A transformer oil quality analysis photoacoustic spectroscopy detection method and device

Abstract

The present application relates to the technical field of transformer on-line monitoring, in particular to a transformer oil quality analysis photoacoustic spectrum detection system and method. The method extracts transformer oil sample through a circulating oil pump, and removes dissolved gas in the oil by using an oil-gas separation membrane assembly under a negative pressure environment. After preheating by a spiral coil, the removed gas is filtered and homogenized by a micro-porous membrane structure, and then sent into a resonant photoacoustic cell. A main controller dynamically adjusts a laser modulation frequency according to a gas flow rate, so as to maintain the stability of an acoustic resonance field in the resonant photoacoustic cell. The modulation laser excites the measured gas to generate a photoacoustic signal, and after collection and processing, the concentration of the multi-component dissolved gas is decoupled and calculated, and the transformer fault early warning is carried out accordingly. The present application has the advantages of stable degassing, high detection sensitivity, strong anti-interference ability, and suitability for multi-component continuous detection.

Description

Technical Field

[0001] This invention relates to the field of online transformer monitoring technology, specifically to a photoacoustic spectroscopy detection system and method for transformer oil quality analysis. Background Technology

[0002] Transformers are critical equipment in power systems, and their operating status directly affects the safety and stability of the power transmission and distribution system. During long-term operation, transformers are affected by fault factors such as localized overheating, discharge, and insulation aging. This leads to the gradual generation and dissolution of characteristic gases such as hydrogen, methane, ethane, ethylene, acetylene, carbon monoxide, and carbon dioxide in the insulating oil. By detecting and analyzing the changes in the types and concentrations of dissolved gases in transformer oil, the development of latent faults within the transformer can be effectively determined. Therefore, dissolved gas analysis in transformer oil has become an important technical means for transformer condition monitoring and fault early warning.

[0003] In existing technologies, gas chromatography or other offline analytical methods are typically used to detect dissolved gases in transformer oil. While these methods offer a certain level of accuracy, they generally suffer from problems such as complex sample pretreatment, long detection cycles, large equipment size, and difficulty in achieving continuous online monitoring. Meanwhile, while photoacoustic spectroscopy offers high sensitivity and multi-component analysis potential for detecting extracted gases, existing solutions still face some shortcomings in practical applications. For example, dissolved gases may not be fully extracted from the oil, oil mist may be easily entrained during the degassing process, and the temperature fluctuations and flow patterns of the gas to be tested before entering the detection chamber can be significant, affecting the stability of the acoustic field and the consistency of the photoacoustic signal within the resonant photoacoustic cell. Furthermore, the coexistence of multiple gas components can easily lead to cross-interference, affecting the accuracy of concentration inversion.

[0004] Therefore, how to provide a transformer oil quality analysis method and apparatus that can take into account degassing efficiency, gas purity, temperature and flow field stability, and is suitable for high-sensitivity detection of multiple components has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a photoacoustic spectroscopy detection method and device for transformer oil quality analysis, which addresses the shortcomings of the prior art.

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

[0007] A photoacoustic spectroscopy method for transformer oil quality analysis, the method comprising:

[0008] Transformer oil samples are drawn by circulating oil pumps and sent into the degassing chamber. Dissolved gases in the oil are removed by the oil-gas separation membrane assembly under negative pressure.

[0009] The gas to be tested is preheated by a spiral coil attached to the outer wall of the annular cavity, filtered and homogenized by a microporous membrane structure set at the air inlet of the resonant photoacoustic cell, and then sent into the resonant photoacoustic cell. The annular cavity covers the laser and the outside of the resonant photoacoustic cell, and a temperature gradient field with a gradually decreasing temperature from the laser end to the resonant photoacoustic cell end is formed inside it.

[0010] The main controller dynamically adjusts the laser modulation frequency according to the gas flow rate entering the resonant photoacoustic cell to maintain the stability of the acoustic resonance field within the resonant photoacoustic cell.

[0011] The modulated laser is incident into the resonant photoacoustic cell to excite the gas to be tested to generate a photoacoustic signal. The photoacoustic signal is captured by a microphone and processed to achieve decoupled calculation of the concentration of multi-component dissolved gases in the oil. Based on the calculation results, transformer fault early warning is provided.

[0012] The process of drawing transformer oil samples via a circulating oil pump and sending them into the degassing chamber, where dissolved gases are removed from the oil through an oil-gas separation membrane assembly under negative pressure, includes:

[0013] Transformer oil samples are drawn by the circulating oil pump, and after impurities are removed by the oil sample filter located upstream of the circulating oil pump, the oil is sent into the degassing chamber through the sampling valve.

[0014] Inside the degassing chamber, the dissolved gases in the oil sample are removed through the oil-gas separation membrane assembly by the negative pressure environment provided by the vacuum pump. The oil-gas separation membrane assembly is made of an oleophobic and gas-friendly semi-permeable membrane material.

[0015] The process involves preheating the extracted test gas through a spiral coil attached to the outer wall of the annular cavity, filtering and homogenizing it through a microporous membrane structure located at the inlet of the resonant photoacoustic cell, and then sending it into the resonant photoacoustic cell. This includes:

[0016] The extracted gas to be tested is introduced into the spiral coil after passing through a multi-channel selection valve. The heat of the annular cavity is used to preheat the gas to be tested, so that the temperature of the gas to be tested is similar to the ambient temperature of the resonant photoacoustic cell.

[0017] The multi-channel selection valve has at least three channels, wherein the first channel is connected to the degassing chamber for gas detection, the second channel is connected to the standard gas bag for self-calibration, and the third channel is connected to the exhaust end of the vacuum pump for gas path purging and self-circulation.

[0018] The preheated gas to be tested is fed into the resonant photoacoustic cell through a microporous membrane structure located at the inlet and outlet of the resonant photoacoustic cell. The microporous membrane structure is made of an oleophobic and anaerobic material, and its porosity gradually increases from the inlet end to the center end of the resonant photoacoustic cell along the gas flow direction, so as to isolate the oil mist carried during the degassing process and to homogenize the diffusion of the gas entering the resonant photoacoustic cell.

[0019] The microporous membrane structure is made of polyethersulfone material, with a membrane thickness of 0.5 mm and a pore size range of 1-10 μm. Its porosity gradually increases from 20% at the inlet end to 50% near the center end of the resonant photoacoustic cell.

[0020] The annular cavity surrounds the laser and the resonant photoacoustic cell, and forms a temperature gradient field inside it that gradually decreases in temperature from the laser end to the resonant photoacoustic cell end, including:

[0021] A flow-guiding fin is provided on the inner wall of the annular cavity. The distribution density of the flow-guiding fin increases along the direction from the laser end to the resonant photoacoustic cell end. The heat conduction efficiency of the corresponding region is controlled by the difference in the distribution density of the flow-guiding fin in different regions, so that a temperature gradient field with the temperature gradually decreasing from the laser end to the resonant photoacoustic cell end is formed inside the annular cavity.

[0022] The step of dynamically adjusting the laser modulation frequency according to the gas flow rate entering the resonant photoacoustic cell by the main controller to maintain the stability of the acoustic resonant field within the resonant photoacoustic cell includes:

[0023] The gas flow rate entering the resonant photoacoustic cell is adjusted and obtained by a mass flow controller located upstream of the air inlet of the resonant photoacoustic cell;

[0024] The main controller is connected to the mass flow controller and the optical chopper respectively. Based on the gas flow rate signal fed back by the mass flow controller, the modulation frequency of the optical chopper is adjusted in real time to keep the laser modulation frequency dynamically matched with the gas flow rate and maintain the stability of the acoustic resonance field in the resonant photoacoustic cell.

[0025] The real-time adjustment of the modulation frequency of the optical chopper to dynamically match the laser modulation frequency with the gas flow rate includes:

[0026] The optical chopper is positioned between the laser and the fiber collimator, and its rotation is driven by a motor to achieve periodic modulation of the laser output from the laser.

[0027] The main controller adjusts the speed of the motor according to the gas flow rate signal fed back by the mass flow controller to change the modulation frequency of the optical chopper, so that the modulation frequency matches the acoustic resonance frequency caused by the change in gas flow rate in the resonant photoacoustic cell.

[0028] The process involves incident a modulated laser into the resonant photoacoustic cell to excite the gas under test to generate a photoacoustic signal, which is then captured by a microphone and processed to achieve decoupled calculation of the concentration of multi-component dissolved gases in the oil. This includes:

[0029] The laser output from the laser is modulated by an optical chopper, collimated by an optical fiber collimator to output a parallel beam, and then focused by a focusing lens and incident into the resonant photoacoustic cell. A resonator is provided inside the resonant photoacoustic cell. The gas to be tested is excited by the laser in the resonator to generate a photoacoustic signal, which is captured by a microphone located on the side wall of the resonant photoacoustic cell.

[0030] The photoacoustic signal captured by the microphone is sequentially amplified by a preamplifier, demodulated by a lock-in amplifier, and acquired by a data acquisition card to achieve decoupled calculation of the concentration of multi-component dissolved gases in oil, and to provide early warning of transformer faults based on the calculation results.

