Intelligent manufacturing process data acquisition method and system

Abstract

The invention relates to the technical field of data processing, in particular to an intelligent manufacturing process data acquisition method and system, and the method comprises the steps: carrying out the industrial data acquisition and industrial data processing of an intelligent manufacturing process; the industrial data collection comprises the following steps: collecting original data in a preset time period in an intelligent manufacturing process; the industrial data processing comprises the following steps: determining a data information enriching stage according to original data; determining the collection authenticity of the data information enriching stage; determining an adaptive de-noising decomposition interval scale of the data information enriching stage by using the collection authenticity and the initial decomposition interval scale of the data information enriching stage; decomposing the original data by using the self-adaptive denoising decomposition interval scale to obtain all component data matrixes; selecting a target component data matrix from all the component data matrixes; and adding the target component data matrixes to obtain target acquisition data. According to the invention, the accuracy and reliability of data acquisition in the intelligent manufacturing process are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for collecting data in an intelligent manufacturing process. Background Art

[0002] Smart manufacturing is the direction of transformation and upgrading of the manufacturing industry. It realizes the intelligence and digitization of the production process through the deep integration of information technology and manufacturing technology. In the process of smart manufacturing, data collection is the basis for realizing intelligence and digitization. By real-time monitoring of various data in the production process, enterprises can timely discover and solve problems, optimize production processes, and improve product quality and production efficiency.

[0003] In some scenarios, multiple sensors are often used to monitor the status of equipment, environmental parameters, and key indicators in the production process. As the number of intelligent manufacturing equipment increases and the production line becomes more complex, the data collection environment becomes more complex. In order to cope with a more complex data collection environment, the number of sensors can be increased and the frequency of data collection can be increased. As the number of sensors increases and the frequency of data collection increases, the data throughput increases dramatically, resulting in a decrease in the timeliness of the data collection system. And due to the complexity of the data collection environment, it is easy to cause data overlap and mutual influence, causing data disorder, thereby generating erroneous data and affecting the extraction of real information for subsequent analysis. As a result, the accuracy and reliability of data collection in the intelligent manufacturing process are low. Summary of the invention

[0004] In order to solve the technical problem of low accuracy and reliability of data collection in the intelligent manufacturing process, the purpose of the present invention is to provide a method and system for data collection in the intelligent manufacturing process. The technical solution adopted is as follows:

[0005] In a first aspect, an embodiment of the present invention provides a method for collecting data in an intelligent manufacturing process, including: collecting industrial data and processing industrial data in an intelligent manufacturing process; industrial data collection includes: collecting raw data within a predetermined time period in the intelligent manufacturing process; industrial data processing includes: determining a data information rich stage based on the raw data; determining the collection authenticity of the data information rich stage based on data maximum points, data minimum points and the window area where each data information rich point in the data information rich stage is located; determining an adaptive denoising decomposition interval scale of the data information rich stage using the collection authenticity and the initial decomposition interval scale of the data information rich stage; decomposing the raw data of the data information rich stage using the adaptive denoising decomposition interval scale to obtain all component data matrices; selecting a target component data matrix with a variance less than a first threshold from all component data matrices; adding the target component data matrices to obtain target collected data.

[0006] Optionally, determining the data information rich stage based on the original data includes: fitting the original data to obtain a fitting curve; constructing a window centered on any original data in the fitting curve to obtain a target window where the original data is located; calculating the variance of the data in the target window where the original data is located, the maximum value of the data in the target window, the minimum value of the data in the target window, and the first time distance between the maximum value of the data and the minimum value of the data; calculating the first difference between the maximum value of the data and the minimum value of the data, and the first ratio between the first difference and the first time distance; determining the first product between the variance and the first ratio as the information detail level of the original data; determining the original data whose information detail level is greater than a second threshold as a data information rich point; marking the interval where continuous data information rich points are located as a data information rich stage.

[0007] Optionally, determining the collection authenticity in the data information rich stage based on the data maximum points, data minimum points in the data information rich stage and the window area where each data information rich point in the data information rich stage is located includes: determining the data fluctuation delay characteristic coefficient in the data information rich stage based on the data maximum points and data minimum points in the data information rich stage; determining the distortion coefficient in the data information rich stage based on the window area where each data information rich point in the data information rich stage is located; determining the collection authenticity in the data information rich stage based on the information detail level, data fluctuation delay characteristic coefficient and distortion coefficient of all data information rich points in the data information rich stage.

[0008] Optionally, determining the data fluctuation delay characteristic coefficient of the data information rich stage based on the data maximum points and data minimum points in the data information rich stage includes: determining a plurality of consecutive data sampling points after the data maximum point in the data information rich stage to constitute a data sampling point set of the data maximum point; determining the delay coefficient of the data maximum point based on the slope of any maximum point in the data sampling point set of the data maximum point on the fitting curve of the original data and a second time distance between any maximum point and the nearest minimum point in the data sampling point set; and averaging the delay coefficients of all data maximum points in the data information rich node to obtain the data fluctuation delay characteristic coefficient of the data information rich stage.

[0009] Optionally, determining the delay coefficient of the data maximum point based on the slope of any maximum point in the data sampling point set of the data maximum point on the fitting curve of the original data and the second time distance between any maximum point and the nearest minimum point in the data sampling point set includes: calculating a second ratio of the slope of any maximum point on the fitting curve of the original data and the corresponding second time distance; multiplying the second ratios corresponding to each maximum point to obtain the delay coefficient of the data maximum point.

[0010] Optionally, determining the distortion coefficient of the data information rich stage according to the window area where each data information rich point in the data information rich stage is located includes: obtaining the data maximum value, the data minimum value and the third time distance between the data maximum value and the data minimum value in the window area where the data information rich point is located; determining the second difference between the means of the original data corresponding to the data information rich points on the left and right sides of the data information rich point; determining the distortion coefficient of the data information rich stage according to the data maximum value, the data minimum value, the third time distance and the second difference of each data information rich point.

[0011] Optionally, determining the distortion coefficient of the data information rich stage based on the data maximum value, data minimum value, third time distance and second difference of each data information rich point includes: calculating the third difference between the data maximum value and the data minimum value of each data information rich point, and the third ratio between the third difference and the third time distance; calculating the second product between the third ratio and the absolute value of the second difference, and calculating the superposition value of the second products of each data information rich point; determining the fourth ratio of the superposition value of the second product to the number of all data information rich points in the data information rich stage as the distortion coefficient of the data information rich stage.

[0012] Optionally, determining the collection authenticity in the data information rich stage based on the information detail level, data fluctuation delay characteristic coefficient and distortion coefficient of all data information rich points in the data information rich stage includes: determining the information detail level of all data information rich points in the data information rich stage as the data information richness in the data information rich stage; determining the third product of the data information richness, the data fluctuation delay characteristic coefficient and the distortion coefficient as the collection authenticity in the data information rich stage.

[0013] Optionally, determining the adaptive denoising decomposition interval scale in the data information rich stage by using the initial decomposition interval scale in the data information rich stage includes: determining the ratio of the initial decomposition interval scale to the acquisition authenticity as the adaptive denoising decomposition interval scale.

[0014] In a second aspect, an embodiment of the present invention provides an intelligent manufacturing process data collection system, comprising: a processor and a memory; wherein the memory is used to store computer programs that can be run on the processor; the processor is used to execute the program stored in the memory to implement the steps of the intelligent manufacturing process data collection method mentioned in the first aspect.

[0015] The present invention has the following beneficial effects: an embodiment of the present invention collects industrial data and processes industrial data in an intelligent manufacturing process; industrial data collection includes: collecting raw data within a predetermined time period in the intelligent manufacturing process; industrial data processing includes: determining a data information-rich stage according to the raw data; determining the collection authenticity of the data information-rich stage according to data maximum points, data minimum points and the window area where each data information-rich point in the data information-rich stage is located; determining an adaptive denoising decomposition interval scale of the data information-rich stage using the collection authenticity and the initial decomposition interval scale of the data information-rich stage; decomposing the raw data of the data information-rich stage using the adaptive denoising decomposition interval scale to obtain all component data matrices; selecting a target component data matrix whose variance is less than a first threshold from all component data matrices; and adding the target component data matrices to obtain target collection data.

[0016] In this way, the embodiment of the present invention can extract data from the data information-rich stage of the original data collected in the intelligent manufacturing process, avoiding the problem of a sharp increase in data throughput leading to a decrease in the timeliness of the data collection system. The authenticity of the data collected in the data information-rich stage is analyzed. Finally, the adaptive denoising decomposition interval scale of the data information-rich stage is determined, and the original data of the data information-rich stage is decomposed. The target component data matrix with a variance less than the first threshold is selected from the component data matrix and added to obtain the target collected data, thereby achieving the purpose of denoising and reducing erroneous data, and improving the accuracy and reliability of data collection in the intelligent manufacturing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A flow chart of a method for collecting data in an intelligent manufacturing process provided by one embodiment of the present invention;

[0019] Figure 2 A flowchart of industrial data processing in an intelligent manufacturing process provided by an embodiment of the present invention;

[0020] Figure 3 A schematic diagram of the structure of an intelligent manufacturing process data acquisition system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the intelligent manufacturing process data acquisition method and system proposed by the present invention, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0022] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0023] The specific scheme of an intelligent manufacturing process data collection method and system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0024] Embodiment 1:

[0025] See also Figure 1 , which shows a flow chart of a method for collecting data in an intelligent manufacturing process provided by an embodiment of the present invention, including:

[0026] S101, collect and process industrial data for the intelligent manufacturing process.

[0027] Specifically, industrial data collection includes: collecting raw data within a predetermined time period in the intelligent manufacturing process. Among them, various sensors are used in the embodiment of the present invention to collect data in the intelligent manufacturing process. The types of sensors include but are not limited to temperature sensors, pressure sensors, humidity sensors, vibration sensors, and smoke sensors. These sensors are fixed to the key parts of the intelligent manufacturing machine tool to collect raw data in real time. The data collected by the sensor can be wirelessly connected to a data acquisition controller, such as an embedded controller and a PLC controller. The data acquisition controller is responsible for reading the sensor signal and performing preliminary processing on the raw signal, such as filtering and amplification, to ensure data accuracy. Furthermore, the data acquisition controller can also set a data sampling frequency of once per second, and store the collected raw data after removing duplication, missing and abnormal values. In order to facilitate industrial data processing in subsequent steps. The predetermined time period can be determined according to actual conditions. In the embodiment of the present invention, the value is 1 hour.

[0028] Further, such as Figure 2 As shown, Figure 2 A flowchart of industrial data processing in an intelligent manufacturing process provided by an embodiment of the present invention includes:

[0029] S201, determining a data information enrichment stage according to original data.

[0030] Specifically, in the process of intelligent manufacturing, the main purpose of data collection is to ensure the accuracy and completeness of the data, so attention needs to be paid to the disordered data in the original data in the intelligent manufacturing process. In actual operation, the more disordered and complex the data in the intelligent manufacturing process is, the more the data in the current stage contains more manufacturing process information or the data in this stage is subject to serious external interference, resulting in inaccurate data. Therefore, when the data fluctuations in a certain stage are more drastic and the amplitude of the change is larger, it means that the intelligent manufacturing process in this stage contains rich information, such as drastic fluctuations may indicate equipment failure, wear and tear, and other abnormal interference conditions, which are very important for the maintenance and timely repair of equipment. Therefore, in the embodiment of the present invention, it is necessary to determine the stage of rich data information in the original data.

[0031] Further, when determining the data information rich stage, as an optional embodiment of the present invention, the original data is first fitted to obtain a fitting curve; then a window is constructed with any original data in the fitting curve as the center to obtain a target window where the original data is located; secondly, the variance of the data in the target window where the original data is located, the maximum value of the data in the target window, the minimum value of the data in the target window, and the first time distance between the maximum value of the data and the minimum value of the data are calculated; and the first difference between the maximum value of the data and the minimum value of the data, and the first ratio between the first difference and the first time distance are calculated; then the first product between the variance and the first ratio is determined to be the information detail level of the original data; and the original data with an information detail level greater than a second threshold is determined to be a data information rich point; finally, the interval where the continuous data information rich points are located is marked as a data information rich stage.

[0032] Specifically, first obtain the original data x within 1 hour of the history of the intelligent manufacturing process, and then use linear fitting and least squares method to obtain the fitting curve f(x) of the original data x. Build a window with any original data in the fitting curve f(x) as the center, and the step length is 5 data points adjacent to the left and right of any original data to obtain the target window of any original data. If there are not enough remaining data points, all the remaining data points are used as a target window. Get the maximum value H of the data in the target window max , data minimum value H min and the first time distance D between the two t . Calculate the variance S of all data points in each target window.

[0033] Furthermore, in the embodiment of the present invention, the following formula is used to calculate the information detail level:

[0034]

[0035] In the above formula, F i represents the information detail of the i-th original data, and i represents the i-th original data in the fitting curve f(x). S i Represents the variance of the data in the target window corresponding to the i-th original data. Represents the maximum value of the data in the target window corresponding to the i-th original data. Represents the minimum value of the data in the target window corresponding to the i-th original data. Represents the i-th original data corresponding to the target window and The first time distance between them, t represents time and has no other special meaning.

[0036] The information details of all original data are obtained in the above manner.

[0037] Furthermore, the second threshold value can be determined according to actual conditions. In the embodiment of the present invention, the second threshold value T1=0.5 is set to perform threshold judgment, and data sampling points corresponding to original data with information detail levels greater than 0.5 are regarded as data information-rich points.

[0038] Furthermore, all adjacent and continuous data information-rich points are regarded as data of the same stage and marked as the data information-rich stage.

[0039] S202, determining the data collection authenticity in the data information enrichment stage according to the data maximum point, the data minimum point and the window area where each data information enrichment point is located in the data information enrichment stage.

[0040] Specifically, during operation, intelligent manufacturing equipment will adjust its state due to changes in load, temperature or operating mode, resulting in significant fluctuations in the collected data, accompanied by certain delay changes, reflecting the time required for the intelligent manufacturing equipment to return to a stable state. In contrast, noise interference from the external environment will cause random fluctuations, but usually will not cause delay changes. Therefore, the embodiment of the present invention determines the authenticity of the data collected during the data information-rich stage by analyzing the data fluctuation delay and distortion coefficient during the data information-rich stage.

[0041] Furthermore, the embodiment of the present invention obtains all data information rich stages in the original collected data of the intelligent manufacturing process. For the collection authenticity of each data information rich stage, the data fluctuation delay characteristic coefficient of the data information rich stage is first determined based on the data maximum points and data minimum points in the data information rich stage; then the distortion coefficient of the data information rich stage is determined based on the window area where each data information rich point in the data information rich stage is located; finally, the collection authenticity of the data information rich stage is determined based on the information detail level, data fluctuation delay characteristic coefficient and distortion coefficient of all data information rich points in the data information rich stage.

[0042] Specifically, in determining the data fluctuation delay in the data information rich stage, as an optional embodiment of the present invention, firstly, a plurality of continuous data sampling points after the data maximum point in the data information rich stage are determined to constitute a data sampling point set of the data maximum point; then, according to the slope of any maximum point in the data sampling point set of the data maximum point on the fitting curve of the original data, and the second time distance between any maximum point and the nearest minimum point in the data sampling point set, the delay coefficient of the data maximum point is determined; finally, the delay coefficients of all data maximum points in the data information rich node are averaged to obtain the data fluctuation delay characteristic coefficient of the data information rich stage.

[0043] Among them, the embodiment of the present invention counts 19 consecutive data sampling points after each data maximum point in each data information rich stage to form a data sampling point set for each data maximum point, and each data sampling point set contains at least one maximum point and at least one minimum point. The slope k of each maximum point in each data sampling point set on the fitting curve f(x) and the time distance L between the maximum point corresponding to the data sampling point set and the nearest minimum point are obtained. When determining the delay coefficient of the data maximum point, the embodiment of the present invention first calculates the second ratio of the slope of any maximum point on the fitting curve of the original data to the corresponding second time distance; then the second ratio corresponding to each maximum point is multiplied to obtain the delay coefficient of the data maximum point.

[0044] Specifically, the embodiment of the present invention uses the following formula to calculate the delay coefficient of the data maximum point:

[0045]

[0046] In the above formula, E represents the delay coefficient of the data maximum point. M represents the set of 19 data sampling points after a data maximum point in the data information rich stage. m represents the mth maximum point in the M set. k m Represents the slope of the mth maximum point on the data fitting curve f(x). mRepresents the second time distance between the mth maximum point and the nearest minimum point.

[0047] Further, after obtaining the delay coefficients of all data maximum points in the data information rich stage, the set N of all data maximum points in the data information rich stage and the delay coefficient corresponding to each data maximum point are obtained. The embodiment of the present invention uses the following formula to calculate the data fluctuation delay characteristic coefficient in the data information rich stage:

[0048]

[0049] In the above formula, R j Represents the data fluctuation delay characteristic coefficient of the jth data information-rich stage. j represents the jth data information-rich stage in the original data of the intelligent manufacturing process. N j Represents the set of all data maximum points in the jth data information enrichment stage. j Represents N j The nth in the set j A data maximum point. Represents the nth j The delay coefficient of the data maximum point.

[0050] Furthermore, the data fluctuation delay characteristic coefficient of the data information rich stage is effectively measured by the average of the delay coefficients of all data maximum points in the data information rich stage. If the data fluctuation delay characteristic coefficient of the data information rich stage is less than the threshold, it means that the data information rich stage is caused by external noise. If the data fluctuation delay characteristic coefficient of the stage is greater than the threshold, it means that the data information rich stage is caused by the intelligent manufacturing equipment itself.

[0051] Furthermore, in the data information-rich stage caused by the state changes of the intelligent manufacturing equipment itself, strong external environmental noise interference may cause the sensor input data to exceed its measurement range. In this case, the sensor output data will not be able to return to the normal value range in time after reaching the maximum value. This abnormal output will cause a pulse train to appear in the data, that is, a series of abnormally high values ​​in a short period of time. When similar pulse trains appear frequently in the fluctuation pattern, the input data of the surface sensor has been saturated, resulting in the output data being unable to return to the normal range in time, which makes the overall output data high, exacerbating the distortion of the collected data. Therefore, it is necessary to analyze it in the subsequent decomposition process.

[0052] Furthermore, when determining the distortion coefficient of the data information rich stage, as an optional embodiment of the present invention, firstly obtain the data maximum value, data minimum value and the third time distance between the data maximum value and the data minimum value in the window area where the data information rich point is located; then determine the second difference between the means of the original data corresponding to the data information rich points on the left and right sides of the data information rich point; finally, determine the distortion coefficient of the data information rich stage according to the data maximum value, data minimum value, third time distance and second difference of each data information rich point.

[0053] Specifically, all data-rich points in the data-rich stage are obtained, and a window area of ​​the data-rich point is constructed starting from any data-rich point in the data-rich stage. The step length of the window area is the next 10 data-rich points. When the remaining data-rich points are insufficient, the remaining data-rich points are used as a window area. The maximum value of the data G in the window area of ​​each data-rich point is obtained. max , data minimum value G min And the third time distance t between the two. Obtain the data means on the left and right sides of any data information-rich point in the data information-rich stage, and the absolute value of the second difference between the data means on both sides μ(G) a,b The absolute value of the second difference μ(G) a,b The larger the value of is, the greater the data difference on both sides of the information-rich point is, that is, the collected original data may have a pulse train phenomenon at the information-rich point, resulting in a larger value for subsequent data acquisition.

[0054] Furthermore, when determining the distortion coefficient of the data information rich stage, as an optional embodiment of the present invention, firstly, the third difference between the data maximum value and the data minimum value of each data information rich point, and the third ratio between the third difference and the third time distance are calculated; then, the second product between the third ratio and the absolute value of the second difference is calculated, and the superposition value of the second products of each data information rich point is calculated; finally, the fourth ratio of the superposition value of the second product to the number of all data information rich points in the data information rich stage is determined as the distortion coefficient of the data information rich stage.

[0055] Specifically, the embodiment of the present invention uses the following formula to calculate the distortion coefficient in the data information rich stage:

[0056]

[0057] In the above formula, P j Represents the distortion coefficient of the jth data information-rich stage. j represents the set of all data information enrichment points in the jth data information enrichment stage.j Represents U j The uth in the set j data points. Represents the u j The maximum value of the data in the window area corresponding to the data point. Represents the u j The data point corresponds to the minimum value of the data in the window area. Represents the u j The absolute value of the second difference between the means of the original data corresponding to the data information-rich points on the left and right sides of the data point in the jth data information-rich stage. Indicates u j The third time distance between the maximum value and the minimum value of the data in the window area corresponding to each data point.

[0058] In the stage where data information is rich, the frequent occurrence of a large number of pulse trains may indicate that the data in the information-rich stage is seriously distorted, and the fluctuations in these pulse trains will mask many important changes. As a result, key events are not recorded, which significantly increases the distortion of the sensor data. In addition, due to the pulse train, the output data cannot quickly return to the normal range after reaching the maximum value. Therefore, in the subsequent data decomposition process, a stronger data decomposition scale is required to eliminate the influence of the gradually increasing fluctuation trend of the collected data. In the embodiment of the present invention, the acquisition authenticity and the initial decomposition interval scale of the data information rich stage are used to determine the final denoising decomposition interval scale.

[0059] Furthermore, when determining the authenticity of collection in the data information rich stage, as an optional embodiment of the present invention, the information detail level of all data information rich points in the data information rich stage is first determined as the data information richness of the data information rich stage; then the third product of the data information richness, the data fluctuation delay characteristic coefficient and the distortion coefficient is determined as the authenticity of collection in the data information rich stage.

[0060] Specifically, the embodiment of the present invention first calculates the mean value ε(F) of the information detail level of all data information rich points in the data information rich stage. j , which is used as the data information richness of the data information rich stage. The greater the data information richness, the more detailed information is contained in the data information rich stage, and analysis is required in the subsequent data collection accuracy analysis. Finally, the data fluctuation delay characteristic coefficient, distortion coefficient and data information richness of the data information rich stage are combined to evaluate the collection authenticity of the data information rich stage.

[0061] Furthermore, the embodiment of the present invention uses the following formula to calculate the authenticity of the data collected during the information enrichment phase:

[0062] KPI j =ε(F) j ×R j ×P j

[0063] In the above formula, KPI j Indicates the authenticity of the collection of the jth data information enrichment stage. ε(F) j Represents the data information richness of the jth data information rich stage. R j represents the data fluctuation delay characteristic coefficient of the jth data information-rich stage. j Represents the distortion coefficient of the jth data information-rich stage.

[0064] Among them, when the data in the data information-rich stage shows obvious delay characteristics, that is, the data fluctuation delay characteristic coefficient is large, it means that the data fluctuation is caused by the state change of the intelligent manufacturing equipment itself, rather than external noise interference. By analyzing the distortion of the data information-rich stage, the data in the data information-rich stage can be corrected, so as to more accurately evaluate the authenticity of the data collection in the data information-rich stage.

[0065] S203, determining an adaptive denoising decomposition interval scale in a data information rich stage by using the acquisition authenticity and the initial decomposition interval scale in a data information rich stage.

[0066] Specifically, in the traditional matrix decomposition algorithm (Robust Low-rank Matrix Decomposition, RLMD), the local mean in the sliding window of each raw data in the intelligent manufacturing process is calculated, and the length of the sliding window can be set to 20 sampling points. The initial decomposition interval scale is determined according to the local mean in the sliding window of each raw data. It is worth noting that the initial decomposition interval scale can refer to the RLMD decomposition algorithm in the prior art, and the embodiment of the present invention will not be repeated here.

[0067] Further, when determining the adaptive denoising decomposition interval scale in the data information rich stage, as an optional embodiment of the present invention, the ratio of the initial decomposition interval scale to the acquisition authenticity is determined as the adaptive denoising decomposition interval scale.

[0068] Specifically, the embodiment of the present invention uses the following formula to calculate the adaptive denoising decomposition interval scale:

[0069]

[0070] In the above formula, Q j Represents the adaptive denoising decomposition interval scale for the jth data information-rich stage. Represents the initial decomposition interval scale of the jth data information-rich stage. KPI j It represents the collection authenticity of the jth data information enrichment stage.

[0071] S204, using an adaptive denoising decomposition interval scale to decompose the original data in the data information rich stage to obtain all component data matrices.

[0072] Specifically, the RLMD decomposition algorithm is used to decompose the original data in the data information rich stage using the adaptive denoising decomposition interval scale to obtain all component data matrices. In the RLMD decomposition algorithm, each component data matrix represents a specific data component of the original data, and is usually sorted from high to low according to the importance of the component data matrix.

[0073] S205 , selecting a target component data matrix whose variance is less than a first threshold from all component data matrices.

[0074] Specifically, a component data matrix with a smaller data variance is selected from among all component data matrices. The first threshold can be determined according to actual conditions, and is not limited in the embodiment of the present invention.

[0075] S206, adding the target component data matrices to obtain target acquisition data.

[0076] Specifically, all target component data matrices are added together to obtain accurate and reliable collected data. The processed collected data is finally transmitted to the user interface, so that relevant personnel can effectively adjust and optimize the intelligent manufacturing process based on the collected data. The user interface can provide users with intuitive data display and operation functions, helping them to monitor production conditions in real time and make timely decisions based on data analysis results, thereby improving manufacturing efficiency and product quality.

[0077] The embodiment of the present invention can extract the data information rich stage from the raw data collected in the intelligent manufacturing process, avoiding the problem of a sharp increase in data throughput leading to a decrease in the timeliness of the data collection system. The authenticity of the data collected in the data information rich stage is analyzed. The adaptive denoising decomposition interval scale of the data information rich stage is finally determined, and the raw data of the data information rich stage is decomposed. The target component data matrix with a variance less than the first threshold is selected from the component data matrix and added to obtain the target collected data, thereby achieving the purpose of denoising and reducing erroneous data, and improving the accuracy and reliability of data collection in the intelligent manufacturing process.

[0078] Embodiment 2:

[0079] Corresponding to the intelligent manufacturing process data collection method provided in the above embodiment, based on the same technical concept, an embodiment of the present invention further provides an intelligent manufacturing process data collection system, which is used to execute the above intelligent manufacturing process data collection method. Figure 3 A schematic diagram of a smart manufacturing process data acquisition system provided by an embodiment of the present invention is shown in FIG. Figure 3 The intelligent manufacturing process data acquisition system may have relatively large differences due to different configurations or performances, and may include one or more processors 301 and memory 302, the memory 302 is used to store computer programs that can be run on the processor 301, and the processor 301 is used to execute the program stored in the memory 302 to achieve the above Figure 1 or Figure 2 The memory 302 may be a temporary storage or a permanent storage. The application stored in the memory 302 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the intelligent manufacturing process data acquisition system.

[0080] Furthermore, the processor 301 can be configured to communicate with the memory 302 to execute a series of computer executable instructions in the memory 302 on the intelligent manufacturing process data acquisition system. The intelligent manufacturing process data acquisition system can also include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input and output interfaces 305, and one or more keyboards 306.

[0081] Specifically in this embodiment, the intelligent manufacturing process data acquisition system includes a processor, a communication interface, a memory and a communication bus; wherein the processor, the communication interface and the memory communicate with each other through the bus; the memory is used to store computer programs; the processor is used to execute the programs stored in the memory to achieve the above Figure 1 or Figure 2 The various steps in the method embodiment are similar to those in the method embodiment, and have the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of the present invention will not be described in detail here.

[0082] It should be noted that the intelligent manufacturing process data acquisition system provided by the embodiment of the present invention and the intelligent manufacturing process data acquisition method provided by the embodiment of the present invention are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned intelligent manufacturing process data acquisition method, and has the same or similar beneficial effects, and the repetitive parts will not be repeated.

[0083] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0084] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for collecting data in an intelligent manufacturing process, characterized in that: The intelligent manufacturing process data collection method comprises: Collect and process industrial data for intelligent manufacturing processes; The industrial data collection includes: collecting raw data within a predetermined time period in the intelligent manufacturing process; The industrial data processing comprises: determining a data information enrichment stage according to the original data; Determine the collection authenticity of the data information enrichment stage according to the data maximum value point, the data minimum value point in the data information enrichment stage and the window area where each data information enrichment point in the data information enrichment stage is located; Determine the adaptive denoising decomposition interval scale of the data information rich stage by using the acquisition authenticity and the initial decomposition interval scale of the data information rich stage; Decomposing the original data in the data information-rich stage by using the adaptive denoising decomposition interval scale to obtain all component data matrices; Selecting a target component data matrix whose variance is less than a first threshold from all the component data matrices; The target component data matrices are added together to obtain target acquisition data.

2. The intelligent manufacturing process data collection method according to claim 1 is characterized in that: The stage of determining data information enrichment according to the original data comprises: Fitting the original data to obtain a fitting curve; Constructing a window with any original data in the fitting curve as the center to obtain a target window where the original data is located; Calculate the variance of the data in the target window where the original data is located, the maximum value of the data in the target window, the minimum value of the data in the target window, and the first time distance between the maximum value of the data and the minimum value of the data; Calculating a first difference between the maximum value of the data and the minimum value of the data, and a first ratio between the first difference and the first time distance; Determine a first product between the variance and the first ratio as the information detail level of the original data; Determining the original data whose information detail level is greater than a second threshold as a data information rich point; The interval where the continuous data information-rich points are located is marked as the data information-rich stage.

3. The intelligent manufacturing process data collection method according to claim 1 is characterized in that: Determining the acquisition authenticity of the data information enrichment stage according to the data maximum value point, the data minimum value point in the data information enrichment stage and the window area where each data information enrichment point in the data information enrichment stage is located includes: Determine the data fluctuation delay characteristic coefficient of the data information rich stage according to the data maximum value point and the data minimum value point in the data information rich stage; Determine the distortion coefficient of the data information rich stage according to the window area where each data information rich point is located in the data information rich stage; The authenticity of the data collection in the data information rich stage is determined according to the information detail level of all data information rich points in the data information rich stage, the data fluctuation delay characteristic coefficient and the distortion coefficient.

4. The intelligent manufacturing process data collection method according to claim 3 is characterized in that: Determining the data fluctuation delay characteristic coefficient of the data information rich stage according to the data maximum value point and the data minimum value point in the data information rich stage includes: Determine a plurality of continuous data sampling points after the data maximum value point in the data information rich stage to form a data sampling point set of the data maximum value point; Determine the delay coefficient of the data maximum point according to the slope of any maximum point in the data sampling point set of the data maximum point on the fitting curve of the original data and the second time distance between any maximum point and the nearest minimum point in the data sampling point set; The delay coefficients of all data maximum points in the data information rich node are averaged to obtain the data fluctuation delay characteristic coefficient of the data information rich stage.

5. The intelligent manufacturing process data collection method according to claim 4 is characterized in that: Determining the delay coefficient of the data maximum point according to the slope of any maximum point in the data sampling point set of the data maximum point on the fitting curve of the original data and the second time distance between any maximum point and the nearest minimum point in the data sampling point set comprises: Calculate a second ratio of the slope of the arbitrary maximum point on the fitting curve of the original data to the corresponding second time distance; The second ratios corresponding to the respective maximum points are multiplied to obtain the delay coefficient of the data maximum point.

6. The intelligent manufacturing process data collection method according to claim 3 is characterized in that: The step of determining the distortion coefficient of the data information rich stage according to the window area where each data information rich point is located in the data information rich stage comprises: Acquire a data maximum value, a data minimum value, and a third time distance between the data maximum value and the data minimum value in the window region where the data information rich point is located; Determine a second difference between the means of the original data corresponding to the data information rich points on the left and right sides of the data information rich point; The distortion coefficient of the data information rich stage is determined according to the data maximum value, the data minimum value, the third time distance and the second difference value of each of the data information rich points.

7. The intelligent manufacturing process data collection method according to claim 6 is characterized in that: The determining of the distortion coefficient of the data information rich stage according to the data maximum value, the data minimum value, the third time distance and the second difference value of each data information rich point includes: Calculating a third difference between the data maximum value and the data minimum value of each of the data information-rich points, and a third ratio between the third difference and the third time distance; Calculating a second product between the third ratio and the absolute value of the second difference, and calculating a superposition value of the second products of each of the data information-rich points; A fourth ratio of the superposition value of the second product to the number of all data information rich points in the data information rich stage is determined as a distortion coefficient of the data information rich stage.

8. The intelligent manufacturing process data collection method according to claim 3 is characterized in that: Determining the acquisition authenticity of the data information rich stage according to the information detail level of all data information rich points in the data information rich stage, the data fluctuation delay characteristic coefficient and the distortion coefficient includes: Determine the information detail level of all data information enrichment points in the data information enrichment stage as the data information enrichment level of the data information enrichment stage; Determine the third product of the data information richness, the data fluctuation delay characteristic coefficient and the distortion coefficient as the collection authenticity of the data information rich stage.

9. The intelligent manufacturing process data collection method according to claim 1, characterized in that: The step of determining the adaptive denoising decomposition interval scale of the data information rich stage by using the acquisition authenticity and the initial decomposition interval scale of the data information rich stage comprises: The ratio of the initial decomposition interval scale to the acquisition authenticity is determined as the adaptive denoising decomposition interval scale.

10. An intelligent manufacturing process data acquisition system, characterized in that: The intelligent manufacturing process data acquisition system includes: a processor and a memory; wherein the memory is used to store computer programs that can be run on the processor; the processor is used to execute the program stored in the memory to implement the steps of the intelligent manufacturing process data acquisition method as described in any one of claims 1-9.

Images