Abrasion prevention and control system and method for low-temperature economizer

Abstract

The invention relates to the field of abrasion prevention and control, and discloses a low-temperature economizer abrasion prevention and control system and method which are used for achieving accurate prediction and active prevention and control of low-temperature economizer abrasion through a multi-model fusion innovative method. The abrasion prevention and control method for the low-temperature economizer comprises the steps that firstly, a time-varying memory kernel function is established to quantify historical working condition influences; a space-time risk map is generated through a cellular automaton in combination with the flow field parameters; phase-space reconstruction and nonlinear kinetic analysis are carried out on the atlas to establish early warning indexes; wear mode characteristics are extracted through topological data analysis; and generating accurate prevention and control parameters based on the atlas and topological characteristics. Fractional calculus, cellular automaton, nonlinear dynamics, topological data analysis and variational optimization are organically fused to form a technical chain, conversion from passive early warning to active guide control is realized, prediction precision, early warning perspectiveness and prevention and control pertinence are improved, the service life of equipment is effectively prolonged, and operation and maintenance cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of wear control, and more particularly to a wear control system and method for a cryogenic economizer. Background Technology

[0002] As a key component of waste heat recovery systems in coal-fired power plants, metallurgy, chemical industries, and other sectors, cryogenic economizers operate under harsh flue gas environments characterized by high ash content and high flow rates. Tube wall wear is a core challenge affecting the safety and economy of equipment operation. In recent years, although many studies have attempted to introduce digital methods, most remain limited to machine learning models based on historical data, exhibiting inherent limitations in terms of principle, interpretability, and the ability to extrapolate operating conditions. The existing technology suffers from the following shortcomings: 1. Most existing wear models are based on integer-order differential equations or linear accumulation assumptions, treating wear as an instantaneous process only related to the current operating conditions. This fails to quantify the nonlinear memory effect of historical operating conditions on current material damage. Consequently, the prediction accuracy of these models decreases significantly under complex operating conditions such as varying operating conditions and intermittent impacts. 2. Traditional methods can typically only predict the total wear amount or average rate at a specific point, lacking the ability to predict the spatial distribution characteristics of wear patterns and their evolution path over time. They cannot anticipate the emergence of high-risk patterns such as wear grooves and localized pits, resulting in a lack of targeted prevention and control measures. 3. Existing early warning systems are mostly based on fixed wear thresholds, which are "post-event" or "in-event" alarms. They lack early, forward-looking identification mechanisms for the transition of the system from stable wear to unstable abrupt changes, and cannot provide sufficient early warning time before wear accelerates and deteriorates.

[0003] Therefore, we propose a low-temperature economizer wear control system and method to solve the above problems. Summary of the Invention

[0004] This invention provides a system and method for preventing and controlling wear of low-temperature economizers, which uses an innovative multi-model fusion approach to achieve accurate prediction and proactive prevention and control of wear in low-temperature economizers.

[0005] The first aspect of this invention provides a method for preventing and controlling wear in a cryogenic economizer. This method includes: constructing and solving equations based on the economizer's history and data to establish a time-varying memory kernel function; combining the time-varying memory kernel function with flow field distribution parameters and generating a spatiotemporal risk map through cellular automata calculations; reconstructing and analyzing the regional evolution sequences in the spatiotemporal risk map to establish a system instability early warning index; performing topological data analysis on the spatiotemporal risk map to extract a set of topological invariants; and solving the wear distribution functional using a variational optimization algorithm based on the spatiotemporal risk map and the set of topological invariants to generate prevention and control parameters.

[0006] Optionally, in the first implementation of the first aspect of the present invention, temperature, pressure, and flow rate data of the economizer sensing system are collected, and particulate matter concentration distribution data are acquired simultaneously. The data is time-aligned and normalized to form multi-physics time series data. Based on fractional calculus theory, the historical cumulative effect is modeled as a dynamic system, and a fractional differential equation framework is established. The fractional differential equation framework is processed to transform it into a convolution integral form, and a weighted kernel function is constructed. According to the characteristics and data of the economizer tube wall material, the key parameters in the weighted kernel function are calibrated to form a time-varying memory kernel function. The multi-physics time series data is input into the time-varying memory kernel function for calculation to obtain the quantified contribution weight and generate a memory weight distribution matrix.

[0007] Optionally, in the second implementation of the first aspect of the present invention, a mesh model of the economizer tube wall is established, and the time-varying memory kernel function is mapped onto the mesh model to form an initial damage field; flow field distribution parameters are obtained, and the parameters are coupled with the initial damage field to construct a local environmental state vector; transformation rules are defined based on the material wear mechanism to establish a state evolution function; parallel iterative calculation of the cellular automaton is initiated to perform evolution, and all cells update their wear states based on the state evolution function; a spatiotemporal risk map is generated through the parallel iterative calculation.

[0008] Optionally, in a third implementation of the first aspect of the present invention, key regions are identified from the spatiotemporal risk map, and the state value changes of the regions are extracted to form a wear evolution time series; the wear evolution time series is reconstructed in phase space, and by determining the embedding dimension and parameters, the one-dimensional time series is reconstructed into a trajectory in a high-dimensional phase space to establish a reconstructed phase space; in the reconstructed phase space, the divergence rate is calculated, and a chaotic characteristic quantification index is obtained through the maximum Lyapunov exponent; based on the chaotic characteristic quantification index and combined with the analysis in the reconstructed phase space, a system instability early warning index is established; the system instability early warning index is compared with a preset safety threshold to generate a system stability assessment result.

[0009] Optionally, in the fourth implementation of the first aspect of the present invention, the wear depth distribution data in the spatiotemporal risk map is converted into three-dimensional spatial point cloud data; a simple complex sequence is constructed from the three-dimensional spatial point cloud data to track the generation and disappearance during the scale change process; the lifetime distribution of topological features in the simple complex sequence is recorded to generate a topological feature persistence map; topological invariants are extracted from the topological feature persistence map to form a set of topological invariants; the set of topological invariants is compared with a topological feature library to establish a topological feature description.

[0010] Optionally, in the fifth implementation of the first aspect of the present invention, based on the wear depth spatial distribution data provided by the spatiotemporal risk map, an optimized functional is constructed to establish a wear distribution optimization model; prevention and control strategy parameters are determined and introduced into the wear distribution optimization model; the wear distribution optimization model is solved using a variational method to establish a mapping relationship; an iterative optimization algorithm is constructed to solve the mapping relationship to generate an optimized parameter solution set; the optimized parameter solution set is converted into a sequence of control instructions for the actuator to form a precise prevention and control parameter set.

[0011] Optionally, in the sixth implementation of the first aspect of the present invention, after the actuator is driven to act based on the precise prevention and control parameter set, multi-physics response data is collected to generate a prevention and control effect verification dataset; the prevention and control effect verification dataset is analyzed with the prediction results of the spatiotemporal risk map to generate a prevention and control effect evaluation report; based on the consistency index in the prevention and control effect evaluation report, the fractional-order parameters and state transition rule parameters are adjusted to generate model parameter optimization instructions; the model parameter optimization instructions are fed back to the fractional-order derivative model and the cellular automata model to achieve online self-correction and establish a wear prediction and prevention and control closed-loop system; based on the corrected model parameters, each step is re-executed to optimize the precise prevention and control parameter set.

[0012] A second aspect of this invention provides a low-temperature economizer wear prevention and control system, comprising: a processing module for constructing and solving equations based on the economizer's history and data, and establishing a time-varying memory kernel function; a prediction module for combining the time-varying memory kernel function with flow field distribution parameters, and generating a spatiotemporal risk map through cellular automata calculations; an evaluation module for reconstructing and analyzing the regional evolution sequences in the spatiotemporal risk map, and establishing system instability early warning indicators; an identification module for performing topological data analysis on the spatiotemporal risk map and extracting a set of topological invariants; and an output module for solving the wear distribution functional using a variational optimization algorithm based on the spatiotemporal risk map and the set of topological invariants, and generating prevention and control parameters.

[0013] The mechanism of this invention is as follows: the memory effect of the wear process is quantified by fractional calculus, breaking through the limitations of traditional instantaneous models; cellular automata are used to simulate the spatiotemporal evolution of wear morphology under multi-physics coupling, realizing the emergence from microscopic rules to macroscopic patterns; nonlinear dynamics are combined to identify the critical state of the system, realizing forward-looking early warning; topological data analysis is used to extract wear pattern features that are insensitive to noise; finally, variational optimization is used to transform the predicted morphological features into optimal control parameters, forming a complete closed loop from "historical memory, morphological prediction, critical early warning, pattern recognition, and active guidance and control".

