CN116258281A - Internet of things fire control monitoring and regulating system based on cloud platform management - Google Patents
Internet of things fire control monitoring and regulating system based on cloud platform management Download PDFInfo
- Publication number
- CN116258281A CN116258281A CN202310530880.2A CN202310530880A CN116258281A CN 116258281 A CN116258281 A CN 116258281A CN 202310530880 A CN202310530880 A CN 202310530880A CN 116258281 A CN116258281 A CN 116258281A
- Authority
- CN
- China
- Prior art keywords
- subsequence
- subsequences
- similar
- data
- sequence
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 29
- 230000001105 regulatory effect Effects 0.000 title claims abstract description 9
- 239000011159 matrix material Substances 0.000 claims abstract description 77
- 230000007704 transition Effects 0.000 claims abstract description 60
- 230000008859 change Effects 0.000 claims abstract description 39
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000000034 method Methods 0.000 claims description 27
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 230000001276 controlling effect Effects 0.000 description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 7
- 238000012423 maintenance Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000005259 measurement Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Mathematical Physics (AREA)
- Development Economics (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Bioinformatics & Computational Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Biology (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Algebra (AREA)
- Entrepreneurship & Innovation (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of electronic data processing, and provides an Internet of things fire monitoring and regulating system based on cloud platform management, which comprises the following components: collecting operation data of terminal equipment of the fire-fighting system; obtaining a plurality of subsequences from the operation data, and clustering to obtain a reference cluster class and a similar subsequence of the current subsequence; acquiring a reference subsequence set and a referenceable degree according to the distance between subsequences in the reference cluster; acquiring a trend change sequence, acquiring a state transition probability matrix of the rightmost data point in each similar subsequence, and acquiring an enhancement matrix of the rightmost data point in each similar subsequence according to the state transition probability matrix, the reference subsequence set and the trend change sequence; and acquiring a state prediction result of the current data point according to the enhancement matrix, and completing monitoring and regulation of the equipment state. The invention aims to improve the accuracy of the equipment state prediction result by combining the historical data trend change relation.
Description
Technical Field
The invention relates to the technical field of electronic data processing, in particular to an Internet of things fire monitoring and regulating system based on cloud platform management.
Background
In an internet of things fire monitoring and controlling system based on cloud platform management, monitoring, early warning and controlling of an abnormal state of equipment are required to be carried out on the operation state of terminal equipment of the fire, and the monitoring of the operation state of the terminal equipment can be carried out by acquiring all historical data of the terminal equipment to train a hidden Markov model (Hidden Markov Model, HMM); each end device in the system maintains its historical operating time series data, with one observation at each point in time, but the state hidden behind those observations is unknown; the use of HMMs allows for the correlation of observation sequences with state sequences for real-time state prediction and identification.
For the operation state early warning of the terminal equipment, a state transition model can be established through the HMM to represent the transition probability between different operation states of the equipment, for example: normal, deviation, abnormality, and other different states requiring maintenance; according to the state probability transition model learned by the HMM, the state of the equipment can be predicted, so that corresponding early warning and maintenance are carried out; however, in the hidden markov model, all information in the historical data is fused to the current state, that is, the state prediction of the next time point of the device is only predicted by the current state and the state transition probability matrix, so that the trend information of the data point is lost in the prediction process; meanwhile, the sum of the probabilities of all states in the state transition probability matrix is 1, so when the transition probability matrix of the next state is obtained for prediction, the predicted deviation exists in the similar state transition probability matrix, and therefore the state transition probability matrix is required to be enhanced through the neighbor relation between the current data point distribution and the historical data points, so that an accurate equipment state prediction result is obtained.
Disclosure of Invention
The invention provides an internet of things fire control monitoring and regulating system based on cloud platform management, which aims to solve the problem that the result is inaccurate due to the fact that equipment state prediction is carried out only through the current state, and adopts the following technical scheme:
the embodiment of the invention provides an Internet of things fire monitoring and controlling system based on cloud platform management, which comprises the following components:
the terminal equipment data acquisition module acquires operation data of terminal equipment of the fire control system and uploads the operation data to the cloud platform data center through the Internet of things;
end device data processing module: acquiring a plurality of subsequences of the operation data through a preset sliding window, taking the subsequence in which the current data point is positioned as the current subsequence, acquiring the distance between any two subsequences according to the data in the subsequences, clustering according to the distance and the state quantity of equipment to obtain a plurality of cluster types, and marking the cluster type in which the current subsequence is positioned as a reference cluster type of the current subsequence and marking other subsequences in the reference cluster type as similar subsequences of the current subsequence;
acquiring a reference subsequence set of each similar subsequence according to the distance between subsequences in the reference cluster, and acquiring the referenceable degree of each reference subsequence set according to the distance between subsequences in the reference subsequence set;
acquiring a trend change sequence of each subsequence in the reference cluster according to the data in each subsequence in the reference cluster, acquiring a state transition probability matrix of the rightmost data point in each similar subsequence, acquiring an intersection ratio of each similar subsequence and each subsequence in the reference subsequence set on the trend change sequence according to the reference subsequence set and the trend change sequence, and acquiring an enhancement matrix of the rightmost data point in each similar subsequence according to the state transition probability matrix, the reference subsequence set and the intersection ratio;
and the equipment state monitoring and regulating module acquires a state prediction result of the current data point according to the enhancement matrix of the right-most data point in each similar subsequence in the reference cluster, and completes monitoring and regulating of the equipment state.
