An online timing preserving rainflow counting method

Abstract

The application relates to the technical field of data processing, and particularly discloses an online time sequence reserved rain flow counting method, which comprises the following steps: S1, acquiring an original overload sequence; S2, reading four continuous data points in the original overload sequence into a new data set, judging whether a full cycle exists in the new data set, if the full cycle exists, deleting the full cycle from the new data set and recording the full cycle into a cycle result list, and then supplementing two new data points from the original overload sequence into the new data set; if the full cycle does not exist, supplementing one new data point from the original overload sequence into the end of the new data set; S3, taking the last four data points in the new data set as the input of the next round, and repeatedly executing S2 until all data points in the original overload sequence are read and no full cycle exists in the new data set; and S4, arranging all full cycles in the cycle result list of the latest round in time sequence. The application can more efficiently and completely identify and extract the full cycle.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to an online time-preserving rainflow counting method. Background Technology

[0002] In aircraft structural fatigue life assessment, cycle counting of load time histories is one of the core steps connecting load input and damage calculation. By counting cycles, complex random load time histories can be transformed into a series of discrete load cycles, thus providing the basic input for subsequent life estimation based on Miner's linear cumulative damage theory.

[0003] Existing cycle counting methods can be broadly categorized into three types: peak-rate, step-through, and variable-range. Among them, the variable-range counting method, which focuses on the load range and its auxiliary parameters (such as the mean value), can more comprehensively reflect the stress amplitude, average stress effect, and the influence of loading sequence on fatigue damage during load cycles, and is therefore widely used in engineering. The rainflow counting method, as a typical representative of the variable-range method, is based on the stress-strain hysteresis loop of the material. It can effectively identify closed hysteresis loops in the load sequence and provide the stress amplitude and mean value for each cycle, making it the currently recognized fatigue cycle counting method with the most clearly defined physical meaning.

[0004] However, traditional rainflow counting methods still have some shortcomings when dealing with complex load histories: First, some implementations rely on geometric criteria and are not sensitive enough to the identification of microscopic closed loops nested in macroscopic loads; second, single-round counting often fails to completely extract all full loops, requiring complex backtracking mechanisms; third, most methods do not support preserving the order information of the full loop in the time domain, making it difficult to meet the needs of subsequent intraspectral loop sorting and sequence structure analysis; fourth, most algorithms are offline batch processing modes, which are poorly adaptable to scenarios that require online continuous processing of new data.

[0005] Therefore, how to efficiently and thoroughly identify and extract the entire cycle while preserving the temporal sequence information of the cycle is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] In view of the above problems, the present invention proposes an online time-preserving rainflow counting method to overcome or at least partially solve the above problems.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides an online time-series preserved rainflow counting method, comprising the following steps: S1. Obtain the original overload sequence of the aircraft during a certain measured mission segment; S2. Read four consecutive data points from the original overload sequence into a new dataset. Check whether the data points in the new dataset are a peak-valley alternation sequence. If there are non-peak-valley intermediate points, remove the intermediate points and supplement the subsequent data points from the original overload sequence until the new dataset forms four consecutive peak-valley value points. Determine if a countable full cycle exists in the new dataset. If it does, record the two intermediate data points of the full cycle in the cycle result list, delete the two intermediate data points of the full cycle in the new dataset, and add two new data points from the original overload sequence to the end of the new dataset. If it does not exist, add one new data point from the original overload sequence to the end of the new dataset. S3. Using the last four data points in the new dataset as input for the next round, repeat S2 until all data points in the original overload sequence have been read and there is no full loop in the new dataset. Use the latest loop result sequence as the final count result of the current task segment. S4. Sort all full loops in the latest loop result list in chronological order.

[0009] Furthermore, in S2, the existence of countable fully cyclic data points in the new dataset is determined according to a preset criterion. The determination method is as follows: Suppose that there are four consecutive peak and trough points in the new dataset ( n zi-1 , n zi , n zi+1 , n zi+2 ),in, zi Indicates the index of the data point; If | n zi+1 - n zi |≤max{| n zi - n zi-1 |, | n zi+2 - n zi+1 |}, then determine the two intermediate data points among the four consecutive peak and valley points ( n zi , n zi+1 ) form a complete loop, and the intermediate data points ( n zi , n zi+1 Record the data to the loop results list, and delete these two intermediate data points in the new dataset. nzi , n zi+1 ); If | n zi+1 - n zi |>max{| n zi - n zi-1 |, | n zi+2 - n zi+1 The |} indicates that these four consecutive peak and valley points do not constitute an extractable full cycle. In this case, these four consecutive peak and valley points in the new dataset are retained.

