CN112686389A - Estimation method and estimation device for optimal value of equipment parameter - Google Patents
Estimation method and estimation device for optimal value of equipment parameter Download PDFInfo
- Publication number
- CN112686389A CN112686389A CN202011566343.6A CN202011566343A CN112686389A CN 112686389 A CN112686389 A CN 112686389A CN 202011566343 A CN202011566343 A CN 202011566343A CN 112686389 A CN112686389 A CN 112686389A
- Authority
- CN
- China
- Prior art keywords
- data
- matrix
- value
- equipment
- estimating
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000012549 training Methods 0.000 claims abstract description 88
- 238000004140 cleaning Methods 0.000 claims abstract description 24
- 239000011159 matrix material Substances 0.000 claims description 106
- 238000012795 verification Methods 0.000 claims description 26
- 238000004422 calculation algorithm Methods 0.000 claims description 23
- 238000003860 storage Methods 0.000 claims description 23
- 230000002159 abnormal effect Effects 0.000 claims description 22
- 230000015654 memory Effects 0.000 claims description 19
- 238000001914 filtration Methods 0.000 claims description 17
- 238000012216 screening Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 238000007670 refining Methods 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 abstract description 8
- 230000005540 biological transmission Effects 0.000 description 6
- 238000004519 manufacturing process Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000010200 validation analysis Methods 0.000 description 3
- 238000005034 decoration Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an estimation method and an estimation device for an optimal value of an equipment parameter, wherein the estimation method for the optimal value of the equipment parameter comprises the following steps: acquiring and storing historical data of equipment parameters; cleaning the historical data to obtain model training data; establishing a training model according to the model training data; and acquiring real-time data of the equipment parameters, and estimating the optimal values of the equipment parameters according to the training model. According to the method, the optimal value of the equipment parameter is estimated, modeling can be performed without predicting the equipment mechanism, and the obtained optimal value is more reasonable and higher in accuracy.
Description
Technical Field
The invention relates to the field of industrial automation control, in particular to an estimation method and an estimation device for an optimal value of an equipment parameter.
Background
In the field of industrial control, when the safety, economy and stability of equipment are researched, the optimal values of various parameters of the equipment under a specific operation condition are often required to be used as basic data for deep analysis. For example, when equipment fault diagnosis is performed, by comparing the difference between the actual value and the estimated optimal value, the abnormal condition of the equipment can be found in time, and larger loss is avoided. Once the estimated value is inaccurate, the false alarm or the false alarm occurs, which affects the reliability of the whole system and causes huge loss or increase of labor cost.
In the prior art, various parameters of equipment, such as temperature, pressure, voltage, current, rotating speed, vibration and the like, are generally acquired through sensors, and the parameter set is { p }1,p2,...,pnP. According to the mechanism of the equipment, selecting several main parameters i with maximum correlation with working conditions1,i2,...,inAnd recording the index as I (I belongs to P) as a working condition index. Determining the value range of each working condition index according to the mechanism model, and adding each index inIs divided into knAnd (4) section. Each working condition index is a section of any value range cell, and as a working condition, the number of all possible working conditions is k1*k2*...*knAnd (4) seed preparation. And under the same working condition, when the performance is optimal, recording the data of each parameter at the moment as the optimal value of the current working condition.
The defects of the prior art are as follows: 1. the working condition index selection and the optimal judgment condition are closely related to a mechanism model of the equipment, sometimes even the most appropriate working condition index is difficult to select, and the optimal judgment condition is difficult to determine; 2. the method is characterized in that the working condition indexes are segmented into a working condition library generated by an exhaustion method, a large number of working conditions can not occur actually, and a lot of unnecessary calculation is performed; 3. since the specific parameters are used as the indexes of the working conditions, the parameters cannot be optimized.
In view of the foregoing, it is desirable to provide an estimation method and an estimation apparatus for an optimal value of a device parameter, which have higher accuracy, universality, operability, and independence and are more optimized.
Disclosure of Invention
The embodiment of the invention provides an estimation method and an estimation device for an optimal value of a device parameter, which are used for solving the technical problems that the prior art is difficult to determine an optimal judgment condition, large in useless calculation amount and the like.
According to an aspect of the embodiments of the present invention, a method for estimating an optimal value of a device parameter is provided, including: acquiring and storing historical data of equipment parameters; cleaning the historical data to obtain model training data; establishing a training model according to the model training data; and acquiring real-time data of the equipment parameters, and estimating the optimal values of the equipment parameters according to the training model. The equipment parameters comprise different parameters of the same equipment, the same or different parameters of equipment with different models and different parameters of different equipment, and can be selected and set according to actual application.
Optionally, the data cleansing comprises: and screening the historical data according to a preset time period and a value-taking interval. Selecting historical data of one or more time periods as source data, and carrying out value taking according to a set value taking interval to obtain model training data. The value space may be a time space, that is, historical data of a preset time period is screened according to the preset time space.
Optionally, the data cleansing further comprises: and filtering the data obtained by screening according to a preset filter, wherein the filter comprises one or more of a constant data filter, a timeout data filter and a parameter limit filter. Due to the fact that data acquisition equipment is abnormal or a network is interrupted, data cannot be updated or updating is overtime possibly caused, or parameter data are seriously deviated from normal values due to the fact that the equipment runs abnormally, in order to filter invalid data, one or more of filters such as a constant data filter, an overtime data filter and a parameter limit filter can be applied to filter the data screened according to a preset time period and a value-taking interval, and model training data are obtained.
Optionally, the data cleansing further comprises: and filtering the abnormal data according to the filtering instruction. The filtering instructions may be manually selected or input by an operator.
Optionally, the method for establishing the training model includes: converting the model training data into a data matrix A, wherein the data of the same equipment parameter at different moments are used as rows of the data matrix A, and the data of different equipment parameters at the same moment are used as columns of the data matrix A; converting the data matrix A into a refined data matrix B by adopting a data refining algorithm; converting the refined data matrix B into a feature matrix C by adopting a feature extraction algorithm; and storing the fine data matrix B and the feature matrix C. The training model includes the culled data matrix B and the feature matrix C.
Optionally, the elements of the data matrix a are arranged in a row order, and the elements are separated by separators, which are denoted as PARAMS, and stored. Wherein the separator is preferably ",".
Optionally, the data culling algorithm comprises: if the number of columns of the data matrix A is smaller than or equal to a preset value, taking the data matrix A as a carefully selected data matrix B; if the number of columns of the data matrix A is larger than the preset value, sorting the data of each equipment parameter according to the size, and selecting a median, an average value, a minimum value and a maximum value until the selected data amount reaches the preset value; if the sum of the numbers of the median, the average value, the minimum value and the maximum value in the data of one or more equipment parameters in the data matrix A is less than the preset value, secondary data selection is carried out according to a preset rule so that the sum of the selected data amount reaches the preset value. The preset rule may be that secondary screening is performed on the remaining data according to a preset value-taking interval, that is, secondary data selection is performed on the remaining data according to the preset value-taking interval under the condition that the sum of the numbers of the median, the average, the minimum and the maximum in the data of one or more device parameters in the data matrix a is less than the preset value, so that the sum of the selected data amounts reaches the preset value. The preset value can be set according to practical application.
Optionally, the feature extraction algorithm is:
C＝(BT×B)-1。
optionally, the pre-estimating comprises: converting the real-time data into a matrix U; obtaining an optimal value matrix V of the equipment parameters by adopting a pre-estimation algorithm, wherein the pre-estimation algorithm comprises the following steps:
v ═ B × (W ÷ sum); wherein W is C (B)T×U)，sum＝ΣWij，WijIs the element of the matrix W, wherein i and j are positive integers, which respectively represent the row number and the column number of the matrix W, i.e. WijRepresenting the element in the ith row and the jth column of the matrix W.
Optionally, the method for estimating the optimal value of the device parameter further includes verifying the training model after the training model is established, where the verifying includes forward verifying and reverse verifying, the forward verifying selects normal data of the device parameter to verify the training model, and the reverse verifying selects abnormal data of the device parameter to verify the training model.
Optionally, the method of verifying includes: converting the selected data for verifying the training model into a matrix X; obtaining a verification matrix Y by adopting a pre-estimation algorithm, wherein the pre-estimation algorithm is as follows:
y ═ B × (Z ÷ sum'); wherein Z is C (B)T×X)，sum′＝ΣZij，ZijIs the element of matrix Z, where i and j are positive integers, and respectively represent the row number and column number of matrix Z, i.e. ZijAn element representing the ith row and the jth column in the matrix Z;
and comparing and analyzing each value in the verification matrix Y with corresponding data in the matrix X.
Optionally, the comparing and analyzing each value in the validation matrix Y and the corresponding data in the matrix X includes: comparing each element in the matrix X with the corresponding element in the verification matrix Y, and respectively counting the number of the elements with larger deviation in each row; comparing the number of elements with larger deviation in each row with the number of columns of the matrix X. The optional technical scheme adopts an abnormal time counting method to verify the training model. Specifically, if the deviation between a certain element in the matrix X and the corresponding element in the verification matrix Y is large, the data of the element in the matrix X is considered to be abnormal, and the count at the abnormal moment is increased by 1; for forward verification, if the number of abnormal moments is far smaller than the column number of the matrix X, the training model is considered to be better; for reverse validation, if the number of abnormal moments is close to the number of columns of matrix X, the training model is considered to be better.
According to another aspect of the embodiments of the present invention, there is also provided an estimation apparatus for an optimal value of a device parameter, including: the acquisition unit is used for acquiring historical data and real-time data of equipment parameters; the storage unit is used for storing the historical data and the real-time data; the cleaning unit is used for cleaning the historical data; the model training unit is used for establishing a training model according to the model training data; and the estimation unit is used for estimating the optimal value of the equipment parameter according to the training model.
Optionally, the cleaning unit includes a screening subunit, configured to screen the historical data according to a preset time period and a value interval.
Optionally, the cleaning unit further includes a filtering subunit, configured to filter the data obtained by the screening according to a preset filter, where the filter includes one or more of a constant data filter, a timeout data filter, and a parameter limit filter.
Optionally, the filtering subunit is further configured to filter the abnormal data according to the filtering instruction.
Optionally, the storage unit is further configured to store the training model.
Optionally, the apparatus for estimating the optimal value of the device parameter further includes a verification unit, configured to verify the training model.
According to another aspect of the embodiments of the present invention, there is also provided a storage medium, where the storage medium includes a stored program, and the program executes the method for estimating the optimal value of the device parameter when running.
According to another aspect of the embodiments of the present invention, there is also provided an electronic apparatus, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method for estimating the optimal value of the device parameter through the computer program.
In the embodiment of the invention, the optimal value of the equipment parameter is estimated by adopting the method, modeling can be carried out without predicting the mechanism of the equipment, and the obtained optimal value is more reasonable and higher in accuracy. In addition, the invention screens normal data for machine training through data cleaning, and the established training model is more accurate and has strong operability. Moreover, the training model of the invention is independent of the equipment mechanism and can be applied to any equipment, thereby greatly improving the universality of the method.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
FIG. 1 is a flow chart of an alternative method for estimating an optimal value of a device parameter according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an alternative apparatus for estimating an optimal value of a device parameter according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "comprises" and "comprising," and any variations thereof, in the description and claims of the present invention and the above-described drawings, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
According to an aspect of the embodiments of the present invention, an embodiment of an estimation method for an optimal value of a device parameter is provided, as shown in fig. 1, the method may include the following steps:
102, acquiring and storing historical data of equipment parameters;
106, establishing a training model according to the model training data;
and step 108, acquiring real-time data of the equipment parameters, and estimating the optimal values of the equipment parameters according to the training model.
In step 104, the data cleansing includes: and screening the historical data according to a preset time period and a value-taking interval. Selecting historical data of one or more time periods as source data, and carrying out value taking according to a set value taking interval to obtain model training data. The value space may be a time space, that is, historical data of a preset time period is screened according to the preset time space.
The equipment parameters comprise different parameters of the same equipment, the same or different parameters of equipment with different models and different parameters of different equipment, and can be selected and set according to actual application. For example, a device parameter may be a parameter of the same device, such as temperature, pressure, voltage, current, rotational speed, vibration, etc.; or different parameters of different devices in the same production line; the same parameters of the same type of equipment of different enterprises can be obtained; and so on.
As a specific embodiment, historical data of the last 6 months or even one year is used as source data, and 10 minutes are used as value intervals for data cleaning to obtain model training data. Alternatively, the preset time period may be a plurality of time periods, or may be one time period.
As a preferred embodiment, in step 104, the data cleansing further includes: and filtering the data obtained by screening according to a preset filter, wherein the filter comprises one or more of a constant data filter, a timeout data filter and a parameter limit filter. Due to the fact that data acquisition equipment is abnormal or a network is interrupted, data cannot be updated or updating is overtime possibly caused, or parameter data are seriously deviated from normal values due to the fact that the equipment runs abnormally, in order to filter invalid data, one or more of filters such as a constant data filter, an overtime data filter and a parameter limit filter can be applied to filter the data screened according to a preset time period and a value-taking interval, and model training data are obtained.
As a preferred embodiment, in step 104, the data cleansing further includes: and filtering the abnormal data according to the filtering instruction. The filtering instructions may be manually selected or input by an operator.
As a specific embodiment, 10 minutes are taken as a value interval, historical data of equipment parameters in the past 6 months are screened, screened data are displayed in a dot-line graph, an operator can manually select macroscopic abnormal data to filter, a time value corresponding to the filtered data is synchronized with a background, and the background deletes corresponding data, so that the data displayed at the front end is consistent with the data stored in the background.
In step 106, the method for establishing the training model includes:
s1061, converting the model training data into a data matrix A, wherein data of the same equipment parameter at different moments are used as rows of the data matrix A, and data of different equipment parameters at the same moment are used as columns of the data matrix A;
specifically, if the model training data includes data of n device parameters at m moments, the model training data is converted into an n × m data matrix a;
s1062, converting the data matrix A into a refined data matrix B by adopting a data refining algorithm;
specifically, if the column number m of the data matrix A is less than or equal to a preset value, taking the data matrix A as a carefully selected data matrix B; if the number m of the columns of the data matrix A is larger than the preset value, sorting the data of each equipment parameter according to the size, and selecting a median, an average value, a minimum value and a maximum value until the selected data amount reaches the preset value; if the sum of the numbers of the median, the average value, the minimum value and the maximum value in the data of one or more equipment parameters in the data matrix A is less than the preset value, secondary data selection is carried out according to a preset rule so that the sum of the selected data amount reaches the preset value. The preset rule may be that secondary screening is performed on the remaining data according to a preset value-taking interval, that is, secondary data selection is performed on the remaining data according to the preset value-taking interval under the condition that the sum of the numbers of the median, the average, the minimum and the maximum in the data of one or more device parameters in the data matrix a is less than the preset value, so that the sum of the selected data amounts (the number of the median, the number of the average, the number of the minimum, the number of the maximum and the data amount obtained by the secondary data selection) reaches the preset value. The preset value can be set according to practical application. The purpose of data refinement in this embodiment is to preserve the main characteristic data (including median, mean, minimum, maximum) of each parameter, and if there are gaps, then select data to fill in the gaps at intervals in the remaining timestamps.
As a preferred embodiment, the preset value is default to 1000, and may be adjusted according to the number n of the device parameters, for example, when a certain multiple of the number of the device parameters is greater than the default value 1000, the preset value is adjusted to a certain multiple of the number of the device parameters, and the multiple may be determined according to actual needs, and is preferably 4 times. The arrangement can ensure that main characteristic data of each parameter, including median, average value, minimum value and maximum value, are reserved in the refined data matrix B;
s1063, converting the refined data matrix B into a feature matrix C by adopting a feature extraction algorithm;
specifically, the feature extraction algorithm is as follows: c ═ BT×B)-1；
S1064, storing the refined data matrix B and the feature matrix C.
Optionally, the elements of the data matrix a are arranged in a row order, and the elements are separated by separators, which are denoted as PARAMS, and stored. Wherein the separator is preferably ",". The training model includes PARAMS, the cull data matrix B, and the feature matrix C.
In step 108, the estimating includes:
s1081, converting the real-time data into a matrix U, wherein data of the same equipment parameter at different moments are used as rows of the matrix U, and data of different equipment parameters at the same moment are used as columns of the matrix U;
s1082, obtaining an optimal value matrix V of the device parameters by adopting a pre-estimation algorithm, wherein the pre-estimation algorithm is as follows:
v ═ B × (W ÷ sum); wherein W is C (B)T×U)，sum＝ΣWij，WijIs the element of the matrix W, wherein i and j are positive integers, which respectively represent the row number and the column number of the matrix W, i.e. WijRepresenting the element in the ith row and the jth column of the matrix W. Each element in the matrix V is the optimal value of each corresponding element in the matrix U.
As a preferred embodiment, step 106 is followed by:
and 106', verifying the training model, wherein the verification comprises forward verification and reverse verification, the forward verification selects normal data of the equipment parameters to verify the training model, and the reverse verification selects abnormal data of the equipment parameters to verify the training model.
Specifically, the method for verifying comprises the following steps:
(1) converting the selected data for verifying the training model into a matrix X;
(2) obtaining a verification matrix Y by adopting a pre-estimation algorithm, wherein the pre-estimation algorithm is as follows:
y ═ B × (Z ÷ sum'); wherein Z is C (B)T×X)，sum′＝ΣZij，ZijIs the element of matrix Z, where i and j are positive integers, and respectively represent the row number and column number of matrix Z, i.e. ZijAn element representing the ith row and the jth column in the matrix Z;
(3) comparing and analyzing each value in the verification matrix Y with corresponding data in the matrix X, specifically:
i) comparing each element in the matrix X with the corresponding element in the verification matrix Y, and respectively counting the number of the elements with larger deviation in each row;
ii) comparing the number of elements with larger deviations per row with the number of columns of the matrix X.
The optional technical scheme adopts an abnormal time counting method to verify the training model. Specifically, if the deviation between a certain element in the matrix X and the corresponding element in the verification matrix Y is large, the data of the element in the matrix X is considered to be abnormal, and the count at the abnormal moment is increased by 1; for forward verification, if the number of abnormal moments is far smaller than the column number of the matrix X, the training model is considered to be better; for reverse validation, if the number of abnormal moments is close to the number of columns of matrix X, the training model is considered to be better. Furthermore, the model is scored according to the forward verification result and the reverse verification result. The higher the score of the model is, the better the model is, the model with the score not reaching the standard cannot be applied, and the training model can be established in the model training process again and scored again until the model reaches the standard and can be applied. According to the embodiment, the training model is verified by the verification method, and the accuracy of the training model is checked, so that the reasonable training model can be obtained, and the more accurate optimal value of the equipment parameter can be obtained.
In the embodiment of the invention, the optimal value of the equipment parameter is estimated by adopting the method, modeling can be carried out without predicting the mechanism of the equipment, and the obtained optimal value is more reasonable and higher in accuracy. In addition, the invention screens normal data for machine training through data cleaning, and the established training model is more accurate and has strong operability. Moreover, the training model of the invention is independent of the equipment mechanism and can be applied to any equipment, thereby greatly improving the universality of the method.
The estimation method for the optimal value of the device parameter in the embodiment has wide application, including but not limited to the following application scenarios:
the hidden danger of the equipment is discovered in time. And estimating the optimal value of the equipment parameter regularly according to the real-time data of the equipment, and considering that the data at a certain moment is abnormal when the deviation between the individual parameter at the moment and the optimal value is overlarge. When the real-time data of the equipment continuously generates abnormity or the abnormity occurs at high frequency, the health condition of the equipment may be in a problem. In view of the above, can discover equipment hidden danger in advance, overhaul equipment, avoid appearing major incident etc..
And evaluating the performance of the equipment. Judging the running stability of the equipment according to the comparison between the real-time data of the equipment and the optimal value of the real-time data of the equipment; in addition, for the same equipment of different enterprises in the same industry, the performance of the equipment can be transversely compared through real-time data and optimal value comparison.
And thirdly, joint debugging of the equipment. According to the real-time data of different equipment of the production line and the corresponding optimal values of the real-time data, joint debugging is carried out on the related equipment, the production line efficiency is improved, and the process cost is reduced.
And fourthly, assisting in intelligent manufacturing. According to the comparative analysis of the real-time data and the corresponding optimal values of the same equipment of the same production line of different enterprises, an auxiliary equipment manufacturer adjusts or improves the equipment design or the manufacturing process, so that the equipment can fully exert the performance when in use, and all parameters of the equipment are closer to the theoretical optimal values.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.
Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
According to another aspect of the embodiment of the present invention, there is further provided an estimation apparatus for estimating an optimal value of a device parameter, where the estimation apparatus for estimating an optimal value of a device parameter may be a server, or may be a terminal device having functions of calculation, storage, communication, display, and the like. Fig. 2 is a schematic diagram of an alternative apparatus for estimating an optimal value of a device parameter according to an embodiment of the present invention, as shown in fig. 2, the apparatus may include: an acquisition unit 201, a storage unit 203, a cleaning unit 205, a model training unit 207, and an estimation unit 209, wherein,
an obtaining unit 201, configured to obtain historical data and real-time data of device parameters;
a storage unit 203, configured to store the historical data and the real-time data;
a cleaning unit 205, configured to perform data cleaning on the historical data;
a model training unit 207, configured to establish a training model based on the model training data;
and the estimating unit 209 is configured to estimate the optimal value of the equipment parameter according to the training model.
It should be noted that the obtaining unit 201 and the storing unit 203 in this embodiment may be configured to execute step 102 in this embodiment, the cleaning unit 205 in this embodiment may be configured to execute step 104 in this embodiment, the model training unit 207 in this embodiment may be configured to execute step 106 in this embodiment, and the cleaning unit 209 in this embodiment may be configured to execute step 108 in this embodiment. The following description is repeated by analogy.
Optionally, the cleaning unit 205 includes a screening subunit, configured to screen the historical data according to a preset time period and a value interval.
As a preferred embodiment, the cleaning unit 205 further comprises a filtering subunit for filtering the data obtained by the screening according to preset filters, wherein the filters include one or more of a constant data filter, a timeout data filter, and a parameter limit filter. And the filtering subunit is also used for filtering the abnormal data according to the filtering instruction.
The storage unit 203 is further configured to store the training model.
As a preferred embodiment, the estimation apparatus for the optimal value of the device parameter further includes a verification unit, configured to verify the training model.
It should be noted here that the modules described above are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the disclosure of the above embodiments. It should be noted that the modules described above as a part of the apparatus may operate in a hardware environment according to the embodiment of the present invention, and may be implemented by software or hardware.
According to another aspect of the embodiments of the present invention, there is also provided a server or a terminal for implementing the method for estimating the optimal value of the device parameter, where the method includes: the one or more processors, memory, and transmission means may also include input-output devices.
The memory may be used to store software programs and modules, such as program instructions/modules corresponding to the method and apparatus for estimating an optimal value of a device parameter in the embodiments of the present invention, and the processor executes various functional applications and data processing by operating the software programs and modules stored in the memory, that is, the method for estimating an optimal value of a device parameter is implemented. The memory may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include memory located remotely from the processor, and these remote memories may be connected to the terminal through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The above-mentioned transmission device is used for receiving or transmitting data via a network, and may also be used for data transmission between a processor and a memory. Examples of the network may include a wired network and a wireless network. In one example, the transmission device includes a Network adapter (NIC) that can be connected to the router via a Network cable and other Network devices to communicate with the internet or a local area Network. In one example, the transmission device is a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.
Wherein the memory is specifically used for storing application programs.
The processor may invoke the memory-stored application program via the transmission means to perform the steps of:
acquiring and storing historical data of equipment parameters;
cleaning the historical data to obtain model training data;
establishing a training model according to the model training data;
and acquiring real-time data of the equipment parameters, and estimating the optimal values of the equipment parameters according to the training model.
Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, and this embodiment is not described herein again.
It will be appreciated by those skilled in the art that a terminal may be any terminal device having computing, storage, communication, display, etc. functionality.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by a program instructing hardware associated with the terminal device, where the program may be stored in a computer-readable storage medium, and the storage medium may include: flash disks, Read-Only memories (ROMs), Random Access Memories (RAMs), magnetic or optical disks, and the like.
The embodiment of the invention also provides a storage medium. Alternatively, in this embodiment, the storage medium may be a program code for executing a method for estimating an optimal value of a device parameter.
Optionally, in this embodiment, the storage medium may be located on at least one of a plurality of network devices in a network shown in the above embodiment.
Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps:
acquiring and storing historical data of equipment parameters;
cleaning the historical data to obtain model training data;
establishing a training model according to the model training data;
and acquiring real-time data of the equipment parameters, and estimating the optimal values of the equipment parameters according to the training model.
Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, and this embodiment is not described herein again.
Optionally, in this embodiment, the storage medium may include, but is not limited to: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
The integrated unit in the above embodiments, if implemented in the form of a software functional unit and sold or used as a separate product, may be stored in the above computer-readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing one or more computer devices (which may be personal computers, servers, network devices, etc.) to execute all or part of the steps of the method according to the embodiments of the present invention.
In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed client may be implemented in other manners. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (12)
1. A method for estimating the optimal value of a device parameter is characterized by comprising the following steps:
acquiring and storing historical data of equipment parameters;
cleaning the historical data to obtain model training data;
establishing a training model according to the model training data;
and acquiring real-time data of the equipment parameters, and estimating the optimal values of the equipment parameters according to the training model.
2. The method for estimating the optimal value of the equipment parameter according to claim 1, wherein the data cleaning comprises:
and screening the historical data according to a preset time period and a value-taking interval.
3. The method of estimating the optimal value of the plant parameter as claimed in claim 2, wherein the data cleaning further comprises:
and filtering the data obtained by screening according to a preset filter, wherein the filter comprises one or more of a constant data filter, a timeout data filter and a parameter limit filter.
4. The method for estimating the optimal value of the equipment parameter according to claim 1, wherein the method for establishing the training model comprises the following steps:
converting the model training data into a data matrix A, wherein the data of the same equipment parameter at different moments are used as rows of the data matrix A, and the data of different equipment parameters at the same moment are used as columns of the data matrix A;
converting the data matrix A into a refined data matrix B by adopting a data refining algorithm;
converting the refined data matrix B into a feature matrix C by adopting a feature extraction algorithm;
and storing the fine data matrix B and the feature matrix C.
5. The method of estimating the optimal value of the plant parameter as claimed in claim 4, wherein the data culling algorithm comprises:
if the number of columns of the data matrix A is smaller than or equal to a preset value, taking the data matrix A as a carefully selected data matrix B;
if the number of columns of the data matrix A is larger than the preset value, sorting the data of each equipment parameter according to the size, and selecting a median, an average value, a minimum value and a maximum value until the selected data amount reaches the preset value; if the sum of the numbers of the median, the average value, the minimum value and the maximum value in the data of one or more equipment parameters in the data matrix A is less than the preset value, secondary data selection is carried out according to a preset rule so that the sum of the selected data amount reaches the preset value.
6. The method for estimating the optimal value of the equipment parameter according to claim 4 or 5, wherein the feature extraction algorithm comprises:
C＝(BT×B)-1。
7. the method of claim 1, wherein the estimating comprises:
converting the real-time data into a matrix U;
obtaining an optimal value matrix V of the equipment parameters by adopting a pre-estimation algorithm, wherein the pre-estimation algorithm comprises the following steps:
V＝B×(W÷sum)；
wherein W is C (B)T×U)，sum＝ΣWij，WijAre the elements of the matrix W.
8. The method for estimating the optimal value of the device parameter according to claim 1, further comprising verifying the training model after establishing the training model, wherein the verifying includes forward verification and reverse verification, the forward verification selects normal data of the device parameter to verify the training model, and the reverse verification selects abnormal data of the device parameter to verify the training model.
9. The method for estimating the optimal value of the equipment parameter according to claim 8, wherein the verifying method comprises:
converting the selected data for verifying the training model into a matrix X;
obtaining a verification matrix Y by adopting a pre-estimation algorithm, wherein the pre-estimation algorithm is as follows:
Y＝B×(Z÷sum′)；
wherein Z is C (B)T×X)，sum′＝ΣZij，ZijIs each element of the matrix Z;
and comparing and analyzing each value in the verification matrix Y with corresponding data in the matrix X.
10. An apparatus for predicting an optimal value of a device parameter, comprising:
the acquisition unit is used for acquiring historical data and real-time data of equipment parameters;
the storage unit is used for storing the historical data and the real-time data;
the cleaning unit is used for cleaning the historical data;
the model training unit is used for establishing a training model according to the model training data;
and the estimation unit is used for estimating the optimal value of the equipment parameter according to the training model.
11. A storage medium, characterized in that the storage medium comprises a stored program, wherein the program when executed performs the method of any of the preceding claims 1 to 9.
12. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the method of any of the preceding claims 1 to 9 by means of the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011566343.6A CN112686389A (en) | 2020-12-25 | 2020-12-25 | Estimation method and estimation device for optimal value of equipment parameter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011566343.6A CN112686389A (en) | 2020-12-25 | 2020-12-25 | Estimation method and estimation device for optimal value of equipment parameter |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112686389A true CN112686389A (en) | 2021-04-20 |
Family
ID=75451807
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011566343.6A Pending CN112686389A (en) | 2020-12-25 | 2020-12-25 | Estimation method and estimation device for optimal value of equipment parameter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112686389A (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104102773A (en) * | 2014-07-05 | 2014-10-15 | 山东鲁能软件技术有限公司 | Equipment fault warning and state monitoring method |
| CN108334674A (en) * | 2018-01-17 | 2018-07-27 | 浙江大学 | A kind of steam turbine high-pressure cylinder method for monitoring operation states based on parameter association intellectual analysis |
| KR20190098484A (en) * | 2018-02-14 | 2019-08-22 | 부산대학교 산학협력단 | System and Method for Fault Isolation in Industrial Processes using CART based variable ranking |
| US20190355152A1 (en) * | 2017-07-28 | 2019-11-21 | Peking University Shenzen Graduate School | Point cloud attribute compression method based on kd tree and optimized graph transformation |
| CN110794305A (en) * | 2019-10-14 | 2020-02-14 | 北京理工大学 | Power battery fault diagnosis method and system |
| CN111275335A (en) * | 2020-01-20 | 2020-06-12 | 华电莱州发电有限公司 | Data-driven slurry circulating pump optimization method and system |
| CN111401565A (en) * | 2020-02-11 | 2020-07-10 | 西安电子科技大学 | DOA estimation method based on machine learning algorithm XGboost |
| US20200242400A1 (en) * | 2019-01-25 | 2020-07-30 | Oath Inc. | Systems and methods for hyper parameter optimization for improved machine learning ensembles |
| CN112001440A (en) * | 2020-08-20 | 2020-11-27 | 苏州鸿哲智能科技有限公司 | Fault diagnosis logic algorithm and system |
| CN112650200A (en) * | 2020-12-25 | 2021-04-13 | 中能融安(北京)科技有限公司 | Equipment fault diagnosis method and device |
-
2020
- 2020-12-25 CN CN202011566343.6A patent/CN112686389A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104102773A (en) * | 2014-07-05 | 2014-10-15 | 山东鲁能软件技术有限公司 | Equipment fault warning and state monitoring method |
| US20190355152A1 (en) * | 2017-07-28 | 2019-11-21 | Peking University Shenzen Graduate School | Point cloud attribute compression method based on kd tree and optimized graph transformation |
| CN108334674A (en) * | 2018-01-17 | 2018-07-27 | 浙江大学 | A kind of steam turbine high-pressure cylinder method for monitoring operation states based on parameter association intellectual analysis |
| KR20190098484A (en) * | 2018-02-14 | 2019-08-22 | 부산대학교 산학협력단 | System and Method for Fault Isolation in Industrial Processes using CART based variable ranking |
| US20200242400A1 (en) * | 2019-01-25 | 2020-07-30 | Oath Inc. | Systems and methods for hyper parameter optimization for improved machine learning ensembles |
| CN110794305A (en) * | 2019-10-14 | 2020-02-14 | 北京理工大学 | Power battery fault diagnosis method and system |
| CN111275335A (en) * | 2020-01-20 | 2020-06-12 | 华电莱州发电有限公司 | Data-driven slurry circulating pump optimization method and system |
| CN111401565A (en) * | 2020-02-11 | 2020-07-10 | 西安电子科技大学 | DOA estimation method based on machine learning algorithm XGboost |
| CN112001440A (en) * | 2020-08-20 | 2020-11-27 | 苏州鸿哲智能科技有限公司 | Fault diagnosis logic algorithm and system |
| CN112650200A (en) * | 2020-12-25 | 2021-04-13 | 中能融安(北京)科技有限公司 | Equipment fault diagnosis method and device |
Non-Patent Citations (3)
| Title |
|---|
| LEE SEOKJIN 等: "Feature Extraction Based on the Non-Negative Matrix Factorization of Convolutional Neural Networks for Monitoring Domestic Activity With Acoustic Signals", 《IEEE ACCESS》, vol. 08, 6 July 2020 (2020-07-06), pages 122384 - 122395, XP011798791, DOI: 10.1109/ACCESS.2020.3007199 * |
| 司加胜 等: "关于机械滚动轴承故障信号诊断仿真", 《计算机仿真》, vol. 36, no. 01, 15 January 2019 (2019-01-15), pages 403 - 408 * |
| 曹海岐: "基于BP神经网络的复合储能式装载机工况识别及模糊控制器优化", 《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》, no. 09, 15 September 2019 (2019-09-15), pages 029 - 27 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111459700B (en) | Equipment fault diagnosis method, diagnosis device, diagnosis equipment and storage medium | |
| CN111427753B (en) | Capacity prediction device based on ARIMA model and control method thereof | |
| CN104866632B (en) | Fault data acquisition methods, device and the terminal of home appliance | |
| CN112650200B (en) | Method and device for diagnosing plant station equipment faults | |
| CN106844171A (en) | Mass operation and maintenance implementation method | |
| CN109818961B (en) | Network intrusion detection method, device and equipment | |
| CN105721187A (en) | Service fault diagnosis method and apparatus | |
| US20090164618A1 (en) | Network system and method of administrating networks | |
| CN114637263B (en) | Abnormal working condition real-time monitoring method, device, equipment and storage medium | |
| CN110932899A (en) | Intelligent fault compression research method and system applying AI | |
| JP2002342182A (en) | Support system for operation management in network system | |
| CN117216481A (en) | Remote monitoring method and system for electric appliance | |
| WO2024104406A1 (en) | Anomaly detection method and cloud network platform | |
| CN116991947B (en) | Automatic data synchronization method and system | |
| CN106951360B (en) | Data statistical integrity calculation method and system | |
| CN112686389A (en) | Estimation method and estimation device for optimal value of equipment parameter | |
| CN111277427B (en) | Data center network equipment inspection method and system | |
| CN116595353A (en) | Remote fault diagnosis and intelligent decision system for camera | |
| CN111538666B (en) | Method, device, computer equipment and storage medium for collecting test results in batches | |
| CN115689320A (en) | Health management method and device for base station operation and computer readable storage medium | |
| CN112015627A (en) | Data acquisition method and related device | |
| CN111740856A (en) | Network communication equipment alarm acquisition abnormity early warning method based on abnormity detection algorithm | |
| WO2020034197A1 (en) | Address identification method, apparatus and system, and storage medium, processor and terminal | |
| CN114511047B (en) | Excavator working mode identification method, computer equipment and medium | |
| CN117221241B (en) | Intelligent switch control process data transmission method and 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 |