CN111667186A - Method and device for determining the energy consumption of a machine for production - Google Patents

Method and device for determining the energy consumption of a machine for production Download PDF

Info

Publication number
CN111667186A
CN111667186A CN202010522303.5A CN202010522303A CN111667186A CN 111667186 A CN111667186 A CN 111667186A CN 202010522303 A CN202010522303 A CN 202010522303A CN 111667186 A CN111667186 A CN 111667186A
Authority
CN
China
Prior art keywords
energy consumption
matrix
determining
target
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010522303.5A
Other languages
Chinese (zh)
Other versions
CN111667186B (en
Inventor
吴远沈
黄铮
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Siemens Power Automation Ltd
Original Assignee
Siemens Power Automation Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Siemens Power Automation Ltd filed Critical Siemens Power Automation Ltd
Priority to CN202010522303.5A priority Critical patent/CN111667186B/en
Publication of CN111667186A publication Critical patent/CN111667186A/en
Application granted granted Critical
Publication of CN111667186B publication Critical patent/CN111667186B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/067—Enterprise or organisation modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/04—Manufacturing
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention relates to a method and a device for determining the energy consumption of a machine for production, the method comprising: acquiring a plurality of original energy consumption data corresponding to production in a target time period, wherein the production in the target time period comprises a plurality of process steps, and the original energy consumption data are machine energy consumption corresponding to a machine which executes the process steps and is acquired by periodic sampling; determining energy consumption modes according to the original energy consumption data, wherein each energy consumption mode corresponds to one process step; and determining the machine energy consumption within a preset time period after each process step is adjusted according to each energy consumption mode.

