Preprocessing and analysis methods and media for energy consumption big data in smart manufacturing workshops

By modeling and fusion analysis of energy consumption data in an intelligent manufacturing workshop, the problem that the existing technology cannot quickly process and analyze real-time energy consumption data is solved, and the rapid processing and accurate analysis of data is realized, providing better energy saving Control decision-making basis.

CN114326635BActive Publication Date: 2024-11-19JIANGNAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111666942.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-31
Publication Date
2024-11-19
Estimated Expiration
2041-12-31

AI Technical Summary

Technical Problem

The prior art is difficult to quickly process and analyze real-time energy consumption data of manufacturing workshops, and the processing process of artificial intelligence methods cannot adapt to the complexity of manufacturing workshops and cannot quickly provide decision-making basis for energy conservation and emission reduction.

Method used

A preprocessing and analysis method for energy consumption big data in intelligent manufacturing workshops is proposed. By modeling the energy consumption-related data of the manufacturing process, the energy consumption data and process data are correlated, and the data is fused using instantaneous characteristic values ​​and cumulative energy consumption characteristic values. Realize rapid processing and accurate analysis of data.

Benefits of technology

It realizes rapid preprocessing and fusion of energy consumption data in the manufacturing workshop, improves data accuracy, can quickly respond to the complex production characteristics of the manufacturing workshop, and provides better decision-making basis for energy saving control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114326635B_ABST
    Figure CN114326635B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and medium for preprocessing and analyzing big data of energy consumption in intelligent manufacturing workshops. The method steps include: modeling data related to energy consumption in the manufacturing process, including process data, processing energy consumption data, processing logistics energy consumption data, and auxiliary material consumption data; associating energy consumption data with process data, and dividing energy consumption data according to the processing process or feed rate; fusing instantaneous characteristic values ​​reflecting the law of cutting power changes, cumulative energy consumption characteristic values ​​reflecting overall energy consumption and energy efficiency, and process time data. The present invention can realize the rapid processing of big data of energy consumption in manufacturing workshops, and better adapt to the complex and changeable production characteristics of discrete manufacturing workshops; at the same time, it realizes the rapid fusion processing of manufacturing process energy consumption data and process data, and the finally obtained process energy consumption characteristic data also provides a decision-making basis for energy-saving control in intelligent manufacturing workshops. The method and medium have practical and feasible practical value.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the combination of intelligent manufacturing and big data processing, and in particular to a method and medium for preprocessing and analyzing energy consumption big data in an intelligent manufacturing workshop. Background Art

[0002] As everyone's environmental awareness increases, energy conservation and low-carbon production are gaining more and more attention in the field of manufacturing. In the past few years, how to reduce energy consumption in the manufacturing process has become an important research hotspot. More than half of the energy consumption in the manufacturing workshop comes from the processing machine tools. Therefore, if the energy consumption of machine tools can be reduced, the carbon emissions of the production process can be significantly reduced. In order to achieve energy conservation and emission reduction of machine tools, researchers have conducted a lot of theoretical and practical research, mainly including processing parameter optimization, machine tool status control, process planning and production scheduling.

[0003] On the other hand, with the development of Internet of Things technology, a manufacturing workshop will generate various types of data. At the same time, these data show some characteristics of big data, such as large amount of data, many types, and fast update speed. Therefore, how to obtain and process these data has become an important issue, and various data preprocessing and analysis methods have been proposed, including fog computing, deep learning, and artificial intelligence methods.

[0004] However, the current processing methods for manufacturing workshop energy consumption data still have two shortcomings. First, many current data collection and analysis models and algorithms focus on historical manufacturing data and cannot process real-time production data; second, the current methods, especially artificial intelligence methods, require a certain amount of time to process, which cannot adapt to the complexity of the manufacturing workshop production process. Therefore, how to quickly pre-process and analyze the data of the manufacturing workshop is still a difficult problem. Summary of the invention

[0005] Purpose of the invention: In view of the shortcomings of the existing technology, a method and medium for preprocessing and analyzing energy consumption big data in an intelligent manufacturing workshop is provided to realize rapid processing and data fusion of energy consumption data in the manufacturing process, and provide a decision-making basis for energy-saving control in intelligent manufacturing workshops.

[0006] Technical solution: A method for preprocessing and analyzing big data on energy consumption in smart manufacturing workshops, including:

[0007] Step 1: Model the energy consumption related data of the manufacturing process, which includes process data, processing energy consumption data, processing logistics energy consumption data and auxiliary material consumption data;

[0008] Step 2: Associate energy consumption data with process data and divide energy consumption data according to processing or feed rate;

[0009] Step 3: Fuse the instantaneous characteristic value reflecting the changing law of cutting power, the cumulative energy consumption characteristic value reflecting the overall energy consumption and energy efficiency with the process time data.

