Oxidation fan operation fault prediction and health management system

Abstract

The invention relates to the technical field of predictive health management of industrial equipment, in particular to an oxidation fan operation fault prediction and health management system. Comprising a data acquisition module used for acquiring operation environment parameters and downstream process loads of an oxidation fan; the first processing module is used for determining a pneumatic stability margin representing the operation stability of the fan; the second processing module is used for determining a corrosion fatigue crack growth rate representing blade damage accumulation; the health assessment module is used for determining a comprehensive health index representing the current state of the equipment and outputting a corresponding health state grade according to a preset health state grading threshold value; and the optimization control module is used for generating a self-adaptive optimization control strategy in response to the comprehensive health index and the health state level. According to the method, the one-sidedness that only a single physical field is concerned in the prior art is overcome, the real health state of the oxidation fan can be evaluated more accurately and comprehensively, and the accuracy of fault prediction is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of predictive health management technology for industrial equipment, specifically to a fault prediction and health management system for oxidation blowers. Background Technology

[0002] In industrial production, oxidation blowers are key power equipment to ensure the continuity of the process flow, and real-time monitoring of their health status is of utmost importance. Existing technologies for health monitoring of equipment such as oxidation fans have significant limitations; these methods typically focus on risks in a single physical domain, such as assessing the risk of aerodynamic instability during fan operation in isolation or analyzing fatigue damage to blade materials independently, without linking the two together. However, this fragmented monitoring approach ignores the strong coupling effect between the macroscopic operating environment and microscopic material damage. Specifically, aerodynamic instability can generate pressure pulsations, which can significantly accelerate the propagation of corrosion fatigue cracks in blade materials. This multi-physics coupling effect is a key factor leading to accelerated equipment damage and reduced lifespan, but it has been overlooked by traditional monitoring methods. Therefore, how to establish a health assessment model that integrates the coupling effects of multiple physical fields such as aerodynamics and materials, overcome the limitations of existing monitoring methods, and achieve accurate assessment and scientific prediction of the health status of wind turbines has become a technical problem that urgently needs to be solved in this field.

[0003] The information disclosed in the background section above is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention discloses an oxidation blower operation fault prediction and health management system. Specifically, the technical solution of this invention is as follows: An oxidation fan operation fault prediction and health management system, comprising: The data acquisition module is used to obtain the operating environment parameters of the oxidation blower and the downstream process load; The first processing module is used to determine the aerodynamic stability margin, which characterizes the operating stability of the wind turbine, based on the operating environment parameters. The second processing module is used to determine the corrosion fatigue crack propagation rate, which characterizes the accumulation of blade damage, based on the aerodynamic stability margin and operating environment parameters. The health assessment module is used to combine aerodynamic stability margin and corrosion fatigue crack propagation rate to determine a comprehensive health index that characterizes the current state of the equipment, and output the corresponding health status level according to the preset health status grading threshold. The optimization control module is used to generate adaptive optimization control strategies in response to the comprehensive health index and health status level.

[0005] Preferably, the first processing module is used to determine the aerodynamic stability margin based on operating environment parameters, including: It retrieves the current outlet pressure and flow rate measured by the sensors, as well as the air density from the operating environment parameters; Based on air density and flow rate, the surge pressure boundary is obtained by querying the surge pressure boundary function pre-set in the system; The aerodynamic stability margin is calculated by comparing the current outlet pressure with the surge pressure boundary.

[0006] Preferably, the second processing module is used to determine the corrosion fatigue crack propagation rate based on aerodynamic stability margin and operating environment parameters, including: The system calls upon the aerodynamic stability margin determined by the first processing module and the concentration of corrosive media in the operating environment parameters obtained by the data acquisition module. Based on a pre-defined crack propagation model coupled with aerodynamic stability, the aerodynamic stability margin is used as a dynamic stress influence factor, and the corrosion fatigue crack propagation rate is calculated by combining the concentration of the corrosive medium.

