Gas analysis method and system for offshore wind power single pile foundation

Abstract

The invention provides an offshore wind power single pile foundation gas analysis method and system.The system comprises gas production equipment deployed in a pile and analysis equipment deployed on an operation platform and applying an MEMS device gas sensor, and the method comprises the steps that S10, gas in the pile is collected through the gas production equipment; s20, analyzing the collected gas concentration through analysis equipment to obtain gas analysis data; s30, collecting the environment in the pile through an environment sensor to obtain environment data; and S40, correcting the gas analysis data through the environment data and the deep learning model to obtain gas real data. On one hand, analysis equipment of a high-sensitivity photon type detector chip integrated with a metamaterial infrared absorber is used for replacing a traditional electrochemical sensor to analyze gas concentration, on the other hand, environmental data are collected, and corresponding error coefficients in all environments are obtained in a deep learning mode; and correcting by using the error coefficient to obtain a true value of the analysis gas concentration.

Description

Technical Field

[0001] The present invention relates to the technical field of offshore wind power, and in particular to a gas analysis method and system for an offshore wind power single pile foundation. Background Art

[0002] At present, the foundation of offshore wind turbines mainly adopts single pile foundation, composite tube foundation and other forms. For offshore single pile foundation, after it penetrates the seabed, the seawater and seabed silt in the original position will continue to remain inside the foundation. These silts and organic substances rich in water exist in the closed space inside the single pile for a long time. Under the decomposition of microorganisms, they gradually corrode and produce a mixed gas rich in hydrogen sulfide gas. Although the single pile foundation is separated from the inside of the tower by a foundation platform, the foundation platform is not completely sealed, so the hydrogen sulfide gas in the single pile foundation will gradually overflow into the internal space of the tower.

[0003] Hydrogen sulfide gas can directly hinder the body's oxygen uptake and transportation, causing the respiratory enzymes in the cells to lose their activity, resulting in cell hypoxia and suffocation and death. When the concentration of hydrogen sulfide gas in the air is 10-15ppm, people can show symptoms of poisoning. GBZ2.1-2019 "Occupational Exposure Limits for Hazardous Factors in the Workplace: Part 1 Chemical Hazardous Factors" stipulates that the concentration of hydrogen sulfide gas is 10mg / m3, that is, below 6.6ppm. If the concentration of hydrogen sulfide in some wind turbines is not handled in time, it can reach 500ppm or even higher, seriously endangering the life safety of operation and maintenance personnel. In addition, hydrogen sulfide gas is a corrosive pollutant with a greater threat to power semiconductors. Among them, the converter is the most seriously affected. The IGBT module usually causes the temperature to rise during the operation of the wind turbine. Under the action of the applied voltage, hydrogen sulfide gas reacts with copper devices exposed to the air to form copper sulfide (CuS) crystals. Copper sulfide crystals can damage the IGBT switch and cause the module to fail, and in the worst case, even cause a short circuit. In addition, it will also cause pitting corrosion of components such as relays and contactors within the unit, resulting in poor contact, seriously affecting the safe and stable operation of the unit. Summary of the invention

[0004] The present invention provides a gas analysis method and system for an offshore wind power single pile foundation, which solve the problems in the above-mentioned wind turbine pile gas detection process.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: A gas analysis method for an offshore wind power monopile foundation comprises the following steps: S10 collects the gas in the pile through gas collection equipment; S20 analyzes the collected gas concentration through an analysis device to obtain gas analysis data; S30 collects the environment inside the pile through the environmental sensor to obtain environmental data; S40 corrects the gas analysis data through environmental data and deep learning models to obtain real gas data.

[0006] In the above S30, the environmental data collected by the environmental sensor include the air pressure, temperature, humidity and salt spray concentration in the pile. After collection, the collected environmental data is normalized: ; Environmental data x According to the above formula, is the mean of the data, σ is the standard deviation of the data.

[0007] In the above S40, the environmental data and the deep learning model are neural network models, and the air pressure, temperature, humidity and salt spray concentration in the pile are used as inputs of the neural network model to obtain the error coefficient.

