A predictive generator corrosion state monitoring system based on isotope tracing and multi-factor fusion analysis

Abstract

The application discloses a predictive generator corrosion state monitoring system based on isotope tracing and multi-factor fusion analysis, comprising a collection mechanism, a leakage tracing unit, a sensing unit and a comprehensive analysis and diagnosis unit. The leakage tracing unit is used for online measurement of deuterium / hydrogen isotope ratio of dissolved hydrogen in the cold water. The sensing unit is used for synchronous measurement of pH value, dissolved oxygen concentration and other water quality parameters of the water sample. The comprehensive analysis and diagnosis unit fuses the above measurement data, and through the built-in model, not only can the leakage hydrogen and the corrosion hydrogen be accurately distinguished, but also the real-time corrosion rate of the internal components of the generator can be quantitatively calculated, and the future corrosion trend can be predicted. The application solves the problems that the traditional monitoring method cannot trace the source of hydrogen and cannot quantitatively evaluate the corrosion state, can effectively avoid the misjudgment of the unit state, provides a decision basis for realizing the preventive maintenance of the generator, and improves the operation safety.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of equipment corrosion monitoring, in particular to a predictive generator corrosion state monitoring system based on isotope tracing and multi-factor fusion analysis. BACKGROUND

[0002] The safe and stable operation of a generator set is the core guarantee of the power system. With the continuous development of the power industry, the application of large hydrogen-cooled generators is becoming more and more widespread. The stator winding of the generator uses cold water for internal cooling, and the water quality of the cold water has a direct and important impact on the service life and safety of the copper components inside the generator. Ensuring the safe and stable operation of the generator is of key significance to maintaining the normal operation of the power system, reducing power outages and reducing economic losses. Therefore, the monitoring technology related to the water quality of the cold water is also continuously developing and improving.

[0003] In the prior art, in order to monitor the operating state of the generator, the online monitoring of parameters such as dissolved hydrogen, pH value, dissolved oxygen and conductivity in the cold water is generally adopted. This monitoring method can reflect the water quality of the cold water to a certain extent. Through the monitoring of dissolved hydrogen, it is expected to understand whether there is hydrogen leakage or corrosion related problems; monitoring the pH value can know whether the acidity or alkalinity of the cold water is in the appropriate range; the detection of dissolved oxygen can judge the influence of oxygen content in the water on the copper components; and the monitoring of conductivity helps to evaluate the ion concentration in the water and the like. These monitoring parameters provide a basis for judging the operating state of the generator from different aspects.

[0004] However, the existing monitoring technology has obvious defects. The traditional dissolved hydrogen monitoring cannot distinguish the source of hydrogen, and the detected hydrogen may come from the leakage of hydrogen cooling system cooling hydrogen or from the electrochemical corrosion of the cold water on the copper wire. This uncertainty often leads to misjudgment of the state of the unit or misses the early warning signal of early corrosion. Moreover, the existing system only lists the independent values of the pH, dissolved oxygen, conductivity and other water quality indicators, and fails to deeply fuse and correlate analyze these parameters with the real equipment corrosion state, and cannot quantitatively evaluate the real-time corrosion rate of the internal components of the generator under the synergistic action of multiple factors. At the same time, the traditional monitoring belongs to "after" or "in-process" monitoring, which can only reflect the current water quality deterioration or corrosion condition that has occurred, lacks the forward-looking prediction ability of the corrosion trend, and cannot provide effective decision guidance for the preventive maintenance and life management of the unit. SUMMARY

[0005] In order to solve the technical problems in the prior art, the present application provides a predictive generator corrosion state monitoring system based on isotope tracing and multi-factor fusion analysis.

[0006] The application provides a predictive generator corrosion state monitoring system based on isotope tracing and multi-factor fusion analysis, which adopts the following technical scheme:

[0007] A predictive generator corrosion state monitoring system based on isotope tracing and multi-factor fusion analysis is used for monitoring a hydrogen-cooled generator using deuterium isotope-labeled hydrogen as a cooling medium, and the generator corrosion state monitoring system comprises:

[0008] A collection mechanism in communication with the generator constant cooling water circuit forms a circulating passage for providing real-time constant cooling water samples for the system, and the collection mechanism comprises a collection pump for driving water sample circulation and a multi-parameter flow cell carrying a subsequent detection unit;

[0009] A leakage tracing unit in fluid communication with the multi-parameter flow cell is used for degassing from the real-time constant cooling water sample and on-line measurement of the deuterium / hydrogen isotope ratio of hydrogen, for tracing the composition of dissolved hydrogen, so as to quantitatively distinguish hydrogen from hydrogen-cooled system leakage and hydrogen generated by electrochemical corrosion of the internal components of the generator in contact with the constant cooling water;

[0010] A sensing unit installed in the multi-parameter flow cell is used for measuring the pH value, dissolved oxygen concentration and conductivity of the real-time constant cooling water sample, for evaluating water quality corrosion; and,

[0011] An integrated analysis and diagnosis unit electrically connected with the leakage tracing unit and the sensing unit is used for determining whether the generator hydrogen-cooled system has leakage and the leakage degree based on real-time data of the deuterium / hydrogen isotope ratio, and quantitatively calculating the real-time corrosion rate of the internal copper components of the generator based on the pH value, dissolved oxygen concentration, conductivity of the real-time constant cooling water sample and the hydrogen concentration generated by internal component corrosion analyzed by the leakage tracing unit according to the isotope ratio, and generating predictive information on future corrosion trends, to provide a decision basis for implementing preventive maintenance.

[0012] In some embodiments, the leakage tracing unit comprises a membrane degassing device, an isotope ratio mass spectrometer and a thermal conductivity detector, the inlet of the membrane degassing device is in communication with the multi-parameter flow cell, for transferring dissolved hydrogen in the real-time constant cooling water sample to a gas phase; the isotope ratio mass spectrometer is in communication with the gas outlet of the membrane degassing device, for measuring the deuterium / hydrogen isotope ratio of the gas phase hydrogen, and the thermal conductivity detector is installed between the membrane degassing device and the isotope ratio mass spectrometer, for measuring the total hydrogen concentration in the gas phase discharged by the membrane degassing device before isotope ratio measurement, to calculate the real-time total dissolved hydrogen concentration.

[0013] In some embodiments, the sensing unit further comprises a temperature-pressure sensor in communication with the multi-parameter flow cell for simultaneous measurement of the temperature and pressure of the real-time cooling water sample, and the comprehensive analysis and diagnosis unit receives the temperature and pressure data measured by the temperature-pressure sensor and uses them as parameters for correction or compensation of the calculation of the real-time corrosion rate.

[0014] In some embodiments, the inlet of the sampling pump is in communication with the cooling water tank of the generator through a sampling inlet line, and the outlet of the sampling pump is in communication with the multi-parameter flow cell through a sampling outlet line.

[0015] In some embodiments, the sensing unit comprises a pH sensor, a dissolved oxygen concentration sensor and a conductivity sensor for real-time measurement of the pH value, the dissolved oxygen concentration and the conductivity of the cooling water sample, respectively.

[0016] In some embodiments, the procedure for determining whether there is a leakage in the hydrogen cooling system of the generator by the comprehensive analysis and diagnosis unit comprises:

[0017] acquiring the real-time measured deuterium / hydrogen isotope ratio R measured ;

[0018] comparing the R measured with the baseline R natural representing the natural abundance isotope ratio of hydrogen generated by pure corrosion stored in the controller;

[0019] when the condition R measured > R natural + δ is met, it is determined that there is a hydrogen leakage, where δ is a preset judgment threshold.

[0020] In some embodiments, the comprehensive analysis and diagnosis unit quantitatively calculates the hydrogen concentration C corr generated by corrosion of the internal components by running the following isotope mass balance equation:

[0021] ,

[0022] where C H is the real-time total dissolved hydrogen concentration obtained by the thermal conductivity detector; R measured is the deuterium / hydrogen isotope ratio measured by the isotope ratio mass spectrometer in real time; R leak is the preset deuterium / hydrogen isotope ratio of deuterium-labeled hydrogen as the cooling medium; and R natural is the preset baseline representing the natural isotope ratio of hydrogen generated by corrosion.

[0023] In some embodiments, the integrated analysis and diagnostic unit is used to calculate the real-time corrosion rate V of the copper components inside the generator. corr The algorithm is based on a multi-parameter electrochemical corrosion model, whose input variables include at least the hydrogen production concentration C during corrosion. corr The dissolved oxygen concentration, pH value, and conductivity σ measured by the sensing unit;

[0024] The multi-parameter electrochemical corrosion model is represented by the following functional relationship:

[0025] ,

[0026] Where k is the comprehensive corrosion coefficient; f, g, h, j, e are empirical functions related to the concentration of hydrogen produced by corrosion, the concentration of dissolved oxygen, the pH value, the conductivity, and the temperature, respectively; and T is the real-time water sample temperature measured by the sensing unit.

[0027] In some embodiments, the comprehensive analysis and diagnostic unit generates predictive information about future corrosion trends by analyzing real-time corrosion rates V from multiple continuously collected historical data points. corr This is achieved by performing time series analysis on the data points, which includes at least calculating the slope of the corrosion rate trend and establishing an extrapolation prediction model.

