Method and system for predicting the degree of deterioration of a press and of key components

Abstract

The application provides a press machine and a key component deterioration degree prediction method and system, and applies to the technical field of press machine fault diagnosis. The method comprises the following steps: obtaining a plurality of abnormal operation data corresponding to a key component of a press machine within a preset time length, and constructing a target failure time sequence according to the failure time of each of the plurality of abnormal operation data; wherein, in the case of a single key component, the target is the key component, and in the case of multiple key components, the target is the press machine; constructing a first failure distribution function according to the failure time sequence, the shape parameter and the scale parameter of the Weibull distribution, and the first failure distribution function is used to predict the failure degree trend of the target; determining the first maximum likelihood estimation value corresponding to the shape parameter and the second maximum likelihood estimation value corresponding to the scale parameter according to the first failure distribution function, and then combining the predicted failure time of the target to predict the deterioration degree of the target. Meet the preventive maintenance needs under the background of modern intelligent manufacturing.

Description

Technical Field

[0001] This application relates to the field of press fault diagnosis technology, and in particular to a method and system for predicting the degree of deterioration of a press and its key components. Background Technology

[0002] In modern intelligent manufacturing systems, presses are core equipment in production lines for metal forming and stamping, and their operational reliability directly affects production continuity, product quality, and equipment safety. Key components of presses, such as the main transmission mechanism, bearings, and slides, are subjected to complex stresses such as impact, torsion, and friction under cyclic alternating loads, leading to frequent progressive failures such as fatigue damage accumulation, accelerated wear, and deterioration of lubrication performance.

[0003] Existing methods for predicting the deterioration of presses and key components often employ threshold alarm methods, empirical life methods, and linear regression prediction methods. However, these methods have their own limitations, resulting in an inability to effectively characterize the nonlinear randomness of press and key component deterioration, low prediction accuracy, and insufficient lead time, which cannot meet the pre-maintenance requirements in the context of modern intelligent manufacturing. Summary of the Invention

[0004] This application provides a method and system for predicting the degradation degree of a press and its key components. By combining the failure time series of the press and key components with the failure distribution function of the Weibull distribution, it achieves an accurate characterization of the nonlinear degradation process of the press and key components, significantly improving the lead time and accuracy of fault prediction, thereby supporting predictive maintenance decisions in a smart manufacturing environment. Compared with traditional threshold alarm methods and empirical life methods, this method can overcome the limitations of linear assumptions through data-driven Weibull modeling, achieving an upgrade from binary alarms to continuous probability prediction, and from static empirical values ​​to dynamic adaptive estimation, significantly improving the robustness and adaptability of press and key component prediction.

[0005] This application provides a method for predicting the deterioration degree of a press and its key components, including: Within a preset time period, multiple abnormal operating data corresponding to the key components of the press are acquired, and a failure time sequence of the target is constructed based on the failure time of each of the multiple abnormal operating data; wherein, when the key component is a single component, the target is the key component, and when the key components are multiple components, the target is the press. Based on the failure time series, the shape parameter and scale parameter of the Weibull distribution, a first failure distribution function of the Weibull distribution is constructed. The first failure distribution function is used to predict the failure trend of the target. Based on the first failure distribution function, determine the first maximum likelihood estimate corresponding to the shape parameter and the second maximum likelihood estimate corresponding to the scale parameter; The degree of degradation of the target is predicted based on the target's predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate.

[0006] According to an embodiment of this application, a method for predicting the deterioration degree of a press and its key components is provided. When the key component is a single unit, the method for acquiring multiple abnormal operating data corresponding to the key component includes: when the key component is a bearing in a transmission system, acquiring axial vibration data, radial vibration data, and bearing temperature data of the bearing at various times, and determining multiple abnormal operating data corresponding to the bearing from the multiple axial vibration data, multiple radial vibration data, and multiple bearing temperature data; when the key component is a slider in a motion mechanism, acquiring displacement data of the slider at various times, and determining multiple abnormal operating data corresponding to the slider from the multiple displacement data; when the key component is a brake hydraulic monitoring system in a flywheel brake system... In the case of a detector, the first oil pressure data and the first oil temperature data of the brake hydraulic monitor at each time moment are acquired, and multiple abnormal operating data corresponding to the brake hydraulic monitor are determined from multiple first oil pressure data and multiple first oil temperature data; in the case where the key component is the hydraulic cylinder of the flywheel brake system, the second oil pressure data and the second oil temperature data of the hydraulic cylinder at each time moment are acquired, and multiple abnormal operating data corresponding to the hydraulic cylinder are determined from multiple second oil pressure data and multiple second oil temperature data; in the case where the key component is the balance bar air supply pipe of the balance system, the air pressure data of the balance bar air supply pipe at each time moment is acquired, and multiple abnormal operating data corresponding to the balance bar air supply pipe are determined from multiple air pressure data.

[0007] According to an embodiment of this application, a method for predicting the deterioration degree of a press and its key components is provided. When there are multiple key components, the method for acquiring multiple abnormal operating data corresponding to the key components of the press includes: acquiring axial vibration data, radial vibration data, and bearing temperature data of the bearing at various times; displacement data of the slider at various times; first oil pressure data and first oil temperature data of the brake hydraulic monitor at various times; second oil pressure data and second oil temperature data of the hydraulic cylinder at various times; and air pressure data of the balance bar air supply pipe at various times; and determining multiple abnormal operating data corresponding to the press from the multiple axial vibration data, multiple radial vibration data, multiple bearing temperature data, multiple displacement data, multiple first oil pressure data, multiple first oil temperature data, multiple second oil pressure data, multiple second oil temperature data, and multiple air pressure data.

[0008] According to an embodiment of this application, a method for predicting the deterioration degree of a press and its key components includes determining a first maximum likelihood estimate corresponding to the shape parameter and a second maximum likelihood estimate corresponding to the scale parameter based on a first failure distribution function. This includes: differentiating the first failure distribution function to obtain a probability density function of a Weibull distribution; constructing a maximum likelihood function based on the probability density function; taking the logarithm of the maximum likelihood function to obtain an objective function; obtaining a first partial derivative function based on the objective function with respect to the shape parameter, and obtaining a second partial derivative function based on the scale parameter; solving for the first maximum likelihood estimate based on the first partial derivative function, and solving for the second maximum likelihood estimate based on the second partial derivative function.