[0010] Furthermore, in S4, the earlier real time corresponding to the peak and valley values ​​in each full loop in the loop result list is used as the time stamp of the corresponding full loop. All full loops are sorted in ascending order according to this stamp to obtain a full loop sequence arranged in time order. The overload sequence that cannot be further decomposed in the new dataset is used as the residual loop sequence of the current task segment.

[0011] Furthermore, if new data points are added to the original overload sequence after S4, and there are countable full cycles in the latest round, then two newly generated peak and valley data points are appended to the end of the current residual cycle sequence. If there are no countable full cycles in the latest round, then one newly generated data point is added to the end of the current residual cycle sequence, and the last four data points are used as new inputs to repeat S2 until no new data points are generated in the original overload sequence.

[0012] Furthermore, in S1, the original overload sequence is the overload sequence after filtering.

[0013] Furthermore, in S2, a filtering window is introduced, and four consecutive data points from the original overload sequence are read into the filtering window. The data points within the filtering window constitute a new dataset.

[0014] Secondly, the present invention provides an online time-series preserved rainflow counting system, which employs the online time-series preserved rainflow counting method described above, including: The data acquisition module is used to acquire the original overload sequence of the aircraft during a specific test mission segment; The judgment module reads four consecutive data points from the original overload sequence into a new dataset. It checks if the data points in the new dataset form an alternating peak-valley sequence. If there are non-peak-valley intermediate points, these are removed, and subsequent data points are added from the original overload sequence until the new dataset forms four consecutive peak-valley points. It then checks if a countable full cycle exists in the new dataset. If it does, the two intermediate data points of this full cycle are recorded in the cycle result list, and these two intermediate data points are deleted from the new dataset. Two new data points are added from the original overload sequence to the end of the new dataset. If no such cycle exists, one new data point is added from the original overload sequence to the end of the new dataset. The loop module is used to take the last four data points in the new dataset as input to the next round of the decision module, until all data points in the original overload sequence have been read and there is no longer a full loop in the new dataset. The latest round of loop result sequence is used as the final count result of the current task segment. The sorting module is used to sort all full loops in the loop result list in chronological order.

[0015] Furthermore, the system also includes an online post-processing module, which is used to determine whether there are newly added data points in the original overload sequence. If so, it determines whether there is a countable full cycle in the latest round. If there is a countable full cycle, two newly generated peak and valley data points are appended to the end of the current residual cycle sequence. If there is no countable full cycle in the latest round, one newly generated data point is added to the end of the current participating cycle sequence, and the last four data points are used as the new input for the determination module until no new data points are generated in the original overload sequence.

[0016] Thirdly, the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor; when the processor executes the computer program, it implements the steps of the online time-series preserved rainflow counting method as described above.

[0017] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the steps of the online time-series preserved rainflow counting method as described above.

[0018] As can be seen from the above technical solution, compared with the prior art, the present invention has the following beneficial effects: 1. Improving the sensitivity of identifying micro-nested loops by replacing the geometric inclusion criterion with the amplitude comparison criterion: In this invention, the amplitude comparison between the two middle points of the range and the adjacent range is used as the loop identification criterion in a local four-point window. This criterion directly matches the physical meaning of the material hysteresis loop closure. It does not require global path simulation and can promptly reveal micro-closed loops nested in macroscopic loads in the first round of local judgment, avoiding missed detections caused by the failure to meet the geometric inclusion conditions.

[0019] 2. A global iterative counting mechanism is adopted to ensure the thoroughness of full cycle extraction: After each round of scanning and counting, the extracted full cycles are deleted from the sequence, and the updated sequence is used as the new input for the next round of iteration, until no more full cycles can be extracted. This mechanism can automatically capture newly generated countable cycles due to sequence concatenation after the deletion of counted cycles, ensuring that the counting results converge and are without omissions at the macroscopic level.

[0020] 3. Introducing a time stamping and sorting mechanism to fully preserve the temporal order information of the cycles: This invention uses the earliest real time corresponding to the peak and valley values ​​in each full cycle as the time stamp of that cycle. After all cycles have been extracted, they are sorted by time to obtain a time-ordered full cycle sequence. This provides crucial temporal dimension information for subsequent load order effect analysis and intraspectral cycle distribution research.

[0021] 4. Supports online incremental processing, adaptable to long-term continuous monitoring scenarios: This invention defines the payload sequence that cannot be further decomposed in the new dataset after counting as the residual cyclic sequence. When new peak and valley data are input, it can be directly appended to the end of the current residual cyclic sequence and the cyclic identification and extraction can continue without re-initialization or global recalculation, significantly reducing the overhead of repeated calculations. It is suitable for long-term continuous monitoring of flight test data and data streaming input scenarios.

[0022] 5. It can be seamlessly cascaded with the front-end preprocessing method to form a complete front-end processing link: This invention directly receives the filtered and compressed overload peak-valley value sequence as input, without changing the original sampling sequence number and time correspondence of the retained points. It can be used in conjunction with the aforementioned four-point sliding window overload sequence filtering method to form a complete "filtering-counting" front-end processing scheme for overload sequences of transport aircraft, which has strong engineering practicality. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0024] Figure 1 This is a flowchart of the online time-series preserved rainflow counting method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a valley-peak-valley-peak type full-cycle waveform provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a peak-valley-peak-valley full-cycle waveform provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of two types of waveforms that cannot be extracted in the full cycle, as provided in an embodiment of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] like Figure 1 As shown in the figure, an embodiment of the present invention discloses an online time-preserving rainflow counting method, including the following steps: S1. Obtain the original overload sequence of the aircraft during a certain measured mission segment; S2. Read four consecutive data points from the original overload sequence into a new dataset. Check whether the data points in the new dataset are a peak-valley alternation sequence. If there are non-peak-valley intermediate points, remove the intermediate points and supplement the subsequent data points from the original overload sequence until the new dataset forms four consecutive peak-valley value points. According to the preset amplitude discrimination criterion, determine whether there is a countable full cycle in the new dataset. If it exists, record the two intermediate data points of the full cycle in the cycle result list, delete the two intermediate data points of the full cycle in the new dataset, and add two new data points from the original overload sequence to the end of the new dataset; if it does not exist, add one new data point from the original overload sequence to the end of the new dataset. S3. Using the last four data points in the new dataset as input for the next round, repeat S2 until all data points in the original overload sequence have been read and there is no full loop in the new dataset. Use the latest loop result sequence as the final count result of the current task segment. S4. Sort all full loops in the latest loop result list in chronological order.

[0027] The core of this invention lies in the four-point online rainflow counting that preserves the time-domain load sequence, used to perform cyclic counting on the filtered overload time history. The aim is to directly extract the complete cycles in the load history of each task segment, while preserving the order information of these complete cycles in the time domain.

[0028] Before performing the counting statistics, peak and valley values ​​are detected in the overload time history of each task segment to identify all peak and valley values, and the corresponding sampling point numbers are retained. If the following formula is satisfied, a peak or valley value is recorded.

[0029] or

[0030] The specific implementation process is as follows: S1. Data preparation: Using the filtered overload sequence of a measured task segment as the counting unit, prepare the input data.

[0031] S2. Loop Identification and Extraction: Read in four consecutive data points of the original overload sequence in sequence. n zi-1 , n zi , n zi+1 , n zi+2 The data points within the filtering window form a new dataset, also known as a buffer. zi This represents the index of the data points. It checks whether the data points within the window form a strict peak-valley alternation sequence. If there are non-peak-valley intermediate points, they are removed, and subsequent data points are added from the original sequence until four consecutive peak-valley values ​​are formed within the window. For example, in (1,0,-1), 0 is an intermediate point and needs to be removed.

[0032] Subsequently, the existence of a countable full cycle is determined according to the following amplitude discrimination criterion: 1) If | n zi+1 - n zi |≤max{| n zi - n zi-1 |, | n zi+2 - n zi+1 |}, then determine the two intermediate data points among the four consecutive peak and valley points ( n zi , n zi+1 This constitutes a complete loop, the magnitude of which is | nzi+1 - n zi | / 2, mean is ( n zi + n zi+1 ) / 2, and the intermediate data points ( n zi , n zi+1 Record the data to the loop results list, and delete these two intermediate data points in the new dataset. n zi , n zi+1 According to the above criteria, six typical envelope-type full-cycle waveforms can be formed during the overload time history, and the corresponding waveforms are as follows: Figures 2-3 As shown.

[0033] 2) If | n zi+1 - n zi |>max{| n zi - n zi-1 |, | n zi+2 - n zi+1 The |} indicates that these four consecutive peak-valley points do not constitute an extractable complete cycle. In this case, these four consecutive peak-valley points in the new dataset are retained. Under this criterion, these four data points that do not form a complete cycle can be classified into two typical waveforms in terms of time history, such as... Figure 4 As shown.

[0034] S3, Window Sliding and Iteration: If a complete cycle is extracted in this round, two new data points are added from the original overload sequence to the end of the new dataset. If a complete cycle is not extracted in this round, one data point is added. Then, return to S2, and use the last four data points in the new dataset as input for the next round. Repeat this process until all data points in the original overload sequence have been read.

[0035] S4. Sequence Reset and Termination: If no more complete cycles can be extracted from the new dataset in the final round, it indicates that all possible cycles in the sequence have been completely counted. At this point, all recorded complete cycles represent the final count result for this task segment.

[0036] Then, the earlier real time corresponding to the peak and valley values ​​in each full loop in the loop result list is used as the time stamp of the corresponding full loop. All full loops are sorted in ascending order according to this stamp to obtain a full loop sequence arranged in time order. The overload sequence that cannot be further decomposed in the new dataset is used as the residual loop sequence of the current task segment.

[0037] S5. Online follow-up processing: If new data points are added to the original overload sequence after S4, and the latest round contains countable full cycles, then two newly generated peak-valley data points are appended to the end of the current residual cyclic sequence. If the latest round does not contain countable full cycles, then one newly generated data point is appended to the end of the current residual cyclic sequence. Using the last four data points as new input, S2 is repeated to perform cycle identification and extraction. After extraction, subsequent new data is appended to the end of the residual cyclic sequence, and the above process is repeated until no new data points are generated in the original overload sequence. This processing method avoids repeated initialization and global recalculation.

[0038] The difference between the method of this invention and the classic rainflow counting method / traditional four-point rainflow counting method is that: (1) The discrimination criterion shifts from geometric inclusion to amplitude comparison, highlighting the efficient identification capability of local windows. Some four-point rainflow counting methods use geometric inclusion as the criterion (requiring the middle two points to fall within the interval of the first and last points). When nesting micro-cycles under macro-loads, this can lead to inconsistencies with physically closed hysteresis loops. The method of this invention focuses on amplitude comparison: within a local four-point window ( n zi-1 , n zi , n zi+1 , n zi+2 If | n zi+1 - n zi |≤max{| n zi - n zi-1 |,| n zi+2 - n zi+1 If |}, it is identified as a complete loop and extracted immediately. This localization criterion not only aligns with the physical meaning of loop closure but also avoids dependence on global path simulation, making it easier to discover and extract microscopic closed loops nested within macroscopic loads locally. This direct window discrimination strategy improves the consistency, interpretability, and computational efficiency of the identification.

[0039] Taking the sequence (0, 1, -2, 10) as an example: First, (1, -2) corresponds to a complete cycle (closed hysteresis loop) in a mechanical sense, with a half-amplitude (amplitude) of |1-(-2)| / 2=1.5 and a mean of (1+(-2)) / 2=-0.5. Since |1-(-2)|=3<|-2-10|=10, according to the amplitude comparison criterion, (1, -2) can be directly determined as a complete cycle and extracted. In contrast, if a four-point implementation with geometric inclusion as the core criterion is used, the cycle may be ignored because -2 exceeds the interval (0, 10). At the same time, in the classic rainflow counting method that emphasizes strict path simulation, the flow path starting from peak 1 will not close at -2, but will continue to flow towards 10. Therefore, this micro-cycle is often not extracted in the initial scanning stage (it may only appear in the subsequent backtracking or iteration process). In contrast, the local amplitude criterion can directly reveal this micro-cycle in the first round of local judgment, demonstrating sensitivity and timeliness to nested cycles.

[0040] (2) The present invention adopts a global iteration method. Even if the counting is not directly triggered in the local window, it can capture the newly emerging countable cycles caused by the reconnection of the sequence due to the deletion of the counted cycles and the dynamic connection of the sequence through the deletion of the counted cycles, until the counting results converge at the macro level.

[0041] Taking the sequence (-1, -3, 1, -2, 3, -1.5, -1.25, -4, -1.35) as an example, the traditional four-point rainflow counting method, if only one round of counting is performed, can only identify and extract the complete cycles (1, -2) and (3, -1.5), without further discovering the new closed cycles induced by sequence concatenation after the deletion of these cycles. After the above two complete cycles are deleted, the sequence is updated to (-1, -3, -1.25, -4, -1.35). At this time, (-3, -1.25) becomes the enveloped range and satisfies the counting condition, thus forming a new extractable complete cycle. In contrast, repeated global iterations can continuously extract such subsequently generated complete cycles until the result converges, avoiding the problems of incomplete extraction and incomplete counting of complete cycles.

[0042] (3) Time order preservation and incremental continuation. This invention introduces a time order preservation mechanism, which maintains the relative order of load cycles in the time domain during the cyclic identification / extraction process, facilitating subsequent intraspectral cyclic sorting and sequence structure analysis. Furthermore, to meet online processing requirements and reduce redundant computation overhead, the load sequence that cannot be further decomposed after the previous processing round is defined as the residual cyclic sequence. When newly generated peak and valley data arrive, they are directly appended to the end of this residual cyclic sequence, and the last four data points are used as new input to continue execution from the cyclic identification and extraction step. This incremental processing method avoids re-initialization and global recalculation.

[0043] In other embodiments, the present invention also provides an online time-series preserved rainflow counting system, which employs the online time-series preserved rainflow counting method described above, including: The data acquisition module is used to acquire the original overload sequence of the aircraft during a specific test mission segment; The judgment module reads four consecutive data points from the original overload sequence into a new dataset. It checks if the data points in the new dataset form an alternating peak-valley sequence. If there are non-peak-valley intermediate points, these are removed, and subsequent data points are added from the original overload sequence until the new dataset forms four consecutive peak-valley points. It then checks if a countable full cycle exists in the new dataset. If it does, the two intermediate data points of this full cycle are recorded in the cycle result list, and these two intermediate data points are deleted from the new dataset. Two new data points are added from the original overload sequence to the end of the new dataset. If no such cycle exists, one new data point is added from the original overload sequence to the end of the new dataset. The loop module is used to take the last four data points in the new dataset as input to the next round of the decision module, until all data points in the original overload sequence have been read and there is no longer a full loop in the new dataset. The latest round of loop result sequence is used as the final count result of the current task segment. The sorting module is used to sort all full loops in the loop result list in chronological order.

[0044] The system also includes an online post-processing module, which is used to determine whether there are newly added data points in the original overload sequence. If so, it determines whether there is a countable full cycle in the latest round. If there is a countable full cycle, it appends two newly generated peak and valley data points to the end of the current residual cycle sequence. If there is no countable full cycle in the latest round, it adds one newly generated data point to the end of the current participating cycle sequence, and uses the last four data points as the new input for the determination module, until no new data points are generated in the original overload sequence.

[0045] In another embodiment, the present invention also provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor; wherein, when the processor executes the computer program, it implements the steps of the online time-retaining rainflow counting method as described above.

[0046] In one embodiment, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the online time-series preserved rainflow counting method as described above.

[0047] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0048] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An online time-series preserved rainflow counting method, characterized in that, Includes the following steps: S1. Obtain the original overload sequence of the aircraft during a certain measured mission segment; S2. Read four consecutive data points from the original overload sequence into a new dataset. Check whether the data points in the new dataset are a peak-valley alternation sequence. If there are non-peak-valley intermediate points, remove the intermediate points and supplement the subsequent data points from the original overload sequence until the new dataset forms four consecutive peak-valley value points. According to the preset amplitude discrimination criterion, determine whether there is a countable full cycle in the new dataset. If it exists, record the two intermediate data points of the full cycle in the cycle result list, delete the two intermediate data points of the full cycle in the new dataset, and add two new data points from the original overload sequence to the end of the new dataset; if it does not exist, add one new data point from the original overload sequence to the end of the new dataset. S3. Using the last four data points in the new dataset as input for the next round, repeat S2 until all data points in the original overload sequence have been read and there is no full loop in the new dataset. Use the latest loop result sequence as the final count result of the current task segment. S4. Sort all full loops in the latest loop result list in chronological order.

2. The online time-series preserved rainflow counting method as described in claim 1, characterized in that, In S2, the method for determining whether there are countable full loops in the new dataset according to the preset amplitude discrimination criterion is as follows: Suppose that there are four consecutive peak and trough points in the new dataset ( n zi-1 , n zi , n zi+1 , n zi+2 ),in, zi Indicates the index of the data point; If | n zi+1 - n zi |≤max{| n zi - n zi-1 |, | n zi+2 - n zi+1 |}, then determine the two intermediate data points among the four consecutive peak and valley points ( n zi , n zi+1 ) form a complete loop, and the intermediate data points ( n zi , n zi+1 Record the data to the loop results list, and delete these two intermediate data points in the new dataset. n zi , n zi+1 ); If | n zi+1 - n zi |>max{| n zi - n zi-1 |, | n zi+2 - n zi+1 The |} indicates that these four consecutive peak and valley points do not constitute an extractable full cycle. In this case, these four consecutive peak and valley points in the new dataset are retained.

3. The online time-series preserved rainflow counting method as described in claim 1, characterized in that, In S4, the earlier real time corresponding to the peak and valley values ​​in each full loop in the loop result list is used as the time stamp of the corresponding full loop. All full loops are sorted in ascending order according to this stamp to obtain a full loop sequence arranged in time order. The overload sequence that cannot be further decomposed in the new dataset is used as the residual loop sequence of the current task segment.

4. The online time-series preserved rainflow counting method as described in claim 3, characterized in that, If, after S4, there are new data points in the original overload sequence and there are countable full cycles in the latest round, then two newly generated peak and valley data points are appended to the end of the current residual cycle sequence. If there are no countable full cycles in the latest round, then one newly generated data point is added to the end of the current residual cycle sequence, and the last four data points are used as new inputs to repeat S2 until no new data points are generated in the original overload sequence.

5. The online time-series preserved rainflow counting method as described in claim 1, characterized in that, In S1, the original overload sequence is the overload sequence after filtering.

6. The online time-series preserved rainflow counting method as described in claim 1, characterized in that, In S2, a filtering window is introduced, and four consecutive data points from the original overload sequence are read into the filtering window. The data points in the filtering window constitute a new dataset.

7. An online time-series preserved rainflow counting system, characterized in that, It employs the online time-series preserved rainflow counting method as described in any one of claims 1-6, comprising: The data acquisition module is used to acquire the original overload sequence of the aircraft during a specific test mission segment; The judgment module reads four consecutive data points from the original overload sequence into a new dataset. It checks if the data points in the new dataset form an alternating peak-valley sequence. If there are non-peak-valley intermediate points, these are removed, and subsequent data points are added from the original overload sequence until the new dataset forms four consecutive peak-valley points. It then checks if a countable full cycle exists in the new dataset. If it does, the two intermediate data points of this full cycle are recorded in the cycle result list, and these two intermediate data points are deleted from the new dataset. Two new data points are added from the original overload sequence to the end of the new dataset. If no such cycle exists, one new data point is added from the original overload sequence to the end of the new dataset. The loop module is used to take the last four data points in the new dataset as input to the next round of the judgment module, until all data points in the original overload sequence have been read and there is no longer a full loop in the new dataset. The latest round of loop result sequence is used as the final count result of the current task segment. The sorting module is used to sort all full loops in the loop result list in chronological order.

8. The online time-series preserved rainflow counting system as described in claim 7, characterized in that, It also includes an online post-processing module, which is used to determine whether there are newly added data points in the original overload sequence. If so, it determines whether there is a countable full cycle in the latest round. If there is a countable full cycle, it appends two newly generated peak and valley data points to the end of the current residual cycle sequence. If there is no countable full cycle in the latest round, it adds one newly generated data point to the end of the current participating cycle sequence, and uses the last four data points as the new input of the judgment module until no new data points are generated in the original overload sequence.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor; characterized in that, when the processor executes the computer program, it implements the steps of the online time-series preserved rainflow counting method as described in any one of claims 1-6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the online time-series preserved rainflow counting method as described in any one of claims 1-6.

Images