A cylinder degradation analysis method and system, device, storage medium

Abstract

The application provides a cylinder degradation analysis method and system, equipment and a storage medium. The cylinder degradation analysis method comprises the following steps: obtaining cylinder score data within a preset time range; predicting a cylinder score at a future time point according to the cylinder score data and a score prediction model; and determining a degradation trend of the cylinder according to the predicted cylinder score. The cylinder score is obtained, the cylinder score at the future time point is predicted in the cylinder score prediction model, the predicted cylinder score is analyzed according to the judgment rule set in the cylinder score prediction model, the degradation trend of the cylinder is determined, a plurality of cylinders can be compared horizontally and side by side, the degradation trend of one cylinder can be observed according to the time dimension, whether the plurality of cylinders appear the degradation trend at the same time can be determined, and sufficient time is provided for maintenance preparation, the influence on production work is reduced as much as possible, and normal operation work on the production line is ensured.

Description

Technical Field

[0001] This invention belongs to the technical field of cylinder testing, specifically relating to a cylinder deterioration analysis method, system, equipment, and storage medium. Background Technology

[0002] On automated production lines, there are a wide variety of cylinders used and they are widely distributed. Cylinder equipment maintenance is generally done by providing real-time warnings to equipment managers based on the threshold of the equipment's operating time. This method can only provide real-time alarms and cannot predict the equipment's status. It is impossible to know from a large amount of historical data whether the equipment is in a deteriorating or good state, nor can it know the possible state of the equipment in the next week or at a specific time in the future. Therefore, it is impossible to perform maintenance on the equipment in advance. Summary of the Invention

[0003] To overcome the shortcomings of the prior art, the present invention provides a cylinder degradation analysis method, system, equipment, and storage medium to solve the problem that the prior art cylinder equipment is only maintained after a fault is discovered, and cannot predict the degradation trend of the cylinder equipment or make maintenance arrangements in advance.

[0004] One embodiment of the present invention provides a cylinder degradation analysis method, comprising: acquiring cylinder score data within a preset time range;

[0005] Based on the cylinder score data and the score prediction model, predict the cylinder score at a future point in time.

[0006] Based on the predicted cylinder score, the cylinder degradation trend is determined.

[0007] In one embodiment, the score prediction model includes a first prediction algorithm, which includes the following formula:

[0008] L t =α(y t -S t-s )+(1-α)(L t-1 +b t-1 );

[0009] b t =β(L t -L t-1 )+(1-β)b t-1 ;

[0010] S t =Υ(y t -L t )+(1-Υ)S t-s ;

[0011] F t+k =L t+kb t +S t+k-s ;

[0012] Where α is the first smoothing coefficient, β is the second smoothing coefficient, and Υ is the third smoothing coefficient, and the values ​​of α, β, and Υ are between 0 and 1.

[0013] L t Let b be the smoothed cylinder score at time t. t S represents the trend value of the cylinder score at time t. t Let y be the seasonal periodic value of the cylinder score at time t. t Let be the cylinder score at time t;

[0014] s is the length of the seasonal cycle, k is the length of the predicted future time, and F t+k This represents the predicted cylinder score at time (t+k).

[0015] In one embodiment, the first smoothing coefficient, the second smoothing coefficient, and the third smoothing coefficient have the same value.

[0016] In one embodiment, the method further includes an accuracy determination step for the first prediction algorithm:

[0017] The cylinder score data within a preset time range is divided into a sample dataset and a verification dataset. The sample dataset consists of historical data for a first time period within the preset time range, and the verification dataset consists of historical data for a second time period within the preset time range, which is located after the first time period.

[0018] The values ​​of the first smoothing coefficient, the second smoothing coefficient, and the third smoothing coefficient are determined using the sample dataset;

[0019] Based on the determined values ​​of the first smoothing coefficient, the second smoothing coefficient, and the third smoothing coefficient, the cylinder score for the second time period is predicted using the first prediction algorithm to obtain the predicted cylinder score for the second time period.

[0020] The predicted cylinder score for the second time period is compared with the cylinder score data in the verification data for the second time period to calculate the prediction deviation value of the first prediction algorithm.

[0021] In one embodiment, the sample dataset is the first X% of historical data within the preset time range, and the verification dataset is the last 1-X% of historical data within the preset time range.

[0022] In one embodiment, the score prediction model further includes a second prediction algorithm. When the prediction deviation value of the first prediction algorithm is less than or equal to a preset deviation value, the first prediction algorithm is used to predict the cylinder score at a future time point. When the prediction deviation value of the second prediction algorithm is greater than the preset deviation value, the second algorithm is used to predict the cylinder score at a future time point.

[0023] In one embodiment, the second prediction algorithm includes the following formula:

[0024] Y = bX + a;

[0025] Where b is the slope of the regression equation, a is the intercept of the regression equation, X is the time value at a future point in time, and Y is the predicted cylinder score.

[0026] In one embodiment, the slope and intercept of the regression equation in the second prediction algorithm are calculated as follows:

[0027]

[0028] Where, x i For any time value within a preset time range, y i The cylinder score for any time value within a preset time range. This is the average time value within a preset time range. This is the average value of the cylinder score within a preset time range.

[0029] In one embodiment, the process of obtaining the cylinder score data includes:

[0030] Acquire cylinder action data, which includes cylinder action duration, number of cylinder actions, and cumulative cylinder action duration;

[0031] The cylinder score is calculated based on the cylinder action duration, the number of cylinder actions, and the cumulative cylinder action duration.

[0032] In one embodiment, the cylinder score is calculated according to the following formula:

[0033]

[0034] Among them, C n For cylinder scores, X n The upper limit alarm value for cylinder action duration, Y is the number of cylinder actions, and T is the cylinder action duration. 总 The cumulative working time of the cylinder.

[0035] One embodiment of the present invention also provides a cylinder degradation analysis system, comprising:

[0036] The data acquisition module is used to acquire cylinder score data within a preset time range;

[0037] The prediction module is used to predict the cylinder score at a future point in time based on the cylinder score data and the score prediction model.

[0038] The processing module determines the degradation trend of the cylinder based on the predicted cylinder score.

[0039] One embodiment of the present invention also provides a cylinder deterioration analysis device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described cylinder deterioration analysis method.

[0040] One embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described cylinder degradation analysis method.

[0041] The cylinder degradation analysis method, system, equipment, and storage medium provided in the above embodiments of the present invention have the following beneficial effects:

[0042] A cylinder score is obtained by simply calculating the cylinder's actuation time. Using this score as a target, a cylinder score prediction model predicts the cylinder score at a future point in time. Existing data acquisition devices can be used for simple data collection, eliminating the need for additional equipment and its associated costs. By analyzing the predicted cylinder scores using established judgment rules within the cylinder score prediction model, the cylinder degradation trend is determined: no degradation trend or degradation trend (which can be categorized as rapid or slow degradation). Cylinders without a degradation trend can continue to be used without further maintenance. For cylinders exhibiting a degradation trend, it is determined whether they are in a rapid or slow degradation trend. Cylinders with a rapid degradation trend are prioritized for maintenance, followed by those with a slow degradation trend. This allows for horizontal comparison of multiple cylinders, enabling the prioritization of cylinder maintenance. It can also observe the deterioration trend of one cylinder based on the time dimension, or prioritize the maintenance of the cylinder showing the deterioration trend based on whether multiple cylinders show the deterioration trend at the same time, so that the maintenance work can be carried out in an orderly manner and sufficient time can be prepared in advance to minimize the impact on production and ensure the normal operation of the production line. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0044] Figure 1 Provided for embodiments of the present invention Figure 1 Flowchart of a cylinder degradation analysis method;

[0045] Figure 2 This is a schematic diagram of a typical cylinder.

[0046] Figure 3 This is a schematic diagram of the predicted cylinder score provided in one embodiment of the present invention;

[0047] Figure 4 This is a flowchart illustrating the accuracy determination of a first prediction algorithm provided in one embodiment of the present invention;

[0048] Figure 5 This is a computational logic diagram in a prediction model provided in one embodiment of the present invention;

[0049] Figure 6 A structural diagram of a cylinder degradation analysis provided in another embodiment of the present invention;

[0050] Figure 7 This is a diagram of the internal structure of a computer provided in another embodiment of the present invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0052] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative positional relationship and movement of the components in a specific posture. If the specific posture changes, the directional indications will also change accordingly.

[0053] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the use of "and / or" or "and / or" throughout the text includes three parallel solutions. For example, "A and / or B" includes solution A, solution B, or a solution where both A and B are satisfied simultaneously. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0054] In a production workshop, numerous cylinder devices are distributed along the production line. Data is collected from the production line by a data acquisition device using a PLC (Programmable Logic Controller). The program written for this controller is the PLC control program. Specifically, a PLC controller is a digital electronic system specifically designed for industrial applications. It uses a programmable memory to store instructions for performing logical operations, sequential control, timing, counting, and arithmetic operations, controlling various types of mechanical equipment or production processes through digital or analog inputs and outputs. The system distinguishes between action data and status data in the PLC data according to rules pre-set by the operator and monitors the action data. However, it lacks the ability to analyze whether cylinders will deteriorate at a future point in time based on simple processing of collected cylinder action data, as it lacks data-driven judgment rules. This application, however, can predict the probability of future cylinder failure risks based on simple processing and transformation of cylinder action data and established judgment rules, thereby promptly identifying potential problems and troubleshooting faulty cylinders.

[0055] Please see Figure 1 One embodiment of the present invention provides a cylinder degradation analysis method, comprising:

[0056] Obtain cylinder score data within a preset time range;

[0057] Based on the cylinder score data and the score prediction model, predict the cylinder score at a future point in time.

[0058] Based on the predicted cylinder score, the cylinder degradation trend is determined.

[0059] The process of obtaining the cylinder score data includes: obtaining cylinder action data, which includes cylinder action duration, number of cylinder actions, and cumulative cylinder action duration;

[0060] The cylinder score is calculated based on the cylinder's actuation duration, the number of cylinder actuations, and the cumulative actuation duration. It should be noted that the preset time range can be in units of years, months, and days, including a year, a month, a week, or a day. The time range can be adjusted according to the actual needs.

[0061] Please see Figure 2 This is a schematic diagram of a typical cylinder 300. The cylinder 300 includes a cylinder body 310 and a piston 320 disposed inside the cylinder body 310. The piston 320 drives the piston rod 330 to reciprocate within the cylinder body 310, thereby realizing the function of the cylinder 300. A top dead center sensor 311 and a bottom dead center sensor 312 are disposed on the cylinder body 310. The top dead center sensor 311 is located at the top dead center of the cylinder, i.e., point B. When the piston 320 moves to the top dead center, the top dead center sensor 311 receives the position information of the piston 320 and sends the information that the piston 320 is at the top dead center to the controller. The bottom dead center sensor 312 is located at the bottom dead center of the cylinder, i.e., point A. When the piston 320 moves to the bottom dead center, the bottom dead center sensor 312 receives the position information of the piston 320 and sends the information that the piston 320 is at the bottom dead center to the controller. Therefore, by setting the top dead center sensor 311 and the bottom dead center sensor 312, the movement of the piston 320 in the cylinder 300 can be effectively detected, thereby providing analytical data for the subsequent operation of the cylinder. In this embodiment, the top dead center sensor 311 and the bottom dead center sensor 312 are Hall sensors. A magnet is provided in the piston 320. When the piston 320 moves to the top dead center B or the bottom dead center A of the cylinder, the Hall sensor can detect the position of the piston 320. In this embodiment, the piston 320 of the cylinder 300 can return to its original position under the action of a spring, gravity, or other external forces. It can be known that the actions of cylinder 300 are opening and clamping. That is, the piston 320 moves from point A to point B, which is the opening action. The time from the signal of the lower dead center sensor 312 to the signal of the upper dead center sensor 311 is the opening action duration of cylinder 300. The piston 320 moves from point B to point A, which is the clamping action. The time from the signal of the upper dead center sensor 311 to the signal of the lower dead center sensor 312 is the clamping action duration of cylinder 300.

[0062] The cylinder score is calculated according to the following formula:

[0063]

[0064] Among them, C n For cylinder scores, X n The upper limit alarm value for cylinder action duration, Y is the number of cylinder actions, and T is the cylinder action duration. 总 The cumulative working time of the cylinder.

[0065] In other words, the upper limit alarm value X for cylinder action time n The product of the number of cylinder actuations Y is the maximum actuation time within the normal range, which is the actual cumulative cylinder working time T. 总 Multiply the ratio by 100 to get the cylinder score C. n It should be noted that there can be more than one type of cylinder action; that is, when more than one type of cylinder action occurs, there are multiple upper limit alarm values ​​for the duration of each cylinder action: X1, X2, ..., X. n Similarly, a one-to-one correspondence calculation yields C1, C2, ..., C n The final average value is the cylinder score of cylinder 300.

[0066] The cylinder's operating time is simply calculated to obtain a cylinder score. Using this score as a target, a cylinder score prediction model predicts the cylinder score at a future point in time. This future point in time can be a month, a week, a day, or an hour. By analyzing the predicted cylinder scores using established judgment rules within the cylinder score prediction model, the cylinder's degradation trend is determined. The model categorizes cylinders as either showing no degradation trend or showing a degradation trend (which can be classified as rapid or slow degradation). Cylinders without a degradation trend can continue to be used without further maintenance. Cylinders showing a degradation trend are further classified as either rapidly or slowly deteriorating. Cylinders exhibiting a rapid degradation trend are prioritized for maintenance, followed by those showing a slow degradation trend. For example, one judgment rule in cylinder score prediction models sets normal and warning scores. When the predicted cylinder score is higher than the normal score, it can be confirmed that the cylinder has not shown a deterioration trend; when the predicted cylinder score is lower than the normal score, it can be confirmed that the cylinder has a deterioration trend. In particular, when the predicted cylinder score is between the warning and normal scores, it can be confirmed that the cylinder has a slow deterioration trend; when the predicted cylinder score is lower than the warning score, it can be confirmed that the cylinder has a rapid deterioration trend. Another example is a judgment rule that compares the predicted cylinder score with historical cylinder scores. When the predicted cylinder score is consistently lower than the historical cylinder score, it can be considered that the cylinder has a deterioration trend.

[0067] This allows for horizontal comparison of multiple cylinders, enabling observation of the degradation trend of a single cylinder over time. It also allows for prioritizing maintenance of cylinders showing degradation trends when multiple cylinders exhibit similar trends simultaneously. This ensures orderly maintenance, allows ample time for advance preparation, minimizes disruption to production, and guarantees normal production line operation. It also prevents defective products from being produced due to the degradation of a single cylinder, thus avoiding quality issues, material waste, and reduced production efficiency.

[0068] In one embodiment, the score prediction model includes a first prediction algorithm, which includes the following formula:

[0069] L t =α(y t -S t-s )+(1-α)(L t-1 +b t-1 );

[0070] b t =β(L t -L t-1 )+(1-β)b t-1 ;

[0071] S t =Υ(y t -L t )+(1-Υ)S t-s ;

[0072] F t+k =L t +kb t +S t+k-s ;

[0073] Where α is the first smoothing coefficient, β is the second smoothing coefficient, and Υ is the third smoothing coefficient, and the values ​​of α, β, and Υ are between 0 and 1.

[0074] L t Let b be the smoothed cylinder score at time t. t S represents the trend value of the cylinder score at time t. t The seasonal periodic value of the cylinder score at time t;

[0075] s is the length of the seasonal cycle, k is the length of the predicted future time, and F t+k This represents the predicted cylinder score at time (t+k).

[0076] Among them, L t =α(y t -S t-s )+(1-α)(Lt-1 +b t-1 ) is the level function, which is the weighted average of the seasonally adjusted observed values ​​and the previous period's unseasonal forecast values.

[0077] b t =β(L t -L t-1 )+(1-β)b t-1 The `trend` function calculates the trend of cylinder scores, predicting the trend at a future point in time from a sufficient number of cylinder scores.

[0078] S t =Υ(y t -L t )+(1-Υ)S t-s The eigenvalue function is the moving average of the current seasonality coefficient and the seasonality coefficient of the same period in the previous cycle, incorporating the cyclical changes of the same cylinder action or the same product. Understandably, for simplicity and ease of understanding, especially in the first prediction algorithm, the initial values ​​L0 = y0 and b0 = y1 - y0 are used to predict the cylinder score at a future point in time. The first prediction algorithm uses a multiplicative method for prediction. Depending on the needs, the first prediction algorithm can also use an additive method for prediction. Generally, for time series with relatively fixed seasonal variations, the prediction results obtained using the additive prediction method are more accurate. For time series where seasonal variations are proportional to the current level, the prediction results obtained using the multiplicative prediction method are more accurate.

[0079] In practice, by substituting a sufficient number of cylinder scores into the first prediction algorithm, the level function is used to calculate the predicted smoothing point. This smoothing point is then substituted into the trend function to obtain the predicted smoothing line. Finally, the smoothing line is substituted into the seasonal function, combined with actual periodic changes, to obtain predicted cylinder scores that reflect changes in actual production. A sufficiently large number of cylinder scores, representing at least seven time points, is necessary to calculate the cylinder score for the first future time point.

[0080] Please see Figure 3 For example, if the prediction time is one day, then you need to know the cylinder scores for the previous 7 days to predict the cylinder scores for the 8th day.

[0081] In one embodiment, the first smoothing coefficient α, the second smoothing coefficient β, and the third smoothing coefficient Y have the same value. The values ​​of the first smoothing coefficient α, the second smoothing coefficient β, and the third smoothing coefficient Y range from 0 to 1, and can be calculated using various combinations of values ​​obtainable from 0 to 1. Verification has shown that setting the three smoothing coefficients to the same value allows for a more accurate prediction of the cylinder score at a future time.

[0082] Please see also Figure 4 In one embodiment, the method further includes an accuracy determination step for the first prediction algorithm:

[0083] The cylinder score data within a preset time range is divided into a sample dataset and a validation dataset. The sample dataset consists of historical data from a first time period within the preset time range, and the validation dataset consists of historical data from a second time period within the preset time range, which is after the first time period. Specifically, it is assumed that the first 90% of the historical data within the preset time range is used as the sample dataset, and the last 10% of the historical data within the preset time range is used as the validation dataset. If the entire year of 2021 data is used, then the sample dataset (the first 328 days) is used for prediction, and the validation dataset (the last 37 days) is used to verify the accuracy of the algorithm model. In this case, the first time period is the first 328 days, and the second time period is the last 37 days. The values ​​of the first and second time periods can be determined according to the actual situation of the production line as needed.

[0084] The sample dataset is used to determine the values ​​of the first smoothing coefficient α, the second smoothing coefficient β, and the third smoothing coefficient Y. In one embodiment, the sample dataset can be used to train a model using deep learning methods to derive the values ​​of the first smoothing coefficient α, the second smoothing coefficient β, and the third smoothing coefficient Y.

[0085] Based on the determined values ​​of the first smoothing coefficient α, the second smoothing coefficient β, and the third smoothing coefficient Y, the cylinder score for the second time period is predicted using the first prediction algorithm to obtain the predicted cylinder score for the second time period.

[0086] The predicted cylinder score for the second time period is compared with the cylinder score data in the verification data for the second time period to calculate the prediction deviation value of the first prediction algorithm.

[0087] In other words, to verify the accuracy of the first prediction algorithm, the difference between the predicted cylinder score obtained by the first prediction algorithm and the actual cylinder score is the prediction deviation. The predicted and actual cylinder scores are calculated using the SMAPE (Symmetric Mean Absolute Percentage Error) method. The SMAPE formula is as follows:

[0088]

[0089] Where η is the prediction deviation, R is the predicted cylinder score, and F is the cylinder score. Generally, the calculated prediction deviation should range from 0% to 200%, with a closer value to 0% indicating a more accurate prediction.

[0090] Understandably, the process involves six steps: First, dividing the cylinder score data within a preset time range into a sample dataset and a validation dataset; second, substituting known historical data (i.e., the sample dataset) into the first prediction algorithm to calculate the smoothing coefficient; third, substituting the smoothing coefficient into the first prediction algorithm to refine it; fourth, substituting the cylinder scores from the sample dataset into the refined first prediction algorithm to calculate the predicted cylinder scores; fifth, comparing the predicted cylinder scores with the cylinder scores in the validation data; and sixth, calculating a sufficient number of predicted cylinder scores and comparing them with the validation data to determine the prediction deviation using the SMAPE method.

[0091] In one embodiment, the sample dataset consists of the first X% of historical data within the preset time range, and the verification dataset consists of the last 1-X% of historical data within the preset time range. Here, X ranges from 0 to 100. In this embodiment, the value of X can be determined according to actual needs. For example, X can be 90, in which case the sample dataset consists of the first 90% of historical data within the preset time range, and the verification dataset consists of the last 10% of historical data within the preset time range. As another example, X can be 70, in which case the sample dataset consists of the first 70% of historical data within the preset time range, and the verification dataset consists of the last 30% of historical data within the preset time range.

[0092] Assume that the first 90% of historical data within the preset time range is used as the sample dataset, and the last 10% of historical data within the preset time range is used as the validation dataset. If the entire year of 2021 data is used, then the sample dataset (the first 328 days) is used for prediction, and the validation dataset (the last 37 days) is used to verify the accuracy of the algorithm model. The cylinder scores from the sample dataset are substituted into the first prediction algorithm to calculate the values ​​of the first smoothing coefficient α, the second smoothing coefficient β, and the third smoothing coefficient Y, thus refining the first prediction algorithm. Then, the cylinder scores from the sample dataset are substituted into the refined first prediction algorithm to calculate a set of predicted cylinder scores, which are then compared with the cylinder scores from the validation dataset.

[0093] Please see Figure 5 In one embodiment, the score prediction model further includes a second prediction algorithm. When the prediction deviation value of the first prediction algorithm is less than or equal to a preset deviation value, the first prediction algorithm is used to predict the cylinder score at a future time point. When the prediction deviation value of the second prediction algorithm is greater than the preset deviation value, the second algorithm is used to predict the cylinder score at a future time point.

[0094] It should be noted that when the prediction deviation exceeds the preset deviation value, it indicates that the predicted cylinder score calculated by the first prediction algorithm has too large an error, and the prediction is deemed invalid. The first prediction algorithm is therefore unsuitable, and the second prediction algorithm is used instead. For example, if the preset deviation is 20% and the predicted deviation is 23%, then the first prediction algorithm is invalid, and the second prediction algorithm is used. As another example, if the preset deviation is 20% and the predicted deviation is 10%, then the first prediction algorithm is confirmed to be accurate, and it continues to be used for prediction. In other words, by switching between the first and second prediction algorithms, the prediction results of the score prediction model are made more accurate.

[0095] In one embodiment, the second prediction algorithm includes the following formula:

[0096] Y = bX + a;

[0097] Where b is the slope of the regression equation, a is the intercept of the regression equation, X is the time value at a future point in time, and Y is the predicted cylinder score. After determining the slope parameter b and the intercept parameter a of the regression equation, the cylinder score Y at any future point in time can be calculated using the time value X. In this embodiment, the prediction data sample uses all historical data within the query time range. For example, to calculate the slope and intercept of the regression equation using data from July 2022, the slope parameter b and the intercept parameter a are calculated using the number of days X and the cylinder score Y for each day in July. The calculated slope parameter b and intercept parameter a then provide the specific linear regression equation.

[0098] In one embodiment, the slope and intercept of the regression equation in the second prediction algorithm are calculated as follows:

[0099]

[0100] Where, x i For any time value within a preset time range, y i The cylinder score for any time value within a preset time range. This is the average time value within a preset time range. This is the average value of the cylinder score within a preset time range.

[0101] In this embodiment, any time value x within a preset time range is used. i The cylinder fraction y at any time value within a preset time range i Average time within a preset time range and the average value of the cylinder score within the preset time range To calculate the slope parameter b of the regression equation. After calculating the slope parameter b, then based on the time average within a preset time range. and the average value of cylinder scores within a preset time range To calculate the intercept parameter 'a' of the regression equation. For example, if you need to calculate the slope and intercept of the regression equation using data from July 2022, then:

[0102]

[0103] in,

[0104] Please see Figure 6 A cylinder degradation analysis system 100, comprising:

[0105] The data acquisition module 110 is used to acquire cylinder score data within a preset time range;

[0106] Prediction module 120 is used to predict the cylinder score at a future point in time based on the cylinder score data and the score prediction model.

[0107] The processing module 130 determines the degradation trend of the cylinder based on the predicted cylinder score.

[0108] The acquisition module 110 acquires the cylinder action time and calculates a cylinder score based on the simple calculation of the cylinder action time. Using this cylinder score as a target, the prediction module 120 predicts the cylinder score at a future point in time using a cylinder score prediction model. This future point in time can be the cylinder score for a month, a week, a day, or an hour. The predicted cylinder score is analyzed using certain judgment rules set in the cylinder score prediction model. The processing module 130 determines the cylinder degradation trend based on the predicted cylinder score, classifying it as either no degradation trend or a degradation trend, which can be categorized as rapid degradation or slow degradation. Cylinders without a degradation trend can continue to be used without further maintenance; for cylinders with a degradation trend, it is determined whether they are in a rapid or slow degradation trend. Cylinders with a rapid degradation trend are prioritized for maintenance, followed by cylinders with a slow degradation trend. This allows for horizontal comparison of multiple cylinders, enabling observation of the degradation trend of a single cylinder over time. It also allows for prioritizing maintenance of cylinders showing degradation trends when multiple cylinders exhibit similar trends simultaneously. This ensures orderly maintenance, allows ample time for advance preparation, minimizes disruption to production, and guarantees normal production line operation. It also prevents defective products from being produced due to the degradation of a single cylinder, thus avoiding quality issues, material waste, and reduced production efficiency.

[0109] Please see Figure 7 A cylinder degradation analysis device 200 includes a memory 220 and a processor 210. The memory 220 stores a computer program 240, and the processor 220 executes the computer program 240 to implement the aforementioned cylinder degradation analysis method. In this embodiment, the processor 210, the memory 220, and the computer program 240 transmit data via a data bus 230.

[0110] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described cylinder degradation analysis method.

[0111] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for analyzing cylinder degradation, characterized in that, include: Obtain cylinder score data within a preset time range; Based on the cylinder score data and the score prediction model, predict the cylinder score at a future point in time. Based on the predicted cylinder score, determine the cylinder's degradation trend; The score prediction model includes a first prediction algorithm, which includes the following formula: L t = α(y t -S t-s )+(1-α)(L t-1 +b t-1 ); b t = β(L t -L t-1 )+(1-β)b t-1 ; S t = ϒ(y t -L t )+(1-ϒ)S t-s ; F t+k = L t + kb t + S t+k-s ; Where α is the first smoothing coefficient, β is the second smoothing coefficient, and ϒ is the third smoothing coefficient, and the values ​​of α, β, and ϒ are between 0 and 1; L t Let b be the smoothed cylinder score at time t. t S represents the trend value of the cylinder score at time t. t Let y be the seasonal periodic value of the cylinder score at time t. t Let be the cylinder score at time t; s is the length of the seasonal cycle, k is the length of the predicted future time, and F t+k The cylinder score at time (t+k) is the predicted value. The process of obtaining the cylinder score data includes: Obtain the cylinder's motion data, and calculate the cylinder score based on the cylinder's motion data.

2. The cylinder degradation analysis method as described in claim 1, characterized in that, The first smoothing coefficient, the second smoothing coefficient, and the third smoothing coefficient all have the same value.

3. The cylinder degradation analysis method as described in claim 1, characterized in that, It also includes the accuracy determination step of the first prediction algorithm: The cylinder score data within a preset time range is divided into a sample dataset and a verification dataset. The sample dataset consists of historical data for a first time period within the preset time range, and the verification dataset consists of historical data for a second time period within the preset time range, which is located after the first time period. The values ​​of the first smoothing coefficient, the second smoothing coefficient, and the third smoothing coefficient are determined using the sample dataset; Based on the determined values ​​of the first smoothing coefficient, the second smoothing coefficient, and the third smoothing coefficient, the cylinder score for the second time period is predicted using the first prediction algorithm to obtain the predicted cylinder score for the second time period. The predicted cylinder score for the second time period is compared with the cylinder score data in the verification data for the second time period to calculate the prediction deviation value of the first prediction algorithm.

4. The cylinder degradation analysis method as described in claim 3, characterized in that, The sample dataset consists of the first X% of historical data within the preset time range, and the verification dataset consists of the last 1-X% of historical data within the preset time range, where X ranges from 0 to 100.

5. The cylinder degradation analysis method as described in claim 3, characterized in that, The score prediction model also includes a second prediction algorithm. When the prediction deviation value of the first prediction algorithm is less than or equal to a preset deviation value, the first prediction algorithm is used to predict the cylinder score at a future time point. When the prediction deviation value of the second prediction algorithm is greater than the preset deviation value, the second algorithm is used to predict the cylinder score at a future time point.

6. The cylinder degradation analysis method as described in claim 5, characterized in that, The second prediction algorithm includes the following formula: Y = bX + a; Where b is the slope of the regression equation, a is the intercept of the regression equation, X is the time value at a future point in time, and Y is the predicted cylinder score.

7. The cylinder degradation analysis method as described in claim 6, characterized in that, The calculation methods for the slope and intercept of the regression equation in the second prediction algorithm include: ； ； Where, x i For any time value within a preset time range, y i The cylinder score for any time value within a preset time range. This is the average time value within a preset time range. This is the average value of the cylinder score within a preset time range.

8. The cylinder degradation analysis method according to any one of claims 1-7, characterized in that, The cylinder's motion data includes cylinder motion duration, number of cylinder motions, and cumulative cylinder motion duration; The cylinder score is calculated based on the cylinder action duration, the number of cylinder actions, and the cumulative cylinder action duration.

9. The cylinder degradation analysis method as described in claim 8, characterized in that, The cylinder score is calculated according to the following formula: Among them, C n For cylinder scores, X n The upper limit alarm value for cylinder action duration, Y is the number of cylinder actions, and T is the cylinder action duration. 总 The cumulative working time for cylinder operation.

10. A cylinder degradation analysis system, characterized in that, include: The data acquisition module is used to acquire cylinder score data within a preset time range; The prediction module is used to predict the cylinder score at a future point in time based on the cylinder score data and the score prediction model. The processing module determines the degradation trend of the cylinder based on the predicted cylinder score. The score prediction model includes a first prediction algorithm, which includes the following formula: L t = α(y t -S t-s )+(1-α)(L t-1 +b t-1 ); b t = β(L t -L t-1 )+(1-β)b t-1 ; S t = ϒ(y t -L t )+(1-ϒ)S t-s ; F t+k = L t + kb t + S t+k-s ; Where α is the first smoothing coefficient, β is the second smoothing coefficient, and ϒ is the third smoothing coefficient, and the values ​​of α, β, and ϒ are between 0 and 1; L t Let b be the smoothed cylinder score at time t. t S represents the trend value of the cylinder score at time t. t Let y be the seasonal periodic value of the cylinder score at time t. t Let be the cylinder score at time t; s is the length of the seasonal cycle, k is the length of the predicted future time, and F t+k The cylinder score at time (t+k) is the predicted value. The process of obtaining the cylinder score data includes: Obtain the cylinder's motion data, and calculate the cylinder score based on the cylinder's motion data.

11. A cylinder degradation analysis device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the cylinder deterioration analysis method according to any one of claims 1-7.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the cylinder deterioration analysis method according to any one of claims 1-7.

Images