Optionally, the obtaining a plurality of subsequences from the operation data through a preset sliding window includes the following specific methods:
and taking the latest acquired data point as a starting point, acquiring a plurality of sequences of the operation data according to the length of the sliding window and the step length of the sliding window through a preset sliding window, and marking each sequence as a subsequence.
Optionally, the method for obtaining the distance between any two subsequences according to the data in the subsequences includes the following specific steps:
for any two subsequences, taking the data points at the same position in the two subsequences as a point pair, and calculating the distance between the two subsequences according to the data values in all the point pairs, namely, regarding each subsequence as a data point in a high-dimensional space, wherein the Euclidean distance between the data points is the distance between the subsequences;
the distance between any two sub-sequences is obtained.
Optionally, the method for obtaining the reference subsequence set of each similar subsequence according to the distance between the subsequences in the reference cluster includes the following specific steps:
will reference similar subsequences in cluster classes
As subsequences to be determined, the determination in KNN
Values, expressed as
Acquisition in reference cluster class
Each subsequence to be determined
Sub-sequences which are nearest neighbors to each other, if the number of sub-sequences which are nearest neighbors to each other is greater than
Then according to the sub-sequence to be judged
Before the distance of (2) is selected from small to large
A sub-sequence, wherein the obtained set of sub-sequences is used as a sub-sequence to be judged
Is defined by a set of reference subsequences;
if the sub-sequence to be judged is in the reference cluster class
The number of mutually nearest neighbor subsequences is insufficient
When it is to be judged with the subsequence
The set composed of all sub-sequences which are mutually nearest neighbors is used as the sub-sequence to be judged
Is defined by a set of reference subsequences;
a set of reference subsequences for each similar subsequence in the reference cluster class is obtained.
Optionally, the obtaining the referenceable degree of each reference subsequence set includes the following specific methods:
will reference similar subsequences in cluster classes
As the subsequence to be determined, the subsequence to be determined
Referenceable degree of reference subsequence set of (c)
The calculation method of (1) is as follows:
wherein ,
representing sub-sequences to be determined
The average distance from all the subsequences in the reference set of subsequences,
representing sub-sequences to be determined
The number of subsequences in the set of reference subsequences,
representing sub-sequences to be determined
In the reference subsequence set of (2)
Subsequence and subsequence to be determined
Is a distance of (2);
represents the average distance maximum of all the subsequences in the reference cluster class and all the subsequences in the corresponding reference set of subsequences,
representing all subsequences in a reference cluster classThe average distance of all the subsequences in the corresponding reference subsequence set is the minimum.
Optionally, the method for acquiring the trend change sequence of each sub-sequence in the reference cluster according to the data in each sub-sequence in the reference cluster includes the following specific steps:
taking any one subsequence in the reference cluster as an example subsequence, subtracting the value of the previous data point from the value of the next data point in the example subsequence to obtain a result which is larger than 0 and is represented by 1, smaller than 0 and is represented by-1 and equal to 0 and is represented by 0, calculating any two adjacent data points in the example subsequence, sequencing the result according to the sequence of the data points, and marking the obtained sequence as a trend change sequence of the example subsequence;
and obtaining a trend change sequence of each subsequence in the reference cluster.
Optionally, the step of obtaining the cross-over ratio of each similar subsequence to each subsequence in the reference subsequence set on the trend change sequence includes the following specific steps:
will be similar to the subsequence
Any one subsequence in the reference subsequence set of (a) is taken as a target subsequence, and is obtained
Intersection and union of trend change sequences of the trend change sequence with the target subsequence;
the intersection is a set formed by elements with the same numerical value at the same position in the sequence; the union is that two elements with different numerical values at the same position are respectively taken as one element in the union, and two elements with the same numerical value at the same position are taken as one element in the union;
acquiring according to the element number of the intersection and the element number of the union
The cross ratio of the trend sequence of the target subsequence to the trend sequence of the target subsequence; obtaining each similar subsequence and its reference subsequence setThe cross-over ratios of each subsequence over the trend sequence.
Optionally, the method for obtaining the enhancement matrix of the right-most data point in each similar subsequence includes the following specific steps:
wherein ,
representing similar subsequences
From the state of the right most data point
To state
Is used for the degree of enhancement of the (c) in the (c),
representing similar subsequences
To a referenceable degree of a set of reference sub-sequences,
representing similar subsequences
The number of subsequences in the set of reference subsequences,
representing similar subsequences
In the reference subsequence set of (2)
Subsequences and similar subsequences
The ratio of the cross over the trend change sequence,
representing similar subsequences
In the reference subsequence set of (2)
Subsequences and similar subsequences
Is used for the distance of (a),
represents an exponential function with a base of a natural constant,
representing similar subsequences
In the reference subsequence set of (2)
State transition probability matrix of right-most data point in subsequence
To state
State transition probabilities of (2);
obtaining similar subsequences
The enhancement degree of all state transitions of the rightmost data point is normalized by softmax, and the obtained result is recorded as the enhancement factor of each state transition, and the enhancement factors of all state transitions form similar subsequences
An enhancement matrix of the right-most data points; obtaining each similarityEnhancement matrix of right-most data points in the subsequence.
The beneficial effects of the invention are as follows: according to the invention, the traditional state transition probability matrix is enhanced when the state of the next data point is predicted according to the enhancement matrix obtained by the subsequence, and compared with the method of directly predicting through the state transition probability matrix, the method can supplement the historical information of the data at the current moment, so that the defect of judging continuous information and trend information change of the data point due to the hidden Markov model is avoided; meanwhile, the similar subsequence of the current subsequence is judged through clustering, the reference subsequence set is obtained through the distance difference between subsequences in the reference cluster, and the enhancement matrix is obtained through the reference subsequence set.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a block diagram of an internet of things fire monitoring and controlling system based on cloud platform management according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a structural block diagram of an internet of things fire monitoring and controlling system based on cloud platform management according to an embodiment of the present invention is shown, where the system includes:
the terminal equipment data acquisition module 101 acquires operation data of terminal equipment of the fire protection system and uploads the operation data to the cloud platform data center through the Internet of things.
The purpose of the embodiment is to monitor and predict the equipment state by combining the real-time operation data of the terminal equipment with the historical operation data and correspondingly regulate and control according to the equipment state prediction result, so that the operation data of the terminal equipment in the fire-fighting system is required to be collected first; the operation data of the terminal equipment comprises the operation data such as outlet pressure, voltage, current, pump motor temperature and the like of the terminal water pump, wherein the terminal water pump comprises an outdoor hydrant pump, an indoor spray pump and a fire-fighting pressure stabilizing pump; according to the embodiment, the operation data of the tail end water pump is acquired through the intelligent acquisition cabinet in the fire pump room, the intelligent acquisition cabinet is an acquisition device for monitoring information such as the switching state, the current, the voltage and the fault state of all equipment in the pump room, and various sensors are used for acquiring various data; in this embodiment, the pressure data of the end water pump is described as an example, and other types of operation data are the same as the pressure data in the subsequent processing and analysis process, so that the pressure data is used as the operation data of the end device to perform the subsequent processing and analysis; it should be noted that, the operation data includes real-time operation data and all historical operation data of the terminal device, the operation data is time sequence data, the real-time operation data is the latest collected data point, all previous data points are historical operation data, and the collection frequency of the operation data is set by the intelligent collection cabinet, which is not described in detail in this embodiment.
Further, the collected operation data are uploaded to the cloud platform data center through the Internet of things, and the cloud platform management system directly processes and analyzes the operation data in the cloud platform data center, so that monitoring and prediction of the operation state of the equipment are completed.
So far, the operation data of the terminal equipment are obtained and uploaded to the cloud platform data center for subsequent processing analysis.
End device data processing module 102:
it should be noted that, in order to enable the state of the device at the next time to be predicted by the hidden markov model, trend information of the data may be considered, and state transition probabilities of a plurality of similar values in the state transition probability matrix may be avoided; firstly, acquiring a reference subsequence of a subsequence where a current data point is located through clustering, and optimizing current data prediction through similar data in historical data; and then, obtaining an enhancement matrix of the state transition probability matrix of the current data point through the continuous state information of the reference subsequences, wherein the enhancement matrix can be used for predicting and enhancing the state transition probability matrix of the current data point obtained by the hidden Markov model.
It should be further noted that, in the process of obtaining the enhancement matrix, since there is a measurement deviation for the trend information in the distance measurement of the clusters, the enhancement matrix needs to be corrected by the deviation of the reference subsequences in the clusters, so that the state transition probability matrix is accurately optimized by the enhancement matrix.
(1) And acquiring a plurality of subsequences of the operation data through a preset sliding window, taking the subsequence in which the current data point is positioned as the current subsequence, clustering all the subsequences, and acquiring a reference cluster class and a similar subsequence of the current subsequence according to a clustering result.
It should be noted that, because the operation data is time sequence data and there is trend change between data points, a sliding window with a fixed length is adopted to intercept the subsequence from the data point which is acquired last, i.e. the current data point, and meanwhile, in order to ensure the integrity of trend information, part of trend information is not lost because of sectionally acquiring the subsequence, so that the step length of the sliding window is smaller than the length of the sliding window, and a plurality of subsequences are obtained; the difference between the same position elements among the subsequences is used as distance measurement among the subsequences, and then the subsequences are clustered according to the distance measurement, so that the situation that data fluctuation is similar among the subsequences in the same cluster exists, and similar trend information can be provided.
Specifically, for operation data, namely pressure time sequence data of the water pump, the latest collected data point is taken as a starting point, the latest collected data point is a real-time current data point, a plurality of sections of sequences are obtained for the operation data according to the length of a sliding window and the step length of the sliding window through a fixed-length preset sliding window, each section of sequence is recorded as a subsequence, the length of the sliding window in the embodiment is 10, and the step length of the sliding window is described by 1; for any two subsequences, taking data points at the same position in the two subsequences as a point pair, and calculating the distance between the two subsequences according to the data values in all the point pairs, namely, regarding each subsequence as a data point in a high-dimensional space, wherein the Euclidean distance between the data points is the distance between the subsequences, and the number of dimensions of the high-dimensional space is equal to the number of elements of the subsequences; obtaining the distance between any two subsequences, carrying out K-means clustering on all the subsequences according to the distance between the subsequences, determining the cluster number, namely the state number predicted by a hidden Markov model through the K value, for example, dividing the equipment state into 5 types, and determining the cluster number K=5; in this embodiment, the states of the end water pump are divided into 4 states including normal, deviation, abnormality and maintenance, and in this embodiment, k=4 is used for description; all sub-sequences are divided into 4 cluster classes.
Further, the subsequence where the current data point is located is marked as a current subsequence, the cluster where the current subsequence is located is marked as a reference cluster of the current subsequence, and other subsequences in the reference cluster are marked as similar subsequences of the current subsequence.
So far, the reference cluster class of the current subsequence and the similar subsequence are obtained, and the enhancement matrix is obtained through the subsequent acquisition of the similar subsequence, so that the state transition probability matrix is optimized.
(2) And acquiring a reference subsequence set of each similar subsequence according to the distance between the subsequences in the reference cluster, and acquiring the referenceable degree of each reference subsequence set.
It should be noted that, when the continuous trend changes in the prediction process, the influence corresponding to the continuous trend needs to be reflected in the target state of the trend in the reference cluster in the enhancement matrix, that is, the state transition probability matrix in the hidden markov model is optimized by referring to the trend of the data points in the subsequence, so that the prediction of the hidden markov model can include the continuous trend information of the data points.
It should be further noted that, for the case that the distance between the subsequences is larger in the reference cluster class, the information reference of the current subsequence cannot be performed through all the similar subsequences, a reference subsequence set needs to be further determined in the cluster class, and for any subsequence in the reference cluster class, the acquisition of the reference subsequence set can be performed through KNN in the reference cluster class, that is, the reference is performed through K nearest neighbor subsequences of the subsequence in the cluster class; but because the nearest neighbor subsequence exists in the subsequence
Is a subsequence
Is the nearest neighbor subsequence of (2) in normal case
Can be used as
Is referred to in (2); however if
Not be
Nearest neighbor subsequence of (2), then
Cannot be used as
Is referred to in (2); then, for the current sub-sequence and each similar sub-sequence in the reference cluster class, the sub-sequences that are nearest neighbors to each other need to be found, so as to obtain a reference sub-sequence set of each sub-sequence.
Specifically, first, similar subsequences in a reference cluster class are referred to
As subsequences to be determined, a KNN is determined
Value of KNN to facilitate discrimination from K employed for clustering
Value this example is expressed as
Wherein KNN is K nearest neighbor classification algorithm, which is known in the art, and is not described in detail in this embodiment, which adopts
Description is made; acquisition in reference cluster class
Each subsequence to be determined
Sub-sequences which are nearest neighbors to each other, if the number of sub-sequences which are nearest neighbors to each other is greater than
Then according to the sub-sequence to be judged
Before the distance of (2) is selected from small to large
A sub-sequence, wherein the obtained set of sub-sequences is used as a sub-sequence to be judged
Is denoted as reference subsequence set of (2)
The method comprises the steps of carrying out a first treatment on the surface of the If the sub-sequence to be judged is in the reference cluster class
The number of mutually nearest neighbor subsequences is insufficient
When it is, it will be associated with the subsequence to be determined
The set composed of all sub-sequences which are mutually nearest neighbors is used as the sub-sequence to be judged
Is also denoted as reference subsequence set of (2)
The method comprises the steps of carrying out a first treatment on the surface of the And acquiring a reference subsequence set of each similar subsequence in the reference cluster according to the method.
It is further noted that in the reference cluster class, a subsequence
Sub-sequences having edges in sub-sequences nearest to each other
All the sub-sequences which are nearest neighbors are edge discrete sub-sequences, so that the reference value of the sub-sequences to the current sub-sequence is lower; so for a sub-sequence to be determined
The evaluation of the referenceable degree of the reference subsequence set is required according to the distribution condition of the subsequences in the reference subsequence set, so that reference is provided when the current subsequence calculates an enhancement matrix of the state transition probability matrix according to the reference subsequence set; and further, the enhancement effect of the enhancement matrix on the state transition probability matrix is most reasonable.
Specifically, the subsequence to be determined
For exampleTo-be-judged subsequence
Reference subsequence set of (a) is
To-be-judged subsequence
Referenceable degree of reference subsequence set of (c)
The calculation method of (1) is as follows:
wherein ,
representing sub-sequences to be determined
The average distance from all the subsequences in the reference set of subsequences,
representing sub-sequences to be determined
The number of subsequences in the set of reference subsequences,
representing sub-sequences to be determined
In the reference subsequence set of (2)
Subsequence and subsequence to be determined
Is a distance of (2);
represents the average distance maximum of all the subsequences in the reference cluster class and all the subsequences in the corresponding reference set of subsequences,
representing the average minimum distance between all subsequences in the reference cluster and all subsequences in the corresponding reference subsequence set; the larger the average distance corresponding to the subsequences to be judged is, the larger the mutual referenceability is, the smaller the average distance between the subsequences is, so that the larger the average distance is, the smaller the referenceable degree of the reference subsequence set of the subsequences to be judged is; and acquiring the referenceable degree of each reference subsequence set in the reference cluster according to the method.
The referenceable degree of each reference subsequence set in the reference cluster is obtained, and a reference basis is provided for the enhancement matrix acquisition of the subsequent state transition probability matrix.
(3) According to the data in each subsequence in the reference cluster, a trend change sequence of each subsequence in the reference cluster is obtained, a state transition probability matrix of the right-most data point in each similar subsequence is obtained, and according to the state transition probability matrix, the reference subsequence set and the trend change sequence, an enhancement matrix of the right-most data point in each similar subsequence is obtained.
It should be noted that, the rightmost data point in the current subsequence is the data point collected most recently, that is, the current data point, and each similar subsequence is to predict the state of the next data point of the rightmost data point, so that the state transition probability matrix of the rightmost data point in each similar subsequence can be obtained; the state transition probability matrix is the state change possibility of the next data point obtained through trend information, so that the state change probability on the trend needs to be obtained through the trend difference among the subsequences in the reference subsequence set, and the enhancement factor is obtained by combining the referenceable degree, and the probability of the state change in the state transition probability matrix is optimized through the enhancement factor.
Specifically, for any one subsequence in the reference cluster, subtracting the value of the previous data point from the value of the next data point to obtain a result which is larger than 0, wherein the result is represented by 1, the result which is smaller than 0 is represented by-1, the result which is equal to 0 is represented by 0, the calculation is performed on any two adjacent data points in the subsequence, the result is ordered according to the sequence of the data points, the obtained sequence is recorded as a trend change sequence of the subsequence, and the element number of the trend change sequence is one less than that of the subsequence; and acquiring a trend change sequence of each subsequence in the reference cluster according to the method.
Further, a state transition probability matrix of the right-most data point in each similar subsequence is obtained, and the state transition probability matrix is obtained as the prior art in the hidden markov model, which is not described in detail in this embodiment; in similar subsequences
For example, for similar subsequences
Any one of the sub-sequences in the reference sub-sequence set of (a) is obtained
Intersection and union of the trend change sequence of the subsequence and the trend change sequence, wherein the intersection is a set formed by elements with the same numerical value at the same position in the sequence, the union is respectively taken as one element in the union by two elements with different numerical values at the same position, the two elements with the same numerical value at the same position are taken as one element in the union, and the acquisition is carried out according to the number of the elements of the intersection and the number of the elements of the union
The cross ratio of the trend sequence of (2) to the trend sequence of the subsequence; and obtaining the cross ratio of each similar subsequence and each subsequence in the reference subsequence set thereof on the trend change sequence according to the method.
Further, according to the cross ratio of the trend change sequences, the reference subsequence set and the referenceable degree thereof, and the state transition probability matrix, the enhancement matrix of the right-most data point in each similar subsequence is obtained to obtain the similar subsequence
For example, it enhances states in a matrix
To state
The calculation method of the enhancement factor of (2) is as follows:
wherein ,
representing similar subsequences
From the state of the right most data point
To state
Is used for the degree of enhancement of the (c) in the (c),
representing similar subsequences
To a referenceable degree of a set of reference sub-sequences,
representing similar subsequences
The number of subsequences in the set of reference subsequences,
representing similar subsequences
In the reference subsequence set of (2)
Subsequences and similar subsequences
The ratio of the cross over the trend change sequence,
representing similar subsequences
In the reference subsequence set of (2)
Subsequences and similar subsequences
Is used for the distance of (a),
represents an exponential function with a base of a natural constant,
representing similar subsequences
In the reference subsequence set of (2)
State transition probability matrix of right-most data point in subsequence
To state
State transition probabilities of (2); obtaining similar subsequences according to the method described above
The enhancement degree of all state transitions of the rightmost data point is normalized by softmax, the obtained result is recorded as the enhancement factor of each state transition, and the enhancement factors of all state transitions form similar subsequences
An enhancement matrix of the right-most data points; since each subsequence in the reference set of subsequences is identical to a similar subsequence
The smaller the distance, the greater the reference, and the corresponding increase in the influence of the subsequence; meanwhile, the larger the cross ratio is, the closer trend information is shown, the influence of the subsequence is increased, the weighted average effect is achieved on the probability of the same state transition in the reference subsequence set through the cross ratio and the distance, and the state transition probability is optimized by combining the referenceable degree of the reference subsequence set; through softmax normalization, the sum of enhancement factors of all state transitions in the enhancement matrix is ensured to be still 1 so as to be used for the state prediction of the subsequent current data point; the enhancement matrix for the right-most data point in each similar subsequence was obtained as described above.
Therefore, the enhancement matrix of the right-most data point in each similar subsequence is obtained, and compared with the state transition probability matrix, the enhancement matrix considers the state transition in the subsequence with similar trend changes in the reference subsequence set more, so that the enhancement factors in the enhancement matrix are more accurate compared with the state transition probability.
The device state monitoring and controlling module 103 obtains the state prediction result of the current data point according to the enhancement matrix of the rightmost data point in each similar subsequence in the reference cluster, and completes the monitoring and controlling of the device state.
After the enhancement matrix of the right-most data point in each similar subsequence in the reference cluster is obtained, performing state marking on the historical data points: 0 is normal, 1 is deviation exists, 2 is abnormal, and 3 is four states to be maintained; a plurality of data points in the operation data are provided with enhancement matrixes, an initial transition probability matrix is obtained according to the enhancement matrix with the first 30% of data points of the enhancement matrix, a hidden Markov model is trained according to the enhancement matrix with the remaining 70% of data points, and the initial transition probability matrix and a data point proportion implementer trained by the model can be set according to actual conditions; in the training process, the state transition probability matrix is in the traditional hidden markov model, and the state transition probability matrix is replaced by the enhancement matrix in the embodiment; taking the maximum state transition probability as the prediction of the next state after training is completed, and acquiring a corresponding state label so as to complete the state prediction of the current data point; when the state prediction result is abnormal, early warning is carried out in the system; and when the state prediction result is that maintenance is needed, a maintenance work order is dispatched through the cloud platform management system, and maintenance of the tail end water pump is performed.
Thus, the state monitoring, prediction and regulation of the fire control system terminal equipment based on cloud platform management are completed.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (8)
1. Internet of things fire monitoring and control system based on cloud platform management, which is characterized in that the system comprises:
the terminal equipment data acquisition module acquires operation data of terminal equipment of the fire control system and uploads the operation data to the cloud platform data center through the Internet of things;
end device data processing module: acquiring a plurality of subsequences of the operation data through a preset sliding window, taking the subsequence in which the current data point is positioned as the current subsequence, acquiring the distance between any two subsequences according to the data in the subsequences, clustering according to the distance and the state quantity of equipment to obtain a plurality of cluster types, and marking the cluster type in which the current subsequence is positioned as a reference cluster type of the current subsequence and marking other subsequences in the reference cluster type as similar subsequences of the current subsequence;
acquiring a reference subsequence set of each similar subsequence according to the distance between subsequences in the reference cluster, and acquiring the referenceable degree of each reference subsequence set according to the distance between subsequences in the reference subsequence set;
acquiring a trend change sequence of each subsequence in the reference cluster according to the data in each subsequence in the reference cluster, acquiring a state transition probability matrix of the rightmost data point in each similar subsequence, acquiring an intersection ratio of each similar subsequence and each subsequence in the reference subsequence set on the trend change sequence according to the reference subsequence set and the trend change sequence, and acquiring an enhancement matrix of the rightmost data point in each similar subsequence according to the state transition probability matrix, the reference subsequence set and the intersection ratio;
and the equipment state monitoring and regulating module acquires a state prediction result of the current data point according to the enhancement matrix of the right-most data point in each similar subsequence in the reference cluster, and completes monitoring and regulating of the equipment state.
2. The cloud platform management-based internet of things fire monitoring and control system according to claim 1, wherein the acquiring the plurality of subsequences of the operation data through the preset sliding window comprises the following specific methods:
and taking the latest acquired data point as a starting point, acquiring a plurality of sequences of the operation data according to the length of the sliding window and the step length of the sliding window through a preset sliding window, and marking each sequence as a subsequence.
3. The cloud platform management-based internet of things fire monitoring and control system according to claim 1, wherein the method for acquiring the distance between any two subsequences according to the data in the subsequences comprises the following specific steps:
for any two subsequences, taking the data points at the same position in the two subsequences as a point pair, and calculating the distance between the two subsequences according to the data values in all the point pairs, namely, regarding each subsequence as a data point in a high-dimensional space, wherein the Euclidean distance between the data points is the distance between the subsequences;
the distance between any two sub-sequences is obtained.
4. The cloud platform management-based internet of things fire monitoring and control system according to claim 1, wherein the specific method for acquiring the reference subsequence set of each similar subsequence according to the distance between the subsequences in the reference cluster comprises the following steps:
will reference similar subsequences in cluster classes
As subsequences to be determined, the +.in KNN is determined>
Value, expressed as->
Acquiring +.>
The individual and the subsequence to be determined->
Sub-sequences which are nearest neighbors to each other, if the number of sub-sequences which are nearest neighbors to each other is greater than +.>
Then according to the sub-sequence to be judged +.>
Distance of (2) from small to large before +.>
A subsequence, wherein the obtained set of subsequences is used as the subsequence to be judged +.>
Is defined by a set of reference subsequences;
if the sub-sequence to be judged is in the reference cluster class
The number of subsequences which are nearest neighbors to each other is less than +.>
In the mean time, will be associated with the subsequence to be determined->
The set of all subsequences that are nearest neighbors to each other is used as the subsequence to be determined +.>
Is defined by a set of reference subsequences;
a set of reference subsequences for each similar subsequence in the reference cluster class is obtained.
5. The cloud platform management-based fire control monitoring and control system of the internet of things according to claim 1, wherein the obtaining the referenceable degree of each reference subsequence set comprises the following specific methods:
will reference similar subsequences in cluster classes
As a subsequence to be determined, the subsequence to be determined +.>
Is +.about.about.>
The calculation method of (1) is as follows:
wherein ,
representing the subsequence to be determined->
Average distance from all subsequences in the reference set of subsequences,/->
Representing the subsequence to be determined->
Number of subsequences in the reference subsequence set, +.>
Representing the subsequence to be determined->
Is +.>
Subsequence and subsequence to be determined->
Is a distance of (2); />
Representing the average distance maximum value of all sub-sequences in the reference cluster class and all sub-sequences in the corresponding reference sub-sequence set,/I>
Representing the average minimum distance between all sub-sequences in the reference cluster class and all sub-sequences in the corresponding reference sub-sequence set.
6. The cloud platform management-based internet of things fire monitoring and control system according to claim 1, wherein the acquiring the trend change sequence of each sub-sequence in the reference cluster according to the data in each sub-sequence in the reference cluster comprises the following specific steps:
taking any one subsequence in the reference cluster as an example subsequence, subtracting the value of the previous data point from the value of the next data point in the example subsequence to obtain a result which is larger than 0 and is represented by 1, smaller than 0 and is represented by-1 and equal to 0 and is represented by 0, calculating any two adjacent data points in the example subsequence, sequencing the result according to the sequence of the data points, and marking the obtained sequence as a trend change sequence of the example subsequence;
and obtaining a trend change sequence of each subsequence in the reference cluster.
7. The cloud platform management-based fire control monitoring and control system of the internet of things according to claim 1, wherein the method for obtaining the cross-over ratio of each similar subsequence to each subsequence in the reference subsequence set thereof on the trend change sequence comprises the following specific steps:
will be similar to the subsequence
Any one of the sub-sequences in the reference sub-sequence set of (a) is taken as the target sub-sequence to obtain +.>
Intersection and union of trend change sequences of the trend change sequence with the target subsequence;
the intersection is a set formed by elements with the same numerical value at the same position in the sequence; the union is that two elements with different numerical values at the same position are respectively taken as one element in the union, and two elements with the same numerical value at the same position are taken as one element in the union;
acquiring according to the element number of the intersection and the element number of the union
The cross ratio of the trend sequence of the target subsequence to the trend sequence of the target subsequence; and obtaining the cross ratio of each similar subsequence to each subsequence in the reference subsequence set on the trend change sequence.
8. The cloud platform management-based internet of things fire monitoring and control system according to claim 1, wherein the method for obtaining the enhancement matrix of the right-most data point in each similar subsequence comprises the following specific steps:
wherein ,
representing a similar subsequence->
The right-most data point is from state->
To state->
Degree of enhancement of->
Representing a similar subsequence->
Reference degree of the reference subsequence set, < ->
Representing a similar subsequence->
Number of subsequences in the reference subsequence set, +.>
Representing a similar subsequence->
Is +.>
Subsequence and similar subsequence->
Cross ratio on trend change sequence, +.>
Representing a similar subsequence->
Is +.>
Subsequence and similar subsequence->
Distance of->
Represents an exponential function based on natural constants, < ->
Representing a similar subsequence->
Is +.>
From the state transition probability matrix of the right-most data point in the subsequenceStatus->
To state->
State transition probabilities of (2);
obtaining similar subsequences
The enhancement degree of all state transitions of the rightmost data point is normalized by softmax, and the obtained result is recorded as the enhancement factor of each state transition, and the enhancement factors of all state transitions form similar subsequence->
An enhancement matrix of the right-most data points; an enhancement matrix is obtained for the right-most data point in each similar subsequence. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310530880.2A CN116258281B (en) | 2023-05-12 | 2023-05-12 | Internet of things fire control monitoring and regulating system based on cloud platform management |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310530880.2A CN116258281B (en) | 2023-05-12 | 2023-05-12 | Internet of things fire control monitoring and regulating system based on cloud platform management |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116258281A true CN116258281A (en) | 2023-06-13 |
| CN116258281B CN116258281B (en) | 2023-07-25 |
Family
ID=86688304
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310530880.2A Active CN116258281B (en) | 2023-05-12 | 2023-05-12 | Internet of things fire control monitoring and regulating system based on cloud platform management |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116258281B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116510223A (en) * | 2023-06-29 | 2023-08-01 | 欣灵电气股份有限公司 | Self-identification fire pump electrical parameter inspection monitoring system based on Internet of things |
| CN116757337A (en) * | 2023-08-18 | 2023-09-15 | 克拉玛依市鼎泰建设(集团)有限公司 | House construction progress prediction system based on artificial intelligence |
| CN117235462A (en) * | 2023-11-09 | 2023-12-15 | 海门市知行包装制品有限公司 | Intelligent fault prediction method for bag type packaging machine based on time sequence data analysis |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103729528A (en) * | 2012-10-15 | 2014-04-16 | 富士通株式会社 | Device and method for processing sequence |
| CN104850916A (en) * | 2015-05-31 | 2015-08-19 | 上海电机学院 | Improved-gray-Markov-model-based power equipment fault prediction method |
| JP2015192502A (en) * | 2014-03-27 | 2015-11-02 | 富士通株式会社 | Power consumption prediction method, power consumption prediction program and power consumption prediction device |
| CN105574669A (en) * | 2015-12-16 | 2016-05-11 | 国网山东省电力公司电力科学研究院 | Space-time union data clustering analysis based abnormal status detection method for power transmission and transformation device |
| JP2017068782A (en) * | 2015-10-02 | 2017-04-06 | 株式会社日立製作所 | Time series data processing device and time series data processing method |
| WO2017212880A1 (en) * | 2016-06-09 | 2017-12-14 | 株式会社日立製作所 | Data prediction system and data prediction method |
| CN108763654A (en) * | 2018-05-03 | 2018-11-06 | 国网江西省电力有限公司信息通信分公司 | A kind of electrical equipment fault prediction technique based on Weibull distribution and hidden Semi-Markov Process |
| CN109376892A (en) * | 2018-06-12 | 2019-02-22 | 电子科技大学 | A kind of equipment state prediction method based on life cycle phase locating for equipment |
| CN110362558A (en) * | 2019-06-12 | 2019-10-22 | 广东工业大学 | A kind of energy consumption data cleaning method based on neighborhood propagation clustering |
| CN110955226A (en) * | 2019-11-22 | 2020-04-03 | 深圳市通用互联科技有限责任公司 | Equipment failure prediction method and device, computer equipment and storage medium |
| US20200258157A1 (en) * | 2019-02-11 | 2020-08-13 | Td Ameritrade Ip Company, Inc. | Time-Series Pattern Matching System |
| CN114065605A (en) * | 2021-09-30 | 2022-02-18 | 云南电网有限责任公司曲靖供电局 | Intelligent electric energy meter running state detection and evaluation system and method |
| CN115186762A (en) * | 2022-07-29 | 2022-10-14 | 江西科骏实业有限公司 | Engine abnormity detection method and system based on DTW-KNN algorithm |
| CN115423158A (en) * | 2022-08-17 | 2022-12-02 | 贵州北盘江电力股份有限公司光照分公司 | Predictive analysis method and system for data trend of hydroelectric generating set |
-
2023
- 2023-05-12 CN CN202310530880.2A patent/CN116258281B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103729528A (en) * | 2012-10-15 | 2014-04-16 | 富士通株式会社 | Device and method for processing sequence |
| JP2015192502A (en) * | 2014-03-27 | 2015-11-02 | 富士通株式会社 | Power consumption prediction method, power consumption prediction program and power consumption prediction device |
| CN104850916A (en) * | 2015-05-31 | 2015-08-19 | 上海电机学院 | Improved-gray-Markov-model-based power equipment fault prediction method |
| JP2017068782A (en) * | 2015-10-02 | 2017-04-06 | 株式会社日立製作所 | Time series data processing device and time series data processing method |
| CN105574669A (en) * | 2015-12-16 | 2016-05-11 | 国网山东省电力公司电力科学研究院 | Space-time union data clustering analysis based abnormal status detection method for power transmission and transformation device |
| WO2017212880A1 (en) * | 2016-06-09 | 2017-12-14 | 株式会社日立製作所 | Data prediction system and data prediction method |
| CN108763654A (en) * | 2018-05-03 | 2018-11-06 | 国网江西省电力有限公司信息通信分公司 | A kind of electrical equipment fault prediction technique based on Weibull distribution and hidden Semi-Markov Process |
| CN109376892A (en) * | 2018-06-12 | 2019-02-22 | 电子科技大学 | A kind of equipment state prediction method based on life cycle phase locating for equipment |
| US20200258157A1 (en) * | 2019-02-11 | 2020-08-13 | Td Ameritrade Ip Company, Inc. | Time-Series Pattern Matching System |
| CN110362558A (en) * | 2019-06-12 | 2019-10-22 | 广东工业大学 | A kind of energy consumption data cleaning method based on neighborhood propagation clustering |
| CN110955226A (en) * | 2019-11-22 | 2020-04-03 | 深圳市通用互联科技有限责任公司 | Equipment failure prediction method and device, computer equipment and storage medium |
| CN114065605A (en) * | 2021-09-30 | 2022-02-18 | 云南电网有限责任公司曲靖供电局 | Intelligent electric energy meter running state detection and evaluation system and method |
| CN115186762A (en) * | 2022-07-29 | 2022-10-14 | 江西科骏实业有限公司 | Engine abnormity detection method and system based on DTW-KNN algorithm |
| CN115423158A (en) * | 2022-08-17 | 2022-12-02 | 贵州北盘江电力股份有限公司光照分公司 | Predictive analysis method and system for data trend of hydroelectric generating set |
Non-Patent Citations (1)
| Title |
|---|
| 冯凯文;孟凡荣;牛强;闫秋艳;: "基于趋势点状态模型的时间序列预测算法", 计算机应用研究, no. 12, pages 510 - 4516 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116510223A (en) * | 2023-06-29 | 2023-08-01 | 欣灵电气股份有限公司 | Self-identification fire pump electrical parameter inspection monitoring system based on Internet of things |
| CN116510223B (en) * | 2023-06-29 | 2023-09-01 | 欣灵电气股份有限公司 | Self-identification fire pump electrical parameter inspection monitoring system based on Internet of things |
| CN116757337A (en) * | 2023-08-18 | 2023-09-15 | 克拉玛依市鼎泰建设(集团)有限公司 | House construction progress prediction system based on artificial intelligence |
| CN116757337B (en) * | 2023-08-18 | 2023-11-21 | 克拉玛依市鼎泰建设(集团)有限公司 | House construction progress prediction system based on artificial intelligence |
| CN117235462A (en) * | 2023-11-09 | 2023-12-15 | 海门市知行包装制品有限公司 | Intelligent fault prediction method for bag type packaging machine based on time sequence data analysis |
| CN117235462B (en) * | 2023-11-09 | 2024-02-13 | 海门市知行包装制品有限公司 | Intelligent fault prediction method for bag type packaging machine based on time sequence data analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116258281B (en) | 2023-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116258281B (en) | Internet of things fire control monitoring and regulating system based on cloud platform management | |
| CN107092582B (en) | Online abnormal value detection and confidence evaluation method based on residual posterior | |
| CN109814527B (en) | Industrial equipment fault prediction method and device based on LSTM recurrent neural network | |
| CN112527788A (en) | Method and device for detecting and cleaning abnormal value of transformer monitoring data | |
| CN108518804A (en) | A kind of machine room humiture environmental forecasting method and system | |
| CN108664009B (en) | Stage division and fault detection method based on correlation analysis | |
| CN106896219B (en) | The identification of transformer sub-health state and average remaining lifetime estimation method based on Gases Dissolved in Transformer Oil data | |
| CN113225209B (en) | Network monitoring real-time early warning method based on time series similarity retrieval | |
| CN116992322B (en) | Smart city data center management system | |
| Wang et al. | Fault detection and diagnosis for multiple faults of VAV terminals using self-adaptive model and layered random forest | |
| CN117113729B (en) | Digital twinning-based power equipment online state monitoring system | |
| CN115423158A (en) | Predictive analysis method and system for data trend of hydroelectric generating set | |
| CN111611961A (en) | Harmonic anomaly identification method based on variable point segmentation and sequence clustering | |
| CN114912640A (en) | Method and system for detecting abnormal mode of generator set based on deep learning | |
| CN117312997A (en) | Intelligent diagnosis method and system for power management system | |
| CN110084301B (en) | Hidden Markov model-based multi-working-condition process working condition identification method | |
| CN113640675B (en) | Aviation lithium battery abnormity detection method based on Snippets characteristic extraction | |
| CN116992246B (en) | Intelligent sensing method and system for underground airflow parameters | |
| CN110689140A (en) | Method for intelligently managing rail transit alarm data through big data | |
| US20160275407A1 (en) | Diagnostic device, estimation method, non-transitory computer readable medium, and diagnostic system | |
| CN107808209B (en) | Wind power plant abnormal data identification method based on weighted kNN distance | |
| CN116361628A (en) | Fault category intelligent analysis method and device based on VFD room | |
| CN112561153A (en) | Scenic spot crowd gathering prediction method based on model integration | |
| CN116757337B (en) | House construction progress prediction system based on artificial intelligence | |
| CN117150244B (en) | Intelligent power distribution cabinet state monitoring method and system based on electrical parameter analysis |
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 |