[0008] The above neural network model is a fully connected layer BP neural network, including an input layer, a hidden layer and an output layer; The number of input layer nodes of the BP neural network is the same as the number of types of environmental data, the number of hidden layer nodes of the BP neural network is greater than or equal to the input layer nodes, and the number of output nodes of the BP neural network is 1, which is the error coefficient; Each environmental data enters the network from the input layer, and is processed layer by layer through the hidden layer until it is output at the output layer. The state of neurons in each layer only affects the state of neurons in the next layer. If the expected output is not obtained in the output layer, the signal transmission process enters the back propagation, and the weights and thresholds are continuously adjusted according to the prediction error, so that the predicted output results of the BP neural network continue to approach the expected output value; then, according to the environmental data, the error coefficient of gas concentration monitoring of the analysis equipment in the offshore environment is obtained, and the measured gas concentration is corrected by the error coefficient.

[0009] When the measured gas concentration is corrected by the error coefficient as mentioned above, simulated gas data is introduced to obtain the real gas data of the previous time period, and the simulated gas data of the next time period is directly obtained through curve simulation. The slope of a certain node of the simulated gas data in the next time period is then compared with the error coefficient to obtain an intermediate value of the slope and the error coefficient as a correction to the gas analysis data to obtain the real gas data.

[0010] The above curve simulation obtains the simulated gas data using the least squares method.

[0011] The above-mentioned process of using the least square method to simulate the curve and obtain the simulated gas data is as follows: For a certain gas, the gas concentration at n different time nodes is known: ( , )( , )...( , ); The change in gas concentration is expressed by the parabolic curve equation: ; in, , , is the unknown coefficient, and ( , ) into the equation, we get: ; That is: ; Then for ( , ),have: ; The combined matrix is: ; Assumptions For A, For T, is X, then AX=T; have: T; Then the equation for simulation curve is obtained.

[0012] A gas analysis system is obtained using the above-mentioned gas analysis method for a single offshore wind power pile foundation. The system includes a gas collection device installed in the pile and an analysis device installed on an operating platform. The gas collection device is used to collect the gas in the pile and send it to the analysis device. The analysis device receives the collected gas and performs gas concentration analysis to obtain gas analysis data.

[0013] The gas collection equipment mentioned above is a collection pump installed in the pile; Alternatively, the gas collection equipment can use the increased pressure caused by the gas generated by the decomposition of microorganisms in the pile as a power source for collecting the gas in the pile; Alternatively, the gas extraction equipment extracts the gas in the pile by means of vacuum negative pressure and transfers it to the analysis equipment.

[0014] The above-mentioned analysis equipment includes a MEMS device gas detection sensor. The MEMS device gas detection sensor adopts a narrow-band infrared detector array formed by a photon detector chip integrating a metamaterial infrared absorber. The narrow-band infrared detector array is designed according to the type of gas to be detected and the infrared characteristic absorption spectrum, and the metamaterial infrared absorber filter array and different types of photon detector arrays are designed to filter out stray infrared light and convert gas-related infrared light signals into electrical signals. After subsequent signal processing and operation module algorithm processing, multi-gas information and concentration are calculated; and then displayed in real time through a digital display module or transmitted to the cloud through a wireless transmission module; The signal processing and operation module contains a signal amplification circuit, a filtering circuit and an analog-to-digital converter for amplifying, filtering and analog-to-digital converting the array electrical signal output by the narrow-band infrared detector array to output a digital electrical signal, and transmits the digital electrical signal to the back-end control circuit for processing.

[0015] The present invention provides a gas analysis method and system for an offshore wind power single pile foundation. On the one hand, the gas concentration is analyzed by using an analysis device with a high-sensitivity photon-type detector chip that integrates a metamaterial infrared absorber instead of a traditional electrochemical sensor. On the other hand, environmental data is collected, and the corresponding error coefficient under each environment is obtained by deep learning. The error coefficient is used to correct the true value of the analyzed gas concentration. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 It is a distribution diagram of the system of the present invention; Figure 2 It is a schematic diagram of the structure of the analysis equipment; Figure 3 Schematic diagram of the packaging of the analysis equipment components.

[0017] Among them: analysis equipment 1, gas extraction equipment 2. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention clearer, the following content will systematically and completely describe the specific technical solutions of the present invention in combination with the drawings provided according to the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0019] Embodiment 1: A gas analysis method and system for an offshore wind power single pile foundation, the system comprising gas extraction equipment deployed in the pile and analysis equipment using a MEMS device gas sensor deployed on an operating platform.

[0020] The method performs steps S10 to S40 according to the system.

[0021] S10 collects the gas in the pile through gas collection equipment.

[0022] S20 analyzes the collected gas concentration through an analysis device to obtain gas analysis data.

[0023] S30 collects the environment inside the pile through environmental sensors to obtain environmental data.

[0024] S40 corrects the gas analysis data through environmental data and deep learning models to obtain real gas data.

[0025] The offshore wind power monopile foundation gas analysis method realizes the application of the analysis system, among which, the following Figure 1 As shown, the offshore wind power monopile foundation gas system includes gas collection equipment and analysis equipment.

[0026] The gas collection equipment can be deployed in the pile to detect the gas in the pile, and the analysis equipment is deployed on the operating platform to receive the gas from the collection equipment in the pile and realize gas detection.

[0027] Steps S10 to S40 are performed based on the offshore wind power monopile foundation gas analysis system.

[0028] S10 collects the gas in the pile through gas collection equipment.

[0029] The gas extraction equipment may be a gas extraction pump or a similar device installed in a pile, and the power source of the gas extraction equipment is provided by the energy storage of the wind power equipment.

[0030] Alternatively, the gas inside the pile can be collected by utilizing the increased pressure caused by the gas produced by the decomposition of microorganisms inside the pile as a power source.

[0031] For example, a separate one-way valve flap may be provided in the pile to connect to the collection area. When the microorganisms decompose and produce gas, causing the pile (bottom area) to be pressurized, the one-way valve flap opens and part of the gas enters the collection area.

[0032] Alternatively, the gas in the pile (bottom area) is extracted by vacuum negative pressure until it is transferred to the collection area.

[0033] S20 analyzes the collected gas concentration by using an analysis device of a MEMS device gas sensor to obtain gas analysis data.

[0034] Among them, the analysis equipment uses MEMS device gas detection sensors.

[0035] For example, Figure 2 As shown, the MEMS gas detection sensor uses a narrow-band infrared detector array formed by a highly sensitive photon detector chip with integrated metamaterial infrared absorbers as the core component. According to the type of gas to be measured and the infrared characteristic absorption spectrum, a metamaterial infrared absorber filter array and different types of photon detector arrays are designed to filter out stray infrared light and convert gas-related infrared light signals into electrical signals. After subsequent signal processing and calculation module algorithm processing, multi-gas information and concentration are calculated; then they are displayed in real time through a digital display module or transmitted to the cloud through a wireless transmission module.

[0036] Among them, the infrared light source can adopt a MEMS light source, which has the advantages of electrical modulation and stable power output.

[0037] Among them, the narrow-band infrared detector array includes a plurality of highly sensitive photon-type detector chips integrated with metamaterial infrared absorbers. The absorption band of the metamaterial infrared absorber and the response band of the photon-type detector chip both cover the infrared characteristic absorption bands of all component gases to be measured in the scene where the gas to be measured is located. The metamaterial infrared absorber can be constructed using an optical antenna array to filter out infrared light (stray infrared light) that is irrelevant to the characteristic absorption band of the gas, and only retain infrared light in the band related to the concentration information of the gas to be measured.

[0038] Among them, the following Figure 3 As shown, a semiconductor cooler module can be added to the bottom of the package of the narrow-band infrared detector array, and a TO package is used on the outside.

[0039] Among them, the signal amplification circuit, filtering circuit and analog-to-digital converter are used to amplify, filter and convert the array electrical signal output by the narrow-band infrared detector array into a digital electrical signal, and transmit the digital electrical signal to the back-end control circuit for processing.

