CN115022348A - High-end battery intelligent factory cloud-level architecture data storage method - Google Patents

High-end battery intelligent factory cloud-level architecture data storage method Download PDF

Info

Publication number
CN115022348A
CN115022348A CN202210614506.6A CN202210614506A CN115022348A CN 115022348 A CN115022348 A CN 115022348A CN 202210614506 A CN202210614506 A CN 202210614506A CN 115022348 A CN115022348 A CN 115022348A
Authority
CN
China
Prior art keywords
cell space
space
data analyzer
local
global
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
CN202210614506.6A
Other languages
Chinese (zh)
Other versions
CN115022348B (en
Inventor
王子赟
程林
王艳
纪志成
杨建芬
宋文龙
高伟伟
施璐
刘红杰
李丹
朱文光
王子杨
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.)
Jiangnan University
Tianneng Battery Group Co Ltd
Original Assignee
Jiangnan University
Tianneng Battery Group Co 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 Jiangnan University, Tianneng Battery Group Co Ltd filed Critical Jiangnan University
Priority to CN202210614506.6A priority Critical patent/CN115022348B/en
Publication of CN115022348A publication Critical patent/CN115022348A/en
Application granted granted Critical
Publication of CN115022348B publication Critical patent/CN115022348B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1095—Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04Q—SELECTING
    • H04Q9/00—Arrangements in telecontrol or telemetry systems for selectively calling a substation from a main station, in which substation desired apparatus is selected for applying a control signal thereto or for obtaining measured values therefrom
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04Q—SELECTING
    • H04Q2209/00—Arrangements in telecontrol or telemetry systems
    • H04Q2209/20—Arrangements in telecontrol or telemetry systems using a distributed architecture
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04Q—SELECTING
    • H04Q2209/00—Arrangements in telecontrol or telemetry systems
    • H04Q2209/80—Arrangements in the sub-station, i.e. sensing device
    • H04Q2209/84—Measuring functions
    • 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
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Testing Or Calibration Of Command Recording Devices (AREA)

Abstract

The invention discloses a high-end battery intelligent factory cloud-level architecture data storage method, and belongs to the field of intelligent storage. According to the method, real-time factory data are transmitted to a cloud end for storage by using each distributed sensor, the problem of data loss under a cloud level is considered, timely data compensation is performed after a local data analyzer detects data loss, meanwhile, each local data analyzer performs local state estimation in the distributed data analyzer by using a collective filtering algorithm under the disturbance of uncertain noise, the real feasible set range of the system state is jointly obtained, and data and state estimation results are transmitted back to an original cloud end in real time and copied to a backup cloud.

Description

High-end battery intelligent factory cloud-level architecture data storage method
Technical Field
The invention relates to a high-end battery intelligent factory cloud-level architecture data storage method, and belongs to the field of intelligent storage.
Background
With the rapid development of science and technology, the current high-end battery intelligent factory workshops are more and more complex, the scale is larger and larger, on an intelligent factory production line, a wireless sensor network is formed by distributed sensors to collect operation data of each production process in real time, and a data analyzer is used for processing the collected operation data to obtain system state estimation of a factory, so that the high-end battery intelligent factory workshop has important significance for subsequent monitoring management.
The traditional data processing mode of a high-end intelligent factory is to transmit data of each distributed sensor to a data analyzer, store the data in the data analyzer and process the data to complete state estimation, all data information can be fully utilized, and the state estimation precision has global optimality. However, under the condition of a large number of distributed sensors, when data storage and processing are performed in one data analyzer, the load of the data analyzer is large, so that the calculation efficiency is low, and meanwhile, if the data analyzer fails, all stored data are lost, so that the state estimation process is seriously influenced; the cloud end is used for data storage, the data analyzer acquires data from the cloud end for processing, the risk of all data loss is greatly reduced by isolating data storage and data processing, meanwhile, the burden of a single data analyzer is reduced, and the calculation efficiency is higher, so that the cloud-level architecture method of the high-end battery intelligent factory has deeper research significance.
However, the cloud end for storage and the data analyzer for data processing also face some problems, such as: when a data analyzer processes data, process noise of a system to be measured and sensor measurement noise are inevitable, and the research of a traditional data processing algorithm assumes that the noise has Gaussian characteristics, such as a Kalman filtering algorithm; however, in the actual production process, on one hand, the measured data transmitted to the cloud stage is insufficient under some special conditions, for example, part of the data is lost, which often results in that accurate statistical characteristics cannot be obtained; on the other hand, process noise and measurement noise are not subjected to certain random probability distribution, so that the problem that accurate storage data cannot be obtained due to data loss exists in the conventional cloud data storage mode is caused.
Disclosure of Invention
In order to solve the problem that accurate storage data cannot be obtained due to data loss in the existing cloud data storage mode, the invention provides a high-end battery intelligent factory cloud architecture data storage method, which is applied to a high-end battery intelligent factory cloud architecture, and the high-end battery intelligent factory cloud architecture comprises the following steps: the system comprises a plurality of sensors, a plurality of local data analyzers and a global data analyzer, wherein the sensors, the local data analyzers and the global data analyzer are distributed on each production process of the high-end battery and are used for acquiring data in real time;
each sensor corresponds to a cloud end used for storing data acquired and processed by the sensor; the local data analyzers are used for acquiring data acquired by the sensors in real time from the cloud and performing local state estimation, and each local data analyzer sends a local state estimation result acquired by the local data analyzer to the global data analyzer; the global data analyzer is used for carrying out global state estimation according to the local state estimation result of each local data analyzer, transmitting the obtained global state estimation result back to each local data analyzer and further transmitting the global state estimation result to the corresponding cloud end for storage by the local data analyzers.
Optionally, the method includes:
step 1: acquiring measurement data of all distributed sensors at the moment k in the high-end battery production process, transmitting the measurement data to corresponding cloud terminals for storage, and downloading the measurement data stored in the cloud terminals by using all local data analyzers;
step 2: acquiring a linear state space expression of a high-end battery intelligent factory workshop, and determining an initial positive multi-cell space and an initial full-symmetry multi-cell space; the state space expression comprises a measurement equation and a state equation;
and step 3: according to a measurement equation in the state space expression, each local data analyzer forms the downloaded measurement data at the k moment into measurement band information; judging whether measurement data loss occurs or not, if the measurement data loss occurs, if k is equal to 1, processing is not needed, and if k is not equal to 1, the measurement information at the k-1 moment is taken as the measurement information at the k moment; and transmitting to the corresponding cloud;
and 4, step 4: and (3) solving local state estimation of each local data analyzer at the k moment according to the state equation of the state space expression in the step 2 and the measurement band information in the step 3:
solving a predicted fully-symmetrical multi-cell space at the moment k +1 by using a state equation in a state space expression, and converting the predicted fully-symmetrical multi-cell space into a predicted positive multi-cell space; then, the positive multi-cell space at the k +1 moment is solved by using the measurement band information at the k moment, and the local state estimation of each local data analyzer is completed;
and 5: transmitting the local state estimation obtained by each local data analyzer to a global data analyzer, calculating the final global state estimation by utilizing Minkowski, transmitting the global state estimation result to each local data analyzer, and transmitting the global state estimation result to each corresponding cloud end by each local data analyzer;
and 6: and each local data analyzer transmits the measurement data acquired by the local data analyzer, the local state estimation result obtained according to the measurement data and the global state estimation result to a backup cloud for backup storage.
Optionally, the step 2 includes:
the method comprises the following steps of obtaining an expression of an n-dimensional linear state space of an intelligent factory workshop of the high-end battery:
x k+1 ＝Ax k +Bu k +w k (1)
y i,k ＝C i x k +v i,k (2)
equations (1) and (2) are a state equation and a measurement equation of the system, respectively, where k represents time, and k is 1, …, N; x is the number of k A system state value representing time k; x is the number of k+1 Represents the system state value at the moment k + 1; u. of k A system input value representing time k; a and B respectively represent a system matrix and an input matrix; w is a k Representing process noise of the system; i represents a distributed sensor number, i is 1, …, and M represents that M distributed sensors in the wireless sensor network perform measurement; y is i,k Representing the data collected by the ith sensor k at the moment; c i An output matrix representing the ith sensor; v. of i,k Representing the measurement noise, v, of the i-th sensor i,k ∈[-δ k ,δ k ]，δ k Is the noise maximum boundary value at time k, -delta k The noise minimum boundary value at the k moment;
defining the positive multi-cellular space as:
Figure BDA0003666994500000031
wherein ,
Figure BDA0003666994500000032
is the central point of the positive multi-cell space, d is the generator matrix of the positive multi-cell space, diag (d) is the diagonal value equal to dDiagonal matrix, x is feasible set range of positive multi-cell space wrapping, m is intermediate variable, | | | | | luminance ∞ Represents an infinite norm;
defining the fully symmetric multi-cell space as:
Figure BDA0003666994500000033
wherein p is the central point of the full-symmetric multi-cell space, H is the generating matrix of the full-symmetric multi-cell space, B n ∈[-1,1] n Is n unit intervals [ -1,1 [)]The unit box is formed by the following components,
Figure BDA0003666994500000034
minkowski and z is an intermediate variable;
from the above definition, the positive multi-cell space at the initial time is determined as
Figure BDA0003666994500000035
And an initial fully symmetric multi-cell space Z (p) 1 ,H 1 ) And make an order
Figure BDA0003666994500000036
Namely that
Figure BDA0003666994500000037
diag{d 1 }＝H 1 (6)
wherein ,
Figure BDA0003666994500000038
is the central point of the initial positive multicellular space, d 1 Generating a matrix for an initial positive multi-cell space; p is a radical of 1 Is the central point of the initial holosymmetric multi-cell space, H 1 Is a generator matrix of an initial fully symmetric multi-cell space.
Optionally, step 3 includes:
defining the measurement band information of the local data analyzer according to the downloaded sensor i as S (p) i,k ,c i,k )；
Figure BDA0003666994500000039
wherein ,ci,k The sensor i measures the center point with information for time k,
Figure BDA00036669945000000310
p i,k the direction vector with information is measured for sensor i at time k,
Figure BDA00036669945000000311
θ s estimating a feasible set range for the measurement band information;
if the loss of the measured data occurs, when k is equal to 1, the order is given
S(p i,k ,c i,k )＝0 (8)
When k ≠ 1, the measurement information at k-1 time is used as the measurement information at k time, so as to
S(p i,k ,c i,k )＝S(p i,k-1 ,c i,k-1 ) (9)。
Optionally, the step 4 includes:
solving a predicted fully-symmetrical multi-cell space at the moment k +1 according to a state equation in the state space expression, and converting the predicted fully-symmetrical multi-cell space into a predicted positive multi-cell space; then, the positive multi-cell space at the k +1 moment is solved by using the measurement band information at the k moment, and the local state estimation of each local data analyzer is completed;
according to the state equation of the state space expression of the formula (1) and the measurement band information shown in the formula (7), local state estimation of each cloud-side corresponding local data analyzer at the k moment is obtained:
in the system state equation of equation (1), the process noise is confined to an unknown but bounded range, i.e., let the process noise w k P (0, g), g being the boundary value of the noise;
using a fully symmetric multi-cell space Z (p) without taking into account process noise interference k ,H k ) Watch (A)Positive multi-cell space showing k time
Figure BDA0003666994500000041
Namely, it is
Figure BDA0003666994500000042
H k ＝diag{d k } (12)
Mixing Z (p) k ,H k ) Method for solving predicted fully-symmetrical multi-cell space at k +1 moment by substituting equation of state in formula (1)
Figure BDA0003666994500000043
Figure BDA0003666994500000044
Figure BDA0003666994500000045
wherein ,
Figure BDA0003666994500000046
to predict the center point of the fully symmetric multi-cellular space,
Figure BDA0003666994500000047
generating a matrix for predicting a fully symmetric multi-cell space;
at this time is
Figure BDA0003666994500000048
Adding process noise w k P (0, g), and using the most compact predictive positive multi-cellular space
Figure BDA0003666994500000049
De-wrapping the predicted fully symmetric multi-cellular space,
Figure BDA00036669945000000410
the calculation formula of (2) is as follows:
Figure BDA00036669945000000411
Figure BDA00036669945000000412
wherein ,
Figure BDA00036669945000000413
to predict the center point of the positive multi-cell space,
Figure BDA00036669945000000414
to predict the generator matrix of the positive multi-cell space,
Figure BDA00036669945000000415
is that the diagonal value is equal to
Figure BDA00036669945000000416
The diagonal matrix of (a) is,
Figure BDA00036669945000000417
for diagonal values equal to
Figure BDA00036669945000000418
Diagonal matrix of n H Is composed of
Figure BDA00036669945000000419
Dimension of the matrix, | | | | represents norm, l, h are intermediate variables;
predicting the positive multi-cellular space determined in the formula (15) and the formula (16)
Figure BDA00036669945000000420
Decomposition into n band-constrained equations
Figure BDA00036669945000000421
Figure BDA00036669945000000422
wherein xj To predict the feasible set variables for the j-th dimension of the positive multi-cell space,
Figure BDA00036669945000000423
to predict the feasible set minimum for the j-th dimension of the positive multi-cell space,
Figure BDA00036669945000000424
j is a dimension variable for predicting the maximum value of the feasible set of the j dimension of the positive multi-cell space;
the measurement band information S (p) of the i-th data analyzer at time k obtained by equation (7) is obtained i,k ,c i,k ) Updating to obtain positive multi-cell space at k +1 time
Figure BDA0003666994500000051
Comprises the following steps:
Figure BDA0003666994500000052
Figure BDA0003666994500000053
wherein ,
Figure BDA0003666994500000054
Figure BDA0003666994500000055
Figure BDA0003666994500000056
wherein xs For restraining in a beltN represents the intersection, α j (k +1) is the maximum value of the jth dimension of the feasible set range,
α j+n (k +1) is the minimum value of the jth dimension of the feasible set range, max represents taking the maximum value, min represents taking the minimum value,
Figure BDA0003666994500000057
representing the transpose of the jth dimension of the unit diagonal matrix,
Figure BDA0003666994500000058
is the value of the jth dimension of the center point of the positive multi-cell space,
Figure BDA0003666994500000059
the value of the jth dimension of the matrix is generated for the positive multi-cell space,
Figure BDA00036669945000000510
is that the diagonal value is equal to
Figure BDA00036669945000000511
The diagonal matrix of (a).
Figure BDA00036669945000000512
Final local state estimation results for each data analyzer.
Optionally, step 5 includes:
local State estimation results for each data Analyzer according to equations (17) to (22)
Figure BDA00036669945000000513
Solving for M positive multi-cell spaces and for global positive multi-cell spaces using Minkowski's sum in a global data analyzer
Figure BDA00036669945000000514
Figure BDA00036669945000000515
Is defined as:
Figure BDA00036669945000000516
positive multi-cell space with M local estimates
Figure BDA00036669945000000517
Decomposition into n constrained equations by equation (17)
Figure BDA00036669945000000518
The band-constrained equation in the global estimation is defined as
Figure BDA00036669945000000519
Figure BDA00036669945000000520
Then
Figure BDA00036669945000000521
Comprises the following steps:
Figure BDA00036669945000000522
Figure BDA00036669945000000523
wherein ,
Figure BDA0003666994500000061
Figure BDA0003666994500000062
Figure BDA0003666994500000063
wherein xsum Is composed of
Figure BDA0003666994500000064
Variable of feasible set of lines, alpha j,sum (k +1) is the maximum value of the jth dimension of the feasible set range;
α j+n,sum (k +1) is the minimum value of the jth dimension of the feasible set range,
Figure BDA0003666994500000065
is the global positive multi-cell spatial center point,
Figure BDA0003666994500000066
is the value of the jth dimension of the central point of the global positive multi-cell space, d k+1 A generator matrix for the global positive multi-cell space,
Figure BDA0003666994500000067
generating the j-th dimension value of the matrix for the positive multi-cell space, the global positive multi-cell space
Figure BDA0003666994500000068
The represented feasible set range is a final global estimation result, the global state estimation result is transmitted to each local data analyzer, and each local data analyzer transmits the global state estimation result to each corresponding cloud.
Optionally, the local data analyzer and the global data analyzer are implemented by using a computer server.
The invention has the beneficial effects that:
according to the method, the measurement data of the high-end battery production process at the moment when the sensors are distributed k are obtained and transmitted to the corresponding cloud for storage, and meanwhile, the data analyzer is used for downloading the measurement data stored in the cloud; acquiring a linear state space expression of a high-end battery intelligent factory workshop, and determining an initial positive multi-cell space and an initial full-symmetry multi-cell space; measuring data at the k moment downloaded by the data analyzer is formed into measuring band information by a measuring equation in the state space expression; if the loss of the measurement data occurs, when k is equal to 1, no processing is needed, and when k is equal to 1, the measurement information at the time of k-1 is taken as the measurement information at the time of k; and transmitting to the original cloud end; according to the state equation of the state space expression and the measurement zone information, local state estimation of the data analyzer k moment corresponding to each cloud is obtained; transmitting the local state estimation obtained by each data analyzer to a total data analyzer, calculating the final global state estimation by utilizing Minkowski, transmitting the global state estimation result to each data analyzer, and transmitting the global state estimation result to each original cloud end by the data analyzers; and transmitting the data of each data analyzer and the global state estimation result to a backup cloud for storage. According to the cloud-level architecture method for the battery intelligent factory, the factory data are transmitted to the cloud end for storage by the aid of the distributed sensors, the problem of data loss under the cloud level is considered, the data analyzer performs timely data compensation after detecting the data loss, meanwhile, local state estimation is performed in the distributed data analyzer by the aid of a collective filtering algorithm under disturbance of uncertain noise, production practice is better met, a real feasible set range of system states is jointly obtained, and data and state estimation results are transmitted back to the original cloud end in real time and are copied to a backup cloud.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flow chart of a high-end battery smart factory cloud architecture method disclosed in one embodiment of the present application.
Fig. 2 is a cloud architecture diagram of a high-end battery smart factory as disclosed in one embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The first embodiment is as follows:
the embodiment provides a high-end battery intelligent factory cloud architecture data storage method, which is applied to a high-end battery intelligent factory cloud architecture, and with reference to fig. 1, the method includes:
step 1: acquiring measurement data of all distributed sensors at the moment k in the high-end battery production process, transmitting the measurement data to corresponding cloud terminals for storage, and downloading the measurement data stored in the cloud terminals by using all local data analyzers;
step 2: acquiring a linear state space expression of a high-end battery intelligent factory workshop, and determining an initial positive multi-cell space and an initial full-symmetry multi-cell space; the state space expression comprises a measurement equation and a state equation;
and step 3: according to a measurement equation in the state space expression, each local data analyzer forms the downloaded measurement data at the k moment into measurement band information; judging whether measurement data loss occurs or not, if the measurement data loss occurs, if k is equal to 1, processing is not needed, and if k is not equal to 1, the measurement information at the k-1 moment is taken as the measurement information at the k moment; and transmitting to the corresponding cloud end;
and 4, step 4: and (3) solving local state estimation of each local data analyzer at the k moment according to the state equation of the state space expression in the step 2 and the measurement band information in the step 3:
solving a predicted fully-symmetrical multi-cell space at the moment k +1 by using a state equation in a state space expression, and converting the predicted fully-symmetrical multi-cell space into a predicted positive multi-cell space; then, the measuring band information at the moment k is utilized to calculate the positive multi-cell space at the moment k +1, and the local state estimation of each local data analyzer is completed;
and 5: transmitting the local state estimation obtained by each local data analyzer to a global data analyzer, calculating the final global state estimation by utilizing Minkowski, transmitting the global state estimation result to each local data analyzer, and transmitting the global state estimation result to each corresponding cloud end by each local data analyzer;
step 6: and each local data analyzer transmits the measurement data acquired by the local data analyzer, the local state estimation result obtained according to the measurement data and the global state estimation result to a backup cloud for backup storage.
Example two:
the embodiment provides a method for storing data of a cloud-level architecture of a high-end battery intelligent factory, the method is applied to the cloud-level architecture of the high-end battery intelligent factory, please refer to fig. 2, and the cloud-level architecture of the high-end battery intelligent factory includes: the system comprises a plurality of sensors, a plurality of local data analyzers and a global data analyzer, wherein the sensors, the local data analyzers and the global data analyzer are distributed on each production process of the high-end battery and are used for acquiring data in real time; each sensor corresponds to a cloud end used for storing data acquired and processed by the sensor; the local data analyzers are used for acquiring data acquired by the sensors in real time from the cloud and performing local state estimation, and each local data analyzer sends a local state estimation result acquired by the local data analyzer to the global data analyzer; the global data analyzer is used for carrying out global state estimation according to the local state estimation result of each local data analyzer, transmitting the obtained global state estimation result back to each local data analyzer and further transmitting the global state estimation result to the corresponding cloud end by the local data analyzers for storage; the method comprises the following steps:
step 101, acquiring measurement data of all distributed sensors at the moment k in a high-end battery production process, transmitting the measurement data to corresponding cloud terminals for storage, and downloading the measurement data stored in the cloud terminals by using all local data analyzers;
each distributed sensor corresponds to a cloud storage, and each cloud corresponds to a local data analyzer for data processing.
102, acquiring a linear state space expression of a high-end battery intelligent factory workshop, and determining an initial positive multi-cell space and an initial full-symmetrical multi-cell space; the state space expression comprises a measurement equation and a state equation;
the method comprises the following steps of obtaining an n-dimensional linear state space expression of a high-end battery intelligent factory workshop:
x k+1 ＝Ax k +Bu k +w k (1)
y i,k ＝C i x k +v i,k (2)
equations (1) and (2) are a state equation and a measurement equation of the system, respectively, where k represents time, and k is 1, …, N; x is the number of k A system state value representing time k; x is the number of k+1 Represents the system state value at the moment k + 1; u. of k A system input value representing time k; a and B respectively represent a system matrix and an input matrix; w is a k Representing process noise of the system; i represents a distributed sensor number, i is 1, …, and M represents that M distributed sensors in the wireless sensor network perform measurement; y is i,k Represents the observed value at the moment of the ith sensor k; c i An output matrix representing the ith sensor; v. of i,k Representing the measurement noise, v, of the i-th sensor i,k ∈[-δ k ,δ k ]，δ k Is the noise maximum boundary value at time k, -delta k The noise minimum boundary value at the k moment;
the system state value refers to the state value of each process of battery production, namely the production state value of the processes of mixing, coating, rolling, baking, assembling and formation on a production line; the system input value refers to the quantity of raw materials for battery production;
defining the positive multi-cellular space as:
Figure BDA0003666994500000081
wherein ,
Figure BDA0003666994500000082
is the central point of the regular multicellular space, d is the generating matrix of the regular multicellular space, diag (d) is the diagonal matrix with the diagonal value equal to d, x is the feasible set range of the regular multicellular space, m is the intermediate variable, | | | | | (| ∞ Represents an infinite norm;
defining the full-symmetric multi-cell space as follows:
Figure BDA0003666994500000091
wherein p is holo-symmetryThe central point of the multi-cell space, H is the generating matrix of the full-symmetrical multi-cell space, B n ∈[-1,1] n Is n unit intervals [ -1,1 [)]The unit box is formed by the following components,
Figure BDA0003666994500000092
for Minkowski and z as an intermediate variable, it can be seen that when H is a diagonal matrix and H ═ diag { d }, the fully symmetric multi-cell space becomes a positive multi-cell space, and thus it can be seen that the positive multi-cell space is a special fully symmetric multi-cell space;
from the above definition, the positive multi-cell space at the initial time is determined as
Figure BDA0003666994500000093
And an initial fully symmetric multi-cell space Z (p) 1 ,H 1 ) And make an order
Figure BDA0003666994500000094
Namely that
Figure BDA0003666994500000095
diag{d 1 }＝H 1 (6)
103, according to a measurement equation in the state space expression, each local data analyzer forms the downloaded measurement data at the time k into measurement band information; judging whether measurement data loss occurs or not, if the measurement data loss occurs, if k is equal to 1, processing is not needed, and if k is not equal to 1, the measurement information at the k-1 moment is taken as the measurement information at the k moment; and transmitting to the corresponding cloud end;
defining the measurement zone information of a sensor i in a time k data analyzer as S (p) i,k ,c i,k )；
Figure BDA0003666994500000096
wherein ,ci,k The sensor i measures the center point with information for time k,
Figure BDA0003666994500000097
p i,k the direction vector with information is measured for sensor i at time k,
Figure BDA0003666994500000098
θ s the feasible set range is estimated for the measurement band information.
If the loss of the measured data occurs, when k is equal to 1, the order is given
S(p i,k ,c i,k )＝0 (8)
When k ≠ 1, the measurement information at k-1 time is used as the measurement information at k time, so as to
S(p i,k ,c i,k )＝S(p i,k-1 ,c i,k-1 ) (9)
And step 104, obtaining local state estimation of each local data analyzer at the k moment according to the state equation of the state space expression in the step 102 and the measurement band information in the step 103:
solving a predicted fully-symmetrical multi-cell space at the moment k +1 by using a state equation in a state space expression, and converting the predicted fully-symmetrical multi-cell space into a predicted positive multi-cell space; then, the positive multi-cell space at the k +1 moment is solved by using the measurement band information at the k moment, and the local state estimation of each local data analyzer is completed;
according to the state equation of the state space expression of the formula (1) and the measurement band information of the formula (7), local state estimation of each cloud-side corresponding local data analyzer at the k moment is obtained:
in the system state equation of equation (1), the process noise is confined to an unknown but bounded range, i.e., let the process noise w k P (0, g), g being the boundary value of the noise;
using a fully symmetric multi-cell space Z (p) without taking into account process noise interference k ,H k ) Positive multi-cell space representing k time
Figure BDA0003666994500000101
Namely, it is
Figure BDA0003666994500000102
H k ＝diag{d k } (12)
Mixing Z (p) k ,H k ) Method for solving predicted fully-symmetrical multi-cell space at k +1 moment by substituting equation of state in formula (1)
Figure BDA0003666994500000103
Figure BDA0003666994500000104
Figure BDA0003666994500000105
wherein ,
Figure BDA0003666994500000106
to predict the center point of the fully symmetric multi-cellular space,
Figure BDA0003666994500000107
generating a matrix for predicting a fully symmetric multi-cell space;
at this time is
Figure BDA0003666994500000108
Adding process noise w k P (0, g), and using the most compact predictive positive multi-cellular space
Figure BDA0003666994500000109
De-wrapping the predicted fully symmetric multi-cellular space,
Figure BDA00036669945000001010
the calculation formula of (2) is as follows:
Figure BDA00036669945000001011
Figure BDA00036669945000001012
wherein ,
Figure BDA00036669945000001013
to predict the center point of the positive multi-cell space,
Figure BDA00036669945000001014
to predict the generator matrix for the positive multi-cell space,
Figure BDA00036669945000001015
is that the diagonal value is equal to
Figure BDA00036669945000001016
The diagonal matrix of (a) is,
Figure BDA00036669945000001017
is that the diagonal value is equal to
Figure BDA00036669945000001018
Diagonal matrix of n H Is composed of
Figure BDA00036669945000001019
The dimension of the matrix, | | | | represents the norm, l, h are intermediate variables.
Predicting the positive multi-cellular space determined in the formula (15) and the formula (16)
Figure BDA00036669945000001020
Decomposition into n band-constrained equations
Figure BDA00036669945000001021
Figure BDA00036669945000001022
wherein xj To predict the feasible set variables for the j-th dimension of the positive multi-cell space,
Figure BDA00036669945000001023
to predict the feasible set minimum for the j-th dimension of the positive multi-cell space,
Figure BDA00036669945000001024
j is a dimension variable for predicting the maximum value of the feasible set of the j dimension of the positive multi-cell space;
the measurement band information S (p) of the i-th data analyzer at time k obtained by equation (7) is obtained i,k ,c i,k ) Updating to obtain positive multi-cell space at k +1 time
Figure BDA00036669945000001025
Comprises the following steps:
Figure BDA00036669945000001026
Figure BDA00036669945000001027
wherein ,
Figure BDA0003666994500000111
Figure BDA0003666994500000112
Figure BDA0003666994500000113
wherein xs For feasible set variables within the constraint, n denotes the intersection, α j (k +1) is the maximum value of the jth dimension of the feasible set range,
α j+n (k +1) is the minimum value of the jth dimension of the feasible set range,max represents taking the maximum value, min represents taking the minimum value,
Figure BDA0003666994500000114
representing the transpose of the jth dimension of the unit diagonal matrix,
Figure BDA0003666994500000115
is the value of the jth dimension of the center point of the positive multi-cell space,
Figure BDA0003666994500000116
the value of the jth dimension of the matrix is generated for the positive multi-cell space,
Figure BDA0003666994500000117
is that the diagonal value is equal to
Figure BDA0003666994500000118
The diagonal matrix of (a).
Figure BDA0003666994500000119
Final local state estimation results for each data analyzer.
Step 105, transmitting the local state estimation obtained by each local data analyzer to a global data analyzer, calculating the final global state estimation by utilizing Minkowski, transmitting the global state estimation result to each local data analyzer, and transmitting each local data analyzer to each corresponding cloud end;
local State estimation results for each data Analyzer according to equations (17) to (22)
Figure BDA00036669945000001110
Solving for M positive multi-cell spaces and for global positive multi-cell spaces using Minkowski's sum in a total data analyzer
Figure BDA00036669945000001111
Figure BDA00036669945000001112
Is defined as:
Figure BDA00036669945000001113
positive multi-cell space with M local estimates
Figure BDA00036669945000001114
Decomposition into n constrained equations by equation (17)
Figure BDA00036669945000001115
The band-constrained equation in the global estimation is defined as
Figure BDA00036669945000001116
Figure BDA00036669945000001117
Then
Figure BDA00036669945000001118
Comprises the following steps:
Figure BDA00036669945000001119
Figure BDA00036669945000001120
wherein ,
Figure BDA00036669945000001121
Figure BDA0003666994500000121
Figure BDA0003666994500000122
wherein xsum Is composed of
Figure BDA0003666994500000123
Variable of feasible set of lines, alpha j,sum (k +1) is the maximum value of the jth dimension of the feasible set range,
α j+n,sum (k +1) is the minimum value of the jth dimension of the feasible set range,
Figure BDA0003666994500000124
is the global positive multi-cell spatial center point,
Figure BDA0003666994500000125
is the value of the jth dimension of the central point of the global positive multi-cell space, d k+1 A generator matrix for the global positive multi-cell space,
Figure BDA0003666994500000126
generating the j-th dimension value of the matrix for the positive multi-cell space, the global positive multi-cell space
Figure BDA0003666994500000127
The represented feasible set range is a final global estimation result, the global state estimation result is transmitted to each data analyzer, and the data analyzer transmits the global state estimation result to each original cloud.
And 106, transmitting the measurement data acquired by each local data analyzer, the local state estimation result obtained according to the measurement data and the global state estimation result to a backup cloud for backup storage by each local data analyzer.
In order to further ensure the safety of data, in the scheme of the application, the data of each local data analyzer uploads the sensor measurement data acquired by the data of each local data analyzer, local state estimation results obtained according to the measurement data and global estimation results obtained from the global data analyzer to a backup-only cloud for storage; the number of backup clouds can be determined according to actual conditions.
Some steps in the embodiments of the present invention may be implemented by software, and the corresponding software program may be stored in a readable storage medium, such as an optical disc or a hard disk.
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 (7)

1. A high-end battery intelligent factory cloud architecture data storage method is applied to a high-end battery intelligent factory cloud architecture, and is characterized in that the high-end battery intelligent factory cloud architecture comprises the following steps: the system comprises a plurality of sensors, a plurality of local data analyzers and a global data analyzer which are distributed on each production process of the high-end battery and are used for acquiring data in real time;
each sensor corresponds to a cloud end used for storing data acquired and processed by the sensor; the local data analyzers are used for acquiring data acquired by the sensors in real time from the cloud and performing local state estimation, and each local data analyzer sends a local state estimation result acquired by the local data analyzer to the global data analyzer; the global data analyzer is used for carrying out global state estimation according to the local state estimation result of each local data analyzer, transmitting the obtained global state estimation result back to each local data analyzer and further transmitting the global state estimation result to the corresponding cloud end for storage by the local data analyzers.
2. The method according to claim 1, characterized in that it comprises:
step 1: measuring data of all sensors k distributed in the high-end battery production process are obtained and transmitted to corresponding cloud terminals for storage, and meanwhile, the measuring data stored in the cloud terminals are downloaded by using all local data analyzers;
step 2: acquiring a linear state space expression of a high-end battery intelligent factory workshop, and determining an initial positive multi-cell space and an initial full-symmetry multi-cell space; the state space expression comprises a measurement equation and a state equation;
and step 3: according to a measurement equation in the state space expression, each local data analyzer forms the downloaded measurement data at the k moment into measurement band information; judging whether measurement data loss occurs or not, if the measurement data loss occurs, if k is equal to 1, processing is not needed, and if k is not equal to 1, the measurement information at the k-1 moment is taken as the measurement information at the k moment; and transmitting to the corresponding cloud end;
and 4, step 4: and (3) solving local state estimation of each local data analyzer at the k moment according to the state equation of the state space expression in the step 2 and the measurement band information in the step 3:
solving a predicted fully-symmetrical multi-cell space at the moment k +1 by using a state equation in a state space expression, and converting the predicted fully-symmetrical multi-cell space into a predicted positive multi-cell space; then, the positive multi-cell space at the k +1 moment is solved by using the measurement band information at the k moment, and the local state estimation of each local data analyzer is completed;
and 5: transmitting the local state estimation obtained by each local data analyzer to a global data analyzer, calculating the final global state estimation by utilizing Minkowski, transmitting the global state estimation result to each local data analyzer, and transmitting the global state estimation result to each corresponding cloud end by each local data analyzer;
step 6: and each local data analyzer transmits the measurement data acquired by the local data analyzer, the local state estimation result obtained according to the measurement data and the global state estimation result to a backup cloud for backup storage.
3. The method of claim 2, wherein step 2 comprises:
the method comprises the following steps of obtaining an n-dimensional linear state space expression of a high-end battery intelligent factory workshop:
x k+1 ＝Ax k +Bu k +w k (1)
y i,k ＝C i x k +v i,k (2)
equations (1) and (2) are a state equation and a measurement equation of the system, respectively, where k represents time, and k is 1, …, N; x is the number of k Representing the system state at time kA value; x is the number of k+1 Represents the system state value at the time k + 1; u. of k A system input value representing time k; a and B respectively represent a system matrix and an input matrix; w is a k Representing process noise of the system; i represents a distributed sensor number, i is 1, …, and M represents that M distributed sensors in the wireless sensor network perform measurement; y is i,k Representing the data collected by the ith sensor k at the moment; c i An output matrix representing the ith sensor; v. of i,k Representing the measurement noise, v, of the i-th sensor i,k ∈[-δ k ,δ k ]，δ k Is the noise maximum boundary value at time k, -delta k The noise minimum boundary value at the k moment;
defining a positive multi-cell space as:
Figure FDA0003666994490000021
wherein ,
Figure FDA0003666994490000022
is the central point of the regular multicellular space, d is the generating matrix of the regular multicellular space, diag (d) is the diagonal matrix with the diagonal value equal to d, x is the feasible set range of the regular multicellular space, m is the intermediate variable, | | | | | (| ∞ Represents an infinite norm;
defining the full-symmetric multi-cell space as follows:
Figure FDA0003666994490000023
wherein p is the central point of the full-symmetric multi-cell space, H is the generating matrix of the full-symmetric multi-cell space, B n ∈[-1,1] n Is n unit intervals [ -1,1 [)]The unit box is formed by the following components,
Figure FDA0003666994490000024
minkowski and z is an intermediate variable;
positive multicellular at the initial moment determined by the above definitionSpace is
Figure FDA0003666994490000025
And an initial fully symmetric multi-cell space Z (p) 1 ,H 1 ) And make an order
Figure FDA0003666994490000026
Namely, it is
Figure FDA0003666994490000027
diag{d 1 }＝H 1 (6)
wherein ,
Figure FDA0003666994490000028
is the central point of the initial positive multicellular space, d 1 Generating a matrix for an initial positive multi-cell space; p is a radical of 1 Is the central point of the initial holosymmetric multi-cell space, H 1 Is a generator matrix of an initial fully symmetric multi-cell space.
4. The method of claim 2, wherein step 3 comprises:
defining the measurement band information of the local data analyzer according to the downloaded sensor i as S (p) i,k ,c i,k )；
Figure FDA0003666994490000029
wherein ,ci,k The sensor i measures the center point with information for time k,
Figure FDA00036669944900000210
p i,k the direction vector with information is measured for sensor i at time k,
Figure FDA0003666994490000031
θ s estimating a feasible set range for the measurement band information;
if the loss of the measured data occurs, when k is equal to 1, the order is given
S(p i,k ,c i,k )＝0 (8)
When k ≠ 1, the measurement information at k-1 time is used as the measurement information at k time, so as to
S(p i,k ,c i,k )＝S(p i,k-1 ,c i,k-1 ) (9)。
5. The method of claim 4, wherein the step 4 comprises:
solving a predicted fully-symmetrical multi-cell space at the moment k +1 according to a state equation in the state space expression, and converting the predicted fully-symmetrical multi-cell space into a predicted positive multi-cell space; then, the positive multi-cell space at the k +1 moment is solved by using the measurement band information at the k moment, and the local state estimation of each local data analyzer is completed;
according to the state equation of the state space expression of the formula (1) and the measurement band information shown in the formula (7), local state estimation of each cloud-side corresponding local data analyzer at the k moment is obtained:
in the system state equation of equation (1), the process noise is confined to an unknown but bounded range, i.e., let the process noise w k P (0, g), g being the boundary value of the noise;
using a fully symmetric multi-cell space Z (p) without taking into account process noise interference k ,H k ) Positive multi-cell space representing k time
Figure FDA0003666994490000032
Namely, it is
Figure FDA0003666994490000033
H k ＝diag{d k } (12)
Mixing Z (p) k ,H k ) Method for solving predicted fully-symmetrical multi-cell space at k +1 moment by substituting equation of state in formula (1)
Figure FDA0003666994490000034
Figure FDA0003666994490000035
Figure FDA0003666994490000036
wherein ,
Figure FDA0003666994490000037
to predict the center point of the fully symmetric multi-cellular space,
Figure FDA0003666994490000038
generating a matrix for predicting a fully symmetric multi-cell space;
at this time is
Figure FDA0003666994490000039
Adding process noise w k P (0, g), and using the most compact predictive positive multi-cellular space
Figure FDA00036669944900000310
De-wrapping predicts a fully symmetric multi-cellular space,
Figure FDA00036669944900000311
the calculation formula of (2) is as follows:
Figure FDA00036669944900000312
Figure FDA00036669944900000313
wherein ,
Figure FDA00036669944900000314
To predict the center point of the positive multi-cell space,
Figure FDA00036669944900000315
to predict the generator matrix for the positive multi-cell space,
Figure FDA00036669944900000316
is that the diagonal value is equal to
Figure FDA00036669944900000317
The diagonal matrix of (a) is,
Figure FDA00036669944900000318
for diagonal values equal to
Figure FDA00036669944900000319
Diagonal matrix of n H Is composed of
Figure FDA0003666994490000041
Dimension of the matrix, | | | | represents norm, l, h are intermediate variables;
predicting the positive multi-cellular space determined in the formula (15) and the formula (16)
Figure FDA0003666994490000042
Decomposition into n band-constrained equations
Figure FDA0003666994490000043
Figure FDA0003666994490000044
wherein xj To predict the feasible set variables for the j-th dimension of the positive multi-cell space,
Figure FDA0003666994490000045
to predict the feasible set minimum for the j-th dimension of the positive multi-cell space,
Figure FDA0003666994490000046
j is a dimension variable for predicting the maximum value of the feasible set of the j dimension of the positive multi-cell space;
the measurement band information S (p) of the i-th data analyzer at time k obtained by equation (7) is obtained i,k ,c i,k ) Updating to obtain the positive multi-cell space at the k +1 moment
Figure FDA0003666994490000047
Comprises the following steps:
Figure FDA0003666994490000048
Figure FDA0003666994490000049
wherein ,
Figure FDA00036669944900000410
Figure FDA00036669944900000411
Figure FDA00036669944900000412
wherein xs For feasible set variables within the constraint, n denotes the intersection, α j (k +1) is the maximum value of the jth dimension of the feasible set range,
α j+n (k +1) is the minimum value of the jth dimension of the feasible set range, and max represents taking the maximum valueThe value, min, represents taking the minimum value,
Figure FDA00036669944900000413
representing the transpose of the jth dimension of the unit diagonal matrix,
Figure FDA00036669944900000414
is the value of the jth dimension of the center point of the positive multi-cell space,
Figure FDA00036669944900000415
the value of the jth dimension of the matrix is generated for the positive multi-cell space,
Figure FDA00036669944900000416
is that the diagonal value is equal to
Figure FDA00036669944900000417
A diagonal matrix of (a);
Figure FDA00036669944900000418
final local state estimation results for each data analyzer.
6. The method of claim 5, wherein the step 5 comprises:
local State estimation results for each data Analyzer according to equations (17) to (22)
Figure FDA00036669944900000419
Solving for M positive multi-cell spaces and for global positive multi-cell spaces using Minkowski's sum in a global data analyzer
Figure FDA00036669944900000420
Figure FDA00036669944900000421
Is defined as:
Figure FDA00036669944900000422
positive multi-cell space with M local estimates
Figure FDA0003666994490000051
Decomposition into n constrained equations by equation (17)
Figure FDA0003666994490000052
The band-constrained equation in the global estimation is defined as
Figure FDA0003666994490000053
Figure FDA0003666994490000054
Then
Figure FDA0003666994490000055
Comprises the following steps:
Figure FDA0003666994490000056
Figure FDA0003666994490000057
wherein ,
Figure FDA0003666994490000058
Figure FDA0003666994490000059
Figure FDA00036669944900000510
wherein xsum Is composed of
Figure FDA00036669944900000511
Variable of feasible set of lines, alpha j,sum (k +1) is the maximum value of the jth dimension of the feasible set range;
α j+n,sum (k +1) is the minimum value of the jth dimension of the feasible set range,
Figure FDA00036669944900000512
is the global positive multi-cell spatial center point,
Figure FDA00036669944900000513
is the value of the jth dimension of the central point of the global positive multi-cell space, d k+1 A generator matrix for the global positive multi-cell space,
Figure FDA00036669944900000514
generating the j-th dimension value of the matrix for the positive multi-cell space, the global positive multi-cell space
Figure FDA00036669944900000515
The represented feasible set range is a final global estimation result, the global state estimation result is transmitted to each local data analyzer, and each local data analyzer transmits the global state estimation result to each corresponding cloud.
7. The method of claim 6, wherein the local data analyzer and the global data analyzer are implemented using a computer server.
CN202210614506.6A 2022-05-27 2022-05-27 Intelligent factory cloud-level architecture data storage method for high-end battery Active CN115022348B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210614506.6A CN115022348B (en) 2022-05-27 2022-05-27 Intelligent factory cloud-level architecture data storage method for high-end battery

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210614506.6A CN115022348B (en) 2022-05-27 2022-05-27 Intelligent factory cloud-level architecture data storage method for high-end battery

Publications (2)

Publication Number Publication Date
CN115022348A true CN115022348A (en) 2022-09-06
CN115022348B CN115022348B (en) 2023-04-28

Family

ID=83071678

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210614506.6A Active CN115022348B (en) 2022-05-27 2022-05-27 Intelligent factory cloud-level architecture data storage method for high-end battery

Country Status (1)

Country Link
CN (1) CN115022348B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170147922A1 (en) * 2015-11-23 2017-05-25 Daniel Chonghwan LEE Filtering, smoothing, memetic algorithms, and feasible direction methods for estimating system state and unknown parameters of electromechanical motion devices
CN109543143A (en) * 2019-01-28 2019-03-29 杭州电子科技大学 The Multi-sensor Fusion estimation method of non-linear belt bias system
CN112260867A (en) * 2020-10-21 2021-01-22 山东科技大学 State estimation method of event-triggered transmission complex network based on collective member estimation
CN113701742A (en) * 2021-08-24 2021-11-26 吕太之 Mobile robot SLAM method based on cloud and edge fusion calculation
CN113950018A (en) * 2021-10-13 2022-01-18 华东理工大学 Asynchronous multi-sensor network system and global ellipsoid state estimation method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170147922A1 (en) * 2015-11-23 2017-05-25 Daniel Chonghwan LEE Filtering, smoothing, memetic algorithms, and feasible direction methods for estimating system state and unknown parameters of electromechanical motion devices
CN109543143A (en) * 2019-01-28 2019-03-29 杭州电子科技大学 The Multi-sensor Fusion estimation method of non-linear belt bias system
CN112260867A (en) * 2020-10-21 2021-01-22 山东科技大学 State estimation method of event-triggered transmission complex network based on collective member estimation
CN113701742A (en) * 2021-08-24 2021-11-26 吕太之 Mobile robot SLAM method based on cloud and edge fusion calculation
CN113950018A (en) * 2021-10-13 2022-01-18 华东理工大学 Asynchronous multi-sensor network system and global ellipsoid state estimation method

Also Published As

Publication number Publication date
CN115022348B (en) 2023-04-28

Similar Documents

Publication Publication Date Title
CN111898247B (en) Landslide displacement prediction method, landslide displacement prediction equipment and storage medium
WO2023134188A1 (en) Index determination method and apparatus, and electronic device and computer-readable medium
CN114936504A (en) Equipment residual life prediction method and system based on Bayesian multi-source data fusion
CN113139605A (en) Power load prediction method based on principal component analysis and LSTM neural network
CN112990587A (en) Method, system, equipment and medium for accurately predicting power consumption of transformer area
CN115908051A (en) Method for determining energy storage capacity of power system
CN116485031A (en) Method, device, equipment and storage medium for predicting short-term power load
CN116930609A (en) Electric energy metering error analysis method based on ResNet-LSTM model
CN117289668B (en) Distributed speed reducer network cooperative control method, device, equipment and storage medium
CN116706907B (en) Photovoltaic power generation prediction method based on fuzzy reasoning and related equipment
CN113886454A (en) Cloud resource prediction method based on LSTM-RBF
CN113222263A (en) Photovoltaic power generation power prediction method based on long-term and short-term memory neural network
CN117113086A (en) Energy storage unit load prediction method, system, electronic equipment and medium
CN116207766B (en) Dynamic threshold control method and system for lithium battery energy storage system
CN116882079A (en) Water pump characteristic curve self-adaptive calibration and prediction method
CN115022348B (en) Intelligent factory cloud-level architecture data storage method for high-end battery
Lin et al. Hpt-rl: Calibrating power system models based on hierarchical parameter tuning and reinforcement learning
CN114093433B (en) Observer-based method and system for evaluating prediction precision of single-ton energy consumption in rectification process
CN115907228A (en) Short-term power load prediction analysis method based on PSO-LSSVM
CN114741822A (en) Method, system and device for predicting power failure probability of power distribution network under natural disasters
CN109753018B (en) Error compensation system and dynamic compensation method based on cloud intelligence
CN115456168B (en) Training method of reinforcement learning model, energy consumption determining method and device
CN117609737B (en) Method, system, equipment and medium for predicting health state of inertial navigation system
CN117521907A (en) Photovoltaic power generation power interval prediction method considering photovoltaic output and meteorological elements
CN117057458A (en) Task sensor precision prediction method based on ensemble learning

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