[0014] Beneficial Effects: This invention constructs a complete wear prevention and control system through the organic synergy of five modules: processing, prediction, evaluation, identification, and output. The time-varying memory kernel function generated by the processing module provides historical accumulated parameters for the prediction module. The spatiotemporal risk map output by the prediction module simultaneously supports the early warning analysis of the evaluation module and the pattern discrimination of the identification module. Finally, the output module generates precise prevention and control parameters through variational optimization. This systematic combination achieves complementary advantages between quantitative early warning and qualitative identification, solves the problem of spatiotemporal scale coupling, enhances adaptability under varying operating conditions, and reduces operation and maintenance costs through a closed-loop self-optimization mechanism. This invention establishes a principle-driven prediction and prevention paradigm that does not rely on big data, solving the inherent defects of machine learning methods in the application of key industrial equipment, such as the "black box" problem, strong data dependence, and poor extrapolation ability of operating conditions. It provides a new technical path for intelligent operation and maintenance of industrial equipment and has important theoretical significance and engineering application value. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of one embodiment of a method for preventing wear in a low-temperature economizer according to the present invention; Figure 2 This is a schematic diagram of one embodiment of a low-temperature economizer wear control system according to the present invention; Detailed Implementation

[0016] This invention provides a method, apparatus, device, and storage medium for preventing and controlling wear in low-temperature economizers, using an innovative multi-model fusion approach to achieve accurate prediction and proactive prevention of wear. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0017] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of a method for preventing wear in a low-temperature economizer according to the present invention includes: 101. Based on historical and real-time multiphysics data of the economizer, construct and solve fractional derivative equations, and establish a time-varying memory kernel function to quantify the influence of historical operating conditions on the current wear state.

[0018] It is understood that the executing entity of this invention can be a low-temperature economizer wear prevention and control device, or it can be a terminal or a server; the specific implementation is not limited here. This embodiment of the invention will be described using a server as an example.

[0019] It should be noted that real-time monitoring data of temperature, pressure and flow rate from the economizer field sensing system are collected, and particulate matter concentration distribution data obtained through computational fluid dynamics simulation are acquired. The data are then time-aligned and normalized to form a standardized multiphysics time series dataset. Based on fractional calculus theory, the historical cumulative effect in the material wear process is modeled as a dynamic system with memory characteristics, and a fractional differential equation framework describing the nonlinear relationship between the current wear state and historical working conditions is established. The fractional differential equation framework is numerically discretized and transformed into a convolution integral form containing time lag terms, thereby constructing a weighted kernel function mathematical expression that reflects the attenuation characteristics of historical influences. Based on the fatigue damage characteristics and experimental data of the economizer tube wall material, the key parameters in the weighted kernel function are calibrated, especially the fractional order that controls the memory decay rate, forming a specific form of the time-varying memory kernel function with clear physical meaning; The multiphysics time series data is input into the time-varying memory kernel function for convolution operation to calculate the quantitative contribution weight of the working condition at each historical moment to the current wear state, and generate a memory weight distribution matrix. This matrix is ​​used as a key parameter input into the subsequent cellular automata model evolution process.

[0020] 102. By combining the time-varying memory kernel function with the flow field distribution parameters, a spatiotemporal risk map predicting the future wear pattern evolution is generated through parallel evolution calculation of cellular automata.

[0021] It should be noted that a discretized digital grid model of the economizer tube wall is established, and the memory weight distribution matrix output by the time-varying memory kernel function is mapped onto each cell of the digital grid model to form an initial damage field that considers the historical cumulative effect. Obtain the current flow field distribution parameters, including local velocity vector, particle impact angle and kinetic energy distribution, and couple these parameters with the initial damage field to construct the local environmental state vector of each cell; Based on the material wear mechanism, a cell state transition rule is defined, and a multi-factor state evolution function is established, which includes the interaction of adjacent cell states, the influence of local flow field parameters, and the superposition of historical damage weights. Initiate parallel iterative computation of the cellular automaton, perform multi-step evolution within a preset future time period, and update the wear state of all cells synchronously based on the state evolution function within each time step; Through the parallel iterative calculation, a spatiotemporal risk map is generated that can display the wear depth distribution, evolution path, and high-risk areas. This map serves as a key input for subsequent nonlinear dynamic analysis and topological data analysis.

[0022] 103. Perform phase space reconstruction and nonlinear dynamic analysis on the evolution sequences of key regions in the spatiotemporal risk map, and establish a system instability early warning index.

[0023] It should be noted that the most active key regions for wear development are identified from the spatiotemporal risk map, and the state value changes of these regions at continuous time steps are extracted to form a time series of wear evolution in key regions. The wear evolution time series of the key area is reconstructed in phase space. By determining the appropriate embedding dimension and delay parameter, the one-dimensional time series is reconstructed into a trajectory in a high-dimensional phase space, and the reconstructed phase space of the system dynamic state is established. In the reconstructed phase space, the divergence rate of adjacent trajectory points is calculated, and the maximum Lyapunov exponent of the system is solved by numerical methods to obtain a quantitative index of chaotic characteristics that characterizes the sensitivity of the system to initial conditions. Based on the aforementioned chaotic characteristic quantification index, combined with the geometric feature analysis of the trajectory in the reconstructed phase space, a system instability early warning index is established that can provide early warning of the transition of a wear system from a stable state to an unstable state. The system instability early warning index is compared with a preset safety threshold to generate a system stability assessment result. This result is used as a key parameter input into the subsequent variational optimization algorithm to adjust the initiative and intervention intensity of the prevention and control strategy.

[0024] 104. Perform topological data analysis on the spatiotemporal risk map to extract a set of topological invariants that characterize the macroscopic wear pattern.

[0025] It should be noted that the wear depth distribution data in the spatiotemporal risk map is converted into three-dimensional spatial point cloud data, where two dimensions represent spatial coordinates and the third dimension represents the wear depth value; A simple complex sequence based on different scales is constructed from the three-dimensional spatial point cloud data, and the generation and disappearance of each topological feature during the scale change process are tracked by the continuous homology theory. The lifetime distribution of topological features in the simple complex sequence is recorded to generate a topological feature persistence map describing the geometry of the worn surface. This map quantifies the stability of the topological features at different scales. Extract topological invariants with significant lifetimes from the topological feature persistence graph, including the number of connected components, the number of holes, and high-dimensional topological features, to form a set of topological invariants; The set of topological invariants is matched and compared with the topological feature library of typical wear modes to establish a topological feature description of the current wear pattern, providing a qualitative classification basis for wear modes for subsequent variational optimization algorithms.

[0026] 105. Based on the spatiotemporal risk map and the set of topological invariants, the wear distribution uniformity functional is solved by variational optimization algorithm to generate precise control parameters to guide the action of the actuator.

[0027] It should be noted that, based on the wear depth spatial distribution data provided by the spatiotemporal risk map, an optimization functional is constructed with the wear distribution uniformity as the objective. Key features in the set of topological invariants are introduced as morphological constraints to establish a wear distribution optimization model that includes spatial morphological constraints. The adjustable control strategy parameter space is determined, including the guide vane angle adjustment range, local flow velocity control range and surface treatment scheme combination, and the control strategy parameters are introduced as optimization variables into the wear distribution optimization model; The wear distribution optimization model is solved using the variational method. The necessary conditions for the objective functional to reach the extreme value are derived through the Euler-Lagrange equation, and the mapping relationship between the control strategy parameters and the wear distribution response is established. An iterative optimization algorithm is constructed to solve the mapping relationship. Under the condition of satisfying engineering constraints, the optimal combination of prevention and control strategy parameters is searched to generate the optimal parameter solution set that makes the wear distribution most uniform. The optimized parameter set is transformed into a specific sequence of control commands for the actuator, including the guide vane adjustment angle value, the flow rate control set value, and the surface treatment implementation scheme, forming a set of precise control parameters that can directly drive the economizer control system.

[0028] 106. It also includes: automatically correcting model parameters based on measured data to achieve a self-optimization closed loop for the wear control system.

[0029] It should be noted that after the actuator is driven to move based on the set of precise prevention and control parameters, the multi-physics response data of the economizer system is collected in real time, including the flow velocity, pressure monitoring data and pipe wall temperature distribution change data after the flow field redistribution, and a dataset for verifying the prevention and control effect is generated. The control effect verification dataset is compared and analyzed with the prediction results of the spatiotemporal risk map. The consistency index between the actual wear distribution and the predicted wear distribution is calculated, and a control effect evaluation report is generated. Based on the consistency index in the prevention and control effect evaluation report, dynamically adjust the fractional order parameter in the time-varying memory kernel function and the state transition rule parameter in the cellular automaton model to generate model parameter optimization instructions. The model parameter optimization instructions are fed back to the fractional derivative model and the cellular automata model to achieve online self-calibration of the model parameters and establish a wear prediction and control closed-loop system with self-learning capabilities. Based on the corrected model parameters, each step is re-executed, and the precise control parameter set is iteratively optimized to achieve continuous performance improvement of the economizer wear control system.

[0030] In this embodiment of the invention, the beneficial effects are as follows: The theoretical models from five fields—fractional calculus, cellular automata, nonlinear dynamics, topological data analysis, and variational optimization—are organically integrated to construct a driving technology chain from "historical memory quantification" to "active morphological guidance and control." This achieves a shift from the traditional "monitoring-early warning" paradigm to a "pre-simulation-guidance and control" paradigm. By quantifying wear memory effects using fractional-order models, predicting spatiotemporal evolution patterns using cellular automata, achieving critical early warning using nonlinear dynamics, identifying wear patterns using topological analysis, and generating precise control parameters using variational optimization, a closed-loop system with self-correcting capabilities is ultimately formed. This effectively solves the inherent defects of traditional methods in terms of prediction accuracy, early warning foresight, and targeted control, providing a new technical path for wear control of industrial equipment.

[0031] Please see Figure 2 An embodiment of the chip detection system based on image analysis in this invention includes: 201. A processing module, which constructs and solves equations based on the history and data of the economizer to establish a time-varying memory kernel function; 202. A prediction module, which combines the time-varying memory kernel function with flow field distribution parameters and generates a spatiotemporal risk map through cellular automata calculations; 203. An evaluation module, which reconstructs and analyzes the regional evolution sequences in the spatiotemporal risk map to establish a system instability early warning index; 204. An identification module, which performs topological data analysis on the spatiotemporal risk map and extracts a set of topological invariants; 205. An output module, which solves the wear distribution functional using a variational optimization algorithm based on the spatiotemporal risk map and the set of topological invariants to generate control parameters.

[0032] In this embodiment of the invention, the beneficial effects are as follows: The processing module establishes a time-varying memory kernel function to quantify the cumulative impact of historical operating conditions; the prediction module combines flow field parameters to generate a spatiotemporal risk map, achieving accurate prediction of wear patterns; the evaluation module establishes instability early warning indicators through nonlinear dynamic analysis, providing a judgment of critical states; the identification module uses topological data analysis to extract wear pattern features, achieving qualitative identification of wear types; and the output module completes closed-loop control from prediction to decision-making. These five modules form a complete technology chain, improving the accuracy and effectiveness of wear prevention and control.

[0033] The present invention also provides a low-temperature economizer wear prevention and control device, which includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor performs the steps of the low-temperature economizer wear prevention and control method in the above embodiments.

[0034] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the method for preventing wear of a low-temperature economizer.

[0035] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0036] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0037] The above-described 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 the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for preventing wear in a low-temperature economizer, characterized in that, The method for preventing wear of a low-temperature economizer includes: Based on the history and data of the economizer, equations were constructed and solved, and a time-varying memory kernel function was established; By combining the time-varying memory kernel function with the flow field distribution parameters, a spatiotemporal risk map is generated through cellular automata calculations. The regional evolution sequences in the spatiotemporal risk map are reconstructed and analyzed to establish system instability early warning indicators; Topological data analysis was performed on the spatiotemporal risk map to extract the set of topological invariants; Based on the spatiotemporal risk map and the set of topological invariants, the wear distribution functional is solved by variational optimization algorithm to generate prevention and control parameters.

2. The method for preventing wear of a low-temperature economizer according to claim 1, characterized in that, include: Temperature, pressure, and flow rate data from the economizer sensing system are collected, and particulate matter concentration distribution data are acquired simultaneously. The data are then time-aligned and normalized to form multiphysics time series data. Based on fractional calculus theory, the historical cumulative effect is modeled as a dynamic system, and a framework of fractional differential equations is established. The fractional differential equation framework is processed and transformed into a convolution integral form to construct a weighted kernel function; Based on the characteristics and data of the economizer tube wall material, the key parameters in the weighted kernel function are calibrated to form a time-varying memory kernel function; The multiphysics time series data is input into the time-varying memory kernel function for calculation, and the quantized contribution weight is calculated to generate the memory weight distribution matrix.

3. The method for preventing wear of a low-temperature economizer according to claim 2, characterized in that, include: A mesh model of the economizer tube wall is established, and the time-varying memory kernel function is mapped onto the mesh model to form an initial damage field; Obtain the flow field distribution parameters, couple the parameters with the initial damage field, and construct a local environmental state vector; Transition rules are defined based on material wear mechanisms, and state evolution functions are established. Initiate parallel iterative computation of the cellular automaton to perform evolution, and update the wear state of all cells based on the state evolution function; The spatiotemporal risk map is generated through the parallel iterative calculation.

4. The method for preventing wear of a low-temperature economizer according to claim 1, characterized in that, include: Key regions are identified from the spatiotemporal risk map, and the changes in the state values ​​of these regions are extracted to form a wear evolution time series. The wear evolution time series is reconstructed in phase space. By determining the embedding dimension and parameters, the one-dimensional time series is reconstructed into a trajectory in a high-dimensional phase space, and the reconstructed phase space is established. In the reconstructed phase space, the divergence rate is calculated, and the quantitative index of chaotic characteristics is obtained through the maximum Lyapunov exponent. Based on the aforementioned chaotic characteristic quantification index, combined with the analysis in the reconstructed phase space, a system instability early warning index is established. The system instability early warning index is compared with a preset safety threshold to generate a system stability assessment result.

5. The method for preventing wear of a low-temperature economizer according to claim 4, characterized in that, include: The wear depth distribution data in the spatiotemporal risk map is converted into three-dimensional spatial point cloud data; A simple complex sequence is constructed from the three-dimensional spatial point cloud data to track the generation and disappearance during the scale change process; Record the lifetime distribution of topological features in the simple complex sequence to generate a topological feature persistence map; Topological invariants are extracted from the persistent graph of topological features to form a set of topological invariants; The set of topological invariants is compared with the topological feature library to establish a topological feature description.

6. The method for preventing wear of a low-temperature economizer according to claim 1, characterized in that, include: Based on the wear depth spatial distribution data provided by the spatiotemporal risk map, an optimized functional is constructed, and a wear distribution optimization model is established. Determine the parameters of the prevention and control strategy, and introduce the parameters of the prevention and control strategy into the wear distribution optimization model; The wear distribution optimization model is solved using the variational method to establish a mapping relationship; An iterative optimization algorithm is constructed to solve the mapping relationship and generate a set of optimization parameters. The optimized parameter set is transformed into a sequence of control instructions for the actuator, forming a precise control parameter set.

7. The method for preventing wear of a low-temperature economizer according to claim 6, characterized in that, Also includes: After driving the actuator to act based on the precise prevention and control parameter set, multi-physics response data is collected to generate a prevention and control effect verification dataset. The prevention and control effectiveness verification dataset is analyzed with the prediction results of the spatiotemporal risk map to generate a prevention and control effectiveness evaluation report; Based on the consistency index in the aforementioned prevention and control effect evaluation report, adjust the fractional order parameter and the state transition rule parameter to generate model parameter optimization instructions. The model parameter optimization instructions are fed back to the fractional derivative model and the cellular automaton model to achieve online self-correction and establish a wear prediction and prevention closed-loop system. Each step is re-executed based on the corrected model parameters to optimize the set of precise prevention and control parameters.

8. A low-temperature economizer wear control system, characterized in that, The low-temperature economizer wear prevention and control system includes: The processing module constructs and solves equations based on the history and data of the economizer, and establishes a time-varying memory kernel function; The prediction module is used to combine the time-varying memory kernel function with the flow field distribution parameters and generate a spatiotemporal risk map through cellular automata calculation; The evaluation module is used to reconstruct and analyze the regional evolution sequences in the spatiotemporal risk map and establish system instability early warning indicators. The identification module is used to perform topological data analysis on the spatiotemporal risk map and extract the set of topological invariants; The output module, based on the spatiotemporal risk map and the set of topological invariants, solves the wear distribution functional using a variational optimization algorithm to generate prevention and control parameters.

Images