cooling system, the relatively clean, low-temperature air enters the dehumidification system. This system uses solid adsorbents such as silica gel or a lithium chloride rotary dehumidifier with a thickness of 5-10 cm. The dehumidification system removes condensate and excess moisture from the air, resulting in cleaner, lower-temperature, dry air. This prevents humid air from entering the high-efficiency fine filtration system and affecting its filtration efficiency.

[0114] After passing through the dehumidification system, the relatively clean, low-temperature, and dry air enters the high-efficiency fine filter, which uses a cylindrical structure and 1m long F9 grade fully synthetic composite fiber to filter out about 95% of dust particles with a diameter ≥0.4μm, turning it into clean, low-temperature, and dry air. This clean air then enters the gas turbine combustion chamber through the gas turbine inlet under the action of the air compressor.

[0115] According to the output power calculation formula in Example 2, this embodiment calculates that for every 100 Pa increase in the gas turbine inlet pressure difference, the gas turbine output power decreases by 62 kW / H, approximately 1.3%. The three-stage inlet filtration system proposed in this invention can effectively reduce the annual average pressure difference by more than 200 Pa, which can increase the annual average output power of the gas turbine by approximately 2.6%.

[0116] When the gas turbine inlet temperature exceeds 22°C in summer, the chilled water system is activated through the control system, and the opening of the solenoid valves or the supply and return water temperatures of the chilled water system are adjusted to lower the gas turbine inlet temperature to 22°C. When the gas turbine inlet temperature is below 10°C in winter, the chilled water system is activated through the control system, the chilled water system is switched to heating mode, and the opening of the solenoid valves or the supply and return water temperatures of the chilled water system are adjusted to raise the gas turbine inlet temperature to 10°C.

[0117] The DCS system enables real-time closed-loop control of the gas turbine intake cooling system intake temperature with a deviation of no more than 1.5%, and enables high-speed information exchange with the DCS with an interaction time of no more than 2 seconds.

[0118] The accompanying drawings of the embodiments disclosed in this invention only involve structures relevant to the embodiments disclosed in this invention. Other structures can be referred to with common designs. Unless otherwise specified, the same embodiment and different embodiments of this invention can be combined with each other.

[0119] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0120] Any aspects not covered in this invention are applicable to existing technologies.

Claims

1. A method for optimizing and improving the power generation capacity of a gas turbine, characterized in that, A three-stage air filtration system consisting of a primary bag filter, a medium-efficiency V-filter, and a high-efficiency fine filter is connected to the gas turbine's intake duct. A cooling system and a dehumidification system are installed between the medium-efficiency V-filter and the high-efficiency fine filter. The air temperature is cooled first, and then the air is dehumidified. Energy is provided by a chilled water system. The enhancement method includes the following: Real-time monitoring data are obtained for the pressure difference at each stage after primary, medium and high efficiency treatment, ambient temperature and humidity, temperature and humidity before and after entering and exiting the cooling and dehumidification system, chilled water supply and return water temperature and flow rate, and gas turbine inlet temperature. Real-time monitoring data of gas turbine output power are also obtained. A gas turbine output power prediction model is constructed. The algorithm is driven by training data with the pressure difference at each stage and the gas turbine inlet air temperature as inputs and the gas turbine output power as output, so as to obtain the gas turbine output power prediction model. Using ambient temperature and humidity, temperature and humidity before and after entering and exiting the cooling and dehumidification system, chilled water supply and return water temperature and flow rate, and pressure difference before and after the primary bag filter, medium-efficiency V filter and high-efficiency fine filter as independent variables, and power generation as dependent variable, a power generation relationship model is constructed. Using the supply and return water temperatures and flow rates of the chilled water system as inputs, the energy consumption under different power generation capacities is calculated to form an energy consumption relationship model. Based on the power generation relationship model and energy consumption relationship model, a multi-objective optimization is performed to maximize power generation, minimize the difference between power generation and predicted power, and minimize energy consumption. The optimal temperature point at the gas turbine inlet is determined by the Pareto front, and the temperature and flow rate of the chilled water system at this time are determined. The gas turbine power generation is controlled by this temperature and flow rate.

2. The method according to claim 1, characterized in that, The acquired real-time monitoring data is subjected to K-means clustering algorithm to identify abnormal data, including missing and mutated data; and machine learning algorithm is used to repair abnormal data to obtain preprocessed data for subsequent model building; the machine learning algorithm is BP neural network or K nearest neighbor classification.

3. The method according to claim 1, characterized in that, The data-driven algorithms include artificial neural networks, support vector regression, random forests, or extreme gradient boosting algorithms.

4. A device for optimizing and enhancing the power generation capacity of a gas turbine, characterized in that, The device performs the method according to any one of claims 1-3.

5. A device for optimizing and enhancing the power generation capacity of a gas turbine, characterized in that, The device performs the method according to any one of claims 1-3, the device comprising: The air intake filtration system is equipped with a three-stage filtration system consisting of a pre-filter bag filter, a medium-efficiency V-filter, and a high-efficiency fine filter. The cooling and dehumidification system consists of a cooling system and a dehumidification system, located between the medium-efficiency V filter and the high-efficiency fine filter. It first cools the air temperature and then dehumidifies the air. The cooling system of the cooling and dehumidification system is powered by a chilled water system. Pressure sensors are used to collect the pressure difference before and after the primary bag filter, medium-efficiency V filter, and high-efficiency fine filter; Temperature sensors are used to monitor the supply and return water temperatures of chilled water systems; Flow sensors monitor the flow rate of chilled water systems; Temperature and humidity sensor to collect ambient temperature and humidity, as well as temperature and humidity before and after entering and exiting the cooling and dehumidification system; The intelligent control system for intake air filtration acquires data from the pressure sensor, temperature sensor, flow sensor, and temperature and humidity sensor mentioned above, and monitors the pressure difference of the three-stage filtration, the temperature and flow of chilled water supply and return water, the ambient temperature and humidity, and the temperature and humidity before and after entering the cooling and dehumidification system, and communicates with the gas turbine DCS system in real time. Finally, the clean, dry, and low-temperature air, after being treated by high-efficiency fine filtration, enters the gas turbine combustion chamber through the gas turbine inlet under the action of the gas turbine air compressor.

