CN116992322B - Smart city data center management system - Google Patents
Smart city data center management system Download PDFInfo
- Publication number
- CN116992322B CN116992322B CN202311235852.4A CN202311235852A CN116992322B CN 116992322 B CN116992322 B CN 116992322B CN 202311235852 A CN202311235852 A CN 202311235852A CN 116992322 B CN116992322 B CN 116992322B
- Authority
- CN
- China
- Prior art keywords
- data sequence
- data
- temperature
- temperature data
- sequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 claims abstract description 98
- 230000008859 change Effects 0.000 claims abstract description 45
- 230000001360 synchronised effect Effects 0.000 claims abstract description 29
- 230000002159 abnormal effect Effects 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 9
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 44
- 229910052760 oxygen Inorganic materials 0.000 claims description 44
- 239000001301 oxygen Substances 0.000 claims description 44
- 238000000034 method Methods 0.000 claims description 40
- 230000008569 process Effects 0.000 claims description 24
- 230000005856 abnormality Effects 0.000 claims description 12
- 238000005315 distribution function Methods 0.000 claims description 8
- 230000014509 gene expression Effects 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 4
- 230000001105 regulatory effect Effects 0.000 claims description 3
- 230000007613 environmental effect Effects 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 4
- 238000009423 ventilation Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 238000005057 refrigeration Methods 0.000 description 2
- 230000003321 amplification Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000000758 substrate 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
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A30/00—Adapting or protecting infrastructure or their operation
- Y02A30/60—Planning or developing urban green infrastructure
Abstract
The invention relates to the field of data processing, in particular to a smart city data center management system, which comprises: collecting each monitoring data sequence; obtaining synchronous information quantity of the temperature data sequences according to probability distribution curves of the monitoring data sequences; obtaining an independence coefficient of the temperature data sequence according to the change degree between adjacent data points in each group of each monitoring data sequence; obtaining iterative search radius of the temperature data sequence by adopting iterative set information of each iteration of a mean shift clustering algorithm, synchronous information quantity and independence coefficient of the temperature data sequence; and (3) adopting a mean shift clustering algorithm to obtain a temperature abnormal data set for the iterative search radius of the temperature data sequence, and completing the monitoring management of the smart city data center. The change and the connection between each monitoring data are analyzed from the whole and local angles, the abnormal change condition of the environment of the data center machine room can be detected more accurately, and the monitoring management efficiency of the smart city data center is improved.
Description
Technical Field
The application relates to the field of data processing, in particular to a management system of a smart city data center.
Background
In the construction of a smart city, the data center is an important basic resource and is an important component for realizing the smart city, so that the data center should be uniformly managed to ensure the normal operation of the data center. The data center management system comprises equipment management, service management and the like, and is convenient for monitoring, early warning, backup and management of the data center. In the construction of smart cities, stable operation of a data center is particularly important, and stability and high efficiency of a system are maintained by monitoring and detecting operation conditions and abnormal conditions of system equipment.
The traditional mean shift clustering algorithm is unreasonable in window setting, so that the clustering of the environmental monitoring data of the data center is inaccurate, the detection and analysis accuracy of the monitoring data is low, and the environmental abnormality of equipment of the smart city data center is caused, so that the equipment is damaged and economic loss is caused.
In summary, the invention provides a smart city data center management system, which collects each monitoring data sequence through a sensor system, combines the relevance and the relative independence among the monitoring data sequences, improves the iterative search radius of each iteration in a mean shift clustering algorithm, and completes the monitoring management of the smart city data center management system.
Disclosure of Invention
In order to solve the above technical problems, the present invention provides a smart city data center management system, the system comprising:
and a data acquisition module: collecting each monitoring data sequence;
and a data processing module: acquiring probability distribution curves of all monitoring data sequences; obtaining synchronous information quantity of the temperature data sequences in the monitoring data according to the similarity of probability distribution curves of all the monitoring data sequences; uniformly dividing each monitoring data sequence into an array; obtaining the relative change rate of each group according to each data point of each group of the temperature data sequence; acquiring the relative change rate of each group of the humidity and oxygen concentration data sequences; obtaining an independence coefficient of the temperature data sequence according to the relative change rate of each group of each monitoring data sequence;
obtaining iteration set information according to two sets before and after each monitoring data sequence iteration; obtaining first, second and third reference radius dividing degrees of the next iteration of the temperature data sequence according to iteration set information of each iteration and synchronous information quantity and independence coefficient of the temperature data sequence; dividing the first, second and third reference radiuses according to the current iteration search radius of the temperature data sequence to obtain the next iteration search radius of the temperature data sequence;
an anomaly monitoring module: adopting a mean shift clustering algorithm to obtain a cluster of the temperature data sequence by combining the iterative search radius of the temperature data sequence; obtaining a temperature abnormal data set according to the cluster of the temperature data sequence; obtaining an abnormal management index of the temperature data sequence according to the temperature abnormal data set in the temperature data sequence;
respectively acquiring abnormal management indexes of a humidity and oxygen concentration data sequence; and completing the monitoring management of the intelligent city data center according to the abnormal management indexes of each monitoring data sequence.
Preferably, the specific step of obtaining the synchronous information amount of the temperature data sequence in the monitoring data according to the similarity of the probability distribution curves of the monitoring data sequences comprises the following steps:
for the temperature, humidity and oxygen concentration data sequences in the monitoring data, the similarity of the probability distribution function of the humidity data sequence and the probability distribution function of the temperature data sequence is recorded as a first synchronous information quantity, and the similarity of the probability distribution function of the oxygen concentration data sequence and the probability distribution function of the temperature data sequence is recorded as a second synchronous information quantity;
the first synchronization information amount and the second synchronization information amount are used as the synchronization information amount of the temperature data sequence.
Preferably, the specific step of obtaining the relative change rate of each group according to each data point of each group of the temperature data sequence is as follows:
for each group of data points of the temperature data sequence, obtaining a difference value of the variation degree of adjacent data points;
the relative rate of change of each group is obtained by summing the absolute values of the differences of all adjacent data points of each group of the temperature data sequence.
Preferably, the expression for obtaining the difference value of the variation degree of the adjacent data points is:
in the method, in the process of the invention,、/>indicate temperature data sequence->Group->Person, th->Values of individual data points>The values in brackets are indicated as a function +.>Ordinate on the curve, +.>The values in brackets are indicated as a function +.>Slope value on curve, +.>Indicate temperature data sequence->Group->The difference in degree of variation between a data point and the last adjacent data point.
Preferably, the specific step of obtaining the independence coefficient of the temperature data sequence according to the relative change rate of each group of each monitoring data sequence is as follows:
the ratio of the relative change rates of the humidity data sequence and the temperature data sequence corresponding to each group is marked as a first ratio, and the average value of the first ratios of all groups is used as a first independence coefficient of the temperature data sequence;
the ratio of the relative change rate of each group corresponding to the oxygen concentration data sequence and the temperature data sequence is marked as a second ratio, and the average value of the second ratios of all groups is used as a second independence coefficient of the temperature data sequence;
and taking the first independence coefficient and the second independence coefficient of the temperature data sequence as the independence coefficients of the temperature data sequence.
Preferably, the specific step of obtaining the iteration set information according to the two sets before and after each monitoring data sequence iteration is as follows:
for two sets before and after each monitoring data sequence iteration, marking the ratio of the intersection of data points in the two sets to the union as the similarity of the two sets; recording absolute values of differences of numerical value means of all data points in the two sets as iteration difference information of the two sets;
and taking the similarity and iteration difference information of the two sets before and after each monitoring data sequence iteration as iteration set information.
Preferably, the expression for dividing the first, second and third reference radius of the next iteration of the temperature data sequence according to the iteration set information of each iteration, the synchronous information quantity and the independence coefficient of the temperature data sequence is:
in the method, in the process of the invention,to eliminate the parameter with denominator 0, +.>、/>Indicate->Second, th->Searching temperature data point set in circle for multiple iterations, +.>、/>Representation according to the collection->、/>A set of humidity data points at a time corresponding to the temperature data points within,、/>representation according to the collection->、/>Oxygen concentration data point set at the time point corresponding to the temperature data point in +.>Representing the similarity of the intersection of two sets calculated using the Jaccard algorithm, +.>Representing the absolute value of the difference between the mean of two aggregate data points,/->、/>Indicating the amount of synchronization information between the humidity data sequence, the oxygen concentration data sequence and the temperature data sequence, respectively,/->、/>The independence coefficient of the humidity data sequence and the oxygen concentration data sequence relative to the temperature data sequence is represented by +.>、/>、/>Dividing the first, second and third reference radii.
Preferably, the expression for dividing the search radius of the next iteration of the temperature data sequence according to the current iteration of the temperature data sequence and the first, second and third reference radii of the next iteration is as follows:
in the method, in the process of the invention,、/>indicate temperature data sequence->Second, th->The iterative search radius size at the time of sub-clustering,for normalization function->For regulating parameters->、/>、/>Dividing the first, second and third reference radii.
Preferably, the specific step of obtaining the temperature anomaly data set according to the cluster of the temperature data sequence includes:
setting a temperature anomaly threshold; calculating the data average value of each cluster data point of the temperature data sequence, and marking the cluster with the data average value of the cluster data points larger than the temperature abnormality threshold value as a temperature abnormality data set.
Preferably, the specific step of obtaining the abnormality management index of the temperature data sequence according to the temperature abnormality data set in the temperature data sequence includes:
setting a temperature anomaly threshold value, calculating a data average value of a temperature anomaly data set, and recording an absolute value of a difference value between the data average value and the temperature anomaly threshold value as a first coefficient;
and recording the ratio of the first coefficient to the data average value as an abnormal management index of the temperature data sequence.
The invention has at least the following beneficial effects:
compared with the traditional mean shift clustering algorithm, the method of the invention considers the characteristic difference of the data collected at different positions of the machine room during the monitoring management of the smart city center, respectively analyzes the relevance of each monitoring data sequence on the whole to obtain the synchronous information quantity among the monitoring data sequences, and is favorable for linking the different variation trend of the data from the whole angle;
meanwhile, the relative independence of each monitoring data sequence in a local range is analyzed to obtain the relative independence coefficient among the monitoring data sequences, the difference among the monitoring data sequences can be analyzed from the local angle of each monitoring data sequence, and the control of local detail information is increased; the difference between the data sets in each iteration process of each monitoring data sequence is obtained, the relativity and the relative independence between the monitoring data sequences are combined, the reference radius division degree after iteration is obtained based on the synchronous information quantity and the relative independence coefficient, the search radius size in each iteration process in the mean shift clustering process is timely adjusted, the acquired monitoring data of the smart city data center machine room environment is provided with a more accurate clustering effect, the abnormal change condition of the data center machine room environment can be accurately detected, and the monitoring management efficiency of the smart city data center is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a smart city data center management system provided by the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description refers to the specific implementation, structure, characteristics and effects of a smart city data center management system according to the present invention with reference to the accompanying drawings and preferred embodiments. 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 a management system for a smart city data center provided by the present invention with reference to the accompanying drawings.
The embodiment of the invention provides a management system of a smart city data center, which comprises a data acquisition module, a data processing module and an abnormality monitoring module.
Specifically, the system for acquiring and processing engineering information based on internet of things according to the present embodiment provides a smart city data center management system, referring to fig. 1, the method includes the following steps:
and the data acquisition module is used for monitoring and managing the intelligent city data center machine room mainly through a data processing technology. The management of the smart city data center is particularly important, since the equipment for storing various core data is stored in the machine room, environmental changes can be caused in the operation process of the equipment, and meanwhile, the ventilation of the machine room and the operation of refrigeration equipment can also influence the environment. Therefore, a sensor system is formed by adopting a temperature sensor, a humidity sensor and an oxygen concentration detection sensor, and a temperature data sequence, a humidity data sequence and an oxygen concentration data sequence of a data center machine room are respectively acquired.
Because the change of the monitoring data of different positions in the machine room is different in the operation process of the equipment, the sensor system is respectively arranged at the top end of the machine room, the rear part of the machine cabinet, the door opening of the machine room and the ventilation opening according to the temperature change characteristics. In the running process of the data center machine room equipment, larger environmental temperature changes are easily caused to the rear part of the cabinet and the top end of the machine room, so that moisture in the machine room is evaporated, and the humidity in the air is increased, and therefore, whether the environment of the data center machine room is abnormal or not can be accurately judged by monitoring the two positions; meanwhile, energy consumption can be generated in the running process of equipment, the energy consumption can cause the change of the oxygen concentration in the machine room, the ventilation condition of the machine room is greatly influenced on the oxygen concentration and the humidity, and due to the safety consideration, people can influence a human body if the oxygen concentration is insufficient when entering the machine room, so that the sensors are also arranged on the ventilation opening in the machine room and the position of the door of the machine room to collect all monitoring data in the environment.
The monitoring data of three sensors at four positions are respectively collected at the same moment, and the monitoring data of three sensors at the same position are respectively collected at the same momentThree types of monitoring data are collected, and the time interval for collecting each data point is thatThe data amount of one cycle of each monitoring data acquisition is +.>。
So far, each monitoring data of the smart city data center room can be acquired through the module, and each monitoring data sequence is obtained.
And the data processing module is used for respectively analyzing each monitoring data sequence acquired by the sensor system at each position according to different positions of the sensor placed in the machine room. The following takes as an example each monitoring data sequence collected by the sensor system at the rear position of the cabinet.
Because the temperature near the cabinet can be relatively high when the cabinet operates, the temperature near the cabinet fluctuates within a certain range under the condition of stable operation through corresponding refrigeration measures; the oxygen concentration near the cabinet is low, and the change of the oxygen concentration is opposite to the change of the temperature; as the temperature increases, a small amount of moisture in the machine room evaporates, resulting in an increase in the humidity in the air.