[0031] The process of sequentially amplifying the photoacoustic signal captured by the microphone, demodulating it with a lock-in amplifier, and acquiring it with a data acquisition card to achieve decoupled calculation of the concentration of multi-component dissolved gases in the oil includes:

[0032] The laser is controlled to sequentially output multiple detection wavelengths corresponding to the characteristic absorption wavelengths of various target dissolved gases, and the photoacoustic signal intensity at each detection wavelength is obtained.

[0033] Construct an absorption coefficient matrix, wherein the elements of the absorption coefficient matrix are the absorption coefficients of each target dissolved gas at each detection wavelength;

[0034] Based on the photoacoustic signal intensity at each detection wavelength and the absorption coefficient matrix, a set of linear equations is established between the photoacoustic signal intensity vector and the gas concentration vector. By solving the set of linear equations, the concentration values ​​of each target dissolved gas are obtained.

[0035] The method further includes:

[0036] A multi-channel selection valve is installed in the gas path between the degassing chamber and the resonant photoacoustic cell. The main controller controls the multi-channel selection valve to perform the following operations sequentially according to a preset cycle:

[0037] Switch to the calibration path, introduce standard gas of known concentration from the standard gas bag into the resonant photoacoustic cell for photoacoustic detection, compare the detection result with the known concentration of the standard gas, and complete the system self-calibration;

[0038] Switch to the purging path and use a vacuum pump to evacuate and purge the resonant photoacoustic cell and gas pipeline to remove residual gas.

[0039] Switch to the detection path and introduce the gas to be tested extracted from the degassing chamber into the resonant photoacoustic cell for photoacoustic detection.

[0040] Compared with the prior art, the beneficial effects of the present invention are:

[0041] 1. This invention achieves continuous and stable extraction of dissolved gases from transformer oil through the combination of circulating oil extraction, negative pressure degassing, and oil-gas separation membrane components; and effectively reduces the impact of oil mist entrainment, gas temperature fluctuations, and flow inhomogeneity on the detection process through spiral coil preheating, microporous membrane filtration homogenization, and temperature gradient field control, thereby improving the stability and cleanliness of the gas to be tested before entering the resonant photoacoustic cell.

[0042] 2. This invention achieves highly sensitive and accurate detection of multi-component dissolved gases in transformer oil by dynamically adjusting the laser modulation frequency according to the gas flow rate to maintain the stability of the acoustic resonant field in the resonant photoacoustic cell, and by combining multi-detection wavelength photoacoustic signal acquisition with multi-component concentration decoupling calculation, which is beneficial to improving the reliability and practicality of transformer fault early warning. Attached Figure Description

[0043] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0044] Figure 1 A schematic flowchart of a photoacoustic spectroscopy detection method for transformer oil quality analysis provided in an embodiment of the present invention;

[0045] Figure 2 This is a schematic diagram of the structure of a photoacoustic spectroscopy detection device for transformer oil quality analysis provided in an embodiment of the present invention;

[0046] Figure 3 This is a schematic diagram of the connection structure of a multi-channel selector valve provided in an embodiment of the present invention;

[0047] Figure 4 This is a schematic diagram of the structure of the light source optical path module and the constant temperature control unit provided in an embodiment of the present invention;

[0048] Figure 5 A schematic diagram of a microporous membrane structure provided in an embodiment of the present invention;

[0049] Figure 6 A schematic diagram of the flow guide fin structure of the annular cavity provided in an embodiment of the present invention;

[0050] Among them, 10-oil circuit degassing module; 11-degassing chamber; 12-circulating oil pump; 13-oil-gas separation membrane assembly; 14-sampling valve; 15-oil sample filter; 16-multi-channel selector valve; 17-standard gas bag; 18-spiral coil; 21-optical chopper; 22-tunable semiconductor laser; 23-fiber collimator; 24-focusing lens; 31-air inlet; 32-resonant photoacoustic cell; 33-microphone; 34-air outlet; 35-micro porous membrane structure; 36-mass flow controller; 37-solenoid valve; 38-vacuum pump; 40-signal processing module; 41-interface unit; 51-annular cavity structure; 52-guide fin. Detailed Implementation

[0051] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0052] Example 1

[0053] Please see Figure 1 The present invention provides an embodiment of a photoacoustic spectroscopy method for transformer oil quality analysis, the specific steps of which are as follows:

[0054] S1: A transformer oil sample is drawn by a circulating oil pump and sent into the degassing chamber. Under negative pressure, the dissolved gas in the oil is removed by the oil-gas separation membrane assembly.

[0055] Specifically, during transformer operation, fault-specific gases such as hydrogen, methane, ethane, ethylene, acetylene, carbon monoxide, and carbon dioxide dissolve in the oil. These gases are typically uniformly distributed within the oil in a molecular dissolved state. If the liquid oil sample is used directly as the detection target, it will not only be affected by the oil's viscosity, fluidity, and background absorption, but the liquid environment will also weaken the response stability of the target gas during subsequent detection, thus hindering the accurate identification of multi-component dissolved gases. Therefore, in this embodiment, before detection, the dissolved gases in the oil sample are released and separated from the liquid phase, transforming the target from a complex oil system into a more stable gaseous component, thereby providing the necessary conditions for subsequent photoacoustic spectroscopy detection.

[0056] In this embodiment, a cyclic extraction method is used to continuously transport transformer oil samples, allowing the oil samples to continuously enter the degassing chamber and maintain a dynamic flow state. Compared to static sampling followed by one-time degassing, cyclic transport reduces oil sample retention and stratification in local areas, ensuring that newly entering oil samples continuously replenish the degassing area, thereby improving the uniformity of the overall oil sample's participation in degassing. Simultaneously, the oil sample undergoes filtration before entering the degassing chamber, removing suspended particles, aged deposits, and other mechanical impurities to reduce the possibility of impurities adhering, clogging, or disturbing the degassing process, ensuring a more stable subsequent gas evolution process.

[0057] Furthermore, after the oil sample enters the degassing chamber, a negative pressure environment below atmospheric pressure is created, altering the original dissolution equilibrium between the dissolved gases and the liquid phase. This increases the partial pressure difference between the gas components within the oil, promoting the migration of dissolved gases from the liquid phase to the gas phase. In other words, the negative pressure environment reduces the retention capacity of gases in the oil, making it easier for trace amounts of gas that were originally stably dissolved in the oil to precipitate, thereby improving degassing efficiency. This process is particularly beneficial for increasing the release of low-concentration fault gases, preventing the residual gases in the oil phase from affecting the sensitivity of subsequent detection.

[0058] During gas evolution, the evolved gas does not simply escape directly, but is further separated into gas and liquid through the selective permeation of the oil-gas separation membrane. Because the oil-gas separation membrane assembly is made of an oleophobic and gas-philic semi-permeable membrane material, it has high permeability to gases but a significant barrier effect on liquid oil. Thus, dissolved gases preferentially pass through the membrane layer into the gas path to be tested under the drive of the pressure difference, while the oil is confined to the original oil path side, thereby reducing the interference of droplet entrainment, oil mist mixing, and liquid phase residue on subsequent detection during degassing. This method not only improves the purity of the evolved gas but also reduces the risk of contamination from oil samples directly entering subsequent detection stages.

[0059] More preferably, through the combined action of circulating oil extraction, negative pressure precipitation, and membrane separation, dissolved gases in the oil sample can be extracted continuously and gradually, rather than being released abruptly and unevenly. This allows for smoother changes in composition and concentration of the analyte gases entering subsequent detection processes at different times, thereby improving the continuity, consistency, and repeatability of the detection process. Especially in scenarios requiring online or quasi-online monitoring of transformer operating status, this continuous degassing method is more conducive to accurately reflecting the dynamic changes in fault gases in the oil.

[0060] S2: The gas to be tested is preheated by a spiral coil attached to the outer wall of the annular cavity, filtered and homogenized by a microporous membrane structure set at the air inlet of the resonant photoacoustic cell, and then sent into the resonant photoacoustic cell. The annular cavity covers the laser and the outside of the resonant photoacoustic cell, and a temperature gradient field with a gradually decreasing temperature from the laser end to the resonant photoacoustic cell end is formed inside it.

[0061] Specifically, the gas to be tested obtained from the degassing step typically still exhibits temperature fluctuations, localized oil mist entrainment, and uneven flow patterns before entering the resonant photoacoustic cell. If directly fed into the resonant photoacoustic cell, it is prone to condensation or thermal drift due to the temperature inconsistency between the gas and the detection environment, affecting the stability of the photoacoustic response. Furthermore, trace amounts of oil mist or localized high-speed airflow entrained during degassing may also interfere with subsequent detection accuracy. Therefore, in this embodiment, the gas to be tested is preheated, filtered, and homogenized sequentially before entering the resonant photoacoustic cell to improve the consistency of the gas state entering the detection area.

[0062] In this embodiment, the gas to be tested is first preheated by a spiral coil attached to the outer wall of the annular cavity. Due to the stable heat distribution around the annular cavity, the gas can exchange heat sufficiently with the outer wall during spiral transport, gradually increasing its temperature to approach the operating temperature of the resonant photoacoustic cell. Using a spiral coil path extends the gas transport path, improving heat exchange efficiency; it also avoids sudden temperature changes caused by rapid gas flow, thus reducing transient disturbances when the gas enters the resonant photoacoustic cell. This method effectively reduces gas condensation, temperature unevenness, and the resulting photoacoustic signal drift.

