Cylinder health degree scoring method and device, equipment and storage medium
By conducting real-time health status assessment and predictive maintenance of the cylinder, the problem of inability to timely detect changes in cylinder performance and potential failures in the prior art is solved, achieving more efficient maintenance and reducing economic losses.
Patent Information
- Application Number
- CN202510329721.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-20
- Publication Date
- 2025-06-06
AI Technical Summary
The existing technology lacks real-time monitoring and quantitative evaluation methods, and cannot promptly detect changes in cylinder performance and potential failures, resulting in increased risk of equipment failure, lack of targeted maintenance plans and economic losses.
By obtaining multiple sets of beat data of the cylinder under stable and healthy conditions, configuring feature data, training scoring models, and then calculating the health scores of the cylinder under real-time operating conditions, real-time and quantitative health status evaluation and predictive maintenance of the cylinder are achieved.
Automatic monitoring and predictive maintenance of cylinder health status is achieved, reducing manpower and material waste, reducing downtime and maintenance costs, and improving economic benefits.
Smart Images
Figure CN120106680A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of cylinder working condition monitoring, and in particular to a cylinder health scoring method, device, equipment and storage medium. Background Art
[0002] As a key actuator in automation equipment, cylinders are widely used in various industrial production equipment. The quality of their working condition directly affects the operating efficiency and product quality of the equipment. At present, the maintenance of cylinders in the industrial field mainly relies on regular inspections and passive repairs and replacements. There is a lack of real-time monitoring mechanisms and quantitative evaluation methods, and it is impossible to detect performance changes and potential failures of cylinders in a timely manner.
[0003] In the photovoltaic industry, string welding machines are key equipment in the production of photovoltaic modules. The health of its components directly affects the stability of the equipment, production efficiency and product quality. The cylinders in the string welding machines undertake important tasks such as precise positioning, traction, cutting, and handling, which are crucial to improving the welding quality and production efficiency of the battery cells.
[0004] However, a health status assessment system for the cylinder of the string welding machine has not yet been established in the prior art, resulting in the following problems: 1) Failure to detect cylinder performance degradation in a timely manner increases the risk of equipment failure; 2) The maintenance plan lacks pertinence, resulting in waste of manpower and material resources; 3) Frequent downtime accidents caused by cylinder failures affect the utilization rate of the production line, thus causing economic losses.
[0005] Although the existing inspection methods can ensure the normal operation of the equipment to a certain extent, the number of cylinders in the string welding machine is huge and distributed in multiple functional areas. There are significant differences in the basic status and usage intensity of cylinders in different areas, resulting in different degrees of wear of each cylinder, which makes the traditional inspection plan both complicated and inefficient.
[0006] The current manual inspection mode often requires shutdown operations, and cannot achieve real-time monitoring and early warning of cylinder status. At the same time, inspection work requires high professional skills and professional qualities of inspectors, and the inspection results are easily affected by subjective factors, resulting in large differences in judgment results between different inspectors, and frequent missed inspections or misjudgments. Existing cylinder health detection methods often rely on experience and intuitive judgment, lack standardized health assessment standards, and are difficult to accurately quantify the degree of cylinder performance degradation.
[0007] At the same time, cylinder maintenance generally adopts the "after-the-fact" maintenance method, which may not only lead to extended unplanned downtime of equipment, but also significantly increase maintenance costs and affect economic benefits. Summary of the invention
[0008] In order to solve at least one technical problem in the prior art, the embodiments of the present invention provide a cylinder health scoring method, device, equipment and storage medium, which can evaluate the health status of the cylinder in real time and quantitatively; when the health score of the cylinder is low, an alarm message can be issued in time, realizing automatic monitoring of the health status of the cylinder and predictive maintenance of the cylinder. In order to achieve the above technical objectives, the technical solution adopted by the embodiments of the present invention is: In a first aspect, an embodiment of the present invention provides a cylinder health scoring method, comprising the following steps: Obtain multiple sets of beat data of the cylinder under stable and healthy working conditions, configure feature data according to the beat data, and train and obtain a scoring model; A plurality of groups of beat data of the cylinder under real-time operating conditions are obtained, characteristic data are configured according to the beat data, and the health score of the cylinder is calculated by the scoring model.
[0009] Furthermore, the method of obtaining multiple groups of beat data of the cylinder under a stable and healthy working condition, configuring feature data according to the beat data, and training and obtaining a scoring model specifically includes the following model training process: Acquire a batch of beat data of the cylinder under a stable and healthy working condition, wherein the batch of beat data includes multiple groups of beat data; Configuring multiple sets of feature data corresponding to the multiple sets of beat data obtained; Configuring a sliding time window size and a sliding step size for performing data analysis on the multiple sets of feature data in the form of a sliding time window; the number of the multiple sets of beat data is greater than the sliding time window size; Sliding the sliding time window according to the sliding step length, performing line segment fitting according to the linear relationship between the feature data and time in each sliding time window, and obtaining fitting line segments corresponding to the feature data in each sliding time window; Calculate the slope of the fitting line segment in each sliding time window and perform ADF test on the characteristic data in each sliding time window; Sort the slopes of the fitting line segments of each sliding time window from small to large; Traverse the slopes of each sliding time window after sorting to determine whether the slope meets the conditions and whether the ADF test meets the requirements; if there is no slope that meets the conditions and the ADF test meets the requirements, re-acquire a batch of beat data of the cylinder under stable and healthy conditions and perform the model training process again; if there is a slope that meets the conditions and the ADF test meets the requirements, calculate the mean and standard deviation of each feature data according to the corresponding sliding time window as the indicator benchmark value; Model storage includes saving the calculated benchmark values of each indicator in the scoring model.
[0010] Optionally, after acquiring a batch of beat data of the cylinder under a stable and healthy working condition and before configuring a plurality of sets of feature data corresponding to the plurality of sets of beat data acquired, the method further includes: Data filtering is performed based on the duration characteristics of the beat data.
[0011] Further, a set of beat data includes the extension time and retraction time of the cylinder within the beat cycle; a corresponding set of characteristic data includes the extension time, retraction time and the extension-retraction time difference of the cylinder within the beat cycle; or, a set of beat data only includes the extension time or retraction time of the cylinder within the beat cycle, then a corresponding set of characteristic data includes the extension time or retraction time of the cylinder within the beat cycle.
[0012] Furthermore, the line segment fitting based on the linear relationship between the feature data and time in the sliding time window adopts a linear regression model or a least squares method for fitting.
[0013] Furthermore, the slope meeting the condition means that the slope is less than a preset slope threshold, and the ADF test meeting the requirement means that the P value in the ADF test is less than a preset P threshold.
[0014] Furthermore, the step of obtaining multiple groups of beat data of the cylinder under real-time operating conditions, configuring feature data according to the beat data, and calculating the health score of the cylinder through the scoring model specifically includes: Acquire multiple groups of beat data of the cylinder under real-time operating conditions; the number of the multiple groups of beat data is greater than or equal to the sliding time window size; According to the obtained multiple groups of beat data of the cylinder under the real-time operating conditions, corresponding multiple groups of feature data are configured; Obtain the benchmark values of each indicator in the scoring model corresponding to the cylinder; Determine the critical value of each indicator according to the benchmark value of each indicator; then perform scoring polynomial fitting according to the benchmark value of the indicator and the critical value of the indicator; Performing data analysis on the multiple sets of feature data in the form of a sliding time window, calculating the mean and standard deviation of each feature data within the sliding time window as an indicator, and obtaining the score of each indicator according to the corresponding scoring polynomial; Configure the weight of each indicator; The health score of the cylinder is calculated based on the score of each indicator and the weight of each indicator.
[0015] In a second aspect, an embodiment of the present invention provides a cylinder health scoring device, comprising: The model training module is used to obtain multiple sets of beat data of the cylinder under stable and healthy working conditions, configure feature data according to the beat data, and train and obtain a scoring model; The model inference module is used to obtain multiple groups of beat data of the cylinder under real-time operating conditions, configure feature data according to the beat data, and calculate the health score of the cylinder through the scoring model.
[0016] In a third aspect, an embodiment of the present invention provides an electronic device, including: a memory storing a computer program; The processor is used to run the computer program, and when the computer program is run, the steps of the cylinder health scoring method as described above are executed.
[0017] In a fourth aspect, an embodiment of the present invention provides a storage medium, in which a computer program is stored, and the computer program is configured to execute the steps of the cylinder health scoring method as described above when running.
[0018] The technical solution provided by the embodiment of the present invention has the following beneficial effects: the cylinder health scoring method proposed in the present application can accurately evaluate the health of different cylinders in the form of quantitative scoring; the method abandons the traditional complex and inefficient inspection work, reduces the waste of manpower and material resources, and reduces the influence of human subjective factors; at the same time, it can realize predictive maintenance of cylinders instead of "after-the-fact" maintenance, which can effectively reduce unplanned downtime and maintenance costs, thereby indirectly or directly improving economic benefits. When a serious abnormality occurs in the cylinder, an alarm message can be issued in time to remind the equipment staff in time, thereby playing an active role in the maintenance, repair and repair process of the machine and effectively avoiding the risk of major failures. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 The figure is a flow chart of a cylinder health scoring method in an embodiment of the present invention.
[0020] Figure 2 4 is a flow chart of a training scoring model in an embodiment of the present invention.
[0021] Figure 3 Schematic diagram of linear fitting of characteristic data in an embodiment of the present invention.
[0022] Figure 4 This is a flow chart of calculating the cylinder health score through a scoring model in an embodiment of the present invention.
[0023] Figure 5 Schematic diagram of scoring polynomial fitting in an embodiment of the present invention.
[0024] Figure 6 Schematic diagram of a cylinder health scoring device in an embodiment of the present invention.
[0025] Figure 7Schematic diagram of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION
[0026] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0027] In the description of the embodiments of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.
[0028] In the description of the embodiments of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, it can also be the internal connection of two components, it can be a wireless connection, or it can be a wired connection. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0029] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.
[0030] like Figure 1 As shown, the embodiment of the present invention proposes a cylinder health scoring method, comprising the following steps: Step S10, obtaining multiple groups of beat data of the cylinder under a stable and healthy working condition, configuring feature data according to the beat data, and training and obtaining a scoring model; Step S20, obtaining multiple groups of beat data of the cylinder under real-time operating conditions, configuring feature data according to the beat data, and calculating the health score of the cylinder through the scoring model.
[0031] The cylinder health scoring method proposed in this application is described in detail below through specific examples.
[0032] like Figure 2As shown, the step S10, obtaining multiple groups of beat data of the cylinder under a stable and healthy working condition, configuring feature data according to the beat data, training and obtaining a scoring model, specifically includes the following model training process: Step S101, obtaining a batch of beat data of the cylinder under a stable and healthy working condition, wherein the batch of beat data includes multiple groups of beat data; The beat data is the beat data of the cylinder in one beat cycle; in one beat cycle, the cylinder completes one extension action and one retraction action; in this embodiment, a set of beat data includes the extension time and retraction time of the cylinder in the beat cycle; The cylinder pushes the piston through the pressure of gas (usually compressed air), generating linear reciprocating motion in the cylinder to complete the mechanical action; the obtained beat data of the cylinder is the time of its linear travel, including the extension time and retraction time of the cylinder; under the stable and healthy working condition of the cylinder, the extension time of each beat cycle is stable and unchanged, and the retraction time is also stable and unchanged. With the passage of time, the cylinder's travel time is gradually prolonged due to various reasons (such as wear and aging, stagnation, air pressure and air tightness, air pipe bending, etc.), which may eventually cause an overtime failure; according to the working mechanism of the cylinder, the evolution of its health from stable operation to overtime failure is a gradual process; therefore, this application scores the health of the cylinder according to its mechanism characteristics; In one embodiment, the beat data of the cylinder can be collected by a data acquisition system and stored on an intermediate server; the data acquisition system mainly includes a PLC and an acquisition sensor; the acquisition frequency can be once every 100ms, and the acquisition protocol can be HostLink; the data acquisition system is not the focus of this application, and its introduction is omitted; the embodiment of this application obtains the collected beat data of the cylinder from the intermediate server; In some other embodiments, due to reasons such as the installation position of the cylinder, a set of beat data may also only include the extension time or retraction time of the cylinder within the beat cycle; in other words, a set of beat data only includes the extension time of the cylinder within the beat cycle or the retraction time of the cylinder within the beat cycle; In order to facilitate better training of the scoring model, multiple sets of beat data of the cylinder under stable and healthy working conditions usually include more than 100 sets of beat data; Optionally, step S102, data filtering is performed according to the duration characteristics of the beat data; In a beat cycle, the extension time or retraction time of various types of cylinders usually ranges from 80ms to 3000ms. A duration threshold is set, such as 5000ms, to filter out beat data that is greater than the duration threshold. Step S103, configuring multiple sets of feature data corresponding to the multiple sets of beat data obtained; In this embodiment, a set of beat data includes the extension time and retraction time of the cylinder in the beat cycle; a corresponding set of characteristic data includes the extension time, retraction time and extension and retraction time difference of the cylinder in the beat cycle; as shown in Table 1;
[0033] In some other embodiments, a set of beat data only includes the extension time or retraction time of the cylinder in the beat cycle, and the corresponding set of characteristic data includes the extension time or retraction time of the cylinder in the beat cycle; the characteristic data does not include the extension-retraction time difference; Each set of beat data corresponds to a set of feature data; Step S104, configuring a sliding time window size and a sliding step size, for performing data analysis on the multiple sets of feature data in the form of a sliding time window; the number of the multiple sets of beat data is greater than the sliding time window size; Suppose the sliding time window size is window size, the sliding step is window step, and the number of the multiple sets of feature data is num sample; the sliding time window size and the sliding step are configured as follows:
[0034] Taking the first case above as an example, when a batch of beat data contains 100 sets of beat data, the corresponding configuration obtains 100 sets of feature data, the size of the sliding time window is set to 50, that is, the sliding time window contains 50 sets of feature data, and the sliding step is set to 1; Step S105, sliding the sliding time window according to the sliding step length, performing line segment fitting according to the linear relationship between the feature data and time in each sliding time window, and obtaining fitting line segments corresponding to the feature data in each sliding time window; According to the working mechanism of the cylinder, for a cylinder under stable and healthy working conditions, the characteristic data (such as extension time and / or retraction time) in the sliding time window has a linear relationship with time t, such as Figure 3 As shown; Figure 3 The vertical axis is the extension time or retraction time of the cylinder, in ms; Specifically, the line segment fitting based on the linear relationship between the feature data and time in the sliding time window adopts a linear regression model or a least squares method for fitting; in this embodiment, a ridge regression model is specifically adopted for fitting, which can prevent overfitting and improve the generalization ability of the model; Step S106, calculating the slope of the fitting line segment in each sliding time window and performing an ADF test on the feature data in each sliding time window; Step S107, sorting the slopes of the fitting line segments of each sliding time window from small to large; Step S108, traverse the slopes of each sliding time window after sorting, and determine whether the slope meets the conditions and whether the ADF test meets the requirements; when there is no slope that meets the conditions and the ADF test meets the requirements, re-acquire a batch of beat data of the cylinder under stable and healthy conditions, and perform the model training process again; when there is a slope that meets the conditions and the ADF test meets the requirements, calculate the mean and standard deviation of each feature data according to the corresponding sliding time window as the indicator benchmark value; In this embodiment, the characteristic data includes three dimensions, namely, the extension time, the retraction time and the extension-retraction time difference, so a total of six index reference values are calculated; in other embodiments, the characteristic data may include only one dimension, namely, the extension time or the retraction time, so two index reference values are calculated; the indexes are shown in Table 2;
[0035] The slope meets the condition when the slope is less than a preset slope threshold, for example, 0.01 to 0.001; The ADF test meets the requirements when the P value in the ADF test is less than a preset P threshold, such as 0.05; Step S109, model storage, includes saving the calculated benchmark values of each indicator in the scoring model.
[0036] The above model training will be triggered by production equipment personnel in actual operation in the production environment; due to the complex and changeable production environment of different projects, and the differences in the basic status and usage intensity of cylinders of different models, machines and regions, each cylinder will be trained independently with its own exclusive scoring model. The model training process is triggered when the cylinder is in a stable and healthy working condition.
[0037] like Figure 4 As shown, the step S20, obtaining multiple groups of beat data of the cylinder under real-time operating conditions, configuring feature data according to the beat data, and calculating the health score of the cylinder through the scoring model, specifically includes: Step S201, obtaining multiple groups of beat data of the cylinder under real-time operating conditions; the number of the multiple groups of beat data is greater than or equal to the sliding time window size; The beat data of the cylinder under real-time operating conditions can be collected by a data collection system and stored on an intermediate server; then the collected beat data is obtained from the intermediate server; in this embodiment, a set of beat data includes the extension time and retraction time of the cylinder in the beat cycle; Step S202, configuring corresponding multiple sets of feature data according to multiple sets of beat data of the cylinder under real-time operating conditions; In this embodiment, a set of beat data includes the extension time and retraction time of the cylinder in the beat cycle; a corresponding set of characteristic data includes the extension time, retraction time and extension and retraction time difference of the cylinder in the beat cycle; Step S203, obtaining the reference value of each indicator in the scoring model corresponding to the cylinder; In this embodiment, a total of six index benchmark values of three-dimensional feature data are obtained; Step S204, determining the critical value of each indicator according to the benchmark value of each indicator; then fitting a scoring polynomial according to the benchmark value of the indicator and the critical value of the indicator; The indicator critical value is configured to be 2 to 3 times the indicator reference value; For example, when the mean indicator benchmark value is less than 1000ms, the mean indicator critical value is configured as 3 times the mean indicator benchmark value; when the mean indicator benchmark value is ≥ 1000ms, the mean indicator critical value is configured as 2 times the mean indicator benchmark value; the standard deviation indicator critical value is configured as 3 times the standard deviation indicator benchmark value; The scoring polynomial can be a linear polynomial or a quadratic polynomial; Figure 5 The result of a first-order polynomial fitting is shown as an example; the horizontal axis is the indicator, and the vertical axis is the indicator score; the indicator score corresponding to the indicator baseline value is 100 points, and the indicator score corresponding to the indicator critical value is 0 points; Step S205, performing data analysis on the multiple sets of feature data in the form of a sliding time window, calculating the mean and standard deviation of each feature data in the sliding time window as an indicator, and obtaining the score of each indicator according to the corresponding scoring polynomial; Step S206, configuring the weight of each indicator; The weight of each indicator can be configured by custom weight, AHP weight method or entropy weight method; Step S207, calculating the health score of the cylinder according to the score of each indicator and the weight of each indicator.
[0038] When the cylinder is in normal working condition, the health score is high; when the cylinder is in abnormal working condition, the health score is low; based on the calculated cylinder health score, predictive maintenance of the cylinder can be achieved; for example, in the actual production environment of a project, the following maintenance strategy can be specified: (1) When the health score is lower than A, the system issues an alarm message; for example, A is 60; (2) When the health score is lower than C for B consecutive times, the system will issue an alarm message; for example, B is 5 and C is 65; (3) The system calculates the health scores for the most recent D times. If E scores are lower than F, the system issues an alarm. For example, D is 20, E is 10, and F is 65.
[0039] The above-mentioned A, B, C, D, E, and F thresholds are calculated through actual working conditions of the project production environment, data analysis, and feedback from production personnel; they will be dynamically adjusted and optimized during the testing phase.
[0040] like Figure 6 As shown, the embodiment of the present invention also provides a cylinder health scoring device, including: The model training module is used to obtain multiple sets of beat data of the cylinder under stable and healthy working conditions, configure feature data according to the beat data, and train and obtain a scoring model; The model inference module is used to obtain multiple groups of beat data of the cylinder under real-time operating conditions, configure feature data according to the beat data, and calculate the health score of the cylinder through the scoring model.
[0041] like Figure 7 As shown, an embodiment of the present invention further proposes an electronic device, comprising: a processor and a memory; the processor and the memory communicate with each other, for example, are connected and communicate with each other through a bus; a computer program is stored in the memory; the processor is used to run the computer program, and when the computer program runs, the steps of the cylinder health scoring method described above are executed; the processor can be a CPU, or other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components and other devices; the memory can include volatile memory, such as random access memory; the memory can also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory can also include a combination of the above-mentioned types of memory.
[0042] An embodiment of the present invention further proposes a storage medium, in which a computer program is stored, and the computer program is configured to execute the steps of the cylinder health scoring method as described above when running; the storage medium includes a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (Flash Memory), a hard disk (HDD) or a solid-state drive (SSD), etc. and any combination thereof.
[0043] Finally, it should be noted that the above specific implementation methods are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.
Claims
1. A cylinder health scoring method, characterized in that: The following steps are involved: Obtain multiple sets of beat data of the cylinder under stable and healthy working conditions, configure feature data according to the beat data, and train and obtain a scoring model; A plurality of groups of beat data of the cylinder under real-time operating conditions are obtained, characteristic data are configured according to the beat data, and the health score of the cylinder is calculated by the scoring model.
2. The cylinder health scoring method according to claim 1, characterized in that: The method of obtaining multiple groups of beat data of the cylinder under stable and healthy working conditions, configuring feature data according to the beat data, and training and obtaining a scoring model specifically includes the following model training process: Acquire a batch of beat data of the cylinder under a stable and healthy working condition, wherein the batch of beat data includes multiple groups of beat data; Configuring multiple sets of feature data corresponding to the multiple sets of beat data obtained; Configuring a sliding time window size and a sliding step size for performing data analysis on the multiple sets of feature data in the form of a sliding time window; the number of the multiple sets of beat data is greater than the sliding time window size; Sliding the sliding time window according to the sliding step length, performing line segment fitting according to the linear relationship between the feature data and time in each sliding time window, and obtaining fitting line segments corresponding to the feature data in each sliding time window; Calculate the slope of the fitting line segment in each sliding time window and perform ADF test on the characteristic data in each sliding time window; Sort the slopes of the fitting line segments of each sliding time window from small to large; Traverse the slopes of each sliding time window after sorting to determine whether the slope meets the conditions and whether the ADF test meets the requirements; if there is no slope that meets the conditions and the ADF test meets the requirements, re-acquire a batch of beat data of the cylinder under stable and healthy conditions and perform the model training process again; if there is a slope that meets the conditions and the ADF test meets the requirements, calculate the mean and standard deviation of each feature data according to the corresponding sliding time window as the indicator benchmark value; Model storage includes saving the calculated benchmark values of each indicator in the scoring model.
3. The cylinder health scoring method according to claim 2, characterized in that: After obtaining a batch of beat data of the cylinder under a stable and healthy working condition, and before configuring a plurality of sets of feature data corresponding to the plurality of sets of beat data obtained, the method further includes: Data filtering is performed based on the duration characteristics of the beat data.
4. The cylinder health scoring method according to claim 2, characterized in that: A set of beat data includes the extension time and retraction time of the cylinder within the beat cycle; a corresponding set of characteristic data includes the extension time, retraction time and extension-retraction time difference of the cylinder within the beat cycle; or, a set of beat data only includes the extension time or retraction time of the cylinder within the beat cycle, then the corresponding set of characteristic data includes the extension time or retraction time of the cylinder within the beat cycle.
5. The cylinder health scoring method according to claim 2, characterized in that: The line segment fitting based on the linear relationship between the characteristic data and time in the sliding time window adopts a linear regression model or a least squares method for fitting.
6. The cylinder health rating method according to claim 2, characterized in that: The slope meets the condition means that the slope is less than a preset slope threshold, and the ADF test meets the requirement means that the P value in the ADF test is less than a preset P threshold.
7. The cylinder health rating method according to any one of claims 2 to 6, characterized in that: The step of obtaining multiple groups of beat data of the cylinder under real-time operating conditions, configuring feature data according to the beat data, and calculating the health score of the cylinder by using the scoring model specifically includes: Acquire multiple groups of beat data of the cylinder under real-time operating conditions; the number of the multiple groups of beat data is greater than or equal to the sliding time window size; According to the obtained multiple groups of beat data of the cylinder under the real-time operating conditions, corresponding multiple groups of feature data are configured; Obtain the benchmark values of each indicator in the scoring model corresponding to the cylinder; Determine the critical value of each indicator according to the benchmark value of each indicator; then perform scoring polynomial fitting according to the benchmark value of the indicator and the critical value of the indicator; Performing data analysis on the multiple sets of feature data in the form of a sliding time window, calculating the mean and standard deviation of each feature data within the sliding time window as an indicator, and obtaining the score of each indicator according to the corresponding scoring polynomial; Configure the weight of each indicator; The health score of the cylinder is calculated based on the score of each indicator and the weight of each indicator.
8. A cylinder health scoring device, characterized in that: include: The model training module is used to obtain multiple sets of beat data of the cylinder under stable and healthy working conditions, configure feature data according to the beat data, and train and obtain a scoring model; The model inference module is used to obtain multiple groups of beat data of the cylinder under real-time operating conditions, configure feature data according to the beat data, and calculate the health score of the cylinder through the scoring model.
9. An electronic device, characterized in that: include: a memory storing a computer program; A processor is used to run the computer program, and when the computer program is run, the steps of the cylinder health scoring method according to any one of claims 1 to 7 are executed.
10. A storage medium, characterized in that: The storage medium stores a computer program, and the computer program is configured to execute the steps of the cylinder health scoring method according to any one of claims 1 to 7 when running.