CN115766383A - Equipment state evaluation early warning method and system based on algorithm fusion technology - Google Patents
Equipment state evaluation early warning method and system based on algorithm fusion technology Download PDFInfo
- Publication number
- CN115766383A CN115766383A CN202211313681.8A CN202211313681A CN115766383A CN 115766383 A CN115766383 A CN 115766383A CN 202211313681 A CN202211313681 A CN 202211313681A CN 115766383 A CN115766383 A CN 115766383A
- Authority
- CN
- China
- Prior art keywords
- vector
- compensation
- determining
- matrix
- current
- 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
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 77
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000011156 evaluation Methods 0.000 title claims abstract description 30
- 238000005516 engineering process Methods 0.000 title claims abstract description 28
- 230000004927 fusion Effects 0.000 title claims abstract description 26
- 239000013598 vector Substances 0.000 claims abstract description 275
- 239000011159 matrix material Substances 0.000 claims abstract description 116
- 238000012549 training Methods 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 17
- 238000010801 machine learning Methods 0.000 claims description 13
- 238000013459 approach Methods 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 description 20
- 230000000875 corresponding effect Effects 0.000 description 16
- 238000010586 diagram Methods 0.000 description 16
- 238000011161 development Methods 0.000 description 3
- 230000002596 correlated effect Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Abstract
The application relates to an equipment state evaluation early warning method and system based on an algorithm fusion technology, wherein the method comprises the following steps: determining a current observation vector and determining an initial prediction vector corresponding to the current observation vector; determining a similarity coefficient vector according to the similarity operator, the mode matrix and the current observation vector; inputting the similar coefficient vector to a residual prediction model to obtain a current residual prediction vector; determining a final prediction vector according to the initial prediction vector and the current residual prediction vector so as to determine whether to trigger equipment fault alarm; according to the method and the device, residual compensation is carried out on actual prediction data based on the prediction value, so that the deviation between the prediction value and the actual value is effectively reduced, the problem that the prediction residual is increased and false alarm is generated due to the fact that the current data and the historical data have the deviation in the device state assessment early warning method based on the nonlinear state assessment algorithm in the related art is solved, and the accuracy of early warning is improved.
Description
Technical Field
The application relates to the technical field of equipment state monitoring, in particular to an equipment state assessment and early warning method and system based on an algorithm fusion technology.
Background
With the development of new generation information technologies such as big data, cloud computing, internet of things and the like, the integration of the traditional physical system and the information system is further promoted, and the intelligent development of the energy production industry is paid attention by various big energy groups. In order to identify the abnormality of the equipment in the early stage of equipment failure and process the equipment in time to prevent the occurrence of events such as non-stop of the equipment, an equipment state assessment early warning technology based on big data analysis is initially developed and applied. The nonlinear state evaluation algorithm is an intelligent early warning modeling method for equipment faults, which is widely applied, and is widely applied to intelligent early warning modeling of power generation equipment such as gas turbine generator sets, coal-electric sets and fan sets in recent years.
The nonlinear state evaluation algorithm carries out system operation state prediction alarm by comparing the similarity degree of the current data and the historical data (namely, monitoring the similarity degree between the multidimensional signals). In the implementation process, on one hand, due to the limitation of the nonlinear state evaluation algorithm, the too large amount of historical data leads to too slow training process of the historical data, and even results cannot be calculated; on the other hand, under the condition of a certain amount of historical data, all boundary environmental conditions and working conditions of the power generation equipment are difficult to exhaust. Therefore, the prediction residual error of the early warning model based on the nonlinear evaluation algorithm is still increased due to the deviation of the current data and the historical data, and the situations of false alarm and low accuracy rate of the early warning model occur.
Aiming at the problem that in the related art, an effective solution is not provided yet because the current data and the historical data have deviation, the prediction residual error is increased and the false alarm is generated in the equipment state evaluation early warning method based on the nonlinear state evaluation algorithm.
Disclosure of Invention
The embodiment of the application provides an equipment state assessment early warning method and system based on an algorithm fusion technology, and at least solves the problem that in the related technology, due to the fact that the current data and the historical data have deviation, prediction residual errors are increased, and therefore false alarm is generated in the equipment state assessment early warning method based on a nonlinear state assessment algorithm.
