A fault prediction method and system for a calcium formate production plant

Abstract

The application discloses a kind of calcium formate production equipment failure prediction method and system, specifically relates to failure prediction technical field, its through gathering calcium formate production equipment in the environmental data and structural stress response data in running process;Based on environmental data analysis metal surface corrosion rate variation trend, output corrosion evolution characteristic parameter;Through analysis structural stress response data, output stress evolution characteristic parameter;Based on corrosion evolution characteristic parameter, extract corrosion dominant type degradation index;Based on stress evolution characteristic parameter, extract stress dominant type degradation index;Combining corrosion dominant type degradation index and stress dominant type degradation index, construct corrosion-stress synergistic failure risk score model, output collaborative prediction result, and combine historical failure data, predict the remaining useful life of calcium formate production equipment, improve the operation reliability of calcium formate production equipment, while reducing maintenance cost.

Description

Technical Field

[0001] This invention relates to the field of fault prediction technology, and more specifically, to a fault prediction method and system for calcium formate production equipment. Background Technology

[0002] Existing calcium formate production equipment mostly operates in environments with high temperature, high humidity, and weak acidity. Key components of the equipment are exposed to complex loads and corrosive media for a long time, making them prone to corrosion-stress coupling failure.

[0003] Current conventional fault diagnosis methods are mostly based on single corrosion or single stress analysis, which makes it difficult to fully identify the non-uniform deterioration process caused by the combined effects of corrosion and stress, leading to sudden equipment failures and affecting the overall production process. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a method and system for predicting the failure of calcium formate production equipment to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for predicting faults in calcium formate production equipment includes the following steps:

[0007] S1: Collect environmental data and structural stress response data of the calcium formate production equipment during operation;

[0008] S2: Based on environmental data, analyze the changing trends of corrosion rate and corrosion distribution on metal surfaces under different operating environmental conditions, and output corrosion evolution characteristic parameters;

[0009] S3: Analyze structural stress response data, simulate the stress concentration area and evolution path of key components of equipment under long-term cyclic loading, and output stress evolution characteristic parameters;

[0010] S4: Based on the corrosion evolution characteristic parameters, and the coupling mapping relationship between local fatigue and corrosion acceleration of the material, extract the corrosion-dominant deterioration index;

[0011] S5: Based on the stress evolution characteristic parameters, extract stress-dominant degradation indices using the finite element dynamic response calculation method;

[0012] S6: Combining corrosion-dominant and stress-dominant degradation indices, a corrosion-stress synergistic failure risk scoring model is constructed, and synergistic prediction results are output;

[0013] S7: Based on the collaborative prediction results and combined with historical fault data, predict the remaining service life of calcium formate production equipment.

[0014] In a preferred embodiment, S1 specifically refers to:

[0015] Environmental parameters include temperature, humidity, and acid / alkali concentration data during the operation of the calcium formate production equipment;

[0016] Structural stress response data includes mechanical load data and thermal stress data borne by key components.

[0017] In a preferred embodiment, S2 specifically refers to:

[0018] Based on temperature, humidity, and acid / alkali concentration data, a statistical analysis was performed on the corrosion rate of metal surfaces under different operating conditions, generating corrosion rate variation curves.

[0019] Images of corrosion zones on metal surfaces under different operating conditions are acquired, and corrosion depth and distribution data are generated through image processing.

[0020] Based on corrosion rate variation curves, corrosion depth and corrosion depth distribution data, statistical analysis methods are applied to calculate corrosion depth growth rate, corrosion area ratio and corrosion region expansion path, and output corrosion evolution characteristic parameters.

[0021] In a preferred embodiment, S3 specifically refers to:

[0022] Based on mechanical load data and thermal stress data, a finite element numerical model of key components of calcium formate production equipment was established.

[0023] The stress distribution of key components under long-term cyclic loading is simulated and calculated based on the finite element numerical model to determine the location and range of stress concentration areas in key components.

[0024] Based on the changes in the location and extent of stress concentration areas in key components, the evolution path of stress concentration areas is generated;

[0025] Stress evolution characteristic parameters are generated based on the location, extent, and evolution path of the stress concentration region.

[0026] In a preferred embodiment, S4 specifically refers to:

[0027] Based on corrosion evolution characteristic parameters, a coupling relationship between local fatigue and corrosion acceleration in metallic materials is established.

[0028] Based on the coupling relationship between local fatigue and accelerated corrosion of metallic materials, the degree of influence of corrosion on the fatigue deterioration of metallic materials is determined.

[0029] Based on the degree of influence of corrosion on the fatigue deterioration of metallic materials, corrosion-dominant deterioration indicators are extracted.

[0030] In a preferred embodiment, S5 specifically refers to:

[0031] The stress evolution characteristic parameters are input into the finite element dynamic response calculation model to calculate the dynamic stress distribution of key components under the combined action of cyclic mechanical load and thermal stress.

[0032] Extract the equivalent stress peak value, stress cycle amplitude, and stress change rate from the dynamic stress distribution results;

[0033] The cumulative fatigue damage of key components is estimated based on the equivalent stress peak value and stress cycle amplitude.

[0034] The cumulative fatigue damage, peak equivalent stress, and stress cycle amplitude were extracted as stress-dominant degradation indicators.

[0035] In a preferred embodiment, S6 specifically refers to:

[0036] A corrosion-stress synergistic failure risk scoring model is established based on corrosion-dominant and stress-dominant degradation indices.

[0037] The corrosion-stress synergistic failure risk scoring model is used to calculate the corrosion-dominant and stress-dominant degradation indices by weighting them to obtain a corrosion-stress synergistic failure risk score, which serves as the synergistic prediction result.

[0038] In a preferred embodiment, S7 specifically refers to:

[0039] Obtain historical fault data of the calcium formate production equipment, including the time and type of historical faults;

[0040] Statistical analysis of historical failure data was conducted to obtain the lifespan distribution parameters of calcium formate production equipment;

[0041] A relationship for estimating remaining service life was constructed based on the life distribution parameters of calcium formate production equipment.

[0042] Substitute the collaborative prediction results into the remaining service life estimation relationship to calculate and output the remaining service life of the calcium formate production equipment.

[0043] On the other hand, the present invention provides a fault prediction system for calcium formate production equipment, comprising:

[0044] Data acquisition module: Collects environmental data and structural stress response data of the calcium formate production equipment during operation;

[0045] Trend Analysis Module: Based on environmental data, this module analyzes the trends in corrosion rate and corrosion distribution on metal surfaces under different operating conditions and outputs corrosion evolution characteristic parameters.

[0046] Coupled mapping analysis module: Analyzes structural stress response data, simulates the stress concentration area and evolution path of key components of equipment under long-term cyclic loading, and outputs stress evolution characteristic parameters;

[0047] Corrosion index extraction module: Based on corrosion evolution characteristic parameters and the coupling mapping relationship between local fatigue and corrosion acceleration of materials, corrosion-dominant deterioration indexes are extracted.

[0048] Stress index extraction module: Based on stress evolution characteristic parameters and the finite element dynamic response calculation method, stress-dominant degradation indices are extracted.

[0049] Collaborative Analysis Module: Combining corrosion-dominant and stress-dominant degradation indices, a corrosion-stress collaborative failure risk scoring model is constructed, and collaborative prediction results are output.

[0050] Lifespan prediction module: Based on collaborative prediction results and combined with historical failure data, predict the remaining lifespan of calcium formate production equipment.

[0051] The technical effects and advantages of the fault prediction method and system for calcium formate production equipment of the present invention are as follows:

[0052] By collecting environmental and structural stress response data in real time during the operation of calcium formate production equipment, key influencing factors are comprehensively identified. Analysis of corrosion rate and distribution trends extracts corrosion evolution characteristic parameters, helping to accurately characterize the corrosion trends of materials in high-temperature, high-humidity, and weakly acidic environments. The stress concentration behavior under long-term cyclic loading is simulated using the finite element method to obtain stress evolution characteristic parameters, enabling dynamic monitoring of potential fatigue failure paths. A synergistic failure risk scoring model is established by integrating corrosion-dominant and stress-dominant degradation indices, enabling scientific prediction of complex coupled failures. Combined with historical fault data, the remaining life of the calcium formate production equipment is assessed, significantly improving its operational reliability and management level under extreme conditions. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of a method for predicting faults in calcium formate production equipment according to the present invention;

[0054] Figure 2 This is a schematic diagram of the structure of a fault prediction system for calcium formate production equipment according to the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. Example 1

[0056] Figure 1 This invention provides a method for predicting faults in calcium formate production equipment, which includes the following steps:

[0057] S1: Collect environmental data and structural stress response data of the calcium formate production equipment during operation;

[0058] S2: Based on environmental data, analyze the changing trends of corrosion rate and corrosion distribution on metal surfaces under different operating environmental conditions, and output corrosion evolution characteristic parameters;

[0059] S3: Analyze structural stress response data, simulate the stress concentration area and evolution path of key components of equipment under long-term cyclic loading, and output stress evolution characteristic parameters;

[0060] S4: Based on the corrosion evolution characteristic parameters, and the coupling mapping relationship between local fatigue and corrosion acceleration of the material, extract the corrosion-dominant deterioration index;

[0061] S5: Based on the stress evolution characteristic parameters, extract stress-dominant degradation indices using the finite element dynamic response calculation method;

[0062] S6: Combining corrosion-dominant and stress-dominant degradation indices, a corrosion-stress synergistic failure risk scoring model is constructed, and synergistic prediction results are output;

[0063] S7: Based on the collaborative prediction results and combined with historical fault data, predict the remaining service life of calcium formate production equipment.

[0064] S1: Collect environmental data and structural stress response data of the calcium formate production equipment during operation, including:

[0065] Environmental parameters include temperature, humidity, and acid / alkali concentration data during the operation of the calcium formate production equipment;

[0066] Structural stress response data includes mechanical load data and thermal stress data borne by key components.

[0067] Calcium formate production equipment is a complete set of specialized equipment used to produce calcium formate, which typically includes reaction vessels, pipelines, valves, pumps, cooling or heating devices, stirring devices, and other equipment components.

[0068] Environmental parameters and structural stress response data fall under the category of operational status monitoring, used to characterize the external environmental influences and internal mechanical responses of production equipment under actual operating conditions. A data acquisition unit can be installed in the production workshop. This unit includes a data acquisition module, a data transmission module, and a data storage module connected to the calcium formate production equipment. The data acquisition module connects to multiple sensors, such as temperature sensors, humidity sensors, acid / alkali concentration sensors, mechanical load sensors, and thermal stress sensors, via a data bus, converting the analog or digital signals detected by the sensors into digital data suitable for processing. The data transmission module uses industrial Ethernet or a wireless network to send the digital data to the data storage module in real time. The data storage module can use a relational database or a time-series database to uniformly manage and archive the received environmental parameter and structural stress response data.

[0069] For example, in a calcium formate production workshop, a temperature sensor is installed on the outer wall of the reactor to record the temperature changes outside the reactor in real time; a humidity sensor is installed in the air near the reactor to record the air humidity; an acid-base concentration sensor is installed in the reaction liquid circulation pipeline to record the pH value of the reaction liquid; a mechanical load sensor is installed on the agitator bearing to measure the radial load borne by the bearing; and a thermal stress sensor is arranged on the heat exchanger tube sheet to measure the thermal stress generated by the thermal expansion of the tube sheet.

[0070] Environmental parameters refer to the external or media conditions at the operating site of the calcium methyl methacrylate (CMMA) production equipment. Key elements include temperature, humidity, and acid / base concentration. Temperature data represents the temperature value of the surface of the reactor or equipment components, or the ambient air, and can be measured using a PT100 platinum resistance thermometer, thermocouple, or infrared thermometer. Humidity data reflects the relative humidity of the workshop air or a localized enclosed space, and can be measured using a capacitive or resistive humidity sensor. Acid / base concentration data represents the pH value of the reaction solution inside the pipeline or reactor, and can be measured using a glass electrode pH sensor or an optical pH sensor.

[0071] For example, when the internal temperature of the reactor is about 80 degrees Celsius, the PT100 probe outputs a corresponding resistance value, which is converted into digital temperature data by the data acquisition module; when the relative humidity of the workshop is 45%, the capacitive humidity sensor outputs a corresponding voltage, which is converted into digital humidity data by analog-to-digital conversion; when the pH of the reaction liquid is about 3.2, the glass electrode sensor outputs a voltage signal, which is converted into a digital pH value by pH transmitter calibration.

[0072] Structural stress response data refers to the stress response of key components inside production equipment under external loads and thermal expansion during operation. Mechanical load data represents the stress state of components under mechanical action, including bearing loads, connecting rod bending moments, gear contact stresses, etc., and can be measured using resistance strain gauge load sensors, piezoelectric load sensors, or torque sensors. Thermal stress data represents the internal stress of structural components caused by temperature differences, and can be obtained using fiber Bragg grating thermo-stress sensors, thermocouple arrays, or strain gauge arrays combined with finite element pre-set calculations.

[0073] For example, at the beginning of the stirring process, the stirrer bearing bears a radial load of 12 kN, which is output as force value after being detected by the load sensor; as the temperature of the heat exchange medium changes, the temperature difference of the heat exchanger tube sheet reaches 60 degrees Celsius, and the thermal expansion of the tube sheet generates a thermal stress of 80 MPa, which is measured by the thermal stress sensor and converted into digital stress data.