[0028] In some embodiments, the specific implementation of calculating the slope of the corrosion rate change trend and establishing an extrapolation prediction model is as follows: N consecutive real-time corrosion rate data point sequences (t) within a preset time window are collected and stored. i V corr,i ), where i is a natural number between 1 and N;

[0029] A least squares linear regression analysis was performed on the data point sequence to obtain a slope m representing the linear change trend of corrosion rate with time;

[0030] Based on this slope m and the latest corrosion rate V corr,now By extrapolation formula:

[0031] V pred =m⋅Δt+V corr,now

[0032] To predict the corrosion rate V after a future time interval Δt pred Based on the comparison between the predicted value and the safety threshold, it provides forward-looking guidance for preventive maintenance.

[0033] In summary, this application includes at least one of the following beneficial technical effects:

[0034] 1. By using isotope tracing technology, the system can accurately quantify and distinguish hydrogen gas from hydrogen cold system leakage and hydrogen gas generated by internal component corrosion. This solves the fundamental problem of unknown hydrogen source in traditional monitoring technology, avoiding misjudgment of the operation state of the generator set.

[0035] 2. The system combines the corrosion hydrogen concentration obtained by isotope analysis with real-time measured pH value, dissolved oxygen concentration, conductivity and other water quality parameters for fusion analysis. Through the built-in electrochemical corrosion model, the real-time corrosion rate of the internal copper components of the generator is quantitatively calculated, and the corrosiveness of the water quality can be more comprehensively evaluated.

[0036] 3. The system can generate predictive information on future corrosion trends by time series analysis of historical real-time corrosion rate data. This provides direct decision-making basis for preventive maintenance, enabling the unit maintenance to change from passive "fault response" mode to active "state prediction" mode.

[0037] 4. Through early and accurate identification of leakage and corrosion risks and future trend prediction, the scheme helps to improve the safety and reliability of the generator operation, prolong the service life of the equipment, and reduce maintenance costs. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 is the overall structure schematic diagram of the predictive generator corrosion state monitoring system based on isotope tracing and multi-factor fusion analysis provided by an embodiment of the present application connected with the generator cooling system;

[0039] Figure 2 is Figure 1 the structure schematic diagram of the hydrogen cooling system and the fixed cooling water system in the generator;

[0040] Figure 3 is Figure 1 the internal structure schematic diagram of the main components of the monitoring system in the generator;

[0041] Figure 4 is Figure 1 the detailed structure schematic diagram of the analysis flow path of the monitoring system in the generator;

[0042] Explanation of reference signs: 1, collecting mechanism; 11, collecting pump; 12, multi-parameter flow cell; 13, sample inlet pipeline; 14, sample outlet pipeline; 2, leakage tracing unit; 21, membrane degassing device; 22, thermal conductivity detector; 23, isotope ratio mass spectrometer; 3, sensing unit; 31, pH sensor; 32, dissolved oxygen concentration sensor; 33, conductivity sensor; 34, temperature and pressure sensor; 4, comprehensive analysis and diagnosis unit; 41, controller; 100, hydrogen-cooled generator; 110, hydrogen-cooling system; 111, hydrogen source; 112, air supplementing pump; 113, buffer tank; 114, fan; 115, cooler; 120, constant cooling water system; 121, constant cooling water tank; 122, cooling water pump. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be further described in detail below with reference to the drawings. The described embodiments are only possible technical implementations of the present application, but the present application is not limited to this. Those skilled in the art can combine the embodiments of the present application without creative labor to obtain other embodiments within the protection scope of the present application.

[0044] Please refer to Figures 1-4 , the present application mainly uses isotope tracing and multi-factor fusion analysis to monitor generator corrosion, which achieves the effects of accurately tracing the source of hydrogen, quantifying the corrosion rate and predicting the corrosion trend, avoiding misjudgment of the unit state, and providing decision basis for preventive maintenance. The present application will be further described in detail below.

[0045] Please refer to Figure 1 and Figure 2 , a large hydrogen-cooled generator usually contains two sets of precise cooling systems which are independent but in contact at a certain interface: one is the hydrogen-cooling system 110 for cooling the rotor and stator core, and the other is the constant cooling water system 120 for cooling the largest heat-generating stator winding.

[0046] The hydrogen-cooling system 110 of the generator and the source of "leaked hydrogen": the hydrogen-cooling system 110 of the generator is a closed high-pressure gas circulation loop. In the technical solutions of the present application, the system uses a deuterium isotope-labeled hydrogen source 111 as the cooling medium. As shown in Figure 1 , the labeled hydrogen is pressurized by the air supplementing pump 112 and sent into the buffer tank 113 for pressure stabilization, and then sent into the sealed generator 100 housing by the fan 114. In the housing, the hydrogen circulates at high speed, flows through the surfaces of the generator rotor, stator core and other components, carries away the heat generated by them, is cooled by the cooler 115, and then is sent into the buffer tank 113 to form a circulation.

