CN110555477A - municipal facility fault prediction method and device - Google Patents
municipal facility fault prediction method and device Download PDFInfo
- Publication number
- CN110555477A CN110555477A CN201910819011.5A CN201910819011A CN110555477A CN 110555477 A CN110555477 A CN 110555477A CN 201910819011 A CN201910819011 A CN 201910819011A CN 110555477 A CN110555477 A CN 110555477A
- Authority
- CN
- China
- Prior art keywords
- state classification
- municipal
- classification features
- data
- fault prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims description 33
- 238000012549 training Methods 0.000 claims description 32
- 230000006870 function Effects 0.000 claims description 13
- 238000010606 normalization Methods 0.000 claims description 9
- 238000004140 cleaning Methods 0.000 claims description 6
- 238000012423 maintenance Methods 0.000 abstract description 12
- 230000008439 repair process Effects 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 7
- 238000003860 storage Methods 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 5
- 238000009826 distribution Methods 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 5
- 238000007689 inspection Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000001364 causal effect Effects 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 241000699670 Mus sp. Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000035622 drinking Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- Strategic Management (AREA)
- Artificial Intelligence (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a municipal facility fault prediction method and a municipal facility fault prediction device, wherein the method comprises the steps of acquiring data of a plurality of state classification characteristics of municipal facilities collected in real time, preprocessing the data, determining whether the state classification characteristics meet condition independence hypothesis and/or weight equality hypothesis, optimizing a preset naive Bayes classifier according to the state classification characteristics through a pierce correlation coefficient and/or an entropy method if the state classification characteristics do not meet the condition independence hypothesis and/or the weight equality hypothesis, inputting the preprocessed data into the optimized naive Bayes classifier to obtain the municipal facility fault prediction probability, and determining the municipal facility fault prediction according to the sequence of the municipal facility fault prediction probability in all the municipal facility fault prediction probabilities. The problem of the inefficiency of present municipal facilities maintenance, with high costs, rely on more citizens initiatively to report to repair and the manual work is patrolled and examined, can't in time match the demand can be solved.
Description
Technical Field
the embodiment of the invention relates to the technical field of fault prediction, in particular to a municipal facility fault prediction method and device.
background
in recent years, with the rapid development of urbanization in China, the number of people using urban municipal facilities is increased, corresponding equipment faults and maintenance requirements are frequent, and further development of cities is restricted and hindered. Many times, simply increasing the facility maintenance investment or increasing the number of people easily causes imbalance between the demand and the supply end, that is, resources configured in the peak period of people flow are likely to be idle in other periods to cause waste. Therefore, how to fundamentally improve the maintenance efficiency of the existing municipal facilities is the key point for solving the problems.
Disclosure of Invention
The embodiment of the invention provides a municipal facility fault prediction method and device, which are used for solving the problems that the municipal facility maintenance efficiency is low, the cost is high, more people rely on active repair and manual inspection, and the requirements cannot be matched in time at present.
In a first aspect, an embodiment of the present invention provides a municipal facility fault prediction method, including:
Acquiring data of a plurality of state classification characteristics of municipal facilities collected in real time;
Pre-processing data for a plurality of status classification features of the municipal facility;
Determining whether the state classification features meet a condition independence hypothesis and/or a weight equality hypothesis, if not, optimizing a preset naive Bayes classifier through a Pierce correlation coefficient and/or an entropy method according to the state classification features, and inputting the preprocessed data of the state classification features of the municipal facilities into the optimized naive Bayes classifier to obtain a failure prediction probability of the municipal facilities; the preset naive Bayes classifier is obtained by training and learning according to historical data of municipal facilities;
And determining the fault prediction of the municipal facilities according to the sequence of the fault prediction probabilities of the municipal facilities in all the municipal facility fault prediction probabilities.
according to the technical scheme, the data are preprocessed and input into a trained and learned preset naive Bayesian classifier to obtain the fault prediction probability according to the data of the classification characteristics of the multiple states of the municipal facilities collected in real time, and then the fault prediction of the municipal facilities is obtained according to the sequencing in all the municipal facility fault prediction probabilities, so that the problems that the municipal facilities are low in maintenance efficiency and high in cost at present, more people rely on active repair and manual inspection, and the demands cannot be matched in time can be solved.
Optionally, the preprocessing the data of the plurality of status classification features of the municipal facility includes:
and performing data cleaning and normalization processing on the data of the plurality of state classification characteristics of the municipal facilities.
optionally, the optimizing a preset naive bayes classifier by the pierce correlation coefficient and/or the entropy method according to the state classification features includes:
Determining the correlation among the state classification features through a Pierce correlation coefficient, and splitting and combining the state classification features in the preset naive Bayes classifier according to the correlation among the state classification features; and/or determining the weight corresponding to the plurality of state classification features in the preset naive Bayes classifier by an entropy method according to the data of the plurality of state classification features.
optionally, the determining, by using the pierce correlation coefficient, a correlation between the plurality of state classification features, and splitting and combining the plurality of state classification features in the preset naive bayes classifier according to the correlation between the plurality of state classification features includes:
Dividing the plurality of state classification features into a plurality of groups;
Determining the correlation among the state classification features in each group through a Pierce correlation coefficient, splitting the state classification features of which the correlation is smaller than a first threshold value in the state classification features in each group in the preset naive Bayes classifier according to the correlation among the state classification features in each group, and combining the state classification features of which the correlation is greater than or equal to the first threshold value in the state classification features in each group in the preset naive Bayes classifier.
optionally, the training and learning according to the historical data of the municipal facilities to obtain the preset naive bayes classifier includes:
Acquiring historical data of all municipal facilities;
Correcting the historical data of all municipal facilities through Laplace to obtain a training set;
And optimizing a naive Bayes classifier by a Pierce correlation coefficient and/or an entropy method according to the state classification characteristics of the municipal facilities in the training set, and training and learning historical data of the state classification characteristics of the municipal facilities in the training set to obtain the preset naive Bayes classifier.
optionally, after determining the failure prediction of the municipal facility, the method further comprises:
Obtaining the accuracy rate and the recall rate of the municipal facilities;
and adjusting the precision of the preset naive Bayes classifier according to the precision and the recall rate of the municipal facilities.
Optionally, the determining the fault prediction of the municipal facility according to the ranking of the fault prediction probabilities of the municipal facility among all the fault prediction probabilities of the municipal facility includes:
according to the fault prediction threshold value, dividing the sequence of the fault prediction probabilities of all the municipal facilities into a plurality of intervals; each interval corresponds to one fault prediction;
and determining the fault prediction of the municipal facilities according to the intervals corresponding to the sequences of the fault prediction probabilities of the municipal facilities in all the municipal facility fault prediction probabilities.
In a second aspect, an embodiment of the present invention provides a municipal facility fault prediction apparatus, including:
The acquisition unit is used for acquiring data of a plurality of state classification characteristics of the municipal facilities collected in real time;
a processing unit for preprocessing data of a plurality of state classification features of the municipal facility; determining whether the state classification features meet a condition independence hypothesis and/or a weight equality hypothesis, if not, optimizing a preset naive Bayes classifier through a Pierce correlation coefficient and/or an entropy method according to the state classification features, and inputting the preprocessed data of the state classification features of the municipal facilities into the optimized naive Bayes classifier to obtain a failure prediction probability of the municipal facilities; the preset naive Bayes classifier is obtained by training and learning according to historical data of municipal facilities; and determining the fault prediction of the municipal facilities according to the sequence of the fault prediction probabilities of the municipal facilities in all the municipal facility fault prediction probabilities.
Optionally, the processing unit is specifically configured to:
And performing data cleaning and normalization processing on the data of the plurality of state classification characteristics of the municipal facilities.
optionally, the processing unit is specifically configured to:
determining the correlation among the state classification features through a Pierce correlation coefficient, and splitting and combining the state classification features in the preset naive Bayes classifier according to the correlation among the state classification features; and/or determining the weight corresponding to the plurality of state classification features in the preset naive Bayes classifier by an entropy method according to the data of the plurality of state classification features.
Optionally, the processing unit is specifically configured to:
Dividing the plurality of state classification features into a plurality of groups;
Determining the correlation among the state classification features in each group through a Pierce correlation coefficient, splitting the state classification features of which the correlation is smaller than a first threshold value in the state classification features in each group in the preset naive Bayes classifier according to the correlation among the state classification features in each group, and combining the state classification features of which the correlation is greater than or equal to the first threshold value in the state classification features in each group in the preset naive Bayes classifier.
Optionally, the processing unit is specifically configured to:
Acquiring historical data of all municipal facilities;
Correcting the historical data of all municipal facilities through Laplace to obtain a training set;
and optimizing a naive Bayes classifier by a Pierce correlation coefficient and/or an entropy method according to the state classification characteristics of the municipal facilities in the training set, and training and learning historical data of the state classification characteristics of the municipal facilities in the training set to obtain the preset naive Bayes classifier.
Optionally, the processing unit is further configured to:
after determining the failure prediction of the municipal facility, obtaining an accuracy rate and a recall rate of the municipal facility;
And adjusting the precision of the preset naive Bayes classifier according to the precision and the recall rate of the municipal facilities.
optionally, the processing unit is specifically configured to:
according to the fault prediction threshold value, dividing the sequence of the fault prediction probabilities of all the municipal facilities into a plurality of intervals; each interval corresponds to one fault prediction;
And determining the fault prediction of the municipal facilities according to the intervals corresponding to the sequences of the fault prediction probabilities of the municipal facilities in all the municipal facility fault prediction probabilities.
In a third aspect, an embodiment of the present invention further provides a computing device, including:
A memory for storing program instructions;
And the processor is used for calling the program instructions stored in the memory and executing the municipal facility fault prediction method according to the obtained program.
in a fourth aspect, embodiments of the present invention further provide a computer-readable non-volatile storage medium, which includes computer-readable instructions, and when the computer-readable instructions are read and executed by a computer, the computer is caused to execute the above-mentioned municipal facility fault prediction method.
drawings
in order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
fig. 1 is a schematic diagram of a system architecture according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a municipal facility fault prediction method according to an embodiment of the invention;
FIG. 3 is a diagram illustrating a state classification feature according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a municipal facility fault prediction device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 illustrates an exemplary system architecture, which may be a server 100, including a processor 110, a communication interface 120, and a memory 130, to which embodiments of the present invention are applicable.
The communication interface 120 is used for communicating with the sensors and the network devices for collecting data of each municipal facility, and receiving and transmitting information transmitted by the sensors and the network devices to realize communication.
The processor 110 is a control center of the server 100, connects various parts of the entire server 100 using various interfaces and routes, performs various functions of the server 100 and processes data by operating or executing software programs and/or modules stored in the memory 130 and calling data stored in the memory 130. Alternatively, processor 110 may include one or more processing units.