[0063] Furthermore, the preheated gas to be tested enters the resonant photoacoustic cell after passing through a microporous membrane structure located at the inlet. This microporous membrane structure serves two purposes: firstly, it blocks oil mist that may be entrained during the degassing process, reducing the entry of liquid particles into the detection space; secondly, its porous diffusion characteristics buffer and redistribute the airflow, improving potential localized flow deviations, jetting, or concentration stratification, thus resulting in a more uniform gas distribution within the resonant photoacoustic cell. This helps improve the consistency of gas distribution within the detection area, preventing changes in localized concentration abruptness or uneven flow rate from affecting the detection results.

[0064] Simultaneously, the annular cavity surrounds the laser and the resonant photoacoustic cell, forming a temperature gradient field that gradually decreases in temperature from the laser end to the resonant photoacoustic cell end. This temperature gradient field allows the gas to undergo a gradual thermal transition as it is transported along the spiral coil and approaches the resonant photoacoustic cell, rather than abruptly switching in temperature. This helps maintain thermal stability during gas transport and also allows the gas entering the resonant photoacoustic cell to be closer to the target detection temperature environment, reducing the impact of external thermal disturbances on the detection process.

[0065] Overall, this step, through the processing method of spiral coil preheating + microporous membrane filtration and homogenization + gradual adjustment of temperature gradient field, ensures that the extracted gas to be tested is in a more stable, clean and uniform state before entering the resonant photoacoustic cell. This provides more reliable gas conditions for subsequent photoacoustic excitation and signal acquisition, and improves the accuracy and stability of multi-component dissolved gas detection.

[0066] S3: The main controller dynamically adjusts the laser modulation frequency according to the gas flow rate entering the resonant photoacoustic cell to maintain the stability of the acoustic resonance field in the resonant photoacoustic cell.

[0067] Specifically, the flow velocity of the gas to be tested after entering the resonant photoacoustic cell affects the sound field distribution and resonant state within the cell. If the laser modulation frequency remains constant while the gas flow velocity fluctuates, the excitation frequency is prone to deviating from the acoustic resonant frequency, resulting in a decrease in photoacoustic signal amplitude, reduced detection sensitivity, and poorer repeatability. Therefore, this embodiment dynamically adjusts the laser modulation frequency based on the change in gas flow velocity entering the resonant photoacoustic cell to ensure that the acoustic excitation process remains in a relatively stable resonant state.

[0068] In this embodiment, the gas flow velocity information entering the resonant photoacoustic cell is first acquired, and this flow velocity information is used as the basis for adjustment. Since changes in gas flow velocity alter the flow state, residence time, and local sound propagation conditions of the gas medium within the resonant photoacoustic cell, gas flow velocity can serve as an important parameter reflecting the current acoustic operating state. When an increase or decrease in gas flow velocity is detected, the laser modulation frequency is adjusted accordingly to ensure a high degree of consistency between the modulation excitation and the current acoustic resonance conditions.

[0069] Furthermore, the purpose of dynamically adjusting the laser modulation frequency is to reduce the adverse effects of airflow disturbance on the photoacoustic detection process. By synchronously changing the modulation frequency with the gas flow rate, the resonant peak shift or response attenuation caused by frequency mismatch can be avoided, allowing the gas under test to still obtain a relatively stable photoacoustic excitation effect under different flow rate conditions. This not only helps to improve the photoacoustic signal intensity but also enhances the data consistency between different detection times.

[0070] This embodiment uses a flow velocity sensing-frequency matching-resonance stabilization processing logic to enable the laser modulation process to adapt to changes in the flow state of the gas under test, thereby maintaining the stability of the acoustic resonance field in the resonant photoacoustic cell and providing a reliable foundation for subsequent photoacoustic signal acquisition and multi-component concentration decoupling.

[0071] S4: The modulated laser is incident into the resonant photoacoustic cell to excite the gas to be tested to generate a photoacoustic signal. The microphone captures the photoacoustic signal and performs signal processing to realize the decoupled calculation of the concentration of multi-component dissolved gases in the oil, and to perform transformer fault early warning based on the calculation results.

[0072] Specifically, the analyte gas entering the resonant photoacoustic cell contains multiple dissolved gas components, each with different absorption characteristics to laser light of different detection wavelengths. When the modulated laser light is incident on the resonant photoacoustic cell, the analyte gas absorbs the light energy and undergoes periodic thermoelastic expansion, thereby generating a corresponding photoacoustic signal. Since this photoacoustic signal is related to the absorption characteristics and concentration levels of the gas components, quantitative analysis of multi-component dissolved gases in oil can be achieved based on the collected photoacoustic response.

[0073] In this embodiment, the modulated laser is first introduced into a resonant photoacoustic cell, causing the gas to be tested to generate an enhanced acoustic response upon excitation. The photoacoustic signal is then captured by a microphone. Compared to directly detecting optical absorption intensity, photoacoustic methods can convert the gas's absorption of the laser into a more easily extracted acoustic signal, thereby improving the sensitivity of trace component detection and reducing interference from complex oil and gas backgrounds on the detection process.

[0074] Furthermore, the photoacoustic signals captured by the microphone are amplified, demodulated, and processed to extract the effective photoacoustic signal intensity corresponding to each detection wavelength. Since the absorption coefficients of different target dissolved gases are different at multiple detection wavelengths, the correspondence between the photoacoustic signal intensity and the concentration of each gas can be established based on the photoacoustic signal intensity corresponding to each detection wavelength and in combination with the pre-constructed absorption coefficient matrix. This allows for the calculation of the concentration values ​​of each target dissolved gas, thus achieving decoupled calculation of the concentration of multi-component dissolved gases.

[0075] Through the above processing, the originally superimposed absorption responses of multiple components can be separated, avoiding cross-interference between different gases under single-wavelength detection and improving the accuracy of component concentration identification. Based on the obtained multi-component dissolved gas concentration results, the changes in fault characteristic gases in transformer oil can be further determined, and transformer fault early warning can be provided accordingly. In other words, this step essentially constructs a processing chain of laser excitation—photoacoustic acquisition—signal demodulation—concentration decoupling—fault discrimination, enabling the spectral absorption information of the gas to be measured to be converted into fault characterization results that can be used for condition assessment.

[0076] Preferably, controlling the laser to sequentially output multiple detection wavelengths corresponding to the characteristic absorption wavelengths of different target dissolved gases can enhance the distinguishability between different components; while solving for the concentration of each component through a system of linear equations can maintain good quantitative analysis capability under conditions of multiple component coexistence. Through this process, highly sensitive detection and accurate decoupling of multiple dissolved gases in transformer oil can be achieved, providing a reliable basis for fault early warning.

[0077] The specific steps of S1 are as follows:

[0078] S1.1: The transformer oil sample is drawn by the circulating oil pump, and after impurities in the oil are removed by the oil sample filter located upstream of the circulating oil pump, it is sent into the degassing chamber through the sampling valve.

[0079] Specifically, during long-term operation, transformer oil may contain not only the dissolved gases to be detected, but also solid particles, suspended impurities, and localized deposits. If it is directly fed into the subsequent degassing step without pretreatment, it can easily cause unstable oil flow and even affect the subsequent release of dissolved gases. Therefore, in this embodiment, the transformer oil sample is first extracted, filtered, and directionally introduced to ensure that the oil sample entering the degassing stage has good cleanliness and flow stability, providing suitable conditions for subsequent gas removal.

[0080] In this embodiment, a circulating oil pump extracts transformer oil samples, continuously transporting them along a predetermined path. This circulating extraction method improves the continuity of oil sample flow, avoiding problems such as localized stagnation, stratification, or insufficient sampling that can occur during static sampling. This allows the oil sample to proceed more stably into subsequent steps. Simultaneously, circulating transport reduces the impact of single-sample fluctuations on the test results, improving the consistency of the entire testing process.

[0081] Furthermore, before the oil sample enters the circulating oil pump, it passes through an oil sample filter to remove impurities. This design aims to preemptively intercept particulate matter and impurities in the oil, reducing their interference with subsequent flow and degassing processes. On one hand, this reduces the risk of impurities clogging or adhering to the oil sample flow path; on the other hand, it helps prevent impurities from entering the degassing chamber and affecting the oil-gas contact state, thereby ensuring the stability and repeatability of subsequent degassing operations.

[0082] After filtration, the oil sample is sent into the degassing chamber via a sampling valve. This valve serves to direct the oil sample and switch the flow rate, ensuring the sample enters the degassing chamber along a pre-defined path and preventing flow fluctuations caused by disordered inflow. This sequential process of extraction, filtration, and introduction makes the oil sample entering the degassing chamber more suitable for subsequent dissolved gas removal in terms of cleanliness, fluidity, and controllability. Overall, this step achieves pretreatment and stable delivery of the original transformer oil sample, laying the foundation for subsequent negative pressure degassing and gas-liquid separation.

[0083] S1.2: In the degassing chamber, the dissolved gas in the oil sample is removed through the oil-gas separation membrane assembly by the negative pressure environment provided by the vacuum pump. The oil-gas separation membrane assembly is made of an oleophobic and gas-friendly semi-permeable membrane material.

[0084] Specifically, although the transformer oil sample after pre-filtration has removed larger particulate impurities, the target gas still exists in a dissolved state dispersed within the oil. If this dissolved gas is not released from the liquid phase first, the target gas for subsequent detection will be unclear and easily affected by the continuous interference from the liquid oil background. Therefore, this embodiment constructs a negative pressure environment in the degassing chamber, transforming the dissolved gas in the oil sample from a stable dissolved state to a free state that is easily precipitated, and further achieving gas-liquid separation through an oil-gas separation membrane assembly, thereby obtaining the target gas suitable for subsequent detection.

[0085] In this embodiment, negative pressure is used to reduce the partial pressure of the gas in the degassing chamber, disrupting the original dissolution equilibrium in the oil sample and causing characteristic gases such as hydrogen, methane, ethylene, and acetylene dissolved in the oil to gradually escape. Compared to natural release under normal pressure, negative pressure degassing improves gas evolution efficiency, shortens gas-liquid separation time, and reduces the problem of insufficient release of some low-concentration dissolved gases, thereby improving the representativeness and accuracy of subsequent detection results.

[0086] Furthermore, the precipitated gas is removed through the oil-gas separation membrane module. Because the oil-gas separation membrane module is made of an oleophobic and gas-loving semi-permeable membrane material, it has a high permeability to gases and a strong barrier effect on liquid oil. This allows the gas precipitated from the oil sample to preferentially pass through the membrane material into the subsequent detection process, while the liquid oil is confined to the original flow path, preventing oil from directly mixing with the gas to be tested. Through this selective permeation process, the impact of oil entrainment, oil mist contamination, and liquid phase residue on the stability of subsequent photoacoustic detection can be effectively reduced.

[0087] Meanwhile, using membrane separation for degassing avoids the uneven gas release problems caused by vigorous bubbling or strong disturbances, allowing dissolved gases to be extracted from the oil sample in a relatively stable and continuous manner. This helps maintain the continuity of the release process of the gas components to be analyzed, reduces the impact of instantaneous fluctuations on subsequent quantitative analysis, and improves the consistency and repeatability of the detection process.

[0088] The specific steps of S2 are as follows:

[0089] S2.1: The gas to be tested is introduced into the spiral coil after passing through the multi-channel selection valve. The heat of the annular cavity is used to preheat the gas to be tested, so that the temperature of the gas to be tested is consistent with the ambient temperature of the resonant photoacoustic cell.

[0090] Specifically, the gas to be tested, extracted from the degassing chamber, typically exhibits low temperature, significant temperature fluctuations, or incompatibility with the operating environment of the resonant photoacoustic cell before entering the subsequent detection area. Directly introducing this gas into the resonant photoacoustic cell can easily lead to condensation, localized thermal disturbances, or changes in gas properties due to temperature differences, thereby affecting the stability of the photoacoustic detection process and the consistency of the measurement results. Therefore, in this embodiment, the gas to be tested is preheated before entering the resonant photoacoustic cell, gradually bringing its temperature closer to the ambient temperature of the cell.

[0091] In this embodiment, the gas to be tested is introduced into the spiral coil after passing through a multi-channel selector valve. This introduction process allows the gas to flow stably along a predetermined gas path and enter the preheating path. Using a spiral coil design extends the flow path of the gas, increasing the heat exchange time between it and the external heat source; it also helps to mitigate the temperature abrupt changes caused by rapid gas flow, making the preheating process smoother and more thorough.

[0092] Furthermore, the spiral coil is attached to the outer wall of the annular cavity. As the gas to be tested flows through the spiral coil, it can exchange heat with the heat transferred from the annular cavity, thereby gradually increasing the gas temperature. Since this preheating process is a gradual heating along the conveying path, rather than instantaneous heating, it can avoid the problems of local overheating or uneven temperature distribution, allowing the gas to be tested to reach a relatively uniform thermal state before entering the resonant photoacoustic cell.

[0093] The preheating process described above ensures that the temperature of the gas being tested is consistent with the ambient temperature of the resonant photoacoustic cell. This reduces the temperature shock when the gas enters the detection area, minimizing the impact of condensation and temperature drift on the photoacoustic signal. Furthermore, once the gas temperature stabilizes, its flow state and acoustic response conditions become more consistent, which is beneficial for maintaining the stability and repeatability of the photoacoustic signal during subsequent detection processes.

[0094] S2.2: The multi-channel selection valve has at least three channels, wherein the first channel is connected to the degassing chamber for gas detection, the second channel is connected to the standard gas bag for self-calibration, and the third channel is connected to the exhaust end of the vacuum pump for gas path purging and self-circulation.

[0095] Specifically, to ensure that subsequent testing processes can perform both normal sample gas testing and calibration / purging, this embodiment employs a multi-channel switching method during the gas delivery process, allowing gases from different sources to enter the same testing flow according to their intended purpose. This avoids the process dispersion issues caused by separate gas paths for testing, calibration, and purging, while also improving switching efficiency and processing consistency between different operating conditions.

[0096] In this embodiment, the first channel is connected to the degassing chamber and is used to guide the degassed gas to be tested into subsequent detection steps. This channel corresponds to normal detection conditions, allowing dissolved gases from the transformer oil sample to enter the subsequent preheating, filtration, and photoacoustic detection processes along a predetermined path, thereby completing the analysis of multi-component dissolved gases in the oil.

[0097] The second channel connects to a standard gas bag for self-calibration. By introducing a standard gas of known concentration, the detection results can be compared to a benchmark to determine if there is a response deviation in the current detection process, thereby improving the accuracy and comparability of subsequent sample gas detection results.

[0098] The third channel connects to the exhaust end of the vacuum pump and is used for gas path purging and self-circulation. After completing a test or calibration, residual gas can be discharged through this channel, reducing the memory effect and cross-interference of the previous gas composition on the next test, and ensuring that the gas entering the subsequent testing process maintains a higher purity.

[0099] S2.3: The preheated gas to be tested is fed into the resonant photoacoustic cell through a microporous membrane structure set at the inlet and outlet of the resonant photoacoustic cell. The microporous membrane structure is made of an oleophobic and anaerobic material, and its porosity gradually increases from the inlet end to the center end of the resonant photoacoustic cell along the gas flow direction, so as to isolate the oil mist carried during the degassing process and to make the gas entering the resonant photoacoustic cell diffuse uniformly.

[0100] Specifically, the gas to be tested may still contain trace amounts of oil mist during degassing and transportation, and the airflow distribution may also exhibit localized deviations or uneven velocity. If it directly enters the resonant photoacoustic cell, it can easily contaminate the detection space and affect the uniformity of gas distribution within the cell, thereby interfering with the stable acquisition of the photoacoustic signal. Therefore, in this embodiment, the gas to be tested is first filtered and diffused homogenized using a microporous membrane structure before entering the resonant photoacoustic cell.

[0101] In this embodiment, the microporous membrane structure is made of an oleophobic and anaerobic material, which allows gas to pass through smoothly while blocking the oil mist entrained during the degassing process. This reduces the possibility of oil mist entering the resonant photoacoustic cell and minimizes the impact of oil residue on subsequent detection results.

[0102] Furthermore, the porosity of the microporous membrane structure gradually increases along the gas flow direction. This gradual pore distribution allows the gas to be measured to gradually transition from a relatively concentrated flow pattern to a more dispersed and uniform flow pattern before entering the resonant photoacoustic cell, preventing localized high-speed airflow from directly impacting the detection area. This improves the consistency of gas distribution within the cell and reduces detection fluctuations caused by uneven airflow.

[0103] Meanwhile, by setting microporous membrane structures at the air inlet and outlet, the gas to be tested can maintain a relatively stable flow state during its entry into and exit from the resonant photoacoustic cell, thereby improving the overall airflow environment inside the cell and enhancing the repeatability and stability of the photoacoustic detection process.

[0104] Furthermore, the microporous membrane structure is made of polyethersulfone material, with a membrane thickness of 0.5 mm and a pore size range of 1-10 μm. Its porosity gradually increases from 20% at the inlet end to 50% near the center end of the resonant photoacoustic cell.

[0105] Specifically, polyethersulfone (PES) materials possess good temperature resistance, film-forming stability, and gas permeability, enabling them to balance oil mist blocking and gas transport efficiency during the passage of the test gas. Therefore, they are suitable as materials for the aforementioned microporous membrane structures. By employing PES materials, the consistency of the filtration and homogenization processes of the test gas can be improved while ensuring the strength and stability of the membrane structure.

[0106] In this embodiment, a membrane thickness of 0.5 mm is set to achieve a balance between filtration efficiency and airflow resistance. If the membrane thickness is too small, the oil mist isolation capability and structural stability may be insufficient; if the membrane thickness is too large, it will increase the gas flow resistance, which is not conducive to the smooth entry of the gas to be tested into the resonant photoacoustic cell. Therefore, using a membrane thickness of 0.5 mm is beneficial to maintain good filtration performance while avoiding excessive impact on gas delivery.

[0107] Furthermore, setting the pore size range to 1-10 μm allows the microporous membrane structure to effectively block tiny oil mist particles entrained during degassing, while ensuring the target gas can pass through smoothly. This reduces the risk of oil mist contaminating the detection environment after entering the resonant photoacoustic cell and also helps maintain the stability of the subsequent photoacoustic detection process.

[0108] Meanwhile, the porosity gradually increases from 20% at the inlet to 50% near the center of the resonant photoacoustic cell along the gas flow direction, allowing the gas to be tested to transition from initial flow restriction and filtration to gradual diffusion and homogenization as it passes through the membrane structure. The lower porosity at the inlet helps to weaken local high-speed airflow and enhance oil mist interception; the higher porosity near the center of the resonant photoacoustic cell helps to reduce subsequent flow resistance, allowing the gas to diffuse into the resonant photoacoustic cell in a more uniform manner.

[0109] Furthermore, the annular cavity surrounds the laser and the resonant photoacoustic cell, and a temperature gradient field with gradually decreasing temperature from the laser end to the resonant photoacoustic cell end is formed inside it. This includes: providing guide fins on the inner wall of the annular cavity, the distribution density of the guide fins increasing along the direction from the laser end to the resonant photoacoustic cell end, and regulating the heat conduction efficiency of the corresponding region by the difference in the distribution density of the guide fins in different regions, so that a temperature gradient field with gradually decreasing temperature from the laser end to the resonant photoacoustic cell end is formed inside the annular cavity.

[0110] Specifically, before the gas to be tested enters the resonant photoacoustic cell, in addition to preheating, it is also necessary to coordinate and control the thermal distribution between the laser region and the resonant photoacoustic cell region. If the temperature field distribution around both is unreasonable, it can easily lead to fluctuations in the laser output state or cause additional thermal disturbances to the gas entering the vicinity of the resonant photoacoustic cell, thereby affecting the stability of subsequent photoacoustic detection. Therefore, this embodiment forms a temperature gradient field that gradually decreases from the laser end to the resonant photoacoustic cell end within the annular cavity, making the thermal environment at different locations transition more smoothly, thus balancing the operational stability of both the laser excitation end and the detection end.

[0111] In this embodiment, the annular cavity surrounds the laser and the resonant photoacoustic cell, placing their peripheries within a continuous heat conduction path. This enclosure allows heat to be transferred gradually along a predetermined direction, rather than accumulating disorderly in localized areas, thus providing a basis for establishing a directional temperature gradient. Consequently, when the gas under test is preheated by the spiral coil, it can also achieve more stable heat exchange thanks to this continuous thermal environment.

[0112] Furthermore, guide fins are arranged on the inner wall of the annular cavity, and the distribution density of the guide fins increases gradually from the laser end to the resonant photoacoustic cell end. This distribution method allows for targeted adjustment of heat conduction efficiency in different regions: near the laser end, where the guide fins are relatively sparsely distributed, heat is more easily retained, forming a relatively high-temperature region; while near the resonant photoacoustic cell end, where the guide fins are denser, local heat conduction and diffusion capabilities are enhanced, causing the temperature in this region to gradually decrease. This results in a continuously varying temperature gradient field from high to low, rather than an abrupt temperature distribution.

[0113] The aforementioned temperature gradient field reduces the thermal shock caused by direct heat transfer from the laser to the vicinity of the resonant photoacoustic cell. It also allows the gas under test to experience a smoother thermal transition as it flows towards the cell. This helps reduce the impact of localized hot spots, temperature drift, and thermal inhomogeneity on photoacoustic detection, resulting in a more stable temperature state for the gas under test when it enters the detection area.

[0114] The specific steps for S3 are as follows:

[0115] S3.1: The gas flow rate entering the resonant photoacoustic cell is adjusted and obtained by a mass flow controller located upstream of the air inlet of the resonant photoacoustic cell.

[0116] Specifically, the flow velocity of the gas to be tested before entering the resonant photoacoustic cell directly affects its residence time, distribution, and the stability of the subsequent photoacoustic response. If the gas flow velocity is too high, it can easily lead to increased airflow disturbance within the cell, affecting the stability of the acoustic field; if the gas flow velocity is too low, it may reduce detection efficiency and hinder timely gas replenishment. Therefore, in this embodiment, the flow velocity of the gas to be tested is adjusted and acquired before entering the resonant photoacoustic cell to ensure that the gas entering the detection area is in a suitable and controllable flow state.

[0117] In this embodiment, by adjusting the gas flow rate entering the resonant photoacoustic cell, the gas to be tested can be stably introduced at a preset rate, avoiding inconsistencies in detection conditions caused by excessive flow rate fluctuations. This ensures that the gas to be tested entering the resonant photoacoustic cell from different batches and at different stages maintains a high degree of consistency in flow conditions, thereby improving the repeatability and comparability of subsequent photoacoustic detection results.

[0118] Furthermore, acquiring the corresponding flow rate information while adjusting the gas flow rate provides a basis for the dynamic matching of the subsequent laser modulation frequency. In other words, this step not only achieves stable control of the input state of the gas under test, but also provides a prerequisite for maintaining the stability of the acoustic resonant field within the photoacoustic cell based on changes in gas flow rate.

[0119] S3.2: The main controller is connected to the mass flow controller and the optical chopper respectively. Based on the gas flow rate signal fed back by the mass flow controller, the modulation frequency of the optical chopper is adjusted in real time to keep the laser modulation frequency dynamically matched with the gas flow rate and maintain the stability of the acoustic resonance field in the resonant photoacoustic cell.

[0120] Specifically, the gas to be tested entering the resonant photoacoustic cell does not maintain a constant flow rate; its velocity may fluctuate during the detection process due to factors such as pre-gas degassing, transport, and gas path switching. If the laser modulation frequency remains fixed, changes in gas flow rate can easily lead to a shift between the excitation conditions and the current acoustic resonance state, resulting in weakened photoacoustic response, signal amplitude fluctuations, and decreased repeatability of the detection results. Therefore, this embodiment adjusts the laser modulation frequency in real time according to changes in gas flow rate to maintain a dynamic matching relationship between the two.

[0121] In this embodiment, the acquired gas flow velocity signal is used as the basis for frequency modulation. When the gas flow velocity increases, the gas renewal rate and local acoustic field conditions in the resonant photoacoustic cell will change accordingly. At this time, the laser modulation frequency is adjusted synchronously to avoid the excitation frequency deviating from the current effective resonance range. When the gas flow velocity decreases, the modulation frequency is reduced or corrected accordingly so that the gas under test can still maintain a good photoacoustic excitation effect under the new flow state. Through this adaptive frequency adjustment with changes in flow velocity, the detection process under different flow conditions can maintain a relatively consistent acoustic response.

[0122] Furthermore, the dynamic matching does not simply maintain a fixed proportional relationship, but rather aims to maintain the stability of the acoustic resonant field within the photoacoustic cell by continuously correcting the modulation frequency. This reduces the resonant peak shift and response attenuation caused by flow velocity disturbances, ensuring that the gas under test receives relatively stable excitation conditions throughout the detection cycle, thereby improving the stability and effectiveness of photoacoustic signal acquisition.

[0123] The specific steps of S3.2 are as follows:

[0124] S3.2.1: The optical chopper is positioned between the laser and the fiber collimator, and its rotation is driven by a motor to achieve periodic modulation of the laser output from the laser.

[0125] Specifically, to enable a detectable periodic photoacoustic response to be generated after laser irradiation of the gas under test, the continuous laser output from the laser needs to be converted into modulated light with certain frequency characteristics. If the gas under test is directly irradiated with an unmodulated continuous laser, it is difficult for the gas to form a stable periodic pressure wave after absorbing the light energy, resulting in poor extraction of the subsequent photoacoustic signal. Therefore, in this embodiment, the laser is periodically modulated in the laser transmission path so that the laser incident on the resonant photoacoustic cell has a clear modulation rhythm.

[0126] In this embodiment, an optical chopper is positioned between the laser and the fiber collimator, allowing the laser to undergo modulation before entering collimated transmission. This ensures that the beam output from the fiber collimator already possesses periodic variation characteristics, thus providing a stable modulated light source for subsequent focused incidence and gas excitation. Compared to modulation in subsequent optical paths, this method helps maintain consistency in the temporal characteristics of the transmitted beam, reducing additional interference from later optical paths on the modulation rhythm.

[0127] Furthermore, by driving the rotation with a motor, the optical chopper can periodically block and allow the laser at a set rotation speed, thereby forming a pulsed or periodically modulated laser. This modulation method essentially converts continuous light energy into an excitation signal distributed according to a time period, causing the gas under test to undergo thermoelastic expansion and contraction at a corresponding frequency after absorbing the light energy, thus forming a photoacoustic signal that can be captured by a microphone.

[0128] Meanwhile, the use of a rotary periodic modulation method facilitates subsequent adjustment of the modulation frequency based on changes in gas flow rate. In other words, changes in the motor drive speed can be directly reflected in changes in the laser modulation frequency, thus giving the laser excitation rhythm good adjustability and providing a foundation for achieving dynamic matching between the laser modulation frequency and gas flow rate.

[0129] S3.2.2: The main controller adjusts the speed of the motor according to the gas flow rate signal fed back by the mass flow controller to change the modulation frequency of the optical chopper, so that the modulation frequency matches the acoustic resonance frequency caused by the change in gas flow rate in the resonant photoacoustic cell.

[0130] Specifically, after the gas to be tested enters the resonant photoacoustic cell, its flow velocity changes, causing changes in the acoustic field conditions within the cell and further affecting the acoustic resonant frequency. If the modulation frequency of the optical chopper remains unchanged, a mismatch between the laser modulation frequency and the current acoustic resonant frequency can easily occur, leading to a decrease in the photoacoustic signal amplitude and a reduction in detection sensitivity. Therefore, in this embodiment, the motor speed is dynamically adjusted according to the change in the gas flow velocity signal to synchronously change the modulation frequency of the optical chopper, ensuring that the laser modulation process always matches the current acoustic resonant state.

[0131] In this embodiment, the motor speed is correlated with the modulation frequency of the optical chopper. Therefore, by adjusting the motor speed, the laser modulation rhythm can be quickly corrected. When the gas flow rate increases, the modulation frequency is adjusted accordingly to adapt to the updated acoustic resonance conditions; when the gas flow rate decreases, the modulation frequency is adjusted back accordingly so that the gas under test can still obtain a strong photoacoustic excitation response under the new flow conditions.

[0132] By using the above method, the laser excitation frequency can be continuously aligned with the effective resonant range within the photoacoustic cell, reducing the resonant offset and response attenuation caused by flow velocity fluctuations, and ensuring the stability and consistency of the photoacoustic signal acquisition process.

[0133] The specific steps for S4 are as follows:

[0134] S4.1: The laser output from the laser is modulated by an optical chopper, collimated by an optical fiber collimator to output a parallel beam, and then focused by a focusing lens and incident into the resonant photoacoustic cell. A resonator is provided inside the resonant photoacoustic cell. The gas to be tested is excited by the laser in the resonator to generate a photoacoustic signal, which is captured by a microphone provided on the side wall of the resonant photoacoustic cell.

[0135] Specifically, to ensure that the absorption of laser energy by the gas under test can be effectively converted into a detectable acoustic signal, this embodiment first shapes the laser propagation state before introducing it into the resonant photoacoustic cell to excite the gas under test. If the laser is directly incident without modulation, collimation, and focusing, problems such as unstable incident energy distribution, dispersed effective area, and insufficient excitation efficiency are likely to occur, which is not conducive to the stable acquisition of subsequent photoacoustic signals. Therefore, this embodiment improves the effective excitation capability of the laser on the gas under test by using a modulation-collimation-focusing-incident method.

[0136] In this embodiment, the laser output from the laser is first modulated by an optical chopper, converting continuous light into modulated light with periodic variation characteristics; then, a parallel beam is output through an optical fiber collimator to reduce the impact of beam divergence on the transmission process; finally, it is further focused by a focusing lens and incident into a resonant photoacoustic cell, allowing the laser energy to act more concentratedly on the gas to be tested. Through the above processing, the absorption efficiency of the gas to be tested by the laser can be improved, and the subsequent photoacoustic excitation effect can be enhanced.

[0137] Furthermore, after the gas under test is excited by a modulated laser within the resonator, it undergoes corresponding thermoelastic expansion and contraction due to periodic absorption of light energy, thereby generating a photoacoustic signal. Since the resonator can enhance the acoustic response at a specific frequency, the photoacoustic signal generated by the gas under test can be made more obvious, which is beneficial to improving the detectability of weak signals.

[0138] Once the photoacoustic signal is generated, it is captured by a microphone positioned on the side wall of the resonant photoacoustic cell. Placing the microphone on the side wall facilitates the reception of the sound pressure change signal generated within the cell and extracts the acoustic response of the gas under test to laser absorption, providing the original signal for subsequent amplification, demodulation, and concentration decoupling calculations.

[0139] S4.2: The photoacoustic signal captured by the microphone is sequentially amplified by a preamplifier, demodulated by a lock-in amplifier, and acquired by a data acquisition card to achieve decoupled calculation of the concentration of multi-component dissolved gases in the oil, and to provide transformer fault early warning based on the calculation results.

[0140] Specifically, the photoacoustic signals captured by microphones are typically weak in amplitude and easily superimposed with environmental noise and airflow disturbances. If used directly for concentration calculations, this can easily affect the accuracy of the results. Therefore, in this embodiment, the photoacoustic signals are first amplified, demodulated, and processed to extract the effective photoacoustic response corresponding to the modulated laser, providing a reliable data foundation for subsequent decoupled calculations of multi-component concentrations.

[0141] In this embodiment, the photoacoustic signal captured by the microphone is first pre-amplified to improve the identifiability of the weak acoustic signal; then, phase-locked demodulation is performed to extract the effective response component consistent with the laser modulation frequency from the mixed signal, suppressing irrelevant noise interference; finally, the demodulated signal is acquired to obtain the photoacoustic signal intensity values ​​corresponding to each detection wavelength. Through the above processing, the original photoacoustic signal can be converted into electrical signal data suitable for subsequent quantitative analysis.

[0142] Furthermore, since different target dissolved gases have different absorption coefficients at different detection wavelengths, a correspondence between the photoacoustic signal intensity at each detection wavelength and the concentration of each component can be established based on the absorption coefficient matrix. The superimposed responses of multiple components can then be decoupled and solved to obtain the concentration results of multiple dissolved gases in the oil. This reduces cross-interference between components and improves the accuracy of multi-component quantitative analysis.

[0143] After obtaining the concentration of each target dissolved gas, the transformer's operating status can be further judged based on the type, concentration level, and changes of the fault characteristic gas, and fault warnings can be issued accordingly.

[0144] The specific steps of S4.2 are as follows:

[0145] S4.2.1: Control the laser to sequentially output multiple detection wavelengths corresponding to the characteristic absorption wavelengths of various target dissolved gases, and obtain the photoacoustic signal intensity at each detection wavelength.

[0146] Specifically, the absorption of laser light by various target dissolved gases in transformer oil is not uniform, with different gases typically corresponding to different characteristic absorption wavelengths. If only a single detection wavelength is used for measurement, overlapping responses from multiple gases can easily occur, making it difficult to accurately distinguish the contributions of each component. Therefore, this embodiment controls the laser to sequentially output laser light at multiple detection wavelengths, causing each target dissolved gas to generate a response within its corresponding absorption band, thus providing a basis for subsequent decoupling of multi-component concentrations.

[0147] In this embodiment, multiple detection wavelengths correspond to the characteristic absorption wavelengths of different target dissolved gases. After the laser is sequentially switched to each detection wavelength, the gas to be tested absorbs the laser at the corresponding wavelength in the resonant photoacoustic cell, generating a corresponding photoacoustic signal. By acquiring the photoacoustic signal intensity at each detection wavelength, a signal set reflecting the absorption characteristics of multiple target gases can be formed, rather than obtaining only a single mixed response.

[0148] Furthermore, since different target dissolved gases exhibit varying response intensities at different detection wavelengths, the intensity of the acquired photoacoustic signals at each wavelength can characterize the differences in the contribution of different components to the overall signal. This provides input for subsequently establishing a system of linear equations based on the absorption coefficient matrix and solving for the concentrations of each component, thereby improving the accuracy of quantitative analysis under conditions of multi-component coexistence.

[0149] S4.2.2: Construct an absorption coefficient matrix, wherein the elements in the absorption coefficient matrix are the absorption coefficients of each target dissolved gas at each detection wavelength.

[0150] Specifically, after acquiring the photoacoustic signal intensity at each detection wavelength, it is necessary to establish a quantitative correspondence between each target dissolved gas and each detection wavelength in order to separate and solve the multi-component mixed signal. Since the absorption capacity of different target dissolved gases varies at different detection wavelengths, this embodiment constructs an absorption coefficient matrix to uniformly characterize this difference, providing a basis for subsequent concentration decoupling calculations.

[0151] In this embodiment, each element in the absorption coefficient matrix represents the absorption coefficient of a target dissolved gas to laser light at a specific detection wavelength. By arranging the absorption coefficients according to the correspondence between gas type and detection wavelength, a matrix structure reflecting the absorption characteristics of multiple target gases under multiple wavelength conditions can be formed. In this way, the originally dispersed absorption characteristics of each component can be transformed into a set of parameters that are easy to calculate and process.

[0152] Furthermore, the purpose of constructing the absorption coefficient matrix is ​​to establish a relationship between the photoacoustic signal intensity acquired at each detection wavelength and the absorption characteristics of each target dissolved gas. Since multiple gases may simultaneously produce responses at the same detection wavelength, and the response degree of the same gas varies at different detection wavelengths, it is difficult to accurately distinguish the contributions of each component based solely on a single wavelength or a single response intensity. By introducing the absorption coefficient matrix, the multi-wavelength photoacoustic response can be mapped to a multi-component concentration solution problem, thereby improving the accuracy of mixed gas analysis.

[0153] Meanwhile, the absorption coefficient matrix essentially forms the parameter basis for subsequently establishing a system of linear equations. That is, the photoacoustic signal intensity at each detection wavelength can be considered as the output quantity, the concentration of each target dissolved gas can be considered as the quantity to be determined, and the absorption coefficient matrix is ​​used to characterize the linear correlation between the two. Using this matrix, the cross-influence of different target gases at different wavelengths can be incorporated into a unified calculation framework, reducing mutual interference under conditions of multi-component coexistence.