Description

Method and device for determining the energy consumption of a machine for production
Technical Field
The invention relates to the field of production, in particular to a method and a device for determining energy consumption of a machine for production.
Background
The production refers to the whole process from the input of raw materials to the output of products, and the process comprises a plurality of process steps, such as cutting, grinding, polishing and the like, and the process steps involved in the production are different according to the actual needs of each product. Each process step requires a corresponding machine to perform, which generates machine energy consumption.
Sometimes, an enterprise needs to adjust process steps to optimize production, for example, to remove certain process steps, and this time, an evaluation is often needed in advance to know the energy consumption of the machine after adjusting the process steps, so as to determine whether the adjustment is needed.
Disclosure of Invention
In view of this, the present invention proposes a method of determining the energy consumption of a machine for production, comprising:
acquiring a plurality of original energy consumption data corresponding to production in a target time period, wherein the production in the target time period comprises a plurality of process steps, and the original energy consumption data are machine energy consumption corresponding to a machine which executes the process steps and is acquired by periodic sampling;
determining energy consumption modes according to the original energy consumption data, wherein each energy consumption mode corresponds to one process step;
and determining the machine energy consumption within a preset time period after each process step is adjusted according to each energy consumption mode.
According to the method as described above, optionally, determining the energy consumption modes from the raw energy consumption data comprises:
determining a target matrix according to the original energy consumption data, wherein the column number of the target matrix is obtained according to the duration corresponding to at least one process step;
determining a first matrix according to the target matrix;
acquiring a feature vector matrix of the first matrix;
and determining an energy consumption matrix according to the characteristic vector matrix and the target matrix, and determining each energy consumption mode according to data of each row of the energy consumption matrix.
According to the method, optionally, the number of columns of the target matrix is the number of samples that can be sampled in the average value of the time lengths corresponding to a plurality of process steps.
According to the method as described above, optionally, determining an objective matrix from the raw energy consumption data comprises:
determining the final column number t' of the target matrix;
determining a sequence x ═ { x ] from the raw energy consumption data1,x2,…xi,…,xn’In which xiRepresenting ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n ' is the maximum sampling number which satisfies that n '/t ' is an integer in a target time period;
determining a target based on the final number of columns tMatrix array
Figure BDA0002532531100000021
According to the method, optionally, the final column number t' of the target matrix is the number that can be sampled in the average value of the time lengths corresponding to a plurality of process steps.
According to the method as described above, optionally, determining an objective matrix from the raw energy consumption data comprises:
determining an initial column number t of a target matrix;
determining an initial sequence x ═ { x ] from the raw energy consumption data1,x2,…xi,…,xnIn which xiRepresenting ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n is the maximum sampling number which satisfies n/t as an integer in a target time period;
determining an initial matrix according to the initial column number t
Figure BDA0002532531100000022
Performing a centralization operation on each original energy consumption data in the initial sequence, wherein the centralization operation comprises: obtaining an average value of the original energy consumption data of each row in the initial matrix, and subtracting the average value of the corresponding row from each original energy consumption data;
determining a secondary matrix XX from the centred initial matrixT；
For the second matrix XXTDecomposing the eigenvalues to obtain a plurality of eigenvalues, and determining corresponding variance occupancy according to the plurality of eigenvalues;
updating the value of the initial column number t and returning to execute the operation of determining an initial matrix according to the initial column number t until the value of the initial column number t is updated to a preset number;
determining an inflection point according to each corresponding variance occupancy and the initial column number t, determining the number t 'of samples which can be sampled in time corresponding to the inflection point, and taking the t' as the final column number of the target matrix;
determining an object matrix
Figure BDA0002532531100000031
n ' is the maximum number of samples for the target time period for which n '/t ' is an integer.
According to the method as described above, optionally, determining a first matrix according to the target matrix comprises:
performing a centralization operation on each original energy consumption data in the target matrix, wherein the centralization operation comprises the following steps: obtaining an average value of the original energy consumption data of each row in the target matrix, and subtracting the average value of the corresponding row from each original energy consumption data;
according to the centralized target matrix XmodeDetermining the first matrix
Figure BDA0002532531100000032
Obtaining a feature vector matrix of the first matrix comprises:
performing eigenvalue decomposition on the first matrix;
sequentially sorting the corresponding eigenvectors according to the descending order of the eigenvalues, and generating an eigenvector matrix W [ W ]1；w2；…；wj；…wk]Wherein w isjJ is more than or equal to 1 and less than or equal to k, and each row of the eigenvector matrix is a value corresponding to one eigenvector;
determining an energy consumption matrix according to the eigenvector matrix and the target matrix comprises:
modes＝W*Xmodewhere mods is the power consumption matrix.
According to the method as described above, optionally, determining each energy consumption mode according to the data of each row of the energy consumption matrix includes:
determining an energy consumption mode corresponding to each process step according to each row of data in the energy consumption matrix;
according to each energy consumption mode, determining the machine energy consumption within a preset time period after each process step is adjusted comprises the following steps:
determining a target mode from various energy consumption modes;
and determining the machine energy consumption within the preset time period after the target mode is removed.
According to the method described above, optionally, determining the target mode from the various energy consumption modes comprises: taking the last h energy consumption modes determined according to the energy consumption matrix as target modes, wherein h is more than or equal to 1 and is less than k;
determining the machine energy consumption within the preset time period after the target pattern is removed comprises:
changing the last h rows of data in the models into 0 as energy consumption matrix models after adjusting the process stepsnewAnd determining the machine energy consumption in the preset time period according to the energy consumption matrix after the process step is adjusted.
According to the method described above, optionally, determining the machine energy consumption within the preset time period according to the energy consumption matrix after the adjusting process step includes:
according to W-1*modesnewAnd the sampling frequency determines the machine energy consumption within a preset time period.
According to the method, optionally, determining an energy consumption mode corresponding to each process step according to each row of data in the energy consumption matrix includes at least one of the following ways:
the first method is as follows: discarding a line of data which cannot be matched with the process steps;
the second method comprises the following steps: and determining a row of data corresponding to the process steps one by one as an energy consumption mode.
In another aspect, the present invention provides an apparatus for determining energy consumption of a machine for production, comprising:
the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a plurality of original energy consumption data corresponding to production in a target time period, the production in the target time period comprises a plurality of process steps, and the original energy consumption data is the machine energy consumption corresponding to a machine which executes the process steps and is acquired by periodic sampling;
a first determining unit for determining energy consumption modes based on said raw energy consumption data, each energy consumption mode corresponding to a process step;
and the second determining unit is used for determining the machine energy consumption within a preset time period after each process step is adjusted according to each energy consumption mode.
According to the apparatus as described above, optionally, the first determining unit includes:
a first determining subunit, configured to determine an objective matrix according to the original energy consumption data, where the number of columns of the objective matrix is obtained according to a duration corresponding to at least one process step;
a second determining subunit, configured to determine a first matrix according to the target matrix;
a first obtaining subunit, configured to obtain a feature vector matrix of the first matrix;
and the third determining subunit is used for determining an energy consumption matrix according to the eigenvector matrix and the target matrix and determining each energy consumption mode according to data of each row of the energy consumption matrix.
According to the apparatus as described above, optionally, the first determining subunit is specifically configured to:
determining the final column number t 'of the target matrix, wherein the final column number t' of the target matrix is the number of samples which can be sampled in the average value of the time lengths corresponding to the multiple process steps;
determining a sequence x ═ { x ] from the raw energy consumption data1,x2,…xi,…,xn’In which xiRepresenting ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n ' is the maximum sampling number which satisfies that n '/t ' is an integer in a target time period;
determining a target matrix according to the final column number t
Figure BDA0002532531100000041
According to the apparatus as described above, optionally, the first determining subunit is specifically configured to:
determining an initial column number t of a target matrix;
determining an initial sequence x ═ { x ] from the raw energy consumption data1,x2,…xi,…,xnIn which xiRepresenting ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n is the maximum sampling number which satisfies n/t as an integer in a target time period;
determining an initial matrix according to the initial column number t
Figure BDA0002532531100000051
Performing a centralization operation on each original energy consumption data in the initial sequence, wherein the centralization operation comprises: obtaining an average value of the original energy consumption data of each row in the initial matrix, and subtracting the average value of the corresponding row from each original energy consumption data;
determining a secondary matrix XX from the centred initial matrixT；
For the second matrix XXTDecomposing the eigenvalues to obtain a plurality of eigenvalues, and determining corresponding variance occupancy according to the plurality of eigenvalues;
updating the value of the initial column number t and returning to execute the operation of determining an initial matrix according to the initial column number t until the value of the initial column number t is updated to a preset number;
determining an inflection point according to each corresponding variance occupancy and the initial column number t, determining the number t 'of samples which can be sampled in time corresponding to the inflection point, and taking the t' as the final column number of the target matrix;
determining an object matrix
Figure BDA0002532531100000052
n ' is the maximum number of samples for the target time period for which n '/t ' is an integer.
According to the apparatus as described above, optionally, the second determining subunit is specifically configured to:
performing a centralization operation on each original energy consumption data in the target matrix, wherein the centralization operation comprises the following steps: obtaining an average value of the original energy consumption data of each row in the target matrix, and subtracting the average value of the corresponding row from each original energy consumption data;
according to the centralized target matrix XmodeDetermining the first matrix
Figure BDA0002532531100000053
The first obtaining subunit is specifically configured to:
performing eigenvalue decomposition on the first matrix;
sequentially sorting the corresponding eigenvectors according to the descending order of the eigenvalues, and generating an eigenvector matrix W [ W ]1；w2；…wj；…；wk]Wherein w isjJ is more than or equal to 1 and less than or equal to k, and each row of the eigenvector matrix is a value corresponding to one eigenvector;
the third determining subunit is specifically configured to:
modes＝W*Xmodewhere mods is the power consumption matrix.
According to the apparatus as described above, optionally, the third determining subunit is specifically configured to:
determining an energy consumption mode corresponding to each process step according to each row of data in the energy consumption matrix;
the second determining unit is specifically configured to:
determining a target mode from various energy consumption modes;
and determining the machine energy consumption within the preset time period after the target mode is removed.
According to the apparatus as described above, optionally, the second determining unit is further specifically configured to:
taking the last h energy consumption modes determined according to the energy consumption matrix as target modes, wherein h is more than or equal to 1 and is less than k;
changing the last h-row data in the models into 0 as an adjustment process stepPost-step energy consumption matrix modelsnewAnd according to W-1*modesnewAnd the sampling frequency determines the machine energy consumption within a preset time period.
The invention also provides an apparatus for determining the energy consumption of a machine for production, comprising:
at least one memory for storing instructions;
at least one processor configured to execute a method of determining machine energy consumption for production in accordance with any of the above in accordance with instructions stored in the memory.
The invention further provides a readable storage medium having stored therein machine readable instructions which, when executed by a machine, perform a method of determining machine energy consumption for production according to any of the above.
According to the invention, by acquiring a plurality of original energy consumption data corresponding to production in a target time period, analyzing the energy consumption of the original data to acquire the energy consumption mode corresponding to each process step, and then acquiring the machine energy consumption in the target time period after the process steps are adjusted through the energy consumption models, simulation evaluation can be carried out before a new production scheme is realized, so as to avoid high investment and low output.
Drawings
The foregoing and other features and advantages of the invention will become more apparent to those skilled in the art to which the invention relates upon consideration of the following detailed description of a preferred embodiment of the invention with reference to the accompanying drawings, in which:
FIG. 1 is a schematic flow diagram of a method of determining energy consumption of a machine for production according to an embodiment of the present invention.
FIG. 2A is a flow diagram illustrating a method for determining energy consumption of a machine for production according to another embodiment of the present disclosure.
FIG. 2B is a diagram illustrating a variance occupancy curve according to another embodiment of the present invention.
FIG. 3 is a graph illustrating machine energy consumption before and after adjustment of process steps.
Fig. 4 is a schematic structural diagram of an apparatus for determining energy consumption of a machine for production according to an embodiment of the present invention.
Fig. 5 is a schematic structural diagram of an apparatus for determining energy consumption of a machine for production according to another embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail by referring to the following examples.
For production, one or more machines are often used. Each machine may perform a process step or multiple process steps. Some of these multiple process steps may be redundant and elimination of these redundant steps is desirable to optimize the overall production. Therefore, the entire production after optimization needs to be evaluated before the actual optimization scheme is implemented to avoid high input and low output. The invention provides a method and a device for determining the energy consumption of a machine for production, which can evaluate the energy consumption of the machine after adjusting a process step.
Example one
The present embodiments provide a method of determining energy consumption of a machine for production. The production comprises a plurality of process steps, for example comprising several, dozens or even more process steps. The process steps of the target time period described below may also include a plurality of process steps, which may be all or a part of the process steps that should be included in one production, and the specific requirement is determined according to the target time period. The main execution body of the scheme is a device for determining the energy consumption of the machine for production, and the device can be integrated in a personal computer or independently arranged, and is not described in detail herein.
Step 101, obtaining a plurality of original energy consumption data corresponding to production in a target time period, wherein the original energy consumption data is machine energy consumption corresponding to a machine which executes a process step and is obtained through periodic sampling.
Each machine has an energy consumption, i.e. the machine energy consumption, when it performs a process step. In this embodiment, the energy consumption of the machine may be obtained through periodic sampling, and specifically, the corresponding energy consumption of the machine may be calculated by collecting data such as current and voltage, and of course, other manners may also be adopted to obtain the energy consumption of the machine, which is not specifically described again. The sampling frequency can be determined according to actual needs, for example, sampling every 1 minute.
The content of the plurality of process steps included within the target time period may be determined empirically. Alternatively, the raw energy consumption data in the target time period may be cleaned, that is, some abnormal values, such as a value less than 0 or an outlier, are removed, and the cleaned raw energy consumption data is used as the raw energy consumption data required in this step 101.
The target time period can be set according to actual needs. Typically, the target time period is required to be long enough to cover enough process steps for subsequent analysis, for example, 24 hours.
Step 102, determining energy consumption modes from the raw energy consumption data, each energy consumption mode corresponding to a process step.
This step determines the energy consumption pattern by analyzing the raw energy consumption data. Each energy consumption pattern is for one process step, that is, the step is for obtaining the energy consumption pattern for each process step.
Each process step corresponds to an energy consumption mode, and the energy consumption of the process step in a certain preset time period is obtained according to the energy consumption mode.
And 103, determining the machine energy consumption within a preset time period after each process step is adjusted according to each energy consumption mode.
Adjusting process steps herein may refer to removing one or more process steps that are desired to be removed, thereby determining how much machine power is saved over a predetermined period of time. For example, after determining the energy consumption mode corresponding to each process step, redundant process steps are removed. In this way, the corresponding machine energy consumption can be determined from the remaining raw energy consumption data. Or, the new production line may include some process steps, and the corresponding machine energy consumption in the preset time period may be determined according to the energy consumption modes corresponding to the some process steps. The preset time period of this embodiment may be determined according to actual needs, for example, less than or equal to the target time period, specifically, for example, the target time period is 24 hours, and the preset time period may be 30 minutes.
Specifically, the process steps that need to be removed may be determined according to the weight sequence corresponding to each energy consumption mode, for example, the process step with the smallest weight is the redundant step. The weight order refers to an order of proportions occupied by the energy consumption patterns, and for example, the weights of the energy consumption patterns are arranged in order of ascending to descending. Of course, some process steps or process steps that are desired to be removed may also be removed based on the expertise of the expert, and will not be described in detail.
According to the invention, by acquiring a plurality of original energy consumption data corresponding to production in a target time period, analyzing the energy consumption of the original data to acquire the energy consumption mode corresponding to each process step, and then acquiring the machine energy consumption in the target time period after the process steps are adjusted through the energy consumption models, simulation evaluation can be carried out before a new production scheme is realized, so as to avoid high investment and low output.
Example two
This example further illustrates the method of example one for determining energy consumption of a machine for production. Since the order of the process steps within the target time period is not fixed, as shown in fig. 2A, it is a schematic flow chart of the method for determining the energy consumption of the machine for production according to the present embodiment. The method for determining the energy consumption of the machine for production comprises the following steps:
step 201, obtaining a plurality of original energy consumption data corresponding to production in a target time period.
The raw energy consumption data is the machine energy consumption corresponding to the machine performing the process step obtained by periodic sampling. The step 201 may be specifically consistent with the step 101, for example, the original energy consumption data is cleaned, and details are not described herein.
In step 202, an objective matrix is determined according to the raw energy consumption data.
Wherein the number of columns of the target matrix may be obtained according to a duration corresponding to at least one process step. Specifically, the number of columns of the target matrix is, for example, the number that can be sampled in the average value of the time lengths corresponding to a plurality of process steps. The possible process steps within the target time may be determined empirically, for example, assuming the target time is 24 hours, then how many process steps may pass within the 24 hours may be determined from the history, and the average of the durations of these process steps may be divided by the sampling period to obtain the number of columns of the target matrix. Of course, the number of columns of the target matrix may also be determined based on expert experience, i.e., how much a possible pattern duration should be determined based on expert experience. The number of columns of the target matrix may be determined in other ways, such as by one of a number of process steps that may correspond to the number of columns of the target matrix. Of course, the information can also be obtained by other methods, which are not described in detail herein. In the case that the number of columns of the target matrix is determined, the number of rows may be determined according to the number of original energy consumption data in the target time period obtained in step 201, for example, the original energy consumption data obtained in step 201 are sequentially filled into a blank target matrix, and until the remaining original energy consumption data cannot fill one row, the remaining original energy consumption data are discarded.
Alternatively, the target matrix may be obtained according to the following manner:
step S1: the final number of columns t' of the target matrix is determined.
The final column number t' may be the number of samples that can be taken in the average of the time periods corresponding to the plurality of process steps.
Step S2: determining a sequence x ═ { x ] from the raw energy consumption data1,x2,…xi,…,xn’In which xiAnd representing the ith original energy consumption data, i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n ' is the maximum sampling number which satisfies n '/t ' as an integer in the target time period.
The number of raw energy consumption data in the sequence x is required to form a matrix with the column number t'. Thus, it is possible to discard several original energy consumption data, for example, the last several original energy consumption data. That is, the largest number of samples that can satisfy the integer of n 'divided by t' can be selected in time order.
Step S3: determining an object matrix
Figure BDA0002532531100000091
n ' is the maximum number of samples for the target time period for which n '/t ' is an integer.
Another way to determine the final number of columns of the destination matrix is described below, as in the method shown in fig. 2B.
In step S11, the initial number of columns t of the target matrix is determined.
The initial number of columns t here can be determined empirically by an expert, for example by the expert on the basis of the duration of one or more process steps. In the method shown in fig. 2B, it is necessary to determine values of a plurality of initial column numbers t, for example, the initial column numbers may be sequentially determined as 5, 6, 7, 8, and so on until the number of values of the initial column number t reaches a preset number.
Of course, in order to reduce the number of values of the initial column number t, the number of times that at least one process step can be sampled within the duration of one or more process steps may be determined according to the duration of the at least one process step, for example, as each initial column number t.
Step S12, determining an initial sequence x ═ { x } according to the original energy consumption data1,x2,…xi,…,xnIn which xiAnd representing the ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n is the maximum sampling number which satisfies n/t as an integer in the target time period.
The number of raw energy consumption data in the initial sequence x is required to form a matrix with the number of columns t. Thus, it is possible to discard several original energy consumption data, for example, the last several original energy consumption data. That is, the largest number of samples that can satisfy the integer division of n by t can be selected in time order.
Step S13, determining an initial matrix according to the initial column number t
Figure BDA0002532531100000101
And step S14, performing a centralization operation on each original energy consumption data in the initial sequence.
The centralizing operation includes: and acquiring the average value of the original energy consumption data of each row in the initial matrix, and subtracting the average value of the corresponding row from each original energy consumption data.
Specifically, in step S14, the average value of the original energy consumption data of each row in the initial matrix may be obtained, and the average value of the corresponding row is subtracted from each original energy consumption data. For example, p1＝(x1+x2…+xt) T, then centered x1Becomes x1-p1. Thus, if pi＝(xi+xi+1…+xi+t)/t，piI is more than or equal to 1 and less than or equal to n-t, then the data of the ith row becomes xi-pi，xi+1-pi，…，xi+t-piThen the initial matrix X after centering becomes:
Figure BDA0002532531100000102
step S15, determining a secondary matrix XX according to the centralized initial matrixT。
XTIs the transpose of the initial matrix X.
Step S16, for the second matrix XXTAnd decomposing the eigenvalues to obtain a plurality of eigenvalues, and determining the corresponding variance occupation ratio according to the plurality of eigenvalues.
The variance occupation ratio probability of variance exposure can be obtained according to the following formula:
Figure BDA0002532531100000111
wherein λ is1Is the maximum eigenvalue, λqRepresenting the qth eigenvalue.
In step S17, the value of the initial column count t is updated and the process returns to step S13 until the value of the initial column count t is updated to a predetermined number of times.
Each time the initial column number t is updated, a variance occupancy is corresponded. Fig. 2B is a diagram showing a relationship between time and a variance occupancy rate, and the initial column number t is a sampling number, which also corresponds to a sampling time, which is an abscissa in fig. 2B. For example, when the sampling frequency is 1 minute and 1 time, and the number of initial columns is 11, the time on the abscissa in fig. 2B is 11 min.
The preset times of the embodiment can be set according to actual needs.
In step S18, an inflection point is determined from the variance occupancy and the initial column number t, and the number t 'of samples that can be sampled within a time period corresponding to the inflection point is determined, and this t' is used as the final column number of the objective matrix.
Determining the inflection point using the occupancy of variance can make the determined inflection point more accurate. The inflection point here refers to a point at which the curve corresponding to the variance occupancy becomes gradually smooth. FIG. 2B is a graph illustrating different occupancy of variance for different Ts. For example, when the initial column number T determines a value, which corresponds to a variance occupancy, and the initial column number T indicates how many samples of power consumption data can be obtained for each row, the corresponding sampling time T is determined. More specifically, the sampling time is 1 minute and 1 time, T is 30, and one row has 30 sampling data, which corresponds to T being 30 min. That is, in addition to the above method for obtaining the initial number of columns t, the final number of columns t' may be determined by sequentially taking values of t as shown in fig. 2B. As shown in fig. 2B, the point corresponding to T of 30min is an inflection point, and the curve of the variance occupancy after that approaches the level, where T represents time. Fig. 2B shows T ═ 30min, that is, T ═ 30.
Step S19, determining an object matrix
Figure BDA0002532531100000112
n ' is the maximum number of samples for the target time period for which n '/t ' is an integer.
Because the number of columns of the target matrix is determined again, the number of the original energy consumption data included in the target matrix may need to be adjusted again, and may be selected from the original energy consumption data in step 201.
Step 203, a first matrix is determined according to the target matrix.
The step may specifically include: performing centralization operation on each original energy consumption data in the target matrix, and determining a first matrix according to the centralized target matrix
Figure BDA0002532531100000121
The specific steps of the centralization operation may be: and acquiring the average value of the original energy consumption data of each row in the target matrix, and subtracting the average value of the corresponding row from each original energy consumption data. The centering operation is the same as step S4, and is not described in detail herein.
Figure BDA0002532531100000122
That is XmodeThe transposed matrix of (2). In this embodiment, the ground matrix after the centering operation is still marked as Xmode。
Step 204, a feature vector matrix of the first matrix is obtained.
The step 201 may specifically include:
performing eigenvalue decomposition on the first matrix;
sequentially ordering the corresponding eigenvectors according to the descending order of the eigenvalues, and generating an eigenvector matrix set W ═ W1；w2；…wj；…；wk]Wherein w isjJ is more than or equal to 1 and less than or equal to k, and represents the j-th feature vector matrix after sorting. Each row in the eigenvector matrix is a value corresponding to an eigenvector. For example,
Figure BDA0002532531100000123
the first row of data in the eigenvector matrix is w11,w12,…,w1t，. That is, each of the feature vector matricesThe rows correspond to the values corresponding to each ordered feature vector in turn, i.e. a column of values that is to be erected is spread out laterally.
How to perform eigenvalue decomposition on a matrix belongs to the prior art, and is not described herein again. And (4) sequentially arranging the characteristic values from large to small, namely sequentially arranging the energy consumption modes corresponding to the process steps from large to small according to the weight.
Step 205, determining an energy consumption matrix according to the eigenvector matrix and the target matrix, and determining each energy consumption mode according to data of each row of the energy consumption matrix.
Thus, the resulting energy consumption matrix may be: mode ═ W × Xmode。
Specifically, an energy consumption mode corresponding to each process step can be determined according to each row of data in the energy matrix. And regarding each row of data in the energy consumption matrix as an energy consumption mode, wherein the energy consumption modes may correspond to the process steps one by one, or may not be matched with the energy consumption modes of the process steps, and the process steps which cannot be matched with the energy consumption modes can be removed.
Determining an energy consumption mode corresponding to each process step according to each row of data in the energy consumption mode comprises at least one of the following modes:
the first method is as follows: and abandoning the energy consumption mode which can not match with the process steps.
The second method comprises the following steps: and determining the energy consumption modes corresponding to the process steps one by one.
The magnitude of the eigenvalue can be regarded as the importance degree of the energy consumption mode, and the eigenvalue corresponds to the eigenvector representing the direction.
And step 206, determining a target mode from the energy consumption modes, and determining the machine energy consumption within a preset time period after the target mode is removed.
The target mode here is the power consumption mode that is desired to be eliminated. For example, the last h energy consumption patterns determined from the energy model are taken as target patterns, where 1 ≦ h < k. That is, the latter h rows of data in the energy consumption matrix models are all changed into 0 as the energy consumption matrix models after the adjustment processnewAccording to the energy consumption momentThe array can acquire the corresponding machine energy consumption. Because the feature vectors are sequentially arranged according to the sequence of the proportion from large to small, the corresponding feature vectors can be removed according to the actual requirement, for example, the feature vectors arranged at the back are removed, which is equivalent to the energy consumption modes arranged at the back. Of course, the models can be used according to actual needsnewEach energy consumption mode in (1) is first obtained, e.g., for energy consumption mode modelsnewEach row corresponds to a power consumption mode, and the matching process step can be determined according to the curve of one row, so that the feature vector which is to be removed can be known.
It can be understood here that, assuming that 10 process steps are required to be performed to produce one product a, a new product B is required to be produced, the product B needs 5 process steps, and the 5 process steps are included in the 10 process steps corresponding to the product a, so that the machine energy consumption for producing the new product B can be simulated according to the historical data of the product a, and further, the advance evaluation can be performed.
The machine energy consumption matrix herein may be based on W-1*modesnewAnd the corresponding sampling frequency. Wherein, W-1Is the inverse matrix of W.
It can be understood that, as for the energy consumption data in the matrix, the energy consumption data are sequentially arranged according to a time sequence, and accordingly, each energy consumption data in the energy consumption matrix can also be regarded as being sequentially arranged according to a time sequence, and the corresponding energy consumption data in a certain time period can be obtained, so that the energy consumption after the process step is adjusted can be determined.
According to the embodiment, a corresponding matrix is established according to the original energy consumption data, the energy consumption matrix is finally obtained, the energy consumption modes corresponding to the corresponding process steps are further obtained, the process steps needing to be removed can be determined, and the machine energy consumption after the process steps are removed is obtained, so that the advance evaluation is carried out, and the high input and low output are avoided.
EXAMPLE III
This example further specifically illustrates the method of determining the energy consumption of the production process of the previous example.
In this embodiment, a production process of a relay protection device includes: cutting, punching, grinding, polishing, wiping, soot blowing, flushing, assembling, packaging and the like. The target time period is 24 hours. The raw energy consumption data was sampled 1 time in 1 minute, and thus 1440 raw energy consumption data were acquired.
First, each raw energy consumption data in the target time period is obtained, x ═ x1,x2,…,x1440}. From the historical data, it was determined that the average of the durations corresponding to the 18 process steps was approximately 10 minutes.
Thus, the initial matrix
Figure BDA0002532531100000141
Performing centralized operation on each original energy consumption data: obtaining the average value p of each original energy consumption data of each linei，pi＝(xi+xi+1…+xi+9)/10。
Thus, the initial matrix X becomes
Figure BDA0002532531100000142
Generating a secondary matrix XX from the centered initial matrixT. For the second matrix XXTDecomposing the eigenvalue to obtain the maximum eigenvalue lambda1And calculating the occupancy of variance
Figure BDA0002532531100000143
And then, reselecting the column number of the initial matrix X, and selecting an inflection point according to a curve corresponding to each difference occupancy rate after reaching a preset number of times. Let the inflection point t' be 12.
Then the target matrix
Figure BDA0002532531100000144
Centralizing each energy consumption data in the target matrix, namely acquiring the target matrix XmodeOf each rowAverage and let each data subtract the average of the row. Let the average value of jth line data be p'jWherein j is more than or equal to 1 and less than or equal to 120, the centralized target matrix is:
Figure BDA0002532531100000145
determining a first matrix from the centralized target matrix
Figure BDA0002532531100000147
Decomposing the eigenvalue of the first matrix, sequencing the corresponding eigenvectors in turn according to the descending order of the eigenvalue, and generating an eigenvector matrix W [ W ]1；w2；…wi；…；wk]Wherein w isiAnd representing the sorted ith characteristic vector, wherein 1 is less than or equal to i and less than or equal to k. The data of each row of W in the eigenvector matrix is spread out laterally for the data in one eigenvector.
Determining an energy consumption matrix from the eigenvector matrix comprises:
Figure BDA0002532531100000146
each eigenvector is used as a column of a matrix formed by the eigenvector matrix, and finally a matrix corresponding to the energy consumption mode is obtained. The power consumption mode matching the process step is determined from each row of power consumption modes. Specifically, each row of data can be matched into a corresponding process step according to historical energy consumption patterns or expert experience, a row of data which cannot be matched with the process step is discarded, some energy consumption patterns correspond to unnecessary process steps, the possibility of optimization exists, and one or more of the energy consumption patterns can be removed as target patterns.
Because the eigenvectors in the eigenvector matrix are correspondingly arranged in sequence according to the order of the eigenvalues from large to small, the h eigenvectors arranged in the back can be removed, so that some process steps with less occurrence times or unimportance can be removed more simply.
After the target mode is removed, the last h rows of data in the models are all changed into 0, and the matrix after the change is taken as the energy consumption matrix after the process step is adjusted in the modelsnewAnd determining the machine energy consumption in the preset time period according to the energy consumption matrix.
The energy consumption of the machine after the process step can be adjusted according to W-1*modesnewThe data in (1) is determined. One data per row represents a predicted energy consumption data sampled every 1 minute. Wherein W-1Is the inverse matrix of W.
As shown in fig. 3, which is a graph illustrating the energy consumption of the machine before and after the adjustment of the process steps according to the embodiment, the preset time period is 30 minutes. The abscissa T is time in minutes, the ordinate E is energy consumption data in Wh, L1 is a graph illustrating the energy consumption of the machine before the adjustment of the process step, and L2 is a graph illustrating the energy consumption of the machine after the adjustment of the process step, and it can be seen from fig. 2 that the energy consumption is saved by 40665 Wh. Specifically, the sampling frequency is once in 1 minute. Thus, the target matrix X can be takenmodeThe first 30 data in. Next, W may be taken-1*modesnewThe first 30 data in the resulting matrix results. This results in a similar comparison of figure 3.
Example four
The embodiment provides a device for determining the energy consumption of a machine for production, which is used for executing the method for determining the energy consumption of the machine for production in the first embodiment.
Fig. 4 is a schematic structural diagram of an apparatus for determining energy consumption of a machine for production according to the present embodiment. The device for determining the energy consumption of a machine for production comprises an acquisition unit 401, a first determination unit 402 and a second determination unit 403.
The obtaining unit 401 is configured to obtain multiple pieces of original energy consumption data corresponding to production in a target time period, where the production in the target time period includes multiple process steps, and the original energy consumption data is machine energy consumption corresponding to a machine that executes the process steps and is obtained by periodic sampling; the first determining unit 402 is configured to determine energy consumption modes according to the original energy consumption data, wherein each energy consumption mode corresponds to one process step; the second determining unit 403 is configured to determine, according to each energy consumption mode, the energy consumption of the machine within a preset time period after each process step is adjusted.
The working method of each unit of this embodiment is the same as that of the previous embodiment, and is not described herein again.
According to the invention, by acquiring a plurality of original energy consumption data corresponding to production in a target time period, analyzing the energy consumption of the original data to acquire the energy consumption mode corresponding to each process step, and then acquiring the machine energy consumption in the target time period after the process steps are adjusted through the energy consumption models, simulation evaluation can be carried out before a new production scheme is realized, so as to avoid high investment and low output.
EXAMPLE five
This example further illustrates the device for determining the energy consumption of the machine for production of example four. Fig. 5 is a schematic structural diagram of an apparatus for determining energy consumption of a machine for production according to the present embodiment.
The first determining unit 402 in the apparatus for determining the energy consumption of a machine for production includes a first determining subunit 4021, a second determining subunit 4022, a first acquiring subunit 4023, and a third determining subunit 4024. The first determining subunit 4021 is configured to determine an object matrix according to the original energy consumption data, where the number of columns of the object matrix is obtained according to a duration corresponding to at least one process step; the second determining subunit 4022 is configured to determine a first matrix according to the target matrix; the first obtaining subunit 4023 is configured to obtain an eigenvector matrix of the first matrix; the third determining subunit 4024 is configured to determine an energy consumption matrix according to the eigenvector matrix and the target matrix, and determine each energy consumption mode according to data of each row of the energy consumption matrix.
Optionally, the first determining subunit 4021 is specifically configured to:
determining the final column number t 'of the target matrix, wherein the final column number t' of the target matrix is the number of samples which can be sampled in the average value of the time lengths corresponding to the multiple process steps;
determining a sequence x ═ { x ] from the raw energy consumption data1,x2,…xi,…,xn’In which xiRepresenting ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n ' is the maximum sampling number which satisfies that n '/t ' is an integer in a target time period;
determining a target matrix according to the final column number t
Figure BDA0002532531100000161
Or, optionally, the first determining subunit is specifically configured to:
determining an initial column number t of a target matrix;
determining an initial sequence x ═ { x } according to the original energy consumption data1,x2,…xi,…,xnIn which xiRepresenting ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n is the maximum sampling number which satisfies n/t as an integer in a target time period;
determining an initial matrix according to the initial column number t
Figure BDA0002532531100000162
Performing a centralization operation on each original energy consumption data in the initial sequence, wherein the centralization operation comprises the following steps: acquiring an average value of the original energy consumption data of each row in the initial matrix, and subtracting the average value of the corresponding row from each original energy consumption data;
determining a secondary matrix XX from the centred initial matrixT；
For the second matrix XXTDecomposing the eigenvalues to obtain a plurality of eigenvalues, and determining corresponding variance occupancy according to the plurality of eigenvalues;
updating the value of the initial column number t and returning to execute the operation of determining an initial matrix according to the initial column number t until the value of the initial column number t is updated to a preset number;
determining an inflection point according to the difference occupancy of each party, determining the number t 'of samples in time corresponding to the inflection point, and taking the t' as the final column number of the target matrix;
determining an object matrix
Figure BDA0002532531100000171
n ' is the maximum number of samples for the target time period for which n '/t ' is an integer.
Optionally, the second determining subunit 4022 is specifically configured to:
performing a centralization operation on each original energy consumption data in the target matrix, wherein the centralization operation comprises the following steps: obtaining the average value of the original energy consumption data of each row in the target matrix, and subtracting the average value of the corresponding row from each original energy consumption data;
determining a first matrix from the centralized target matrix
Figure BDA0002532531100000172
The first obtaining subunit 4023 is specifically configured to:
performing eigenvalue decomposition on the first matrix;
sequentially sorting the corresponding eigenvectors according to the descending order of the eigenvalues, and generating an eigenvector matrix W [ W ]1；w2；…wj；…；wk]Wherein w isjJ is more than or equal to 1 and less than or equal to k;
the third determining subunit 4024 is specifically configured to:
modes＝W*Xmodewhere modes is a power consumption matrix.
Optionally, the third determining subunit 4024 is specifically configured to:
determining an energy consumption mode corresponding to each process step according to each row of data in the energy consumption matrix;
the second determining unit 403 is specifically configured to:
determining a target mode from various energy consumption modes;
and determining the machine energy consumption within a preset time period after the target mode is removed.
Optionally, the second determining unit 403 is further specifically configured to:
taking the last h energy consumption modes determined according to the energy consumption matrix as target modes, wherein h is more than or equal to 1 and is less than k;
changing the last h rows of data in the models into 0 as energy consumption matrix models after adjusting the process stepsnewAnd according to W-1*modesnewAnd the sampling frequency determines the machine energy consumption within a preset time period.
Optionally, the determining the machine energy consumption within the preset time period according to the energy consumption matrix after the adjusting process step includes: according to W-1*modesnewAnd the sampling frequency determines the machine energy consumption within a preset time period.
The working method of each unit of this embodiment is the same as that of the previous embodiment, and is not described herein again.
According to the embodiment, a corresponding matrix is established according to the original energy consumption data, the energy consumption matrix is finally obtained, the energy consumption modes corresponding to the corresponding process steps are further obtained, the process steps needing to be removed can be determined, and the machine energy consumption after the process steps are removed is obtained, so that the advance evaluation is carried out, and the high input and low output are avoided.
The invention also provides an apparatus for determining energy consumption of a machine for production, comprising at least one memory and at least one processor. Wherein the memory is to store instructions. The processor is configured to execute the method for determining the energy consumption of the machine for production described in any of the preceding embodiments according to instructions stored in the memory.
Embodiments of the present invention also provide a readable storage medium. The readable storage medium has stored therein machine readable instructions which, when executed by a machine, cause the machine to perform the method of determining machine energy consumption for production described in any of the preceding embodiments.
The readable medium has stored thereon machine readable instructions which, when executed by a processor, cause the processor to perform any of the methods previously described. In particular, a system or apparatus may be provided which is provided with a readable storage medium on which software program code implementing the functionality of any of the embodiments described above is stored and which causes a computer or processor of the system or apparatus to read and execute machine-readable instructions stored in the readable storage medium.
In this case, the program code itself read from the readable medium can realize the functions of any of the above-described embodiments, and thus the machine-readable code and the readable storage medium storing the machine-readable code form part of the present invention.
Examples of the readable storage medium include floppy disks, hard disks, magneto-optical disks, optical disks (e.g., CD-ROMs, CD-R, CD-RWs, DVD-ROMs, DVD-RAMs, DVD-RWs, DVD + RWs), magnetic tapes, nonvolatile memory cards, and ROMs. Alternatively, the program code may be downloaded from a server computer or from the cloud via a communications network.
It will be understood by those skilled in the art that various changes and modifications may be made in the above-disclosed embodiments without departing from the spirit of the invention. Accordingly, the scope of the invention should be determined from the following claims.
It should be noted that not all steps and units in the above flows and system structure diagrams are necessary, and some steps or units may be omitted according to actual needs. The execution order of the steps is not fixed and can be adjusted as required. The apparatus structures described in the above embodiments may be physical structures or logical structures, that is, some units may be implemented by the same physical entity, or some units may be implemented by a plurality of physical entities, or some units may be implemented by some components in a plurality of independent devices.
In the above embodiments, the hardware unit may be implemented mechanically or electrically. For example, a hardware unit or processor may comprise permanently dedicated circuitry or logic (such as a dedicated processor, FPGA or ASIC) to perform the corresponding operations. The hardware units or processors may also include programmable logic or circuitry (e.g., a general purpose processor or other programmable processor) that may be temporarily configured by software to perform the corresponding operations. The specific implementation (mechanical, or dedicated permanent, or temporarily set) may be determined based on cost and time considerations.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (20)

1. A method of determining energy consumption of a machine for production, comprising:
acquiring a plurality of original energy consumption data corresponding to production in a target time period, wherein the production in the target time period comprises a plurality of process steps, and the original energy consumption data are machine energy consumption corresponding to a machine which executes the process steps and is acquired by periodic sampling;
determining energy consumption modes according to the original energy consumption data, wherein each energy consumption mode corresponds to one process step;
and determining the machine energy consumption within a preset time period after each process step is adjusted according to each energy consumption mode.
2. The method of claim 1, wherein determining the energy consumption modes from the raw energy consumption data comprises:
determining a target matrix according to the original energy consumption data, wherein the column number of the target matrix is obtained according to the duration corresponding to at least one process step;
determining a first matrix according to the target matrix;
acquiring a feature vector matrix of the first matrix;
and determining an energy consumption matrix according to the characteristic vector matrix and the target matrix, and determining each energy consumption mode according to data of each row of the energy consumption matrix.
3. The method of claim 2, wherein the number of columns of the target matrix is the number of samples that can be taken within an average of the time durations corresponding to the plurality of process steps.
4. The method of claim 2, wherein determining an objective matrix based on the raw energy consumption data comprises:
determining the final column number t' of the target matrix;
determining a sequence x ═ { x ] from the raw energy consumption data1，x2，...xi，...，xn′In which xiRepresenting ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n ' is the maximum sampling number which satisfies that n '/t ' is an integer in a target time period;
determining a target matrix according to the final column number t
Figure FDA0002532531090000011
5. The method of claim 4, wherein the final column number t' of the target matrix is the number of samples that can be taken within the average of the time durations corresponding to the plurality of process steps.
6. The method of claim 2, wherein determining an objective matrix based on the raw energy consumption data comprises:
determining an initial column number t of a target matrix;
determining an initial sequence x ═ { x ] from the raw energy consumption data1，x2，...xi，...，xnIn which xiRepresenting ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n is the maximum sampling number which satisfies n/t as an integer in a target time period;
determining an initial matrix according to the initial column number t
Figure FDA0002532531090000021
Performing a centralization operation on each original energy consumption data in the initial sequence, wherein the centralization operation comprises: obtaining an average value of the original energy consumption data of each row in the initial matrix, and subtracting the average value of the corresponding row from each original energy consumption data;
determining a secondary matrix XX from the centred initial matrixT；
For the second matrix XXTDecomposing the eigenvalues to obtain a plurality of eigenvalues, and determining corresponding variance occupancy according to the plurality of eigenvalues;
updating the value of the initial column number t and returning to execute the operation of determining an initial matrix according to the initial column number t until the value of the initial column number t is updated to a preset number;
determining an inflection point according to each corresponding variance occupancy and the initial column number t, determining the number t 'of samples which can be sampled in time corresponding to the inflection point, and taking the t' as the final column number of the target matrix;
determining an object matrix
Figure FDA0002532531090000022
n ' is the maximum number of samples for the target time period for which n '/t ' is an integer.
7. The method of claim 2, wherein determining a first matrix based on the target matrix comprises:
performing a centralization operation on each original energy consumption data in the target matrix, wherein the centralization operation comprises the following steps: obtaining an average value of the original energy consumption data of each row in the target matrix, and subtracting the average value of the corresponding row from each original energy consumption data;
according to the centralized target matrix XmodeDetermining the first matrix
Figure FDA0002532531090000023
Obtaining the first matrix
Figure FDA0002532531090000024
Comprises:
for the first matrix
Figure FDA0002532531090000025
Carrying out eigenvalue decomposition;
sequentially sorting the corresponding eigenvectors according to the descending order of the eigenvalues, and generating an eigenvector matrix W [ W ]1；W2；...wj；...；Wk]Wherein w isjJ is more than or equal to 1 and less than or equal to k, and each row of the eigenvector matrix is a value corresponding to one eigenvector;
determining an energy consumption matrix according to the eigenvector matrix and the target matrix comprises:
modes＝W*Xmodewhere mods is the power consumption matrix.
8. The method of claim 7, wherein determining the energy consumption modes according to the data of the columns of the energy consumption matrix comprises:
determining an energy consumption mode corresponding to each process step according to each row of data in the energy consumption matrix;
according to each energy consumption mode, determining the machine energy consumption within a preset time period after each process step is adjusted comprises the following steps:
determining a target mode from various energy consumption modes;
and determining the machine energy consumption within the preset time period after the target mode is removed.
9. The method of claim 8, wherein determining the target pattern from the various energy consumption patterns comprises: taking the last h energy consumption modes determined according to the energy consumption matrix as target modes, wherein h is more than or equal to 1 and is less than k;
determining the machine energy consumption within the preset time period after the target pattern is removed comprises:
changing the last h rows of data in the models into 0 as energy consumption matrix models after adjusting the process stepsnewAnd determining the machine energy consumption in the preset time period according to the energy consumption matrix after the process step is adjusted.
10. The method of claim 9, wherein determining the machine energy consumption within the predetermined time period based on the energy consumption matrix after the adjusting process step comprises:
according to W-1*modesnewAnd the sampling frequency determines the machine energy consumption within a preset time period.
11. The method of claim 9, wherein determining a power consumption pattern corresponding to each process step based on the data for each row in the power consumption matrix comprises at least one of:
the first method is as follows: discarding a line of data which cannot be matched with the process steps;
the second method comprises the following steps: and determining a row of data corresponding to the process steps one by one as an energy consumption mode.
12. Apparatus for determining the energy consumption of a machine for production, comprising:
the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a plurality of original energy consumption data corresponding to production in a target time period, the production in the target time period comprises a plurality of process steps, and the original energy consumption data is the machine energy consumption corresponding to a machine which executes the process steps and is acquired by periodic sampling;
a first determining unit for determining energy consumption modes based on said raw energy consumption data, each energy consumption mode corresponding to a process step;
and the second determining unit is used for determining the machine energy consumption within a preset time period after each process step is adjusted according to each energy consumption mode.
13. The apparatus of claim 12, wherein the first determining unit comprises:
a first determining subunit, configured to determine an objective matrix according to the original energy consumption data, where the number of columns of the objective matrix is obtained according to a duration corresponding to at least one process step;
a second determining subunit, configured to determine a first matrix according to the target matrix;
a first obtaining subunit, configured to obtain a feature vector matrix of the first matrix;
and the third determining subunit is used for determining an energy consumption matrix according to the eigenvector matrix and the target matrix and determining each energy consumption mode according to data of each row of the energy consumption matrix.
14. The apparatus according to claim 13, wherein the first determining subunit is specifically configured to:
determining the final column number t 'of the target matrix, wherein the final column number t' of the target matrix is the number of samples which can be sampled in the average value of the time lengths corresponding to the multiple process steps;
determining a sequence x ═ { x ] from the raw energy consumption data1，x2，...xi，...，xn′In which xiRepresenting ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n ' is the maximum sampling number which satisfies that n '/t ' is an integer in a target time period;
determining a target matrix according to the final column number t
Figure FDA0002532531090000041
15. The method according to claim 13, wherein the first determining subunit is specifically configured to:
determining an initial column number t of a target matrix;
according to the original energy consumptionData determination of an initial sequence x ═ x1，x2，...xi，...，xnIn which xiRepresenting ith original energy consumption data, wherein i is more than or equal to 1 and less than or equal to n, i is a positive integer, and n is the maximum sampling number which satisfies n/t as an integer in a target time period;
determining an initial matrix according to the initial column number t
Figure FDA0002532531090000042
Performing a centralization operation on each original energy consumption data in the initial sequence, wherein the centralization operation comprises: obtaining an average value of the original energy consumption data of each row in the initial matrix, and subtracting the average value of the corresponding row from each original energy consumption data;
determining a secondary matrix XX from the centred initial matrixT；
For the second matrix XXTDecomposing the eigenvalues to obtain a plurality of eigenvalues, and determining corresponding variance occupancy according to the plurality of eigenvalues;
updating the value of the initial column number t and returning to execute the operation of determining an initial matrix according to the initial column number t until the value of the initial column number t is updated to a preset number;
determining an inflection point according to each corresponding variance occupancy and the initial column number t, determining the number t 'of samples which can be sampled in time corresponding to the inflection point, and taking the t' as the final column number of the target matrix;
determining an object matrix
Figure FDA0002532531090000051
n ' is the maximum number of samples for the target time period for which n '/t ' is an integer.
16. The apparatus according to claim 13, wherein the second determining subunit is specifically configured to:
performing a centralization operation on each original energy consumption data in the target matrix, wherein the centralization operation comprises the following steps: obtaining an average value of the original energy consumption data of each row in the target matrix, and subtracting the average value of the corresponding row from each original energy consumption data;
according to the centralized target matrix XmodeDetermining the first matrix
Figure FDA0002532531090000052
The first obtaining subunit is specifically configured to:
for the first matrix
Figure FDA0002532531090000053
Carrying out eigenvalue decomposition;
sequentially sorting the corresponding eigenvectors according to the descending order of the eigenvalues, and generating an eigenvector matrix W [ W ]1；w2；...wj；...；Wk]Wherein w isjJ is more than or equal to 1 and less than or equal to k, and each row of the eigenvector matrix is a value corresponding to one eigenvector;
the third determining subunit is specifically configured to:
modes＝W*Xmodewhere mods is the power consumption matrix.
17. The apparatus according to claim 16, wherein the third determining subunit is specifically configured to:
determining an energy consumption mode corresponding to each process step according to each row of data in the energy consumption matrix;
the second determining unit is specifically configured to:
determining a target mode from various energy consumption modes;
and determining the machine energy consumption within the preset time period after the target mode is removed.
18. The apparatus of claim 17, wherein the second determining unit is further specifically configured to:
taking the last h energy consumption modes determined according to the energy consumption matrix as target modes, wherein h is more than or equal to 1 and is less than k;
changing the last h rows of data in the models into 0 as energy consumption matrix models after adjusting the process stepsnewAnd according to W-1*modesnewAnd the sampling frequency determines the machine energy consumption within a preset time period.
19. Apparatus for determining the energy consumption of a machine for production, comprising:
at least one memory for storing instructions;
at least one processor configured to perform the method of determining machine energy consumption for production of any of claims 1-11 in accordance with instructions stored by the memory.
20. Readable storage medium having stored therein machine readable instructions, which when executed by a machine, perform the method of determining machine energy consumption for production according to any of claims 1-11.
CN202010522303.5A 2020-06-10 2020-06-10 Method and device for determining the energy consumption of a machine for production Active CN111667186B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010522303.5A CN111667186B (en) 2020-06-10 2020-06-10 Method and device for determining the energy consumption of a machine for production

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010522303.5A CN111667186B (en) 2020-06-10 2020-06-10 Method and device for determining the energy consumption of a machine for production

Publications (2)

Publication Number Publication Date
CN111667186A true CN111667186A (en) 2020-09-15
CN111667186B CN111667186B (en) 2023-05-05

Family

ID=72386412

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010522303.5A Active CN111667186B (en) 2020-06-10 2020-06-10 Method and device for determining the energy consumption of a machine for production

Country Status (1)

Country Link
CN (1) CN111667186B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112578178A (en) * 2020-12-07 2021-03-30 珠海格力电器股份有限公司 Energy information monitoring method, energy monitoring device, production line and storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109034497A (en) * 2018-08-31 2018-12-18 广东工业大学 Prediction technique, system, medium and the equipment of polycrystalline reduction process energy consumption value
CN110991043A (en) * 2019-12-03 2020-04-10 北京中计开元科技有限公司 Energy consumption modeling and evaluating method of energy consumption system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109034497A (en) * 2018-08-31 2018-12-18 广东工业大学 Prediction technique, system, medium and the equipment of polycrystalline reduction process energy consumption value
CN110991043A (en) * 2019-12-03 2020-04-10 北京中计开元科技有限公司 Energy consumption modeling and evaluating method of energy consumption system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112578178A (en) * 2020-12-07 2021-03-30 珠海格力电器股份有限公司 Energy information monitoring method, energy monitoring device, production line and storage medium

Also Published As

Publication number Publication date
CN111667186B (en) 2023-05-05

Similar Documents

Publication Publication Date Title
CN106647262B (en) Differential evolution method for agile satellite multi-target task planning
JP5362823B2 (en) Recommendation system by fast matrix factorization using infinite dimensions
CN110334580A (en) The equipment fault classification method of changeable weight combination based on integrated increment
JPWO2015040790A1 (en) Payout amount prediction device, payout amount prediction method, program, and payout amount prediction system
WO2015040789A1 (en) Product recommendation device, product recommendation method, and recording medium
CN112417028A (en) Wind speed time sequence characteristic mining method and short-term wind power prediction method
CN113283426B (en) Embedded target detection model generation method based on multi-target neural network search
WO2015040791A1 (en) Order-volume determination device, order-volume determination method, recording medium, and order-volume determination system
CN112766537B (en) Short-term electric load prediction method
CN108022123B (en) Automatic adjustment method and device for business model
JP7267966B2 (en) Information processing device and information processing method
CN107783998A (en) The method and device of a kind of data processing
CN111667186A (en) Method and device for determining the energy consumption of a machine for production
CN111931990B (en) Power load prediction method based on optimized least square support vector machine
CN111400964B (en) Fault occurrence time prediction method and device
CN116543229A (en) Image classification method of deep learning optimizer based on self-adaptive accumulation coefficient
CN110322055A (en) A kind of method and system improving data risk model scoring stability
CN112419098B (en) Power grid safety and stability simulation sample screening and expanding method based on safety information entropy
US20220261648A1 (en) Gradient pruning for efficient training of machine learning models
CN110737773B (en) Information classification method and system based on neural network
CN113688989A (en) Deep learning network acceleration method, device, equipment and storage medium
CN108121993A (en) A kind of data processing method and device
CN117893216B (en) User behavior analysis method and system based on big data
CN115204324B (en) Equipment power consumption abnormity detection method and device based on IFOA-DBN-ELM
CN114898152B (en) Embedded elastic self-expanding universal learning system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant