Production data collaborative collection method for digital control platform of wind power equipment manufacturing

Abstract

The present invention relates to the field of data acquisition technology, and in particular to a method for collaboratively acquiring production data of a digital management and control platform for wind power equipment manufacturing. The method comprises the following steps: determining data change trends of various production data within a plurality of consecutive reference time windows during a historical wind power equipment production stage, combining the data correlations of any two production data to obtain linkage reactions of the various production data within the historical wind power equipment production stage, fusing the data change trends and linkage reactions of the various production data to obtain the production status of wind power equipment manufacturing, and adjusting the reference time window in combination with the adaptation of the reference time window so that the size of the adjusted time window is adapted to the production status of wind power equipment manufacturing, thereby effectively solving the problem of low reliability of production data acquisition caused by adopting the same and fixed time window for acquisition of various types of production data in the wind power equipment manufacturing process, thereby improving the reliability of production data acquisition.

Description

Technical Field

[0001] The present invention relates to the technical field of data acquisition, and in particular to a method for collaboratively acquiring production data on a digital management and control platform for wind power equipment manufacturing. Background Art

[0002] The manufacturing process of wind turbine equipment (such as wind turbine blades, towers, hubs, main shafts, gearboxes, etc.) involves a large number of complex process flows and a variety of heterogeneous equipment. Production data is scattered across different systems and equipment, resulting in inconsistent collection methods, diverse data formats, and poor real-time performance. This makes it difficult to achieve transparent and traceable management of the entire production process. In the field of wind turbine equipment manufacturing, integrating the Internet of Things, cloud computing, big data analysis, and artificial intelligence technologies, and building a cross-functional, full-process digital management and control platform—the Digital Management and Control Platform for Wind Power Equipment Manufacturing—can achieve collaborative collection of production data from multiple processes, multiple equipment, and multiple systems on the manufacturing site, providing a high-quality data foundation for achieving lean manufacturing, quality traceability, and predictive equipment maintenance.

[0003] In existing technologies, data from each collection point is synchronously uploaded to a central server or cloud platform based on preset timestamps, enabling the collaborative collection of multi-source data throughout the entire wind power equipment manufacturing process. However, due to factors such as varying collection frequencies, the data from different devices may not accurately correspond in the time dimension, making it difficult to accurately analyze the relationship between them and impacting the consistency of production data. Furthermore, due to the varying importance of different types of production data to the production process, different collection frequencies (e.g., high or low) are required. Forcibly standardizing the sampling frequencies for collaboratively collecting multiple types of production data—that is, collecting all types of production data within the same, fixed time window throughout the entire production process—would compromise the reliability of production data collection, as well as the integrity and effectiveness of analysis. For example, uniformly adopting high-frequency collection would result in a large amount of invalid, redundant data, increasing system burden. Conversely, uniformly adopting low-frequency collection could miss transient characteristics or abnormal signals in critical operating conditions. Summary of the Invention

[0004] In order to solve the technical problem that the reliability of data collection is affected when using a fixed time window to collect all types of production data in the wind power equipment manufacturing process, the purpose of the present invention is to provide a production data collaborative collection method for a digital management and control platform for wind power equipment manufacturing. The technical solution adopted is as follows:

[0005] The present invention provides a method for collaboratively collecting production data of a digital management and control platform for wind power equipment manufacturing, comprising:

[0006] Determine the data change trends of various production data in a number of consecutive benchmark time windows during the historical wind power equipment production stage;

[0007] Obtaining data correlation between any two production data, and combining the data change trends to obtain linkage reactions of various production data in historical wind power equipment production stages, wherein the linkage reactions represent associations with various other production data;

[0008] Integrate the data change trends and linkage reactions of various production data to obtain the production status of wind power equipment manufacturing;

[0009] According to the production status and the adaptation of the reference time window, the reference time window is adjusted to collect production data according to the adjusted time window.

[0010] In an exemplary embodiment, the collaborative production data collection method further includes a process of acquiring the reference time window:

[0011] Presetting a candidate time window set, wherein the candidate time window set includes multiple candidate time windows of different lengths;

[0012] Obtaining the adaptation of the time window to be selected according to the number of types of production data existing in the time window to be selected and the sampling period of all types of production data;

[0013] All candidate time windows are traversed to obtain the adaptation status of each candidate time window, and the candidate time window corresponding to the best adaptation status is selected as the reference time window.

[0014] In an exemplary embodiment, the process of obtaining the adaptation status of the candidate time window includes:

[0015] Obtaining the difference between the sampling period of all types of production data and the duration of the selected time window to obtain the duration adaptability of the sampling period of all types of production data;

[0016] The duration adaptability of the sampling periods of all types of production data is integrated to obtain the comprehensive duration adaptability;

[0017] The number of types of production data existing in the candidate time window and the comprehensive duration adaptability are integrated to obtain the adaptation of the candidate time window. The adaptation of the candidate time window is proportional to the number of types and the comprehensive duration adaptability.