[0154] S4.2.3: Based on the photoacoustic signal intensity at each detection wavelength and the absorption coefficient matrix, establish a set of linear equations between the photoacoustic signal intensity vector and the gas concentration vector. By solving the set of linear equations, the concentration values ​​of each target dissolved gas are obtained.

[0155] Specifically, after acquiring the photoacoustic signal intensity at each detection wavelength and constructing the absorption coefficient matrix, it is necessary to further convert the multi-wavelength detection results into quantitative concentration results of each target dissolved gas. Since the photoacoustic signal at each detection wavelength is usually formed by the combined contribution of multiple target gases, it is difficult to directly obtain the actual concentration of the corresponding gas by observing the signal intensity at a single wavelength. Therefore, in this embodiment, the photoacoustic signal intensity at each detection wavelength is taken as the observed quantity, and the concentration of each target dissolved gas is taken as the unknown quantity, and a linear equation system between the two is established by combining the absorption coefficient matrix.

[0156] In this embodiment, the photoacoustic signal intensity vector is used to characterize the actual response results at each detection wavelength, the gas concentration vector is used to characterize the concentration distribution of each target dissolved gas, and the absorption coefficient matrix is ​​used to characterize the response weight relationship of different target gases at different detection wavelengths. By establishing the aforementioned linear equation set, the mixed response of multiple gases superimposed under multiple wavelength conditions can be transformed into a computable concentration separation problem.

[0157] Furthermore, by solving the linear equations, the concentration values ​​of each target dissolved gas can be obtained, thereby achieving concentration decoupling of multi-component dissolved gases. This separates the contributions of multiple gases that were originally coupled in the same photoacoustic response, reduces cross-interference between different components, and improves the accuracy of concentration identification results.

[0158] Furthermore, the method also includes installing a multi-channel selection valve in the gas path between the degassing chamber and the resonant photoacoustic cell, and controlling the multi-channel selection valve to perform the following operations sequentially according to a preset cycle via the main controller:

[0159] Switch to the calibration path, introduce standard gas of known concentration from the standard gas bag into the resonant photoacoustic cell for photoacoustic detection, compare the detection result with the known concentration of the standard gas, and complete the system self-calibration;

[0160] Switch to the purging path and use a vacuum pump to evacuate and purge the resonant photoacoustic cell and gas pipeline to remove residual gas.

[0161] Switch to the detection path and introduce the gas to be tested extracted from the degassing chamber into the resonant photoacoustic cell for photoacoustic detection.

[0162] Specifically, to ensure the accuracy, continuity, and repeatability of subsequent photoacoustic detection results, this embodiment introduces a calibration path and a purging path in addition to the normal detection process. This ensures that the gas to be tested is in a calibrable and cleanable gas path state before and after detection. If only the gas to be tested is continuously introduced for detection, the stability of the detection results is easily reduced due to detection drift, residual gas retention, or cross-influence between batches of gas. Therefore, this embodiment achieves a cyclical process of calibration-purging-detection by periodically switching different paths.

[0163] In this embodiment, when switching to the calibration path, a standard gas of known concentration from the standard gas bag is introduced into the resonant photoacoustic cell for photoacoustic detection, and the obtained detection result is compared with the known concentration of the standard gas. This method allows for the determination of whether there is a response deviation in the current detection link, and enables self-calibration, thereby improving the accuracy and comparability of subsequent gas detection results.

[0164] Furthermore, when switching to the purge path, a vacuum pump is used to evacuate and purge the resonant photoacoustic cell and related gas paths, removing any residual gas from the previous round of testing or calibration. This effectively reduces the memory effect and cross-interference of residual gas on the next round of testing, avoids cross-contamination between different gas samples, and thus maintains the cleanliness of the testing environment.

[0165] After calibration and purging, the system switches to the detection path, and the gas to be tested extracted from the degassing chamber is introduced into the resonant photoacoustic cell for formal testing. Since the detection link has completed reference calibration and residual gas has been removed at this time, the gas to be tested entering the detection step can complete photoacoustic detection under relatively stable and clean conditions, which is beneficial to improving the authenticity and repeatability of the detection results.

[0166] Example 2

[0167] Please see Figure 2 This invention provides an embodiment of a transformer oil quality analysis photoacoustic spectroscopy detection device. The device is arranged vertically and, from bottom to top, includes an oil degassing module 10, a photoacoustic detection gas path module, and a light source optical path module. A signal processing module 40 is located on the right side of the device, and a constant temperature control unit is located around the periphery. This vertical, layered arrangement allows the gas to be tested to be transported from bottom to top, facilitating degassing, preheating, and entry into the resonant photoacoustic cell 32 for detection.

[0168] The oil degassing module 10 includes a degassing chamber 11, a circulating oil pump 12, an oil-gas separation membrane assembly 13, a sampling valve 14, and an oil sample filter 15. The circulating oil pump 12 is located at the lower inlet of the degassing chamber 11 and is used to extract transformer oil samples and drive the oil samples to circulate. The oil sample filter 15 is located upstream of the circulating oil pump 12 and is used to filter out impurities in the oil sample. The sampling valve 14 is located on the side wall of the degassing chamber 11 and is used to control the entry and exit of the oil sample. The oil-gas separation membrane assembly 13 is located inside the degassing chamber 11 and is used to remove dissolved gases from the transformer oil from the oil sample under negative pressure conditions.

[0169] Please see Figure 3 The schematic diagram of the connection structure of the multi-channel selection valve provided in this embodiment of the invention shows that the gas output end of the degassing chamber 11 is connected to the photoacoustic detection gas path module through the multi-channel selection valve 16. The multi-channel selection valve 16 is connected to the branch containing the degassing chamber 11, the standard gas bag 17, and the vacuum pump 38, respectively, and is used to switch between the gas to be tested, the standard gas, and the vacuum / self-circulation channel, thereby realizing the device's detection, self-calibration, and cyclic operation functions. The connecting pipeline between the multi-channel selection valve 16 and the resonant photoacoustic cell 32 is preferably constructed as a spiral coil 18. The spiral coil 18 is attached to the outer wall of the annular cavity structure 51 to preheat the gas before it enters the resonant photoacoustic cell 32 using the heat from the constant temperature control unit, thereby reducing the temperature difference between the gas to be tested and the photoacoustic detection area and reducing condensation interference.

[0170] Please see Figure 4 The schematic diagram of the light source optical path module and the constant temperature control unit provided in this embodiment of the invention shows that the light source optical path module is located at the top of the device and includes a tunable semiconductor laser 22, an optical chopper 21, an optical fiber collimator 23, and a focusing lens 24. The laser emitted by the tunable semiconductor laser 22 is first modulated by the optical chopper 21, then collimated by the optical fiber collimator 23, and finally focused by the focusing lens 24 and injected into the resonant photoacoustic cell 32 along the optical axis to excite the gas to be measured to generate a photoacoustic signal.

[0171] The photoacoustic detection gas path module is located in the middle of the device and includes an air inlet 31, a resonant photoacoustic cell 32, a microphone 33, an air outlet 34, a microporous membrane structure 35, a mass flow controller 36, a solenoid valve 37, and a vacuum pump 38. The air inlet 31 is connected to the output of a multi-channel selector valve 16 and is used to introduce the gas to be tested into the resonant photoacoustic cell 32. The mass flow controller 36 is located upstream of the air inlet 31 and is used to precisely control the gas flow rate entering the resonant photoacoustic cell 32. Please refer to [link to relevant documentation]. Figure 5The schematic diagram of the microporous membrane structure provided in this embodiment of the invention shows that microporous membrane structures 35 are provided at both the air inlet 31 and the air outlet 34 to isolate oil mist and improve the uniformity of gas diffusion; a microphone 33 is installed on the side wall of the resonant photoacoustic cell 32 to collect the photoacoustic signal generated after the gas to be tested is irradiated by a modulated laser; a solenoid valve 37 and a vacuum pump 38 are provided downstream of the air outlet 34 to realize the opening and closing control of the gas path and negative pressure suction.

[0172] The constant temperature control unit is configured as a ring-shaped cavity structure 51, which surrounds the light source optical path module and the resonant photoacoustic cell 32, and its inner wall is provided with guide fins 52. Please refer to [link / reference]. Figure 6 The schematic diagram of the flow-guiding fin structure of the annular cavity provided in this embodiment of the invention shows that the annular cavity structure 51 can form a predetermined temperature gradient field between the high-temperature end where the tunable semiconductor laser 22 is located and the low-temperature end where the resonant photoacoustic cell 32 is located. The flow-guiding fin 52 is used to optimize heat transfer and temperature field distribution in the cavity, so as to improve the temperature stability of the laser and the resonant photoacoustic cell 32 during long-term operation.

[0173] The signal processing module 40 is electrically connected to the main controller 41 via the optical chopper 21, the mass flow controller 36, and the microphone 33. The main controller 41 coordinates the laser modulation frequency with the gas flow rate; the signal processing module 40 amplifies, demodulates, acquires, and analyzes the photoacoustic signal output from the microphone 33 to obtain information on the composition and concentration of dissolved gases in the transformer oil, and uses this information to provide fault warnings.

[0174] In this embodiment, the working process of the device can be summarized as follows: the circulating oil pump 12 drives the oil sample into the degassing chamber 11, and the dissolved gas in the oil sample is extracted by the oil-gas separation membrane assembly 13 and then passes through the multi-channel selection valve 16, the spiral coil 18, the mass flow controller 36 and the air inlet 31 in sequence before entering the resonant photoacoustic cell 32; at the same time, the laser output by the tunable semiconductor laser 22 is modulated by the optical chopper 21, collimated by the fiber collimator 23 and focused by the focusing lens 24 and then incident into the resonant photoacoustic cell 32, exciting the gas to be tested to generate a photoacoustic signal; the microphone 33 collects the photoacoustic signal and sends it to the signal processing module 40 for processing, thereby completing the detection of the transformer oil quality status.

[0175] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A photoacoustic spectroscopy method for transformer oil quality analysis, characterized in that, The method includes: Transformer oil samples are drawn by circulating oil pumps and sent into the degassing chamber. Dissolved gases in the oil are removed by the oil-gas separation membrane assembly under negative pressure. The gas to be tested is preheated by a spiral coil attached to the outer wall of the annular cavity, filtered and homogenized by a microporous membrane structure set at the air inlet of the resonant photoacoustic cell, and then sent into the resonant photoacoustic cell. The annular cavity covers the laser and the outside of the resonant photoacoustic cell, and a temperature gradient field with a gradually decreasing temperature from the laser end to the resonant photoacoustic cell end is formed inside it. The main controller dynamically adjusts the laser modulation frequency according to the gas flow rate entering the resonant photoacoustic cell to maintain the stability of the acoustic resonance field within the resonant photoacoustic cell. The modulated laser is incident into the resonant photoacoustic cell to excite the gas to be tested to generate a photoacoustic signal. The photoacoustic signal is captured by a microphone and processed to achieve decoupled calculation of the concentration of multi-component dissolved gases in the oil. Based on the calculation results, transformer fault early warning is provided.

2. The photoacoustic spectroscopy method for transformer oil quality analysis according to claim 1, characterized in that, The process of drawing transformer oil samples via a circulating oil pump and sending them into the degassing chamber, where dissolved gases are removed from the oil through an oil-gas separation membrane assembly under negative pressure, includes: Transformer oil samples are drawn by the circulating oil pump, and after impurities are removed by the oil sample filter located upstream of the circulating oil pump, the oil is sent into the degassing chamber through the sampling valve. Inside the degassing chamber, the dissolved gases in the oil sample are removed through the oil-gas separation membrane assembly by the negative pressure environment provided by the vacuum pump. The oil-gas separation membrane assembly is made of an oleophobic and gas-friendly semi-permeable membrane material.

3. The photoacoustic spectroscopy method for transformer oil quality analysis according to claim 1, characterized in that, The process involves preheating the extracted test gas through a spiral coil attached to the outer wall of the annular cavity, filtering and homogenizing it through a microporous membrane structure located at the inlet of the resonant photoacoustic cell, and then sending it into the resonant photoacoustic cell. This includes: The extracted gas to be tested is introduced into the spiral coil after passing through a multi-channel selection valve. The heat of the annular cavity is used to preheat the gas to be tested, so that the temperature of the gas to be tested is similar to the ambient temperature of the resonant photoacoustic cell. The multi-channel selection valve has at least three channels, wherein the first channel is connected to the degassing chamber for gas detection, the second channel is connected to the standard gas bag for self-calibration, and the third channel is connected to the exhaust end of the vacuum pump for gas path purging and self-circulation. The preheated gas to be tested is fed into the resonant photoacoustic cell through a microporous membrane structure located at the inlet and outlet of the resonant photoacoustic cell. The microporous membrane structure is made of an oleophobic and anaerobic material, and its porosity gradually increases from the inlet end to the center end of the resonant photoacoustic cell along the gas flow direction, so as to isolate the oil mist carried during the degassing process and to homogenize the diffusion of the gas entering the resonant photoacoustic cell.

4. The photoacoustic spectroscopy method for transformer oil quality analysis according to claim 3, characterized in that, The microporous membrane structure is made of polyethersulfone material, with a membrane thickness of 0.5 mm and a pore size range of 1-10 μm. Its porosity gradually increases from 20% at the inlet end to 50% near the center end of the resonant photoacoustic cell.

5. The photoacoustic spectroscopy method for transformer oil quality analysis according to claim 3, characterized in that, The annular cavity surrounds the laser and the resonant photoacoustic cell, and forms a temperature gradient field inside it that gradually decreases in temperature from the laser end to the resonant photoacoustic cell end, including: A flow-guiding fin is provided on the inner wall of the annular cavity. The distribution density of the flow-guiding fin increases along the direction from the laser end to the resonant photoacoustic cell end. The heat conduction efficiency of the corresponding region is controlled by the difference in the distribution density of the flow-guiding fin in different regions, so that a temperature gradient field with the temperature gradually decreasing from the laser end to the resonant photoacoustic cell end is formed inside the annular cavity.

6. The photoacoustic spectroscopy method for transformer oil quality analysis according to claim 1, characterized in that, The step of dynamically adjusting the laser modulation frequency according to the gas flow rate entering the resonant photoacoustic cell by the main controller to maintain the stability of the acoustic resonant field within the resonant photoacoustic cell includes: The gas flow rate entering the resonant photoacoustic cell is adjusted and obtained by a mass flow controller located upstream of the air inlet of the resonant photoacoustic cell; The main controller is connected to the mass flow controller and the optical chopper respectively. Based on the gas flow rate signal fed back by the mass flow controller, the modulation frequency of the optical chopper is adjusted in real time to keep the laser modulation frequency dynamically matched with the gas flow rate and maintain the stability of the acoustic resonance field in the resonant photoacoustic cell.

7. The photoacoustic spectroscopy method for transformer oil quality analysis according to claim 6, characterized in that, The real-time adjustment of the modulation frequency of the optical chopper to dynamically match the laser modulation frequency with the gas flow rate includes: The optical chopper is positioned between the laser and the fiber collimator, and its rotation is driven by a motor to achieve periodic modulation of the laser output from the laser. The main controller adjusts the speed of the motor according to the gas flow rate signal fed back by the mass flow controller to change the modulation frequency of the optical chopper, so that the modulation frequency matches the acoustic resonance frequency caused by the change in gas flow rate in the resonant photoacoustic cell.

8. The photoacoustic spectroscopy method for transformer oil quality analysis according to claim 1, characterized in that, The process involves incident a modulated laser into the resonant photoacoustic cell to excite the gas under test to generate a photoacoustic signal, which is then captured by a microphone and processed to achieve decoupled calculation of the concentration of multi-component dissolved gases in the oil. This includes: The laser output from the laser is modulated by an optical chopper, collimated by an optical fiber collimator to output a parallel beam, and then focused by a focusing lens and incident into the resonant photoacoustic cell. A resonator is provided inside the resonant photoacoustic cell. The gas to be tested is excited by the laser in the resonator to generate a photoacoustic signal, which is captured by a microphone located on the side wall of the resonant photoacoustic cell. The photoacoustic signal captured by the microphone is sequentially amplified by a preamplifier, demodulated by a lock-in amplifier, and acquired by a data acquisition card to achieve decoupled calculation of the concentration of multi-component dissolved gases in oil, and to provide early warning of transformer faults based on the calculation results.

9. The photoacoustic spectroscopy method for transformer oil quality analysis according to claim 8, characterized in that, The process of sequentially amplifying the photoacoustic signal captured by the microphone, demodulating it with a lock-in amplifier, and acquiring it with a data acquisition card to achieve decoupled calculation of the concentration of multi-component dissolved gases in the oil includes: The laser is controlled to sequentially output multiple detection wavelengths corresponding to the characteristic absorption wavelengths of various target dissolved gases, and the photoacoustic signal intensity at each detection wavelength is obtained. Construct an absorption coefficient matrix, wherein the elements of the absorption coefficient matrix are the absorption coefficients of each target dissolved gas at each detection wavelength; Based on the photoacoustic signal intensity at each detection wavelength and the absorption coefficient matrix, a set of linear equations is established between the photoacoustic signal intensity vector and the gas concentration vector. By solving the set of linear equations, the concentration values ​​of each target dissolved gas are obtained.

10. The photoacoustic spectroscopy method for transformer oil quality analysis according to claim 1, characterized in that, The method further includes: A multi-channel selection valve is installed in the gas path between the degassing chamber and the resonant photoacoustic cell. The main controller controls the multi-channel selection valve to perform the following operations sequentially according to a preset cycle: Switch to the calibration path, introduce standard gas of known concentration from the standard gas bag into the resonant photoacoustic cell for photoacoustic detection, compare the detection result with the known concentration of the standard gas, and complete the system self-calibration; Switch to the purging path and use a vacuum pump to evacuate and purge the resonant photoacoustic cell and gas pipeline to remove residual gas. Switch to the detection path and introduce the gas to be tested extracted from the degassing chamber into the resonant photoacoustic cell for photoacoustic detection.

Images