[0007] Preferably, the health assessment module is used to determine a comprehensive health index by combining aerodynamic stability margin and corrosion fatigue crack propagation rate, including: The preset minimum permissible aerodynamic stability margin and maximum permissible crack propagation rate are used as evaluation benchmarks; Based on the weighted Euclidean distance, the aerodynamic stability margin determined by the first processing module and the deviation of the corrosion fatigue crack propagation rate determined by the second processing module from the ideal safe point are calculated, thereby determining the comprehensive health index.

[0008] Preferably, the health assessment module is used to output a corresponding health status level based on a preset health status grading threshold, including: The comprehensive health index determined by the health assessment module is compared with the preset first and second warning thresholds; If the comprehensive health index is greater than or equal to the first warning threshold, the health status level will be determined as a safe state. If the comprehensive health index is less than the first warning threshold but greater than or equal to the second warning threshold, the health status level will be determined as a Level 1 warning. If the overall health index is less than the second warning threshold, the health status level will be determined as a level two warning.

[0009] Preferably, the optimization control module includes a damage evolution trend prediction unit, used for: Based on the corrosion fatigue crack propagation rate determined by the second processing module, the damage accumulation in the future prediction time domain is integrated over time to obtain the predicted future crack size. Based on the future crack size, the predicted health index for future moments is recalculated.

[0010] Preferably, the optimization control module includes a control cost function construction unit, used for: Identify the process performance items used to measure process load deviation; Based on the predicted health index determined by the damage evolution trend prediction unit, a health cost item is determined to measure the degree of equipment health degradation. By combining the process performance item and the health cost item, a dual-objective cost function is constructed.

[0011] Preferably, the optimization control module includes an adaptive strategy generation unit, used for: When the health status level is safe, a control strategy is generated that prioritizes meeting process load commands. When the health status level is at Level 1 warning, a control strategy is generated to smooth the rate of change of control commands. When the health status level is at the second-level warning level, a protective priority strategy based on health cost items is generated, and the wind turbine operating parameters are restricted within the dynamic safety boundary.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention establishes a coupling model between macroscopic aerodynamic instability and microscopic material damage, overcoming the limitations of existing technologies that focus only on a single physical domain. This model enables a more accurate and comprehensive assessment of the true health status of oxidation fans, significantly improving the accuracy of fault prediction. 2. This invention not only assesses the current state of the equipment but also scientifically predicts future health trends based on a damage accumulation model. This represents a shift from traditional post-fault maintenance to proactive predictive health management, providing a forward-looking basis for maintenance decisions. 3. This invention constructs a complete closed-loop management system from health assessment and trend prediction to optimized control. It can dynamically generate adaptive control strategies based on real-time health levels, achieving a dynamic balance between meeting production process requirements and ensuring long-term equipment health. 4. This invention proactively avoids high-risk operating conditions through forward-looking optimized control, effectively slowing down the damage accumulation rate of key components such as fan blades. This extends the overall service life of the equipment, improves operational reliability, and strongly guarantees the continuous and stable operation of industrial production. Attached Figure Description

[0013] The present invention will be further explained below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart of the system of the present invention. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. Example 1:

[0015] Please see Figure 1 An oxidation fan operation fault prediction and health management system, comprising: The data acquisition module is used to obtain the operating environment parameters of the oxidation blower and the downstream process load; The first processing module is used to determine the aerodynamic stability margin, which characterizes the operating stability of the wind turbine, based on the operating environment parameters. The second processing module is used to determine the corrosion fatigue crack propagation rate, which characterizes the accumulation of blade damage, based on the aerodynamic stability margin and operating environment parameters. The health assessment module is used to combine aerodynamic stability margin and corrosion fatigue crack propagation rate to determine a comprehensive health index that characterizes the current state of the equipment, and output the corresponding health status level according to the preset health status grading threshold. The optimization control module is used to generate adaptive optimization control strategies in response to the comprehensive health index and health status level. This invention provides a fault prediction and health management system for oxidation fans. This system aims to address the one-sidedness of existing technologies in monitoring the health status of critical equipment like oxidation fans, which focuses only on a single physical domain while neglecting the accelerating effect of multi-physics coupling on equipment damage. This system constructs a complete causal chain model from the macroscopic operating environment to microscopic material damage, enabling accurate assessment of the fan's health status, scientific prediction of future trends, and closed-loop optimized control. In a specific implementation scenario, this system includes the following core modules: The data acquisition module aims to provide real-time and accurate raw data input for all subsequent analysis and decision-making. In this embodiment, this module achieves this through multiple sensors deployed at the air inlet and outlet of the oxidation blower and in the surrounding environment. For example, it continuously acquires the operating environment parameters of the oxidation blower, such as calculating the intake air density by combining temperature and pressure sensors with altitude information. Measure the concentration of corrosive media using chemical sensors Temperature sensors are used to monitor the operating temperature of key components. At the same time, it also monitors the downstream process load. This parameter reflects the production demand for the fan output; in addition, the real-time speed of the fan is obtained through a speed sensor to determine the principal stress cycle frequency. These multi-dimensional data constitute the dynamic input source of the system model. The first processing module aims to quantify the macroscopic stability of the wind turbine operation in real time. In this embodiment, based on the operating environment parameters, the module focuses on evaluating the distance between the wind turbine and the aerodynamic instability boundary, and finally determines a core indicator: the aerodynamic stability margin that characterizes the wind turbine's operational stability. This margin value is a key prerequisite for subsequent coupling analysis. The second processing module aims to reveal the intrinsic mechanism of macroscopic operational instability on microscopic material damage. In this embodiment, the innovation of this module is that it does not calculate material fatigue in isolation, but determines the corrosion fatigue crack propagation rate characterizing the accumulation of blade damage through a coupled model based on the aerodynamic stability margin calculated by the first processing module, the operating environment parameters, and the current crack size tracked by the system. The purpose of the health assessment module is to integrate risk information from different physical domains into a single, intuitive health indicator for maintenance personnel and control systems to quickly understand and respond to. In this embodiment, the module combines aerodynamic stability margin and corrosion fatigue crack propagation rate to calculate a comprehensive health index characterizing the current state of the equipment. Based on a preset health status grading threshold, the module transforms the complex index value into a clear health status level. The purpose of the optimized control module is to proactively adjust the operation strategy of the fan based on the current health assessment results, thereby shifting from passively responding to faults to proactive health management. In this embodiment, the module responds to the comprehensive health index and health status level by solving a bi-objective optimization problem and automatically generating an adaptive optimization control strategy to achieve a dynamic balance between meeting process requirements and ensuring equipment health. The system in this embodiment, through the collaborative work of the above modules, establishes for the first time a complete closed loop of environment-aerodynamics-material damage-control in the field of oxidation blowers. It can not only assess the current health status, but also predict future evolution trends and automatically optimize control strategies accordingly. Compared with the prior art, this invention deeply couples the originally isolated aerodynamic instability risk and material damage risk, greatly improving the accuracy of fault prediction and the precision of health management, thereby effectively extending the service life of equipment and ensuring production continuity. Example 2:

[0016] The first processing module is used to determine the aerodynamic stability margin based on operating environment parameters, including: It retrieves the current outlet pressure and flow rate measured by the sensors, as well as the air density from the operating environment parameters; Based on air density and flow rate, the surge pressure boundary is obtained by querying the surge pressure boundary function pre-set in the system; The aerodynamic stability margin is calculated by comparing the current outlet pressure with the surge pressure boundary. Based on Example 1, the first processing module in this embodiment aims to provide a real-time, quantitative assessment of the aerodynamic stability risk of the wind turbine. In this embodiment, the module determines the aerodynamic stability margin through the following steps: To determine the aerodynamic stability margin, the module calls upon the current outlet pressure and flow rate measured by the sensors, as well as the air density in the operating environment parameters; This refers to the current outlet pressure of the fan, which represents the fan's ability to perform work on the downstream system. It is measured in real time by a pressure sensor installed in the fan's outlet pipe. This refers to the current volumetric flow rate of the fan, which characterizes the fan's conveying capacity. It is measured in real time by a flow meter installed in the fan's duct. This refers to the air density at the fan inlet. Its function is to correct the fan's performance benchmark under different environmental conditions, based on the ambient temperature. Parameters such as altitude are obtained through calculations using a standard atmospheric physics model; The module is based on air density. With traffic Query the surge pressure boundary function pre-set in the system Obtain the surge pressure boundary; Surge pressure boundary This refers to a specific air density and traffic Under certain conditions, the highest outlet pressure limit under which the fan can operate stably; this functional relationship is predetermined through performance testing or high-precision computational fluid dynamics simulation before the equipment leaves the factory, and stored in the system memory in the form of a data table or function to ensure efficient querying; based on the surge pressure boundary, the module uses the current outlet pressure Boundary with surge pressure By comparing the results, the aerodynamic stability margin was calculated. This calculation follows the standard definition in the field of aerodynamics and is specifically achieved through the following formula:

[0017] in, This refers to the aerodynamic stability margin, which is a dimensionless percentage. Its physical meaning is the relative distance between the current operating point and the surge boundary. The smaller the value, the closer the fan is to the instability state. This value is calculated in this step and passed to the second processing module as the core output. Example 3:

[0018] The second processing module is used to determine the corrosion fatigue crack propagation rate based on aerodynamic stability margin and operating environment parameters, including: The system calls upon the aerodynamic stability margin determined by the first processing module and the concentration of corrosive media in the operating environment parameters obtained by the data acquisition module. Based on a pre-defined crack propagation model coupled with aerodynamic stability, the aerodynamic stability margin is used as a dynamic stress influence factor, and the corrosion fatigue crack propagation rate is calculated by combining the concentration of the corrosive medium. Based on Example 2, the core innovation of the second processing module in this embodiment is that it does not use the traditional fatigue model that only considers mechanical stress, but establishes a damage accumulation model that couples macroscopic aerodynamic behavior; its purpose is to accurately simulate the accelerating effect of the combined effect of airflow pulsation and corrosive environment on the propagation of microcracks in the blade under real operating conditions. The specific implementation method of this module is as follows: The module calls the aerodynamic stability margin determined by the first processing module. And the concentration of corrosive media in the operating environment parameters acquired by the data acquisition module.

[0019] Based on the aforementioned inputs, the module employs a pre-defined crack propagation model coupled with aerodynamic stability, using the aerodynamic stability margin as a dynamic stress influence factor, and combining it with the concentration of the corrosive medium to calculate the corrosion fatigue crack propagation rate. Its underlying logic is that when the wind turbine's operating point approaches the surge boundary, i.e. As the pressure is reduced, the airflow pulsation in the blade passage will increase dramatically. This high-frequency pressure pulsation will generate an additional dynamic alternating stress on the blade surface, which is a key damage factor that has not been considered in traditional models. This embodiment improves upon the classic Paris formula and proposes the following coupling model:

[0020] in, The corrosion fatigue crack propagation rate, with dimensions of length / cycle, is calculated using this formula; it characterizes the propagation length of the microcrack tip on the blade under one stress cycle. The material constants, whose dimensions need to be consistent with the overall formula, are calibrated by fitting the experimental data through standard corrosion fatigue tests on the materials used in wind turbine blades. Mechanical stress intensity factor, with dimensions of It is mainly determined by the centrifugal force generated by the blade rotation and the stable aerodynamic load; it is calculated based on the current fan speed and operating conditions, through a finite element analysis response surface model pre-installed in the system or the following standard empirical formula: in It is the average stress of key parts of the blade calculated based on the rotational speed. It depends on the correction factor for the crack and blade geometry. This is the current crack size; Aerodynamic stability margin, dimensionless, is calculated in real time by the first processing module; Aerodynamic-damage coupling factor, dimensions and Same as Its physical meaning is when the aerodynamic stability margin The magnitude of the additional dynamic stress intensity factor excited by airflow pulsation when the temperature drops to one reference unit; this parameter is a key innovative parameter of the present invention, and its value needs to be calibrated for a specific type of fan. The concentration of the corrosive medium, such as mg / m³, is measured in real time by the chemical sensors in the data acquisition module; The corrosion sensitivity coefficient, whose dimensions are related to the concentration of the corrosive medium. The dimensions of the terms are reciprocals of each other to ensure that the product terms are reciprocals. It is a dimensionless value, determined from material experimental data; To further clarify the calibration process of the model parameters and ensure its feasibility, the following supplementary explanation is provided in this embodiment: To calibrate the aerodynamic-damage coupling factor It can perform a series of fluid-structure interaction simulations; in each simulation, a specific, constant aerodynamic stability margin is set. The additional dynamic stress intensity factor induced by this was calculated. By analyzing multiple sets of data points Perform regression analysis, such as fitting the data using the least squares method. The relationship can be determined by the formula. The value; It should be noted that the inverse relationship here... It is a first-order approximation of complex aeroelastic effects, aiming to capture This reduces the core trend that accelerates damage; in scenarios requiring higher precision, this term can also be replaced with a more complex nonlinear function calibrated experimentally or through simulation. At the same time, aerodynamic influence items With mechanical stress terms Linear superposition is also an approximation method designed to simplify calculations. In scenarios requiring higher accuracy, it can be replaced by more complex nonlinear coupling functions calibrated through experiments or simulations.

[0021] To ensure the numerical stability of the model, The value sets an actual minimum threshold. For example, 0.5%; when calculated When the value is less than this threshold, the model uniformly adopts... It performs calculations and directly triggers the highest level stall warning. This avoids... The model singularity problem caused when it approaches zero; To calibrate material constants Corrosion sensitivity coefficient A series of standard corrosion fatigue tests need to be conducted on the blade material samples; during the tests, different ranges of constant stress intensity factor are controlled. and constant concentration of corrosive media And record the corresponding crack propagation rate. Based on a large number of experimental data points, the combination of parameters that best describes the material properties can be fitted through multivariate nonlinear regression analysis.

[0022] In more sophisticated models, material constants and corrosion sensitivity coefficient Can be regarded as temperature The function, i.e. Its functional relationship needs to be calibrated through material experiments conducted at different temperatures.

[0023] Example 4: The health assessment module, which combines aerodynamic stability margin and corrosion fatigue crack propagation rate, determines a comprehensive health index, including: The preset minimum permissible aerodynamic stability margin and maximum permissible crack propagation rate are used as evaluation benchmarks; Based on the weighted Euclidean distance, the aerodynamic stability margin determined by the first processing module and the deviation of the corrosion fatigue crack propagation rate determined by the second processing module from the ideal safe point are calculated, thereby determining the comprehensive health index. The health assessment module is used to output the corresponding health status level based on preset health status grading thresholds, including: The comprehensive health index determined by the health assessment module is compared with the preset first and second warning thresholds; If the comprehensive health index is greater than or equal to the first warning threshold, the health status level will be determined as a safe state. If the comprehensive health index is less than the first warning threshold but greater than or equal to the second warning threshold, the health status level will be determined as a Level 1 warning. If the overall health index is less than the second warning threshold, the health status level will be determined as a level two warning. Based on Example 3, this embodiment transforms multi-dimensional risk assessment results into unified, decision-making-friendly indicators. The health assessment module calculates a comprehensive health index and classifies the status. The calculation process of this module calls upon the preset minimum permissible aerodynamic stability margin. With the maximum allowable crack propagation rate As an evaluation benchmark; This refers to the minimum safe threshold for aerodynamic stability margin set according to design specifications and safe operating experience. It is set according to industry standards or equipment design manuals, for example, 15%. This refers to the maximum allowable value of crack propagation rate determined based on the material's fracture toughness, design life, and safety factor. It is determined through fracture mechanics analysis combined with safety margin requirements. Based on this, the module calculates the aerodynamic stability margin determined by the first processing module using weighted Euclidean distance. The corrosion fatigue crack propagation rate determined by the second processing module. With the ideal safe point The degree of deviation is used to determine the comprehensive health index. The technical motivation behind this algorithm lies in its ability to intuitively measure the distance between the current state and the absolutely safe state in a multi-dimensional risk space; the calculation formula is as follows:

[0024] By adopting The function ensures the overall health index. The calculation result will not be lower than 0, ensuring that its value range is strictly maintained within the range specified by the calculation. Within the range; in, The comprehensive health index is dimensionless, and its value ranges from 0 to 1. The closer it is to 1, the healthier the equipment is. The weighting coefficients are dimensionless and satisfy the following conditions: It is set based on the experience of operation and maintenance experts or specific operating conditions; for example, during periods of drastic fluctuations in downstream process load, the setting can be dynamically increased. The weighting is to emphasize the focus on aerodynamic stability; After calculating the comprehensive health index The module then immediately compares it with a preset threshold to output a discrete and clear health level; the module will then integrate the health index. The warning is compared with a preset first warning threshold and a second warning threshold. In this embodiment, the first warning threshold is set to 0.8 and the second warning threshold is set to 0.5. These thresholds are set based on statistical analysis of a large amount of historical operating data and failure cases, and are intended to balance the sensitivity and reliability of the warning. The hierarchical logic is as follows: If the comprehensive health index Greater than or equal to the first warning threshold The health status level will then be determined as a safe state. If the comprehensive health index Less than the first warning threshold but greater than or equal to the second warning threshold The health status level will then be set as Level 1 warning; If the comprehensive health index Less than the second warning threshold The health status level will then be set as Level 2 warning.

[0025] Example 5: The optimization control module includes a damage evolution trend prediction unit, used for: Based on the corrosion fatigue crack propagation rate determined by the second processing module, the damage accumulation in the future prediction time domain is integrated over time to obtain the predicted future crack size. Based on the future crack size, the predicted health index for future moments is recalculated; The optimization control module includes a control cost function construction unit, used for: Identify the process performance items used to measure process load deviation; Based on the predicted health index determined by the damage evolution trend prediction unit, a health cost item is determined to measure the degree of equipment health degradation. By combining process performance and health cost items, a dual-objective cost function is constructed. The optimization control module includes an adaptive policy generation unit, used for: When the health status level is safe, a control strategy is generated that prioritizes meeting process load commands. When the health status level is at Level 1 warning, a control strategy is generated to smooth the rate of change of control commands. When the health status level is a Level 2 warning, a protective priority strategy based on health cost items is generated, and the wind turbine operating parameters are restricted within the dynamic safety boundary. After determining the current health status level, the optimization control module initiates a forward-looking, closed-loop health management process; this process logically includes three closely linked links: prediction of future health status, quantitative construction of control costs, and generation of adaptive control strategies. In the prediction phase, the damage evolution trend prediction unit aims to proactively extrapolate the impact of specific control strategies on the future health status of the equipment; this unit is based on the current corrosion fatigue crack propagation rate determined by the second processing module. For future prediction time domain By integrating the accumulated damage over time, the predicted future crack size can be obtained. The physical basis of this step is to obtain the cumulative quantity by integrating the rate over time:

[0026] in, The predicted crack size, in length, is the core output of this step. The crack size at the current moment, in units of length; when the system is first run, it is set to a standard initial micro-defect size determined based on material purity and manufacturing process. In subsequent operation, The value is periodically updated by a data fusion unit: this unit uses the model's prediction from the previous period as a basis, and periodically, or upon receiving new nondestructive testing data, performs Kalman filtering or a similar state estimation method to fuse the actual measured value with the model's prediction, thereby calibrating and updating the current crack size. This effectively avoids the accumulation of errors in pure model predictions; The instantaneous crack propagation rate is calculated in real time by the model of the second processing module based on the predicted future working conditions; Principal stress cycle frequency, in units of Its physical meaning is the number of stress cycles that the wind turbine blades endure per second, which is obtained from control system commands or speed sensors; The prediction time domain, measured in time, refers to the length of future time that needs to be predicted. Based on predicted future crack size In addition to forecasting future operating conditions, this unit re-substitutes these forecasts into the aforementioned model to recalculate and obtain the predicted health index for future times.

[0027] To achieve proactive optimization, this module incorporates a dynamic performance response model for the oxidation blower; this model can respond to input control command sequences. For example, a series of future target speeds for wind turbines can be used to predict the operating parameters of the wind turbines in the future time domain, such as outlet pressure. and traffic And the final downstream process load Based on these predicted operating parameters, the system can further invoke the models of the first and second processing modules to calculate the predicted health index. These predictions provide the basis for calculating the cost function; the dynamic performance response model can be a first- or second-order transfer function model based on physical mechanisms, or it can be a nonlinear autoregressive model or a neural network model trained using historical operating data through system identification methods. In the quantitative construction phase, the control cost function construction unit is based on the predicted future health index. A mathematical framework for evaluating the merits of different alternative control strategies is created; this unit constructs a bi-objective cost function using the following formula.

[0028]

[0029] It evaluates an alternative control sequence u in the future prediction time domain. The lower the overall performance score, the better; among them, It is a predictive time domain used to evaluate control sequences, and its unit is time. Its value is set according to the response speed of the control system and the characteristics of the process flow. in, It is a process performance item used to measure process load deviation, and it penalizes any behavior that deviates from the production order; This is the target load value for downstream process requirements, while In the control sequence Under the influence of the model, the actual process load output is predicted. The process load reference value used for normalization, such as rated load or maximum load, serves to make the process performance terms dimensionless; Based on predictive health index A defined health cost item used to measure the degree of equipment health degradation penalizes any control actions that lead to future deterioration of health. To achieve adaptive control, weighting coefficients and Designed to align with current health indices Dynamic association, and satisfying in It can be a constant; for example, it can be set to... and It is a preset constant; when the device is healthy, that is... When the value is high, When the health condition is relatively small, the cost function primarily satisfies the process requirements; however, when health deteriorates, i.e. When the value is low, The sharp increase forces optimization algorithms to prioritize health cost terms. In the strategy generation stage, the adaptive strategy generation unit solves the cost function in each control cycle using numerical optimization algorithms such as Model Predictive Control (MPC), Sequential Quadratic Programming (SQP), or Genetic Algorithm. Minimum optimal control sequence And based on the current health status level, implement differentiated control outputs; When the health status level is safe. With lower weights, the optimization results are generated to meet the process load instructions. Priority control strategy; When the health status level is at Level 1 warning By increasing the weights, the optimization algorithm generates control strategies that smooth the rate of change of control commands. For example, the adjustment slope of the fan speed is limited to 50%-80% of the normal value to avoid sudden drops due to drastic parameter changes. value; When the health status level is a Level 2 warning Weights dominate the cost function, and the optimization algorithm generates a protective priority strategy dominated by the health cost term; the system will forcibly limit the wind turbine operating parameters to those based on the predicted health index. Within a dynamic safety boundary calculated in real time, for example, the maximum permissible speed or load response capability can be temporarily reduced by 20%-40%, and an alarm can be issued. The model framework proposed in this embodiment focuses on the core damage mechanism of aerodynamic-corrosion-fatigue coupling, providing an innovative solution for achieving refined health management. It should be noted that this model is an extensible framework, and in future deployments, other key influencing factors can be further integrated; for example, by using the corrosion sensitivity coefficient... and material constants Modeled as operating temperature The function can incorporate thermal effects into damage assessment; by introducing vibration sensor data, it is possible to identify and quantify the additional impact of mechanical vibration sources other than pneumatic pulsation on fatigue life; this modular scalability ensures the practicality and comprehensiveness of this system in dealing with more complex industrial scenarios.

[0030] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention; any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0031] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A fault prediction and health management system for an oxidation blower, characterized in that, include: The data acquisition module is used to obtain the operating environment parameters of the oxidation blower and the downstream process load; The first processing module is used to determine the aerodynamic stability margin, which characterizes the operating stability of the wind turbine, based on the operating environment parameters. The second processing module is used to determine the corrosion fatigue crack propagation rate, which characterizes the accumulation of blade damage, based on the aerodynamic stability margin and operating environment parameters. The health assessment module is used to combine aerodynamic stability margin and corrosion fatigue crack propagation rate to determine a comprehensive health index that characterizes the current state of the equipment, and output the corresponding health status level according to the preset health status grading threshold. The optimization control module is used to generate adaptive optimization control strategies in response to the comprehensive health index and health status level.

2. The oxidation fan operation fault prediction and health management system according to claim 1, characterized in that, The first processing module is used to determine the aerodynamic stability margin based on operating environment parameters, including: It retrieves the current outlet pressure and flow rate measured by the sensors, as well as the air density from the operating environment parameters; Based on air density and flow rate, the surge pressure boundary is obtained by querying the surge pressure boundary function pre-set in the system; The aerodynamic stability margin is calculated by comparing the current outlet pressure with the surge pressure boundary.

3. The oxidation fan operation fault prediction and health management system according to claim 2, characterized in that, The second processing module is used to determine the corrosion fatigue crack propagation rate based on aerodynamic stability margin and operating environment parameters, including: The system calls upon the aerodynamic stability margin determined by the first processing module and the concentration of corrosive media in the operating environment parameters obtained by the data acquisition module. Based on a pre-defined crack propagation model coupled with aerodynamic stability, the aerodynamic stability margin is used as a dynamic stress influence factor, and the corrosion fatigue crack propagation rate is calculated by combining the concentration of the corrosive medium.

4. The oxidation fan operation fault prediction and health management system according to claim 3, characterized in that, The health assessment module is used to determine a comprehensive health index by combining aerodynamic stability margin and corrosion fatigue crack propagation rate, including: The preset minimum permissible aerodynamic stability margin and maximum permissible crack propagation rate are used as evaluation benchmarks; Based on the weighted Euclidean distance, the aerodynamic stability margin determined by the first processing module and the deviation of the corrosion fatigue crack propagation rate determined by the second processing module from the ideal safe point are calculated, thereby determining the comprehensive health index.

5. The oxidation fan operation fault prediction and health management system according to claim 4, characterized in that, The health assessment module is used to output a corresponding health status level based on a preset health status grading threshold, including: The comprehensive health index determined by the health assessment module is compared with the preset first and second warning thresholds; If the comprehensive health index is greater than or equal to the first warning threshold, the health status level will be determined as a safe state. If the comprehensive health index is less than the first warning threshold but greater than or equal to the second warning threshold, the health status level will be determined as a Level 1 warning. If the overall health index is less than the second warning threshold, the health status level will be determined as a level two warning.

6. The oxidation fan operation fault prediction and health management system according to claim 5, characterized in that, The optimization control module includes a damage evolution trend prediction unit, used for: Based on the corrosion fatigue crack propagation rate determined by the second processing module, the damage accumulation in the future prediction time domain is integrated over time to obtain the predicted future crack size. Based on the future crack size, the predicted health index for future moments is recalculated.

7. The oxidation fan operation fault prediction and health management system according to claim 6, characterized in that, The optimization control module includes a control cost function construction unit, used for: Identify the process performance items used to measure process load deviation; Based on the predicted health index determined by the damage evolution trend prediction unit, a health cost item is determined to measure the degree of equipment health degradation. By combining the process performance item and the health cost item, a dual-objective cost function is constructed.

8. The oxidation fan operation fault prediction and health management system according to claim 7, characterized in that, The optimization control module includes an adaptive strategy generation unit, used for: When the health status level is safe, a control strategy is generated that prioritizes meeting process load commands. When the health status level is at Level 1 warning, a control strategy is generated to smooth the rate of change of control commands. When the health status level is at the second-level warning level, a protective priority strategy based on health cost items is generated, and the wind turbine operating parameters are restricted within the dynamic safety boundary.

Images