6. The device according to claim 5, characterized in that, The intelligent control system for intake air filtration includes: Local storage units and cloud platforms are used to store normal data during device operation; The analysis unit obtains the total output power and total heat consumption of the gas turbine under factory conditions, denoted as the standard reference value P for output power. (gt) and the standard reference value for heat consumption HR (gt) Under actual operating conditions, the total output power P of the gas turbine is... (gtc) It is obtained from the following formula: In the formula: C P1 —Correction factor for output power based on ambient temperature; C P2 —The correction factor for the output power due to pressure difference; C P3 —Correction factor for relative humidity on output power; C P4 —Correction factor for the low calorific value of fuel on output power; C P5 —The correction factor for the output power of the exhaust pressure; Under actual operating conditions, the total heat consumption of the gas turbine is calculated using the following formula: In the formula: C H1 —Correction factor for heat loss due to ambient temperature; C H2 —The correction factor for heat loss due to atmospheric pressure; C H3 —The correction factor for heat loss due to relative humidity; C H4 —Correction factor for low calorific value of fuel on heat consumption; C H5 —The correction factor for heat loss due to exhaust pressure; Based on the ambient temperature, the pressure difference of the high-efficiency fine filter, and the humidity before and after entering and exiting the cooling and dehumidification system, establish the relationship curves between output power and heat consumption, and obtain the correction coefficients for output power and heat consumption under their respective conditions. Optimize the gas turbine inlet temperature to maximize output power and minimize heat consumption; The display unit is used to display the relationship curves between various indicators and output power, the relationship curves between indicators and heat consumption, and the optimization results.

7. The device according to claim 5, characterized in that, The intelligent control system for intake air filtration is loaded with a large language model based on Transformers. The large language model based on Transformers takes the pressure difference of each stage after primary, medium and high efficiency treatment, the ambient temperature and humidity, the temperature and humidity before and after entering and exiting the cooling and dehumidification system, the supply and return water temperature and flow rate of the chilled water system, and the gas turbine inlet temperature as inputs, and output power and heat consumption as outputs. The output of the large language model based on Transformers is compared with the set normal range of output power and heat consumption to determine whether there is a fault in the gas turbine.

8. The device according to claim 5, characterized in that, The intake filtration system, while ensuring intake pressure differential, optimizes the structure and effective filtration area of ​​each filter element, setting up three filter elements in sequence: a pre-filter bag filter, a medium-efficiency V filter, and a high-efficiency fine filter. The pre-filter bag filter adopts a bag structure with at least 10 bags, and the filter material is M6 grade high-efficiency glass fiber. The medium-efficiency V filter adopts a 5V structure, and the filter material is F8 grade high-efficiency glass fiber. The high-efficiency fine filter adopts a cylindrical structure, and the diameter and length are adjusted according to the intake space. The filter material is F9 grade fully synthetic composite fiber.

9. The device according to claim 5, characterized in that, The cooling system is set up in a layered and grouped manner, grouped by top and bottom, and layered by inside and outside. The cooling coils in the group are connected in series, and the cooling coils between different layers are independent of each other. The cooling coils include heat dissipation channels, and two heat dissipation fins are set in each heat dissipation channel. The two ends of the heat dissipation pipes of each layer are connected to the chilled water system through chilled water supply pipes and chilled water return pipes, respectively. At least four separate cooling coils, each with at least three layers, with the angle between the two heat dissipation fins being 60° to 120°, and the heat dissipation fins being at an angle of 30° to 60° to the airflow direction. The two heat dissipation fins need to overlap 1-2cm horizontally. The thickness of each cooling coil is 10-30cm. The heat dissipation pipes are made of copper or stainless steel, and the heat dissipation fins are made of corrosion-resistant aluminum foil or stainless steel sheets. The dehumidification system uses a water mist filter, a solid adsorbent, or a lithium chloride rotary dehumidification system with a thickness of 5-10 cm and a cross-sectional area adjusted according to the air intake space.

10. The device according to claim 5, characterized in that, In summer, hot, humid outdoor air containing dust enters the air intake filtration system. It passes through a pre-filter bag filter with 10 bags of M6 grade filter media, which removes about 60% of dust particles with a diameter of ≥0.4μm. Then it passes through a medium-efficiency V filter with 5V structure F8 filter media, which removes about 90% of dust particles with a diameter of ≥0.4μm, resulting in cleaner air. When the outdoor temperature exceeds 22°C, the air passes through a medium-efficiency V filter and enters the cooling system connected to the chilled water system. The cooling system consists of at least four sets of cooling coils. When the chilled water system is turned on, the chilled water enters the cooling coils in the cooling system to cool the high-temperature air to 22°C. The cooled air enters the dehumidification system after being processed by the cooling system. The dehumidification system removes condensate and excess moisture from the air, turning it into cleaner, cooler, and drier air, which then enters the high-efficiency filter. The high-efficiency fine filter uses a cylindrical structure and 1m long F9 grade fully synthetic composite fiber to filter out 95% of dust particles with a diameter ≥0.4μm, turning them into clean, low-temperature, and dry air. This clean air then enters the gas turbine combustion chamber through the gas turbine inlet under the action of the air compressor.

Images