[0040] Then, when conducting gas detection, the gas to be tested enters the equipment in a circulation or diffusion manner. According to the infrared spectrum absorption principle, each gas in the gas to be tested corresponds to a different infrared characteristic absorption peak. At this time, the infrared light source emits a wide-spectrum infrared light covering the detection band of the gas to be tested. When the infrared light passes through the gas to be tested, characteristic infrared absorption of multiple bands occurs. The absorption band corresponds to the sum of the infrared characteristic absorption bands of all gases to be tested. The infrared light absorbed by the gas to be tested reaches the narrow-band infrared detector array, which is composed of a metamaterial infrared absorber and a photon-type detector chip array. The receiver filters out infrared light (stray light) that is irrelevant to the gas concentration information to be measured, and the photon detector chip is used to convert infrared light carrying the characteristic absorption band information of the gas into an electrical signal; after the array electrical signal output by the narrow-band infrared detector array reaches the signal processing and operation module, in order to reduce the noise influence in the sensor system and improve the detection sensitivity, the output signal of the narrow-band infrared detector array is amplified, filtered and analog-to-digital converted and then enters the control circuit, and the gas concentration information is calibrated according to the Lambert-Beer law and the infrared characteristic absorption principle, and a mathematical model of the gas concentration is established through multiple sampling, and the gas component and concentration information to be measured is output.

[0041] S30 collects the environment inside the pile through environmental sensors to obtain environmental data.

[0042] Among them, the air pressure, temperature, humidity, salt spray concentration, etc. inside the pile are collected through a variety of environmental sensors.

[0043] Among them, the collected environmental data are normalized.

[0044] For example, , each environmental data x is calculated according to the above formula, is the mean of the data, and σ is the standard deviation of the data.

[0045] S40 corrects the gas analysis data by using the environmental data to obtain the real gas data.

[0046] Among them, the air pressure, temperature, humidity, salt spray concentration, etc. inside the pile are used as inputs of the neural network model to obtain the error coefficient.

[0047] For example, the neural network model is a fully connected layer BP neural network including an input layer, a hidden layer, and an output layer. The number of nodes in the input layer of the BP neural network is the same as the number of types of environmental data, the number of nodes in the hidden layer of the BP neural network is greater than or equal to the number of nodes in the input layer, and the number of output nodes of the BP neural network is 1, which is the error coefficient. Each environmental data enters the network from the input layer, and is processed layer by layer through the hidden layer until it is output in the output layer. The state of neurons in each layer only affects the state of neurons in the next layer. If the expected output is not obtained in the output layer, the signal transmission process is transferred to back propagation, and the weights and thresholds are continuously adjusted according to the prediction error, so that the predicted output result of the BP neural network continues to approach the expected output value.

[0048] Then, based on the environmental data, the error coefficient of the gas concentration monitoring of the analysis equipment in the offshore environment can be analyzed, and the measured gas concentration can be corrected by the error coefficient to improve the monitoring accuracy.

[0049] Among them, in addition to using the error coefficient to correct the monitored gas concentration represented by the gas analysis data, simulated gas data can be introduced. That is, the real gas data of the previous period is obtained, and the simulated gas data of the next period is directly obtained through curve simulation. Then, the slope of a certain node of the simulated gas data in the next period is compared with the error coefficient, and an intermediate value of the slope and the error coefficient is obtained as a correction to the gas analysis data to obtain the real gas data.

[0050] For the simulated gas data obtained by curve simulation, the least square method can be used.

[0051] For example, for a certain gas, the gas concentrations at n different time nodes are known: (x1, y1) (x2, y2)... (xn, yn).

[0052] Through research and analysis, it was found that the change in the gas concentration of acetylene C2H2 can be expressed by a curve equation that approximates a parabola. . Among them, a0, a1, a2 are unknown coefficients. If (x1, y1) is substituted into the equation, we can get , that is, . Then (xi,yi), we have: .

[0053] The combined matrix is: , assuming For A, For T, is X, then AX=T. Then we have T, and then get the equation for the simulation curve. Then the simulated gas data of any node can be simulated through the equation.

Claims

1. Gas analysis method for offshore wind power monopile foundation, characterized in that: The following steps are involved: S10 collects the gas in the pile through gas collection equipment; S20 analyzes the collected gas concentration through an analysis device to obtain gas analysis data; S30 collects the environment inside the pile through the environmental sensor to obtain environmental data; S40 corrects the gas analysis data through environmental data and deep learning models to obtain real gas data.

2. The offshore wind power monopile foundation gas analysis method according to claim 1, characterized in that: In the above S30, the environmental data collected by the environmental sensor include the air pressure, temperature, humidity and salt spray concentration in the pile. After the collection, the collected environmental data is normalized: ； Environmental data x According to the above formula, is the mean of the data, σ is the standard deviation of the data.