In a first aspect, an embodiment of the present application provides an apparatus state assessment early warning method based on an algorithm fusion technology: forming a mode matrix, a compensation matrix and a compensation observation vector based on a plurality of interrelated variables at different moments when the equipment normally operates; determining a similarity coefficient matrix according to a similarity operator, the mode matrix and the compensation matrix, determining a compensation prediction vector according to a nonlinear evaluation algorithm, the mode matrix and the compensation observation vector, and further generating a compensation residual vector; training to obtain a residual prediction model based on the similarity coefficient matrix and the compensation residual vector by using a machine learning algorithm;
the method comprises the following steps:
determining a current observation vector and determining an initial prediction vector corresponding to the current observation vector; determining a current similarity coefficient vector according to the similarity operator, the mode matrix and the current observation vector;
inputting the current similarity coefficient vector to a preset residual prediction model to obtain a current residual prediction vector; and determining a final prediction vector according to the initial prediction vector and the current residual prediction vector so as to determine whether to trigger equipment fault alarm.
In some embodiments, the forming a pattern matrix, a compensation matrix, and a compensated observation vector comprises:
forming an observation vector at any moment by a plurality of variables which are observed at any moment and are mutually associated, and determining a mode observation vector and the compensation observation vector according to the observation vectors at different historical moments under the normal running state of the equipment; and the mode observation vector and the compensation observation vector correspondingly form a mode matrix and a compensation matrix.
In some embodiments, the training, by using a machine learning algorithm, to obtain a residual prediction model based on the similarity coefficient matrix and the compensated residual vector includes:
and inputting each vector in the similar coefficient matrix into a machine learning algorithm, and training a residual prediction vector output by the algorithm to gradually approach the compensation residual vector to obtain the residual prediction model.
In some embodiments, the determining a final prediction vector to determine whether to trigger a device failure alarm comprises:
and determining the final prediction vector, determining a deviation data vector according to the final prediction vector and the current observation vector, and triggering equipment fault alarm related to parameters under the condition that the numerical value of any parameter in the deviation data vector exceeds a corresponding threshold value.
In some embodiments, determining a compensated prediction vector according to a non-linear evaluation algorithm, the mode matrix and the compensated observation vector, and generating a compensated residual vector, the process comprises:
determining a compensation prediction vector corresponding to the compensation observation vector according to a nonlinear evaluation algorithm, a mode matrix and the compensation observation vector; and determining the compensation residual error vector according to the compensation observation vector and the compensation prediction vector.
In some embodiments, determining a current observation vector and determining an initial prediction vector corresponding to the current observation vector comprises:
acquiring data at the current moment and determining a current observation vector; and determining an initial prediction vector corresponding to the current observation vector according to a nonlinear evaluation algorithm, the mode matrix and the current observation vector.
In a second aspect, an embodiment of the present application provides an apparatus state assessment and early warning system based on an algorithm fusion technology, where the system includes:
the device comprises a presetting module, a compensation module and a compensation observation vector, wherein the presetting module is used for forming a mode matrix, a compensation matrix and a compensation observation vector based on a plurality of interrelated variables at different moments when the device normally operates; determining a similarity coefficient matrix according to a similarity operator, the mode matrix and the compensation matrix, determining a compensation prediction vector according to a nonlinear evaluation algorithm, the mode matrix and the compensation observation vector, and further generating a compensation residual vector; training to obtain a residual prediction model based on the similarity coefficient matrix and the compensation residual vector by using a machine learning algorithm;
the initial determination module is used for determining a current observation vector and determining an initial prediction vector corresponding to the current observation vector; determining a current similarity coefficient vector according to a similarity operator, the mode matrix and the current observation vector;
a final determining module, configured to input the current similarity coefficient vector to a preset residual prediction model to obtain a current residual prediction vector; and determining a final prediction vector according to the initial prediction vector and the current residual prediction vector so as to determine whether to trigger equipment fault alarm.
In some embodiments, in the preset module, the forming a pattern matrix, a compensation matrix and a compensation observation vector comprises:
forming an observation vector at any moment by a plurality of variables which are observed at any moment and are mutually associated, and determining a mode observation vector and the compensation observation vector according to the observation vectors at different historical moments under the normal running state of the equipment; and the mode observation vector and the compensation observation vector correspondingly form a mode matrix and a compensation matrix.
In a third aspect, an embodiment of the present application provides an electronic apparatus, which includes a memory and a processor, where the memory stores a computer program, and the processor is configured to execute the computer program to execute the method for device state assessment and warning based on an algorithm fusion technique.
In a fourth aspect, an embodiment of the present application provides a storage medium, where a computer program is stored in the storage medium, where the computer program is configured to execute the method for device state assessment and early warning based on an algorithm fusion technique when running.
Compared with the problem that the prediction residual error is increased and false alarm is generated due to the fact that the current data and the historical data have deviation in the equipment state assessment early warning method based on the nonlinear state assessment algorithm in the related technology, the model residual error based on the nonlinear state assessment algorithm under different similarity coefficients is predicted by introducing the similarity coefficient matrix and utilizing machine learning algorithms such as a neural network, residual error compensation is conducted on actual prediction data based on the prediction value, therefore, the deviation of the prediction value and the actual value is effectively reduced, and the early warning accuracy is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a schematic diagram of a trained residual prediction model according to a first embodiment of the present application;
FIG. 2 is a schematic diagram of training a BP neural network according to a first embodiment of the present application;
fig. 3 is a schematic diagram of an apparatus state assessment and early warning method based on an algorithm fusion technology according to a first embodiment of the present application;
fig. 4 is a schematic diagram of an apparatus state assessment and early warning method based on an algorithm fusion technology according to a second embodiment of the present application;
fig. 5 is a schematic diagram of an apparatus state assessment and early warning method based on an algorithm fusion technology according to a third embodiment of the present application;
fig. 6 is a block diagram of a device state assessment and early warning system based on an algorithm fusion technology according to a fourth embodiment of the present application;
fig. 7 is an internal structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments provided in the present application without any inventive step are within the scope of protection of the present application.
It is obvious that the drawings in the following description are only examples or embodiments of the present application, and that it is also possible for a person skilled in the art to apply the present application to other similar contexts on the basis of these drawings without inventive effort. Moreover, it should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another.
Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the specification. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of ordinary skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments without conflict.
Unless defined otherwise, technical or scientific terms referred to herein shall have the ordinary meaning as understood by those of ordinary skill in the art to which this application belongs. The use of the terms "a" and "an" and "the" and similar referents in the context of describing the invention (including a single reference) are to be construed in a non-limiting sense as indicating either the singular or the plural. The present application is directed to the use of the terms "including," "comprising," "having," and any variations thereof, which are intended to cover non-exclusive inclusions; for example, a process, method, system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to the listed steps or elements, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. Reference to "connected," "coupled," and the like in this application is not intended to be limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The term "plurality" as used herein means two or more. "and/or" describes the association relationship of the associated object, indicating that there may be three relationships, for example, "a and/or B" may indicate: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. Reference herein to the terms "first," "second," "third," and the like, are merely to distinguish similar objects and do not denote a particular ordering for the objects.
The application provides an equipment state evaluation early warning method based on an algorithm fusion technology, before the method is executed, a residual prediction model needs to be obtained through pre-training, fig. 1 is a schematic diagram of the residual prediction model obtained through training according to a first embodiment of the application, and as shown in fig. 1, the process comprises the following steps:
s101, forming an observation vector at any moment by a plurality of interrelated variables observed at any moment, determining a mode observation vector and a compensation observation vector according to the observation vectors at different historical moments under the normal running state of equipment, wherein the mode observation vector and the compensation observation vector correspondingly form a mode matrix and a compensation matrix;
for example, assume that a certain early warning model includes n correlated variables, and at a certain time i, the observed n variables are denoted as i-time observation vectors X (i), that is:
X(i)＝[X 1 (i)，X 2 (i)，…，X n (i)] T
assuming that n associated measuring points are provided, historical observation vectors of m + k normal operation states at different moments are divided into a mode matrix D consisting of the historical observation vectors of the normal operation states at m different moments (marked as moment 1, moment 2, \8230;, moment m) and a compensation matrix E consisting of the historical observation vectors of the normal operation states at k different moments (marked as moment m +1, moment m +2, \8230;, moment m + k), namely:
the current time data vector is denoted as X (obs), that is:
X(obs)＝[X 1 (obs)，X 2 (obs)，......，X n (obs)] T
step S102, determining a similarity coefficient matrix according to a similarity operator, the mode matrix and the compensation matrix, and determining a compensation residual vector according to a nonlinear evaluation algorithm, the mode matrix and the compensation observation vector; inputting each vector in the similarity coefficient matrix into a machine learning algorithm, and training a residual prediction vector output by the algorithm to gradually approach the compensation residual vector to obtain a residual prediction model; wherein the machine learning prediction algorithm includes but is not limited to a BP neural network algorithm;
for example, the similarity coefficient matrix represents the similarity degree between the n-dimensional vector data at each time in the compensation matrix E and the n-dimensional vector data at each time in the history matrix D, which is denoted as S:
in the formula
For the similarity operator, take n-dimensional row vector O and n-dimensional column vector P as examples:
compensating k time data vectors X (m + j) in a matrix E, wherein:
j＝1，2，…，k，
and substituting the values of j into the following formula respectively to obtain the corresponding predicted data vector Y (m + j), namely:
a residual vector corresponding to the data vector X (m + j) in the compensation matrix E is denoted as R (m + j), that is:
R(m+j)＝X(m+j)-Y(m+j)＝[R 1 (m+j)，R 2 (m+j)，…，R n (m+j)] T
the residual matrix corresponding to the k time data vectors in the compensation matrix E is denoted as R, that is:
in some embodiments, fig. 2 is a schematic diagram of a BP neural network training method according to a first embodiment of the present application, and as shown in fig. 2, the BP neural network includes an input layer, a hidden layer, and an output layer, a BP neural network residual prediction model is constructed by taking similar coefficient vectors in a similar coefficient matrix as inputs and residual vectors in a residual matrix as outputs, and in a model training process, the inputs of the input layer of the BP neural network are similar coefficient vectors:
S(h)＝[S 1h ，S 2h ，…，S mh ] T ，h＝1，2，…，k
the output layer is a prediction residual vector:
R′(m+h)＝[R′ 1 (m+h)，R′ 2 (m+h)，…，R′ n (m+h)] T
during training, the output value R' (m + h) of the BP neural network residual prediction model approaches infinitely to satisfy the following conditions:
R′(m+h)＝[R′ 1 (m+h)，R′ 2 (m+h)，…，R′ n (m+h)] T ＝R(m+h)＝[R 1 (m+h)，R 2 (m+h)，…，R n (m+h)] T
namely, the direction of the neural network is adjusted through the error delta = R' (m + h) -R (m + h), and when the error delta approaches 0, the training of the neural network is completed, so that the trained BP neural network residual prediction model is obtained.
On the basis of completing residual prediction model training, fig. 3 is a schematic diagram of a device state assessment and early warning method based on an algorithm fusion technology according to a first embodiment of the present application, and as shown in fig. 3, the process includes the following steps:
step S301, determining a current observation vector and determining an initial prediction vector corresponding to the current observation vector; determining a similarity coefficient vector according to the similarity operator, the mode matrix and the current observation vector;
for example, the device state preliminary prediction data vector based on the pattern matrix D and the current data X (obs) is denoted as X' (est), that is:
the current data vector X (obs) similarity coefficient vector is denoted as S (obs), i.e.:
and (3) inputting S (obs) as a trained BP neural network residual prediction model to obtain a model output residual prediction vector R' (obs), namely:
R′(obs)＝[R′ 1 (obs)，R′ 2 (obs)，…，R′ n (obs)] T
step S302, inputting the similar coefficient vector to a preset residual prediction model to obtain a current residual prediction vector; determining a final prediction vector according to the initial prediction vector and the current residual prediction vector so as to determine whether to trigger equipment failure alarm or not;
for example, based on the device state initial prediction vector X '(est) and the residual prediction vector R' (obs), a final prediction data vector is obtained, denoted as X (est), that is:
X(est)＝X′(est)+R′(obs)＝[X′ 1 (est)+R′ 1 (obs)，X′ 2 (est)+R′ 2 (obs)，…，X′ n (est)+R′ n (obs)] T
determining a deviation data vector according to the current observation vector and the final prediction data vector; under the condition that the value of any parameter in the deviation data vector exceeds a corresponding threshold value, triggering equipment fault alarm related to the parameter, wherein the final deviation data vector is as follows:
X dif ＝X(est)-X(obs)＝[X 1 (dif)，X 2 (dif)，......，X n (dif)] T
let the residual threshold vector of the n-dimensional parameter be ε, i.e.:
ε＝[ε 1 ，ε 2 ，……，ε n ] T
when X is present dif Any one of the parameters | X f (dif)|＞|ε f When l, this parameter triggers an alarm.
Through the steps S301 to S302, compared with the problem that the prediction residual increases and false alarms are generated due to the deviation between the current data and the historical data in the device state assessment early warning method based on the nonlinear state assessment algorithm in the related art, in the embodiment of the present application, the model residual based on the nonlinear state assessment algorithm under different similar coefficients is predicted by introducing the similar coefficient matrix and using the machine learning algorithm such as the neural network, and the residual compensation is performed on the actual prediction data based on the prediction value, so that the problem that the prediction residual increases and false alarms are generated due to the deviation between the current data and the historical data in the early warning model based on the single nonlinear state assessment algorithm is effectively solved.
In some embodiments, fig. 4 is a schematic diagram of a device state assessment and early warning method based on an algorithm fusion technology according to a second embodiment of the present application, and as shown in fig. 4, the process includes the following steps:
step S401, acquiring a mode matrix and a compensation matrix: marking a plurality of variables observed at any moment as an observation vector of the moment, wherein the variables are related to each other; dividing a multi-time historical observation vector into two parts to respectively form a mode matrix and a compensation matrix;
step S402, obtaining a similar coefficient matrix and a residual error matrix: generating a similarity coefficient matrix by using a similarity operator based on the mode matrix and the compensation matrix, generating a prediction data vector based on a nonlinear state evaluation algorithm, and further generating a residual matrix;
step S403, constructing a neural network residual prediction model: taking the similar coefficient vector in the similar coefficient matrix as input, taking the residual vector in the residual matrix as output, carrying out model training, and constructing a BP neural network residual prediction model;
step S404, acquiring a data vector at the current time: determining a current observation data vector, wherein a plurality of variables observed at any moment are marked as the observation vectors at the moment, and the variables are associated with each other;
step S405, generating a device state initial prediction vector: generating an initial prediction data vector according to a nonlinear state evaluation algorithm based on the current data vector and the mode matrix;
step S406, residual prediction: generating a current similar coefficient vector based on a current data vector and a mode matrix, substituting the current similar coefficient vector into an input layer of a BP neural network residual prediction model, and obtaining model output, namely a residual prediction data vector based on a BP neural network algorithm;
step S407, final prediction data generation and state discrimination: and generating final prediction data based on the initial prediction vector of the equipment state and the residual prediction data vector of the BP neural network algorithm, and judging whether to trigger early warning according to a parameter residual threshold value.
In some embodiments, fig. 5 is a schematic diagram of a device state assessment and early warning method based on an algorithm fusion technology according to a third embodiment of the present application, and as shown in fig. 5, the process includes the following steps:
step S501, a mode matrix and a compensation matrix are obtained, a similarity coefficient matrix is generated by utilizing a similarity operator based on the mode matrix and the compensation matrix, a prediction data vector is generated based on a nonlinear state evaluation algorithm, and then a residual error matrix is generated;
step S502, performing model training by taking similar coefficient vectors in a similar coefficient matrix as input and residual vectors in a residual matrix as output, and constructing a BP neural network residual prediction model;
step S503, acquiring a data vector at the current moment, generating an initial prediction data vector according to a nonlinear state evaluation algorithm based on the current data vector and a mode matrix, generating a current similar coefficient vector based on the current data vector and the mode matrix, bringing the current similar coefficient vector into an input layer of a BP neural network residual prediction model, and acquiring a residual prediction data vector;
and step S504, generating final prediction data based on the initial prediction vector of the equipment state and the residual prediction data vector, and judging whether to trigger early warning according to a parameter residual threshold value.
Fig. 6 is a block diagram of a device state assessment and early-warning system based on an algorithm fusion technology according to a fourth embodiment of the present disclosure, and as shown in fig. 6, the system includes a first preset module 601, a second preset module 602, an initial determining module 603, and a final determining module 604, where:
the first preset module 601 is configured to form an observation vector at any time from a plurality of correlated variables observed at any time, determine a mode observation vector and a compensation observation vector according to observation vectors at different historical times in a normal operation state of the device, where the mode observation vector and the compensation observation vector correspondingly form a mode matrix and a compensation matrix;
the second preset module 602 is configured to determine a similarity coefficient matrix according to the similarity operator, the mode matrix, and the compensation matrix, and determine a compensation residual vector according to a nonlinear evaluation algorithm, the mode matrix, and the compensation observation vector; inputting each vector in the similar coefficient matrix into a machine learning algorithm, and training a residual prediction vector output by the algorithm to gradually approach the compensation residual vector to obtain a residual prediction model;
the initial determining module 603 is configured to determine a current observation vector and determine an initial prediction vector corresponding to the current observation vector; determining a similarity coefficient vector according to the similarity operator, the mode matrix and the current observation vector;
the final determining module 604 is configured to input the similar coefficient vector to the residual prediction model to obtain a current residual prediction vector; and determining a final prediction vector according to the initial prediction vector and the current residual prediction vector so as to determine whether to trigger equipment failure alarm.
In some embodiments, in the final determining module 604, the determining a final prediction vector to determine whether to trigger a device failure alarm includes:
determining the final prediction vector, determining a deviation data vector according to the final prediction vector and the current observation vector, and triggering equipment fault alarm related to any parameter in the deviation data vector under the condition that the value of the parameter exceeds a corresponding threshold value.
In some embodiments, in the second default module 602, the determining a compensated residual vector according to the non-linear evaluation algorithm, the mode matrix and the compensated observation vector comprises:
determining a compensation prediction vector corresponding to the compensation observation vector according to a nonlinear evaluation algorithm, a mode matrix and the compensation observation vector; and determining the compensation residual error vector according to the compensation observation vector and the compensation prediction vector.
In some embodiments, in the initial determining module 603, determining a current observation vector and determining an initial prediction vector corresponding to the current observation vector includes:
acquiring data at the current moment and determining a current observation vector; and determining an initial prediction vector corresponding to the current observation vector according to a nonlinear evaluation algorithm, the mode matrix and the current observation vector.
In combination with the method for evaluating and warning the state of the device based on the algorithm fusion technology in the above embodiments, the embodiments of the present application can be implemented by providing a storage medium. The storage medium having stored thereon a computer program; when being executed by a processor, the computer program realizes any one of the above-mentioned embodiments of the method for evaluating and warning the state of the device based on the algorithm fusion technology.
In one embodiment, a computer device is provided, which may be a terminal. The computer device comprises a processor, a memory, a network interface, a display screen and an input device which are connected through a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program to run on the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to realize the equipment state evaluation early warning method based on the algorithm fusion technology. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
In one embodiment, fig. 7 is a schematic diagram of an internal structure of an electronic device according to an embodiment of the present application, and as shown in fig. 7, there is provided an electronic device, which may be a server, and an internal structure diagram of which may be as shown in fig. 7. The electronic device comprises a processor, a network interface, an internal memory and a non-volatile memory connected by an internal bus, wherein the non-volatile memory stores an operating system, a computer program and a database. The processor is used for providing calculation and control capabilities, the network interface is used for communicating with an external terminal through network connection, the internal memory is used for providing an environment for an operating system and the running of a computer program, the computer program is executed by the processor to realize the device state evaluation early warning method based on the algorithm fusion technology, and the database is used for storing data.
It will be appreciated by those skilled in the art that the structure shown in fig. 7 is a block diagram of only a portion of the structure associated with the present application, and does not constitute a limitation on the electronic devices to which the present application may be applied, and that a particular electronic device may include more or fewer components than shown in the drawings, or may combine certain components, or have a different arrangement of components.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It will be understood by those skilled in the art that for simplicity of description, not all possible combinations of the various features of the embodiments described above have been described, but such combinations should be considered within the scope of the present disclosure as long as there is no conflict between such features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, and these are all within the scope of protection of the present application. Therefore, the protection scope of the present patent application shall be subject to the appended claims.
Claims (10)
1. An equipment state assessment early warning method based on algorithm fusion technology is characterized in that:
forming a mode matrix, a compensation matrix and a compensation observation vector based on a plurality of interrelated variables at different moments when the equipment normally operates; determining a similarity coefficient matrix according to a similarity operator, the mode matrix and the compensation matrix, determining a compensation prediction vector according to a nonlinear evaluation algorithm, the mode matrix and the compensation observation vector, and further generating a compensation residual vector; training to obtain a residual prediction model based on the similarity coefficient matrix and the compensation residual vector by using a machine learning algorithm;
the method comprises the following steps:
determining a current observation vector and determining an initial prediction vector corresponding to the current observation vector; determining a current similarity coefficient vector according to a similarity operator, the mode matrix and the current observation vector;
inputting the current similarity coefficient vector to a preset residual prediction model to obtain a current residual prediction vector; and determining a final prediction vector according to the initial prediction vector and the current residual prediction vector so as to determine whether to trigger equipment fault alarm.
2. The method of claim 1, wherein the forming a pattern matrix, a compensation matrix, and a compensated observation vector comprises:
forming an observation vector at any moment by a plurality of interrelated variables observed at any moment, and determining a mode observation vector and the compensation observation vector according to observation vectors at different historical moments under the normal operation state of equipment; and the mode observation vector and the compensation observation vector correspondingly form a mode matrix and a compensation matrix.
3. The method according to claim 1, wherein the training of the residual prediction model based on the similarity coefficient matrix and the compensated residual vector by using a machine learning algorithm comprises:
inputting each vector in the similarity coefficient matrix into a machine learning algorithm, and training a residual prediction vector output by the algorithm to gradually approach the compensation residual vector to obtain the residual prediction model.
4. The method of claim 1, wherein determining the final prediction vector to determine whether to trigger a device failure alarm comprises:
and determining the final prediction vector, determining a deviation data vector according to the final prediction vector and the current observation vector, and triggering equipment fault alarm related to the parameter under the condition that the value of any parameter in the deviation data vector exceeds a corresponding threshold value.
5. The method of claim 1, wherein determining a compensated prediction vector from the non-linear estimation algorithm, the mode matrix, and the compensated observation vector to generate a compensated residual vector comprises:
determining a compensation prediction vector corresponding to the compensation observation vector according to a nonlinear evaluation algorithm, a mode matrix and the compensation observation vector; and determining the compensation residual error vector according to the compensation observation vector and the compensation prediction vector.
6. The method of claim 1, wherein determining a current observation vector and determining an initial prediction vector corresponding to the current observation vector comprises:
acquiring data at the current moment and determining a current observation vector; and determining an initial prediction vector corresponding to the current observation vector according to a nonlinear evaluation algorithm, the mode matrix and the current observation vector.
7. An equipment state assessment early warning system based on algorithm fusion technology is characterized by comprising:
the device comprises a presetting module, a compensation module and a compensation observation vector, wherein the presetting module is used for forming a mode matrix, a compensation matrix and a compensation observation vector based on a plurality of interrelated variables at different moments when the device normally operates; determining a similarity coefficient matrix according to a similarity operator, the mode matrix and the compensation matrix, determining a compensation prediction vector according to a nonlinear evaluation algorithm, the mode matrix and the compensation observation vector, and further generating a compensation residual vector; training to obtain a residual prediction model based on the similarity coefficient matrix and the compensation residual vector by using a machine learning algorithm;
the initial determination module is used for determining a current observation vector and determining an initial prediction vector corresponding to the current observation vector; determining a current similarity coefficient vector according to a similarity operator, the mode matrix and the current observation vector;
a final determining module, configured to input the current similarity coefficient vector to a preset residual prediction model to obtain a current residual prediction vector; and determining a final prediction vector according to the initial prediction vector and the current residual prediction vector so as to determine whether to trigger equipment fault alarm.
8. The system according to claim 7, wherein in the preset module, the composition mode matrix, the compensation matrix and the compensation observation vector comprise:
forming an observation vector at any moment by a plurality of interrelated variables observed at any moment, and determining a mode observation vector and the compensation observation vector according to observation vectors at different historical moments under the normal operation state of equipment; and the mode observation vector and the compensation observation vector correspondingly form a mode matrix and a compensation matrix.
9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to perform a device state assessment and warning method based on an algorithm fusion technology according to any one of claims 1 to 6.
10. A storage medium, wherein a computer program is stored in the storage medium, and the computer program is configured to execute the method for device state assessment and warning based on algorithm fusion technology according to any one of claims 1 to 6 when running.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211313681.8A CN115766383B (en) | 2022-10-25 | 2022-10-25 | Equipment state evaluation early warning method and system based on algorithm fusion technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211313681.8A CN115766383B (en) | 2022-10-25 | 2022-10-25 | Equipment state evaluation early warning method and system based on algorithm fusion technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115766383A true CN115766383A (en) | 2023-03-07 |
| CN115766383B CN115766383B (en) | 2024-02-06 |
Family
ID=85353165
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211313681.8A Active CN115766383B (en) | 2022-10-25 | 2022-10-25 | Equipment state evaluation early warning method and system based on algorithm fusion technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115766383B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117490002A (en) * | 2023-12-28 | 2024-02-02 | 成都同飞科技有限责任公司 | Water supply network flow prediction method and system based on flow monitoring data |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108009063A (en) * | 2017-11-30 | 2018-05-08 | 厦门理工学院 | The method of a kind of electronic equipment fault threshold detection |
-
2022
- 2022-10-25 CN CN202211313681.8A patent/CN115766383B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108009063A (en) * | 2017-11-30 | 2018-05-08 | 厦门理工学院 | The method of a kind of electronic equipment fault threshold detection |
Non-Patent Citations (1)
| Title |
|---|
| 常澍平;郭江龙;吕玉坤;高明;: "非线性状态估计(NSET)建模方法在故障预警系统中的应用", 软件, no. 07 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117490002A (en) * | 2023-12-28 | 2024-02-02 | 成都同飞科技有限责任公司 | Water supply network flow prediction method and system based on flow monitoring data |
| CN117490002B (en) * | 2023-12-28 | 2024-03-08 | 成都同飞科技有限责任公司 | Water supply network flow prediction method and system based on flow monitoring data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115766383B (en) | 2024-02-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112907781B (en) | System fault diagnosis method and device, computer equipment and storage medium | |
| US9489619B2 (en) | Method for the computer-assisted modeling of a technical system | |
| CN111325159A (en) | Fault diagnosis method and device, computer equipment and storage medium | |
| CN110533294B (en) | Nuclear power plant operation fault alarm method based on artificial intelligence technology | |
| CN112100575A (en) | Equipment state trend prediction method and device, computer equipment and storage medium | |
| CN112699946B (en) | Anomaly detection method and device for sensing monitoring data and related equipment | |
| CN113944888B (en) | Gas pipeline leakage detection method, device, equipment and storage medium | |
| CN115766383A (en) | Equipment state evaluation early warning method and system based on algorithm fusion technology | |
| CN111625516A (en) | Method and device for detecting data state, computer equipment and storage medium | |
| CN113760670A (en) | Cable joint abnormity early warning method and device, electronic equipment and storage medium | |
| CN112611584A (en) | Fatigue failure detection method, device, equipment and medium for wind generating set | |
| CN112884199A (en) | Method and device for predicting faults of hydropower station equipment, computer equipment and storage medium | |
| Yang et al. | A study on software reliability prediction based on support vector machines | |
| CN113758652B (en) | Oil leakage detection method and device for converter transformer, computer equipment and storage medium | |
| CN113759724A (en) | Robot control method and device based on data driving and computer equipment | |
| CN113534669A (en) | Unmanned vehicle control method and device based on data driving and computer equipment | |
| JP3307095B2 (en) | Control system diagnosis / analysis apparatus and method | |
| CN115618751B (en) | Steel plate mechanical property prediction method | |
| CN113808727B (en) | Device monitoring method, device, computer device and readable storage medium | |
| CN113377564B (en) | Fault diagnosis method and device, computer equipment and storage medium | |
| CN115499289B (en) | Equipment state evaluation early warning method and system | |
| CN115496384A (en) | Monitoring management method and device for industrial equipment and computer equipment | |
| US20230147470A1 (en) | Failure prediction system | |
| CN114722025A (en) | Data prediction method, device and equipment based on prediction model and storage medium | |
| CN112395774A (en) | Equipment hierarchical health state evaluation method and device based on hardware structure tree |
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 |