[0018] In an exemplary embodiment, the process of acquiring the data change trend includes:

[0019] Determine the degree of data fluctuation of various production data within each benchmark time window;

[0020] Obtaining the changing trend characteristics of the data fluctuation degree of various production data for the consecutive reference time windows, as well as the changing degree of the data fluctuation degree;

[0021] By integrating the change trend characteristics and change degrees, the data change trends of various production data are obtained.

[0022] In an exemplary embodiment, the data fluctuation degree of various production data in each reference time window is obtained by fusing the data variance and data range of various production data in each reference time window.

[0023] In an exemplary embodiment, the process of acquiring the change trend feature includes:

[0024] Calculate the difference in data fluctuation between the latter and the former reference time windows of various production data in two adjacent reference time windows;

[0025] By fusing the differences, the change trend characteristics of the data fluctuation degree of various production data are obtained;

[0026] The process of obtaining the degree of change includes: obtaining a fitting straight line of the data fluctuation degree of the production data for the consecutive reference time windows, wherein the degree of change is the slope of the fitting straight line.

[0027] In an exemplary embodiment, the process of obtaining the linkage reaction includes:

[0028] According to the data correlation between various production data and other various production data, the data change trends of various production data are weighted averaged to obtain the linkage reaction of various production data.

[0029] In an exemplary embodiment, the process of obtaining the production status of wind power equipment manufacturing includes:

[0030] Determine the production status performance of various production data, wherein the production status performance is obtained by integrating the data change trend and linkage reaction;

[0031] The production status of wind power equipment manufacturing is obtained by integrating the production status of all types of production data.

[0032] In an exemplary embodiment, the specific process of adjusting the reference time window includes:

[0033] Obtaining a production status matching degree according to the production status and the adaptation of the reference time window; the production status matching degree is inversely proportional to the production status and directly proportional to the adaptation;

[0034] The reference time window is adjusted according to the production status matching degree.

[0035] In an exemplary embodiment, adjusting the reference time window according to the production status matching degree includes: multiplying the reference time window by an adjustment coefficient, and the obtained product is the adjusted time window; the adjustment coefficient is obtained by the production status matching degree.

[0036] The present invention has the following beneficial effects: the present invention takes the reference time window as the starting point, and obtains the production status of wind power equipment manufacturing by analyzing the data change trend of various production data in several consecutive reference time windows and the data correlation of any two production data, thereby realizing the adjustment of the reference time window, so that the adjusted time window size is adapted to the production status of wind power equipment manufacturing, and effectively solves the problem of low reliability of production data collection caused by using the same and fixed time window for collecting various types of production data in the wind power equipment manufacturing process, thereby improving the reliability of production data collection and the consistency and integrity of production data, as well as the effectiveness and accuracy of data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flow chart of a method for collaboratively collecting production data of a digital management and control platform for wind power equipment manufacturing provided by one embodiment of the present invention;

[0038] Figure 2 is a flowchart of obtaining a reference time window provided by an embodiment of the present invention;

[0039] Figure 3 This is a flow chart for obtaining the adaptation status of a candidate time window provided by an embodiment of the present invention;

[0040] Figure 4 is a waveform diagram of raw vibration data of a main shaft provided by one embodiment of the present invention;

[0041] Figure 5 This is a flow chart for obtaining data change trends provided by one embodiment of the present invention;

[0042] Figure 6 This is a flowchart for obtaining the production status of wind power equipment manufacturing provided by one embodiment of the present invention;

[0043] Figure 7 is a flowchart of adjusting the reference time window provided by one embodiment of the present invention;

[0044] Figure 8 This is a waveform diagram of spindle vibration data after reference time window adjustment provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] To further illustrate the technical means and effects employed by the present invention to achieve its intended objectives, the following detailed description of the specific embodiments, structures, features, and effects of the present invention is provided in conjunction with the accompanying drawings and preferred embodiments. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which this invention pertains. The data and information collected in this application were obtained with full consent and authorization.

[0047] This embodiment provides a collaborative production data collection method for a digital management and control platform for wind power equipment manufacturing. This method is suitable for collecting various types of production data during wind power equipment manufacturing, specifically the various process flows and heterogeneous equipment involved in wind power equipment manufacturing. Specifically, data from various equipment, process steps, and information systems is collected through industrial gateways, PLCs, sensors, edge computing devices, and the like. This data includes structural component processing data, welding temperature, motor vibration, assembly torque, ambient temperature and humidity, quality inspection results, and worker operation records. The various process flows and heterogeneous equipment involved in wind power equipment manufacturing together constitute the digital management and control platform for wind power equipment manufacturing.

