CN116343953B - Intelligent community management system based on artificial intelligence - Google Patents
Intelligent community management system based on artificial intelligence Download PDFInfo
- Publication number
- CN116343953B CN116343953B CN202310620549.XA CN202310620549A CN116343953B CN 116343953 B CN116343953 B CN 116343953B CN 202310620549 A CN202310620549 A CN 202310620549A CN 116343953 B CN116343953 B CN 116343953B
- Authority
- CN
- China
- Prior art keywords
- data
- data point
- optimization
- trend
- point
- 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.)
- Active
Links
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 19
- 238000005457 optimization Methods 0.000 claims abstract description 97
- 238000012544 monitoring process Methods 0.000 claims abstract description 87
- 238000000034 method Methods 0.000 claims abstract description 56
- 238000001514 detection method Methods 0.000 claims abstract description 5
- 238000010276 construction Methods 0.000 claims abstract 2
- 230000008859 change Effects 0.000 claims description 50
- 230000008569 process Effects 0.000 claims description 22
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 239000004973 liquid crystal related substance Substances 0.000 claims description 10
- 230000007704 transition Effects 0.000 claims description 8
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 3
- 238000007619 statistical method Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 2
- 230000007613 environmental effect Effects 0.000 description 9
- 230000000630 rising effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000003915 air pollution Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 208000023504 respiratory system disease Diseases 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/06—Investigating concentration of particle suspensions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- 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
- 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
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/20—Identification of molecular entities, parts thereof or of chemical compositions
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/90—Programming languages; Computing architectures; Database systems; Data warehousing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Business, Economics & Management (AREA)
- Chemical & Material Sciences (AREA)
- Data Mining & Analysis (AREA)
- Probability & Statistics with Applications (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Fuzzy Systems (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Mathematical Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Medical Informatics (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Emergency Management (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
Abstract
The invention relates to the technical field of electronic digital data processing, in particular to an artificial intelligence-based intelligent community management system, which comprises: and performing outlier factor detection on the acquired data, performing first optimization on the data points according to outlier factor values of the data points and slopes in the time sequence data, performing second optimization on the data points according to similarity between the data points and data contained in local time intervals of adjacent data points, obtaining optimized historical monitoring time sequence data, and performing hidden Markov model construction according to the optimized historical monitoring time sequence data to realize data prediction. According to the method and the device for predicting the data, the historical data are processed, so that the influence of noise data in the historical monitoring time sequence data on the prediction result is avoided, the hidden state corresponding to the data can be accurately obtained by the hidden Markov model when the data is predicted, and the accuracy and the robustness of the prediction result are improved.
Description
Technical Field
The invention relates to the technical field of electronic digital data processing, in particular to an intelligent community management system based on artificial intelligence.
Background
In the intelligent community management system based on artificial intelligence, including security protection monitoring module, entrance guard control module, vehicle management module, energy management module and environmental monitoring module etc. wherein to environmental monitoring module, can prevent and handle environmental risk through environmental factors such as monitoring air quality, noise, temperature and humidity, improve resident quality of life. Wherein to PM 2.5's real-time supervision and early warning can ensure resident's health, through reminding resident's PM 2.5's condition and impel resident's wearing the gauze mask can avoid resident's respiratory disease's emergence to can effectively ensure resident's health, reduce pollution sources and discharge, improve community management efficiency, rational utilization resource, for example reasonable arrangement outdoor activity time, adjustment public facilities live time etc. are in order to reduce resident's contact PM 2.5's risk.
The existing method for early warning according to environment monitoring data in a community is that a hidden Markov model is built for environment monitoring historical data, the most likely hidden state sequence is obtained according to real-time monitoring data through a dimension and thus algorithm, the hidden state sequence in the current scene is the environment state in the community, and the pollution condition of PM2.5 in the community can be divided into according to the requirement of prediction: preferably, the method is good, generally, the method is light in pollution, moderate in pollution and heavy in pollution, marks are made according to the six states in PM2.5 monitoring historical time sequence data in the community, and transition probabilities among the states are obtained.
When the state labeling is carried out on the data points in the historical monitoring time sequence data in the community, the monitored historical time sequence data contains noise data points, and when the labeling is carried out according to the numerical values of the monitored data points, the prediction can only be carried out through the numerical value change in the historical data in the hidden Markov model. In practical situations, when a data point has a severe rising trend in a changing process, an abnormality or noise of the rising trend needs to be judged according to information in historical data, and a state prediction result of the data point for a subsequent data point is accurately obtained.
According to the method, when the data point state labeling is carried out on historical monitoring time sequence data of the community air PM2.5, noise data points are screened according to the local change states of the data points, trend change analysis is carried out on the screened data points, so that the state of each data point contains current numerical information and current change trend information, a hidden Markov model is built to predict a subsequent hidden state sequence, prediction of air quality in the community can be achieved according to a prediction result, and better travel advice is provided for community residents.
Disclosure of Invention
The invention provides an intelligent community management system based on artificial intelligence to solve the existing problems.
The intelligent community management system based on artificial intelligence adopts the following technical scheme:
the invention provides an artificial intelligence-based intelligent community management system, which comprises the following modules:
and a data acquisition module: acquiring historical monitoring time sequence data of PM2.5 concentration in a community;
and a data trend analysis module: taking the COF outlier factor value as the noise degree of the data points, marking a sequence formed by all the data points in the SBN path corresponding to each data point as a path sequence, obtaining a trend optimization coefficient according to the slope of the data points in the path sequence in the historical monitoring time sequence data, and obtaining a new trend optimization factor according to the adjustment result of the trend optimization coefficient on the slope of the data at two sides of any data point;
and a data first optimization module: according to the noise degree and the new trend optimization factor, the data points are adjusted, and a first optimization result of the numerical value corresponding to the data points is obtained;
and a data similarity analysis module: recording any data point as a target data point, recording a time interval corresponding to a path sequence of the target data point in historical monitoring time sequence data as a local time interval of the data point, respectively acquiring data points corresponding to a local time interval with a left end point corresponding to the local time interval of the target data point as a right end point and a local time interval with a right end point as a left end point as left data points and right data points, respectively recording the data points as left data points and right data points, acquiring the path sequence corresponding to the left data point and the right data point, and acquiring the overall trend change similarity degree of the target data point according to the difference of trend optimization coefficients between the target data point and the left data point and the right data point respectively;
and a data secondary optimization module: adjusting the first optimization result of the data points according to the overall trend change similarity degree to obtain the second optimization result of the data points in the historical monitoring time sequence data;
the community management module: and constructing a hidden Markov model according to the optimized historical monitoring time sequence data obtained by the second optimization to obtain an air quality prediction result, and broadcasting and early warning community residents.
Further, the new trend optimization factor is obtained by the following steps:
acquiring COF outlier factor values of all data points in historical monitoring time sequence data by using an outlier factor detection algorithm based on connectivity, performing linear normalization, and marking the normalized result of the COF outlier factor values of the ith data point asIndicating the noise level of the i-th data point;
acquiring slopes of all data points in the path sequence in historical monitoring time sequence data, acquiring the number of data points with continuous same positive and negative directions of all the slopes in time sequence, marking the number as a first number, marking the duty ratio of the first number on the number of the data points in the path sequence as a trend characteristic coefficient, taking the normalized slopes in the sequence of the path sequence corresponding to any data point arranged according to the time sequence as a trend optimization coefficient of the data points, adjusting the average value of the data point slopes of the target data point at the left side and the right side by utilizing the trend optimization coefficient of the data points at the left side and the right side of any data point, marking the average value as a first average value of the data points, multiplying the first average value of the data points according to the trend characteristic coefficient, and acquiring a new trend optimization factor.
Further, the first optimization result is obtained by the following steps:
and continuously multiplying the noise degree and the new trend optimization factor by the numerical value of the data point to obtain a first optimization result of the data point.
Further, the left data point and the right data point are acquired by the following steps:
taking any data point as a target data point, taking the target data point as a starting point, respectively traversing the data points corresponding to the left and right time sequences in the historical monitoring time sequence data one by one, obtaining a local time interval of each data point in the traversing process, and taking a data point when the right end point of the first local time interval in the traversing process is the same as the left end point T of the local time interval [ T, T ] of the target data point as the left data point of the target data point;
meanwhile, the data point when the left end point of the first local time interval in the traversal process is the same as the right end point T of the local time interval [ T, T ] of the target data point is recorded as the right data point of the target data point.
Further, the overall trend change similarity degree is obtained by the following steps:
firstly, according to a trend optimization coefficient acquisition method, an ith data point and a left data point are acquiredZb i Right data pointYb i Trend optimization coefficients of any data points in the corresponding path sequence are respectively recorded asξ i 、ξ(Zb i ) Andξ(Yb i ) The ith data point is respectively related to the left data pointZb i Right data pointYb i The absolute value of the difference between the trend optimization coefficients of the jth data point in the corresponding path sequence is respectively recorded as the left trend difference deltaZξ ij And right trend difference deltaYξ ij The method comprises the steps of carrying out a first treatment on the surface of the Acquiring the ith data point and the left data point respectivelyZb i Right data pointYb i The positive and negative differences of the slopes between the jth data points in the corresponding path sequence are recorded asη ij Andwhen the positive and negative signs of the slopes are the sameη ij And sum->1, when differentη ij And->Is 0; then, the degree of similarity of the global trend change of the ith data point is obtained +.>：
;
Wherein delta isZξ ij Representing the i-th data point and the left data pointZb i Absolute value of difference between trend optimization coefficients of jth data point in corresponding path sequence, deltaYξ ij Representing the i-th data point and the right data pointYb i The absolute value of the difference between trend optimization coefficients of the jth data point in the corresponding path sequence;η ij andrespectively representing the ith data point and the left data pointZb i Right data pointYb i The positive and negative differences of slopes between the jth data points in the corresponding path sequence;NKrepresentation ofKNumber of data points in the distance neighborhood.
Further, the second optimization result is obtained by the following steps:
and (3) regulating the product of the overall trend change similarity degree on the first optimizing result to serve as second optimizing for the data points in the historical monitoring time sequence data, and obtaining the second optimizing result for the data points in the historical monitoring time sequence data.
Further, the method comprises the following specific steps of:
the method comprises the steps of (1) forming optimized historical monitoring time sequence data by using second optimization results corresponding to all data points, marking the environment state corresponding to the data points according to the threshold value of PM2.5 set by an air quality index AQI standard, and obtaining abnormal state information in the historical monitoring time sequence data by taking the data exceeding the threshold value of PM2.5 as abnormal state information;
step (2), after the hidden state of each data point in PM2.5 historical monitoring time sequence data in the community is obtained, a probability transition matrix of the air quality state can be obtained in the optimized historical monitoring time sequence data through a statistical method, so that a hidden Markov model for carrying out real-time monitoring and early warning on the air quality in the community is established;
step (3), for each real-time monitoring time sequence data, obtaining a prediction result of a hidden state sequence according to a Viterbi algorithm through a probability transition matrix obtained by a hidden Markov model, wherein the prediction result is the most likely subsequent hidden state sequence obtained according to the hidden Markov model and the current data point, and obtaining a prediction result of PM2.5 content in a community and a corresponding air quality state;
and (4) after the air quality state corresponding to the PM2.5 content in the community is obtained, broadcasting the air quality state prediction result of the community through an intelligent voice system according to the intelligent community management system, and then early warning the air quality prediction result in the community to residents and providing travel advice.
The technical scheme of the invention has the beneficial effects that:
(1) Based on the method for screening noise data points through PM2.5 historical monitoring time sequence data in community air and optimizing the hidden state division of each data point according to the trend change of the screened data points, compared with the traditional clustering method for clustering the states of the data points, the method can eliminate the influence of noise data points which can cause abnormal trend judgment in the historical monitoring time sequence data on the basis of data point state determination according to the trend change of the data points, and accurately acquire the hidden state division result of the data points at one time.
(2) According to the method, path distance analysis is carried out on the local traversal path of each data point in the community PM2.5 historical monitoring time sequence data, and the numerical classification result of the data point is corrected through the local data point change trend on the basis of eliminating the influence of noise data points in the time sequence data, so that accurate hidden state determination is obtained. If the value of a data point should be classified as slightly contaminated, but in the local trend of the sampled point, the trend is rising sharply, then the data point should be judged as slightly contaminated, so that the change of the hidden state can be predicted in advance when the data point becomes severely contaminated in the subsequent time point.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a block flow diagram of an artificial intelligence based intelligent community management system of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of the specific implementation, structure, characteristics and effects of the intelligent community management system based on artificial intelligence according to the invention with reference to the attached drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the intelligent community management system based on artificial intelligence provided by the invention with reference to the accompanying drawings.
Referring to FIG. 1, a block flow diagram of an artificial intelligence based intelligent community management system according to one embodiment of the invention is shown, the system comprising the following blocks:
and a data acquisition module: historical air quality time sequence data is acquired through PM2.5 sensors installed in the intelligent community and is recorded as historical monitoring time sequence data.
And a data trend analysis module:
after the historical monitoring time sequence data is collected, in order to establish a hidden Markov model for community PM2.5 real-time monitoring and early warning, each data point in the historical monitoring time sequence data is required to be divided into corresponding air pollution states.
The existing method for determining the hidden state of the data point comprises the following steps: setting a PM2.5 threshold value through an air quality index AQI standard of an environmental protection department of a country or region where a community is located, and determining the corresponding air pollution state of each data point according to the interval where the PM2.5 content corresponding to each data point is located.
However, in order to more accurately predict the hidden state sequence in the process of establishing the prediction model, it is necessary to further analyze the change trend of the data points, so as to adjust the hidden state of the historical monitoring time sequence data to be smooth, so that the hidden markov model can predict according to the value and the change trend of the current monitoring data, instead of predicting only through the value of the current monitoring data.
In the process of extracting trend information of a data point, because the trend is sensitive to variation fluctuation of the data point, namely when a noise data point or abnormal variation data point appears, the trend variation of the data point before and after the data point is interrupted, the influence of the noise point in the historical monitoring time sequence data needs to be eliminated in the process of extracting trend characteristics of the data point, so that the trend judgment of the data point is accurately carried out.
Therefore, local trend analysis is required to be carried out on data points in collected community historical monitoring time sequence data, K distance neighborhood parameters of an outlier factor detection algorithm based on connectivity are preset, noise degree judgment of the data points is carried out according to path changes of the data points in the historical monitoring time sequence data, and relevant paths of the noise data points are optimized according to the local path changes of the data points, so that influence of the noise data points on trend changes of the local data points is reduced, and a result of extracting data point trend information is more accurate;
it should be noted that, in this embodiment, the outlier factor detection algorithm based on connectivity is also referred to as COF algorithm;
the method for dividing the hidden state of the data point according to the embodiment comprises the following steps:
firstly, presetting a K-distance neighborhood parameter of a COF algorithm for detecting a data point local outlier factor, carrying out trend change judgment according to a data point minimum traversal path in the K-distance neighborhood, taking the outlier factor obtained according to the minimum traversal path distance as the noise degree of the data point, and carrying out optimization factor judgment of the noise degree according to the trend information of the path change.
And then, after the noise degree and the optimization factor of the data points are obtained, screening the data points in the time sequence data according to the noise degree of the data points, and carrying out numerical optimization on the screened data points through the optimization factor.
Finally, after optimizing the noise data points, carrying out data point state division optimization according to data point trend change information in the K-distance neighborhood of each data point, carrying out hidden state optimization in a local range according to the state change of the local data point on the numerical value, and obtaining accurate data point hidden state division.
For the collected historical monitoring time sequence data, for each data point, the information acquisition of the data point is required to be carried out through the local information of the data point in the time sequence data, so that the time sequence of data information extraction is ensured.
In this embodiment, the size of a K-distance neighborhood parameter of the COF algorithm is preset to be 10, and the K-distance neighborhood value can be adjusted according to an actual scene, that is, for each data point, 10 data points closest to each data point are found in the time sequence data, where the distance is the euclidean distance in the historical monitoring time sequence data, that is, the euclidean distance between the corresponding data points is obtained by using the time stamp and the value size of the data points.
After determining the values of the K-distance neighborhood used for noise analysis and trend optimization of the data points, acquiring a minimum traversal path according to the data points in the K-distance neighborhood of each data point to acquire COF outlier values of the data points, and taking the COF outlier values as the abnormal degrees of the data points; marking a sequence formed by the data points contained in the minimum traversal path corresponding to any data point as a path sequence according to the minimum traversal path, and then, each data point in the historical monitoring time sequence data corresponds to one path sequence;
it should be noted that, the minimum traversal path for each data point is the SBN path in the COF algorithm, and the COF algorithm and the SBN path in the COF algorithm are all existing algorithms, which are not described in detail in this embodiment.
So far, the COF outlier factor value of each data point in the historical monitoring time sequence data is obtained by utilizing a COF algorithm, and the path sequence corresponding to any data point in the obtained historical monitoring time sequence data is combined with the K-distance neighborhood.
After the minimum traversal path of each data point is obtained, the outlier factor can be calculated according to the local average link distance of the COF algorithm, and the noise degree of the data point is judged according to the calculated outlier factor, but the noise degree of the noise data point in the historical monitoring time sequence data is higher, and when the PM2.5 content in the air changes sharply, the data point with the numerical value changed sharply is considered as the noise data point in the analysis and judgment process according to the local outlier factor in the COF algorithm, so that the erroneous judgment on the state of the data point is caused.
In the judging process of the noise degree, the distance of the minimum traversal path of the data point is needed to be measured, but when the data point is in a local area with frequent trend change, the PM2.5 content of the community is indicated to have fluctuation, and the monitoring value under normal conditions is indicated to have continuous trend change and is in a stable value fluctuation state, so that the judgment of the noise data point of the data point and the judgment of the abnormal data point can be distinguished through the path direction change of the minimum traversal path of the data point, and the part is the trend optimization factor of the data point.
In summary, for a group of historical monitoring time series data of the community PM2.5, the data points to be optimized need to be screened according to the noise degree of the data points, and the data points to be optimized need to be subjected to numerical optimization according to the trend change analysis of the minimum traversal path of the K-distance neighborhood of the data points.
COF outlier values of all data points are obtained, linear normalization is carried out, and the normalized result of the COF outlier values of the ith data point is recorded asIndicating the noise level of the i-th data point;
in addition, the slope formed by the ith-2 and ith-1 data points is obtained and is recorded as the left slope of the ith data pointZk i The method comprises the steps of carrying out a first treatment on the surface of the Acquiring the slope formed by the (i+1) th and (i+2) th data points, and recording the slope as the right slope of the (i) th data pointYk i ；
Acquisition of the ith data point atKThe number of data points with continuous trend change in the range of the distance neighborhood, namely acquiring the corresponding slopes of the data points in the range of the K distance neighborhood, counting the number of the data points when signs of all slopes are continuous in time sequence, and recording as；
Then according to atKObtaining the slope of data points in the range of the distance neighborhoodiTrend optimization factor epsilon for data points i ：
;
Wherein, the liquid crystal display device comprises a liquid crystal display device,indicating that the ith data point is atKThe number of data points with a continuous trend change within the distance neighborhood,NKrepresentation ofKThe number of data points contained in the distance neighborhood,Zk i the left slope of the i-th data point is shown,Yk i the right slope of the i-th data point is shown.
In addition, the value of the ith data point is recorded asWhen the data point in the historical monitoring time sequence data is subjected to trend optimization judgment, the ith data point is required to be subjected to continuous trend change judgment according to the time sequence of the data point in a path sequence corresponding to the data point, the data point in the region is required to be judged without considering the value of the ith data point, when the values of the data points at two sides of the ith data point are in a continuous state, if the value of the ith data point deviates, the noise influence appears in the value of the ith data point, and the value of the ith data point is required to be adjusted according to the trend continuous change of the data points at two sides.
In this judging process, becauseThe adjustment tendency of (2) is based onKContinuous trend changes on both sides of the ith data point in the neighborhood are judged so as +.>The greater the number of (2), the greater the number of (2) for>The more necessary the adjustment is. For the adjusted size, this is based on +.>Data points twoThe determination of the arrival slope and departure slope of the data points on the sides is made only when there is a direction of change of the continuous trend, so that only the slopes on the sides are considered to be the same direction, and the value of the i-th data point is adjusted based on the average value of the slopes on the sides as a target.
The noise elimination and trend correction of the data points in the historical monitoring time sequence data can be primarily performed through the process, but in the process, because the data points are measured according to continuous trend change, when the PM2.5 content of the community reaches a critical value degree in the rising process, the trend change of the data points is required to be slowed down, so that each data point in the optimized historical monitoring time sequence data can acquire an accurate continuous historical state sequence when environmental state division is performed according to the setting of the PM2.5 threshold value of the air quality index AQI standard of the environmental protection department.
Then the slope change between the data points in the continuous trend change is optimized so that the value of the adjusted i-th data point can be matched to the current change in air PM2.5 content.
In addition, the trend optimization coefficient xi of the ith data point is obtained by using the COF algorithm i ：
；
Wherein, the liquid crystal display device comprises a liquid crystal display device,k i the path sequence representing the ith data point is arranged according to the time sequence and then corresponds to the slope,NKrepresentation ofKThe number of data points contained in the distance neighborhood, e, represents a natural constant.
And judging whether the adjustment target size of the ith data point accords with the continuous trend change intensity or not according to the comparison of the trend change intensity among the data points for the path sequence of the ith data point, so that the value of the ith data point can accord with the continuous trend change direction and the continuous trend change intensity of the data point in the local area after being adjusted.
By means ofTrend optimization factor vs trend optimization factor epsilon i Optimizing the acquisition method of (1) to obtain optimized trend optimization factors, and recording the optimized trend optimization factors as new trend optimization factorsThe specific acquisition method comprises the following steps:
；
wherein, the liquid crystal display device comprises a liquid crystal display device,new trend optimization factor representing the ith data point, +.>Indicating that the ith data point is atKThe number of data points with a continuous trend change within the distance neighborhood,NKrepresentation ofKThe number of data points contained in the distance neighborhood,Zk i the left slope of the i-th data point is shown,Yk i the right slope of the i-th data point is shown,ξ i-1 represents the trend optimization coefficient for the i-1 data point,ξ i+1 trend optimization coefficients representing the i+1st data point;
and a data first optimization module: and carrying out numerical adjustment optimization on the data points according to the noise degree of the data points and the new trend optimization factors.
After the trend optimization factor and the new trend optimization factor of the data point are obtained, the numerical value of the data point can be adjusted according to the two optimization factors, and the first time in the historical monitoring time sequence data is obtainediNumerical value of data pointsPerforming a first optimization, and marking the optimization result as +.>The specific acquisition method comprises the following steps:
；
wherein, the liquid crystal display device comprises a liquid crystal display device,first optimization result representing the value of the ith data point,/->Represents the noise level of the ith data point, +.>New trend optimization factor representing the ith data point, +.>Represent the firstiA value of a data point;
compared with the traditional clustering method for classifying the states of the data points, the method for classifying the data points by clusters can eliminate the influence of noise data points which can cause abnormal trend judgment in the historical monitoring time sequence data on the basis of determining the data point states according to the trend changes of the data points, so that an accurate data point hiding state classification result is obtained.
After the trend optimization factor of the ith data point is obtained, the numerical value of the data point can be adjusted through the trend optimization factor, so that the continuous trend of the data point is corrected, and the purposes of eliminating the influence of noise data points which can cause abnormal trend judgment in historical monitoring time sequence data and obtaining the accurate hiding state of the data points are achieved.
And a data similarity analysis module:
in the foregoing process, the purpose is to detect deviation of a single sensor, and in the process of monitoring and early warning the content of PM2.5 in air in a smart community through fusion of a plurality of sensors in the smart community, because there is transition of concentration content states of PM2.5 under different days, data point values in the plurality of sensors can be caused to deviate.
In the denoising process, the noise of the data points in the local range of each sensor needs to be optimized, when the overall environment changes, the data points in the single sensor need to be measured according to the difference between the local integers of the data points, so that whether similar trend changes exist or not is determined, when trend changes exist among the local areas, the change of the PM2.5 content of the area in the community where the sensor exists is normal, and on the whole, the adjustment of the data points can be correspondingly improved, so that the accuracy of the state division of the overall PM2.5 content in time sequence data is ensured.
When the difference between the data segments is large, the overall environment change is indicated, and then the adjustment of the noise data points in the time sequence data can be adjusted to a small extent, because the data points in the area have the overall environment deviation, the adjustment can cause the air PM2.5 early warning model to be over-fitted, and the accuracy of the prediction of the normal environment change is reduced.
For the first time in the historical monitoring time sequence dataiData points because it is necessary to correspond to by not including the ith data pointKComparing the data in the adjacent data points with the data in the adjacent local areas, acquiring the data of the adjacent local areas compared with the data in the path sequence of the ith data point, and acquiring the overall trend change similarity degree of the ith data point according to the acquired data, wherein the method comprises the following steps of:
step (1), for the path sequence of the ith data point, the maximum time stamp T corresponding to the data point in the path sequence i With minimum timestamp t i The interval formed is recorded as the local time interval t of the ith data point i ,T i ]；
Step (2), using the ith data point as a starting point, traversing the data points corresponding to the left and right time sequences in the historical monitoring time sequence data one by one respectively to obtain a local time interval of each data point in the traversing process, wherein the local time interval is traversedThe right end point of the first local time interval and the local time interval t of the ith data point in the process i ,T i ]Is t at the left end point of (2) i The data points at the same time, the left data point marked as the ith data pointZb i The method comprises the steps of carrying out a first treatment on the surface of the At the same time, the left end point of the first local time interval in the traversal process is combined with the local time interval t of the ith data point i ,T i ]Right end point T of (2) i The data points at the same time are recorded as the right data point of the ith data pointYb i ；
Step (3) of determining data points adjacent to the local time zone of the ith data point in the data on both sides of the ith data pointZb i AndYb i and thereby forming a corresponding relation between the ith data point and the left data point and the right data point thereof, for the ith data point, obtaining the data point for comparison with the ith data pointZb i AndYb i in the followingKAnd a path sequence corresponding to the distance neighborhood range.
Step (4), according to the trend optimization coefficient acquisition method, the ith data point and the left data point are acquiredZb i Right data pointYb i Trend optimization coefficients of any data points in the corresponding path sequence are respectively recorded asξ i 、ξ(Zb i ) Andξ(Yb i ) The ith data point is respectively related to the left data pointZb i Right data pointYb i In the corresponding path sequence, the absolute value of the difference between the trend optimization coefficients of the jth data point is respectively recorded as the left trend difference deltaZξ ij And right trend difference deltaYξ ij The method comprises the steps of carrying out a first treatment on the surface of the Acquiring the ith data point and the left data point respectivelyZb i Right data pointYb i The positive and negative differences of the slopes between the jth data points in the corresponding path sequence are recorded asη ij Andwhen the positive and negative signs of the slopes are the sameη ij And->1, when differentη ij And->Is 0;
obtaining the overall trend change similarity of the ith data point：
;
Wherein delta isZξ ij Representing the i-th data point and the left data pointZb i Absolute value of difference between trend optimization coefficients of jth data point in corresponding path sequence, deltaYξ ij Representing the i-th data point and the right data pointYb i In the corresponding path sequence, the absolute value of the difference value between trend optimization coefficients of the jth data point;η ij andrespectively representing the ith data point and the left data pointZb i Right data pointYb i In the corresponding path sequence, the positive and negative differences of the slopes between the jth data points are obtained;NKrepresentation ofKNumber of data points in the distance neighborhood.
For the overall trend difference between the sections for judgment, counting the trend change direction of each data point, judging the trend magnitude difference through the trend intensity difference, thereby obtaining the overall difference degree of the two sections, and for the obtained front-back overall difference degree, the above pair can be obtained by the valueAnd (3) judging the magnitude of the numerical adjustment of the (c).
And a data secondary optimization module: for the value of the ith data point after the first optimizationBy usingPerforming a second optimization of the same, wherein +.>Second optimization result representing the ith data point,/->Indicating the degree of similarity of the global trend change of the ith data point, +.>The value representing the i-th data point after the first optimization.
The data points in the acquired historical environment monitoring time sequence data are adjusted and optimized to obtain numerical valuesAnd carrying out mean value fusion according to historical monitoring time sequence data acquired by a plurality of sensors installed in the community, and constructing a hidden Markov model by taking the mean value as an observation value.
To this end, the noise level of the data point is obtained through the COF outlier factor of the data point, so that the data point screening is carried out, and the data point is used for the data pointKAnd optimizing the values of the noise data points by path distances and trend changes in the distance neighborhood to obtain optimized historical environment monitoring time sequence data.
The community management module: and predicting the air quality state in the community according to the optimized historical environment monitoring time sequence data, and realizing intelligent management of the community according to a prediction result.
After the optimized historical environment monitoring time sequence data is obtained, the environment state corresponding to the data point can be marked according to the threshold value set by the air quality index AQI standard of the environmental protection department of the country or region where the community is located, wherein the environment state corresponding to the data point is set, the data exceeding the PM2.5 threshold value is used as the abnormal state information, and the abnormal state information in the historical environment monitoring time sequence data is obtained.
Establishing a hidden Markov prediction model through abnormal state information in historical monitoring time sequence data, pre-warning environmental abnormality of communities by using the established hidden Markov model, and broadcasting the pre-warning condition through an intelligent voice system of an intelligent community, wherein the method comprises the following specific steps of:
and (1) after the hidden state of each data point in the PM2.5 historical monitoring time sequence data in the community is obtained, obtaining a probability transition matrix of the air quality state in the optimized historical monitoring time sequence data through a statistical method, so that a hidden Markov model for carrying out real-time monitoring and early warning on the air quality in the community is established.
And (2) for each real-time monitoring time sequence data, obtaining a prediction result of a hidden state sequence according to a Viterbi algorithm through a probability transition matrix obtained by a hidden Markov model, wherein the prediction result is the most likely subsequent hidden state sequence obtained according to the hidden Markov model and the current data point, and obtaining a prediction result of PM2.5 content in a community and a corresponding air quality state.
And (3) after the air quality state corresponding to the PM2.5 content in the community is obtained, broadcasting the air quality state prediction result of the community through an intelligent voice system according to the intelligent community management system, and then early warning the air quality prediction result in the community to residents and providing travel advice.
It should be noted that, in this embodiment, the exp (-x) model is only used to indicate that the result output by the negative correlation and constraint model is within the interval of [0,1 ], and when implemented, other models with the same purpose may be replaced, and this embodiment only uses the exp (-x) model as an example, and is not limited to this, where x refers to the input of the model.
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 (7)
1. An artificial intelligence-based intelligent community management system is characterized by comprising the following modules:
and a data acquisition module: acquiring historical monitoring time sequence data of PM2.5 concentration in a community;
and a data trend analysis module: taking the COF outlier factor value as the noise degree of the data points, and marking a sequence formed by all the data points in an SBN path corresponding to each data point as a path sequence, wherein the SBN path is the minimum traversal path for each data point in a COF algorithm; obtaining a trend optimization coefficient according to the slope of the data points in the path sequence in the historical monitoring time sequence data, and obtaining a new trend optimization factor according to the adjustment result of the trend optimization coefficient on the slope of the data at two sides of any data point;
and a data first optimization module: according to the noise degree and the new trend optimization factor, the data points are adjusted, and a first optimization result of the numerical value corresponding to the data points is obtained;
and a data similarity analysis module: recording any data point as a target data point, recording a time interval corresponding to a path sequence of the target data point in the historical monitoring time sequence data as a local time interval of the target data point, acquiring a left data point and a right data point according to the local time interval of the target data point, acquiring a path sequence corresponding to the left data point and the right data point, and acquiring the overall trend change similarity degree of the target data point according to the difference of trend optimization coefficients between the target data point and the left data point and the right data point respectively;
and a data secondary optimization module: adjusting the first optimization result of the data points according to the overall trend change similarity degree to obtain the second optimization result of the data points in the historical monitoring time sequence data;
the community management module: and constructing a hidden Markov model according to the optimized historical monitoring time sequence data obtained by the second optimization to obtain an air quality prediction result, and broadcasting and early warning community residents.
2. The intelligent community management system based on artificial intelligence of claim 1, wherein the new trend optimization factor is obtained by the following method:
acquiring COF outlier factor values of all data points in historical monitoring time sequence data by using an outlier factor detection algorithm based on connectivity, performing linear normalization, and marking the normalized result of the COF outlier factor values of the ith data point asIndicating the noise level of the i-th data point;
acquiring slopes of all data points in a path sequence in historical monitoring time sequence data, acquiring the number of continuous and same data points in the positive and negative directions of all the slopes in time sequence, marking the number as a first number, marking the duty ratio of the first number on the number of the data points in the path sequence as a trend characteristic coefficient, taking the normalized slopes in the sequence of the path sequence corresponding to any data point arranged according to the time sequence as a trend optimization coefficient of the data points, adjusting the average value of the data point slopes of the target data point on the left and right sides by utilizing the trend optimization coefficients of the data points on the left and right adjacent sides of any data point, marking the average value as a first average value of the data points, multiplying the first average value of the data points according to the trend characteristic coefficient, and acquiring a new trend optimization factor.
3. The intelligent community management system based on artificial intelligence of claim 1, wherein the first optimization result is obtained by the following method:
and continuously multiplying the noise degree and the new trend optimization factor by the numerical value of the data point to obtain a first optimization result of the data point.
4. The intelligent community management system based on artificial intelligence of claim 1, wherein the left data point and the right data point are acquired by the following method:
taking any data point as a target data point, taking the target data point as a starting point, respectively traversing the data points corresponding to the left and right time sequences in the historical monitoring time sequence data one by one, obtaining a local time interval of each data point in the traversing process, and taking a data point when the right end point of the local time interval appears for the first time in the traversing process and the left end point T of the local time interval [ T, T ] of the target data point are the same as the left end point T of the target data point as the left data point of the target data point;
meanwhile, the data point when the left end point of the local time interval appears for the first time in the traversal process is the same as the right end point T of the local time interval [ T, T ] of the target data point is recorded as the right data point of the target data point.
5. The intelligent community management system based on artificial intelligence of claim 1, wherein the overall trend change similarity is obtained by the following method:
firstly, according to a trend optimization coefficient acquisition method, an ith data point and a left data point Zb are acquired i Right data point Yb i Trend optimization coefficients of any data points in the corresponding path sequence are respectively recorded as、/>And->The ith data point is respectively matched with the left data point Zb i Right data point Yb i In the corresponding path sequence, the absolute value of the difference between the trend optimizing coefficients of the jth data point is respectively marked as left trend difference +.>And right trend difference->The method comprises the steps of carrying out a first treatment on the surface of the Acquiring the ith data point and the left data point Zb respectively i Right data point Yb i In the corresponding path sequence, the positive and negative differences of the slopes between the jth data points are marked as +.>And->When the positive sign and the negative sign of the slopes are the same->And->1, not identical +.>And->Is 0;
then, the overall trend change similarity degree of the ith data point is obtained：
；
Wherein, the liquid crystal display device comprises a liquid crystal display device,representing the i-th data point and the left data point Zb i Absolute value of difference between trend optimization coefficients of jth data point in corresponding path sequence, +.>Representing the ith data point and the right data point Yb i The absolute value of the difference between trend optimization coefficients of the jth data point in the corresponding path sequence; />And->Respectively representing the ith data point and the left data point Zb i Right data point Yb i The positive and negative differences of slopes between the jth data points in the corresponding path sequence; NK denotes the number of data points in the K-distance neighborhood.
6. The intelligent community management system based on artificial intelligence of claim 1, wherein the second optimization result is obtained by the following method:
and carrying out product adjustment on the first optimization result by using the overall trend change similarity degree as a second optimization for the data points in the historical monitoring time sequence data, and obtaining a second optimization result for the data points in the historical monitoring time sequence data.