For this case, the correlation between different monitoring data sequences of the data center room can be calculated. Based on a sequence of acquired temperature data over a periodHumidity data sequence->Oxygen concentration data sequence->And respectively obtaining probability distribution curves of three groups of data sequences of the same period corresponding to the time points, and calculating the similarity between the curves by adopting a KL divergence algorithm. The KL divergence algorithm is a known technique, and this embodiment will not be described in detail.
In the method, in the process of the invention,representing the similarity of two probability distribution curves calculated by KL divergence algorithm,/for the two probability distribution curves>Probability distribution curve representing a sequence of humidity data, +.>Probability distribution curve representing a temperature data sequence, +.>Probability distribution curve representing an oxygen concentration data sequence, < >>Representing a first amount of synchronization information between the sequence of humidity data and the sequence of temperature data,representing a second amount of synchronization information between the oxygen concentration data sequence and the temperature data sequence.
It should be noted that, when the ambient temperature at the rear part of the cabinet in the data center machine room is abnormal, the closer the change rate of the humidity and the temperature is to the synchronous change, the first synchronous information amount between the calculated humidity and the temperature is calculatedThe larger; similarly, if the change of the oxygen concentration and the change of the temperature are more nearly synchronous, the second synchronous information amount of the oxygen concentration and the temperature is +.>The larger. And taking the first synchronous information quantity between the humidity and the temperature and the second synchronous information quantity between the oxygen concentration and the temperature as the synchronous information quantity of the temperature data sequence.
Since the changes between the three monitoring data are close to synchronous, the changes between the different data sequences actually have respective characteristics. The change of the environmental monitoring data of the data center machine room is often caused by the equipment unit, so that the temperature change is caused by the abnormal operation of the equipment, firstly, the oxygen concentration is suddenly reduced, and the humidity is continuously reduced along with the increase of the environmental temperature.
For this case, each monitoring data sequence at the rear part of the cabinet in one period is uniformly divided intoGroups, each group having->Data, in this embodiment +.>，/>。
To better reflect the rate of change of data, a function is usedThe change in slope reflects the extent to which the data changes. By the>Relative rate of change of group data sequences +.>For example, the calculation formula is as follows:
in the method, in the process of the invention,、/>indicate temperature data sequence->Group->Person, th->Values of individual data points>Is the>The number of data points in the group data, +.>The values in brackets are indicated as a function +.>Ordinate on the curve, +.>The values in brackets are indicated as a function +.>Slope value on curve, +.>Representing temperature data sequence NoRelative rate of change of group data.
It should be noted that the functionThe values of the data points in each group can be used for evaluating the change degree among the data points in the group after being corrected; summing all the first differences in the group according to the first difference of the variation degree of two adjacent data points to obtain the relative variation rate of the group, < >>The larger the description temperature data sequence +.>The relative rate of change of the set of data is large, i.e., each data point in the set of data has a large range of fluctuations.
The above steps are repeated, and the relative change rates of the temperature data sequence, the humidity data sequence and the oxygen concentration data sequence in each group can be obtained.
And according to the relation among the monitoring data sequences among the groups in the same time period, the independence coefficient of the humidity data sequence and the oxygen concentration data sequence relative to the temperature data sequence can be obtained.
In the method, in the process of the invention,as a linear normalization function>For dividing the data collected in one period into groups, +.>、、/>Respectively represent humidity, oxygen concentration and temperature data sequence +.>Relative rate of change of group data, +.>A first coefficient of independence of the humidity data sequence relative to the temperature data sequence, +>A second coefficient of independence of the oxygen concentration data sequence from the temperature data sequence is represented.
It should be noted that, the first independent coefficient and the second independent coefficient are both independent coefficients of the temperature data sequence; when the environment in the data center room changes, the relative change rate of the oxygen concentration data sequence increases due to the sudden drop of the oxygen concentration,relatively increase, thereby making +.>The larger the independence of the oxygen concentration data sequence relative to the temperature data sequence is; similarly, if the humidity changes are increased successively, the relative change rate of the humidity data sequence is increased, +.>Relatively increase, thereby making +.>Larger indicates greater independence of the humidity data sequence from the temperature data sequence.
Through the analysis, the relevance and the relative independence of each monitoring data sequence in three data center environments can be obtained. Clustering analysis is carried out on temperature data sequences in the environment by adopting a mean shift clustering algorithm, input data are temperature data sequences of the environment of a data center machine room, an initial search radius is set by an operator, and the initial search radius is set to be. And each iteration can obtain a data set in the search circle, and the reference radius dividing degree affecting the search radius of the next iteration is calculated according to the intersection of the data sets before and after the iteration and the change of the data values before and after the iteration.
In the method, in the process of the invention,0.01, for eliminating the case of 0 denominator,/->、/>Indicate->Second, th->Searching temperature data point set in circle for multiple iterations, +.>、/>Representation according to the collection->、/>Humidity data point set at the time point corresponding to the temperature data point in +.>、/>Representation according to the collection->、/>A set of oxygen concentration data points at a time corresponding to the temperature data points within,representing the similarity of the intersection of two sets calculated using the Jaccard algorithm, +.>Representing the absolute value of the difference between the mean of two aggregate data points,/->、/>Indicating the amount of synchronization information between the humidity data sequence, the oxygen concentration data sequence and the temperature data sequence, respectively,/->、/>The independence coefficient of the humidity data sequence and the oxygen concentration data sequence relative to the temperature data sequence is represented by +.>、/>、/>Dividing the first, second and third reference radii.
It should be noted that the larger the difference between the average values of the two sets of data points before and after the iteration, namelyThe larger the value of (a) indicates that the two sets before and after iteration are greatly different in value, which means that local information is excessively focused at the moment and the whole is ignoredThe data state is that the reference radius dividing degree is required to be enlarged, the searching radius of the next clustering is indirectly enlarged, and the difference of data point values among the sets is reduced;
the larger the intersection between two set data points before and after the iteration, i.eThe larger the value of (2) is, the less the change of the number of the sets before and after iteration is indicated, which means that the difference between the local detail data information is ignored by considering the whole more at the moment, so that the reference radius dividing degree is required to be reduced, the search radius of the next clustering is indirectly reduced, the intersection of the data between the sets is reduced, and the analysis of the local detail is increased.
And meanwhile, the correlation and the relative independence of the humidity data sequence and the oxygen concentration data sequence relative to the temperature data sequence are combined, when the correlation is larger and the relative independence is smaller, the reference radius division degree of the humidity and oxygen concentration data is subjected to amplification correction, namely the humidity and the oxygen concentration have higher similarity to the temperature data sequence, namely the influence of the data sequence on the next clustering search radius of the temperature data sequence is increased.
And according to the division of the first, second and third reference radiuses of the next iteration respectively obtained by the temperature, humidity and oxygen concentration data sequences, the search radius of each iterative clustering in the traditional mean shift clustering algorithm is adjusted to a corresponding degree, so that the clustering of the environmental monitoring data of the data center is more accurate.
In the method, in the process of the invention,、/>indicate temperature data sequence->Next time,First->The iterative search radius size at the time of sub-clustering,for normalization function->For regulating parameters->、/>、/>Dividing the first, second and third reference radii.
It should be noted that the number of the substrates,the value is 0.5, the implementer can select the value by himself, when the difference between the set before and after iteration is small, namelySmaller and at [0.5,1 ]]When the range is in the range, the radius is reduced, and local detail information is focused; when the difference between the set before and after iteration is large, namely +.>The larger and at [1,1.5 ]]When the data is within the range, the radius is enlarged, and the whole information of the data is focused.
The steps are repeated, the search radius of each monitoring data sequence of each monitoring position in each iteration can be obtained in a self-adaptive mode, and the accuracy of the clustering result is improved.
So far, according to the self-adaptive search radius obtained in the calculation process of each monitoring data sequence of each monitoring position, a corresponding clustering result can be obtained by adopting a mean shift algorithm.
The abnormality monitoring module is used for clustering temperature data sequences at the rear part of the cabinetFor example, the temperature anomaly threshold value of the temperature anomaly data set is set to beCalculating cluster mean value of data in each cluster, and selecting cluster mean value larger than +.>Is recorded as a temperature anomaly data set +.>。/>The value of (2) can be set by the operator himself, and the value of (4) is set in this embodiment.
In the method, in the process of the invention,data mean value of temperature anomaly data set in temperature data sequence at rear part of cabinet>Threshold value representing a judgment of a temperature anomaly data set in a temperature data sequence,/->An abnormality management index indicating a cabinet rear temperature data sequence.
And repeating the steps to obtain the abnormal management indexes of each monitoring data sequence at each position of the data center machine room.
Setting the early warning threshold value asAnd monitoring and early warning are carried out when the abnormal management index is larger than the early warning threshold value. Therefore, the influence range of the environment abnormality and the position of the abnormality can be judged, and the abnormal condition can be monitored and managed rapidly. />The value of (2) can be set by the operator, and the value of (4) is set to 0.1 in the embodiment.
Thus, the monitoring and management of the smart city data center can be completed according to each module in the system of the embodiment.
In summary, the embodiment of the invention provides a smart city data center management system, which collects environmental monitoring data through a sensor system, combines the relevance and the relative independence of the monitoring data, improves the iterative search radius of each iteration in a mean shift clustering algorithm, and completes the monitoring management of the smart city data center management system.
Compared with the traditional mean shift clustering algorithm, the method provided by the embodiment of the invention considers the characteristic difference of the data collected at different positions of a machine room during the monitoring management of the smart city center, and analyzes the relevance of each monitoring data on the whole to obtain the synchronous information quantity among the monitoring data, thereby being beneficial to linking the variation trend of the data from the whole;
meanwhile, the relative independence of each monitoring data in a local range is respectively analyzed to obtain the relative independence coefficient among the monitoring data, the difference among the monitoring data can be analyzed from the local angle of each group of data, and the control of local detail information is increased; the difference between the data sets in each iteration process of each monitoring data is calculated, the correlation and the relative independence between the monitoring data are combined, the reference radius division degree after iteration is obtained based on the synchronous information quantity and the relative independence coefficient, the search radius size in each iteration process in the mean shift clustering process is timely adjusted, the acquired intelligent city data center machine room environment monitoring data are more accurately clustered, the abnormal change condition of the data center machine room environment can be more accurately detected, and the monitoring management efficiency of the intelligent city data center is improved.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
The foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, any modifications, equivalents, improvements, etc. that fall within the principles of the present invention are intended to be included within the scope of the present invention.
Claims (7)
1. A smart city data center management system, the system comprising:
and a data acquisition module: collecting each monitoring data sequence;
and a data processing module: acquiring probability distribution curves of all monitoring data sequences; obtaining synchronous information quantity of the temperature data sequences in the monitoring data according to the similarity of probability distribution curves of all the monitoring data sequences; uniformly dividing each monitoring data sequence into an array; obtaining the relative change rate of each group according to each data point of each group of the temperature data sequence; acquiring the relative change rate of each group of the humidity and oxygen concentration data sequences; obtaining an independence coefficient of the temperature data sequence according to the relative change rate of each group of each monitoring data sequence;
obtaining iteration set information according to two sets before and after each monitoring data sequence iteration; obtaining first, second and third reference radius dividing degrees of the next iteration of the temperature data sequence according to iteration set information of each iteration and synchronous information quantity and independence coefficient of the temperature data sequence; dividing the first, second and third reference radiuses according to the current iteration search radius of the temperature data sequence to obtain the next iteration search radius of the temperature data sequence;
an anomaly monitoring module: adopting a mean shift clustering algorithm to obtain a cluster of the temperature data sequence by combining the iterative search radius of the temperature data sequence; obtaining a temperature abnormal data set according to the cluster of the temperature data sequence; obtaining an abnormal management index of the temperature data sequence according to the temperature abnormal data set in the temperature data sequence;
respectively acquiring abnormal management indexes of a humidity and oxygen concentration data sequence; completing monitoring management of the smart city data center according to the abnormal management indexes of each monitoring data sequence;
the specific steps for obtaining the synchronous information quantity of the temperature data sequence in the monitoring data according to the similarity of the probability distribution curves of the monitoring data sequences are as follows:
for the temperature, humidity and oxygen concentration data sequences in the monitoring data, the similarity of the probability distribution function of the humidity data sequence and the probability distribution function of the temperature data sequence is recorded as a first synchronous information quantity, and the similarity of the probability distribution function of the oxygen concentration data sequence and the probability distribution function of the temperature data sequence is recorded as a second synchronous information quantity;
taking the first synchronous information quantity and the second synchronous information quantity as synchronous information quantities of the temperature data sequence;
the specific steps for obtaining the independence coefficient of the temperature data sequence according to the relative change rate of each group of each monitoring data sequence are as follows:
the ratio of the relative change rates of the humidity data sequence and the temperature data sequence corresponding to each group is marked as a first ratio, and the average value of the first ratios of all groups is used as a first independence coefficient of the temperature data sequence;
the ratio of the relative change rate of each group corresponding to the oxygen concentration data sequence and the temperature data sequence is marked as a second ratio, and the average value of the second ratios of all groups is used as a second independence coefficient of the temperature data sequence;
taking the first independence coefficient and the second independence coefficient of the temperature data sequence as the independence coefficients of the temperature data sequence;
the first, second and third reference radius dividing expressions of the next iteration of the temperature data sequence are obtained according to the iteration set information of each iteration, the synchronous information quantity and the independence coefficient of the temperature data sequence:
in the method, in the process of the invention,to eliminate the parameter with denominator 0, +.>、/>Indicate->Second, th->Searching temperature data point set in circle for multiple iterations, +.>、/>Representation according to the collection->、/>Humidity data point set at the time point corresponding to the temperature data point in +.>、/>Representation according to the collection->、/>Oxygen concentration data point set at the time point corresponding to the temperature data point in +.>Representing the similarity of the intersection of two sets calculated using the Jaccard algorithm, +.>Representing the absolute value of the difference between the mean of two aggregate data points,/->、/>Indicating the amount of synchronization information between the humidity data sequence, the oxygen concentration data sequence and the temperature data sequence, respectively,/->、/>The independence coefficient of the humidity data sequence and the oxygen concentration data sequence relative to the temperature data sequence is represented by +.>、/>、/>Dividing the first, second and third reference radii.
2. The smart city data center management system of claim 1, wherein the steps of obtaining the relative rates of change of the sets based on the sets of data points of the temperature data sequence are:
for each group of data points of the temperature data sequence, obtaining a difference value of the variation degree of adjacent data points;
the relative rate of change of each group is obtained by summing the absolute values of the differences of all adjacent data points of each group of the temperature data sequence.
3. The smart city data center management system of claim 2, wherein the expression for obtaining the difference in the degree of change of adjacent data points is:
in the method, in the process of the invention,、/>indicate temperature data sequence->Group->Person, th->The value of the individual data points is calculated,/>the values in brackets are indicated as a function +.>Ordinate on the curve, +.>The values in brackets are indicated as a function +.>Slope value on curve, +.>Indicate temperature data sequence->Group->The difference in degree of variation between a data point and the last adjacent data point.
4. The smart city data center management system of claim 1, wherein the iterative set information is obtained by iterating the two sets before and after each monitored data sequence, comprising the steps of:
for two sets before and after each monitoring data sequence iteration, marking the ratio of the intersection of data points in the two sets to the union as the similarity of the two sets; recording absolute values of differences of numerical value means of all data points in the two sets as iteration difference information of the two sets;
and taking the similarity and iteration difference information of the two sets before and after each monitoring data sequence iteration as iteration set information.
5. The smart city data center management system of claim 1, wherein the expression for dividing the search radius of the next iteration of the temperature data sequence based on the current iteration search radius of the temperature data sequence and the first, second, and third reference radii of the next iteration is:
in the method, in the process of the invention,、/>indicate temperature data sequence->Second, th->The iterative search radius size at the time of sub-clustering,for normalization function->For regulating parameters->、/>、/>Dividing the first, second and third reference radii.
6. The smart city data center management system of claim 1, wherein the clustering of the temperature data sequences to obtain the temperature anomaly data set comprises the steps of:
setting a temperature anomaly threshold; calculating the data average value of each cluster data point of the temperature data sequence, and marking the cluster with the data average value of the cluster data points larger than the temperature abnormality threshold value as a temperature abnormality data set.
7. The smart city data center management system of claim 1, wherein the specific step of obtaining the anomaly management index of the temperature data sequence from the temperature anomaly data set in the temperature data sequence comprises:
setting a temperature anomaly threshold value, calculating a data average value of a temperature anomaly data set, and recording an absolute value of a difference value between the data average value and the temperature anomaly threshold value as a first coefficient;
and recording the ratio of the first coefficient to the data average value as an abnormal management index of the temperature data sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311235852.4A CN116992322B (en) | 2023-09-25 | 2023-09-25 | Smart city data center management system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311235852.4A CN116992322B (en) | 2023-09-25 | 2023-09-25 | Smart city data center management system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116992322A CN116992322A (en) | 2023-11-03 |
| CN116992322B true CN116992322B (en) | 2024-01-16 |
Family
ID=88532364
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311235852.4A Active CN116992322B (en) | 2023-09-25 | 2023-09-25 | Smart city data center management system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116992322B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117250133B (en) * | 2023-11-16 | 2024-02-20 | 国建大数据科技(辽宁)有限公司 | Smart city large-scale data acquisition method and system |
| CN117435874B (en) * | 2023-12-21 | 2024-03-12 | 河北雄安睿天科技有限公司 | Abnormal data detection method and system for water supply and drainage equipment |
| CN117608499B (en) * | 2024-01-23 | 2024-04-05 | 山东华夏高科信息股份有限公司 | Intelligent traffic data optimal storage method based on Internet of things |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110610254A (en) * | 2019-07-29 | 2019-12-24 | 中国南方电网有限责任公司超高压输电公司广州局 | Multi-dimensional monitoring system and prediction evaluation method for pollution degree of equipment surface |
| CN110619263A (en) * | 2019-06-12 | 2019-12-27 | 河海大学 | Hyperspectral remote sensing image anomaly detection method based on low-rank joint collaborative representation |
| CN111913081A (en) * | 2020-07-14 | 2020-11-10 | 上海电力大学 | Mean shift clustering-based abnormal detection method for insulation state of switch cabinet |
| CN112713652A (en) * | 2020-12-24 | 2021-04-27 | 南京岁卞智能设备有限公司 | Intelligent power distribution room comprehensive monitoring management system based on cloud computing platform |
| JP2022082277A (en) * | 2020-11-20 | 2022-06-01 | 富士通株式会社 | Detection program, detection device, and detection method |
| CN115861011A (en) * | 2023-02-15 | 2023-03-28 | 山东优嘉环境科技有限公司 | Smart city optimization management method and system based on multi-source data fusion |
| CN116186634A (en) * | 2023-04-26 | 2023-05-30 | 青岛新航农高科产业发展有限公司 | Intelligent management system for construction data of building engineering |
-
2023
- 2023-09-25 CN CN202311235852.4A patent/CN116992322B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619263A (en) * | 2019-06-12 | 2019-12-27 | 河海大学 | Hyperspectral remote sensing image anomaly detection method based on low-rank joint collaborative representation |
| CN110610254A (en) * | 2019-07-29 | 2019-12-24 | 中国南方电网有限责任公司超高压输电公司广州局 | Multi-dimensional monitoring system and prediction evaluation method for pollution degree of equipment surface |
| CN111913081A (en) * | 2020-07-14 | 2020-11-10 | 上海电力大学 | Mean shift clustering-based abnormal detection method for insulation state of switch cabinet |
| JP2022082277A (en) * | 2020-11-20 | 2022-06-01 | 富士通株式会社 | Detection program, detection device, and detection method |
| CN112713652A (en) * | 2020-12-24 | 2021-04-27 | 南京岁卞智能设备有限公司 | Intelligent power distribution room comprehensive monitoring management system based on cloud computing platform |
| CN115861011A (en) * | 2023-02-15 | 2023-03-28 | 山东优嘉环境科技有限公司 | Smart city optimization management method and system based on multi-source data fusion |
| CN116186634A (en) * | 2023-04-26 | 2023-05-30 | 青岛新航农高科产业发展有限公司 | Intelligent management system for construction data of building engineering |
Non-Patent Citations (2)
| Title |
|---|
| "A Novel Mean-Shift Algorithm for Data Clustering";Claude Cariou et al;《IEEE Access》(第10期);第14575-14585页 * |
| "Hybridization of Mean Shift Clustering and Deep Packet Inspected Classification for Network Traffic Analysis";Sathish A.P. Kumar et al;《Wireless Personal Communications》;第217-233页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116992322A (en) | 2023-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116992322B (en) | Smart city data center management system | |
| Wang et al. | A novel weighted sparse representation classification strategy based on dictionary learning for rotating machinery | |
| WO2022110557A1 (en) | Method and device for diagnosing user-transformer relationship anomaly in transformer area | |
| CN107092582B (en) | Online abnormal value detection and confidence evaluation method based on residual posterior | |
| CN111723925B (en) | Fault diagnosis method, device, equipment and medium for on-road intelligent train air conditioning unit | |
| CN117093879A (en) | Intelligent operation management method and system for data center | |
| CN117034043B (en) | Intelligent building comprehensive energy consumption monitoring method and system based on multi-energy Internet of things | |
| CN116320042B (en) | Internet of things terminal monitoring control system for edge calculation | |
| CN116258281B (en) | Internet of things fire control monitoring and regulating system based on cloud platform management | |
| CN111156657A (en) | Air conditioner frosting state determining method and device | |
| CN115510302A (en) | Intelligent factory data classification method based on big data statistics | |
| CN115933787A (en) | Indoor multi-terminal intelligent control system based on indoor environment monitoring | |
| CN116308963B (en) | Government affair data analysis method and system | |
| CN116307944A (en) | Distribution box remote monitoring system based on artificial intelligence and Internet of things | |
| CN116821809A (en) | Vital sign data acquisition system based on artificial intelligence | |
| CN117216484A (en) | Environmental data monitoring method based on multidimensional data analysis | |
| CN110702438A (en) | Centrifugal water chilling unit and fresh air system combined fault diagnosis method | |
| CN112413814B (en) | Online renewable heating ventilation air conditioner sensor fault detection method based on comprehensive distance | |
| CN115526275A (en) | Remote multi-source data monitoring and analyzing method and system for electricity safety | |
| CN115359857A (en) | Method for predicting PM2.5 concentration based on CART model and optimized RVFLNN model | |
| CN110781538B (en) | Windowing simulation algorithm based on field monitoring | |
| CN116757337B (en) | House construction progress prediction system based on artificial intelligence | |
| CN117289778B (en) | Real-time monitoring method for health state of industrial control host power supply | |
| CN117251749A (en) | Data processing method of Internet of things based on incremental analysis | |
| CN117807551A (en) | Heart rate abnormality capturing method and system based on intelligent ring |
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 |