[0010] Furthermore, the process data modeling formula is as follows:

[0011] PD={PDID,PID,PName,WID,WName,MID,ptws,ptwe} (1)

[0012] Among them, PD represents process data, PDID represents the ID of process data, PID represents process ID, PName represents process name, WID represents the workpiece number ID, WName represents the workpiece name, MID represents machine tool ID, ptws and ptwe represent the start time and end time of the process respectively.

[0013] Furthermore, the processing energy consumption data modeling formula is as follows:

[0014] MECD={MECDID,MID,p,T} (2)

[0015] Among them, MECD represents the machine tool energy consumption data, MECDID represents the machine tool energy consumption data ID, MID represents the machine tool ID, p represents the machine tool power value, and T represents time.

[0016] Furthermore, the modeling formula for processing logistics energy consumption data is as follows:

[0017] LECD={LECDID,PID,LType,LEData,T} (3)

[0018] Among them, LECD represents the energy consumption of logistics equipment, LECDID represents the energy consumption data ID, LType represents the equipment type, LEData represents the energy consumption, PID represents the process ID, and T represents time.

[0019] Furthermore, the auxiliary material consumption data modeling formula is as follows:

[0020] SMD={SMDID,PID,MID,SType,SMData,T} (4)

[0021] Among them, SMD represents auxiliary material consumption data, SMDID represents auxiliary material consumption ID, SType represents auxiliary material type, SMData represents auxiliary material consumption, MID represents machine tool ID, PID represents process ID, and T represents time.

[0022] Further, step 2 specifically includes: dividing the continuous energy data under different states based on the energy consumption state characteristic curve of the machine tool, and using the functional data of the cutting state to identify different working steps;

[0023] Each time 10 power data are collected, that is, Delete the maximum and minimum values ​​and obtain the remaining 8 data The average value of , i = 1, ..., 8, and the average power is used as the judgment standard:

[0024]

[0025] The energy consumption data segmentation process specifically includes:

[0026] Step 2.1: According to the process plan, obtain the start time ptws of the i-th process i and end time ptws i ;

[0027] Step 2.2: The start time of the jth step of the process is ptws i,j ,

[0028] Step 2.3: Obtain the average power MeanCE according to formula (5) t ;

[0029] Step 2.4: If The machine tool enters the cutting state. The power of the cutting process is if Currently still in the empty cutting state, P aircutting Indicates the machine tool air cutting power;

[0030] Step 2.5: If The machine tool enters another cutting process, and the end time of the current step is ptwe i,j , which is the start time of the next step ptws i,j+1 ;

[0031] Step 2.6: Repeat steps 2.3 to 2.5 until the processing is completed. The time data of all steps is:

[0032] ST i ={PID,PName,ptws i,j ,ptwe i,j} (6).

[0033] Furthermore, in step 3, the instantaneous characteristic value is constructed according to the statistical characteristics, specifically including:

[0034]

[0035]

[0036]

[0037]

[0038] Among them, MaxP i Indicates the maximum energy consumption of this process, MinP i Indicates the minimum energy consumption, MeanP i Represents the average energy consumption, VarP i Represents the standard deviation of energy consumption, N i Indicates the number of energy consumption data;

[0039] The instantaneous eigenvalue modeling formula is as follows:

[0040] IEC i ={ptws i ,ptwe i ,MaxP i ,MinP i ,MeanP i ,VarP i} (11).

[0041] Furthermore, in step 3, the total energy consumption and energy efficiency formula of the process are as follows:

[0042]

[0043]

[0044]

[0045]

[0046] Among them, TotalE i Indicates the total energy consumption of the process, TotalSMD i Indicates the total auxiliary material consumption of the process, EnergyEff i Indicates the energy utilization rate of the process, mres i Indicates the start time of the cutting state of this process, mree i Indicates the end time of the cutting state of this process, ProcessEff i Represents the energy efficiency of the process, V i Indicates the volume of the workpiece removed in this process;

[0047] The cumulative energy consumption characteristic value modeling formula is as follows:

[0048] CEC i ={ptwsi ,ptwe i ,TotalE i ,TotalSMD i ,EnergyEff i ,PorcessEffi i} (16).

[0049] Furthermore, in step 3, the process energy consumption fusion data is obtained by using the relational operation method:

[0050]

[0051] Among them, PECD i Represents the process energy consumption fusion data of the process, Indicates connection operation, SMD i Indicates the auxiliary material consumption data of this process.

[0052] In a specific implementation, there is a computer-readable storage medium, including one or more programs for execution by one or more processors, and the one or more programs include instructions for performing any of the methods described above.

[0053] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:

[0054] (1) It can quickly pre-process the energy consumption big data of the manufacturing workshop, solving the problem that previous technical methods could not respond quickly, and better adapting to the complex and changeable production characteristics of discrete manufacturing workshops;

[0055] (2) Data segmentation is performed based on the energy consumption characteristics of the machine tool itself, taking into account the differences between different machine tools to improve data accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is the energy consumption characteristic curve of the machine tool processing process in the present invention;

[0057] Figure 2 The present invention provides a flow chart for the segmentation of energy consumption big data. DETAILED DESCRIPTION

[0058] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0059] In this embodiment, big data processing is combined with production energy consumption in the manufacturing workshop to assist workshop managers in making decisions on energy conservation and emission reduction. Various machine tools and equipment in the intelligent manufacturing workshop will generate energy consumption data. For this reason, it is necessary to collect various data and pre-process the relevant data to facilitate the evaluation and analysis of the production process and assist in production decision-making. The specific process mainly includes discrete manufacturing process data modeling, energy consumption big data segmentation, and energy consumption big data and production data fusion analysis.

[0060] Step 1: Real-time data modeling of smart workshop

[0061] By configuring the manufacturing data acquisition network, different types of data can be obtained. In order to realize data processing and storage, data modeling is first carried out. The discrete manufacturing process mainly includes four types of energy-saving related data: process data, processing energy consumption data, processing logistics energy consumption data and auxiliary material consumption data.

[0062] Process data includes basic information of each processing step, such as process name, number, machine tool, and processing time. Since all manufacturing activities are carried out according to the process plan, process data is the core data of the discrete manufacturing workshop and can be modeled and expressed by formula (1). This data can be obtained from the process plan and the status of the work-in-progress.

[0063] PD={PDID,PID,PName,WID,WName,MID,ptws,ptwe} (1)

[0064] Among them, PD represents process data, PDID represents the ID of process data, PID represents process ID, PName represents process name, WID represents the workpiece number ID, WName represents the workpiece name, MID represents machine tool ID, ptws and ptwe represent the start time and end time of the process respectively.

[0065] For discrete manufacturing workshops, machine tools consume a lot of energy during the processing process, so it is necessary to model the machine tool energy consumption data, as shown in the following formula (2), which mainly includes machine tool information, energy consumption data and the corresponding time.

[0066] MECD={MECDID,MID,p,T} (2)

[0067] Among them, MECD represents the machine tool energy consumption data, MECDID represents the machine tool energy consumption data ID, MID represents the machine tool ID, p represents the machine tool power value, and T represents time.

[0068] The logistics equipment between machine tools in the smart workshop (such as AGV carts and loading and unloading manipulators) will also generate certain energy consumption. Its modeling is shown in formula (3), which is mainly related to the process information.

[0069] LCD={LECDID,PID,LType,LEData,T} (3)

[0070] Among them, LECD represents the energy consumption of logistics equipment, LECDID represents the energy consumption data ID, LType represents the equipment type (AGV or robot), and LEData represents the energy consumption.

[0071] In addition, in addition to energy consumption, the processing process may also consume some auxiliary materials that may be related to energy consumption, such as compressed air, coolant, etc. Therefore, the auxiliary material data can be expressed as formula (4).

[0072] SMD={SMDID,PID,MID,SType,SMData,T} (4)

[0073] Among them, SMD represents auxiliary material consumption data, SMDID represents auxiliary material consumption ID, SType represents auxiliary material type, SMData represents auxiliary material consumption, MID represents machine tool ID, PID represents process ID, and T represents time.

[0074] Step 2: Data segmentation method for processing energy consumption characteristics

[0075] For a manufacturing process, process data is discrete, while machine tool energy consumption data MECD is collected continuously. In order to associate energy consumption data with process data, it is necessary to divide the energy consumption data according to the processing process or feed rate. To this end, a machine tool energy consumption data segmentation method based on processing energy consumption characteristics is proposed. For a single process, its time node (i.e., start time and end time) can be obtained from the enterprise process planning. However, each process often includes multiple work steps. For example, the turning process can include external cylindrical turning, end face turning, and internal cylindrical turning. Each processing process includes different states such as standby, idle, empty cutting, and cutting, such as Figure 1 As shown, the energy consumption in different states is also different.

[0076] In order to analyze the energy characteristics of different work steps or machine tool states, the continuous energy data under different states are divided based on the machine tool energy consumption state characteristic curve. Since the power in the air cutting state is generally low, it can be used to determine whether the machine tool is in the cutting state. The functional data of the cutting state is used to identify different work steps. In order to eliminate the influence of random factors on the energy consumption of the machine tool, 10 power data are collected each time, that is, Then, delete the maximum and minimum values, and the remaining 8 data The average value is taken as the judgment standard, that is,

[0077]

[0078] The energy consumption data segmentation process is as follows: Figure 2 As shown, specifically including:

[0079] Step 1: According to the process planning, obtain the start time ptws of a certain process i and end time ptwe i ;

[0080] Step 2: The start time of the jth step of this process is ptws i,j ,

[0081] Step 3: Obtain the average energy efficiency value MeanCE according to formula (5) t ;

[0082] Step 4: If The machine tool enters the cutting state. The power of the cutting process is if The current state is still empty.

[0083] Step 5: If The machine tool enters another cutting process, and the end time of the current step is ptwe i,j That is the start time of the next step ptws i,j+1 ;

[0084] Step 6: Repeat the operations from step 3 to step 5 until the processing is completed. The time data of the process step is:

[0085] ST i ={PID,PName,ptws i,j ,ptwe i,j} (6)

[0086] Step 3: Manufacturing process energy consumption characteristic data fusion processing method

[0087] For a machining process, the original energy consumption data cannot be directly used to evaluate or optimize the machining process. Important characteristic values ​​must be derived. These characteristic values ​​are mainly divided into two categories: instantaneous and cumulative.

[0088] The instantaneous characteristics mainly reflect the changing law of cutting power, such as the maximum, minimum and average power of a process or feed rate, as follows:

[0089]

[0090]

[0091]

[0092]

[0093] Among them, MaxP i Indicates the maximum energy consumption of this process, MinP i Indicates the minimum energy consumption, MeanP i Represents the average energy consumption, VarP i Represents the standard deviation of energy consumption, N i Indicates the number of energy consumption data;

[0094] The instantaneous characteristic variables are shown in equation (11). These instantaneous characteristics can reflect the operating status of the manufacturing system and are used to detect hidden production anomalies.

[0095] IEC i ={ptws i ,ptwe i ,MaxP i ,MinP i ,MeanP i ,VarP i} (11)

[0096] On the other hand, the cumulative characteristics mainly show the overall energy consumption and energy efficiency of a process. These characteristics can be obtained as follows:

[0097]

[0098]

[0099]

[0100]

[0101] Among them, TotalE i Indicates the total energy consumption of the process, TotalSMD i Indicates the total auxiliary material consumption of the process, EnergyEff i Indicates the energy utilization rate of the process, mres i Indicates the start time of the cutting state of this process, mree i Indicates the end time of the cutting state of this process, ProcessEff i Represents the energy efficiency of the process, V i Indicates the volume of the workpiece removed in this process;

[0102] The cumulative energy consumption characteristic variable is:

[0103] CEC i ={ptws i ,ptwe i ,TotalEi ,TotalSMD i ,EnergyEff i ,PorcessEff i} (16)

[0104] After completing the above feature data, fuse it with the process time data, and use the relational operation method to obtain the process energy consumption fusion data:

[0105]

[0106] The fused data can be used for energy consumption big data analysis, production process failure analysis, etc.

[0107] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0108] In the specific implementation manner of the present invention, the parts not mentioned in the description belong to the common knowledge in the art and can be implemented with reference to the common knowledge.

Claims

1. A method for preprocessing and analyzing big data on energy consumption in smart manufacturing workshops. It is characterized in that include: Step 1: Modeling the energy consumption related data of the manufacturing process, wherein the energy consumption related data includes process data, processing energy consumption data, processing logistics energy consumption data and auxiliary material consumption data; Step 2: Associating energy consumption data with process data, and dividing energy consumption data according to processing or feed rate; specifically, dividing continuous energy data under different states based on the energy consumption state characteristic curve of the machine tool, and identifying different work steps according to the power data changes of the cutting state; Each time 10 power data are collected, that is, Delete the maximum and minimum values ​​and obtain the remaining 8 data The average value of , i = 1, ..., 8, takes the average power as the judgment standard: The energy consumption data segmentation process specifically includes: Step 2.1: According to the process plan, obtain the start time ptws of the i-th process i and end time ptwe i ; Step 2.2: The start time of the jth step of the process is ptws i,j , Step 2.3: Obtain the average power MeanCE according to formula (5) t ; Step 2.4: If The machine tool enters the cutting state, and the power of the cutting process is if Currently still in the empty cutting state, P aircutting Indicates the machine tool air cutting power; Step 2.5: If The machine tool enters another cutting process, and the end time of the current step is ptwe i,j , which is the start time of the next step ptws i,j+1 ; Step 2.6: Repeat steps 2.3 to 2.5 until the processing is completed. The time data of all steps is: ST i ={PID,PName,ptws i,j ,ptwe i,j } (6); Step 3: Fuse the instantaneous characteristic value reflecting the changing law of cutting power, the cumulative energy consumption characteristic value reflecting the total energy consumption and energy efficiency with the process time data.

2. According to the method for preprocessing and analyzing big data of energy consumption of intelligent manufacturing workshops according to claim 1, It is characterized in that The process data modeling formula is as follows: PD={PDID, PID, PName, WID, WName, MID, ptws, ptwe} (1) Among them, PD represents process data, PDID represents the ID of process data, PID represents process ID, PName represents process name, WID represents the workpiece number ID, WName represents the workpiece name, MID represents machine tool ID, ptws and ptwe represent the start time and end time of the process respectively.

3. According to claim 1, the method for preprocessing and analyzing big data of energy consumption in intelligent manufacturing workshops, It is characterized in that The processing energy consumption data modeling formula is as follows: MECD={MECDID,MID,p,T} (2) Among them, MECD represents the machine tool energy consumption data, MECDID represents the machine tool energy consumption data ID, MID represents the machine tool ID, p represents the machine tool power value, and T represents time.

4. According to claim 1, the method for preprocessing and analyzing big data of energy consumption in intelligent manufacturing workshops, It is characterized in that The modeling formula for the processing logistics energy consumption data is as follows: LECD={LECDID, PID, LType, LEData, T} (3) Among them, LECD represents the energy consumption of logistics equipment, LECDID represents the energy consumption data ID, LType represents the equipment type, LEData represents the energy consumption, PID represents the process ID, and T represents time.

5. According to claim 1, the method for preprocessing and analyzing big data of energy consumption in intelligent manufacturing workshops, It is characterized in that The auxiliary material consumption data modeling formula is as follows: SMD={SMDID, PID, MID, SType, SMData, T} (4) Among them, SMD represents auxiliary material consumption data, SMDID represents auxiliary material consumption ID, SType represents auxiliary material type, SMData represents auxiliary material consumption, MID represents machine tool ID, PID represents process ID, and T represents time.

6. The method for preprocessing and analyzing big data on energy consumption of intelligent manufacturing workshops according to claim 1, It is characterized in that The instantaneous characteristic value in step 3 is constructed according to the statistical characteristics, specifically including: Among them, MaxP i Indicates the maximum energy consumption of the process, MinP i Indicates the minimum energy consumption, MeanP i Represents the average energy consumption, VarP i Represents the standard deviation of energy consumption, N i Indicates the number of energy consumption data; The instantaneous eigenvalue modeling formula is as follows: IEC i ={ptws i ,ptwe i ,MaxP i ,MinP i ,MeanP i ,VarP i } (11)。 7. The method for preprocessing and analyzing big data on energy consumption of intelligent manufacturing workshops according to claim 1, It is characterized in that In step 3, the total energy consumption and energy efficiency formula of the process are as follows: Among them, TotalE i Indicates the total energy consumption of the process, TotalSMD i Indicates the total auxiliary material consumption of the process, EnergyEff i Indicates the energy utilization rate of the process, mres i Indicates the start time of the process cutting state, mree i Indicates the end time of the process cutting state, ProcessEff i Represents the energy efficiency of the process, V i Indicates the volume of the workpiece removed during the process; The cumulative energy consumption characteristic value modeling formula is as follows: CEC i ={ptws i ,ptwe i ,TotalE i ,TotalSMD i ,EnergyEff i ,PorcessEff i } (16)。 8. The method for preprocessing and analyzing big data on energy consumption of intelligent manufacturing workshops according to claim 1, It is characterized in that In step 3, the instantaneous characteristic value reflecting the change law of cutting power, the cumulative energy consumption characteristic value reflecting the overall energy consumption and energy efficiency, and the process time data are fused by using the relational operation method to obtain the process energy consumption fusion data: Among them, PECD i The process energy consumption fusion data representing the process, Represents a concatenation operation.

9. A computer-readable storage medium, It is characterized in that The method comprises one or more programs for execution by one or more processors, wherein the one or more programs comprise instructions for executing the method according to any one of claims 1 to 8.