CN116663306A

CN116663306A - Equipment life prediction method, device, equipment and medium based on curve fitting

Info

Publication number: CN116663306A
Application number: CN202310679202.2A
Authority: CN
Inventors: 黄贵发; 褚智伟; 王娟; 王智
Original assignee: Tangzhi Science & Technology Hunan Development Co ltd; Beijing Tangzhi Science & Technology Development Co ltd
Current assignee: Tangzhi Science & Technology Hunan Development Co ltd; Beijing Tangzhi Science & Technology Development Co ltd
Priority date: 2023-06-09
Filing date: 2023-06-09
Publication date: 2023-08-29

Abstract

The application discloses a method, a device, equipment and a medium for predicting equipment life based on curve fitting, which belong to the technical field of equipment life prediction. The method comprises the steps of constructing a target model by using a curve fitting method, and when predicting the current state index, taking actual operation data of equipment, the current health state of the equipment and a state index reference value of the equipment as inputs to ensure that the predicted current state index refers to the previous state of the equipment, also refers to the actual operation quantity and the current state of the equipment, thereby improving the prediction accuracy of the state index; under the condition that the current state index does not need to be corrected, the target model further refers to the current state index and the current state grade of the equipment to predict and obtain the current residual operation quantity of the equipment, and the obtained residual operation quantity predicted value is more accurate. Correspondingly, the device and the readable storage medium for predicting the service life of the equipment based on curve fitting have the technical effects.

Description

Equipment life prediction method, device, equipment and medium based on curve fitting

Technical Field

The application relates to the technical field of equipment life prediction, in particular to a method, a device, equipment and a medium for predicting equipment life based on curve fitting.

Background

The performance and the remaining service life of the equipment determine whether the whole system can normally and stably work or not, and also determine the service life of the equipment. The residual service life is one of the core bases for evaluating the reliable operation of the equipment, and the equipment failure or the failure can be effectively avoided by evaluating the residual service life of the equipment in the working state in real time, so that the equipment failure can be discovered and maintained as soon as possible. But due to the intricacies of the equipment operating environment, such as: the train axle bears the environmental influence such as external temperature, and also has the influence such as abrasion among the train bearings, so that the failure and fault modes of the equipment are complex and various, the real degradation physical process is difficult to describe, and the accurate life prediction is difficult to carry out.

Therefore, how to improve the accuracy of prediction of the remaining life of the device is a problem that needs to be solved by those skilled in the art.

Disclosure of Invention

In view of the above, the present application aims to provide a method, a device and a medium for predicting the life of a device based on curve fitting, so as to improve the accuracy of predicting the remaining life of the device. The specific scheme is as follows:

In a first aspect, the present application provides a method for predicting equipment lifetime based on curve fitting, comprising:

acquiring a previous state grade, a state index reference value, a current state grade and an actual running amount of the equipment;

calculating the state index reference value, the actual running quantity, the state index interval corresponding to the current state grade and the limit running quantity by using a target model constructed by a curve fitting method to obtain the current state index of the equipment;

and if the current state index is determined to be free from correction according to the current state grade, the previous residual operation quantity predicted value and the limit operation quantity corresponding to the current state grade, calculating a state index section and the limit operation quantity corresponding to the current state grade and the current state index by using the target model to obtain the residual operation quantity predicted value of the equipment.

Optionally, the calculating, by using the target model constructed by using the curve fitting method, the state index reference value, the actual running quantity, and the state index interval and the limit running quantity corresponding to the current state level to obtain the current state index of the device includes:

Calculating the state index reference value, the actual running quantity, the state index interval corresponding to the current state grade and the limit running quantity by using the target model according to a first prediction formula to obtain the current state index; the first prediction formula is:

wherein HI is the current state index; a. b, c, d are parameters of a function y in the target model; y is ^-1 An inverse function representing the function y; HI (high intensity polyethylene) _f A reference value for the state index; HI_min _i and HI_max _i Two end point values of the state index section corresponding to the current state level are obtained; _i the limit operation quantity corresponding to the current state grade is obtained; usem is the actual run-size.

Optionally, the calculating, by using the target model, the state index interval and the limit running amount corresponding to the current state level and the current state index to obtain a predicted value of the current residual running amount of the device includes:

calculating a state index interval and a limit running quantity corresponding to the current state grade by using the target model and the current state index by using a second prediction formula to obtain a predicted value of the current residual running quantity; the second predictive formula is:

Wherein RUL is the predicted value of the current residual operation quantity; a. b, c, d are parameters of a function y in the target model;HI_min _i and HI_max _i Two end point values of the state index section corresponding to the current state level are obtained; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; HI is the current state index.

Optionally, the determining that the current state index does not need correction according to the current state level, the previous residual operation amount predicted value and the limit operation amount corresponding to the current state level includes:

acquiring the predicted value of the previous residual operation amount or calculating a state index obtained through previous prediction to obtain the predicted value of the previous residual operation amount;

and if the fault degree of the current state grade is greater than that of the previous state grade and the predicted value of the previous residual operation quantity is not less than the limit operation quantity corresponding to the current state grade, determining that the current state index does not need to be corrected.

Optionally, the method further comprises:

and if the current state index is determined to be corrected according to the current state grade, the previous residual operation quantity predicted value and the limit operation quantity corresponding to the current state grade, correcting the current state index, and calculating a state index section and the limit operation quantity corresponding to the current state grade and the corrected current state index by utilizing the target model to obtain the current residual operation quantity predicted value.

Optionally, the determining, by the current state level, the previous predicted value of the remaining operation amount, and the limit operation amount corresponding to the current state level, that the current state index needs to be corrected includes:

and if the fault degree of the current state grade is greater than that of the previous state grade and the predicted value of the previous residual operation quantity is smaller than the limit operation quantity corresponding to the current state grade, determining that the current state index needs to be corrected.

Optionally, the correcting the current state index includes:

calculating the limit running quantity corresponding to the current state level, the state index section corresponding to the previous state level, the limit running quantity corresponding to the previous state level and the state index reference value by using the target model according to a first correction formula to obtain a correction quantity; the first correction formula is:

wherein adu is the correction amount; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; a. b, c, d are parameters of a function y in the target model; HI_ _i-1 And HI/u _i-1 Two end point values of the state index section corresponding to the previous state level; max (max) _i-1 For the previous state level pairLimit running amount; HI (high intensity polyethylene) _pre A state index obtained for the previous prediction;

calculating a state index section corresponding to the current state grade, the actual running quantity, the correction quantity and a limit running quantity corresponding to the current state grade by using the target model according to a second correction formula to obtain a corrected current state index; the second correction formula is:

wherein HI ^j The state index is the corrected state index; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; a. b, c, d are parameters of a function y in the target model; y is ^-1 An inverse function representing the function y;

HI_min _i and HI_max _i Two end point values of the state index section corresponding to the current state level are obtained; _i the limit operation quantity corresponding to the current state grade is obtained; usem is the actual running amount; adu is the correction amount.

Optionally, the method further comprises:

obtaining a state index interval, a limit running amount and a running amount interval corresponding to at least one state grade by inquiring preset state grade classification information; the at least one status level includes: normal, sub-healthy, mild fault, moderate fault, and severe fault.

Optionally, the method further comprises:

collecting actual operating data of the device;

and adjusting preset state grade classification information according to the actual operation data.

Optionally, the actual operation data includes: the operation amount of the equipment under each state level;

correspondingly, the adjusting the preset state grade classification information according to the actual operation data comprises the following steps:

and adjusting state index intervals and/or limit operation amounts corresponding to the state levels according to the operation amounts of the equipment under the state levels to obtain the adjusted state level classification information.

Optionally, the method further comprises:

fitting parameters of a function y in the target model according to preset state grade classification information or adjusted state grade classification information.

Optionally, the method further comprises:

if the previous state level is an invalid level or the previous state level is an valid level but different from the current state level, determining the state index reference value as a maximum endpoint value of a state index section corresponding to the current state level; otherwise, the state index reference value is determined to be the state index obtained by the previous prediction.

In a second aspect, the present application provides a device life prediction apparatus based on curve fitting, comprising:

The acquisition module is used for acquiring the previous state grade, the previous residual running quantity predicted value, the state index reference value, the current state grade and the actual running quantity of the equipment;

the first prediction module is used for calculating the state index reference value, the actual running quantity, the state index interval corresponding to the current state grade and the limit running quantity by utilizing a target model constructed by a curve fitting method to obtain the current state index of the equipment;

and the second prediction module is used for calculating a state index interval and a limit running amount corresponding to the current state grade and the current state index by using the target model if the current state index is determined to be not corrected according to the current state grade, the previous residual running amount predicted value and the limit running amount corresponding to the current state grade, so as to obtain the current residual running amount predicted value of the equipment.

Optionally, the first prediction module is specifically configured to:

Wherein HI is the current state index; a. b, c, d are parameters of a function y in the target model; y is ^-1 An inverse function representing the function y; HI (high intensity polyethylene) _f A reference value for the state index; HI_ _i and HI/u _i Two end point values of the state index section corresponding to the current state level are obtained; _i the limit operation quantity corresponding to the current state grade is obtained; usem is the actual run-size.

Optionally, the second prediction module is specifically configured to:

wherein RUL is the predicted value of the current residual operation quantity; a. b, c, d are parameters of a function y in the target model;HI_ _i and HI/u _i Two end point values of the state index section corresponding to the current state level are obtained; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; HI is the current state index.

Optionally, the second prediction module is specifically configured to:

Optionally, the second prediction module is further configured to:

wherein adu is the correction amount; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; a. b, c, d are parameters of a function y in the target model; HI_min _i-1 and HI_max _i-1 Two end point values of the state index section corresponding to the previous state level; max (max) _i-1 The limit operation quantity corresponding to the previous state level is obtained; HI (high intensity polyethylene) _pre A state index obtained for the previous prediction;

HI_min _i and HI_max _i For the current state, etcTwo end point values of the state index interval corresponding to the stage; _i the limit operation quantity corresponding to the current state grade is obtained; usem is the actual running amount; adu is the correction amount.

Optionally, the method further comprises:

the query module is used for obtaining a state index interval, a limit running amount and a running amount interval corresponding to at least one state grade by querying preset state grade classification information; the at least one status level includes: normal, sub-healthy, mild fault, moderate fault, and severe fault.

Optionally, the method further comprises:

the collection module is used for collecting actual operation data of the equipment;

and the state grade adjusting module is used for adjusting preset state grade classification information according to the actual operation data.

Optionally, the actual operation data includes: the operation amount of the equipment under each state level; correspondingly, the adjusting module is specifically configured to: and adjusting state index intervals and/or limit operation amounts corresponding to the state levels according to the operation amounts of the equipment under the state levels to obtain the adjusted state level classification information.

Optionally, the method further comprises:

the model adjustment module is used for fitting parameters of the function y in the target model according to preset state grade classification information or adjusted state grade classification information.

Optionally, the method further comprises:

the value determining module is configured to determine the state index reference value as a maximum endpoint value of a state index section corresponding to the current state level if the previous state level is an invalid level or the previous state level is an valid level but is different from the current state level; otherwise, the state index reference value is determined to be the state index obtained by the previous prediction.

In a third aspect, the present application provides an electronic device, comprising:

a memory for storing a computer program;

a processor for executing the computer program to implement the previously disclosed curve fitting based equipment life prediction method.

In a fourth aspect, the present application provides a readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the previously disclosed curve fitting based device lifetime prediction method.

According to the scheme, the application provides a method for predicting the service life of equipment based on curve fitting, which comprises the following steps: acquiring a previous state grade, a state index reference value, a current state grade and an actual running amount of the equipment; calculating the state index reference value, the actual running quantity, the state index interval corresponding to the current state grade and the limit running quantity by using a target model constructed by a curve fitting method to obtain the current state index of the equipment; and if the current state index is determined to be free from correction according to the current state grade, the previous residual operation quantity predicted value and the limit operation quantity corresponding to the current state grade, calculating a state index section and the limit operation quantity corresponding to the current state grade and the current state index by using the target model to obtain the residual operation quantity predicted value of the equipment.

The method comprises the steps of constructing a target model by using a curve fitting method, wherein the target model takes a state index reference value, an actual running quantity, a state index interval corresponding to the current state grade and a limit running quantity of equipment as inputs, and predicting to obtain the current state index of the equipment; and then determining whether the current state index needs to be corrected or not according to the current state grade, the previous residual operation quantity predicted value and the limit operation quantity corresponding to the current state grade of the equipment, and if not, further enabling the target model to take the state index interval and the limit operation quantity corresponding to the current state grade of the equipment and the current state index of the equipment as inputs to predict and obtain the current residual operation quantity of the equipment. In the scheme, the target model constructs a curve logic relation formed by the equipment operation data, the actual state data and the health state of the equipment, and can predict the trend change of the health state of the equipment according to the equipment operation data, the actual state data and other multi-aspect parameters; under the condition that the current state index does not need to be corrected, the target model further refers to the current state index and the current state grade of the equipment to predict and obtain the current residual operation quantity of the equipment, and the obtained residual operation quantity predicted value is more accurate.

Correspondingly, the device and the readable storage medium for predicting the service life of the equipment based on curve fitting have the technical effects.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present application, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method for predicting equipment life based on curve fitting;

FIG. 2 is a schematic diagram of a state level classification according to the present disclosure;

FIG. 3 is a flow chart of a state index correction method according to the present disclosure;

FIG. 4 is a flowchart of another method for predicting equipment life based on curve fitting according to the present disclosure;

FIG. 5 is a schematic diagram of a device for predicting equipment life based on curve fitting according to the present disclosure;

FIG. 6 is a schematic diagram of an electronic device according to the present disclosure;

FIG. 7 is a diagram illustrating a server configuration according to the present application;

Fig. 8 is a diagram of a terminal structure according to the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

At present, because the working environment of the equipment is complicated, the failure and fault modes of the equipment are complex and various, and the real degradation physical process is difficult to describe, so that the accurate life prediction is difficult to carry out. Therefore, the application provides a device life prediction scheme based on curve fitting, which can construct a curve logic relation formed by device operation data, actual state data and the health state of the device, and predict trend changes of the health state of the device according to various parameters such as the device operation data, the actual state data and the like so as to improve the prediction accuracy of the residual life of the device.

Referring to fig. 1, the embodiment of the application discloses a method for predicting equipment life based on curve fitting, which comprises the following steps:

S101, acquiring a previous state level, a state index reference value, a current state level and an actual running amount of the equipment.

In this embodiment, the device for predicting the remaining life may be a train, an aero-engine, a wind-driven generator, or the like, or may be a certain train bearing on a train, an aero-engine, or a wind-driven generator, such as: and predicting the service lives of the train bearing, the aero-engine bearing and the wind driven generator bearing.

Wherein, the residual life of the same device can be repeatedly predicted according to the embodiment, such as: the prediction method provided by the present embodiment is performed every 24 hours. For equipment in a working state, the health state of the equipment is generally monitored in real time by using a health monitoring system, and corresponding health state grades are output in real time, so that the state grades of the equipment at all times can be obtained from the health monitoring system of the equipment in real time each time when the residual life of the equipment is predicted, for example: the health status level at the current prediction time (i.e., the current status level) and the health status level at the previous prediction time (i.e., the previous status level), and the like.

It should be noted that the state levels may be divided into a plurality of types, and each state level corresponds to a corresponding state index value range (i.e., a state index interval) and a limit running amount. As shown in table 1, the status hierarchies are of 5 types: normal, sub-healthy, mild fault, moderate fault, and severe fault, each corresponding to a state index interval and a limit running volume. The values in table 1 may be set based on experience, or may be set based on actual operating data of the present apparatus or the like. Where the devices are different, the units of the run-size may be different, such as: when the device is a train bearing, the unit of the running amount may be a unit of length, such as: kilometers to describe the operating mileage of the train; when the device is an engine bearing, the unit of the running amount may be a unit of time, such as: hours to describe the length of time the engine is running.

TABLE 1

Status level	State index interval	Limit running quantity
			Normal state	100～81	245
Sub-health	81～40	70
			Minor malfunction	40～30	28
Moderate failure	30～26	25
			Severe failure	26～20	1

If the equipment is brand new equipment or the equipment is subjected to life prediction for the first time after running for a period of time or is subjected to prediction for the first time after equipment maintenance is finished, the equipment is considered to be predicted for the first time. If the current time is the first prediction, the previous state grade is an invalid grade; if the current prediction is not the first prediction, the previous state level is the valid level. Accordingly, if the previous state level is an invalid level or the previous state level is an valid level but different from the current state level, determining the state index reference value as the maximum endpoint value of the state index section corresponding to the current state level; otherwise, the state index reference value is determined as the state index obtained by the previous prediction. If the previous state grade is an invalid grade; or collecting maintenance information existing between the previous time and the current time of the equipment; or the previous state grade is an effective grade but is different from the current state grade, and the actual operation quantity is determined as the operation quantity from the previous time to the current time of the equipment.

S102, calculating a state index reference value, an actual running quantity, a state index section corresponding to the current state grade and a limit running quantity by using a target model constructed by a curve fitting method to obtain the current state index of the equipment.

When the target model is constructed, a function conforming to the degradation rule and mechanism of the current equipment can be selected for curve fitting. Such as: the tan function, sin function or con function is selected as the function y in the object model.

In one embodiment, a method for calculating a state index reference value, an actual operation amount, a state index section corresponding to a current state level, and a limit operation amount by using a target model constructed by a curve fitting method to obtain a current state index of an apparatus includes: and calculating a state index reference value, an actual running quantity, a state index interval corresponding to the current state grade and a limit running quantity by using a target model according to a first prediction formula to obtain the current state index.

The first prediction formula is as follows:wherein HI is the current state index; a. b, c and d are parameters of a function y in the target model; y is ^-1 An inverse function representing the function y; HI (high intensity polyethylene) _f Is a state index reference value; /> HI_min _i And HI_max _i Two end point values of a state index section corresponding to the state level are obtained; _i the limit operation quantity corresponding to the current state grade is obtained; usem is the actual run-size. Wherein t_beg _i And_end _i The two end points of the running quantity interval are calculated by the two end points of the state index interval corresponding to the same state level.

Each state level corresponds to a state index section, a limit operation amount and an operation amount section. The state index section, the limit running amount and the running amount section corresponding to all the state levels can be recorded in the state level classification information, so that in one embodiment, the state index section, the limit running amount and the running amount section corresponding to at least one state level are obtained by inquiring the preset state level classification information; the at least one status level includes: normal, sub-healthy, light fault, medium fault and severe fault, wherein the degree of fault magnitude of each state level is as follows: normal < sub-health < slight fault < moderate fault < severe fault.

Referring to fig. 2, 5 state levels shown in table 1 are shown in a curve form, and then a state index section corresponding to the 5 state levels shown in table 1 can be determined by the Y-axis coordinates shown in fig. 2, and a running amount section can be determined by the X-axis coordinates shown in fig. 2.

And S103, if the state index of the time is determined to be unnecessary to correct through the state grade of the time, the state grade of the last time, the predicted value of the residual operation quantity of the last time and the limit operation quantity corresponding to the state grade of the time, calculating the state index section, the limit operation quantity and the state index corresponding to the state grade of the time by using a target model, and obtaining the predicted value of the residual operation quantity of the equipment.

In one embodiment, calculating a state index section and a limit running amount corresponding to the current state level and the current state index by using a target model to obtain a predicted value of the current residual running amount of the device includes: and calculating a state index section, a limit running quantity and the current state index corresponding to the current state grade by using the target model according to a second prediction formula by using the target model to obtain a predicted value of the current residual running quantity.

The second prediction formula is as follows: wherein RUL is the predicted value of the current residual operation quantity; a. b, c and d are parameters of a function y in the target model;

HI_min _i and HI_max _i Two endpoints of the state index section corresponding to the current state levelA value; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; HI is the current state index.

The predicted value of the previous residual operation amount can be calculated by using a second prediction formula through the previous state index, the residual operation amount obtained through the previous prediction can be stored, and whether the state index is corrected or not is judged, and the stored residual operation amount is directly used for avoiding repeated prediction. In one embodiment, determining that the current state index does not need correction by the current state level, the previous remaining operation amount predicted value, and the limit operation amount corresponding to the current state level includes: and obtaining a predicted value of the previous residual operation quantity or calculating a state index obtained through the previous prediction to obtain the predicted value of the previous residual operation quantity.

If the failure degree of the current state grade is greater than that of the previous state grade and the predicted value of the previous residual operation amount is not less than the limit operation amount corresponding to the current state grade; or the failure degree of the current state grade is not more than that of the previous state grade and the predicted value of the previous residual operation quantity is not less than the limit operation quantity corresponding to the current state grade; or the failure degree of the current state grade is not greater than the failure degree of the previous state grade and the predicted value of the previous residual operation quantity is smaller than the limit operation quantity corresponding to the current state grade, and the current state index is determined to be unnecessary to correct.

Correspondingly, if the state index of the time is determined to be corrected through the state grade of the time, the state grade of the last time, the predicted value of the residual operation quantity of the last time and the limit operation quantity corresponding to the state grade of the time, the state index of the time is corrected, and the state index section and the limit operation quantity corresponding to the state grade of the time and the corrected state index of the time are calculated by utilizing a target model, so that the predicted value of the residual operation quantity of the time is obtained. In one embodiment, determining that the current state index needs to be corrected by the current state level, the previous remaining operation amount predicted value, and the limit operation amount corresponding to the current state level includes: obtaining a predicted value of the previous residual operation quantity or calculating a state index obtained through the previous prediction to obtain the predicted value of the previous residual operation quantity; if the failure degree of the current state grade is larger than that of the previous state grade and the predicted value of the previous residual operation quantity is smaller than the limit operation quantity corresponding to the current state grade, determining that the current state index needs to be corrected.

Referring to fig. 3, the specific steps for correcting the current state index include:

s301, calculating the limit operation amount corresponding to the current state level, the state index section corresponding to the previous state level, the limit operation amount corresponding to the previous state level and the state index reference value by using a first correction formula by using the target model to obtain a correction amount.

S302, calculating a state index section corresponding to the current state grade, an actual operation amount, a correction amount and a limit operation amount corresponding to the current state grade by using a target model according to a second correction formula, and obtaining a corrected current state index.

The first correction formula is as follows: wherein adu is the correction amount; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; a. b, c and d are parameters of a function y in the target model; /> HI_min _i-1 And HI_max _i-1 Two end point values of the state index section corresponding to the previous state level are obtained; max (max) _i-1 The limit operation quantity corresponding to the previous state level is obtained; HI (high intensity polyethylene) _pre The state index obtained for the previous prediction.

Wherein, the second correction formula is: wherein HI ^j The state index is the corrected state index; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; a. b, c and d are parameters of a function y in the target model; y is ^-1 An inverse function representing the function y; /> HI_min _i And HI_max _i Two end point values of a state index section corresponding to the state level are obtained; _i the limit operation quantity corresponding to the current state grade is obtained; usem is the actual run size; adu is the correction amount.

It should be noted that, the present embodiment can predict in a data-model hybrid driving manner, that is: after the target model is built, model parameters or functions can be adjusted in time. In one embodiment, the method further comprises: collecting actual operation data of the equipment; and adjusting preset state grade classification information according to the actual operation data. Wherein, the actual operation data includes: the running amount of the equipment under each state level; correspondingly, the preset state grade classification information is adjusted according to the actual operation data, and the method comprises the following steps: and adjusting state index intervals and/or limit operation amounts corresponding to the state levels according to the operation amounts of the equipment in the state levels to obtain the adjusted state level classification information. The adjusted state level classification information may be used to adjust model parameters.

In one embodiment, the parameters of the function y in the object model are fitted according to preset state-level classification information or adjusted state-level classification information. And fitting parameters of a function y in the target model according to preset state grade classification information, namely, a model building process. Fitting parameters of a function y in the target model according to the adjusted state grade classification information to obtain a model updating process.

As can be seen, in this embodiment, a curve fitting method is used to construct a target model, so that the target model takes a state index reference value, an actual running amount, a state index interval corresponding to a current state level and a limit running amount of the device as inputs, and predicts to obtain a current state index of the device; and then determining whether the current state index needs to be corrected or not according to the current state grade, the previous residual operation quantity predicted value and the limit operation quantity corresponding to the current state grade of the equipment, and if not, further enabling the target model to take the state index interval and the limit operation quantity corresponding to the current state grade of the equipment and the current state index of the equipment as inputs to predict and obtain the current residual operation quantity of the equipment. In the scheme, the target model constructs a curve logic relation formed by the equipment operation data, the actual state data and the health state of the equipment, and can predict the trend change of the health state of the equipment according to the equipment operation data, the actual state data and other multi-aspect parameters; under the condition that the current state index does not need to be corrected, the target model further refers to the current state index and the current state grade of the equipment to predict and obtain the current residual operation quantity of the equipment, and the obtained residual operation quantity predicted value is more accurate.

The following examples describe the model construction and prediction process using the train bearing as an example. And selecting a tan tangent function to predict the residual life of the train bearing for fitting the degradation rule and mechanism of the train bearing.

Referring to fig. 4, the remaining life prediction process of the train bearing includes:

step one: and (5) data acquisition.

The method specifically obtains data such as the operated mileage (accumulated operation mileage) of the train bearing, the application mileage (operation mileage between the current time and the previous time), the limit operation mileage value of each health grade of the train bearing, the current health grade, the previous health index, the maintenance information of the train bearing and the like.

The method comprises the steps of acquiring the operated mileage and application mileage data of a train bearing through a train-mounted monitoring system. And acquiring data such as a limit operation mileage value, a current health grade, a previous health index, train bearing maintenance information and the like of each health grade of the train bearing through a ground monitoring system of the train.

Step two: and (5) calibrating data.

If the failure class (sub-health/mild/moderate/severe) changes to normal or there is train bearing maintenance information for the first prediction or health level, the operated mileage value of the train bearing is reset to be equal to the application mileage value.

Step three: and (5) calculating a health index.

First, a model is constructed based on a curve fitting manner. According to the design expectations: (1) trend of degradation trend of model fitting equipment; (2) The model maps the operated mileage of the train bearing to a health index. General expression for designing health index: y=a×f (b×t+c) +d; a. b, c, d are parameters of a function f, y represents a health index, and t represents data related to the operated mileage of the train bearing.

Since the tan inverse function (arctangent function) is more consistent with the train bearing degradation law and mechanism, then the general expression can be: y=a×arctan (b×t+c) +d. Wherein, parameter a influences the overall amplitude of the health index, parameter b influences the steepness of the large data segment with a large slope in the middle of the health index, parameter c influences the situation that the model deviates from the dependent variable y=0, and parameter d influences the situation that the model deviates from the independent variable t=0.

In order to improve the prediction accuracy, more actual data of the train is involved in the remaining life prediction process, and the embodiment finally transforms y=a×arctan (b×t+c) +d into: it is used as a predictive formula for the health index in the model. In the formula, a is preset to be 37, b is preset to be-3.9 and c is presetD is preset to be 7.2 and is 55, which are all preset experience parameters.

And then calculating the health index of the current state of the train bearing according to the established model.

Wherein t_beg _i And t_end _i The calculation formula is as follows:

HI is the train bearing health index in the current state; the health index in the previous state (if the first prediction or the health grade is changed, the preHI value is the maximum value of the health index in the current health grade); usem is the train bearing application mileage from the previous time to the current time (if the first prediction or the health level is changed from the fault class to normal or maintenance information is provided, usem is the train bearing operated mileage value); max (max) _i For the current health grade i lower limit operation mileage value, HI/u _i The health index is the minimum value of the current health grade i, HI/u _i The health index is the maximum value of the health index under the current health grade i; t_beg _i For the mileage start value (corresponding to the starting abscissa of fig. 2) at the current health level i, t_end _i The mileage end value (corresponding to the end point abscissa of fig. 2) at the current health level i; each number represents a health grade of normal/sub-health/light fault/moderate fault/severe fault, respectively, =1, 2,3,4,5 health grades can be referred to table 1 and fig. 2.

Step four: and (5) calibrating and judging the health index.

Judging whether the health index of the current state of the train bearing needs to be calibrated or not. When the health grade of the train bearing becomes serious, and when the result of the previous life prediction is smaller than the limit operation mileage value of the current health grade, the health index needs to be calibrated, otherwise, the life prediction calculation of the train bearing is directly carried out.

Step five: and (5) calibrating a health index.

The previously calculated health index is calibrated. In particular, healthThe index calibration formula is: wherein adu is the train bearing application mileage increment required to be calibrated, and the calculation method is as follows: />

Step six: and (5) life prediction.

Using formulas in the built model And predicting the remaining mileage of the train bearing. And RUL is the residual operation mileage value of the train bearing in the current state.

Therefore, in the embodiment, the key data such as the health grade and mileage of the train bearing are utilized in a curve fitting mode, meanwhile, the fault development mechanism of the train bearing is fully considered, a physical-data hybrid model is established, accurate residual life prediction is realized, data support is provided for realizing efficient maintenance of mechanical equipment, and safe and efficient operation of the mechanical equipment can be ensured. In the specific implementation process, determining model parameters according to the fault development trend of the train bearing and case data to finish the description of the fault development rule of the train bearing; then a curve fitting mode is adopted to obtain a general function expression of the model; and then, referring to more actual data of the train bearing, constructing an actual health index and residual life prediction calculation formula so as to establish a logical relationship between the parameter value of the model, the health grade of the train bearing and the trend of the model.

Of course, as the train bearing case data is updated continuously, the model parameters and/or health class classifications may be dynamically adjusted to achieve dynamic updating of the thresholds. The process can adjust model parameters from a failure mechanism, so that the model can measure the change rule of gradual performance degradation in the running process of equipment, and the life prediction of the train bearing is accurately realized. The physical-data mixed life prediction model constructed in the way can integrate the fault degradation rule and the fault mechanism information of the mechanical equipment into the model, and compared with the traditional physical-data mixed model, the life prediction model is more flexible and has higher accuracy rate of life prediction results.

In the following, a device life prediction apparatus based on curve fitting provided in the embodiments of the present application is described, and the device life prediction apparatus based on curve fitting described in the following may refer to other embodiments described herein.

Referring to fig. 5, an embodiment of the present application discloses a device for predicting equipment lifetime based on curve fitting, including:

an obtaining module 501, configured to obtain a previous state level, a previous predicted value of a remaining operation amount, a state index reference value, a current state level, and an actual operation amount of the device;

The first prediction module 502 is configured to calculate a state index reference value, an actual operation amount, a state index interval corresponding to the current state level, and a limit operation amount by using a target model constructed by a curve fitting method, so as to obtain a current state index of the device;

and the second prediction module 503 is configured to calculate, if it is determined that the current state index does not need to be corrected according to the current state level, the previous predicted value of the residual operation amount, and the limit operation amount corresponding to the current state level, the state index interval and the limit operation amount corresponding to the current state level, and the current state index by using the target model, so as to obtain the predicted value of the residual operation amount of the device.

In one embodiment, the first prediction module is specifically configured to:

calculating a state index reference value, an actual running quantity, a state index interval corresponding to the current state grade and a limit running quantity by using a target model according to a first prediction formula to obtain a current state index; the first predictive formula is:

wherein HI is the current state index; a. b, c and d are parameters of a function y in the target model; y is ^-1 An inverse function representing the function y; HI (high intensity polyethylene) _f Is a state index reference value; HI_ _i And HI/u _i Two end point values of a state index section corresponding to the state level are obtained; _i the limit operation quantity corresponding to the current state grade is obtained; usem is the actual run-size.

In one embodiment, the second prediction module is specifically configured to:

calculating a state index interval, a limit running quantity and a current state index corresponding to the current state grade by using the target model according to a second prediction formula by using the target model to obtain a predicted value of the current residual running quantity; the second predictive formula is:

wherein RUL is the predicted value of the current residual operation quantity; a. b, c and d are parameters of a function y in the target model;

HI_ _i and HI/u _i Two end point values of a state index section corresponding to the state level are obtained; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; HI is the current state index.

In one embodiment, the second prediction module is specifically configured to:

obtaining a predicted value of the previous residual operation quantity or calculating a state index obtained through the previous prediction to obtain the predicted value of the previous residual operation quantity;

if the failure degree of the current state grade is larger than that of the previous state grade and the predicted value of the previous residual operation amount is not smaller than the limit operation amount corresponding to the current state grade, determining that the current state index does not need to be corrected.

In one embodiment, the second prediction module is further configured to:

if the state index of the time is determined to be corrected through the state grade of the time, the state grade of the previous time, the predicted value of the residual operation quantity of the time and the limit operation quantity corresponding to the state grade of the time, the state index of the time is corrected, and the state index interval and the limit operation quantity corresponding to the state grade of the time and the corrected state index of the time are calculated by utilizing a target model, so that the predicted value of the residual operation quantity of the time is obtained.

In one embodiment, the second prediction module is further configured to:

if the failure degree of the current state grade is larger than that of the previous state grade and the predicted value of the previous residual operation quantity is smaller than the limit operation quantity corresponding to the current state grade, determining that the current state index needs to be corrected.

In one embodiment, the second prediction module is further configured to:

calculating the limit running quantity corresponding to the current state level, the state index section corresponding to the previous state level, the limit running quantity corresponding to the previous state level and the state index reference value by using a first correction formula by using the target model to obtain a correction quantity; the first correction formula is:

Wherein, the liquid crystal display device comprises a liquid crystal display device,adu is the correction amount; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; a. b, c and d are parameters of a function y in the target model; HI_min _i-1 and HI_max _i-1 Two end point values of the state index section corresponding to the previous state level are obtained; max (max) _i-1 The limit operation quantity corresponding to the previous state level is obtained; HI (high intensity polyethylene) _pre A state index obtained for the previous prediction;

calculating a state index section corresponding to the current state grade, an actual running amount, a correction amount and a limit running amount corresponding to the current state grade by using a target model according to a second correction formula to obtain a corrected current state index; the second correction formula is:

wherein HI ^j The state index is the corrected state index; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; a. b, c and d are parameters of a function y in the target model; y is ^-1 An inverse function representing the function y;

HI_min _i and HI_max _i Two end point values of a state index section corresponding to the state level are obtained; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; usem is the actual run size; adu is the correction amount.

In one embodiment, the method further comprises:

In one embodiment, the actual operational data includes: the running amount of the equipment under each state level; correspondingly, the adjusting module is specifically used for: and adjusting state index intervals and/or limit operation amounts corresponding to the state levels according to the operation amounts of the equipment in the state levels to obtain the adjusted state level classification information.

In one embodiment, the method further comprises:

the model adjustment module is used for fitting parameters of a function y in the target model according to preset state grade classification information or adjusted state grade classification information.

In one embodiment, the method further comprises:

the value determining module is used for determining the state index reference value as the maximum endpoint value of the state index section corresponding to the state grade if the previous state grade is an invalid grade or the previous state grade is an effective grade but is different from the current state grade; otherwise, the state index reference value is determined as the state index obtained by the previous prediction.

The more specific working process of each module and unit in this embodiment may refer to the corresponding content disclosed in the foregoing embodiment, and will not be described herein.

It can be seen that the present embodiment provides a device for predicting service life of a device based on curve fitting, which can construct a curve logic relationship formed by device operation data, actual state data and its health state, and predict trend changes of the health state of the device according to multiple parameters such as the device operation data, the actual state data, etc., so as to improve the accuracy of predicting the remaining service life of the device.

An electronic device provided in the embodiments of the present application is described below, and an electronic device described below may refer to other embodiments described herein.

Referring to fig. 6, an embodiment of the present application discloses an electronic device, including:

a memory 601 for storing a computer program;

a processor 602 for executing the computer program to implement the method disclosed in any of the embodiments above.

Further, the embodiment of the application also provides electronic equipment. The electronic device may be the server 50 shown in fig. 7 or the terminal 60 shown in fig. 8. Fig. 7 and 8 are each a block diagram of an electronic device according to an exemplary embodiment, and the contents of the drawings should not be construed as any limitation on the scope of use of the present application.

Fig. 7 is a schematic structural diagram of a server according to an embodiment of the present application. The server 50 may specifically include: at least one processor 51, at least one memory 52, a power supply 53, a communication interface 54, an input output interface 55, and a communication bus 56. Wherein the memory 52 is configured to store a computer program that is loaded and executed by the processor 51 to implement the relevant steps in the monitoring of a publishing application as disclosed in any of the foregoing embodiments.

In this embodiment, the power supply 53 is configured to provide an operating voltage for each hardware device on the server 50; the communication interface 54 can create a data transmission channel between the server 50 and an external device, and the communication protocol to be followed is any communication protocol applicable to the technical solution of the present application, which is not specifically limited herein; the input/output interface 55 is used for acquiring external input data or outputting external output data, and the specific interface type thereof may be selected according to the specific application needs, which is not limited herein.

The memory 52 may be a carrier for storing resources, such as a read-only memory, a random access memory, a magnetic disk, or an optical disk, and the resources stored thereon include an operating system 521, a computer program 522, and data 523, and the storage may be temporary storage or permanent storage.

The operating system 521 is used for managing and controlling various hardware devices on the Server 50 and the computer program 522 to implement the operation and processing of the data 523 in the memory 52 by the processor 51, which may be Windows Server, netware, unix, linux, etc. The computer program 522 may further include a computer program that can be used to perform other specific tasks in addition to the computer program that can be used to perform the monitoring method of the publishing application disclosed in any of the foregoing embodiments. The data 523 may include data such as application program developer information in addition to data such as application program update information.

Fig. 8 is a schematic structural diagram of a terminal according to an embodiment of the present application, and the terminal 60 may specifically include, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, or the like.

Generally, the terminal 60 in this embodiment includes: a processor 61 and a memory 62.

Processor 61 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 61 may be implemented in at least one hardware form of DSP (Digital Signal Processing ), FPGA (Field-Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array ). The processor 61 may also include a main processor, which is a processor for processing data in an awake state, also called a CPU (Central Processing Unit ), and a coprocessor; a coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 61 may integrate a GPU (Graphics Processing Unit, image processor) for rendering and drawing of content required to be displayed by the display screen. In some embodiments, the processor 61 may also include an AI (Artificial Intelligence ) processor for processing computing operations related to machine learning.

Memory 62 may include one or more computer-readable storage media, which may be non-transitory. Memory 62 may also include high-speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In this embodiment, the memory 62 is at least used for storing a computer program 621, where the computer program is loaded and executed by the processor 61, and then can implement relevant steps in the method for monitoring a distribution application executed by the terminal side as disclosed in any of the foregoing embodiments. In addition, the resources stored by the memory 62 may also include an operating system 622, data 623, and the like, and the storage manner may be transient storage or permanent storage. The operating system 622 may include Windows, unix, linux, among others. The data 623 may include, but is not limited to, update information of the application.

In some embodiments, the terminal 60 may further include a display 63, an input-output interface 64, a communication interface 65, a sensor 66, a power supply 67, and a communication bus 68.

Those skilled in the art will appreciate that the structure shown in fig. 8 is not limiting of the terminal 60 and may include more or fewer components than shown.

A readable storage medium provided by embodiments of the present application is described below, and the readable storage medium described below may be referred to with respect to other embodiments described herein.

A readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the curve fitting based equipment life prediction method disclosed in the foregoing embodiments. The readable storage medium is a computer readable storage medium, and can be used as a carrier for storing resources, such as read-only memory, random access memory, magnetic disk or optical disk, wherein the resources stored on the readable storage medium comprise an operating system, a computer program, data and the like, and the storage mode can be transient storage or permanent storage.

In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, so that the same or similar parts between the embodiments are referred to each other.

The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of readable storage medium known in the art.

The principles and embodiments of the present application have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present application and the core ideas thereof; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.

Claims

1. A method for predicting equipment life based on curve fitting, comprising:

2. The method according to claim 1, wherein the calculating, by using the target model constructed by using the curve fitting method, the state index reference value, the actual operation amount, and the state index interval and the limit operation amount corresponding to the current state level to obtain the current state index of the device includes:

wherein HI is the current state index; a. b, c, d are parameters of a function y in the target model; y is ^-1 An inverse function representing the function y; HI (high intensity polyethylene) _f A reference value for the state index; HI_min _i and HI_max _i Two end point values of the state index section corresponding to the current state level are obtained; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; usem is the actual run-size.

3. The method according to claim 1, wherein the calculating, by using the target model, the state index interval and the limit running amount corresponding to the current state level and the current state index to obtain the predicted value of the current residual running amount of the device includes:

wherein RUL is the predicted value of the current residual operation quantity; a. b, c, d are parameters of a function y in the target model;

HI_min _i and HI_max _i Two end point values of the state index section corresponding to the current state level are obtained; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; HI is the current state index.

4. The method according to claim 1, wherein said determining that the current state index does not need correction by the current state level, the previous state level, a previous remaining operation amount predicted value, and a limit operation amount corresponding to the current state level includes:

5. The method as recited in claim 1, further comprising:

6. The method according to claim 5, wherein the determining that the current state index needs to be corrected by the current state level, the previous remaining operation amount predicted value, and the limit operation amount corresponding to the current state level includes:

7. The method of claim 5, wherein said correcting said current state index comprises:

HI_min _i and HI_max _i Two end point values of the state index section corresponding to the current state level are obtained; max (max) _i The limit operation quantity corresponding to the current state grade is obtained; usem is the actual running amount; adu is the correction amount.

8. The method according to any one of claims 1 to 7, further comprising:

9. The method according to any one of claims 1 to 7, further comprising:

collecting actual operating data of the device;

10. The method of claim 9, wherein the actual operating data comprises: the operation amount of the equipment under each state level;

11. The method according to any one of claims 1 to 7, further comprising:

12. The method according to any one of claims 1 to 7, further comprising:

13. A device for predicting equipment life based on curve fitting, comprising:

14. An electronic device, comprising:

a memory for storing a computer program;

a processor for executing the computer program to implement the method of any one of claims 1 to 12.

15. A readable storage medium for storing a computer program, wherein the computer program when executed by a processor implements the method of any one of claims 1 to 12.