[0048] There are several main ways to collect various production data in the wind power equipment manufacturing process: PLC and industrial bus collection: real-time reading of key process parameters from control systems such as CNC machine tools, welding robots, and blade forming equipment; sensor collection: deploying temperature, pressure, vibration, current, voltage and other sensors to obtain environmental and equipment status; video and image collection: used for quality inspection or auxiliary analysis of manual assembly, etc., so as to obtain various production data in the wind power equipment manufacturing process.

[0049] When collecting data, various production data can also be preprocessed. For example, edge computing devices can be used to preprocess, convert protocols, and compress data to improve data transmission efficiency. Communication protocols and data formats used by different devices can be parsed and data structures standardized. Data with packet loss, abnormal fluctuations, or zero values ​​can be interpolated, filtered, or eliminated to ensure data stability. Methods such as smoothing, sliding mean, and Kalman filtering can be used to remove high-frequency noise from sensor signals.

[0050] The entire manufacturing process of wind power equipment, i.e., the full production process, is divided into several production stages, which are defined as various wind power equipment production stages. For example, the manufacturing process can be divided into component manufacturing process, component assembly process, and final assembly and integration process. Furthermore, each manufacturing sub-process within these three manufacturing processes can be further subdivided. For example, the component assembly process can be divided into large component production process, small component production process, control system production process, etc., thereby dividing the entire wind power equipment production process into several production stages. As another implementation method, the production stages can also be divided by the staff based on several key nodes in the wind power equipment manufacturing process, such as the key nodes of blade manufacturing, tower manufacturing, transmission system core component manufacturing, hub and key structural component manufacturing, and final assembly and integration. Alternatively, the staff can divide the production stages based on changes in the complexity of the manufacturing process. For example, if the complexity of the manufacturing process changes significantly before and after a certain time point, then this time point can be used as the dividing point to divide the entire wind power equipment production process into several production stages.

[0051] For any wind power equipment production stage, the wind power equipment production stage of the wind power equipment that has been manufactured in the past is defined as the historical wind power equipment production stage. Various production data generated in the wind power equipment manufacturing process during the historical wind power equipment production stage are obtained. This embodiment analyzes the various production data in the historical wind power equipment production stage to obtain the final required time window, and collects data according to the obtained time window for various production data in the same wind power equipment production stage in the future wind power equipment manufacturing process. It should be understood that when switching to a different wind power equipment production stage, data processing is required according to the corresponding historical wind power equipment production stage to obtain the final required time window corresponding thereto, and the final time windows corresponding to different historical wind power equipment production stages may be different.

[0052] like Figure 1 As shown, the present embodiment provides a method for collaboratively collecting production data of a digital management and control platform for wind power equipment manufacturing, comprising the following steps:

[0053] Step S1: determining the data change trend of various production data in a historical wind power equipment production stage within a number of consecutive benchmark time windows;

[0054] Step S2: Obtain data correlation between any two production data, and combine the data change trend to obtain the linkage reaction of various production data in the historical wind power equipment production stage;

[0055] Step S3: integrating the data change trends and linkage reactions of various production data to obtain the production status of wind power equipment manufacturing;

[0056] Step S4: According to the production status and the adaptation of the reference time window, the reference time window is adjusted to collect production data according to the adjusted time window.

[0057] Each step is described in detail below with reference to the accompanying drawings.

[0058] Step S1: determining the data change trends of various production data in a historical wind power equipment production stage within a number of consecutive benchmark time windows.

[0059] During wind power equipment manufacturing, various production data have varying sampling frequencies due to their varying importance to the production process. During collaborative collection, the various production data obtained within a fixed time window may become out of sync and inconsistent. Therefore, this embodiment first determines a reference time window, then adjusts it through data analysis to obtain the desired time window.

[0060] In an exemplary embodiment, Figure 2 As shown, a specific process of obtaining the benchmark time window is given as follows:

[0061] Step S1-1: presetting a candidate time window set, wherein the candidate time window set includes a plurality of candidate time windows of different lengths.

[0062] A set of candidate time windows is preset, comprising multiple candidate time windows of varying lengths. The number of candidate time windows is set based on actual needs. Furthermore, the actual lengths of the longest and shortest candidate time windows in the set are set based on actual needs. In one exemplary embodiment, due to the varying importance of different types of production data in wind power equipment manufacturing, sampling frequencies vary. Therefore, the longest candidate time window in the set is the time interval corresponding to production data collected at the lowest frequency, i.e., the sampling period corresponding to the lowest frequency, and the shortest candidate time window in the set is the time interval corresponding to production data collected at the highest frequency, i.e., the sampling period corresponding to the highest frequency. In one exemplary embodiment, a candidate time window of the shortest length can be initially set, and then a candidate time window can be obtained by incrementing the number of candidate time windows at a fixed increment until the longest candidate time window is reached, thereby obtaining multiple candidate time windows.

[0063] The set of candidate time windows is essentially a time window range, which allows a certain time window deviation to exist during the collaborative data collection process. However, by limiting the set of candidate time windows, this deviation is controlled within an acceptable range. Different types of production data within the same time window are integrated to form a complete time window data set. Moreover, the purpose of determining the time window is to make each type of production data correspond to a data value within the time window, wherein high-frequency data (for example, the duration of the time window is several multiples of the sampling period) is downsampled (average, median, etc.), and low-frequency data (for example, the duration of the time window is less than the sampling period), if it cannot be obtained within the time window, can be interpolated, thereby realizing the collaborative collection and analysis of production data of wind power equipment.

[0064] Step S1-2: Obtaining the adaptation status of the time window to be selected according to the number of types of production data existing in the time window to be selected and the sampling period of all types of production data.

[0065] Each candidate time window in the candidate time window set is traversed, and the adaptation of the candidate time window is obtained according to the number of types of production data existing in the candidate time window and the sampling period of each production data. The more types of production data existing in the candidate time window, and the smaller the difference between the sampling period of various production data appearing in the historical wind power equipment production stage and the candidate time window, the better the adaptation of the candidate time window. Finally, the candidate time window with the best adaptation is selected as the benchmark time window, so that the high-frequency data in the candidate time window can capture the rapidly changing details of the production process and reduce the interpolation error of the low-frequency data.

[0066] In an exemplary embodiment, Figure 3 As shown, a specific process of obtaining the adaptation status of the candidate time window is given as follows:

[0067] Step S1-2-1: Obtain the difference between the sampling period of all types of production data and the duration of the selected time window, and obtain the duration adaptability of the sampling period of all types of production data.

[0068] Set the first The sampling period of production data, that is, the sampling time interval is , set the The length of the selected time window is , get the Sampling period of production data With the The duration of the selected time window The difference According to the difference , get the The time adaptation of the sampling period of production data is Inversely proportional, difference The smaller the time length, the higher the adaptability. In an exemplary embodiment, Indicates the The adaptability of the sampling period of the production data is calculated. Where exp is an exponential function with the natural constant e as the base. represents negative correlation normalization.

[0069] According to the above process, the adaptability of the sampling period of each production data existing in the historical wind power equipment production stage to the duration of each candidate time window is obtained.

[0070] Step S1-2-2: The duration adaptability of the sampling periods of all types of production data is integrated to obtain a comprehensive duration adaptability.

[0071] The sampling period for integrating all types of production data is The duration adaptability of the selected time windows In an exemplary embodiment, the sampling period for calculating all types of production data is for the first The duration adaptability of the selected time windows The greater the overall time adaptation, the better the sampling time interval of all types of production data is. The closer the duration of the selected time windows is, the The more appropriate the selected time window is.

[0072] Step S1-2-3: Integrate the number and types of production data in the candidate time window and the comprehensive duration adaptability to obtain the adaptability of the candidate time window.

[0073] No. The production data in the candidate time window is the production data of all types, and the sampling time interval is less than Production data of the selected time window length. The more types of production data there are in the first candidate time window, the The better the adaptation of the selected time window. The higher the comprehensive duration adaptation degree corresponding to the selected time window, the The better the adaptation of the first candidate time window. The adaptation of the first candidate time window is the same as that of the The number of types of production data in the candidate time window and the The comprehensive duration adaptation corresponding to each candidate time window is proportional.

[0074] In an exemplary embodiment, the total number N of production data types appearing in the historical wind power equipment production stage is obtained, and then the first The number of types of production data in the candidate time window , combined with the total number of species N, we get The number of types of production data in the candidate time window The larger the proportion of quantity, the higher the The easier it is to match the production data within a selected time window, the better the adaptation.

[0075] According to The number of types of production data in the selected time window, and the The comprehensive duration adaptation degree corresponding to the selected time window is obtained The adaptation of the candidate time window is given as follows: A specific quantitative formula for the adaptation of the candidate time windows:

[0076] ;

[0077] in, Indicates the The adaptation of the candidate time windows.

[0078] Step S1-3: traverse all candidate time windows, obtain the adaptation status of each candidate time window, and select the candidate time window corresponding to the best adaptation status as the reference time window.

[0079] Traverse all candidate time windows to obtain the adaptation conditions of each candidate time window, and select the candidate time window corresponding to the best adaptation condition, that is, select the candidate time window corresponding to the adaptation condition with the largest value, and this candidate time window is the reference time window.

[0080] The production rhythm varies at different stages of wind power equipment manufacturing, and critical events such as equipment startup and shutdown, and process parameter adjustments may occur. These events can significantly change the production status, making them key periods of data collection critical. The time window needs to adapt to the different stages and states of the production process. By adjusting the time window and narrowing it when critical events occur, these changes can be captured more accurately.

[0081] By analyzing the changing trends and characteristics of data within a time window, the production status and stability of wind power equipment manufacturing can be monitored in real time. The production data within the reference time window can be analyzed to obtain the production status of wind power equipment manufacturing.

[0082] It should be understood that when the overall production process or different production links in the wind power equipment manufacturing process change due to various reasons, the corresponding production data will also fluctuate. Figure 4 As shown, the original vibration data of the main shaft during the manufacturing process of wind power equipment is shown. The horizontal axis represents time, the vertical axis represents amplitude, the rectangular box represents the original vibration data collected within the reference time window, and the width of the rectangular box is the length of the reference time window.

[0083] In this embodiment, to facilitate subsequent processing, various types of production data are standardized to remove dimension and normalize them to a numerical range of [0, 1]. In one exemplary embodiment, the various types of production data are standardized using a maximum-minimum normalization method. For example, for any type of production data, the maximum and minimum values ​​of that type of production data during the historical wind power equipment production phase are obtained, and then each type of production data of that type is normalized using the maximum-minimum normalization method. All types of production data used below are standardized production data.

[0084] Several consecutive reference time windows are divided within the historical wind power equipment production stage. In an exemplary embodiment, starting from the starting time point of the historical wind power equipment production stage, multiple consecutive reference time windows are divided in sequence. The number of reference time windows is set according to actual needs. For the reliability of data processing, the number of reference time windows can be more. The purpose of dividing multiple consecutive reference time windows is to obtain the data change trend of various production data based on the change rules of production data in these consecutive reference time windows. In an exemplary embodiment, Figure 5 As shown, a specific process of obtaining data change trends is given below:

[0085] Step S1-4: Determine the data fluctuation degree of various production data within each reference time window.

[0086] Taking any reference time window as an example, for any production data, obtain the data fluctuation degree of the production data of this production data within the reference time window. In an exemplary embodiment, the data fluctuation degree consists of two parts: data variance and data range. Then, obtain the data variance and data range of the production data of this production data within the reference time window. The data range is the difference between the maximum production data and the minimum production data of this production data within the reference time window. Then calculate the product of the data variance and the data range as the data fluctuation degree. For example: Figure 4 Taking the vibration signal shown as an example, the variance and range of each vibration value in the reference time window are obtained, and then the product of the variance and the range is calculated as the data fluctuation degree of the vibration signal in the reference time window.

[0087] Step S1-5: Obtain the change trend characteristics of the data fluctuation degree of various production data for a number of consecutive reference time windows, as well as the change degree of the data fluctuation degree.

[0088] For any kind of production data, for several consecutive benchmark time windows, the data fluctuation degree corresponding to each benchmark time window can be obtained, and then the changing trend characteristics of the data fluctuation degree of this kind of production data for several consecutive benchmark time windows, as well as the changing degree of the data fluctuation degree, can be obtained.

[0089] Among them, the process of obtaining the change trend characteristics includes: calculating the difference in data fluctuation degree between the latter reference time window and the former reference time window in two adjacent reference time windows of the production data, thereby obtaining the data fluctuation degree difference corresponding to each two adjacent reference time windows, fusing these data fluctuation degree differences, and obtaining the change trend characteristics of the data fluctuation degree of the production data. When the data fluctuation degree difference is a positive number, it means that the data fluctuation degree of the latter reference time window is greater than the data fluctuation degree of the former reference time window, and the data fluctuation degree is increasing. When the data fluctuation degree difference is a negative number, it means that the data fluctuation degree of the latter reference time window is less than the data fluctuation degree of the former reference time window, and the data fluctuation degree is decreasing. Then, for several consecutive reference time windows, the greater the increase in the data fluctuation degree, the more obvious the change trend characteristics. In an exemplary embodiment, a specific quantification method of the change trend characteristics is given as follows:

[0090] ;

[0091] in, Indicates the The changing trend characteristics of production data, Indicates the number of benchmark time windows in a series of benchmark time windows. Indicates the Production data in the The degree of data fluctuation within a benchmark time window, Indicates the Production data in the The degree of data fluctuation within a benchmark time window.

[0092] when Greater than hour, Greater than 0, when Less than hour, is less than 0. Therefore, It may be positive, negative or 0. It may be positive, negative or 0. The larger the value, the The more obvious the changing trend characteristics of the production data are.

[0093] In order to facilitate subsequent processing, the change trend characteristics of various production data are normalized. Here, the maximum and minimum normalization method is adopted. Specifically: the maximum and minimum values ​​in the change trend characteristics of various production data are obtained, and then the maximum and minimum normalization method is adopted to normalize the change trend characteristics of various production data so that the change trend characteristics of various production data are within [0,1]. The larger the value of the change trend characteristic, the more obvious the data change trend.

[0094] The process of obtaining the degree of change in data fluctuation includes: for any type of production data, each of a number of consecutive reference time windows corresponds to a data fluctuation degree. Thus, a data fluctuation degree set is obtained. A two-dimensional coordinate system is constructed with the reference time window as the horizontal axis and the data fluctuation degree as the vertical axis. The data fluctuation degree of each reference time window is mapped to the two-dimensional coordinate system to obtain a number of discrete points. These discrete points are fitted with a straight line to obtain a fitted line. The slope of the fitted line is obtained, and the slope of the fitted line is the degree of change in the data fluctuation degree of the production data. Therefore, when the data fluctuation degree of the production data shows a decreasing trend, the degree of change is negative; when the data fluctuation degree of the production data shows an increasing trend, the degree of change is positive; and when the data fluctuation degree of the production data shows a stable trend, the degree of change is zero. The larger the value of the degree of change in the data fluctuation degree of the production data, the more obvious the data change trend. In order to facilitate subsequent processing, the degree of change of various production data is normalized. Here, the maximum and minimum normalization method is adopted. Specifically: the maximum and minimum values ​​of the degree of change of various production data are obtained, and then the maximum and minimum normalization method is adopted to normalize the degree of change of various production data so that the degree of change of various production data is within [0,1]. The larger the value of the degree of change, the more obvious the data change trend.

[0095] Step S1-6: Integrate the change trend characteristics and the change degree to obtain the data change trends of various production data.

[0096] For the The normalized change trend characteristics and normalized change degree of the production data are calculated. The product of the normalized change trend characteristics of the production data and the normalized change degree is the product of the normalized change trend characteristics of the production data and the normalized change degree. The data change trends of various production data are obtained by using the above process. The greater the data change trend, the greater the degree of change in the production data.

[0097] It should be understood that for low-frequency production data that does not exist or has only one data point in the reference time window, the data change trends of these types of production data are not obtained in the above manner, but are directly set to 0.

[0098] Step S2: Obtain the data correlation between any two types of production data, and combine the data change trend to obtain the linkage reaction of various production data in the historical wind power equipment production stage.

[0099] In order to determine the dynamic adjustment of the reference time window size according to the change of production status, the data correlation of any two production data is obtained. For any two production data, the production data sequences in the historical wind power equipment production stage are obtained respectively, and any two production data sequences are obtained, and the data correlation of the two production data sequences is obtained. In an exemplary embodiment, the DTW (Dynamic Time Warping) distance of the two production data sequences is obtained, and then the DTW distance is negatively normalized, and the result obtained is the data correlation of the two production data sequences. Among them, the negative correlation normalization can be adopted .

[0100] For any type of production data, the data correlation between this type of production data and various other production data is obtained, and then combined with the data change trends of various other production data, the linkage reaction of this type of production data in the historical wind power equipment production stage is obtained. The linkage reaction represents the relationship between this type of production data and various other production data.

[0101] In an exemplary embodiment, the process of obtaining the linkage reaction includes: performing weighted averaging on the data change trends of various production data based on the data correlation between various production data and other various production data, to obtain the linkage reaction of various production data. The calculation formula is as follows:

[0102] ;

[0103] in, Indicates the A linkage reaction of production data; Indicates the The first production data among other various production data Data relevance of production data; Indicates the The data change trend of production data, Indicates that except The number of types of production data other than this type of production data.

[0104] Characterization The relationship between this production data and other various production data, The bigger, The stronger the linkage reaction that a type of production data may bring about in the production process, the greater its importance. The linkage reaction of various production data is obtained by the above method.

[0105] Step S3: Integrate the data change trends and linkage reactions of various production data to obtain the production status of wind power equipment manufacturing.

[0106] If the data change trend of various production data is greater and the linkage reaction of various production data is greater, the production status of wind power equipment manufacturing in the historical wind power equipment production stage is more unstable and the production status value of wind power equipment manufacturing is higher. Therefore, the data change trend and linkage reaction of various production data are integrated to obtain the production status of wind power equipment manufacturing. In an exemplary embodiment, Figure 6 As shown, a specific process for obtaining the production status of wind power equipment manufacturing is given below:

[0107] Step S3-1: Determine the production status performance of various production data, which is obtained by integrating data change trends and linkage reactions.

[0108] For the Production data, integrated The data change trend of the production data and the The linkage reaction of production data is obtained In an exemplary embodiment, the production status of the production data is calculated. The data change trend of the production data and the The product of the linkage reaction of the production data is used as the first The production status of the production data is displayed.

[0109] Step S3-2: Integrate the production status representations of all types of production data to obtain the production status of wind power equipment manufacturing.

[0110] The production status performance of all types of production data is integrated. Specifically, the average value of the production status performance of all types of production data is calculated. The result is the production status of wind power equipment manufacturing:

[0111] ;

[0112] in, Indicates the production status of wind power equipment manufacturing within the historical wind power equipment production stage; Indicates the The production status of the production data is displayed.

[0113] The greater the production status of wind power equipment manufacturing in the historical wind power equipment production stage, the more unstable the production status of wind power equipment manufacturing is, and the more frequent and collaborative collection of all production data is required.

[0114] Step S4: According to the production status and the adaptation of the reference time window, the reference time window is adjusted to collect production data according to the adjusted time window.

[0115] By adjusting the size of the reference time window, you can adapt to different stages and states of the production process. During periods of rapid change or when data synchronization is required, reduce the reference time window to improve data synchronization accuracy. During periods of relatively stable production or when data synchronization is required less, maintain the reference time window.

[0116] The more stable the production status of wind power equipment manufacturing, that is, the smaller the value, the smaller the reference time window is adjusted. When the production status of wind power equipment manufacturing is more unstable, that is, the larger the value, the larger the reference time window is adjusted. Strengthen the real-time monitoring of production data so as to quickly respond and take measures to make adjustments, avoid further expansion of problems, and reduce production losses. Therefore, the reference time window is adjusted according to the production status of wind power equipment manufacturing in the historical wind power equipment production stage and the adaptation of the reference time window. In an exemplary embodiment, if Figure 7 As shown, a specific adjustment process of the benchmark time window is given as follows:

[0117] Step S4-1: Obtain the production status matching degree according to the production status and the adaptation of the reference time window.

[0118] The greater the production status of wind power equipment manufacturing in the historical wind power equipment production stage, the smaller the production status matching degree, and the larger the reduction amplitude of the reference time window. The better the adaptation of the reference time window, the less the reference time window needs to be reduced, and the greater the production status matching degree, the smaller the reduction amplitude of the reference time window. Then, the production status matching degree of wind power equipment manufacturing in the historical wind power equipment production stage is inversely proportional to the production status of wind power equipment manufacturing in the historical wind power equipment production stage, and is directly proportional to the adaptation of the reference time window. In an exemplary embodiment, a specific quantitative method of the production status matching degree is given as follows:

[0119] ;

[0120] in, Indicates the production status matching degree of wind power equipment manufacturing in the historical wind power equipment production stage, Indicates the adaptation status of the base time window.

[0121] Step S4-2: Adjust the reference time window according to the production status matching degree.

[0122] The smaller the production status matching degree of wind power equipment manufacturing in the historical wind power equipment production stage, the larger the reduction amplitude of the reference time window; the larger the production status matching degree of wind power equipment manufacturing in the historical wind power equipment production stage, the smaller the reduction amplitude of the reference time window. Therefore, the adjustment coefficient is obtained according to the production status matching degree. In an exemplary embodiment, the production status matching degree is directly used as the adjustment coefficient, and then the duration of the reference time window is multiplied by the adjustment coefficient, and the product obtained is the duration of the adjusted time window. Figure 8 As shown in the figure, the width of the rectangular box is the adjusted time window size, which can quickly detect sudden abnormal changes in vibration data and adjust the production process in time.

[0123] After obtaining the adjusted time window corresponding to the historical wind power equipment production stage, all future production data for the same wind power equipment production stage will be collected according to the adjusted time window. It should be understood that the historical wind power equipment production stage can also be adjusted to obtain adjusted time windows for other historical wind power equipment production stages, thereby collecting various production data for the same wind power equipment production stages according to the adjusted time windows.

[0124] In subsequent practical applications, for the same wind power equipment production stage in the future corresponding to the historical wind power equipment production stage, various production data are collected according to the determined adjusted time window. Among them, when the production status of wind power equipment manufacturing is large, for example, greater than the preset value 0.5, it means that the production of wind power equipment is in a fluctuating state, and the adjusted time window is smaller, and priority is given to ensuring the complete collection of high-frequency key features. For high-frequency sampling data, representative features can be extracted through downsampling methods (such as averaging, maximum value, standard deviation, etc.) within the adjusted time window; for low-frequency sampling data, since it may not have been updated within the adjusted time window, the data cache and latest value retention mechanism can be used, that is, the latest value in the previous adjusted time window is used as the value in the current adjusted time window to achieve value completion, and mark whether it is a real-time value or a historical completion value in the data to prevent subsequent misuse.

[0125] When the production status of wind power equipment manufacturing is relatively low, for example, less than or equal to the preset value of 0.5, it means that the production of wind power equipment is relatively stable, and the adjusted time window is larger. The high-frequency sampling data is obtained by downsampling. If the low-frequency data is still missing in the adjusted time window, linear interpolation can be performed based on its stable sampling period. The data of two adjacent sampling moments can be used to fill in the missing values ​​according to time calculation to achieve data compression and structural alignment, ensuring both efficiency and accuracy of collaborative collection.

[0126] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0127] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for collaboratively collecting production data for a digital management and control platform for wind power equipment manufacturing, characterized in that: include: Determine the data change trends of various production data in a number of consecutive benchmark time windows during the historical wind power equipment production stage; The process of obtaining the reference time window is as follows: presetting a candidate time window set, which includes multiple candidate time windows of different lengths; obtaining the adaptation of the candidate time windows based on the number of types of production data existing in the candidate time windows and the sampling period of all types of production data; traversing all candidate time windows to obtain the adaptation of each candidate time window, and selecting the candidate time window corresponding to the best adaptation as the reference time window; The process of obtaining the adaptability of the time window to be selected includes: obtaining the difference between the sampling period of all types of production data and the duration of the time window to be selected, and obtaining the duration adaptability of the sampling period of all types of production data; fusing the duration adaptability of the sampling period of all types of production data to obtain the comprehensive duration adaptability; fusing the number of types of production data existing in the time window to be selected and the comprehensive duration adaptability to obtain the adaptability of the time window to be selected, and the adaptability of the time window to be selected is proportional to the number of types and the comprehensive duration adaptability; Obtaining data correlation between any two production data, and combining the data change trends to obtain linkage reactions of various production data in historical wind power equipment production stages, wherein the linkage reactions represent associations with various other production data; Integrate the data change trends and linkage reactions of various production data to obtain the production status of wind power equipment manufacturing; According to the production status and the adaptation of the reference time window, the reference time window is adjusted to collect production data according to the adjusted time window; the specific process of adjusting the reference time window includes: obtaining the production status matching degree according to the production status and the adaptation of the reference time window; the production status matching degree is inversely proportional to the production status and directly proportional to the adaptation; the reference time window is multiplied by the adjustment coefficient, and the product obtained is the adjusted time window; the adjustment coefficient is obtained from the production status matching degree.

2. The method for collaboratively collecting production data of a digital management and control platform for wind power equipment manufacturing according to claim 1, characterized in that: The process of obtaining the data change trend includes: Determine the degree of data fluctuation of various production data within each benchmark time window; Obtaining the changing trend characteristics of the data fluctuation degree of various production data for the consecutive reference time windows, as well as the changing degree of the data fluctuation degree; By integrating the change trend characteristics and change degrees, the data change trends of various production data are obtained.

3. The method for collaboratively collecting production data for a digital management and control platform for wind power equipment manufacturing as described in claim 2 is characterized in that the data fluctuation degree of various production data within each reference time window is obtained by fusing the data variance and data range of various production data within each reference time window.

4. The method for collaboratively collecting production data of a digital management and control platform for wind power equipment manufacturing according to claim 2, characterized in that: The process of obtaining the change trend characteristics includes: Calculate the difference in data fluctuation between the latter and the former reference time windows of various production data in two adjacent reference time windows; By fusing the differences, the change trend characteristics of the data fluctuation degree of various production data are obtained; The process of obtaining the degree of change includes: obtaining a fitting straight line of the data fluctuation degree of the production data for the consecutive reference time windows, wherein the degree of change is the slope of the fitting straight line.

5. The method for collaboratively collecting production data of a digital management and control platform for wind power equipment manufacturing according to claim 1, characterized in that: The process of obtaining the linkage reaction includes: According to the data correlation between various production data and other various production data, the data change trends of various production data are weighted averaged to obtain the linkage reaction of various production data.

6. The method for collaboratively collecting production data of a digital management and control platform for wind power equipment manufacturing according to claim 1, characterized in that: The process of obtaining the production status of wind power equipment manufacturing includes: Determine the production status performance of various production data, wherein the production status performance is obtained by integrating the data change trend and linkage reaction; The production status of wind power equipment manufacturing is obtained by integrating the production status representation of all types of production data.

Images