[0047] It is particularly important to note that the stator winding cooling water connection system (detailed below) is completely immersed in this high-pressure hydrogen environment. This connection system comprises hundreds of brazed joints connecting the insulated water pipes to the copper conductors. Due to long-term exposure to vibration, thermal expansion and contraction, and material aging, these brazed joints are potential weak points. When a microcrack appears at a joint, the high-pressure hydrogen inside the housing can seep into the lower-pressure stator cooling water through the crack. This infiltrated hydrogen, marked with deuterium, is the "leaked hydrogen" that this application seeks to monitor.

[0048] Generator stator cooling water system 120 and its source of "corrosive hydrogen": The generator's stator cooling water system 120 is an independent liquid circulation loop used for direct internal cooling of the stator windings. Its core technology is called "internal water cooling." Specifically, the stator windings are not solid copper bars, but are composed of strands containing hollow copper wires. For example... Figure 1 As shown, the high-purity cooling water in the constant cooling water tank 121 is driven by the cooling water pump 122 and pumped directly into the internal channels of these hollow copper wires. The water flow undergoes direct and sufficient heat exchange with the inner wall of the copper wires, thus efficiently removing heat at its source. Due to this direct contact between the cooling water and the copper material, a physical interface is provided for electrochemical corrosion. Once the water quality (such as pH value, dissolved oxygen, etc.) deteriorates, a corrosion reaction will occur on the inner wall of the copper, producing hydrogen gas. This hydrogen gas produced inside the constant cooling water system due to material corrosion is the "corrosion hydrogen" that this application aims to monitor.

[0049] In summary, hydrogen from two sources may coexist in the generator's stator cooling water: "leaked hydrogen" indicates a problem with the sealing integrity of the water circuit connection system, while "corrosive hydrogen" indicates problems with water quality and the health of the winding materials. The monitoring system protected in this application achieves accurate differentiation and quantitative analysis of these two types of hydrogen by performing online, bypass sampling and precise analysis of water samples in the stator cooling water circuit.

[0050] Please refer to Figure 1 , Figure 3 and Figure 4The application embodiment provides a predictive generator corrosion state monitoring system based on isotope tracing and multi-factor fusion analysis, which comprises a collecting mechanism 1, a leakage tracing unit 2, a sensing unit 3 and a comprehensive analysis and diagnosis unit 4. The collecting mechanism 1 is in communication with a generator constant cooling water circuit to form a circulating passage, thereby providing real-time constant cooling water samples for the system. The leakage tracing unit 2 is in fluid communication with a multi-parameter flow cell 12 of the collecting mechanism 1, degasses the water sample and measures the deuterium / hydrogen isotope ratio of hydrogen, thereby distinguishing between leakage hydrogen and corrosion hydrogen. The sensing unit 3 is installed in the multi-parameter flow cell 12 and measures the pH value, dissolved oxygen concentration and conductivity of the water sample. The comprehensive analysis and diagnosis unit 4 is electrically connected with the leakage tracing unit 2 and the sensing unit 3, determines the leakage condition of the hydrogen cooling system, calculates the real-time corrosion rate and predicts the future corrosion trend, thereby achieving the beneficial effects of accurately tracing the hydrogen source, quantifying the corrosion rate and making a forward-looking prediction on the corrosion trend. The reason lies in that the isotope tracing can determine the hydrogen source and avoid misjudgment, the multi-factor fusion analysis of water quality parameters can accurately calculate the corrosion rate, and the analysis of historical corrosion rate data can predict the future trend.

[0051] Specifically, the collecting mechanism 1 comprises a collecting pump 11 and a multi-parameter flow cell 12.

[0052] The collecting pump 11 is usually a pump body capable of stably providing power, such as a peristaltic pump or a centrifugal pump. The peristaltic pump has good sealing performance and flow stability, and pushes the liquid to flow through peristaltic mode, avoids the contact between the liquid and other parts inside the pump body, and reduces the possibility of pollution. The centrifugal pump relies on the rotation of the impeller to generate centrifugal force to transport the liquid, and has the characteristics of large flow, high lift and the like. The inlet of the collecting pump 11 is in communication with a constant cooling water tank 121 of the generator through a sample inlet pipeline 13, and the outlet thereof is in communication with the multi-parameter flow cell 12 through a sample outlet pipeline 14. The collecting pump 11 drives the constant cooling water sample to flow in the circulating passage, so that the water sample can smoothly enter the subsequent detection unit. The sample inlet pipeline 13 and the sample outlet pipeline 14 generally adopt corrosion-resistant pipe materials, such as polytetrafluoroethylene pipe, to prevent the pipeline from being corroded by the constant cooling water and affect the quality of the water sample and the detection result.

[0053] The multi-parameter flow cell 12 is a container for carrying subsequent detection units, which is usually made of transparent corrosion-resistant materials, such as organic glass or quartz glass, to facilitate observation of the internal water sample. The multi-parameter flow cell 12 serves as an integrated platform and analysis cavity for all subsequent online analysis units, ensuring that all measurements are performed synchronously on the same water sample. It has good sealing performance and stability, and can ensure the stable flow of the water sample, thereby providing good conditions for subsequent detection. The multi-parameter flow cell 12 is connected with the outlet of the collecting pump 11 and receives the constant cooling water sample delivered by the collecting pump.

[0054] In use, the sampling pump 11 draws water sample in the constant cold water tank 121 out and delivers it to the multi-parameter flow cell 12 through the sampling pipeline 13, forming a stable circulation path to provide real-time and continuous constant cold water sample for subsequent detection. Such combination ensures that the water sample can be timely and accurately delivered to each detection unit, ensuring the normal operation of the entire monitoring system.

[0055] Specifically, the leakage tracing unit 2 includes a membrane degassing device 21, an isotope ratio mass spectrometer 23, and a thermal conductivity detector 22.

[0056] The inlet of the membrane degassing device 21 is in communication with the multi-parameter flow cell 12, and it mainly consists of a degassing membrane and a shell. The degassing membrane is usually a kind of selective permeable membrane material, such as a polydimethylsiloxane membrane, which can allow hydrogen dissolved in water to pass through while preventing other substances from passing through. The shell is usually made of corrosion-resistant metal or plastic material to protect the degassing membrane from the external environment. When the constant cold water sample flows through the membrane degassing device 21, the dissolved hydrogen in the water will transfer to the gas phase through the degassing membrane.

[0057] The isotope ratio mass spectrometer 23 is in communication with the gas outlet of the membrane degassing device 21, and it is a high-precision analytical instrument that measures the deuterium / hydrogen isotope ratio of gas-phase hydrogen by analyzing the isotopes in the gas. The isotope ratio mass spectrometer 23 consists of an ion source, a mass analyzer, and a detector. The ion source converts gas-phase hydrogen into an ion beam, the mass analyzer separates different ions according to their mass and charge ratio, and the detector detects the separated ion signals to obtain the deuterium / hydrogen isotope ratio. In addition to the isotope ratio mass spectrometer 23, other instruments that can measure the isotope ratio, such as a laser absorption spectrometer, can also be used as a substitute.

[0058] The thermal conductivity detector 22 is installed between the membrane degassing device 21 and the isotope ratio mass spectrometer 23, and is used to measure the total hydrogen gas concentration in the gas phase discharged by the membrane degassing device before the isotope ratio measurement, to calculate the real-time total dissolved hydrogen concentration.

[0059] In use, when the cold water sample flows through the membrane degassing device 21 in communication with the multi-parameter flow cell 12, all the dissolved hydrogen in the water, including the leakage hydrogen and the corrosion hydrogen, is efficiently and non-fractionally transferred to the gas phase to form a gas stream carrying all the hydrogen information. The gas stream first enters the thermal conductivity detector 22. The thermal conductivity of hydrogen is extremely high, far exceeding that of the carrier gas (such as nitrogen). The thermal conductivity detector 22 can instantly calculate the total hydrogen volume concentration in the gas stream by accurately measuring the change in thermal conductivity of the mixed gas. The comprehensive analysis and diagnosis unit 4 then converts the gas and water flow parameters into the real-time total dissolved hydrogen concentration required by the subsequent formula. After measuring the concentration, the gas stream is immediately introduced into the isotope ratio mass spectrometer 23. The mass spectrometer performs isotope analysis on the gas stream to measure the deuterium / hydrogen isotope ratio online and with high precision.

[0060] Specifically, the sensing unit 3 includes a pH sensor 31, a dissolved oxygen concentration sensor 32, a conductivity sensor 33, and a temperature and pressure sensor 34.

[0061] The pH sensor 31 is used to measure the pH value of the cold water sample in real time. It generally uses a glass electrode or a composite electrode. The glass electrode has the characteristics of fast response speed and high measurement accuracy. The composite electrode integrates the reference electrode and the measurement electrode, and is more convenient to use. The pH sensor 31 measures the hydrogen ion concentration in the cold water sample by contacting it, thereby obtaining the pH value.

[0062] The dissolved oxygen concentration sensor 32 is used to measure the dissolved oxygen concentration in the cold water sample. Common ones are polarographic dissolved oxygen sensors and fluorescence dissolved oxygen sensors. The polarographic dissolved oxygen sensor is based on electrochemical principle, and determines the dissolved oxygen concentration by measuring the current change on the electrode surface; the fluorescence dissolved oxygen sensor measures the dissolved oxygen concentration by using the quenching effect of fluorescent substances on oxygen, and has the advantages of high measurement accuracy and strong anti-interference ability.

[0063] The conductivity sensor 33 is used to measure the conductivity of the cold water sample. It is usually composed of two electrodes, and the conductivity is calculated by measuring the resistance between the electrodes. The electrodes of the conductivity sensor 33 are generally made of platinum or stainless steel, etc., which have good conductivity and corrosion resistance.

[0064] The temperature and pressure sensor 34 is used to measure the temperature and pressure of the cold water sample synchronously. It can be an integrated sensor, or the temperature sensor and the pressure sensor can be separately arranged. The integrated temperature and pressure sensor 34 has the advantages of small size and easy installation. The temperature sensor generally uses thermocouple or thermistor elements, and the pressure sensor has various types according to different measurement principles, such as piezoelectric pressure sensor, capacitive pressure sensor, etc.

[0065] In use, each sensor is installed in or in communication with the multi-parameter flow cell 12 to measure the pH, dissolved oxygen concentration, conductivity, temperature and pressure of the once-through cooling water. These parameters reflect the water quality of the once-through cooling water from different aspects and provide rich data support for subsequent comprehensive analysis. The comprehensive measurement of multiple parameters can more comprehensively evaluate the corrosiveness of the water quality and improve the accuracy of the internal corrosion state monitoring of the generator.

[0066] In particular, the comprehensive analysis and diagnosis unit 4 is usually implemented by an industrial computer or an embedded controller 41 with a pre-installed program.

[0067] Please refer to Figure 3 , when determining whether the hydrogen cooling system of the generator has a leak, the comprehensive analysis and diagnosis unit 4 first obtains the real-time measured deuterium / hydrogen isotope ratio R measured , and compares it with the natural abundance isotope ratio baseline R natural representing the hydrogen generated by pure corrosion stored in the controller. When the condition R measured > R natural + δ (where δ is a pre-set judgment threshold value for preventing signal fluctuation interference) is met, it is determined that there is a hydrogen leak. This judgment process takes advantage of the difference in isotope ratios of hydrogen from different sources and can accurately determine whether the hydrogen cooling system has a leak.

[0068] When calculating the hydrogen concentration C corr generated by internal component corrosion, the comprehensive analysis and diagnosis unit 4 uses the known isotope ratio R leak , the natural abundance ratio R natural , and the real-time measured R measured and total hydrogen concentration C H to accurately calculate the hydrogen concentration C corr generated by internal component corrosion only. This calculation process excludes the interference of leaked hydrogen, making the subsequent corrosion analysis more accurate. The isotope mass balance equation is:

[0069] ,

[0070] where C H is the real-time total dissolved hydrogen concentration obtained by the thermal conductivity detector; R measured is the real-time deuterium / hydrogen isotope ratio measured by the isotope ratio mass spectrometer; R leak is the pre-set deuterium / hydrogen isotope ratio of deuterium-labeled hydrogen as the cooling medium; and R natural is the pre-set natural isotope ratio baseline representing corrosion-produced hydrogen.

[0071] In calculating the real-time corrosion rate, the comprehensive analysis and diagnosis unit calls a multi-parameter electrochemical corrosion model. The input variables of the model include the corrosion hydrogen concentration C corr , the dissolved oxygen concentration, the pH value, the conductivity σ and the temperature T measured by the sensing unit. The model can be an empirical or semi-empirical functional relationship, and the final output characterizes the real-time corrosion rate V corr of the copper components inside the generator. The temperature and pressure data measured by the temperature and pressure sensor 34 can be used to correct and compensate the model calculation or the sensor readings, improving the accuracy of the calculation.

[0072] The multi-parameter electrochemical corrosion model is a functional relationship as follows:

[0073] ,

[0074] wherein k is a comprehensive corrosion coefficient; f, g, h, j and e are empirical functions related to the corrosion hydrogen concentration, the dissolved oxygen concentration, the pH value, the conductivity and the temperature, respectively; and T is the real-time water temperature measured by the sensing unit.

[0075] In predicting the future corrosion trend, the comprehensive analysis and diagnosis unit performs time series analysis on a plurality of historical real-time corrosion rate Vcorr data points collected at continuous time instants. The specific implementation is to collect and store a sequence of N continuous real-time corrosion rate data points (t i , V corr,i ) within a preset time window, perform least square linear regression analysis on the data point sequence, obtain the slope m representing the linear change trend of the corrosion rate with time, and based on the slope m and the latest corrosion rate V corr,now , predict the corrosion rate V

[0076] V pred = m·Δt + V corr,now

[0077] after a future time interval Δt by an extrapolation formula pred , and provide prospective guidance for preventive maintenance according to the comparison result of the prediction value and the safety threshold.

[0078] The combination logic of the comprehensive analysis and diagnosis unit 4 receives the measurement data of the leakage tracing unit 2 and the sensing unit 3, processes and analyzes the data through a series of algorithms and models, realizes the judgment of the hydrogen cooling system leakage, the calculation of the corrosion hydrogen concentration, the calculation of the real-time corrosion rate and the prediction of the future corrosion trend. This combination mode fuses and analyzes multiple data, fully excavates the information behind the data, and provides a strong decision basis for the preventive maintenance of the generator.

[0079] The beneficial effects of the technical solutions provided in the application include:

[0080] (1) By using isotope tracing technology, the system can accurately distinguish between hydrogen gas from hydrogen cooling system leakage and hydrogen gas generated by internal component corrosion. This solves the fundamental problem of unknown hydrogen sources in traditional monitoring techniques, avoiding misjudgments about the operation status of the generator set.

[0081] (2) The system combines the corrosion hydrogen concentration obtained by isotope analysis with real-time measured pH, dissolved oxygen concentration, conductivity and other water quality parameters for fusion analysis. Through the built-in electrochemical corrosion model, the real-time corrosion rate of the internal copper components of the generator is quantitatively calculated, and the corrosiveness of the water quality can be more comprehensively evaluated.

[0082] (3) The system can generate predictive information on future corrosion trends by performing time series analysis on historical real-time corrosion rate data. This provides direct decision-making basis for implementing preventive maintenance, enabling the unit maintenance to be transformed from a passive "fault response" mode to an active "state prediction" mode.

[0083] (4) Through early and accurate identification of leakage and corrosion risks and future trend prediction, the present scheme helps to improve the safety and reliability of the generator operation, prolong the service life of the equipment, and reduce maintenance costs.

[0084] The specific embodiments of the present application described above do not constitute a limitation on the scope of protection of the present application. Any various other corresponding changes and modifications made in accordance with the technical concept of the present application shall be included within the scope of protection of the present application.

Claims

1. A predictive generator corrosion condition monitoring system based on isotopic tracing and multi-factor fusion analysis for monitoring a hydrogen-cooled generator (100) employing deuterium isotope-labeled hydrogen gas as a cooling medium, characterized in that, The generator corrosion state monitoring system comprises: a collection mechanism (1) in communication with a generator constant cooling water circuit, forming a circulating path for providing real-time constant cooling water samples for the system, the collection mechanism (1) comprising a collection pump (11) for driving water sample circulation and a multi-parameter flow cell (12) carrying subsequent detection units; a leakage tracing unit (2) in fluid communication with the multi-parameter flow cell (12), for degassing and online measuring deuterium / hydrogen isotope ratio of hydrogen from the real-time constant cooling water samples, for tracing the composition of dissolved hydrogen, to quantitatively distinguish hydrogen from hydrogen cooling system (110) leakage and hydrogen generated by electrochemical corrosion of internal components in contact with constant cooling water; the leakage tracing unit (2) comprising a membrane degassing device (21), an isotope ratio mass spectrometer (23) and a thermal conductivity detector (22); a sensing unit (3) installed in the multi-parameter flow cell (12), for measuring pH value, dissolved oxygen concentration and conductivity of the real-time constant cooling water samples, for evaluating water quality corrosion; and a comprehensive analysis and diagnosis unit (4) electrically connected with the leakage tracing unit (2) and the sensing unit (3), for determining whether there is leakage and the degree of leakage of the generator hydrogen cooling system (110) based on real-time data of the deuterium / hydrogen isotope ratio, quantitatively calculating real-time corrosion rate of internal copper components of the generator by algorithm based on pH value, dissolved oxygen concentration, conductivity of the real-time constant cooling water samples and hydrogen concentration generated by internal component corrosion analyzed by the leakage tracing unit (2) according to isotope ratio, and generating predictive information on future corrosion trend, to provide decision basis for preventive maintenance; The comprehensive analysis and diagnosis unit (4) quantitatively calculates the hydrogen concentration C generated due to corrosion of the internal member by operating the following isotope mass balance equation corr : wherein C H is the real-time total dissolved hydrogen concentration obtained by the thermal conductivity detector (22); R measured is the deuterium / hydrogen isotope ratio measured in real time by the isotope ratio mass spectrometer (23); R leak is the preset deuterium / hydrogen isotope ratio of the deuterium-labeled hydrogen gas as the cooling medium; R natural is the preset natural isotope ratio baseline representing the corrosion-produced hydrogen.