7. The intelligent community management system based on artificial intelligence according to claim 1, wherein the construction of the hidden markov model to obtain the air quality prediction result and broadcast and pre-warn community residents according to the optimized historical monitoring time sequence data obtained by the second optimization comprises the following specific steps:
the method comprises the steps of (1) forming optimized historical monitoring time sequence data by using second optimization results corresponding to all data points, marking the environment state corresponding to the data points according to the threshold value of PM2.5 set by an air quality index AQI standard, and obtaining abnormal state information in the historical monitoring time sequence data by taking the data exceeding the threshold value of PM2.5 as abnormal state information;
step (2), after the hidden state of each data point in PM2.5 historical monitoring time sequence data in the community is obtained, a probability transition matrix of the air quality state can be obtained in the optimized historical monitoring time sequence data through a statistical method, so that a hidden Markov model for carrying out real-time monitoring and early warning on the air quality in the community is established;
step (3), for each real-time monitoring time sequence data, obtaining a prediction result of a hidden state sequence according to a Viterbi algorithm through a probability transition matrix obtained by a hidden Markov model, wherein the prediction result is the most likely subsequent hidden state sequence obtained according to the hidden Markov model and the current data point, and obtaining a prediction result of PM2.5 content in a community and a corresponding air quality state;
and (4) after the air quality state corresponding to the PM2.5 content in the community is obtained, broadcasting the air quality state prediction result of the community through an intelligent voice system according to the intelligent community management system, and then early warning the air quality prediction result in the community to residents and providing travel advice.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310620549.XA CN116343953B (en) | 2023-05-30 | 2023-05-30 | Intelligent community management system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310620549.XA CN116343953B (en) | 2023-05-30 | 2023-05-30 | Intelligent community management system based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116343953A CN116343953A (en) | 2023-06-27 |
| CN116343953B true CN116343953B (en) | 2023-08-01 |
Family
ID=86893391
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310620549.XA Active CN116343953B (en) | 2023-05-30 | 2023-05-30 | Intelligent community management system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116343953B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116561535B (en) * | 2023-07-11 | 2023-09-19 | 安徽建筑大学 | Individualized building interaction design processing method |
| CN116612641B (en) * | 2023-07-19 | 2023-09-22 | 天津中德应用技术大学 | Vehicle queue control data processing method based on intelligent network connection |
| CN116933292B (en) * | 2023-08-17 | 2024-01-30 | 华南师范大学 | Authority authentication method and system for job title review |
| 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 |
| CN116910595B (en) * | 2023-09-14 | 2023-12-08 | 山东省地质矿产勘查开发局八〇一水文地质工程地质大队(山东省地矿工程勘察院) | Efficient storage method for hydraulic circular ecological restoration data |
| 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 |
| CN117318310B (en) * | 2023-11-28 | 2024-02-09 | 浙江隆业电气科技有限公司 | Remote unmanned intelligent monitoring system for explosion-proof power distribution cabinet |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112308414A (en) * | 2020-10-30 | 2021-02-02 | 中移（杭州）信息技术有限公司 | Income abnormity detection method and device, electronic equipment and storage medium |
| CN112527788A (en) * | 2020-12-17 | 2021-03-19 | 北京中恒博瑞数字电力科技有限公司 | Method and device for detecting and cleaning abnormal value of transformer monitoring data |
| CN113484813A (en) * | 2021-06-11 | 2021-10-08 | 国网山东省电力公司营销服务中心(计量中心) | Intelligent ammeter fault rate estimation method and system under multi-environment stress |
| CN113505935A (en) * | 2021-07-26 | 2021-10-15 | 上海东方低碳科技产业股份有限公司 | Electric power abnormal fluctuation detection and prediction calculation method based on integrated algorithm |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115775562B (en) * | 2023-02-13 | 2023-04-07 | 深圳市深羽电子科技有限公司 | Sound leakage detection method for Bluetooth headset |
| CN116011686B (en) * | 2023-03-28 | 2023-06-20 | 山东创宇环保科技有限公司 | Charging shed photovoltaic power generation reserve prediction method based on multi-data fusion |
-
2023
- 2023-05-30 CN CN202310620549.XA patent/CN116343953B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112308414A (en) * | 2020-10-30 | 2021-02-02 | 中移（杭州）信息技术有限公司 | Income abnormity detection method and device, electronic equipment and storage medium |
| CN112527788A (en) * | 2020-12-17 | 2021-03-19 | 北京中恒博瑞数字电力科技有限公司 | Method and device for detecting and cleaning abnormal value of transformer monitoring data |
| CN113484813A (en) * | 2021-06-11 | 2021-10-08 | 国网山东省电力公司营销服务中心(计量中心) | Intelligent ammeter fault rate estimation method and system under multi-environment stress |
| CN113505935A (en) * | 2021-07-26 | 2021-10-15 | 上海东方低碳科技产业股份有限公司 | Electric power abnormal fluctuation detection and prediction calculation method based on integrated algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116343953A (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116343953B (en) | Intelligent community management system based on artificial intelligence | |
| CN112394152A (en) | Water quality real-time intelligent monitoring analysis management system based on big data | |
| CN110865418B (en) | Sewage illegal discharge monitoring method and device and computer readable storage medium | |
| CN117456428B (en) | Garbage throwing behavior detection method based on video image feature analysis | |
| CN114518143A (en) | Intelligent environment sensing system | |
| CN115372571A (en) | Intelligent water environment monitoring system | |
| CN115876258A (en) | Livestock and poultry breeding environment abnormity monitoring and alarming system based on multi-source data | |
| CN116522156A (en) | Equipment state data analysis system and method based on energy management platform | |
| CN117522632A (en) | Water quality index prediction method based on deep learning | |
| CN117538503A (en) | Real-time intelligent soil pollution monitoring system and method | |
| CN116900412B (en) | Steel plate equidistant automatic cutting regulation and control method based on visual processing | |
| CN116721550B (en) | Traffic management system based on grid computing | |
| CN117370813A (en) | Atmospheric pollution deep learning prediction method based on K line pattern matching algorithm | |
| CN103310088A (en) | Automatic detecting method of abnormal illumination power consumption | |
| CN117033935A (en) | Prediction method of rainfall characteristic under statistics and monitoring based on Bayesian fusion | |
| CN116702588A (en) | Wind power plant key weather factor forecasting method and system based on multi-source data | |
| CN115931041A (en) | Novel regional environment-friendly monitoring and early warning method | |
| CN117251749B (en) | Data processing method of Internet of things based on incremental analysis | |
| CN114429148A (en) | Power equipment state detection method based on multi-source data fusion | |
| CN117540327B (en) | Enterprise environment autonomous management data acquisition and processing system | |
| CN116757898B (en) | Flood season pollution intensity accounting method and system based on predictive comparison | |
| CN116029605A (en) | Water environment monitoring method and system | |
| CN113257000B (en) | Intelligent detection early warning system and method for road black ice | |
| CN116029999A (en) | Smoke and flame segmentation method and system based on u2net | |
| CN116665043A (en) | Intelligent fire point identification method integrating physical optimization model and deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |