Thermal power plant rotating machinery health management method and system based on Internet of Things

Abstract

The invention relates to the technical field of mechanical equipment health management, in particular to a thermal power plant rotating machine health management method and system based on the Internet of Things, and the method comprises the following steps: S1, synchronously collecting a vibration acoustic emission signal, a lubricating oil dielectric constant and metal surface micro strain field data; s2, calculating a bearing alloy grain boundary corrosion rate, and inverting a material dislocation density change rate through micro strain field data; s3, establishing an incidence matrix; s4, constructing a three-dimensional health state coordinate system based on the incidence matrix in the S3; s5, when the slope of the track exceeds the fatigue limit curve of the material, automatically matching a maintenance action; and S6, verifying the effectiveness of the measure by comparing vector direction changes before and after maintenance, and dynamically correcting the weight coefficient of the incidence matrix. According to the invention, through multi-modal data fusion and three-dimensional health state modeling, the health state of the rotating machinery of the thermal power plant is monitored in real time, and the fault risk is accurately predicted, so that scientific maintenance decision is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical equipment health management, and in particular to a rotating machinery health management method and system for a thermal power plant based on the Internet of Things. Background Art

[0002] In thermal power plants, rotating machinery, such as steam turbines and fans, operates under high loads, high temperature, and high humidity for extended periods. Their health directly impacts the equipment's operating efficiency and the safety and economic viability of the entire power plant. Traditional rotating machinery health management methods rely primarily on regular inspections and manual measurements, typically relying on a single type of sensor to monitor equipment status. This approach has significant limitations, such as insufficient detection accuracy, an inability to reflect dynamic equipment changes in real time, and maintenance decisions often based on experience rather than precise scientific evidence. This leads to unpredictable equipment failures and wastes or delays in maintenance resources.

[0003] With the development of IoT technology, intelligent monitoring systems based on the IoT have become an important direction for improving the efficiency of equipment health management. However, despite the use of multiple sensors and data acquisition systems, existing technologies generally lack systematic solutions for comprehensive health assessment of rotating machinery equipment, and it is difficult to accurately assess the changing trends of equipment health status through collaborative analysis of multimodal data. Therefore, how to effectively integrate different types of sensor data and dynamically evaluate equipment health status through scientific evaluation models to accurately guide maintenance decisions has become a difficult problem that needs to be solved in current technology. Summary of the Invention

[0004] Based on the above objectives, the present invention provides a method and system for health management of rotating machinery in a thermal power plant based on the Internet of Things.

[0005] A method for health management of rotating machinery in a thermal power plant based on the Internet of Things comprises the following steps: S1: Multimodal sensors are deployed at predetermined locations on the turbine bearings and wind turbine gearboxes to simultaneously collect vibration and acoustic emission signals, lubricating oil dielectric constant, and metal surface microstrain field data; S2: Input the time series data of the dielectric constant of the lubricating oil obtained in S1 into the metal ion mobility analysis model based on quantum dot sensing to calculate the grain boundary corrosion rate of the bearing alloy, and at the same time invert the material dislocation density change rate through the microstrain field data; S3: Establish the correlation matrix of the power spectrum of the vibration acoustic emission signal, the grain boundary corrosion rate of the bearing alloy, and the change rate of the material dislocation density; S4: Construct a three-dimensional health status coordinate system based on the correlation matrix of S3, where the horizontal axis is the mechanical wear degree, the vertical axis is the material degradation degree, and the vertical axis is the lubrication failure degree; S5: Based on the motion trajectory of the healthy state point in S4 in the three-dimensional healthy state coordinate system, when the trajectory slope exceeds the material fatigue limit curve, maintenance actions are automatically matched. Specific maintenance actions include: Action 1: If lubrication failure is dominant, online oil filtration is triggered; Action 2: If mechanical wear is dominant, start laser cladding repair; Action 3: If the material degradation exceeds the limit, a spare parts replacement warning is issued; S6: After executing the S5 maintenance action, re-collect the S1 data stream and calculate the displacement vector of the healthy state point. Verify the effectiveness of the measures by comparing the changes in the vector direction before and after maintenance, and dynamically correct the weight coefficient of the correlation matrix.

[0006] Optionally, the S1 specifically includes: S11: Install the sensor bracket on the outer side of the turbine bearing and the side wall of the fan gearbox by mechanical fixing. The sensor bracket is made of high-temperature resistant alloy material. S12: Three types of sensors are installed on each sensor bracket, including a vibration sensor, a lubricating oil dielectric constant sensor, and a micro-strain sensor. The vibration sensor uses an accelerometer with a frequency response range of 20 Hz to 2 MHz and a sampling frequency of no less than 2 MHz. The lubricating oil dielectric constant sensor uses a capacitive sensor to monitor changes in the dielectric constant of the lubricating oil. The micro-strain sensor uses fiber Bragg grating technology to collect real-time micro-strain field data on the metal surface. S13: Data acquisition of each sensor collects vibration signals, lubricating oil dielectric constant data and micro-strain signals at a frequency of 1000 times per second, and ensures signal synchronization.

[0007] Optionally, the S2 specifically includes: S21: The time series data of the dielectric constant of the lubricating oil obtained in S1 is transmitted to the data processing platform, and the time series data is filtered by the data processing module to remove high-frequency noise; S22: Based on quantum dot sensing technology, a metal ion mobility analysis model is constructed. This analysis model uses quantum dot surface modification technology to improve the sensitivity to metal ions, and combines liquid conductivity and temperature parameters to dynamically calculate the mobility of metal ions. The calculation formula is: ,in, is the current intensity measured by the sensor, is the initial metal ion concentration, is the time interval, is the metal ion mobility; S23: Calculate the intergranular corrosion rate of bearing alloys based on metal ion mobility data and changes in the dielectric constant of lubricating oil , the formula is: ,in, is a constant related to the bearing material and lubricant properties, is the change in dielectric constant of lubricating oil; S24: Using microstrain field data, calculate the material dislocation density change rate through inversion algorithm , the formula is: ,in, and are the dislocation densities of the material at the current moment and the initial moment, is the time interval.

[0008] Optionally, the S3 specifically includes: S31: Perform fast Fourier transform processing on the vibration acoustic emission signal collected in S1, convert it into a frequency domain signal, and calculate the power spectrum; S32: Based on the bearing alloy intergranular corrosion rate and material dislocation density change rate calculated in S2, the numerical range of these two values ​​is unified to interval; S33: Combine the power spectrum data calculated in S31 and the corrosion rate and dislocation density change rate data normalized in S32 to establish a correlation matrix between the three. The structure of the correlation matrix is: ; in, Frequency The power spectrum value under For the moment The intergranular corrosion rate of bearing alloys under For the moment The material dislocation density change rate under For different frequency values, The index of the sampling point.

[0009] Optionally, the S4 specifically includes: S41: Based on the correlation matrix in S3, extract the normalized values ​​of the vibration signal power spectrum, the bearing alloy grain boundary corrosion rate, and the material dislocation density change rate; and use these values ​​as the three basic dimensions in the coordinate system, defined as mechanical wear, material degradation, and lubrication failure. S42: Determine the axial scale of the three-dimensional health status coordinate system, where the horizontal axis is set to mechanical wear, using the integral result of the vibration signal power spectrum to characterize the degree of mechanical wear; the vertical axis is set to material degradation, which is comprehensively reflected by combining the weighted sum of the bearing alloy grain boundary corrosion rate and the material dislocation density change rate; the vertical axis is set to lubrication failure, which represents the change in lubrication status by monitoring the change in the dielectric constant of the lubricating oil; S43: Input the numerical values ​​of each dimension obtained in S41 and S42 into the three-dimensional coordinate system to construct a three-dimensional health status coordinate system. The three-dimensional health status coordinate system can display the data of mechanical wear, material degradation and lubrication failure in three dimensions. By marking the health status points in the three-dimensional coordinate system, the current health status position of the equipment can be analyzed.

[0010] Optionally, the S42 specifically includes: S421: The integral of the vibration signal power spectrum reflects the energy consumption of the machine in different frequency bands and can measure the overall level of mechanical wear. The integral result of the vibration signal power spectrum is calculated using the following formula: ,in, is the integration result of the vibration signal, Frequency The power spectrum value under and are the frequency ranges for calculating the integral respectively; S422: The corrosion rate and dislocation density change rate reflect two different degradation mechanisms of the material. By taking the weighted average of these two indicators, the comprehensive material degradation degree is obtained. The calculation formula is: ,in, is the material degradation degree, is the intergranular corrosion rate of the bearing alloy, is the material dislocation density change rate, and are weight coefficients, reflecting the influence ratios of corrosion rate and dislocation density change rate on material degradation; S423: The change in the dielectric constant of the lubricating oil reflects the deterioration of the lubricating oil. The dielectric constant changes with the aging and contamination of the lubricating oil. The degree of lubrication failure is calculated using the following formula: ,in, is the lubrication failure degree, is the dielectric constant value of the current lubricating oil, is the dielectric constant of the lubricating oil at the initial moment, It is the dielectric constant value when the lubricating oil reaches the maximum degradation.

[0011] Optionally, the S5 specifically includes: S51: Based on the three-dimensional health status coordinate system constructed in S4, the movement trajectory of the health status point in the coordinate system is tracked. Each health status point corresponds to the current health status of the mechanical equipment. The position of the health status point is updated by continuous real-time data, which can form a dynamic trajectory of the health status point. S52: Calculating a trajectory slope based on the motion trajectory of the health status point. The slope represents a change speed of the health status point along the time axis in the three-dimensional coordinate system, and is used to reveal the acceleration of the change of the health status of the device. S53: By comparing the trajectory slope of the healthy state point with the slope of the fatigue limit curve, it is determined whether the fatigue bearing capacity of the material has been exceeded. When the trajectory slope exceeds the material fatigue limit curve, it indicates that the equipment has entered the critical fatigue state and maintenance action is required; S54: Automatically matching corresponding maintenance actions based on the trajectory slope judgment result of the health status point.

[0012] Optionally, the S52 specifically includes: S521: The position of each health status point in the three-dimensional health status coordinate system is determined by the real-time health status data of the mechanical equipment. The motion trajectory of the health status point is formed through continuous data update. The position coordinates of the health status point are expressed as ,in Indicates time, ,and Respectively represent the horizontal, vertical and vertical axis coordinates of the health state point in the three-dimensional coordinate system; S522: Calculate the rate of change of the health status point at different time points, and calculate the displacement between two adjacent moments according to the position change of the health status point in the time series. ; S523: Calculate the movement speed of the healthy state point based on the displacement and time difference between adjacent time points ; S524: Calculate the slope of the motion trajectory of the health state point. The slope reflects the acceleration of the health state point along the time axis. The slope of the trajectory Calculated by the following formula: ,in, and Time points and The speed value below.

[0013] Optionally, the S6 specifically includes: S61: After performing the maintenance operation, the position of the health status point is updated by re-collecting the sensor data stream obtained in S1; S62: Verify the effectiveness of maintenance measures by comparing the direction changes of the displacement vectors of the healthy state points before and after maintenance; the change of the vector is calculated by calculating the angle between the two to judge; S63: Evaluate the effectiveness of maintenance measures based on the size of the angle. If the angle is less than 5 degrees, it means that the changes in the health status points are relatively consistent, indicating that the maintenance measures are successful and changing in the expected direction; if the angle If it is greater than 15 degrees, it means that the maintenance effect is not obvious or there are problems; S64: Based on the change of the displacement vector of the health state point, automatically adjust the weight coefficient of the correlation matrix established in S3. Specifically, if the angle If the angle is less than 5 degrees, it means that the changes in health status points are relatively consistent, and the weight coefficients between related factors in the correlation matrix should be increased by 5% to 10%; if the angle is less than 5 degrees, it means that the changes in health status points are relatively consistent, and the weight coefficients between related factors in the correlation matrix should be increased by 5% to 10%; If it is greater than 15 degrees, it means that the health status of the equipment has not improved significantly. The weight coefficient of the relevant parameters in the correlation matrix should be reduced by 5% to 10%.

[0014] An IoT-based health management system for rotating machinery in thermal power plants, used to implement the aforementioned IoT-based health management method for rotating machinery in thermal power plants, includes the following modules: Multimodal sensing module: used to deploy sensors at predetermined locations on turbine bearings and wind turbine gearboxes to collect vibration and acoustic emission signals, lubricating oil dielectric constant, and metal surface microstrain field data; Data processing and analysis module: used to input the time series data of the lubricating oil dielectric constant into the metal ion mobility analysis model based on quantum dot sensing to calculate the grain boundary corrosion rate of the bearing alloy; at the same time, it inverts the material dislocation density change rate through microstrain field data; Health status assessment module: Based on the bearing alloy grain boundary corrosion rate and material dislocation density change rate output by the data processing and analysis module, combined with the vibration acoustic emission signal power spectrum, a correlation matrix is ​​established; Health status modeling module: used to construct a three-dimensional health status coordinate system based on the correlation matrix output by the health status assessment module, where the horizontal axis is the mechanical wear degree, the vertical axis is the material degradation degree, and the vertical axis is the lubrication failure degree; Maintenance decision module: This module calculates the trajectory slope of the health point in the three-dimensional health coordinate system output by the health modeling module and compares it with the fatigue limit curve to determine whether the equipment has entered a critical fatigue state. When the trajectory slope exceeds the material fatigue limit curve, the module automatically selects the optimal maintenance action. Maintenance feedback and correction module: This module is used to re-collect multimodal sensor module data after performing maintenance actions, update the position of the health status point, and verify the effectiveness of the maintenance measures by comparing the changes in the displacement vector of the health status point before and after maintenance. At the same time, the association matrix weight coefficient is dynamically corrected according to the verification results.

[0015] Beneficial effects of the present invention: The present invention can comprehensively and accurately monitor the health status of rotating mechanical equipment in thermal power plants by collecting multiple sensor data in real time, including information such as vibration, lubricating oil dielectric constant, and metal surface microstrain. By deeply fusing these different types of sensor data and combining them with quantum dot sensing technology to analyze metal ion mobility, a comprehensive assessment of equipment wear, corrosion, and dislocation density changes can be achieved, overcoming the limitation of traditional methods that can only rely on a single signal, and ensuring multi-dimensional, real-time monitoring of equipment health status.

[0016] The present invention, by establishing a three-dimensional health status coordinate system and combining it with the dynamic motion trajectory analysis of the equipment health status points, can timely discover the potential failure risks of the equipment; by accurately calculating the changing trend of the equipment health status, it can automatically trigger the corresponding maintenance actions before the equipment enters the critical fatigue state, ensuring that the maintenance decision is scientific and reasonable, and avoiding the blindness and lag of traditional experience-driven methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a schematic diagram of a method for health management of rotating machinery in a thermal power plant according to an embodiment of the present invention; Figure 2 Schematic diagram of a rotating machinery health management system for a thermal power plant according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. It is also noted that, to provide a more detailed description, the following embodiments are best and preferred embodiments, and those skilled in the art may employ alternative methods for implementing certain known technologies. Furthermore, the accompanying drawings are intended only to provide a more detailed description of the embodiments and are not intended to limit the present invention.

[0020] It should be noted that references in the specification to "one embodiment," "an embodiment," "exemplary embodiments," "some embodiments," etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not necessarily every embodiment will include such specific features, structures, or characteristics. Furthermore, when specific features, structures, or characteristics are described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement such features, structures, or characteristics in conjunction with other embodiments (whether or not explicitly described).

[0021] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0022] like Figure 1 As shown, a method for health management of rotating machinery in a thermal power plant based on the Internet of Things includes the following steps: S1: Multimodal sensors are deployed at predetermined locations on steam turbine bearings and wind turbine gearboxes to simultaneously collect vibration and acoustic emission signals, lubricating oil dielectric constant, and metal surface microstrain field data. The sampling frequency of the vibration and acoustic emission signals must be no less than 2 MHz to capture the impact waveforms of micron-sized wear particles. S2: Input the time series data of the dielectric constant of the lubricating oil obtained in S1 into the metal ion mobility analysis model based on quantum dot sensing to calculate the grain boundary corrosion rate (CR) of the bearing alloy, and at the same time invert the material dislocation density change rate (DR) through the microstrain field data; S3: Establish a correlation matrix between the power spectrum of the vibration acoustic emission signal, the grain boundary corrosion rate of the bearing alloy, and the material dislocation density change rate. When a sudden increase in the acoustic emission energy in the 100-300kHz frequency band is detected and the ratio of the bearing alloy grain boundary corrosion rate or the material dislocation density change rate exceeds a preset threshold, it is determined that the system has entered the dangerous stage of combined wear. S4: Based on the correlation matrix of S3, a three-dimensional health status coordinate system is constructed, where the horizontal axis is the mechanical wear (calculated by the integral of acoustic emission energy), the vertical axis is the material degradation (weighted synthesis of CR and DR), and the vertical axis is the lubrication failure (represented by the gradient of dielectric constant change); S5: Based on the motion trajectory of the healthy state point in S4 in the three-dimensional healthy state coordinate system, when the trajectory slope exceeds the material fatigue limit curve, maintenance actions are automatically matched. Specific maintenance actions include: Action 1: If lubrication failure is dominant, online oil filtration is triggered; Action 2: If mechanical wear is dominant, start laser cladding repair; Action 3: If the material degradation exceeds the limit, a spare parts replacement warning is issued; S6: After executing the S5 maintenance action, re-collect the S1 data stream and calculate the displacement vector of the healthy state point. Verify the effectiveness of the measures by comparing the changes in the vector direction before and after maintenance, and dynamically correct the weight coefficient of the correlation matrix.

[0023] S1 specifically includes: S11: The sensor bracket is mechanically fixed on the outer side of the turbine bearing and the side wall of the fan gearbox. The sensor bracket is made of high-temperature resistant alloy material to ensure long-term stability in high-temperature environments. S12: Three types of sensors are installed on each sensor bracket, including a vibration sensor, a lubricating oil dielectric constant sensor, and a micro-strain sensor. The vibration sensor uses an accelerometer with a frequency response range of 20 Hz to 2 MHz and a sampling frequency of no less than 2 MHz. The lubricating oil dielectric constant sensor uses a capacitive sensor to monitor changes in the dielectric constant of the lubricating oil. The micro-strain sensor uses fiber Bragg grating technology to collect real-time micro-strain field data on the metal surface. S13: Each sensor collects vibration signals, lubricating oil dielectric constant data, and microstrain signals at a frequency of 1000 times per second, and ensures signal synchronization. The above steps achieve real-time monitoring of vibration, lubricating oil conditions, and microscopic changes on metal surfaces by deploying multimodal sensors in key locations such as turbine bearings and wind turbine gearboxes, providing reliable data support for subsequent health status assessments and maintenance decisions.

[0024] S2 specifically includes: S21: The lubricating oil dielectric constant time series data obtained in S1 is transmitted to the data processing platform, and the time series data is filtered through the data processing module to remove high-frequency noise and ensure the stability and accuracy of the data; S22: Based on quantum dot sensing technology, a metal ion mobility analysis model is constructed. This model uses quantum dot sensors to monitor the migration process of metal ions in lubricating oil in real time. This analysis model uses quantum dot surface modification technology to improve the sensitivity to metal ions and combines liquid conductivity and temperature parameters to dynamically calculate the mobility of metal ions. The calculation formula is: ,in, is the current intensity measured by the sensor, is the initial metal ion concentration, is the time interval, is the metal ion mobility; S23: Calculate the intergranular corrosion rate of bearing alloys based on metal ion mobility data and changes in the dielectric constant of lubricating oil , the formula is: ,in, is a constant related to the bearing material and lubricant properties, is the change in dielectric constant of the lubricating oil; through this formula, the corrosion rate is obtained It can reflect the corrosion degree of bearing alloy in actual working environment; S24: Using microstrain field data, calculate the material dislocation density change rate through inversion algorithm , the formula is: ,in, and are the dislocation densities of the material at the current moment and the initial moment, The time interval is 10 ...

[0025] S3 specifically includes: S31: Perform fast Fourier transform (FFT) on the vibration acoustic emission signal collected in S1, convert it into a frequency domain signal, and calculate the power spectrum. The power spectrum of the vibration acoustic emission signal is represented by the energy distribution of the signal at different frequency points; the formula is: ,in, For the frequency The power spectrum value under For the moment The vibration signal value under is the number of sampling points, is the frequency value, is the index of the sampling point, is an imaginary unit. This process can extract the frequency features in the vibration signal, especially the frequency band related to wear; S32: Based on the bearing alloy intergranular corrosion rate and material dislocation density change rate calculated in S2, the numerical range of these two values ​​is unified to The normalized corrosion rate and dislocation density change rate can be combined with the vibration signal power spectrum to make a horizontal comparison. S33: Combine the power spectrum data calculated in S31 and the corrosion rate and dislocation density change rate data normalized in S32 to establish a correlation matrix between the three. The structure of the correlation matrix is: ; in, Frequency The power spectrum value under For the moment The intergranular corrosion rate of bearing alloys under For the moment The material dislocation density change rate under For different frequency values, is the index of the sampling point; through this correlation matrix, the different information of vibration signal, corrosion rate and dislocation density change rate can be integrated together to reflect the relationship between them; the above steps establish an accurate correlation matrix by combining the power spectrum of vibration acoustic emission signal, bearing alloy grain boundary corrosion rate and material dislocation density change rate. This matrix can reveal the relationship between different monitoring indicators and provide a more effective technical means for the health management of rotating machinery in thermal power plants.

[0026] S4 specifically includes: S41: Based on the correlation matrix in S3, extract the normalized values ​​of the vibration signal power spectrum, the bearing alloy grain boundary corrosion rate, and the material dislocation density change rate. These values ​​are then used as the three basic dimensions in the coordinate system, defined as mechanical wear, material degradation, and lubrication failure. These dimensions represent different health status information of the equipment, and are standardized to keep their numerical ranges consistent to facilitate subsequent comprehensive evaluation. S42: Determine the axial scale of the three-dimensional health status coordinate system, where the horizontal axis is set to mechanical wear, using the integral result of the vibration signal power spectrum to characterize the degree of mechanical wear; the vertical axis is set to material degradation, which is comprehensively reflected by combining the weighted sum of the bearing alloy grain boundary corrosion rate and the material dislocation density change rate; the vertical axis is set to lubrication failure, which represents the change in lubrication status by monitoring the change in the dielectric constant of the lubricating oil. The scales of each dimension will be adjusted according to different working environments and equipment types; S43: Input the numerical values ​​of each dimension obtained in S41 and S42 into the three-dimensional coordinate system to construct a three-dimensional health status coordinate system. The three-dimensional health status coordinate system can display the data of mechanical wear, material degradation and lubrication failure in three dimensions to achieve an intuitive and comprehensive evaluation of the health status of rotating mechanical equipment. By marking the health status points in the three-dimensional coordinate system, the current health status position of the equipment is analyzed; according to the position changes of the health status points in the three-dimensional coordinate system, the health evolution trend of the mechanical equipment is tracked in real time. In actual operation, through dynamic observation of the health status points, the degradation mode and potential faults of the mechanical equipment can be identified, thereby providing data support for subsequent maintenance decisions.

[0027] S42 specifically includes: S421: The integral result of the vibration signal power spectrum is used to characterize the degree of mechanical wear. Specifically, the integral of the vibration signal power spectrum reflects the degree of energy consumption of the machine in different frequency bands and can measure the overall level of mechanical wear. The integral result of the vibration signal power spectrum is calculated using the following formula: ,in, is the integration result of the vibration signal, Frequency The power spectrum value under and are the frequency ranges used for calculating the integral, respectively. This result represents the total amount of energy in the vibration signal and can be used to evaluate the degree of mechanical wear. S422: The degree of material degradation is comprehensively reflected by combining the weighted sum of the bearing alloy grain boundary corrosion rate and the material dislocation density change rate. The corrosion rate and the dislocation density change rate reflect two different degradation mechanisms of the material. By taking the weighted average of these two indicators, the comprehensive material degradation degree is obtained. The calculation formula is: ,in, is the material degradation degree, is the intergranular corrosion rate of the bearing alloy, is the material dislocation density change rate, and is the weight coefficient, which reflects the influence ratio of corrosion rate and dislocation density change rate on material degradation; by adjusting and , and can flexibly adjust the evaluation criteria according to actual working conditions and material properties; S423: Changes in lubrication status are indicated by monitoring changes in the dielectric constant of the lubricating oil. Changes in the dielectric constant of the lubricating oil reflect the deterioration of the lubricating oil. The dielectric constant changes with aging and contamination of the lubricating oil. The degree of lubrication failure is calculated using the following formula: ,in, is the lubrication failure degree, is the dielectric constant value of the current lubricating oil, is the dielectric constant of the lubricating oil at the initial moment, is the dielectric constant value when the lubricating oil reaches maximum deterioration; this calculation can reflect the aging process of the lubricating oil and provide a basis for early warning of lubrication failure; the above steps quantify multiple indicators such as the vibration signal power spectrum, the bearing alloy grain boundary corrosion rate, the material dislocation density change rate, and the lubricating oil dielectric constant change into dimensions in a three-dimensional health status coordinate system through specific calculation formulas. This method can accurately assess the wear, material degradation and lubrication failure of rotating machinery in thermal power plants, providing a scientific basis for equipment maintenance.

[0028] S5 specifically includes: S51: Based on the three-dimensional health status coordinate system constructed in S4, the trajectory of the health status points in this coordinate system is tracked. Each health status point corresponds to the current health status of the mechanical equipment. By continuously updating the position of the health status points with real-time data, a dynamic trajectory of the health status points can be formed. The trajectory of the health status points represents the health evolution of the equipment over a certain period of time and can reveal trends in equipment wear, degradation, and lubrication failure. S52: Calculating a trajectory slope based on the motion trajectory of the health status point. The slope represents the speed of change of the health status point along the time axis in the three-dimensional coordinate system and is used to reveal the acceleration of the change in the health status of the equipment. When the speed of movement of the health status point in the coordinate system increases, the slope of the trajectory increases, which may indicate accelerated wear of the equipment or intensified material degradation. S53: By comparing the trajectory slope of the healthy state point with the slope of the fatigue limit curve, it is determined whether the fatigue bearing capacity of the material has been exceeded. When the trajectory slope exceeds the material fatigue limit curve, it indicates that the equipment has entered the critical fatigue state and maintenance action is required; S54: Automatically match the corresponding maintenance action based on the trajectory slope judgment result of the health status point; the above steps can quickly identify the accelerating trend of the equipment health status change through real-time tracking and slope judgment of the health status point movement trajectory, especially at the critical moment when the fatigue limit is about to be reached, and automatically match the most appropriate maintenance action. This method significantly improves the intelligent level of health management of rotating machinery in thermal power plants and reduces the risk of sudden failures.

[0029] The calculation of the trajectory slope in S52 specifically includes: S521: The position of each health status point in the three-dimensional health status coordinate system is determined by the real-time health status data of the mechanical equipment. The motion trajectory of the health status point is formed through continuous data update. The position coordinates of the health status point are expressed as ,in Indicates time, ,and Respectively represent the horizontal, vertical and vertical axis coordinates of the health state point in the three-dimensional coordinate system; S522: Calculate the rate of change of the health status point at different time points, and calculate the displacement between two adjacent moments according to the position change of the health status point in the time series. , get the speed of the healthy state point, set the adjacent time point to and , then the displacement of the healthy state point The calculation formula is: ,in, ,and Time points The coordinates of the health status point under ,and For time point The coordinates of the health status point under S523: Calculate the movement speed of the healthy state point based on the displacement and time difference between adjacent time points , the formula is: ,in, Indicates the time interval, The movement speed of health status points; S524: Calculate the slope of the motion trajectory of the health state point. The slope reflects the acceleration of the health state point along the time axis. The slope of the trajectory Calculated by the following formula: ,in, and Time points and The speed value under the condition of fatigue; through this calculation, the acceleration of the healthy state point can be obtained, thereby evaluating the acceleration of the change in the health state of the mechanical equipment; the above steps can accurately quantify the rate and acceleration of the change in the health state of the equipment by clearly calculating the slope of the motion trajectory of the healthy state point, helping to evaluate the wear and degradation trend of the equipment in real time. This method provides a scientific basis for subsequent maintenance decisions, enabling the equipment to take timely measures before entering the critical fatigue state to reduce the risk of failure.

[0030] S6 specifically includes: S61: After performing the maintenance operation, the position of the health status point is updated by re-collecting the sensor data stream obtained in S1; based on the new sensor data, the coordinate position of the health status point after maintenance is calculated. , and update the position of the health state point in the three-dimensional health state coordinate system; S62: Verify the effectiveness of maintenance measures by comparing the direction changes of the displacement vectors of the healthy state points before and after maintenance; the change of the vector is calculated by calculating the angle between the two To judge, a small angle indicates a good maintenance effect, while a large angle may indicate a poor maintenance effect; the calculation formula is: ,in, represents the dot product of the displacement vector of the healthy state point, and Respectively represent the modulus of the displacement vector of the healthy state point; the angle of the calculation result Can be used to judge the effect after maintenance; S63: Evaluate the effectiveness of maintenance measures based on the size of the angle. If the angle is less than 5 degrees, it means that the changes in the health status points are relatively consistent, indicating that the maintenance measures are successful and changing in the expected direction; if the angle If it is greater than 15 degrees, it means that the maintenance effect is not obvious or there are problems, and the change of the health status point is not developing in the expected direction; S64: Based on the change of the displacement vector of the health state point, automatically adjust the weight coefficient of the correlation matrix established in S3. Specifically, if the angle If the angle is less than 5 degrees, it means that the changes in health status points are relatively consistent, and the weight coefficients between related factors in the correlation matrix should be increased by 5% to 10%; if the angle is less than 5 degrees, it means that the changes in health status points are relatively consistent, and the weight coefficients between related factors in the correlation matrix should be increased by 5% to 10%; If the angle is greater than 15 degrees, it means that the health status of the equipment has not improved significantly. The weight coefficients of the relevant parameters in the correlation matrix should be reduced by 5% to 10%. In this way, the sensitivity and accuracy of the model can be dynamically optimized. The above steps compare the changes in the displacement vectors of the health status points before and after maintenance, and use specific angle values ​​(such as 5 degrees and 15 degrees) to judge the maintenance effect, which can accurately quantify the degree of improvement in the health status of the equipment. At the same time, the weight coefficients of the correlation matrix are dynamically adjusted according to the changes in the angle, which further improves the adaptability and accuracy of the health status assessment model, thereby making equipment maintenance decisions more scientific and accurate.

[0031] like Figure 2 As shown, a thermal power plant rotating machinery health management system based on the Internet of Things is used to implement the above-mentioned thermal power plant rotating machinery health management method based on the Internet of Things, including the following modules: Multimodal sensing module: used to deploy sensors at predetermined locations on turbine bearings and wind turbine gearboxes to collect vibration and acoustic emission signals, lubricating oil dielectric constant, and metal surface microstrain field data; Data processing and analysis module: used to input the time series data of the lubricating oil dielectric constant into the metal ion mobility analysis model based on quantum dot sensing to calculate the grain boundary corrosion rate of the bearing alloy; at the same time, it inverts the material dislocation density change rate through microstrain field data; Health status assessment module: Based on the bearing alloy grain boundary corrosion rate and material dislocation density change rate output by the data processing and analysis module, combined with the vibration acoustic emission signal power spectrum, a correlation matrix is ​​established; Health status modeling module: used to construct a three-dimensional health status coordinate system based on the correlation matrix output by the health status assessment module, where the horizontal axis is the mechanical wear degree, the vertical axis is the material degradation degree, and the vertical axis is the lubrication failure degree; Maintenance decision module: This module calculates the trajectory slope of the health point in the three-dimensional health coordinate system output by the health modeling module and compares it with the fatigue limit curve to determine whether the equipment has entered a critical fatigue state. When the trajectory slope exceeds the material fatigue limit curve, the module automatically selects the optimal maintenance action. Maintenance feedback and correction module: This module is used to re-collect multimodal sensor module data after performing maintenance actions, update the position of the health status point, and verify the effectiveness of the maintenance measures by comparing the changes in the displacement vector of the health status point before and after maintenance. At the same time, the association matrix weight coefficient is dynamically corrected according to the verification results.

[0032] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0033] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for health management of rotating machinery in thermal power plants based on the Internet of Things, characterized in that: The following steps are involved: S1: Multimodal sensors are deployed at predetermined locations on the turbine bearings and wind turbine gearboxes to simultaneously collect vibration and acoustic emission signals, lubricating oil dielectric constant, and metal surface microstrain field data; S2: Input the time series data of the dielectric constant of the lubricating oil obtained in S1 into the metal ion mobility analysis model based on quantum dot sensing to calculate the grain boundary corrosion rate of the bearing alloy, and at the same time invert the material dislocation density change rate through the microstrain field data; S3: Establish the correlation matrix of the power spectrum of the vibration acoustic emission signal, the grain boundary corrosion rate of the bearing alloy, and the change rate of the material dislocation density; S4: Construct a three-dimensional health status coordinate system based on the correlation matrix of S3, where the horizontal axis is the mechanical wear degree, the vertical axis is the material degradation degree, and the vertical axis is the lubrication failure degree; S5: Based on the motion trajectory of the healthy state point in S4 in the three-dimensional healthy state coordinate system, when the trajectory slope exceeds the material fatigue limit curve, maintenance actions are automatically matched. Specific maintenance actions include: Action 1: If lubrication failure is dominant, online oil filtration is triggered; Action 2: If mechanical wear is dominant, start laser cladding repair; Action 3: If the material degradation exceeds the limit, a spare parts replacement warning is issued; S6: After executing the S5 maintenance action, re-collect the S1 data stream and calculate the displacement vector of the healthy state point. Verify the effectiveness of the measures by comparing the changes in the vector direction before and after maintenance, and dynamically correct the weight coefficient of the correlation matrix.

2. The method for health management of rotating machinery in a thermal power plant based on the Internet of Things according to claim 1, characterized in that: Said S1 specifically includes: S11: Install the sensor bracket on the outer side of the turbine bearing and the side wall of the fan gearbox by mechanical fixing. The sensor bracket is made of high-temperature resistant alloy material. S12: Three types of sensors are installed on each sensor bracket, including a vibration sensor, a lubricating oil dielectric constant sensor, and a micro-strain sensor. The vibration sensor uses an accelerometer with a frequency response range of 20 Hz to 2 MHz and a sampling frequency of no less than 2 MHz. The lubricating oil dielectric constant sensor uses a capacitive sensor to monitor changes in the dielectric constant of the lubricating oil. The micro-strain sensor uses fiber Bragg grating technology to collect real-time micro-strain field data on the metal surface. S13: Data acquisition of each sensor collects vibration signals, lubricating oil dielectric constant data and micro-strain signals at a frequency of 1000 times per second, and ensures signal synchronization.

3. The method for health management of rotating machinery in a thermal power plant based on the Internet of Things according to claim 1, characterized in that: The S2 specifically includes: S21: The time series data of the dielectric constant of the lubricating oil obtained in S1 is transmitted to the data processing platform, and the time series data is filtered by the data processing module to remove high-frequency noise; S22: Based on quantum dot sensing technology, a metal ion mobility analysis model is constructed. This analysis model uses quantum dot surface modification technology to improve the sensitivity to metal ions, and combines liquid conductivity and temperature parameters to dynamically calculate the mobility of metal ions. The calculation formula is: ,in, is the current intensity measured by the sensor, is the initial metal ion concentration, is the time interval, is the metal ion mobility; S23: Calculate the intergranular corrosion rate of bearing alloys based on metal ion mobility data and changes in the dielectric constant of lubricating oil , the formula is: ,in, is a constant related to the bearing material and lubricant properties, is the change in dielectric constant of lubricating oil; S24: Using microstrain field data, calculate the material dislocation density change rate through inversion algorithm , the formula is: ,in, and are the dislocation densities of the material at the current moment and the initial moment, is the time interval.

4. The method for health management of rotating machinery in a thermal power plant based on the Internet of Things according to claim 1, characterized in that: The S3 specifically includes: S31: Perform fast Fourier transform processing on the vibration acoustic emission signal collected in S1, convert it into a frequency domain signal, and calculate the power spectrum; S32: Based on the bearing alloy intergranular corrosion rate and material dislocation density change rate calculated in S2, the numerical range of these two values ​​is unified to interval; S33: Combine the power spectrum data calculated in S31 and the corrosion rate and dislocation density change rate data normalized in S32 to establish a correlation matrix between the three. The structure of the correlation matrix is: ； in, Frequency The power spectrum value under For the moment The intergranular corrosion rate of bearing alloys under For the moment The material dislocation density change rate under For different frequency values, The index of the sampling point.

5. The method for health management of rotating machinery in a thermal power plant based on the Internet of Things according to claim 1, characterized in that: The S4 specifically includes: S41: Based on the correlation matrix in S3, extract the normalized values ​​of the vibration signal power spectrum, the bearing alloy grain boundary corrosion rate, and the material dislocation density change rate; and use these values ​​as the three basic dimensions in the coordinate system, defined as mechanical wear, material degradation, and lubrication failure. S42: Determine the axial scale of the three-dimensional health status coordinate system, where the horizontal axis is set to mechanical wear, using the integral result of the vibration signal power spectrum to characterize the degree of mechanical wear; the vertical axis is set to material degradation, which is comprehensively reflected by combining the weighted sum of the bearing alloy grain boundary corrosion rate and the material dislocation density change rate; the vertical axis is set to lubrication failure, which represents the change in lubrication status by monitoring the change in the dielectric constant of the lubricating oil; S43: Input the numerical values ​​of each dimension obtained in S41 and S42 into the three-dimensional coordinate system to construct a three-dimensional health status coordinate system. The three-dimensional health status coordinate system can display the data of mechanical wear, material degradation and lubrication failure in three dimensions. By marking the health status points in the three-dimensional coordinate system, the current health status position of the equipment can be analyzed.

6. The method for health management of rotating machinery in a thermal power plant based on the Internet of Things according to claim 5, characterized in that: The S42 specifically includes: S421: The integral of the vibration signal power spectrum reflects the energy consumption of the machine in different frequency bands and can measure the overall level of mechanical wear. The integral result of the vibration signal power spectrum is calculated using the following formula: ,in, is the integration result of the vibration signal, Frequency The power spectrum value under and are the frequency ranges for calculating the integral respectively; S422: The corrosion rate and dislocation density change rate reflect two different degradation mechanisms of the material. By taking the weighted average of these two indicators, the comprehensive material degradation degree is obtained. The calculation formula is: ,in, is the material degradation degree, is the intergranular corrosion rate of the bearing alloy, is the material dislocation density change rate, and are weight coefficients, reflecting the influence ratios of corrosion rate and dislocation density change rate on material degradation; S423: The change in the dielectric constant of the lubricating oil reflects the deterioration of the lubricating oil. The dielectric constant changes with the aging and contamination of the lubricating oil. The degree of lubrication failure is calculated using the following formula: ,in, is the lubrication failure degree, is the dielectric constant value of the current lubricating oil, is the dielectric constant of the lubricating oil at the initial moment, It is the dielectric constant value when the lubricating oil reaches the maximum degradation.

7. The method for health management of rotating machinery in a thermal power plant based on the Internet of Things according to claim 1, characterized in that: The S5 specifically includes: S51: Based on the three-dimensional health status coordinate system constructed in S4, the movement trajectory of the health status point in the coordinate system is tracked. Each health status point corresponds to the current health status of the mechanical equipment. The position of the health status point is updated by continuous real-time data, which can form a dynamic trajectory of the health status point. S52: Calculating a trajectory slope based on the motion trajectory of the health status point. The slope represents a change speed of the health status point along the time axis in the three-dimensional coordinate system, and is used to reveal the acceleration of the change of the health status of the device. S53: By comparing the trajectory slope of the healthy state point with the slope of the fatigue limit curve, it is determined whether the fatigue bearing capacity of the material has been exceeded. When the trajectory slope exceeds the material fatigue limit curve, it indicates that the equipment has entered the critical fatigue state and maintenance action is required; S54: Automatically matching corresponding maintenance actions based on the trajectory slope judgment result of the health status point.

8. The method for health management of rotating machinery in a thermal power plant based on the Internet of Things according to claim 7, characterized in that: The S52 specifically includes: S521: The position of each health status point in the three-dimensional health status coordinate system is determined by the real-time health status data of the mechanical equipment. The motion trajectory of the health status point is formed through continuous data update. The position coordinates of the health status point are expressed as ,in Indicates time, ,and Respectively represent the horizontal, vertical and vertical axis coordinates of the health state point in the three-dimensional coordinate system; S522: Calculate the rate of change of the health status point at different time points, and calculate the displacement between two adjacent moments according to the position change of the health status point in the time series. ; S523: Calculate the movement speed of the healthy state point based on the displacement and time difference between adjacent time points ; S524: Calculate the slope of the motion trajectory of the health state point. The slope reflects the acceleration of the health state point along the time axis. The slope of the trajectory Calculated by the following formula: ,in, and Time points and The speed value below.

9. The method for health management of rotating machinery in a thermal power plant based on the Internet of Things according to claim 1, characterized in that: The S6 specifically includes: S61: After performing the maintenance operation, the position of the health status point is updated by re-collecting the sensor data stream obtained in S1; S62: Verify the effectiveness of maintenance measures by comparing the direction changes of the displacement vectors of the healthy state points before and after maintenance; the change of the vector is calculated by calculating the angle between the two to judge; S63: Evaluate the effectiveness of maintenance measures based on the size of the angle. If the angle is less than 5 degrees, it means that the changes in the health status points are relatively consistent, indicating that the maintenance measures are successful and changing in the expected direction; if the angle If it is greater than 15 degrees, it means that the maintenance effect is not obvious or there are problems; S64: Based on the change of the displacement vector of the health state point, automatically adjust the weight coefficient of the correlation matrix established in S3. Specifically, if the angle If the angle is less than 5 degrees, it means that the changes in health status points are relatively consistent, and the weight coefficients between related factors in the correlation matrix should be increased by 5% to 10%; if the angle is less than 5 degrees, it means that the changes in health status points are relatively consistent, and the weight coefficients between related factors in the correlation matrix should be increased by 5% to 10%; If it is greater than 15 degrees, it means that the health status of the equipment has not improved significantly. The weight coefficient of the relevant parameters in the correlation matrix should be reduced by 5% to 10%.

10. A thermal power plant rotating machinery health management system based on the Internet of Things, used to implement a thermal power plant rotating machinery health management method based on the Internet of Things according to any one of claims 1 to 9, characterized in that: Includes the following modules: Multimodal sensing module: used to deploy sensors at predetermined locations on turbine bearings and wind turbine gearboxes to collect vibration and acoustic emission signals, lubricating oil dielectric constant, and metal surface microstrain field data; Data processing and analysis module: used to input the time series data of the lubricating oil dielectric constant into the metal ion mobility analysis model based on quantum dot sensing to calculate the grain boundary corrosion rate of the bearing alloy; at the same time, it inverts the material dislocation density change rate through microstrain field data; Health status assessment module: Based on the bearing alloy grain boundary corrosion rate and material dislocation density change rate output by the data processing and analysis module, combined with the vibration acoustic emission signal power spectrum, a correlation matrix is ​​established; Health status modeling module: used to construct a three-dimensional health status coordinate system based on the correlation matrix output by the health status assessment module, where the horizontal axis is the mechanical wear degree, the vertical axis is the material degradation degree, and the vertical axis is the lubrication failure degree; Maintenance decision module: This module calculates the trajectory slope of the health point in the three-dimensional health coordinate system output by the health modeling module and compares it with the fatigue limit curve to determine whether the equipment has entered a critical fatigue state. When the trajectory slope exceeds the material fatigue limit curve, the module automatically selects the optimal maintenance action. Maintenance feedback and correction module: This module is used to re-collect multimodal sensor module data after performing maintenance actions, update the position of the health status point, and verify the effectiveness of the maintenance measures by comparing the changes in the displacement vector of the health status point before and after maintenance. At the same time, the association matrix weight coefficient is dynamically corrected according to the verification results.

Images