The memory 130 may be used to store software programs and modules, and the processor 110 executes various functional applications and data processing by operating the software programs and modules stored in the memory 130. The memory 130 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to a business process, and the like. Further, the memory 130 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.
it should be noted that the structure shown in fig. 1 is only an example, and the embodiment of the present invention is not limited thereto.
based on the above description, fig. 2 exemplarily shows a flow of a municipal facility fault prediction method provided by the embodiment of the present invention, where the flow may be performed by a municipal facility fault prediction apparatus, which may be located in the server 100 shown in fig. 1, or may be the server 100.
as shown in fig. 2, the process specifically includes:
step 201, data of a plurality of state classification characteristics of the municipal facilities collected in real time are acquired.
in the embodiment of the invention, the urban municipal facility state is judged according to the sensor of the Internet of things, and the sensor is used for collecting and storing attribute variable data of relevant state classification characteristics by taking municipal facilities as a reference. The municipal facility sensor collects and stores data at 10min data sampling intervals. The data acquired by detection are as follows: the service life, the facility utilization rate, the environmental temperature, the humidity, the service life, the maintenance times, the delivery time and the regional people stream density; other required data are: user evaluation index, and facility brand quality index. That is, the status classification characteristics include a use duration, a facility utilization rate, an ambient temperature, humidity, a use duration, a number of maintenance times, a factory time, a regional crowd density, a user evaluation index, a facility brand quality index, and the like. The user evaluation index and the facility brand quality index are obtained by capturing internet data through a related e-commerce platform. Specific details are shown in table 1.
TABLE 1
Wherein x is1,x2,…,x10Corresponding to 10 state classification features respectively.
Step 202, preprocessing data of a plurality of state classification features of the municipal facility.
during preprocessing, data cleaning and normalization processing are mainly carried out on data of a plurality of state classification characteristics of the municipal facilities.
wherein the data cleansing may include:
1) And judging and removing abnormal values.
The abnormal value refers to that the data acquired in the sample is obviously deviated from other observed values, for example, when the detection environment temperature is higher than 80 degrees or the equipment utilization rate is greater than 1, the data is more likely to be caused by the recording deviation of the sensor, and the data needs to be distinguished and eliminated so as to avoid influencing the accuracy of the model.
The extreme deviation values can be screened by establishing a confidence interval and a confidence limit, and in a normal distribution, the range within three standard deviations from the average value is generally 99% in total, and the calculation formula is as follows by using a 3 sigma method:
standard difference sigma in the above formula is a method for measuring data discrete degreexiIt is typical to detect data for the moment.Is the average of the overall test data.
2) And (5) judging a high lever point.
high leverage points represent points that have a large impact on model results and, although we need to prevent extreme values from interfering with model building fits in the general sense, care must be taken if high leverage points exist in the data.
3) And supplementing missing data values.
And (3) supplementing missing data and maintaining continuity of the sequence and growth change of the trend based on an arithmetic mean value in mathematical knowledge and an algorithm prediction model (a proximity algorithm and an R language Mice function).
and (4) screening data.
In actual data, a detected duplicate value often occurs, and data with duplicate redundancy can be removed by performing sorting filtering according to the number of times the ID number of the data occurs.
The normalization processing of the data comprises:
when data are analyzed, in order to avoid that indexes with larger values weaken other indexes due to different measured dimensions and further influence the accuracy of the model, data normalization processing can be performed through methods such as interval replacement (converting the characteristic value range of original data into the range of [0,1 ]), and the like.
Step 203, determining whether the multiple state classification features meet a condition independence hypothesis and/or a weight equality hypothesis, if not, optimizing a preset naive Bayes classifier through a Pierce correlation coefficient and/or an entropy method according to the multiple state classification features, and inputting the preprocessed data of the multiple state classification features of the municipal facility into the optimized naive Bayes classifier to obtain the failure prediction probability of the municipal facility.
In the embodiment of the invention, the preset naive Bayes classifier is obtained by training and learning according to the historical data of the municipal facilities, specifically, the historical data of all the municipal facilities can be obtained firstly, then the historical data of all the municipal facilities are corrected through Laplace to obtain a training set, and finally the naive Bayes classifier is optimized through a Pierce correlation coefficient and/or an entropy method according to the state classification characteristics of all the municipal facilities in the training set, and the historical data of the state classification characteristics of all the municipal facilities in the training set is trained and learned to obtain the preset naive Bayes classifier.
The naive Bayes algorithm is a classic algorithm in the field of machine learning, and for given training data, a priori/posterior joint probability distribution is learned firstly; based on the model, for a given input x, the output P (c | x) with the maximum posterior probability is obtained by Bayes ' theorem, and the naive Bayes classifier adopts ' attribute condition independence hypothesis ', that is, for known classification features, all the features are assumed to be independent. As shown in FIG. 3, X represents an observed state value of a municipal facility, X1,X2,…,X10Representing the corresponding ten attribute types, namely the state classification features. Each being differentXiContains a set of observations over a period of time.
Wherein, Bayes theorem and attribute condition independent hypothesis:
P(x|c)＝∏P(xi|c)(4)
In the above formula, c represents the occurrence of a random event, and x represents the relevant influencing factor.
after historical data of all municipal facilities are obtained, firstly, splitting and combining state classification features of a naive Bayes classifier according to a Pierce correlation coefficient, secondly, calculating the weight of the state classification features by using an entropy method, then, carrying out sample classification through an optimized Bayes optimal classifier, so that expected loss generated by classification errors is minimum, and finally, constructing a prediction model, wherein the specific formula can be shown as formula (17):
in the above formula, P (c)1| x) represents a random event c under the action of a relevant influence factor x (facility utilization rate, ambient temperature.)1(municipal facility failure) occurrence. P (x)i|c1) Represents when the random event c1When it occurs, factor x is affectediThe occurrence probability, namely the ratio of various influencing factors as fault causes to municipal facility fault accidents. P (c)1) Representing the occurrence of random events c without taking into account the relevant factors1the probability of the condition. Independent assumption according to attribute conditions, P (x | c)1)＝∏P(xi|c1). Wherein P (x) is independent of c, and the size sequence of the final result is not affected, so it is ignored. Pi P (x)i|c1) The multiplication of a plurality of conditional probabilities is expressed, and the problem that the conditional joint probability is difficult to calculate is solved. P (c)1) As a priori probability, it can usually represent the failure rate of the equipment when it leaves the factory or the failure rate caused by other irrelevant factorsthe failure rate.
When the historical data of all municipal facilities are corrected, the following processing is carried out:
In calculating P (x | c)1) When x isifor discrete value states, we need only calculate the proportion of the number of samples taken, such as P (if the number of municipal utility repairs exceeds the limit | if the utility fails or not |.A set of sample data representing the time when a facility fails,a set of sample data components representing the ith influencing factor when a facility fails.
the continuous-value state parameters are mainly determined by a Gaussian distribution, i.e. a probability density function, under the assumption that they conform to a normal distributionandRespectively represent the mean and standard deviation of the sample data of the ith influencing factor when a facility fails. In order to maintain the continuity and stability of the function, the data obtained by actual detection needs to be subjected to standard/normalization processing, so that the stability of operation is ensured, and the interference of random factors and volatility is reduced.
In the formula, the conditional probability P (x)i|C＝c1) Because of the probability density function, the value is not limited to [0,1]]within the interval.
It should be noted that, in the early stage of training a model, a cold start problem is usually encountered, i.e. how to construct a relatively accurate model without a large amount of data. Meanwhile, when sample attribute values which never appear in the training set are met, although various influence factors show that the municipal facility is likely to be in failure, the result of the naive Bayes objective function is still 0. For example, when the weather is suddenly cooled or cooled, the environmental temperature does not appear in the previous sample data, which may cause the classifier to perform partitioning incorrectly, and the prediction probability value may be smoothly corrected by using the laplacian method, which is specifically shown in formula (20).
Wherein N isiRepresents the possible number of values of the ith attribute, and represents the estimated value of the ith influencing factor.
After the preset naive bayes classifier is obtained through the method, whether the state classification features meet the condition independence hypothesis and/or the weight equality hypothesis or not needs to be determined, if not, the preset naive bayes classifier is optimized through a pierce correlation coefficient and/or an entropy method according to the state classification features, and then the data of the preprocessed state classification features of the municipal facilities are input into the optimized naive bayes classifier to obtain the failure prediction probability of the municipal facilities, and the method can be specifically realized through one of the following three methods:
the first mode is as follows:
When the condition independence hypothesis and the weight equal hypothesis are determined that the plurality of state classification features do not meet the condition independence hypothesis, determining the correlation among the plurality of state classification features through a Pierce correlation coefficient, and splitting and combining the plurality of state classification features in a preset naive Bayes classifier according to the correlation among the plurality of state classification features; and determining the weight corresponding to the plurality of state classification features in the preset naive Bayes classifier by an entropy method according to the data of the plurality of state classification features. And then inputting the preprocessed data of the plurality of state classification characteristics of the municipal facilities into an optimized naive Bayes classifier to obtain the failure prediction probability of the municipal facilities.
the second mode is as follows:
When the condition independence assumption is not met by the multiple state classification features, the correlation among the multiple state classification features is determined through a pierce correlation coefficient, and the multiple state classification features in the preset naive Bayes classifier are split and combined according to the correlation among the multiple state classification features. And then inputting the preprocessed data of the plurality of state classification characteristics of the municipal facilities into an optimized naive Bayes classifier to obtain the failure prediction probability of the municipal facilities.
The third mode is as follows:
And when the condition that the plurality of state classification features do not meet the weight equality assumption is determined, determining the weights corresponding to the plurality of state classification features in the preset naive Bayes classifier by an entropy method according to the data of the plurality of state classification features. And then inputting the preprocessed data of the plurality of state classification characteristics of the municipal facilities into an optimized naive Bayes classifier to obtain the failure prediction probability of the municipal facilities.
The splitting and the combining of the plurality of state classification features in the preset naive Bayes classifier mainly comprise the steps of firstly splitting the plurality of state classification features into a plurality of groups, then determining the correlation among the state classification features in each group through a Pierce correlation coefficient, splitting the state classification features of which the correlation is smaller than a first threshold value in the state classification features in each group in the preset naive Bayes classifier according to the correlation among the state classification features in each group, and combining the state classification features of which the correlation is larger than or equal to the first threshold value in the state classification features in each group in the preset naive Bayes classifier.
In the concrete implementation process, when the correlation among the plurality of state classification features is determined through the pierce correlation coefficient, and the plurality of state classification features are split and combined according to the correlation among the plurality of state classification features, the fact that although the sensitivity of different types of municipal facilities to the attribute features is different is mainly considered, for example, fire fighting facilities are less sensitive to the change of the environmental temperature relative to garbage recycling equipment, and drainage facilities are more resistant to the humidity, the degree of correlation among the attribute features still needs to be known so as to prevent the condition independent assumption from being influenced too much. Therefore, it is necessary to screen out the state classification features with large correlation through the pierce correlation coefficient, and the splitting of the joint conditional probability is not performed any more. Wherein the correlation between the state classification features can be determined by the following formula (5).
Wherein C represents a possible occurrence, in this context C1Representing the occurrence of a municipal facility failure in a random event, i.e., a municipal facility failure condition.
first, each state classification feature was divided into 4 groups as shown in table 2.
TABLE 2
| Grouping | Status classification features |
| Facility maintenance group | Facility utilization rate and maintenance times |
| Environment group | ambient temperature, humidity, regional density of people stream |
| time group | Long service life, service life and delivery time |
| Evaluation group | User evaluation index and facility brand quality index |
Within each group, there may be some correlation between the status classification features, such as age and time of departure, ambient temperature and humidity, etc. Because having a correlation does not represent the existence of a causal relationship (potential impact), by grouping the state classification features, the possibility of falsely generalizing the correlation into a causal relationship can be avoided to some extent.
Before:P(x|c)＝P(x1,x2,x3,…,x10|c)＝∏P(xi|c)(6)
After:P(x|c)＝P(x1|c)*P(x2|c)*P(xi,xj|c)*…*P(x10|c)(7)
in the above formula, if the feature x is foundiAnd feature xjThe correlation coefficient between the two is large, and the mutual influence can be presumed to exist in certain possibility, so that the joint conditional probability is reserved and the splitting is not carried out any more.
In the embodiment of the present invention, when determining the weight of each state classification feature by using an entropy method, the method may specifically be:
Assuming a public drinking facility, there is no doubt that it has a strong resistance to ambient humidity, but using the facility for a long time certainly results in a high probability of failure. If a wet weather is maintained, then P (humidity is high, facility fault is yes) is recorded each time a facility fault occurs, and the weight of the influence of P (use time is high, facility fault is yes) on the final result P (x | c) is the same for each fault corresponding to the record, so that the conditional probability higher than the actual value is calculated.
The key to solving this problem is to determine the weight of each attribute feature under different conditions, and generally speaking, the larger the information content of an index observation value is, the larger the potential influence thereof may be, and thus the larger the weight is. In information theory, entropy can be used as a measure of uncertainty, which is smaller when the larger the amount of information it contains, and thus the smaller the entropy value. By the entropy method, the discrete degree of the index can be judged, the higher the discrete degree of the index is, the smaller the entropy value is, the larger the information entropy redundancy is, and the larger the weight is. By building weights, situations like "always wet weather" always generalized too high to be the "cause of facility failure" can be avoided.
firstly, because the measurement units of the data are not uniform or the numerical values have larger difference, the data are normalized:
the forward direction index is as follows:
negative direction index:
wherein xijIs the ith value of the jth attribute feature (such as the detector sensing value of the ambient temperature of the attribute feature j at the ith moment).
Secondly, calculating the proportion P of the ith item in the jth attribute feature in the indexij：
Then is the entropy value of the attribute feature:
From the entropy values, the information entropy redundancy can be calculated:
dj＝1-ej (12)
finally, the related weight w of each characteristic attribute is calculatedij：
variable optimization and weight calculation are carried out through a pierce correlation coefficient and an entropy method, so that the assumption that the attribute condition is independent and the weight is equal can be better met, and the precision of the naive Bayes classifier is improved. In the specific implementation process, if the state classification features are independent or contain the same information quantity, one of the state classification features can be simplified moderately to prevent excessive calculation steps from influencing the output efficiency of the naive Bayes classifier.
In the calculation of the weight w of the input features by means of entropyithen, training the Bayes optimal classifier by comprehensively measuring the influence of the state classification features to minimize expected loss caused by classification errors, and finally constructing a prediction model as shown in formula (21).
And step 204, determining the municipal facility fault prediction according to the sequence of the municipal facility fault prediction probability in all the municipal facility fault prediction probabilities.
Specifically, the sequence of the fault prediction probabilities of all the municipal facilities may be divided into a plurality of sections according to a fault prediction threshold, where each section corresponds to one fault prediction. And then determining the fault prediction of the municipal facilities according to the sections corresponding to the sequences of the fault prediction probabilities of the municipal facilities in all the municipal facility fault prediction probabilities.
In the embodiment of the invention, a method of combining online data and offline data is adopted, unlike the traditional method which relies on the detector data of a single node. The data captured by the sensor and the internet are led into the cloud, so that the municipal facility state can be judged and known more accurately and comprehensively due to diversified data sources, the facility state can be classified, screened and data mined through a big data technology, for example, in 1 ten thousand running facilities, 2% of failure rate and 18% of sub-health rate can be generated, the rule is searched to provide basis for facility failure state prediction and division, and the method is equivalent to the fact that the failure prediction threshold value between the division areas is determined through data mining.
according to this, the main division of municipal utility failure prediction should be by the probability of a possible failure. Based on this consideration, the municipal utility status is classified into four grades of severe, important, urgent and mild according to the municipal utility status grade and the interval value of the status classification. Where P isrank(c|xi) Ranking of probabilities representing a prediction of failure of a particular municipality over the probabilities of predictions of all municipalities, e.g. Prank(c|xi) The 80% represents a higher prediction probability of the facility failing than eighty percent of all municipal facilities, and the facility failure predictions can be partitioned and categorized according to the failure prediction thresholds for the municipal facilities of the city derived from prior data mining, as shown in table 3.
TABLE 3
through the technical scheme, the prediction and release of the municipal facility faults of the Internet of things of the city block can be realized, and the method can be widely applied to aspects such as intelligent blocks, routing inspection planning and the like. It should be noted that the failure prediction probability threshold in table 3 above is only an exemplary function.
The embodiment of the invention provides a municipal facility fault prediction method, which realizes real-time acquisition of facility information through an Internet of things sensor. The prediction data is obtained by updating and integrating the historical data and the current real-time data, has the advantages of simplicity, clearness, quickness, high speed and stability, and can well describe the change characteristics and the real-time trend of the historical fault state. The naive Bayes classifier is improved, prediction accuracy is improved, and meanwhile the problems of high operation complexity and relatively complex attribute dependence in a tree enhancement or TAN classifier are avoided. The algorithm is simple and efficient, has high convergence speed and relatively requires less sample data. The method can simultaneously deal with the problem of multiple classifications, simplify the complexity of the system and facilitate flexible use in engineering.
The embodiment of the invention discloses a municipal facility fault prediction method and a municipal facility fault prediction device. Then, the assumption that the attribute conditions of the input variables (state classification characteristic data) are independent and equal in weight is checked, and if the assumption conditions are not satisfied, optimization is respectively carried out through a Pearson correlation coefficient and an entropy method. Firstly, training a naive Bayesian classifier through preprocessed data, obtaining initial fault prediction probabilities of municipal facilities, then sequencing the initial fault prediction probabilities of all the municipal facilities, mining the obtained urban municipal facility state big data distribution according to the previous data, correspondingly dividing the initial fault prediction probabilities and obtaining four types of fault prediction types of emergency, important, general and safe. Finally, the problems that the existing municipal facilities are low in maintenance efficiency and high in cost, more citizens are relied on to report and repair initiatively and the manual inspection is carried out, and the demands cannot be matched timely are solved.
The embodiment shows that the method comprises the steps of acquiring data of a plurality of state classification features of the municipal facilities collected in real time, preprocessing the data, inputting the preprocessed data into a preset naive Bayes classifier according to the correlation among the state classification features to obtain the failure prediction probability of the municipal facilities, and determining the failure prediction of the municipal facilities according to the sequence of the failure prediction probability of the municipal facilities in the failure prediction probabilities of all the municipal facilities. According to the method, the problem that the municipal facilities are low in maintenance efficiency and high in cost, more people depend on active repair and manual inspection, and the requirements cannot be matched in time can be solved.
Based on the same technical concept, fig. 4 exemplarily shows a structure of a municipal facility fault prediction apparatus provided by an embodiment of the present invention, which may perform a municipal facility fault prediction process, and the apparatus may be located in the server 100 shown in fig. 1, or may be the server 100.
As shown in fig. 4, the apparatus specifically includes:
The acquiring unit 401 is configured to acquire data of a plurality of state classification features of the municipal facility acquired in real time;
A processing unit 402 for preprocessing data of a plurality of status classification features of the municipality; determining whether the state classification features meet a condition independence hypothesis and/or a weight equality hypothesis, if not, optimizing a preset naive Bayes classifier through a Pierce correlation coefficient and/or an entropy method according to the state classification features, and inputting the preprocessed data of the state classification features of the municipal facilities into the optimized naive Bayes classifier to obtain a failure prediction probability of the municipal facilities; the preset naive Bayes classifier is obtained by training and learning according to historical data of municipal facilities; and determining the fault prediction of the municipal facilities according to the sequence of the fault prediction probabilities of the municipal facilities in all the municipal facility fault prediction probabilities.
optionally, the processing unit 402 is specifically configured to:
and performing data cleaning and normalization processing on the data of the plurality of state classification characteristics of the municipal facilities.
Optionally, the processing unit 402 is specifically configured to:
Determining the correlation among the state classification features through a Pierce correlation coefficient, and splitting and combining the state classification features in the preset naive Bayes classifier according to the correlation among the state classification features; and/or determining the weight corresponding to the plurality of state classification features in the preset naive Bayes classifier by an entropy method according to the data of the plurality of state classification features.
Optionally, the processing unit 402 is specifically configured to:
Dividing the plurality of state classification features into a plurality of groups;
Determining the correlation among the state classification features in each group through a Pierce correlation coefficient, splitting the state classification features of which the correlation is smaller than a first threshold value in the state classification features in each group in the preset naive Bayes classifier according to the correlation among the state classification features in each group, and combining the state classification features of which the correlation is greater than or equal to the first threshold value in the state classification features in each group in the preset naive Bayes classifier.
Optionally, the processing unit 402 is specifically configured to:
Acquiring historical data of all municipal facilities;
Correcting the historical data of all municipal facilities through Laplace to obtain a training set;
And optimizing a naive Bayes classifier by a Pierce correlation coefficient and/or an entropy method according to the state classification characteristics of the municipal facilities in the training set, and training and learning historical data of the state classification characteristics of the municipal facilities in the training set to obtain the preset naive Bayes classifier.
Optionally, the processing unit 402 is further configured to:
After determining the failure prediction of the municipal facility, obtaining an accuracy rate and a recall rate of the municipal facility;
And adjusting the precision of the preset naive Bayes classifier according to the precision and the recall rate of the municipal facilities.
optionally, the processing unit 402 is specifically configured to:
According to the fault prediction threshold value, dividing the sequence of the fault prediction probabilities of all the municipal facilities into a plurality of intervals; each interval corresponds to one fault prediction;
and determining the fault prediction of the municipal facilities according to the intervals corresponding to the sequences of the fault prediction probabilities of the municipal facilities in all the municipal facility fault prediction probabilities.
based on the same technical concept, an embodiment of the present invention further provides a computing device, including:
A memory for storing program instructions;
and the processor is used for calling the program instructions stored in the memory and executing the municipal facility fault prediction method according to the obtained program.
based on the same technical concept, the embodiment of the invention also provides a computer-readable non-volatile storage medium, which comprises computer-readable instructions, and when the computer-readable instructions are read and executed by a computer, the computer is enabled to execute the municipal facility fault prediction method.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
these computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A method of municipal facility fault prediction, comprising:
acquiring data of a plurality of state classification characteristics of municipal facilities collected in real time;
Pre-processing data for a plurality of status classification features of the municipal facility;
determining whether the state classification features meet a condition independence hypothesis and/or a weight equality hypothesis, if not, optimizing a preset naive Bayes classifier through a Pierce correlation coefficient and/or an entropy method according to the state classification features, and inputting the preprocessed data of the state classification features of the municipal facilities into the optimized naive Bayes classifier to obtain a failure prediction probability of the municipal facilities; the preset naive Bayes classifier is obtained by training and learning according to historical data of municipal facilities;
And determining the fault prediction of the municipal facilities according to the sequence of the fault prediction probabilities of the municipal facilities in all the municipal facility fault prediction probabilities.
2. The method of claim 1, wherein the pre-processing of the data for the plurality of status classification features of the municipality facility comprises:
and performing data cleaning and normalization processing on the data of the plurality of state classification characteristics of the municipal facilities.
3. the method of claim 1, wherein said optimizing a preset naive bayes classifier by said pierce correlation coefficients and/or entropy method as a function of said plurality of state classification features comprises:
Determining the correlation among the state classification features through a Pierce correlation coefficient, and splitting and combining the state classification features in the preset naive Bayes classifier according to the correlation among the state classification features; and/or determining the weight corresponding to the plurality of state classification features in the preset naive Bayes classifier by an entropy method according to the data of the plurality of state classification features.
4. the method of claim 3, wherein said determining correlations among the plurality of state classification features by a Pierce correlation coefficient, and based on the correlations among the plurality of state classification features, splitting and combining the plurality of state classification features in the preset naive Bayes classifier comprises:
Dividing the plurality of state classification features into a plurality of groups;
Determining the correlation among the state classification features in each group through a Pierce correlation coefficient, splitting the state classification features of which the correlation is smaller than a first threshold value in the state classification features in each group in the preset naive Bayes classifier according to the correlation among the state classification features in each group, and combining the state classification features of which the correlation is greater than or equal to the first threshold value in the state classification features in each group in the preset naive Bayes classifier.
5. The method of claim 1, wherein training and learning from municipal utility historical data to obtain the preset naive bayes classifier comprises:
acquiring historical data of all municipal facilities;
correcting the historical data of all municipal facilities through Laplace to obtain a training set;
and optimizing a naive Bayes classifier by a Pierce correlation coefficient and/or an entropy method according to the state classification characteristics of the municipal facilities in the training set, and training and learning historical data of the state classification characteristics of the municipal facilities in the training set to obtain the preset naive Bayes classifier.
6. The method of claim 1, after determining the prediction of the municipal utility's failure, further comprising:
Obtaining the accuracy rate and the recall rate of the municipal facilities;
And adjusting the precision of the preset naive Bayes classifier according to the precision and the recall rate of the municipal facilities.
7. The method of any one of claims 1 to 6 wherein determining the municipal facility fault prediction based on the ranking of the municipal facility fault prediction probabilities among all the municipal facility fault prediction probabilities comprises:
According to the fault prediction threshold value, dividing the sequence of the fault prediction probabilities of all the municipal facilities into a plurality of intervals; each interval corresponds to one fault prediction;
And determining the fault prediction of the municipal facilities according to the intervals corresponding to the sequences of the fault prediction probabilities of the municipal facilities in all the municipal facility fault prediction probabilities.
8. a utility fault prediction device, comprising:
The acquisition unit is used for acquiring data of a plurality of state classification characteristics of the municipal facilities collected in real time;
a processing unit for preprocessing data of a plurality of state classification features of the municipal facility; determining whether the state classification features meet a condition independence hypothesis and/or a weight equality hypothesis, if not, optimizing a preset naive Bayes classifier through a Pierce correlation coefficient and/or an entropy method according to the state classification features, and inputting the preprocessed data of the state classification features of the municipal facilities into the optimized naive Bayes classifier to obtain a failure prediction probability of the municipal facilities; the preset naive Bayes classifier is obtained by training and learning according to historical data of municipal facilities; and determining the fault prediction of the municipal facilities according to the sequence of the fault prediction probabilities of the municipal facilities in all the municipal facility fault prediction probabilities.
9. The apparatus as claimed in claim 8, wherein said processing unit is specifically configured to:
And performing data cleaning and normalization processing on the data of the plurality of state classification characteristics of the municipal facilities.
10. the apparatus as claimed in claim 8, wherein said processing unit is specifically configured to:
Determining the correlation among the state classification features through a Pierce correlation coefficient, and splitting and combining the state classification features in the preset naive Bayes classifier according to the correlation among the state classification features; and/or determining the weight corresponding to the plurality of state classification features in the preset naive Bayes classifier by an entropy method according to the data of the plurality of state classification features.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910819011.5A CN110555477A (en) | 2019-08-30 | 2019-08-30 | municipal facility fault prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910819011.5A CN110555477A (en) | 2019-08-30 | 2019-08-30 | municipal facility fault prediction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110555477A true CN110555477A (en) | 2019-12-10 |
Family
ID=68738645
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910819011.5A Pending CN110555477A (en) | 2019-08-30 | 2019-08-30 | municipal facility fault prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110555477A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111080039A (en) * | 2020-03-17 | 2020-04-28 | 浙江上风高科专风实业有限公司 | Fan cluster fault prediction method and system |
| CN111458630A (en) * | 2020-04-09 | 2020-07-28 | 广东电网有限责任公司 | Fault diagnosis method for high-voltage circuit breaker |
| CN111638458A (en) * | 2020-06-23 | 2020-09-08 | 广州小鹏汽车科技有限公司 | Method and device for analyzing battery cell fault |
| CN112035286A (en) * | 2020-08-25 | 2020-12-04 | 海尔优家智能科技（北京）有限公司 | Method and device for determining fault cause, storage medium and electronic device |
| CN112990580A (en) * | 2021-03-17 | 2021-06-18 | 中海石油(中国)有限公司 | Drilling overflow early warning method and system based on Bayesian algorithm and storage medium |
| CN113065580A (en) * | 2021-03-17 | 2021-07-02 | 国能大渡河大数据服务有限公司 | Power plant equipment management method and system based on multi-information fusion |
| CN116976862A (en) * | 2023-09-20 | 2023-10-31 | 山东国研自动化有限公司 | Factory equipment informatization management system and method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105425777A (en) * | 2015-12-11 | 2016-03-23 | 渤海大学 | Chemical process fault monitoring method based on active learning |
| CN108764637A (en) * | 2017-06-30 | 2018-11-06 | 亳州中药材商品交易中心有限公司 | Chinese medicine price composite index assessment method, equipment and medium |
| CN109036568A (en) * | 2018-09-03 | 2018-12-18 | 浪潮软件集团有限公司 | Method for establishing prediction model based on naive Bayes algorithm |
| CN109409620A (en) * | 2018-12-29 | 2019-03-01 | 国网福建省电力有限公司 | A kind of intelligent electric meter accident analysis and forecasting system based on big data platform |
| CN110070118A (en) * | 2019-04-10 | 2019-07-30 | 广东电网有限责任公司 | A kind of multi-space data fusion method |
-
2019
- 2019-08-30 CN CN201910819011.5A patent/CN110555477A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105425777A (en) * | 2015-12-11 | 2016-03-23 | 渤海大学 | Chemical process fault monitoring method based on active learning |
| CN108764637A (en) * | 2017-06-30 | 2018-11-06 | 亳州中药材商品交易中心有限公司 | Chinese medicine price composite index assessment method, equipment and medium |
| CN109036568A (en) * | 2018-09-03 | 2018-12-18 | 浪潮软件集团有限公司 | Method for establishing prediction model based on naive Bayes algorithm |
| CN109409620A (en) * | 2018-12-29 | 2019-03-01 | 国网福建省电力有限公司 | A kind of intelligent electric meter accident analysis and forecasting system based on big data platform |
| CN110070118A (en) * | 2019-04-10 | 2019-07-30 | 广东电网有限责任公司 | A kind of multi-space data fusion method |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111080039A (en) * | 2020-03-17 | 2020-04-28 | 浙江上风高科专风实业有限公司 | Fan cluster fault prediction method and system |
| CN111080039B (en) * | 2020-03-17 | 2023-10-20 | 浙江上风高科专风实业有限公司 | Fan cluster fault prediction method and system |
| CN111458630A (en) * | 2020-04-09 | 2020-07-28 | 广东电网有限责任公司 | Fault diagnosis method for high-voltage circuit breaker |
| CN111458630B (en) * | 2020-04-09 | 2022-09-06 | 广东电网有限责任公司 | Fault diagnosis method for high-voltage circuit breaker |
| CN111638458A (en) * | 2020-06-23 | 2020-09-08 | 广州小鹏汽车科技有限公司 | Method and device for analyzing battery cell fault |
| CN111638458B (en) * | 2020-06-23 | 2022-08-16 | 广州小鹏汽车科技有限公司 | Method and device for analyzing battery cell fault |
| CN112035286A (en) * | 2020-08-25 | 2020-12-04 | 海尔优家智能科技（北京）有限公司 | Method and device for determining fault cause, storage medium and electronic device |
| CN112990580A (en) * | 2021-03-17 | 2021-06-18 | 中海石油(中国)有限公司 | Drilling overflow early warning method and system based on Bayesian algorithm and storage medium |
| CN113065580A (en) * | 2021-03-17 | 2021-07-02 | 国能大渡河大数据服务有限公司 | Power plant equipment management method and system based on multi-information fusion |
| CN113065580B (en) * | 2021-03-17 | 2024-04-16 | 国能大渡河大数据服务有限公司 | Power plant equipment management method and system based on multi-information fusion |
| CN116976862A (en) * | 2023-09-20 | 2023-10-31 | 山东国研自动化有限公司 | Factory equipment informatization management system and method |
| CN116976862B (en) * | 2023-09-20 | 2024-01-02 | 山东国研自动化有限公司 | Factory equipment informatization management system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110555477A (en) | municipal facility fault prediction method and device | |
| KR101984730B1 (en) | Automatic predicting system for server failure and automatic predicting method for server failure | |
| CN105653444B (en) | Software defect fault recognition method and system based on internet daily record data | |
| CN110400022B (en) | Cash consumption prediction method and device for self-service teller machine | |
| CN116467674B (en) | Intelligent fault processing fusion updating system and method for power distribution network | |
| CN116579768B (en) | Power plant on-line instrument operation and maintenance management method and system | |
| CN113177857B (en) | Resource elastic allocation method based on typhoon disaster estimation | |
| CN110288035A (en) | A kind of online autonomous learning method and system of intelligent garbage bin | |
| CN107426019A (en) | Network failure determines method, computer equipment and computer-readable recording medium | |
| CN116843071B (en) | Transportation network operation index prediction method and device for intelligent port | |
| CN112415331A (en) | Power grid secondary system fault diagnosis method based on multi-source fault information | |
| CN113988210A (en) | Method and device for restoring distorted data of structure monitoring sensor network and storage medium | |
| CN111813644B (en) | Evaluation method and device for system performance, electronic equipment and computer readable medium | |
| CN112192318A (en) | Machining tool state monitoring method and system | |
| KR101960755B1 (en) | Method and apparatus of generating unacquired power data | |
| CN114118793A (en) | Local exchange risk early warning method, device and equipment | |
| CN115858606A (en) | Method, device and equipment for detecting abnormity of time series data and storage medium | |
| CN115114124A (en) | Host risk assessment method and device | |
| CN110413482B (en) | Detection method and device | |
| CN116484219A (en) | Water supply network water quality abnormal pollution source identification method based on gate control graph neural network | |
| CN115310693A (en) | Comprehensive energy data cleaning method, system, equipment and medium | |
| CN115598459A (en) | Power failure prediction method for 10kV feeder line fault of power distribution network | |
| Tchórzewska-Cieślak et al. | Bayesian inference in the analysis of the failure risk of the water supply network | |
| CN112580260A (en) | Method and device for predicting water flow of pipe network and computer readable storage medium | |
| CN117592865B (en) | Equipment spare part quality state prediction method and device |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200811 Address after: 266555 Qingdao economic and Technological Development Zone, Shandong, Hong Kong Road, No. 218 Applicant after: QINGDAO HISENSE ELECTRONIC INDUSTRY HOLDING Co.,Ltd. Address before: 266071 No. 151, Zhuzhou Road, Laoshan District, Shandong, Qingdao Applicant before: QINGDAO HISENSE TRANSTECH Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201127 Address after: Donghai West Road 266071 Shandong city of Qingdao province No. 17 Applicant after: HISENSE Co.,Ltd. Address before: 266555 Qingdao economic and Technological Development Zone, Shandong, Hong Kong Road, No. 218 Applicant before: QINGDAO HISENSE ELECTRONIC INDUSTRY HOLDING Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191210 |
|
| RJ01 | Rejection of invention patent application after publication |