CN111782706B - Jitter-free real-time rain flow counting method for structural fatigue life analysis - Google Patents

Abstract

The invention provides a jitter-free real-time rain flow counting method for structural fatigue life analysis, which temporarily stores data points which are likely to generate cycle counting jitter in the future and corresponding cycle values thereof by means of a constructed database, and can correct cycle counting results once the data has the jitter phenomenon, so that the jitter phenomenon of cycle counting is avoided; when the method is used for processing structural load data, the first two data points A and B are read firstly, and then according to A < B or A > B, a counting method for data points K which are numbered as odd numbers after the 3 rd data point and data points L which are numbered as even numbers is given according to two conditions. The method can process all load history data including 'divergent convergent waves' in real time on the premise of not adjusting the load history data sequence and not missing any data point; and the cycle counting result is not influenced by future data, so that the jitter phenomenon of cycle counting is avoided, the counting result is more accurate, and the calculation efficiency is high.

Description

Jitter-free real-time rain flow counting method for structural fatigue life analysis

Technical Field

The invention relates to the technical field of structural load history data processing, cycle counting and data storage, in particular to a jitter-free real-time rain flow counting method for structural fatigue life analysis.

Background

Under the repeated loading and unloading action of the structure bearing the cyclic load, even if the internal stress level of the structure is far lower than the allowable stress of the material, when the cyclic number of the loading is enough, the structure still generates fatigue failure. Unlike the strength failure of a structure, which is not predictable when the structure will fail early in its operation, the occurrence of fatigue failure often requires the structure to be in operation for a period of time (even a long time). Furthermore, before the structure fails in fatigue, the material performance is obviously deteriorated, even fatigue cracks are generated, and the stress or strain has strong concentration effect on the local part of the material, so that the fatigue failure is often sudden. The occurrence of fatigue damage to a structure is related to a plurality of factors of the structure in the whole service life, such as thermal load, mechanical load, accidental sudden load and the like, which all lead to the extreme complexity of the fatigue analysis of the structure.

In engineering practice, the fatigue analysis of the structure comprises the following steps: (1) recording the change history of the load of the structure along with the time in the whole service life; (2) converting the load history into a load cycle amplitude value for fatigue analysis and a corresponding cycle number by using a cycle counting method; (3) and obtaining the damage value and the residual fatigue life of the structure through the experimental result of the load amplitude-fatigue failure cycle times of the material.

With the development of communication technology, monitoring data of the structure can be collected in real time, and the counting result of the load process can be processed into the fatigue analysis result of the structure in real time. Therefore, the accuracy and real-time performance of the fatigue analysis method depend on whether the cycle counting method is accurate or not and whether the load history can be processed in real time or not.

In the development history of the fatigue analysis method, the cycle counting method applied in the engineering comprises the following steps: cross counting, peak counting, simple range counting, range pair counting, maximum and minimum counting, rain flow counting, and the like. In the national standard of pressure vessels to be issued in China, the recommended cycle counting method comprises the following steps: maximum-minimum counting and rain flow counting.

For the same load process, different cycle counting methods are adopted, and the obtained counting results are slightly different. Among them, the rain flow counting method was first proposed in 1968 by Matsuishi and Endo, scholars in japan, and the rule of counting was formulated based on the stress-strain behavior of the material, and specifically, the rain flow counting method counts a closed curve in a hysteresis loop formed by the material under cyclic load. As shown in (a) and (B) of fig. 1 and 2, the two basic types of load history data and the stress-strain hysteresis loops corresponding to the two basic types of load history data, the stress-strain σ -epsilon curves corresponding to the structure both form a closed hysteresis loop, the range of the hysteresis loop is the range between the load point B and the load point C, i.e., | B-C |, and the two basic types are recorded as 1 cycle. The counting rule gives physical significance to a rain flow counting method, the counting result of the counting rule obtains better prediction accuracy after being used for fatigue analysis, and the counting rule is gradually popularized in the engineering field and becomes a cycle counting method which is the most widely applied method in the current fatigue analysis.

Rain flow counting methods are also constantly being developed and changed during use. For example, the rain flow counting method counts a half cycle for certain load histories, such as "convergence waves" of increasingly smaller amplitude. In the fatigue test of the material, the whole cycle with loading and unloading paired is taken as a unit, while the half cycle has no clear physical meaning and is only taken as a numerical means to introduce the existing rain flow counting method. In engineering, in order to pair half cycles two by two as much as possible, counting was performed in units of one cycle, and in the last 60 th century, a totally enclosed counting model was developed. The model is characterized in that the load history data needs to be readjusted before counting, so that the load history data starts from the maximum peak value (or the minimum valley value) and ends, and an example of the adjustment is shown in fig. 3. FIG. 3(a) shows a "divergent convergent wave" formed by load history data points A-J, with increasing amplitude followed by decreasing amplitude, finding the maximum point F in the load history data; dividing the load data into two parts at a data point F, wherein the first part is a load point A-F, and the second part is a load point F-J; the second portion of the payload data is moved ahead of the first portion of the payload data and recombined to form a new payload history, i.e., a payload history consisting of the sequence of data points F-G- … -J-a-B- … -F, as shown in fig. 3 (B). For the "divergent convergent wave" shown in fig. 3(c), a minimum point K of load data is newly added as compared with fig. 3 (a); the same adjustment sequence as that of fig. 3(b) is used to divide the load data into two parts at the maximum point F in the load history data, and the two parts are recombined after exchanging the sequence, as shown in fig. 3 (d). By means of the adjustment, the stress-strain curve of the material corresponding to the adjusted load history data is guaranteed to be closed, and the occurrence of half cycle is avoided when counting is further performed by adopting a rain flow counting method. However, the newly added data point K is just in the middle of the unloading stage of J-a, and according to the convention that the load data only includes the peak-to-valley value, the load data point K is ignored, that is, when the fully-closed counting model adjusts the sequence of the load data, the situation that a certain load data point is missed in the cycle counting process occurs. Therefore, the totally enclosed counting model has the following disadvantages: (1) the adjustment before counting needs to ensure that all load history data are obtained before counting, so that the counting method loses real-time property; (2) when adjusting the load data, it sometimes happens that some load points are missed, such as the new load data point K in (c) of fig. 3 is ignored when adjusting the order of the load data, as shown in (d) of fig. 3.

In addition, when the totally enclosed type counting model is used for counting, if the maximum peak value or the minimum valley value in the load history data collected at a certain moment is taken for processing, since the maximum value or the minimum value cannot be guaranteed to be the maximum value or the minimum value in the whole load history, when larger or smaller load data occurs in the loading process, many cyclic records are mistaken, that is, the cyclic counting result is influenced by future data, which is called the jitter phenomenon of cyclic counting for short. Specific examples are as follows:

with a "diverging wave" as shown in fig. 4 (a), the fully enclosed counting model processes the load history data as shown in fig. 4 (a) to obtain a load spectrum as shown in fig. 4 (b). The load spectrum shown in FIG. 4 (B) is counted to obtain 1 cycle each of the amplitude values | A-B |, | C-D |, and | E-F |. After the new data point G, H arrives, the load history data in fig. 4 (a) is updated to the load history data shown in fig. 4 (c), and the load history data in fig. 4 (c) is processed by using the totally enclosed counting model to obtain the load spectrum shown in fig. 4 (d). From (D) of FIG. 4, 1 cycle of amplitude | H-A |, | B-C |, | D-E | and | F-G | can be derived. Comparing fig. 4 (b), it is found that according to the counting rule of the rain flow counting method, not only the number of cycles is increased by one, but also each cycle is different from fig. 4 (b), which causes the cycle recorded before to be wrong, and needs to be counted again. This phenomenon indicates that the cycle cannot be accurately recorded by adjusting the load spectrum according to the local maximum and minimum values, and a jitter phenomenon of cycle counting occurs.

In order to solve the real-time problem, the existing research improves a rain flow counting method, the core idea is to count load course data in two steps, the first step is to count data which can count cycles in the traditional rain flow counting method and mark data points which cannot be recorded as the whole cycles, and the load course formed by the data points forms a 'divergent convergent wave'. The second step is to adjust according to the totally enclosed counting model and count the data separately. Although the processing method avoids the situation that the rain flow counting method records the half cycle and can realize a relatively accurate real-time counting function to a certain extent, the 'divergent convergent wave' in the load process data can not be processed in real time.

In view of the foregoing, there is a need for a real-time rain flow counting method that can overcome the jitter phenomenon in the loop recording.

Disclosure of Invention

Aiming at the problems that the prior art cannot realize real-time performance and jitter occurs during cycle counting, the invention provides a jitter-free real-time rain flow counting method for structural fatigue life analysis.

In order to achieve the purpose, the invention adopts the following technical scheme:

the invention provides a jitter-free real-time rain flow counting method for structural fatigue life analysis, which is characterized by comprising the following steps of:

1) constructing a database for storing structural load data, and initializing the database and a cycle counting result to be both empty sets; respectively defining the first two structural load data in the existing structural load data as a data point A and a data point B, and storing the two data points into a database; respectively defining the last but one and last structural load data in the existing structural load data as a data point I and a data point J, and adding a cycle I-J I in a cycle counting result; the following judgment is made: if A < B is met, entering the step 2), wherein the data point I is a minimum value, and the data point J is a maximum value; if A is larger than B, entering step 5), wherein the data point I is a maximum value, and the data point J is a minimum value;

2) reading the next minimum data point and defining the next minimum data point as a data point K; data point K was judged as follows:

if the data point K is larger than the last minimum value I in the current database or no earlier minimum value exists in the current database, storing the data point K into the database, and executing the step 3);

if the data point K < I is met, deleting the data points I and J in the current database, re-defining the last but one structural load data and the last structural load data existing in the current database as the data point I and the data point J respectively, when no structural load data exists in the database after the data point I and the data point J are deleted, further judging whether the data point K > I is established, if not, continuously deleting the re-defined data points I and J, continuously repeating the process until the data point K > I is established or no structural load data is recorded in the database, storing the data point K in the database, and executing the step 3);

3) detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result, and finishing the cycle counting process; if new structural load data comes, reading the next maximum value data point, defining the next maximum value data point as L, and judging the data point L as follows:

if the data point L is smaller than the last maximum value J in the current database or no earlier maximum value exists in the database, storing the data point L into the database, adding a cycle | K-L | in a cycle counting result, and executing the step 4);

if L > J is satisfied, deleting the loop | I-J | recorded in the step 1) in the counting result for the case of executing the step 3) for the first time; for the case that step 3) is not performed for the first time, deleting the loop | I-J | recorded when step 3) was performed in the previous loop; deleting the cycle | I-J |, then adding a cycle | J-K |, deleting data points J and K in the database, defining the data point I in the current database as a data point K, and defining the last minimum value and the last maximum value of the data point K as the data points I and J respectively, or if no other structural load data exists in the database before the data point K, temporarily not pointing to any structural load data; continuously judging the data points L according to the steps until the data points L are smaller than the last maximum value or have no earlier maximum value in the current database, storing the data points L into the database, respectively defining the last but one structural load data in the database as a data point I and a data point J, adding a cycle | I-J | in a cycle counting result, and executing the step 4);

4) detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result, and finishing the cycle counting process; if a new data point arrives, returning to the step 2);

5) reading the next maximum data point and defining it as data point K, and judging the data point K as follows:

if the data point K is smaller than the last maximum value I in the current database or no earlier maximum value exists in the current database, storing the data point K into the database, and executing the step 6);

if the data point K > I is met, deleting the data points I and J in the current database, re-defining the last but one structural load data and the last structural load data existing in the current database as the data point I and the data point J respectively, when no structural load data exists in the database after the data points I and J are deleted, the re-defined data point I and the data point J do not point to any data, further judging whether K < I is satisfied, if not, continuously deleting the re-defined data points I and J, continuously repeating the process until the data point K < I is satisfied or no data is recorded in the current database, storing the data point K in the database, and executing the step 6);

6) detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result, and finishing the cycle counting process; if new structural load data comes, reading the next minimum value data point, defining the next minimum value data point as L, and judging the data point L as follows:

if the data point L is larger than the last minimum value J in the current database or no earlier minimum value exists in the current database, storing the data point L into the database, adding a cycle | K-L | in a cycle counting result, and executing the step 7);

if L < J is satisfied, deleting the loop | I-J | recorded in the step 1) in the counting result for the case of executing the step 6) for the first time; for the case that step 6) is not performed for the first time, deleting the loop | I-J | recorded when step 6) was performed in the previous loop; deleting the cycle | I-J |, then adding a cycle | J-K |, deleting data points J and K in the database, defining the data point I in the database as a data point K, and respectively defining the last maximum value and the last minimum value of the data point K as the data points I and J, or if no other structural load data exists in the database before the data point K, temporarily not pointing to any structural load data; continuously judging the data point L according to the steps until the data point L is larger than the last minimum value in the current database or no earlier minimum value is met, storing the data point L into the database, respectively defining the last but one and the last structural load data in the database as data points I and J, adding a cycle | I-J | in a cycle counting result, and executing the step 7);

7) detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result, and finishing the cycle counting process; and returning to the step 5) if new structural load data arrives.

The invention has the following characteristics and beneficial effects:

1. the method can realize real-time counting processing on all data points of the load process data collected in real time;

2. when the method is used for carrying out cycle counting, the counting is carried out by taking the whole cycle as a unit, no half cycle exists in a counting result, the counting result corresponds to a closed hysteresis loop of a stress-strain curve, and the method has clear physical significance;

3. by adopting the method for counting, the sequence of the load process data does not need to be adjusted, and the counting process is ensured to accord with the real loading and unloading history of the structure;

4. by adopting the method for counting, any load data point cannot be missed, and the counting result is more accurate; meanwhile, the method can complete the counting function under the condition of not interrupting the load process data, thereby avoiding the error caused by interrupting the load data for counting;

5. the method temporarily stores data points which are likely to generate cycle counting jitter in the future and cycle values corresponding to the data points by means of the constructed database, can correct cycle counting results once the data have the jitter, avoids the cycle counting jitter, specifically, temporarily records the data points into the database when the structural load data meet the convergence wave with smaller and smaller amplitude, and deletes the structural load data from the database if the structural load data recorded in the database are judged not to generate the cycle counting jitter. Each data point is corrected at most once, the speed of the increase of the total processing time is only linear along with the increase of the data volume, the occupation of a program CPU and the access pressure of a disk I/O are greatly reduced, and the operating efficiency of a fatigue analysis program can be obviously improved;

6. by adopting the rain flow counting method provided by the invention, a real-time and accurate fatigue analysis method is convenient to customize, and the operation efficiency and the human-computer interaction experience of a fatigue program can be greatly improved.

Drawings

Fig. 1 (a) and (b) show a first type of load data types and corresponding stress-strain hysteresis loops of a full cycle in a conventional rain flow counting method, respectively.

Fig. 2 (a) and (b) show a second type of load data of a full cycle and a stress-strain hysteresis loop corresponding thereto in a conventional rain flow counting method, respectively.

FIGS. 3(a) and (b) are schematic diagrams of load data satisfying the "divergent convergent waves" characteristic formed by load data points A-J before and after adjustment by using a conventional totally enclosed counting model, respectively; fig. 3(c) and (d) are schematic diagrams of the load data satisfying the "divergent convergent wave" characteristic formed by the load data points a to K before and after adjustment by using the conventional totally enclosed counting model, respectively.

FIGS. 4 (a) and (b) are schematic diagrams of load data satisfying the "divergent wave" characteristic formed by load data points A-F before and after adjustment by using a conventional totally enclosed counting model, respectively; fig. 4 (c) and (d) are schematic diagrams of load data satisfying the "divergent wave" characteristic formed by the load data points a to H before and after adjustment by using the conventional totally enclosed counting model, respectively.

Fig. 5 is a flow chart of a jitter-free real-time rain flow counting method for structural fatigue life analysis according to the present invention.

FIG. 6 is a graph illustrating the classification of the convergent wave when data points A < B are present in the method of the present invention; fig. 6 (a) and (b) show the influence of different newly added data points K on the load spectrum waveform; fig. 6 (c) and (d) show the influence of different newly added data points L on the load spectrum waveform.

FIG. 7 is a diagram illustrating the classification of the convergent wave when the data points A > B are determined according to the method of the present invention; fig. 7 (a) and (b) show the influence of different newly added data points K on the load spectrum waveform; fig. 7 (c) and (d) show the influence of different newly added data points L on the load spectrum waveform.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the detailed description and specific examples, while indicating the scope of the invention, are intended for purposes of illustration only and are not intended to limit the scope of the invention.

For better understanding of the present invention, an application example of a jitter-free real-time rain flow counting method for structural fatigue life analysis proposed by the present invention is described in detail below.

The invention provides a jitter-free real-time rain flow counting method for structural fatigue life analysis, the whole flow of which is shown in figure 5, when structural load data is processed, the method firstly reads the first two data points A and B, and then according to A < B or A > B, the counting method for data points K with odd number and data points L with even number after the 3 rd data point is provided according to two conditions, and the method specifically comprises the following steps:

1) and constructing a database for storing the convergent waves obtained by counting the structural load data by a rain flow counting method. And (3) initializing the database and the cycle counting result to be empty sets, respectively defining the first two pieces of structural load data which are sequenced according to time in the existing structural load data as a data point A and a data point B, and storing the two data points in the database. Defining the penultimate structural load data and the last structural load data in the existing structural load data as a data point I and a data point J respectively, and adding a cycle | I-J | in a cycle counting result. At this time, data point I and data point J in the database correspond to data point a and data point B, respectively. It is determined whether data point a and data point B satisfy a < B. If A < B is satisfied, the step 2) is entered, at which time the data point I is a minimum value and the data point J is a maximum value, and if A > B is satisfied, the step 5) is entered, at which time the data point I is a maximum value and the data point J is a minimum value.

2) The next minimum data point is read and defined as data point K, as required by the structural load data containing only the peak-to-valley values. Data point K was judged as follows: if the data point K is greater than the last minimum value I in the current database or there is no earlier minimum value in the current database, the feature of the convergent wave is still satisfied between the newly read data point K and the structural load data already stored in the database at this time, as shown in fig. 6 (a) (in the data processing process, the structural load data does not necessarily exist before the data point I recorded in the database shown in fig. 6, and the structural load data before the data point I, that is, the structural load data not labeled in fig. 6 is shown in the figure for convenience of understanding, the remainder), the data point K is stored in the database, and step 3 is performed). If the data point K < I, the convergent wave feature formed by the data point K and the structural load data already stored in the current database is destroyed, as shown in (b) of fig. 6, the cycle | I-J | already recorded does not generate the jitter phenomenon of the cycle count, so the data points I and J in the current database are deleted. At this time, newly defining the last but one structural load data and the last structural load data existing in the current database as a data point I and a data point J respectively, continuously judging whether K > I is satisfied, if not, continuously deleting the data points I and J just defined, continuously repeating the process until the data point K > I is satisfied or no structural load data is recorded in the database, storing the data point K in the database, and executing the step 3). It should be noted that data points I and J do not point to any data point for the moment when no structural load data is recorded in the database.

3) And detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result and finishing the cycle counting process, wherein no new data come (except the case that new data exist but do not come yet). And if new structural load data arrives, reading the next maximum value data point according to the requirement that the structural load data only comprises the peak-valley value, and defining the next maximum value data point as L. Data point L was judged as follows: if the data point L is smaller than the previous maximum J in the current database or there is no earlier maximum in the database, the feature of the convergent wave is still satisfied between the newly read data point L and the structural load data stored in the database, as shown in (c) of fig. 6, the data point L is stored in the database, and a cycle | K-L | is added to the cycle count result, and step 4 is executed. If L > J is satisfied, i.e., the data point L is greater than the last maximum value J in the database, then the convergent wave feature formed by the data point L and the structural load data already stored in the database is destroyed, as shown in fig. 6 (d). At this time, according to the counting rule of the conventional rain flow counting method, a cycle | J-K | should be recorded, and the cycle recorded in step 1) of the present invention or step 3) of the previous cycle is | I-J |, and at this time, a jitter phenomenon of the cycle counting result occurs, and the following processing needs to be performed: for the case of performing step 3) for the first time, the loop | I-J | recorded in step 1) needs to be deleted in the counting result; for the case where step 3) is not performed for the first time, loop | I-J | recorded when step 3) was performed in the previous loop needs to be deleted. After the loop | I-J | is deleted, a loop | J-K | is added and the data points J and K are deleted in the database. At this time, the data point I in the current database is defined as the data point K, and the last minimum value and the last maximum value of the data point K are respectively defined as the data points I and J, or if there is no other structural load data in the database before the data point K, the data points I and J temporarily do not point to any structural load data. And continuously judging the data point L according to the steps until the data point L is smaller than the last maximum value or has no earlier maximum value in the current database. At this time, the data point L is stored in the database, the second last and last structural load data in the database at this time are respectively defined as a data point I and a data point J, a loop | I-J | is added to the loop counting result, and step 4) is executed.

4) And detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result and finishing the cycle counting process, wherein no new structural load data arrive (except the situation that new structural load data exist but do not arrive). And if a new data point arrives, returning to the step 2).

5) The next maximum data point is read and defined as data point K, as required by the structural load data containing only the peak-to-valley values. Data point K was judged as follows: if the data point K is smaller than the last maximum value I in the current database or there is no earlier maximum value in the current database, the feature of the convergent wave is still satisfied between the newly read data point K and the stored structural load data in the database, as shown in (a) of fig. 7 (in the data processing process, in the database, the structural load data does not necessarily exist before the data point I recorded in the database shown in fig. 7, and for convenience of understanding, the structural load data before the data point I, that is, the structural load data not labeled in fig. 7 is shown in the figure, the same is used), the data point K is stored in the database, and step 6 is performed). If the data point K > I, the convergent wave feature formed by the data point K and the structural load data already stored in the current database is destroyed, and as shown in (b) in fig. 7, the cycle | I-J | recorded has no jitter phenomenon of the cycle count, so the data points I and J in the current database are deleted. At this time, newly defining the last but one structural load data and the last structural load data existing in the current database as a data point I and a data point J respectively, continuously judging whether K < I is satisfied, if not, continuously deleting the data points I and J just defined, continuously repeating the process until the data point K < I is satisfied or no data is recorded in the current database, storing the data point K in the database, and executing the step 6); it should be noted that data points I and J do not point to any data point for the moment when no structural load data is recorded in the database.

6) And detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result and finishing the cycle counting process, wherein no new structural load data arrive (except the situation that new structural load data exist but do not arrive). And if a new structural load data point arrives, reading the next minimum value data point according to the requirement that the structural load data only comprises the peak-valley value, and defining the next minimum value data point as a data point L. If the data point L is larger than the last minimum value J in the current database or no earlier minimum value exists in the current database, the feature of the convergent wave is still satisfied between the newly read data point L and the structural load data stored in the current database, as shown in (c) in fig. 7, the data point L is stored in the database, and a cycle | K-L | is added to the cycle count result, and step 7 is executed). If L < J is satisfied, i.e., the data point L is smaller than the last minimum value J in the database, the convergent wave characteristic formed by the data point L and the structural load data already stored in the database is destroyed, as shown in fig. 7 (d). At this time, according to the counting rule of the conventional rain flow counting method, a cycle | J-K | should be recorded, and the cycle recorded in step 1) of the present invention or step 6) of the previous cycle is | I-J |, and at this time, a jitter phenomenon of the cycle counting result occurs, and the following processing needs to be performed: for the case of performing step 3) for the first time, the loop | I-J | recorded in step 1) needs to be deleted in the counting result; for the case where step 3) is not performed for the first time, loop | I-J | recorded when step 3) was performed in the previous loop needs to be deleted. After the loop | I-J | is deleted, a loop | J-K | is added and the data points J and K are deleted in the database. At this time, the data point I in the database is defined as the data point K, and the last maximum and minimum values of the data point K are defined as the data points I and J, respectively, or if there is no other structural load data in the database before the new data point K, the data points I and J do not point to any structural load data temporarily. And continuously judging the data point L according to the steps until the data point L is larger than the last minimum value in the current database or no earlier minimum value is met. At this time, the data point L is stored in the database, the last but one structure load data in the database at this time is defined as the data points I and J, respectively, and a loop | I-J | is added to the loop count result, and step 7) is performed.

7) And detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result and finishing the cycle counting process, wherein no new structural load data arrive (except the situation that new structural load data exist but do not arrive). And returning to the step 5) if new structural load data arrives.

The following is a specific example of the process of the invention:

1. the counting process of the method is exemplified in detail by the structural load history data (assuming that new structural load data continuously come in the subsequent process) shown in the table 1;

TABLE 1 example load data

Numbering	Value of load
		1	-12.0
2	7.0
		3	-7.0
4	5.0
		5	-3.0
6	6.0

2. According to the flow chart 5 of the method, the data corresponding to the numbers 1 and 2 are respectively marked as data points A and B, and when A < B is satisfied, the data points A and B are stored in the database for future reference. At this time, the penultimate data point a and the penultimate data point B in the database are defined as I and J, respectively, and a loop is recorded with a magnitude of | I-J | ═ 19. Reading the next structural load data with the number of 3 as a data point K, wherein the K is larger than the last minimum value, and storing the data point K value into a database for future reference; at this time, the database stores 3 data points, namely three points numbered 1,2, and 3, the maximum value point numbered 2 is labeled as data point I, and the minimum value point numbered 3 is labeled as data point J.

3. Continuously reading new structural load data, namely a maximum value, and marking the maximum value as a data point L (numbered 4), wherein the data point L is smaller than the last maximum value, storing the data point L into a database, respectively defining the last to last structural load data in the database as data points I and J, adding a cycle in a cycle counting result, and setting the amplitude as | I-J | ═ 12;

4. at this time, new structural load data still comes, the penultimate data point at this time in the database, namely the data point with the number of 3, is marked as a data point I, and the last data point, namely the data point with the number of 4, is marked as a data point J; continuously reading new structural load data, namely a minimum value, recording the new structural load data as a data point K, namely a data point numbered as 5, wherein the data point K is larger than a last minimum value point, and storing the data point K into a database;

5. continuously reading a new structural load data, namely a maximum value, and marking the maximum value as a data point L, namely a data point with the number of 6; at this time, the data point L is larger than the last maximum value, that is, the data point numbered 4, the cycle count has a jitter phenomenon, a cycle with the amplitude of | I-J | ═ 12 is deleted from the cycle count result, a cycle with the amplitude of | J-K | ═ 8 is added, and at the same time, the data points J and K, that is, two data points numbered 4 and 5, are deleted from the database; at this time, there are 4 data stored in the database, which are respectively 4 points numbered 1,2,3, and 6, the data point numbered 3 is redefined as K, and the data points numbered 1 and 2 are respectively I and J; further judging that the data point L is less than J, storing the data point L into a database at the moment, defining the last but one structural load data in the database as data points I and J respectively, and adding a cycle with the amplitude of | I-J | ═ 9 into the cycle counting result;

6. so far, the data numbered 1-6 in table 1 have all been counted, and if other data still come in real time subsequently, the counting is performed according to the flowchart 5 of the method.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (1)

1. A jitter-free real-time rain flow counting method for structural fatigue life analysis is characterized by comprising the following steps:

1) constructing a database for storing structural load data, and initializing the database and a cycle counting result to be both empty sets; respectively defining the first two structural load data in the existing structural load data as a data point A and a data point B, and storing the two data points into a database; respectively defining the last but one and last structural load data in the existing structural load data as a data point I and a data point J, and adding a cycle I-J I in a cycle counting result; the following judgment is made: if A < B is met, entering the step 2), wherein the data point I is a minimum value, and the data point J is a maximum value; if A is larger than B, entering step 5), wherein the data point I is a maximum value, and the data point J is a minimum value;

2) reading the next minimum data point and defining the next minimum data point as a data point K; data point K was judged as follows:

if the data point K is larger than the last minimum value I in the current database or no earlier minimum value exists in the current database, storing the data point K into the database, and executing the step 3);

if the data point K < I is met, deleting the data points I and J in the current database, re-defining the last but one structural load data and the last structural load data existing in the current database as the data point I and the data point J respectively, when no structural load data exists in the database after the data point I and the data point J are deleted, further judging whether the data point K > I is established, if not, continuously deleting the re-defined data points I and J, continuously repeating the process until the data point K > I is established or no structural load data is recorded in the database, storing the data point K in the database, and executing the step 3);

3) detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result, and finishing the cycle counting process; if new structural load data comes, reading the next maximum value data point, defining the next maximum value data point as L, and judging the data point L as follows:

if the data point L is smaller than the last maximum value J in the current database or no earlier maximum value exists in the database, storing the data point L into the database, adding a cycle | K-L | in a cycle counting result, and executing the step 4);

if L > J is satisfied, deleting the loop | I-J | recorded in the step 1) in the counting result for the case of executing the step 3) for the first time; for the case that step 3) is not performed for the first time, deleting the loop | I-J | recorded when step 3) was performed in the previous loop; deleting the cycle | I-J |, then adding a cycle | J-K |, deleting data points J and K in the database, defining the data point I in the current database as a data point K, and defining the last minimum value and the last maximum value of the data point K as the data points I and J respectively, or if no other structural load data exists in the database before the data point K, temporarily not pointing to any structural load data; continuously judging the data points L according to the steps until the data points L are smaller than the last maximum value or have no earlier maximum value in the current database, storing the data points L into the database, respectively defining the last but one structural load data in the database as a data point I and a data point J, adding a cycle | I-J | in a cycle counting result, and executing the step 4);

4) detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result, and finishing the cycle counting process; if a new data point arrives, returning to the step 2);

5) reading the next maximum data point and defining it as data point K, and judging the data point K as follows:

if the data point K is smaller than the last maximum value I in the current database or no earlier maximum value exists in the current database, storing the data point K into the database, and executing the step 6);

if the data point K > I is met, deleting the data points I and J in the current database, re-defining the last but one structural load data and the last structural load data existing in the current database as the data point I and the data point J respectively, when no structural load data exists in the database after the data points I and J are deleted, the re-defined data point I and the data point J do not point to any data, further judging whether K < I is satisfied, if not, continuously deleting the re-defined data points I and J, continuously repeating the process until the data point K < I is satisfied or no data is recorded in the current database, storing the data point K in the database, and executing the step 6);

6) detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result, and finishing the cycle counting process; if new structural load data comes, reading the next minimum value data point, defining the next minimum value data point as L, and judging the data point L as follows:

if the data point L is larger than the last minimum value J in the current database or no earlier minimum value exists in the current database, storing the data point L into the database, adding a cycle | K-L | in a cycle counting result, and executing the step 7);

if L < J is satisfied, deleting the loop | I-J | recorded in the step 1) in the counting result for the case of executing the step 6) for the first time; for the case that step 6) is not performed for the first time, deleting the loop | I-J | recorded when step 6) was performed in the previous loop; deleting the cycle | I-J |, then adding a cycle | J-K |, deleting data points J and K in the database, defining the data point I in the database as a data point K, and respectively defining the last maximum value and the last minimum value of the data point K as the data points I and J, or if no other structural load data exists in the database before the data point K, temporarily not pointing to any structural load data; continuously judging the data point L according to the steps until the data point L is larger than the last minimum value in the current database or no earlier minimum value is met, storing the data point L into the database, respectively defining the last but one and the last structural load data in the database as data points I and J, adding a cycle | I-J | in a cycle counting result, and executing the step 7);

7) detecting whether new structural load data exist or not, if the structural load data collection process is finished, outputting a cycle counting result, and finishing the cycle counting process; and returning to the step 5) if new structural load data arrives.

Images