2. The predictive generator corrosion condition monitoring system based on isotopic provenance and multi-factor fusion analysis according to claim 1, characterized in that, the inlet of the membrane degassing device (21) is in communication with the multi-parameter flow cell (12), for transferring dissolved hydrogen in the real-time constant cooling water samples to gas phase; the isotope ratio mass spectrometer (23) is in communication with the gas outlet of the membrane degassing device (21), for measuring deuterium / hydrogen isotope ratio of the gas phase hydrogen, and the thermal conductivity detector (22) is installed between the membrane degassing device (21) and the isotope ratio mass spectrometer (23), for measuring total hydrogen concentration in the gas phase discharged by the membrane degassing device (21) before isotope ratio measurement, to calculate real-time total dissolved hydrogen concentration.

3. The predictive generator corrosion condition monitoring system based on isotopic provenance and multi-factor fusion analysis according to claim 2, characterized in that, The sensing unit (3) further comprises a temperature and pressure sensor (34) in communication with the multi-parameter flow cell (12), for synchronously measuring temperature and pressure of the real-time constant cooling water samples, and the comprehensive analysis and diagnosis unit (4) receives temperature and pressure data measured by the temperature and pressure sensor (34) and uses them as parameters for correcting or compensating calculation of the real-time corrosion rate.

4. The predictive generator corrosion condition monitoring system based on isotopic provenance and multi-factor fusion analysis according to claim 1, characterized in that, The inlet of the collection pump (11) is in communication with the constant cooling water tank (121) of the generator through a sample inlet pipeline (13), and the outlet is in communication with the multi-parameter flow cell (12) through a sample outlet pipeline (14).

5. The predictive generator corrosion condition monitoring system based on isotopic provenance and multi-factor fusion analysis of claim 1, wherein, The sensing unit (3) comprises a pH sensor (31), a dissolved oxygen concentration sensor (32) and a conductivity sensor (33) for respectively determining the pH value, dissolved oxygen concentration and conductivity of the cold water sample in real time.

6. The predictive generator corrosion condition monitoring system based on isotopic provenance and multi-factor fusion analysis of claim 1, wherein, The process for determining whether the generator hydrogen cooling system (110) has a leakage by the comprehensive analysis and diagnosis unit (4) comprises the following steps: Obtaining real-time measured deuterium / hydrogen isotope ratio R measured ; The R measured is compared to a baseline R natural representative of the natural abundance isotopic ratio of hydrogen produced by pure corrosion, stored in the controller (41). When condition R is satisfied measured >R natural When +δ is reached, a hydrogen leak is determined to exist, where δ is a preset judgment threshold.

7. The predictive generator corrosion condition monitoring system based on isotopic provenance and multi-factor fusion analysis according to claim 3, characterized in that, Said integrated analysis and diagnostic unit (4) is configured to calculate the real-time corrosion rate V of the copper components inside the generator corr The algorithm is based on a multi-parameter electrochemical corrosion model whose input variables include at least the hydrogen concentration C corr generated by the corrosion of the internal components, the dissolved oxygen concentration, the pH value and the conductivity σ measured by said sensing unit (3); The multi-parameter electrochemical corrosion model is a function relationship as follows: V corr = k - f(C corr ) - g([O2]) - h(pH) - j(σ) - e(T) wherein k is a global corrosion coefficient; f, g, h, j, e are empirical functions related to C corr , dissolved oxygen concentration, pH, conductivity σ and temperature; T is the real-time water sample temperature measured by the sensor unit (3).

8. The predictive generator corrosion condition monitoring system based on isotopic provenance and multi-factor fusion analysis according to claim 7, characterized in that, Said integrated analysis and diagnostic unit (4) generates predictive information on future corrosion trends by performing a time series analysis on the data points of the real-time corrosion rate V corr obtained at a plurality of historical time instants, continuously acquired, which analysis comprises at least the calculation of the trend slope of the corrosion rate and the establishment of an extrapolation prediction model.

9. The predictive generator corrosion condition monitoring system based on isotopic provenance and multi-factor fusion analysis according to claim 8, characterized in that, The specific implementation of calculating the change trend slope of the corrosion rate and establishing an extrapolation prediction model is: collecting and storing N continuous real-time corrosion rate data point sequences (t i , corr,i ) in a preset time window, where i is a natural number between 1 and N. Least square linear regression analysis is performed on the data point sequence to obtain a slope m representing a linear change trend of the corrosion rate with time; Based on this slope m and the current latest corrosion rate V corr,now by extrapolation formula: V pred = m · Δt + V corr,now to predict the corrosion rate V after a future time interval Δt pred and provide proactive guidance for preventive maintenance based on the comparison of the predicted value with a safety threshold.

Images