[0009] According to an embodiment of this application, a method for predicting the degradation degree of a press and its key components includes: predicting the degradation degree of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate; predicting the reliability of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with a second failure distribution function; predicting the reliability of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with a reliability function; predicting the failure degree of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with a failure degree function; and determining the degradation degree of the target based on the failure degree, the reliability, and the failure degree.

[0010] According to an embodiment of this application, a method for predicting the deterioration degree of a press and its key components is provided, wherein the failure degree is in the range [0, 0.1], the reliability is in the range [0.9, 1], and the fault degree is in the range [0, 0.1]. In the case of the specified range, the degree of degradation of the target meets the preset requirements; wherein, ; This represents the average fault rate within the preset time period; This represents the standard deviation of the failure rate within the preset time period.

[0011] According to the method for predicting the deterioration degree of a press and its key components provided in the embodiments of this application, the calculation formula of the first failure distribution function is as follows: ;in, This represents the failure time series; This indicates the degree of failure of the target; Indicates the shape parameters; This represents the scale parameter.

[0012] According to the embodiment of this application, a method for predicting the deterioration degree of a press and its key components is provided, wherein the probability density function is calculated using the following formula: ; Indicates the degree of failure The derivative of the maximum likelihood function is given by the following formula: ; Denotes the maximum likelihood function; Indicates the total number of times failure occurred; Indicates the first time in the failure time series The failure time; the formula for calculating the objective function is: +( ; The derivative result represents... The logarithm of the result.

[0013] According to an embodiment of this application, a method for predicting the deterioration degree of a press and its key components includes: obtaining a first maximum likelihood estimate based on a first partial derivative function and obtaining a second maximum likelihood estimate based on a second partial derivative function; wherein, the calculation formula for the first partial derivative function is: setting the first partial derivative function equal to 0 to obtain the first maximum likelihood estimate, and setting the second partial derivative function equal to 0 to obtain the second maximum likelihood estimate; wherein, the calculation formula for the first partial derivative function is: The formula for calculating the second partial derivative is: .

[0014] This application also provides a system for predicting the deterioration degree of a press and its key components, including: The data processing module is used to acquire multiple abnormal operating data corresponding to the key components of the press within a preset time period; A time series construction module is used to construct a failure time series of the target based on the failure times of the multiple abnormal operation data; wherein, when the key component is a single component, the target is the key component, and when the key components are multiple components, the target is the press. The target degradation prediction module is used to construct a first failure distribution function of the Weibull distribution based on the failure time series, the shape parameter and the scale parameter of the Weibull distribution, and the first failure distribution function is used to predict the failure trend of the target; based on the first failure distribution function, determine a first maximum likelihood estimate corresponding to the shape parameter and a second maximum likelihood estimate corresponding to the scale parameter; and predict the degree of degradation of the target based on the target's expected failure time, the first maximum likelihood estimate and the second maximum likelihood estimate.

[0015] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the degradation prediction method for the press and key components as described above.

[0016] This application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the degradation prediction method for the press and key components as described above.

[0017] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the degradation prediction method for the press and key components as described above.

[0018] The method and system for predicting the degradation degree of a press and its key components provided in this application acquire multiple abnormal operating data corresponding to the key components of the press within a preset time period, and construct a failure time series of the target based on the failure time of each of the multiple abnormal operating data. Wherein, if the key component is single, the target is the key component; if the key components are multiple, the target is the press. A first failure distribution function of the Weibull distribution is constructed based on the failure time series, the shape parameter and the scale parameter of the Weibull distribution. This first failure distribution function is used to predict the failure degree trend of the target. A first maximum likelihood estimate corresponding to the shape parameter and a second maximum likelihood estimate corresponding to the scale parameter are determined based on the first failure distribution function. The degradation degree of the target is predicted based on the predicted failure time of the target, the first maximum likelihood estimate, and the second maximum likelihood estimate. This method, by combining the failure time series of the press and its key components with the failure distribution function of the Weibull distribution, achieves an accurate characterization of the nonlinear degradation process of the press and its key components, significantly improving the lead time and accuracy of fault prediction, thereby supporting predictive maintenance decisions in an intelligent manufacturing environment. Compared to traditional threshold alarm methods and empirical life methods, this method can overcome the limitations of linear assumptions through data-driven Weibull modeling, achieving an upgrade from binary alarms to continuous probability prediction and from static empirical values ​​to dynamic adaptive estimation, significantly improving the robustness and adaptability of prediction for presses and key components. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the method for predicting the deterioration degree of a press and its key components provided in the embodiments of this application. Figure 2 This is a schematic diagram showing the prediction curve results of the failure trend and reliability trend provided in the embodiments of this application; Figure 3 This is a schematic diagram showing the trend of the failure degree as the prediction curve result provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of the deterioration prediction system for the press and key components provided in the embodiments of this application; Figure 5 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] It should be noted that the execution subject involved in the embodiments of this application can be a deterioration prediction system for presses and key components, or it can be an electronic device. Optionally, the electronic device may include: a computer / laptop, a mobile terminal, a server, electronic assembly equipment, and electrical production equipment, etc.

[0023] The following section uses electronic equipment as an example to illustrate in detail the method for predicting the deterioration degree of the press and key components provided in the embodiments of this application: Figure 1 This is a flowchart illustrating the method for predicting the deterioration level of a press and its key components provided in an embodiment of this application. Figure 1 As shown, the method includes the following steps 101-104.

[0024] Step 101: Within a preset time period, acquire multiple abnormal operating data corresponding to the key components of the press, and construct the failure time sequence of the target based on the failure time of each of the multiple abnormal operating data; wherein, when there is a single key component, the target is the key component, and when there are multiple key components, the target is the press.

[0025] The preset duration refers to the data acquisition time window set according to the operating characteristics and maintenance cycle of the press. The specific value of the preset duration is determined according to the equipment type, working system and usage frequency. For example, a press that produces continuously can be set to an analysis cycle of 30 days (i.e., the preset duration).

[0026] A press is a piece of equipment used in the metal forming and manufacturing industry to generate enormous pressure through hydraulic, mechanical, or servo drives for processes such as stamping, forging, and extrusion.

[0027] Key components refer to the core subsystem components of a press that directly affect the equipment's function, safety, or accuracy.