3. The offshore wind power monopile foundation gas analysis method according to claim 2, characterized in that: In the above-mentioned S40, the environmental data and the deep learning model are neural network models, and the air pressure, temperature, humidity and salt spray concentration inside the pile are used as inputs of the neural network model to obtain the error coefficient.

4. The offshore wind power monopile foundation gas analysis method according to claim 3, characterized in that: The neural network model is a fully connected BP neural network, including an input layer, a hidden layer and an output layer; The number of input layer nodes of the BP neural network is the same as the number of types of environmental data, the number of hidden layer nodes of the BP neural network is greater than or equal to the input layer nodes, and the number of output nodes of the BP neural network is 1, which is the error coefficient; Each environmental data enters the network from the input layer, and is processed layer by layer through the hidden layer until it is output at the output layer. The state of neurons in each layer only affects the state of neurons in the next layer. If the expected output is not obtained in the output layer, the signal transmission process enters the back propagation, and the weights and thresholds are continuously adjusted according to the prediction error, so that the predicted output results of the BP neural network continue to approach the expected output value; then, according to the environmental data, the error coefficient of gas concentration monitoring of the analysis equipment in the offshore environment is obtained, and the measured gas concentration is corrected by the error coefficient.

5. The offshore wind power monopile foundation gas analysis method according to claim 4, characterized in that: When the measured gas concentration is corrected by the error coefficient, simulated gas data is introduced to obtain the real gas data of the previous time period, and the simulated gas data of the next time period is directly obtained through curve simulation. The slope of a certain node of the simulated gas data in the next time period is then compared with the error coefficient to obtain an intermediate value of the slope and the error coefficient as a correction to the gas analysis data to obtain the real gas data.

6. The offshore wind power monopile foundation gas analysis method according to claim 5, characterized in that: The curve simulation obtains simulated gas data using the least squares method.

7. The offshore wind power monopile foundation gas analysis method according to claim 6, characterized in that: The process of using the least square method to simulate the curve and obtain the simulated gas data is as follows: For a certain gas, the gas concentration at n different time nodes is known: ( , )( , )...( , ); The change in gas concentration is expressed by the parabolic curve equation: ； in, , , is the unknown coefficient, and ( , ) into the equation, we get: ； That is: ； Then for ( , ),have: ； The combined matrix is: ； Assumptions For A, For T, is X, then AX=T; have: T； Then the equation for simulation curve is obtained.

8. A gas analysis system using the offshore wind power monopile foundation gas analysis method according to claim 7, characterized in that: The system includes a gas collection device installed in the pile and an analysis device installed on the operating platform. The gas collection device is used to collect the gas in the pile and send it to the analysis device. The analysis device receives the collected gas and performs gas concentration analysis to obtain gas analysis data.

9. The gas analysis system according to claim 8, characterized in that: The gas collection equipment is a collection pump installed in the pile; Alternatively, the gas collection equipment can use the increased pressure caused by the gas generated by the decomposition of microorganisms in the pile as a power source for collecting the gas in the pile; Alternatively, the gas extraction equipment extracts the gas in the pile by means of vacuum negative pressure and transfers it to the analysis equipment.

10. The gas analysis system according to claim 9, characterized in that: The analysis equipment includes a MEMS device gas detection sensor. The MEMS device gas detection sensor adopts a narrow-band infrared detector array formed by a photon detector chip integrating a metamaterial infrared absorber. The narrow-band infrared detector array is designed according to the type of gas to be detected and the infrared characteristic absorption spectrum, and the metamaterial infrared absorber filter array and different types of photon detector arrays are designed to filter out stray infrared light and convert gas-related infrared light signals into electrical signals. After subsequent signal processing and operation module algorithm processing, multi-gas information and concentration are calculated; and then displayed in real time through a digital display module or transmitted to the cloud through a wireless transmission module; The signal processing and operation module contains a signal amplification circuit, a filtering circuit and an analog-to-digital converter for amplifying, filtering and analog-to-digital converting the array electrical signal output by the narrow-band infrared detector array to output a digital electrical signal, and transmits the digital electrical signal to the back-end control circuit for processing.

Images