[0074] S2: Based on environmental data, analyze the changing trends of corrosion rate and corrosion distribution on metal surfaces under different operating environmental conditions, and output corrosion evolution characteristic parameters, including:

[0075] Based on temperature, humidity, and acid / alkali concentration data, a statistical analysis was performed on the corrosion rate of metal surfaces under different operating conditions, generating corrosion rate variation curves.

[0076] Temperature data, humidity data, and acid / alkali concentration data constitute a set of environmental parameters used to reflect the differences in corrosion behavior of metal surfaces under different temperatures, relative humidity levels, and pH values. In this embodiment, multiple metal samples (such as 304 stainless steel sheets from the same batch) can be fixed to the outer side of the inner wall of the reaction vessel. The samples are grouped according to temperature sensing area, humidity sensing area, and acid / alkali sensing area, and numbered as samples A1 to An. Samples A1 to An are installed in the environmental chamber, with a set temperature range of 20–100 degrees Celsius, relative humidity range of 20%–80%, and pH value range of 1–7.

[0077] Under each operating condition, a sample A was disassembled every 24 hours for mass difference measurement. The mass loss value Δm (accuracy 0.1 mg) was measured using a high-precision electronic balance. The corrosion rate was then calculated based on the metal sheet area: Corrosion rate (µm / year) = (Δm / (ρ×S×T))×10^6, where ρ is the density of the metal material (g / cm³), S is the surface area of ​​the metal sheet (cm²), and T is the exposure time (hours). A series of corrosion rate numerical sequences were obtained. The corrosion rate was fitted with a least-squares linear regression or nonlinear regression model as a function of time or environmental parameters to obtain the corrosion rate variation curve.

[0078] For example, under conditions of 80 degrees Celsius, 60% humidity, and pH 3, the mass change of sample A3 was measured every 24 hours, and the mass loss data were continuously measured as 0.12 mg, 0.15 mg, and 0.18 mg. The corresponding corrosion rates were calculated to be 0.05 μm / year, 0.06 μm / year, and 0.07 μm / year, respectively. The data were plotted on a two-dimensional coordinate system, with the horizontal axis representing exposure time and the vertical axis representing corrosion rate. Regression analysis was used to fit the curve, generating a corrosion rate change curve that reflects the change of corrosion rate over time.

[0079] Corrosion rate variation curves can be used to visually demonstrate the increasing or decreasing trend of corrosion rate on metal surfaces under different combinations of environmental parameters.

[0080] Images of corrosion zones on metal surfaces under different operating conditions are acquired, and corrosion depth and distribution data are generated through image processing.

[0081] The image acquisition device includes a high-resolution industrial camera, a 3D scanner for metal surfaces, or an optical profilometer, installed directly above or to the side of the corrosion test area to acquire images of the corrosion sample surface. A CCD camera with a resolution of 2592×1944 pixels is used to take timed photos of the sample surfaces numbered A1 to An, with images in TIFF or RAW format to ensure that no image details are lost.

[0082] Image processing includes:

[0083] Image preprocessing: The acquired raw image is subjected to grayscale conversion, noise filtering, and edge enhancement.

[0084] Corrosion region segmentation: Using threshold segmentation, Otsu's method, or deep learning-based semantic segmentation algorithms, the corroded region is distinguished from the uncorroded region, and a binary mask image is generated.

[0085] Corrosion depth measurement: The surface normal height information is obtained by using a structured light 3D scanner, the height difference is converted into corrosion depth value, and the depth matrix of each pixel is output.

[0086] Corrosion distribution calculation: Based on the binary mask image and the depth matrix, the number of pixels corresponding to each depth interval within the corrosion area is counted to generate a corrosion distribution data table, including depth interval, number of pixels, and corrosion area.

[0087] For example, after capturing a high-resolution image of sample A3, the Canny edge detection function using the OpenCV library was used to identify the corrosion boundaries. A height map was obtained using a scanner, with the pixel location height value being 0 micrometers (reference plane) and the bottom height of the corrosion pit being -200 micrometers, resulting in a calculated corrosion depth of 200 micrometers. For the 521×421 pixel corrosion region within the image, 100,000 pixels were identified for depths of 0–50 micrometers, and 80,000 pixels were identified for depths of 50–100 micrometers, generating depth distribution data.

[0088] Corrosion depth and corrosion distribution data reflect the spatial and depth distribution characteristics of the corroded area.

[0089] Based on corrosion rate change curves, corrosion depth and corrosion depth distribution data, statistical analysis methods are applied to calculate corrosion depth growth rate, corrosion area ratio and corrosion region expansion path, and output corrosion evolution characteristic parameters.

[0090] Corrosion evolution characteristic parameters include: corrosion depth growth rate, corrosion area ratio, and corrosion zone expansion path.

[0091] Based on the values ​​corresponding to multiple time points on the corrosion rate change curve, linear or polynomial fitting methods are used to calculate the slope of the corrosion depth as a function of time. The slope is expressed in micrometers per day or micrometers per hour, representing the average rate of increase in corrosion depth on the metal surface over time. The fitting method can use the least squares method, solving for the slope k to minimize ∑(yi–(k·xi + b))^2, where xi is time and yi is the corresponding corrosion rate.

[0092] For example, under conditions of 80 degrees Celsius, 60% humidity, and pH 3, the set of points of the corrosion rate change curve (1 day, 0.05), (2 days, 0.06), and (3 days, 0.07) are fitted to obtain a straight line with a slope of 0.01 micrometers / day, so the corrosion depth growth rate is 0.01 micrometers / day.

[0093] Using a binarized mask image, the total number of pixels in the eroded area (N_cor) and the total number of pixels in the image (N_tot) are counted. The erosion area ratio is then calculated as N_cor ÷ N_tot. The erosion area ratio reflects the proportion of the corroded area to the total surface area of ​​the metal surface and can be expressed as a percentage or a decimal.

[0094] For example, the resolution of sample A3 image is 2592×1944, totaling 5,038,848 pixels. The total number of pixels in the eroded area is 150,000, and the erosion area ratio is approximately 150,000 ÷ 5038,848 ≈ 2.98%.

[0095] Based on multi-time-step corrosion masking, the corrosion boundary contour at each time point is extracted. The boundary movement vector field is obtained by calculating the Hausdorff distance or contour point matching displacement between boundary contours at adjacent time points. Cluster analysis is performed on all vectors to obtain the main propagation directions and paths. The path can be represented as a set of ordered geographic coordinates or boundary point index sequences, used to describe the diffusion trajectory of corrosion from the starting point to each direction on the metal surface.

[0096] For example, in the images of day 1 and day 3, the erosion boundary contours are denoted as C1 and C3, respectively. The contour point movement vector {(x1→x1′),(x2→x2′)} is calculated by least squares matching. The main expansion vector is obtained by fitting and the direction is horizontal, with an expansion path length of 10 mm.

[0097] The corrosion depth growth rate, corrosion area ratio, and corrosion region expansion path are summarized as corrosion evolution characteristic parameters.

[0098] S3: Analyze structural stress response data, simulate the stress concentration areas and evolution paths of key components under long-term cyclic loading, and output stress evolution characteristic parameters, including:

[0099] Based on mechanical load data and thermal stress data, a finite element numerical model of key components of calcium formate production equipment was established.

[0100] Mechanical load data and thermal stress data together reflect the internal stress state of key components of equipment during production, caused by external mechanical forces and temperature changes. The finite element numerical model provides a numerical means to analyze the stress and deformation fields of key components under complex load conditions.

[0101] Acquire 3D geometric information of key components. For key components such as the reaction vessel support frame, heat exchanger tube bundle, and stirrer blades, acquire point cloud data of their shapes using a 3D laser scanner or industrial CT scanning equipment. The point cloud data is then processed by post-processing software for denoising, registration, and surface fitting to generate a closed triangular mesh geometric model, with the model accuracy controlled within 0.1 mm.

[0102] The constitutive model parameters are assigned based on the metal material grade and heat treatment state of the key components. For stainless steel components, tensile tests are used to obtain the elastic modulus, Poisson's ratio, and yield strength; for carbon steel components, high-temperature tensile tests are used to obtain stress-strain curves at different temperatures; and a laboratory thermal expansion tester is used to measure the coefficient of linear expansion. The elastic modulus (e.g., 2.0 × 10^11 Pa), Poisson's ratio (e.g., 0.3), and coefficient of linear expansion are input into the material database.

[0103] The geometric model is meshed. Tetrahedral or hexahedral elements are used, and the mesh size is refined according to the local curvature and stress gradient distribution characteristics of the component. For blade roots and weld areas subject to high stress concentration, the mesh size can be controlled at 0.5 mm; for areas with relatively low overall load, the mesh size can be relaxed to 2.0 mm. Mesh quality inspection ensures that the element distortion rate is less than 0.2 and the element size change rate is smooth.

[0104] Mechanical load data is mapped onto model nodes or element surfaces. This data includes bearing radial load, agitator torque, and pipe pressure. For example, a radial load of 12 kN at the agitator bearing is applied to the bearing support surface through node concentration or surface distribution. Thermal stress data is converted into a temperature field or thermal expansion constraints. A 60°C temperature difference between the two sides of the tube sheet is applied to the tube sheet surface, and corresponding thermal displacement boundary conditions are set in the thermal expansion degree of freedom direction.

[0105] Static or thermo-mechanical coupled analysis of the model is performed using an open-source solver. A nonlinear solver is selected to consider large deformation effects and nonlinear constitutive relations of materials, ensuring iterative convergence accuracy. The stress, strain, and displacement values ​​of key components under given loads are obtained through the solution process.

[0106] Through the above steps, a finite element numerical model was established to reflect the dual effects of mechanical load and thermal stress caused by temperature difference on key components.

[0107] The stress distribution of key components under long-term cyclic loading is simulated and calculated based on the finite element numerical model to determine the location and range of stress concentration areas in key components.

[0108] The stress distribution simulation under long-term cyclic loading takes into account the combined effects of torque, bending moment, axial load and temperature cycling.

[0109] A time-history load condition was established based on the mechanical load data sequence. The load curves collected from the start-up, steady-state, and shutdown of the agitator were segmented, with each segment representing a cycle. Taking 10 starts per day, each lasting 2 hours, followed by a 1-hour rest, as an example, the cycle was repeated 1000 times to simulate a long-term loop.

[0110] Within each cycle, the temperature field and stress field are coupled and iteratively calculated. First, a mechanical load is applied to solve the static stress field, then a temperature field is applied to solve the thermal stress distribution. The resulting thermal stress is superimposed on the mechanical stress field to form a comprehensive stress field, completing one coupled iteration.

[0111] In the comprehensive stress field, high-stress areas are screened using a stress threshold. The threshold can be set to 80% of the maximum stress value as the critical value. The set of mesh cells with stress values ​​higher than the threshold is denoted as the stress concentration region. A connected component analysis algorithm is used to identify the spatial connected components of each high-stress region. The location and extent of each connected component are described by an envelope surface or a minimum bounding sphere.

[0112] For each high-stress connected domain, the coordinates of the center point, the maximum envelope radius, and shape features in the spatial coordinate system are extracted. The center point coordinates can be calculated by averaging the coordinates of all element nodes in the connected domain, and the envelope radius is the distance from the maximum node to the center point. Shape features include aspect ratio, flatness, etc., to describe the geometry of the region.

[0113] Through the above steps, the spatial location and three-dimensional range of the stress concentration zone inside each key component under long-term cyclic loading were determined.

[0114] Based on the changes in the location and extent of stress concentration areas in key components, the evolution path of stress concentration areas is generated;

[0115] The evolution path describes the spatial movement and shape change of high-stress regions with the number of cycles or running time.

[0116] At the end of each cycle, the stress concentration region identification step is repeated to obtain the center point and range matrix of the high stress connected domain after the cycle; the center point sequences of different cycles are matched to construct the center point trajectory sequence.

[0117] The extent of each connected component is recorded, generating a sequence of extent changes over a period. The change in aspect ratio over the period is extracted to assess the distortion direction and velocity of the region shape.

[0118] Spline interpolation or Kalman filtering algorithms are applied to the center point trajectory sequence for smoothing, removing analog noise and generating a continuous curved trajectory. The interpolated trajectory can be refined to a node every hour or every ten minutes for high-precision path representation.

[0119] The smoothed 3D trajectory is imported into visualization software, and different colors or line thicknesses are used to represent the path time process and range changes, forming an intuitive stress evolution path diagram.

[0120] Based on the above process, the spatial movement curve and shape change record of the high-stress region are obtained, forming a complete evolution path of the stress concentration region, which can reflect the trend of fatigue crack initiation and propagation.

[0121] Based on the location, extent, and evolution path of the stress concentration region, stress evolution characteristic parameters are generated;

[0122] Stress evolution characteristic parameters are numerical indicators that quantitatively describe the evolution behavior of high-stress zones, including maximum displacement rate, range growth rate, and migration direction angle.

[0123] Based on the trajectory sequence of the center point, calculate the Euclidean distance between two adjacent trajectory nodes, divide it by the corresponding time interval to obtain the displacement rate, and take the maximum value as the maximum displacement rate.

[0124] Based on the range sequence, the difference between adjacent ranges is calculated and divided by the time interval to obtain the range growth rate. The range growth rate sequence is used to describe the speed of regional expansion, and the average range growth rate is selected as one of the feature parameters.

[0125] Based on the angle between the trajectory node vector and the reference coordinate system axis, the average value of the angle is used to describe the main migration direction by calculating the angle using the inverse cosine function.

[0126] Output the maximum displacement rate, the average range growth rate, and the average angle in vector form or as a scalar list.

[0127] Through the above calculations, stress evolution characteristic parameters that characterize the spatial movement speed, expansion speed, and main migration direction of the stress concentration region are obtained.

[0128] S4: Based on corrosion evolution characteristic parameters and the coupling mapping relationship between local fatigue and corrosion acceleration, corrosion-dominant degradation indices are extracted, including:

[0129] Based on corrosion evolution characteristic parameters, a coupling relationship between local fatigue and corrosion acceleration in metallic materials is established.

[0130] Corrosion evolution characteristic parameters include corrosion depth growth rate, corrosion area ratio, and corrosion zone propagation path, reflecting the corrosion development law of metal surfaces under the influence of various environmental factors. When metallic materials are subjected to cyclic loading, the local fatigue process often interacts with corrosion. Corrosion weakens the deformation tolerance of the material surface and accelerates the initiation and propagation of fatigue cracks, while fatigue can induce higher corrosion rates at corrosion pits or surface defects. Therefore, it is necessary to establish the coupling relationship between local fatigue and corrosion acceleration in order to accurately describe the degradation mechanism under the combined effect.

[0131] The coupling relationship employs a hybrid constitutive model containing corrosion-promoting factors and fatigue damage factors. First, standard fatigue specimens of the metallic material are prepared, with geometric dimensions conforming to GB / T 3075-2015 standards. Corrosion pits of different depths (e.g., 100 μm, 200 μm, and 300 μm) are artificially prepared on the surface according to certain rules to simulate different stages of corrosion evolution. The standard fatigue specimens are then subjected to constant-amplitude cyclic fatigue tests at room temperature and under simulated salt spray conditions (35℃, 95% relative humidity, 5% NaCl concentration). The test frequency is 10 Hz, the stress ratio is 0.1, and the maximum stresses are 200 MPa, 300 MPa, and 400 MPa, respectively.

[0132] During fatigue testing, the relationship between the number of fatigue cycles and crack length was recorded in real time, and crack propagation was periodically measured using a digital microscope or optical microscope. Changes in corrosion evolution characteristic parameters, such as corrosion depth growth rate (v_corr), corrosion area ratio (A_corr), and corrosion zone propagation path length (L_corr), were recorded in parallel. Based on the experimental data, the functional relationship between the fatigue crack propagation rate and the cyclic load stress amplitude and corrosion evolution characteristic parameters was fitted using multiple regression analysis or neural network regression methods to establish a coupled constitutive equation. The best fit value was determined using the least squares method or a genetic algorithm. The coupled model considers both the pure fatigue crack propagation mechanism and the corrosion acceleration factor in the crack propagation rate calculation, forming a coupled relationship between local fatigue and corrosion acceleration.

[0133] Taking a standard fatigue specimen of 304 stainless steel as an example, the experimental results of the artificial corrosion pit depth group of 100 micrometers show that, under the condition of a maximum stress of 300 MPa, when the corrosion evolution characteristic parameters v_corr = 0.02 micrometers / day, A_corr = 0.015, and L_corr = 5 millimeters, the fitting regression coefficients C1 = 1.2 × 10⁻¹² and C2 = 3.5 × 10⁻¹². 6 C3 = 2.1 × 10⁻³, C4 = 1.8 × 10⁻², exponents m = 2.5, p = 1.2, q = 0.8, r = 0.5. The crack propagation rate curve predicted by the model agrees with the experimentally measured value by more than 95%, verifying the ability of the coupled relationship model to describe the combined effect of fatigue and corrosion.

[0134] Based on the coupling relationship between local fatigue and accelerated corrosion of metallic materials, the degree of influence of corrosion on the fatigue deterioration of metallic materials is determined.

[0135] The degree of corrosion's influence on the fatigue degradation of metallic materials is represented by an acceleration factor, defined as the reciprocal of the ratio of fatigue life under corroded conditions to fatigue life under non-corroded conditions at the same load level. When the acceleration factor is greater than 1, it indicates that corrosion accelerates fatigue degradation; the larger the value, the higher the degree of acceleration. An acceleration factor of 1 indicates no acceleration effect.

[0136] The method for calculating the degree of impact is as follows:

[0137] Non-corrosion fatigue life determination: Stainless steel samples of the same batch were subjected to conventional fatigue tests, and the average fatigue life was measured under the same maximum stress and stress ratio conditions.

[0138] Corrosion fatigue life determination: Using standard samples with corrosion pits, cyclic fatigue tests are conducted in a salt spray environment to measure the number of fatigue cycles at which fatigue failure occurs.

[0139] Acceleration factor calculation: Based on multiple sets of experimental data, the acceleration factor for each set is calculated. Then, the acceleration factors under different corrosion pit depths and different stress levels are statistically analyzed to obtain the functional relationship between the acceleration factor and the corrosion evolution characteristic parameters, and to quantitatively describe the influence of solidification corrosion on fatigue deterioration.

[0140] Based on the degree of influence of corrosion on the fatigue deterioration of metallic materials, corrosion-dominant deterioration indicators are extracted.

[0141] Corrosion-dominant degradation indicators are used to quantify the dominant weight of corrosion in the overall failure risk, and comprehensively reflect the severity of corrosion-accelerated fatigue degradation and the criticality of failure.

[0142] The method for extracting degradation indicators is as follows:

[0143] Map the acceleration factor to the range [0,1] using a normalization function:

[0144] The corrosion weight is calculated based on the ratio of the fitted regression coefficients C2, C3, C4 to the fitted regression coefficients C1, C2, C3, C4 in the fatigue-corrosion coupling model.

[0145] Multiplying the normalized acceleration factor by the corrosion weight yields the corrosion-dominant degradation index.

[0146] The closer the corrosion-dominant degradation index is to 1, the more dominant the corrosion effect. The corrosion-dominant degradation index reflects the degree to which corrosion dominates the fatigue failure process under given operating conditions.

[0147] S5: Based on stress evolution characteristic parameters and the finite element dynamic response calculation method, stress-dominant degradation indices are extracted, including:

[0148] The stress evolution characteristic parameters are input into the finite element dynamic response calculation model to calculate the dynamic stress distribution of key components under the combined action of cyclic mechanical load and thermal stress.

[0149] Stress evolution characteristic parameters include the evolution path of stress concentration zones in key components, maximum displacement rate, range growth rate, and migration direction angle, which are used to characterize the evolution law of stress concentration zones during operation. The finite element dynamic response calculation model is a numerical simulation platform that introduces time history analysis capabilities based on static or thermo-mechanical coupled finite element models. The model can simulate the instantaneous stress distribution and stress fluctuations generated in key components under the combined action of periodic mechanical loads and thermal stress caused by temperature cycles in the time domain.

[0150] For key components at the blade root of the agitator, the initial position and evolution path data of the stress concentration zone are loaded based on a three-dimensional geometric mesh model of the blade root. Cyclic mechanical loads are applied to the bearing support nodes at the blade root via time history curves. The cyclic mechanical load curves can be obtained from actual operating condition monitoring of torque-speed-time relationships, using a time history of three daily start-steady-state-stop cycles, each lasting two hours followed by a half-hour stop. Thermal stress is applied to the blade root surface via temperature history curves, based on the temperature difference changes in the heat exchanger tube sheet, with temperature cycles of ±30 degrees Celsius applied. The finite element dynamic response calculation model simultaneously solves for the instantaneous stress field generated by mechanical loads and thermal expansion at each time step (e.g., every 0.1 seconds), and outputs the nodal stress values ​​and element stress values ​​within the time history. The simulation results generate a dynamic stress distribution cloud map of the key components throughout the entire cycle.

[0151] Extract the equivalent stress peak value, stress cycle amplitude, and stress change rate from the dynamic stress distribution results;

[0152] The dynamic stress distribution result is the stress value of each node or element changing with time under the time history. The equivalent stress peak value refers to the maximum stress value among all nodes or elements within one cycle. The stress cycle amplitude refers to the difference between the highest and lowest stress values ​​of a certain node or element within one cycle. The stress change rate refers to the maximum slope of stress change with time in the time domain; the peak value of the slope of the tangent line on the stress-time curve is taken as the stress change rate.

[0153] From the dynamic response analysis results, for the high-stress element set at the blade root, the stress maxima and minima of each element within one cycle (e.g., 7200 seconds) are calculated. The equivalent stress peak value is directly determined from the stress maxima; the stress cycle amplitude is obtained by calculating the difference between the stress maxima and the stress minima; the stress-time curve is differentiated during its rising and falling phases, and the local maximum slope is extracted as the stress change rate. Taking a key element at the blade root as an example, after the cycle starts, the stress rises from 20 MPa to 120 MPa, then drops to 25 MPa, and rises again to 115 MPa; the equivalent stress peak value is taken as 120 MPa; the stress cycle amplitude is taken as 120 – 25 = 95 MPa; the stress change rate is taken as the peak slope of approximately 0.0138 MPa / second (100 MPa / 7200 seconds) during the stress increase from 20 MPa to 120 MPa. The equivalent stress peak value, stress cycle amplitude, and stress change rate constitute a dynamic stress feature set, providing basic data for fatigue damage assessment.

[0154] The cumulative fatigue damage of key components is estimated based on the equivalent stress peak value and stress cycle amplitude.

[0155] The cumulative fatigue damage value is estimated using linear damage accumulation theory or a modified nonlinear fatigue accumulation model. Linear accumulation theory defines the damage value for each stress cycle as the ratio of the number of cycles to the material's durability at that stress amplitude; the cumulative fatigue damage value is obtained by summing the damage values ​​from each cycle. The material durability-stress amplitude relationship can be obtained through the SN curve. The cumulative fatigue damage value reflects the degree of damage to critical components under actual cyclic operating conditions.

[0156] The cumulative fatigue damage value, peak equivalent stress, and stress cycle amplitude were extracted as stress-dominant degradation indicators.

[0157] The stress-dominated degradation index is used to describe the dominant role of internal stress evolution in the overall equipment degradation. The cumulative fatigue damage value characterizes the degree of cumulative damage under cyclic loading; the peak equivalent stress reflects the level of extreme load stimulation; and the stress cycle amplitude reflects the intensity of fatigue cycles. The stress-dominated degradation index is obtained by normalizing and weighting these three characteristic indices.

[0158] S6: Combining corrosion-dominant and stress-dominant degradation indices, a corrosion-stress synergistic failure risk scoring model is constructed, and synergistic prediction results are output, including:

[0159] A corrosion-stress synergistic failure risk scoring model is established based on corrosion-dominant and stress-dominant degradation indices.

[0160] Corrosion-dominated degradation indices and stress-dominated degradation indices reflect the degree of degradation of production equipment under corrosion and cyclic loading, respectively. The corrosion-stress synergistic failure risk scoring model uses both corrosion-dominated and stress-dominated degradation indices as input variables. Through multiple regression analysis, machine learning algorithms, or rule-based expert systems, a numerical model is constructed to comprehensively assess the failure risk of calcium formate production equipment.

[0161] Taking the logistic regression model as an example:

[0162] A sample set containing historical failure and normal operation records was collected. The sample set includes corrosion-dominated and stress-dominated degradation indicators for each calcium formate production unit under the same operating conditions, as well as actual failure labels (0 indicates no failure, 1 indicates failure). The sample size should include annual operating records from at least 100 units to ensure statistical significance.

[0163] Both corrosion-dominant and stress-dominant degradation indices were subjected to min-max normalization, mapping their values ​​to the [0,1] interval to eliminate the influence of dimensional differences. The normalized corrosion-dominant and stress-dominant degradation indices were defined as X3 and X4, respectively.

[0164] Choose the logistic regression model form: logit(P) = B0 + B1·X3 + B2·X4;

[0165] Where P represents the failure risk probability (ranging from 0 to 1), B0 is the intercept term, and B1 and B2 are the regression coefficients corresponding to the corrosion-dominated degradation index and the stress-dominated degradation index, respectively.

[0166] The maximum likelihood estimation method was used to fit regression coefficients B0, B1, and B2 to all sample data. Cross-validation was employed during training, randomly dividing the sample set into five parts for training and validation separately, in order to select model parameters with strong generalization ability.

[0167] After training, regression coefficients B0, B1, and B2 are obtained. In actual operation, X1 and X2 for each device are calculated and substituted into the logit equation to calculate the failure risk probability P.

[0168] The corrosion-stress synergistic failure risk scoring model is used to calculate the corrosion-dominant degradation index and the stress-dominant degradation index by weighting, and the corrosion-stress synergistic failure risk score is obtained as the synergistic prediction result.

[0169] The corrosion-stress co-failure risk score is generated by post-processing the failure risk probability output by the regression model, reflecting the current overall failure risk level of the equipment.

[0170] The failure risk probability is mapped to a corrosion-stress synergistic failure risk score. The mapping method can be linear mapping or graded mapping. For example, using linear mapping, the corrosion-stress synergistic failure risk score = 100 × failure risk probability. Then, the corrosion-stress synergistic failure risk score ranges from 0 to 100 points, with higher scores indicating greater risk.

[0171] Based on historical data and maintenance experience, the corrosion-stress synergistic failure risk score is divided into three levels: low risk (0 ≤ corrosion-stress synergistic failure risk score < 30), medium risk (30 ≤ corrosion-stress synergistic failure risk score < 70), and high risk (70 ≤ corrosion-stress synergistic failure risk score ≤ 100), to assist in maintenance decision-making. The scoring process intuitively reflects the weighted impact of corrosion and fatigue as two deterioration mechanisms into an operable numerical classification, enabling real-time monitoring and prediction of the synergistic failure risk of calcium formate production equipment.

[0172] S7: Based on the collaborative prediction results and combined with historical failure data, predict the remaining service life of the calcium formate production equipment, including:

[0173] Obtain historical fault data of the calcium formate production equipment, including the time and type of historical faults;

[0174] Historical fault data refers to all equipment faults recorded during the actual operation of the calcium formate production equipment. Historical fault data includes the time of occurrence and the type of fault.

[0175] The time of failure refers to the specific moment or date when the calcium formate production equipment experiences a failure or malfunction during actual production. This time of failure is typically obtained from the historical operating records of the calcium formate production equipment, such as workshop equipment management logs, equipment maintenance record databases, and real-time records stored on remote monitoring platforms.

[0176] The fault type refers to the specific form or phenomenon of failure that occurs in the calcium nail bond production equipment, such as wear failure, corrosion failure, fatigue crack propagation, leakage failure, bearing damage, etc.

[0177] The acquisition of historical fault data should cover the operation records of multiple calcium formate production equipment over many years. For example, in a specific embodiment, the operation fault records of calcium formate production equipment in a chemical plant over the past 5 years can be comprehensively compiled to obtain a dataset containing at least 50 fault events, with the specific time and type of each fault marked in the dataset.

[0178] Statistical analysis of historical failure data was conducted to obtain the lifespan distribution parameters of calcium formate production equipment;

[0179] Historical failure data needs to be processed using statistical analysis methods to extract life distribution parameters that reflect the life characteristics of calcium formate production equipment. Life distribution parameters are mathematical parameters used to describe the pattern of failure time or failure interval time of equipment under a specific failure mode. Common life distributions include exponential distribution, Weibull distribution, normal distribution, or log-normal distribution.

[0180] Specifically, the time of each failure in the historical failure data is first differentially calculated to obtain the interval between the initial operation of the equipment and the first failure, as well as the interval between the first and second failures. This process is repeated to obtain a set of time interval data for failures occurring under normal operating conditions. Using the Weibull distribution as an example, the shape and scale parameters of the Weibull distribution, which reflect the lifespan pattern of the equipment, are obtained by fitting the time interval data.

[0181] For example, the historical failure data of calcium formate production equipment includes 30 corrosion failure events that have occurred since the equipment was put into use. After data processing, 30 equipment failure interval data were obtained, which are a series of values ​​such as 200 days, 250 days, 300 days, and 150 days. The maximum likelihood estimation method or the least squares method was used for fitting analysis, and the shape parameter β of the Weibull distribution was determined to be 1.8, and the scale parameter η was 275 days. This shows that the life of calcium formate production equipment gradually decreases over time. The reliability of calcium formate production equipment is relatively high in the early stage of operation. As the operating time increases, the probability of corrosion failure increases significantly.

[0182] By calculating the lifetime distribution parameters, the patterns and trends of specific types of equipment failures can be described.

[0183] A relationship for estimating remaining service life was constructed based on the life distribution parameters of calcium formate production equipment.

[0184] A remaining service life estimation relationship for calcium formate production equipment was established using the equipment's lifespan distribution parameters. This remaining service life estimation relationship is a mathematical model or functional relationship that can predict the remaining length of time the equipment can safely operate, based on its current operating time, current operating status, and lifespan distribution parameters.

[0185] For example, using Weibull distribution parameters (shape parameter β = 1.8, scale parameter η = 275 days), a relationship for estimating the remaining service life of equipment can be established:

[0186] Assuming the time the calcium formate production equipment has been running is U, its remaining expected lifespan can be calculated using a conditional probability function, such as the conditional expected value formula: L(U)=η·Γ(1+1 / β)−U. Here, Γ represents the Gamma function, and the expected time the equipment can safely continue operating under its current state can be quantified through the lifespan estimation relationship.

[0187] By estimating the lifespan of equipment, managers can monitor the operational safety margin of equipment in real time, better control the timing of equipment maintenance or replacement plans, and reduce losses caused by unexpected downtime and failures.

[0188] Substitute the collaborative prediction results into the remaining useful life estimation relationship to calculate and output the remaining useful life of the calcium formate production equipment.

[0189] The collaborative prediction results are converted into equivalent operating time or correction coefficients, which are then substituted into the life estimation relationship to predict the actual remaining life of the equipment in real time.

[0190] For example, assuming the current corrosion-stress synergistic failure risk score is 70 (high risk), the relationship between the score and the equivalent life reduction factor can be established based on historical data and empirical formulas (for example, the life reduction factor corresponding to a risk score of 70 is 0.7). In this case, the actual effective operating time of the equipment is equivalent to 1 / 0.7 times the original time. Substituting this into the life estimation relationship, the true remaining life of the equipment can be calculated.

[0191] After 200 days of operation, the calcium formate equipment has a risk score of 70 and a lifespan reduction factor of 0.7. The effective operating time is then adjusted to approximately 200 / 0.7 = 286 days. Substituting this into the lifespan relationship again, the remaining lifespan is calculated to be 40 days. After determining the remaining lifespan to be 40 days, on-site personnel are immediately instructed to complete necessary maintenance or component replacement within 40 days to prevent more serious malfunctions and production interruptions. Example 2

[0192] The difference between Embodiment 2 and Embodiment 1 is that this embodiment introduces a fault prediction system for calcium formate production equipment.

[0193] Figure 2 A schematic diagram of a fault prediction system for calcium formate production equipment according to the present invention is provided. The fault prediction system for calcium formate production equipment includes:

[0194] Data acquisition module: Collects environmental data and structural stress response data of the calcium formate production equipment during operation;

[0195] Trend Analysis Module: Based on environmental data, this module analyzes the trends in corrosion rate and corrosion distribution on metal surfaces under different operating conditions and outputs corrosion evolution characteristic parameters.

[0196] Coupled mapping analysis module: Analyzes structural stress response data, simulates the stress concentration area and evolution path of key components of equipment under long-term cyclic loading, and outputs stress evolution characteristic parameters;

[0197] Corrosion index extraction module: Based on corrosion evolution characteristic parameters and the coupling mapping relationship between local fatigue and corrosion acceleration of materials, corrosion-dominant deterioration indexes are extracted.

[0198] Stress index extraction module: Based on stress evolution characteristic parameters and the finite element dynamic response calculation method, stress-dominant degradation indices are extracted.

[0199] Collaborative Analysis Module: Combining corrosion-dominant and stress-dominant degradation indices, a corrosion-stress collaborative failure risk scoring model is constructed, and collaborative prediction results are output.

[0200] Lifespan prediction module: Based on collaborative prediction results and combined with historical failure data, predict the remaining lifespan of calcium formate production equipment.

[0201] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0202] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0203] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0204] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0205] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0206] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0207] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0208] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion 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 this application. 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.

[0209] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0210] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting faults in calcium formate production equipment, characterized in that, Includes the following steps: S1: Collect environmental data and structural stress response data of the calcium formate production equipment during operation, specifically: Environmental data includes temperature, humidity, and acid / alkali concentration data during the operation of the calcium formate production equipment; Structural stress response data includes mechanical load data and thermal stress data borne by key components; S2: Based on environmental data, analyze the changing trends of corrosion rate and corrosion distribution on metal surfaces under different operating environmental conditions, and output corrosion evolution characteristic parameters; S3: Analyze structural stress response data, simulate the stress concentration areas and evolution paths of key components under long-term cyclic loading, and output stress evolution characteristic parameters, specifically: Based on mechanical load data and thermal stress data, a finite element numerical model of key components of calcium formate production equipment was established. The stress distribution of key components under long-term cyclic loading is simulated and calculated based on the finite element numerical model to determine the location and range of stress concentration areas in key components. Based on the changes in the location and extent of stress concentration areas in key components, the evolution path of stress concentration areas is generated; Based on the location, extent, and evolution path of the stress concentration region, stress evolution characteristic parameters are generated; S4: Based on the corrosion evolution characteristic parameters, and the coupling mapping relationship between local fatigue and corrosion acceleration of the material, extract the corrosion-dominant deterioration index; S5: Based on the stress evolution characteristic parameters and the finite element dynamic response calculation method, stress-dominant degradation indices are extracted, specifically: The stress evolution characteristic parameters are input into the finite element dynamic response calculation model to calculate the dynamic stress distribution of key components under the combined action of cyclic mechanical load and thermal stress. Extract the equivalent stress peak value, stress cycle amplitude, and stress change rate from the dynamic stress distribution results; The cumulative fatigue damage of key components is estimated based on the equivalent stress peak value and stress cycle amplitude. The cumulative fatigue damage value, peak equivalent stress, and stress cycle amplitude were extracted as stress-dominant degradation indicators. S6: Combining corrosion-dominant and stress-dominant degradation indices, a corrosion-stress synergistic failure risk scoring model is constructed, and synergistic prediction results are output; S7: Based on the collaborative prediction results and combined with historical fault data, predict the remaining service life of calcium formate production equipment.

2. The method for predicting faults in calcium formate production equipment according to claim 1, characterized in that, S2 includes: Based on temperature, humidity, and acid / alkali concentration data, a statistical analysis was performed on the corrosion rate of metal surfaces under different operating conditions, generating corrosion rate variation curves. Images of corrosion zones on metal surfaces under different operating conditions are acquired, and corrosion depth and corrosion depth distribution data are generated through image processing. Based on corrosion rate variation curves, corrosion depth and corrosion depth distribution data, statistical analysis methods are applied to calculate corrosion depth growth rate, corrosion area ratio and corrosion region expansion path, and output corrosion evolution characteristic parameters.

3. The method for predicting faults in calcium formate production equipment according to claim 2, characterized in that, S4 includes: Based on corrosion evolution characteristic parameters, a coupling relationship between local fatigue and corrosion acceleration in metallic materials is established. Based on the coupling relationship between local fatigue and accelerated corrosion of metallic materials, the degree of influence of corrosion on the fatigue deterioration of metallic materials is determined. Based on the degree of influence of corrosion on the fatigue deterioration of metallic materials, corrosion-dominant deterioration indicators are extracted.

4. The method for predicting faults in calcium formate production equipment according to claim 3, characterized in that, S6 includes: A corrosion-stress synergistic failure risk scoring model is established based on corrosion-dominant and stress-dominant degradation indices. The corrosion-stress synergistic failure risk scoring model is used to calculate the corrosion-dominant and stress-dominant degradation indices by weighting them to obtain a corrosion-stress synergistic failure risk score, which serves as the synergistic prediction result.

5. The method for predicting faults in calcium formate production equipment according to claim 4, characterized in that, S7 includes: Obtain historical fault data of the calcium formate production equipment, including the time and type of historical faults; Statistical analysis of historical failure data was conducted to obtain the lifespan distribution parameters of calcium formate production equipment; A relationship for estimating remaining service life was constructed based on the life distribution parameters of calcium formate production equipment. Substitute the collaborative prediction results into the remaining service life estimation relationship to calculate and output the remaining service life of the calcium formate production equipment.

6. A fault prediction system for calcium formate production equipment, used to implement the fault prediction method for calcium formate production equipment according to any one of claims 1-5, characterized in that, include: Data acquisition module: Collects environmental data and structural stress response data of the calcium formate production equipment during operation, specifically: Environmental data includes temperature, humidity, and acid / alkali concentration data during the operation of the calcium formate production equipment; Structural stress response data includes mechanical load data and thermal stress data borne by key components; Trend Analysis Module: Based on environmental data, this module analyzes the trends in corrosion rate and corrosion distribution on metal surfaces under different operating conditions and outputs corrosion evolution characteristic parameters. Coupled mapping analysis module: Analyzes structural stress response data, simulates the stress concentration regions and evolution paths of key components under long-term cyclic loading, and outputs stress evolution characteristic parameters, specifically: Based on mechanical load data and thermal stress data, a finite element numerical model of key components of calcium formate production equipment was established. The stress distribution of key components under long-term cyclic loading is simulated and calculated based on the finite element numerical model to determine the location and range of stress concentration areas in key components. Based on the changes in the location and extent of stress concentration areas in key components, the evolution path of stress concentration areas is generated; Based on the location, extent, and evolution path of the stress concentration region, stress evolution characteristic parameters are generated; Corrosion index extraction module: Based on corrosion evolution characteristic parameters and the coupling mapping relationship between local fatigue and corrosion acceleration of materials, corrosion-dominant deterioration indexes are extracted. Stress index extraction module: Based on stress evolution characteristic parameters and the finite element dynamic response calculation method, it extracts stress-dominant degradation indices, specifically: The stress evolution characteristic parameters are input into the finite element dynamic response calculation model to calculate the dynamic stress distribution of key components under the combined action of cyclic mechanical load and thermal stress. Extract the equivalent stress peak value, stress cycle amplitude, and stress change rate from the dynamic stress distribution results; The cumulative fatigue damage of key components is estimated based on the equivalent stress peak value and stress cycle amplitude. The cumulative fatigue damage value, peak equivalent stress, and stress cycle amplitude were extracted as stress-dominant degradation indicators. Collaborative Analysis Module: Combining corrosion-dominant and stress-dominant degradation indices, a corrosion-stress collaborative failure risk scoring model is constructed, and collaborative prediction results are output. Lifespan prediction module: Based on collaborative prediction results and combined with historical failure data, predict the remaining lifespan of calcium formate production equipment.

Images