[0028] Abnormal operating data refers to sensor monitoring data that exceeds the normal operating range.

[0029] The failure time refers to the time when the fault occurs corresponding to the abnormal operating data.

[0030] Failure time series refers to the set of failure moments arranged in chronological order, used for Weibull distribution modeling.

[0031] In step 101, when there is only one critical component, within a preset time period, the electronic device can acquire multiple operational status data corresponding to this single critical component based on its operational characteristics, thereby identifying multiple abnormal operational data. Based on the failure times of each of these abnormal operational data, a failure time series for the critical component is constructed. This entire process, by focusing on the abnormal data of a single critical component to construct a failure time series, achieves a precise characterization of the individual degradation pattern of that critical component, providing a reliable basis for targeted maintenance.

[0032] When there are multiple critical components, within a preset time period, the electronic device can acquire multiple operating status data corresponding to each critical component based on the operating characteristics of the press. This allows it to determine at least one abnormal operating data point for each critical component, and then construct a failure time series for the press based on the failure times of all acquired abnormal operating data. The entire process, by fusing abnormal data from multiple critical components to construct the overall machine failure time series, achieves a comprehensive evaluation of the press and supports optimized decision-making for overall maintenance strategies.

[0033] In some embodiments, when there is only one critical component, the electronic device acquires multiple abnormal operating data corresponding to the critical component of the press, which may include one of the following implementation methods: Implementation Method 1: When the key component is the bearing of the transmission system, the electronic device acquires the axial vibration data, radial vibration data and bearing temperature data of the bearing at various times, and determines the multiple abnormal operating data corresponding to the bearing from multiple axial vibration data, multiple radial vibration data and multiple bearing temperature data.

[0034] In implementation method 1, a vibration sensor is installed axially and radially on each of the two motor bearing housings of the bearing to collect axial and radial vibration data of the bearing at various times. A temperature sensor is installed near the bearing housing and bearing casing to collect bearing temperature data at various times. Then, the electronic equipment identifies data exceeding a first vibration threshold from the multiple axial vibration data, a second vibration threshold from the multiple radial vibration data, and a preset maximum allowable temperature from the multiple bearing temperature data as multiple abnormal operating data corresponding to the bearing. The second vibration threshold is 1.5 times the baseline value. This entire process, by setting quantitative abnormal thresholds for axial vibration data, radial vibration data, and bearing temperature data, can automatically and accurately identify the abnormal operating state of the bearing, providing a reliable data foundation for modeling the degradation degree based on the bearing's failure time series.

[0035] Implementation Method 2: When the key component is a slider of a motion mechanism, the electronic device acquires the displacement data of the slider at various times and determines multiple abnormal operation data corresponding to the slider from multiple displacement data.

[0036] In implementation method 2, a vibration sensor is installed on each of the four corner columns of the press body to measure the vibration data of the four corners of the press body. A displacement sensor is installed on the inner surface of the columns to measure the displacement data of the slider relative to the theoretical running trajectory at various times. Then, the electronic equipment identifies data exceeding the slider's allowable parallelism tolerance (e.g., ±0.02 mm / m), or where the displacement fluctuation amplitude (the difference in displacement between adjacent cycles) exceeds a set threshold (e.g., 0.5% of the rated stroke), and where the displacement data shows a strong correlation with the vibration data of the four corners of the press body at the corresponding time (e.g., abnormal displacement in a specific direction accompanied by a sudden increase in vibration of the corresponding side column), as multiple abnormal operating data corresponding to the slider. The entire process, by monitoring the offset and fluctuation of the slider's motion trajectory and its correlation with the press body vibration, can accurately identify abnormal slider operating states caused by guide rail wear, increased clearance, or motion mechanism failure, providing a reliable data foundation for modeling the degradation degree based on the slider's failure time series.

[0037] Implementation Method 3: When the key component is the brake hydraulic monitor of the flywheel brake system, the electronic device acquires the first oil pressure data and the first oil temperature data of the brake hydraulic monitor at each time, and determines the multiple abnormal operation data corresponding to the brake hydraulic monitor from multiple first oil pressure data and multiple first oil temperature data.

[0038] In implementation method 3, a pressure sensor and a temperature sensor are installed at the brake hydraulic monitor to collect the first oil pressure data and the first oil temperature data at various times. Then, the electronic equipment identifies data from multiple first oil pressure data points that are below the brake's minimum operating pressure threshold (e.g., 80% of the rated pressure) or above the system safety valve's set pressure; and data from multiple first oil temperature data points that exceed the upper limit of the hydraulic oil's allowable operating temperature (e.g., 70°C) or whose temperature rise rate per unit time exceeds a set value (e.g., 5°C / minute). These are identified as multiple abnormal operating data points corresponding to the brake hydraulic monitor. The entire process, by monitoring the pressure stability and oil temperature changes of the brake hydraulic fluid, can promptly identify brake performance degradation or delays caused by pressure leakage, oil contamination, or overheating in the flywheel brake system. This provides key monitoring indicators for ensuring flywheel brake safety and response reliability, and also provides a reliable data foundation for modeling the degradation degree based on the failure time series of the brake hydraulic monitor.

[0039] Implementation Method 4: When the key component is the hydraulic cylinder of the flywheel brake system, the electronic device acquires the second oil pressure data and the second oil temperature data of the hydraulic cylinder at each moment, and determines the multiple abnormal operation data corresponding to the hydraulic cylinder from multiple second oil pressure data and multiple second oil temperature data.

[0040] In implementation method 4, a pressure sensor and a temperature sensor are installed inside the hydraulic cylinder to collect the second oil pressure data and second oil temperature data of the hydraulic cylinder at various times. Then, the electronic equipment identifies data from multiple second oil pressure data that show abnormal pressure surges (e.g., pressure peak exceeding 150% of the rated value), abnormally prolonged pressure build-up time, or pressure imbalance between the two chambers exceeding the allowable value (e.g., 10%) during the pressure holding phase. It also identifies data from multiple second oil temperature data that exceed the allowable temperature of the cylinder seals or have an excessively large temperature difference between the oil temperature and the main oil circuit temperature. These are identified as multiple abnormal operating data corresponding to the hydraulic cylinder. The entire process, by monitoring the dynamic pressure characteristics and local temperature of the hydraulic cylinder during operation, can effectively identify early faults such as internal leakage, seal aging, or cylinder wear, providing a reliable data foundation for modeling the degradation degree based on the failure time series of the hydraulic cylinder.

[0041] Implementation Method 5: When the key component is the balance bar air supply pipe of the balance system, the electronic device acquires the air pressure data of the balance bar air supply pipe at various times, and determines multiple abnormal operation data corresponding to the balance bar air supply pipe from multiple air pressure data.

[0042] In implementation method 5, a pressure sensor is installed at the balance bar air supply pipe to collect air pressure data at various times. Then, the electronic equipment identifies data points that are lower than the minimum balancing pressure required for the slider's weight, or whose pressure fluctuations exceed a set range (e.g., ±0.05 MPa), and data showing abnormal pressure following behavior during the slider's operating cycle (e.g., pressure changes lagging behind the crankshaft phase for more than a set time), as multiple abnormal operating data corresponding to the balance bar air supply pipe. This entire process, by monitoring the stability and dynamic response performance of the balancing system's air supply pressure, can accurately diagnose problems such as air circuit leaks, pressure regulating valve malfunctions, or response delays, ensuring the smooth operation of the slider and indirectly protecting the transmission mechanism. It provides a quantitative means for evaluating the balancing system's performance and a reliable data foundation for modeling the degradation degree based on the failure time series of the balance bar air supply pipe.

[0043] In some embodiments, when there are multiple key components, the electronic device acquires multiple abnormal operating data corresponding to the key components of the press. This may include: when the key components include the bearings of the transmission system, the slider of the motion mechanism, the brake hydraulic monitor and hydraulic cylinder of the flywheel brake system, and the balance bar air supply pipe of the balance system, the electronic device acquires axial vibration data, radial vibration data, and bearing temperature data of the bearings at various times, displacement data of the slider at various times, first oil pressure data and first oil temperature data of the brake hydraulic monitor at various times, second oil pressure data and second oil temperature data of the hydraulic cylinder at various times, and air pressure data of the balance bar air supply pipe at various times; the electronic device determines multiple abnormal operating data corresponding to the press from multiple axial vibration data, multiple radial vibration data, multiple bearing temperature data, multiple displacement data, multiple first oil pressure data, multiple first oil temperature data, multiple second oil pressure data, multiple second oil temperature data, and multiple air pressure data.

[0044] In this embodiment of the application, from the perspective of the entire press, after the electronic device acquires the axial vibration data, radial vibration data, and bearing temperature data of the bearing at various times, the displacement data of the slider at various times, the first oil pressure data and first oil temperature data of the brake hydraulic monitor at various times, the second oil pressure data and second oil temperature data of the hydraulic cylinder at various times, and the air pressure data of the balance bar air supply pipe at various times, it can combine the processes of the above-described implementation methods 1-5 to determine the multiple abnormal operating data corresponding to the press by identifying at least one abnormal data among the multiple axial vibration data, at least one abnormal data among the multiple radial vibration data, at least one abnormal data among the multiple bearing temperature data, at least one abnormal data among the multiple displacement data, at least one abnormal data among the multiple first oil pressure data, at least one abnormal data among the multiple first oil temperature data, at least one abnormal data among the multiple second oil pressure data, at least one abnormal data among the multiple second oil temperature data, and at least one abnormal data among the multiple air pressure data. The entire process involves comprehensively judging the anomalies in multi-source sensor data of key components of the press, thereby achieving systematic monitoring of the overall health status of the machine. This provides comprehensive and reliable data support for constructing the failure time series of the entire machine and achieving accurate degradation modeling.

[0045] Optionally, the electronic device constructs a failure time series of the target based on the failure times of each of the multiple abnormal operating data. This may include: for each abnormal operating data, the electronic device performs noise reduction, smoothing, and normalization processing on the abnormal operating data to obtain the target abnormal operating data, and takes the failure time of the target abnormal operating data as the failure time of the target abnormal operating data; the electronic device constructs a failure time series of the target based on the failure times of each of the multiple target abnormal operating data.

[0046] In this embodiment, the abnormal operating data collected by the electronic device may contain noise interference, outliers, and missing data, affecting its accuracy and precision. This can lead to inaccurate determination of the subsequent failure time. Therefore, after identifying the abnormal operating data, the electronic device can perform denoising (e.g., using wavelet thresholding), smoothing (e.g., using moving average or exponential weighted smoothing), and normalization (to eliminate the influence of dimensions) to obtain target abnormal operating data with better quality and higher accuracy. The failure time of this target abnormal operating data is then used as its failure time, ensuring high accuracy. Then, when there is only one critical component, the electronic device can construct a failure time series for that single critical component based on the failure times of multiple target abnormal operating data. When there are multiple critical components, the electronic device can construct a failure time series for the entire press based on the failure times of the target abnormal operating data. The entire process aims to improve data quality, reduce the impact of noise on the modeling results, and form a key failure time series that can be used for degradation degree modeling. This provides reliable input for subsequent degradation degree modeling based on Weibull distribution.

[0047] Step 102: Based on the failure time series, the shape parameter and scale parameter of the Weibull distribution, construct the first failure distribution function of the Weibull distribution. The first failure distribution function is used to predict the failure trend of the target.

[0048] Among them, the shape parameter is used to describe the acceleration properties of the failure rate.

[0049] The scale parameter is used to describe the target's average lifetime level.

[0050] The first failure distribution function is a reliability prediction model that enables reliability prediction of the degradation degree of presses and key components. It can effectively characterize the risk growth process of presses and key components from initial stability to accelerated aging.

[0051] In some embodiments, the formula for calculating the first failure distribution function is: .

[0052] in, Represents the failure time series. , This represents the total number of times the failure occurred. Indicates the first time series of failures One failure time; Indicates the degree of failure of the target; Indicates shape parameters; Indicates the scale parameter.

[0053] Step 103: Based on the first failure distribution function, determine the first maximum likelihood estimate corresponding to the shape parameter and the second maximum likelihood estimate corresponding to the scale parameter.

[0054] Among them, the largest likelihood estimate is the optimal estimate of the shape parameter, which quantitatively describes the acceleration pattern of the target failure risk over time.

[0055] The second largest likelihood estimate is the optimal estimate of the scale parameter, which quantitatively describes the characteristic lifetime scale of the target.

[0056] The following section elaborates on how electronic devices determine the first maximum likelihood estimate of the shape parameter and the second maximum likelihood estimate of the scale parameter based on the first failure distribution function: In some embodiments, the electronic device determines a first maximum likelihood estimate corresponding to the shape parameter and a second maximum likelihood estimate corresponding to the scale parameter based on a first failure distribution function. This may include: the electronic device taking the derivative of the first failure distribution function to obtain the probability density function of the Weibull distribution, and constructing a maximum likelihood function based on the probability density function; the electronic device taking the logarithm of the maximum likelihood function to obtain an objective function; the electronic device taking the partial derivative of the objective function with respect to the shape parameter to obtain a first partial derivative function, and taking the partial derivative of the objective function with respect to the scale parameter to obtain a second partial derivative function; the electronic device solving for the first maximum likelihood estimate based on the first partial derivative function, and solving for the second maximum likelihood estimate based on the second partial derivative function.

[0057] In some embodiments, the electronic device responds to the first failure distribution function Taking the derivative, we obtain the probability density function, which is calculated using the following formula: ; Indicates the degree of failure The derivative of .

[0058] In some embodiments, the electronic device constructs a maximum likelihood function based on the probability density function described above. The formula for calculating the maximum likelihood function is as follows: ; This represents the maximum likelihood function.

[0059] The construction process of the maximum likelihood function is as follows: for the observed... n Each independent failure time is as follows: Assume that all failure times follow a parametric distribution as follows ( , If the distribution follows a Weibull distribution, then the first... Failure time The probability of occurrence is determined by the probability density function. , Given that the observations are independent, the joint probability density of the failure time series is the product of the probabilities of occurrence at each failure time, which is defined with respect to the parameter ( , Maximum likelihood function .

[0060] At this point, the expression for the maximum likelihood function is: .

[0061] Next, the formula for calculating the probability density function will be presented. Substitute into the expression of the maximum likelihood function The formula for calculating the maximum likelihood function can then be obtained. .

[0062] It should be noted that the maximum likelihood function It intuitively reflects the situation under given parameters ( , Under this condition, the "probability" of observing this specific set of failure time data is determined. This is achieved by finding parameters (…). , Maximize the parameter value ( , This allows us to obtain the optimal estimate of the overall parameters. Among them, Represents shape parameters The corresponding first maximum likelihood estimate; Representing scale parameters The corresponding second maximum likelihood estimate; In some embodiments, the electronic device takes the logarithm of the maximum likelihood function to obtain the objective function, which is calculated as follows: +( ; Represent the result of differentiation The logarithm of the result.

[0063] In some embodiments, the electronic device obtains a first maximum likelihood estimate based on a first partial derivative function and a second maximum likelihood estimate based on a second partial derivative function. This may include: the electronic device sets the first partial derivative function to 0 to obtain the first maximum likelihood estimate and sets the second partial derivative function to 0 to obtain the second maximum likelihood estimate.

[0064] The formula for calculating the first partial derivative is as follows: ; The formula for calculating the second partial derivative is: .

[0065] In this embodiment of the application, after the electronic device obtains a first partial derivative function by taking the partial derivative of the objective function with respect to the shape parameter, and obtains a second partial derivative function by taking the partial derivative of the objective function with respect to the scale parameter, the electronic device then... and order Then, by solving the system of equations, the first maximum likelihood estimate can be obtained. Second maximum likelihood estimate .

[0066] It should be noted that the first maximum likelihood estimate... Second maximum likelihood estimate The data is updated based on facts and does not rely on specific equipment experience curves, and can adapt to different loads, frequencies, speeds, and lubrication conditions.

[0067] Step 104: Based on the target's predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, predict the target's degree of degradation.

[0068] The predicted failure time refers to a specific point in time used to assess the future degradation state of the target.

[0069] In step 104, the electronic device can estimate the degree of degradation of the target at any future time based on the identified model parameters (i.e., the first maximum likelihood estimate and the second maximum likelihood estimate) and the target's predicted failure time.

[0070] Understandably, when an electronic device constructs a predictable failure time series based on multiple predictable failure times of a target, the degree of degradation of the target at the corresponding predictable failure time can be predicted based on each predictable failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate in the predictable failure time series. Then, based on all the degrees of degradation, the degradation trend of the target can be determined, and a quantifiable degradation trend prediction curve can be generated.

[0071] The following section elaborates on how electronic devices predict the degree of degradation of a target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate: In some embodiments, the electronic device predicts the degree of degradation of the target based on the predicted failure time, a first maximum likelihood estimate, and a second maximum likelihood estimate. This may include: the electronic device predicting the degree of failure of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with a second failure distribution function; the electronic device predicting the reliability of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with a reliability function; the electronic device predicting the degree of failure of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with a failure degree function; and the electronic device determining the degree of degradation of the target based on the degree of failure, the reliability, and the failure degree.

[0072] Optionally, the formula for calculating the second failure distribution function is: ; The reliability function is: ; The fault degree function is: ; in, Indicates the time to be predicted for failure; Indicates the degree of failure; Indicates reliability; This indicates the failure rate, which can also be called the failure rate.

[0073] It is understandable that when an electronic device constructs a time series of predicted failures based on multiple predicted failure times of a target, it can generate failure rate trends, reliability trends, and fault rate trends, and obtain quantifiable prediction curve results for health management (such as remaining life estimation).

[0074] For example, Figure 2 This is a schematic diagram showing the prediction curve results of the failure trend and reliability trend provided in the embodiments of this application; Figure 3 This is a schematic diagram illustrating the trend of the failure rate based on the prediction curve results provided in the embodiments of this application. From Figure 2 As can be seen, the curve corresponding to the second failure distribution function (i.e., the cumulative failure probability curve) monotonically increases with the increase of failure time, asymptotically approaching 1 from 0, intuitively showing the cumulative failure probability of the target at various future time points; while the curve of the reliability function (the other curve) monotonically decreases with the increase of failure time, asymptotically approaching 0 from 1, reflecting the decay process of the probability of the target maintaining normal operation. The two curves, in a complementary relationship, jointly depict the overall picture of the reliability evolution of the target. Figure 3 As can be seen, the curve corresponding to the failure rate function (i.e., the failure rate function curve) exhibits a non-constant trend as it changes with failure time, and its specific shape is determined by the shape parameter. Decision, specifically: if >1, the curve shows an upward trend, indicating that the target has entered the wear and aging stage, and the risk of failure increases rapidly over time; if =1, the curve remains horizontal, indicating a constant failure risk; if A value less than 1 indicates a declining curve, suggesting an early failure phase. This curve directly reflects the instantaneous failure risk intensity of the target at any given time.

[0075] Specifically, the remaining lifetime estimation process is as follows: based on the current time... and preset reliability threshold (For example, 0.1, meaning the allowable failure probability is 90%), by solving the reliability function. = The predicted failure time is obtained. The formula for calculating the remaining useful life (RUL) is: RUL = - Combining the above reliability function, we get: Solving for the problem yields the following: This allows us to obtain a quantitative estimate of the remaining lifespan.

[0076] It should be noted that the entire process enables accurate prediction of remaining lifespan.

[0077] Optionally, the electronic device determines the target preset lifespan interval corresponding to the quantitative estimate of the remaining lifespan from multiple preset lifespan intervals, and outputs the warning information of the target preset lifespan interval.

[0078] Understandably, different preset lifespan intervals correspond to different early warning information. By mapping the quantitative estimate of the remaining lifespan to discrete preset lifespan intervals and outputting the corresponding standardized early warning information, scientific early warnings can be provided for press operation. This realizes the transformation from complex prediction results to concise maintenance instructions, significantly improving the operability and execution efficiency of predictive maintenance decisions.

[0079] In some embodiments, when the failure degree is in the range [0, 0.1], the reliability is in the range [0.9, 1], and ...]. In the case of the range, the degree of degradation of the target meets the preset requirements.

[0080] in, ; This represents the average failure rate over a preset time period; This represents the standard deviation of the failure rate over a preset time period.

[0081] It should be noted that the entire process described in steps 101-104 achieves high-precision prediction of the degradation trend of the press and its key components by constructing a degradation model, estimating model parameters, and updating and predicting the remaining life in real time. Compared with linear trend extrapolation, the prediction error of the above process can be reduced by 30%–60%.

[0082] In this embodiment, the technical solution described in steps 101-104 above, by combining the failure time series of the press and key components with the failure distribution function of the Weibull distribution, achieves an accurate characterization of the nonlinear degradation process of the press and key components, significantly improving the lead time and accuracy of fault prediction, thereby supporting predictive maintenance decisions in an intelligent manufacturing environment. Compared with traditional threshold alarm methods and empirical life methods, this method can overcome the limitations of linear assumptions through data-driven Weibull modeling, achieving an upgrade from binary alarms to continuous probability prediction, and from static empirical values ​​to dynamic adaptive estimation, significantly improving the robustness and adaptability of press and key component prediction.

[0083] It should be noted that the embodiments of this application can also use Autoregressive Integrated Moving Average (ARIMA) to directly model and predict the time series of degradation indicators (such as vibration trends, temperature growth rates, etc.), rather than modeling the failure time itself. Specifically, by constructing an ARIMA(p,d,q) model, the future degradation trajectory is directly predicted by identifying the autocorrelation and moving average characteristics of the sequence, and the remaining lifetime is calculated accordingly, avoiding the limitations of preset failure distribution forms. Here, p represents the autoregressive order; d represents the difference order; and q represents the moving average order.

[0084] The embodiments of this application can also predict remaining lifetime based on Bayesian methods. Specifically, in Bayesian methods, the shape and size parameters of the Weibull distribution are treated as random variables. By fusing prior knowledge with observational data, the posterior distribution of the parameters is dynamically updated, thereby providing a probability distribution estimate and confidence interval for remaining lifetime, which significantly enhances the reliability of the prediction results. This is particularly suitable for scenarios with small samples or where data arrives gradually.

[0085] The following describes the deterioration prediction system for presses and key components provided in the embodiments of this application. The deterioration prediction system for presses and key components described below can be referred to in correspondence with the deterioration prediction method for presses and key components described above.

[0086] Figure 4 This is a schematic diagram of the structure of the deterioration prediction system for the press and key components provided in the embodiments of this application. Figure 4As shown, the system includes: a data processing module 401, a time series construction module 402, and a target degradation prediction module 403.

[0087] The data processing module 401 is used to acquire multiple abnormal operation data corresponding to the key components of the press within a preset time period; The time series construction module 402 is used to construct the failure time series of the target based on the failure times of the multiple abnormal operation data; wherein, when the critical component is a single component, the target is the critical component, and when the critical component is multiple components, the target is the press. The target degradation prediction module 403 is used to construct a first failure distribution function of the Weibull distribution based on the failure time series, the shape parameter and the scale parameter of the Weibull distribution, and the first failure distribution function is used to predict the failure trend of the target; based on the first failure distribution function, determine the first maximum likelihood estimate corresponding to the shape parameter and the second maximum likelihood estimate corresponding to the scale parameter; and predict the degree of degradation of the target based on the target's expected failure time, the first maximum likelihood estimate and the second maximum likelihood estimate.

[0088] Optionally, when the key component is a single unit, the data processing module 401 is specifically used to: acquire axial vibration data, radial vibration data, and bearing temperature data of the bearing at various times when the key component is a bearing in the transmission system; and determine multiple abnormal operating data corresponding to the bearing from multiple axial vibration data, multiple radial vibration data, and multiple bearing temperature data; acquire displacement data of the slider at various times when the key component is a slider in a motion mechanism; and determine multiple abnormal operating data corresponding to the slider from multiple displacement data; and acquire the brake hydraulic monitoring data of the flywheel brake system when the key component is a brake hydraulic monitoring device. The system acquires first oil pressure and first oil temperature data at each given time, and determines multiple abnormal operating data corresponding to the brake hydraulic monitor from multiple first oil pressure and first oil temperature data. When the key component is the hydraulic cylinder of the flywheel brake system, the system acquires second oil pressure and second oil temperature data of the hydraulic cylinder at each given time, and determines multiple abnormal operating data corresponding to the hydraulic cylinder from multiple second oil pressure and second oil temperature data. When the key component is the balance bar air supply pipe of the balance system, the system acquires air pressure data of the balance bar air supply pipe at each given time, and determines multiple abnormal operating data corresponding to the balance bar air supply pipe from multiple air pressure data.

[0089] Optionally, when there are multiple key components, the data processing module 401 is specifically used to acquire, when the key components include the bearing of the transmission system, the slider of the motion mechanism, the brake hydraulic monitor and hydraulic cylinder of the flywheel brake system, and the balance bar air supply pipe of the balance system, the axial vibration data, radial vibration data, and bearing temperature data of the bearing at each moment; the displacement data of the slider at each moment; the first oil pressure data and first oil temperature data of the brake hydraulic monitor at each moment; the second oil pressure data and second oil temperature data of the hydraulic cylinder at each moment; and the air pressure data of the balance bar air supply pipe at each moment; and determine multiple abnormal operating data corresponding to the press from multiple axial vibration data, multiple radial vibration data, multiple bearing temperature data, multiple displacement data, multiple first oil pressure data, multiple first oil temperature data, multiple second oil pressure data, multiple second oil temperature data, and multiple air pressure data.

[0090] Optionally, the target degradation prediction module 403 is specifically used to: differentiate the first failure distribution function to obtain the probability density function of the Weibull distribution; construct a maximum likelihood function based on the probability density function; take the logarithm of the maximum likelihood function to obtain the objective function; obtain a first partial derivative function based on the objective function with respect to the shape parameter; obtain a second partial derivative function based on the objective function with respect to the scale parameter; solve for the first maximum likelihood estimate based on the first partial derivative function; and solve for the second maximum likelihood estimate based on the second partial derivative function.

[0091] Optionally, the target degradation prediction module 403 is specifically used to predict the failure degree of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with the second failure distribution function; predict the reliability of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with the reliability function; predict the failure degree of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with the failure degree function; and determine the degree of degradation of the target based on the failure degree, the reliability, and the failure degree.

[0092] Optionally, the failure degree is in the range [0, 0.1], the reliability is in the range [0.9, 1], and the fault degree is in the range [0, 0.1]. In the case of the specified range, the degree of degradation of the target meets the preset requirements; among which, ; This represents the average failure rate over the preset time period; This represents the standard deviation of the failure rate within the preset time period.

[0093] Optionally, the formula for calculating the first failure distribution function is: ;in, This represents the failure time series; This indicates the degree of failure of the target; This indicates the shape parameter; This indicates the scale parameter.

[0094] Optionally, the probability density function can be calculated as follows: ; This indicates the degree of failure. The derivative of the maximum likelihood function is given by the formula: ; This represents the maximum likelihood function; Indicates the total number of times failure occurred; This indicates the first time in the failure time series. The failure time is specified; the formula for calculating the objective function is: +( ; This indicates the result of the derivative. The logarithm of the result.

[0095] Optionally, the target degradation prediction module 403 is specifically used to set the first partial derivative function equal to 0, solve for the first maximum likelihood estimate, and set the second partial derivative function equal to 0, solve for the second maximum likelihood estimate; wherein, the calculation formula for the first partial derivative function is: The formula for calculating the second partial derivative is as follows: .

[0096] Figure 5 This is a schematic diagram of the structure of the electronic device provided in an embodiment of this application. For example... Figure 5As shown, the electronic device may include: a processor 510, a communications interface 520, a memory 530, and a communications bus 540, wherein the processor 510, the communications interface 520, and the memory 530 communicate with each other through the communications bus 540. The processor 510 can call logic instructions in the memory 530 to execute a method for predicting the degradation degree of a press and its key components. This method includes: acquiring multiple abnormal operating data corresponding to the key components of the press within a preset time period, and constructing a failure time series of the target based on the failure times of each of the multiple abnormal operating data; wherein, if the key component is single, the target is the key component; if the key components are multiple, the target is the press; constructing a first failure distribution function of a Weibull distribution based on the failure time series, the shape parameter and the scale parameter of the Weibull distribution, the first failure distribution function being used to predict the failure degree trend of the target; determining a first maximum likelihood estimate corresponding to the shape parameter and a second maximum likelihood estimate corresponding to the scale parameter based on the first failure distribution function; and predicting the degradation degree of the target based on the predicted failure time of the target, the first maximum likelihood estimate, and the second maximum likelihood estimate.

[0097] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, 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.

[0098] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the degradation prediction method for the press and key components provided by the above methods. The method includes: acquiring multiple abnormal operating data corresponding to the key components of the press within a preset time period, and constructing a failure time series of the target based on the failure time of each of the multiple abnormal operating data; wherein, when the key component is a single component, the target is the key component, and when the key components are multiple components, the target is the press; constructing a first failure distribution function of the Weibull distribution based on the failure time series, the shape parameter and the scale parameter of the Weibull distribution, the first failure distribution function being used to predict the failure trend of the target; determining a first maximum likelihood estimate corresponding to the shape parameter and a second maximum likelihood estimate corresponding to the scale parameter based on the first failure distribution function; and predicting the degradation degree of the target based on the predicted failure time of the target, the first maximum likelihood estimate and the second maximum likelihood estimate.

[0099] In another aspect, embodiments of this application also provide a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a method for predicting the degradation degree of a press and its key components provided by the methods described above. This method includes: acquiring multiple abnormal operating data corresponding to key components of the press within a preset time period, and constructing a failure time series of a target based on the failure times of each of the multiple abnormal operating data; wherein, when the key component is single, the target is the key component, and when the key components are multiple, the target is the press; constructing a first failure distribution function of a Weibull distribution based on the failure time series, the shape parameter and the scale parameter of the Weibull distribution, the first failure distribution function being used to predict the failure degree trend of the target; determining a first maximum likelihood estimate corresponding to the shape parameter and a second maximum likelihood estimate corresponding to the scale parameter based on the first failure distribution function; and predicting the degradation degree of the target based on the predicted failure time of the target, the first maximum likelihood estimate, and the second maximum likelihood estimate.

[0100] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0101] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 this application.

Claims

1. A method of predicting the degree of deterioration of a press and key components, characterized by, include: Within a preset time period, multiple abnormal operating data corresponding to the key components of the press are acquired, and a failure time sequence of the target is constructed based on the failure time of each of the multiple abnormal operating data; wherein, when the key component is a single component, the target is the key component, and when the key components are multiple components, the target is the press. Based on the failure time series, the shape parameter and scale parameter of the Weibull distribution, a first failure distribution function of the Weibull distribution is constructed. The first failure distribution function is used to predict the failure trend of the target. Based on the first failure distribution function, determine the first maximum likelihood estimate corresponding to the shape parameter and the second maximum likelihood estimate corresponding to the scale parameter; The degree of degradation of the target is predicted based on the target's predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate.

2. The press machine and key component deterioration degree prediction method according to Claim 1, characterized by, When the critical component is a single entity, acquiring multiple abnormal operating data corresponding to the critical component of the press includes: When the key component is a bearing in the transmission system, axial vibration data, radial vibration data, and bearing temperature data of the bearing are acquired at various times, and multiple abnormal operating data corresponding to the bearing are determined from multiple axial vibration data, multiple radial vibration data, and multiple bearing temperature data. In the case where the key component is a slider of a motion mechanism, the displacement data of the slider at each time moment is obtained, and multiple abnormal operation data corresponding to the slider are determined from multiple displacement data. In the case where the key component is the brake hydraulic monitor of the flywheel brake system, the first oil pressure data and the first oil temperature data of the brake hydraulic monitor at each time point are obtained, and multiple abnormal operating data corresponding to the brake hydraulic monitor are determined from multiple first oil pressure data and multiple first oil temperature data. When the key component is the hydraulic cylinder of the flywheel brake system, the second oil pressure data and the second oil temperature data of the hydraulic cylinder at each time are obtained, and multiple abnormal operation data corresponding to the hydraulic cylinder are determined from multiple second oil pressure data and multiple second oil temperature data. In the case where the key component is the balance bar air supply pipe of the balance system, the air pressure data of the balance bar air supply pipe at each time point is obtained, and multiple abnormal operation data corresponding to the balance bar air supply pipe are determined from the multiple air pressure data.

3. The press machine and key component deterioration degree prediction method according to Claim 1, characterized by, When there are multiple key components, the acquisition of multiple abnormal operating data corresponding to the key components of the press includes: With the key components including the bearings of the transmission system, the slider of the motion mechanism, the brake hydraulic monitor and hydraulic cylinder of the flywheel brake system, and the balance bar air supply pipe of the balance system, the axial vibration data, radial vibration data and bearing temperature data of the bearing at each moment, the displacement data of the slider at each moment, the first oil pressure data and the first oil temperature data of the brake hydraulic monitor at each moment, the second oil pressure data and the second oil temperature data of the hydraulic cylinder at each moment, and the air pressure data of the balance bar air supply pipe at each moment are obtained. From multiple axial vibration data, multiple radial vibration data, multiple bearing temperature data, multiple displacement data, multiple first oil pressure data, multiple first oil temperature data, multiple second oil pressure data, multiple second oil temperature data, and multiple air pressure data, the corresponding abnormal operating data of the press are determined.

4. The press machine and key component deterioration degree prediction method according to any one of claims 1 to 3, characterized by, The step of determining the first maximum likelihood estimate corresponding to the shape parameter and the second maximum likelihood estimate corresponding to the scale parameter based on the first failure distribution function includes: The probability density function of the Weibull distribution is obtained by taking the derivative of the first failure distribution function, and the maximum likelihood function is constructed based on the probability density function. Taking the logarithm of the maximum likelihood function yields the objective function; The first partial derivative function is obtained by taking the partial derivative of the objective function with respect to the shape parameter, and the second partial derivative function is obtained by taking the partial derivative of the objective function with respect to the scale parameter. The first maximum likelihood estimate is obtained by solving the first partial derivative function, and the second maximum likelihood estimate is obtained by solving the second partial derivative function.

5. The press machine and key component deterioration degree prediction method according to any one of claims 1 to 3, characterized by, The step of predicting the degree of degradation of the target based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate includes: Based on the target's predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with the second failure distribution function, the failure degree of the target is predicted. Based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, and combined with the reliability function, the reliability of the target is predicted. Based on the predicted failure time, the first maximum likelihood estimate, and the second maximum likelihood estimate, combined with the failure degree function, the failure degree of the target is predicted. The degree of degradation of the target is determined based on the degree of failure, the degree of reliability, and the degree of failure.

6. The method for predicting the deterioration degree of the press and key components according to claim 5, characterized in that, In a case where the failure degree is located in an interval [0, 0.1], the reliability is located in an interval [0.9, 1], and the fault degree is located in an interval [0, ], the target degradation degree meets a preset requirement. in, ; This represents the average fault rate within the preset time period; This represents the standard deviation of the failure rate within the preset time period.

7. The method for predicting the deterioration degree of the press and key components according to claim 4, characterized in that, The formula for calculating the first failure distribution function is: ; in, This represents the failure time series; This indicates the degree of failure of the target; Indicates the shape parameters; This represents the scale parameter.

8. The method for predicting the deterioration degree of the press and key components according to claim 7, characterized in that, The formula for calculating the probability density function is as follows: ; Indicates the degree of failure The derivative result; The formula for calculating the maximum likelihood function is as follows: ; Denotes the maximum likelihood function; Indicates the total number of times failure occurred; Indicates the first time in the failure time series One failure time; The formula for calculating the objective function is as follows: +( ; The derivative result represents... The logarithm of the result.

9. The method for predicting the deterioration degree of the press and key components according to claim 8, characterized in that, The step of obtaining the first maximum likelihood estimate based on the first partial derivative function and obtaining the second maximum likelihood estimate based on the second partial derivative function includes: Set the first partial derivative function to 0 and solve for the first maximum likelihood estimate; set the second partial derivative function to 0 and solve for the second maximum likelihood estimate. The formula for calculating the first partial derivative function is as follows: ; The formula for calculating the second partial derivative is: .

10. A system for predicting the deterioration degree of a press and its key components, characterized in that, include: The data processing module is used to acquire multiple abnormal operating data corresponding to the key components of the press within a preset time period; A time series construction module is used to construct a failure time series of the target based on the failure times of the multiple abnormal operation data; wherein, when the key component is a single component, the target is the key component, and when the key components are multiple components, the target is the press. The target degradation prediction module is used to construct a first failure distribution function of the Weibull distribution based on the failure time series, the shape parameter and the scale parameter of the Weibull distribution, and the first failure distribution function is used to predict the failure trend of the target; based on the first failure distribution function, determine a first maximum likelihood estimate corresponding to the shape parameter and a second maximum likelihood estimate corresponding to the scale parameter; and predict the degree of degradation of the target based on the target's expected failure time, the first maximum likelihood estimate and the second maximum likelihood estimate.

Images