CN116955117A - Computer radiator performance analysis system based on data visualization enhancement - Google Patents
Computer radiator performance analysis system based on data visualization enhancement Download PDFInfo
- Publication number
- CN116955117A CN116955117A CN202311197159.2A CN202311197159A CN116955117A CN 116955117 A CN116955117 A CN 116955117A CN 202311197159 A CN202311197159 A CN 202311197159A CN 116955117 A CN116955117 A CN 116955117A
- Authority
- CN
- China
- Prior art keywords
- data
- heat dissipation
- time
- sequence
- acquiring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 50
- 238000013079 data visualisation Methods 0.000 title claims abstract description 25
- 230000017525 heat dissipation Effects 0.000 claims abstract description 176
- 238000000034 method Methods 0.000 claims abstract description 66
- 230000008859 change Effects 0.000 claims abstract description 43
- 238000012216 screening Methods 0.000 claims abstract description 35
- 230000008569 process Effects 0.000 claims abstract description 32
- 230000005855 radiation Effects 0.000 claims abstract description 32
- 239000011159 matrix material Substances 0.000 claims abstract description 24
- 230000000007 visual effect Effects 0.000 claims abstract description 24
- 238000012800 visualization Methods 0.000 claims abstract description 13
- 238000004364 calculation method Methods 0.000 claims description 11
- 230000035772 mutation Effects 0.000 claims description 9
- 238000000513 principal component analysis Methods 0.000 claims description 5
- 238000005259 measurement Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000009877 rendering Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 abstract description 6
- 230000002159 abnormal effect Effects 0.000 abstract description 4
- 238000012545 processing Methods 0.000 abstract description 4
- 238000011156 evaluation Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 5
- 239000000047 product Substances 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 3
- 241001048891 Jatropha curcas Species 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3452—Performance evaluation by statistical analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/3476—Data logging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- 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
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Probability & Statistics with Applications (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
Abstract
The invention relates to the technical field of digital data processing, and provides a computer radiator performance analysis system based on data visualization enhancement, which comprises the following components: acquiring a heat dissipation data sequence at each moment; acquiring information entropy change amount according to the statistical result of the value range of each type of heat dissipation data, and acquiring main data and secondary data according to the information entropy change amount; acquiring data criticality according to the criticality of heat dissipation data in the performance process of the visual radiator; acquiring a clustering result of the heat dissipation data according to the data criticality; acquiring a heat radiation characteristic matrix according to first main components of all types of heat radiation data in each cluster; acquiring a performance association index according to association degrees of different performance states of the radiator at each moment; and acquiring screening data for visualization according to the performance association index. According to the invention, through analyzing the data of different states of the radiator, the interaction quality between personnel and the radiating data is improved, and errors caused by evaluation and analysis of abnormal data on the performance of the radiator are avoided.
Description
Technical Field
The invention relates to the technical field of digital data processing, in particular to a computer radiator performance analysis system based on data visualization enhancement.
Background
With the increase of user experience requirements, the demand standards for notebook computers or desktop computers have increased, such as the running speed, heat dissipation performance, light and thin body types of computers. The computer radiator is mainly divided into a water-cooling radiator and an air-cooling radiator, the better the performance of the radiator is, the shorter the radiating time of a large amount of heat energy generated by the processor during working is, and the more stable the working performance of the computer is; the worse the performance of the radiator, a great amount of heat energy generated during the operation of the processor may not be timely emitted, so that the processor is reduced in frequency and speed, the aging of the CPU is accelerated, and even the phenomenon of blue screen crash of the computer occurs.
The performance of the computer radiator at the present stage is mainly based on the relevant parameters of the radiator and the monitoring parameters in the using process of the computer for data analysis and visualization, for example, the relevant parameters of the heat radiation such as the rotation speed of the fan, the noise of the fan, the power of the fan, the caliber of the fan, the temperature and the frequency of the CPU, etc., but the obtained relevant parameters of the heat radiation often lack representativeness because the computer radiator has the characteristics of multiple models and various scenes. However, for data visualization, the clear data visualization chart should only contain key data and corresponding visualization results, so as to avoid the attraction of irrelevant factors to the attention of people in the visualization chart, and therefore, the removal of irrelevant contents is a key step in the data visualization process. In order to achieve accurate analysis of the performance of a computer heat sink, it is desirable to make the visualization of heat dissipation related parameters highly efficient and attractive.
Disclosure of Invention
The invention provides a computer radiator performance analysis system based on data visualization enhancement, which aims to solve the problem of low interaction quality between an analyst and radiating data, and adopts the following technical scheme:
an embodiment of the invention is based on a computer radiator performance analysis system with data visualization enhancement, and the method comprises the following steps:
the data acquisition module acquires a heat dissipation data sequence at each moment in the test process;
the data classification module is used for acquiring information entropy change according to the statistical result of the value range of each type of heat dissipation data; acquiring main data and secondary data according to the information entropy change amount; acquiring data criticality according to the criticality of heat dissipation data in the performance process of the visual radiator; acquiring data distances between acquisition moments according to the data criticality, and acquiring a clustering result of heat dissipation data by using a K-means algorithm by taking the data distances as measurement distances between the acquisition moments during clustering;
the data screening module is used for respectively acquiring first principal components of each type of heat dissipation data in each cluster by using a principal component analysis PCA algorithm, and taking a matrix formed by the first principal components of all types of heat dissipation data in each cluster as a data characteristic matrix of each cluster; acquiring a performance association index according to a data feature matrix corresponding to different performance states of the radiator and a heat radiation data sequence at each moment; acquiring screening data for visualization according to the performance association index;
and the performance analysis module is used for acquiring a visual result of the heat dissipation data according to the screening data of each cluster, and acquiring a performance analysis result of the radiator according to the visual result.
Preferably, the method for obtaining the information entropy change according to the statistical result of the value range of each type of heat dissipation data comprises the following steps:
for each type of heat dissipation data, counting the value ranges of each type of heat dissipation data at all acquisition moments, and taking each unequal acquisition data as one data level in each type of heat dissipation data;
for any one data level, taking the information entropy of heat dissipation data of the data level as a first calculation factor, taking the information entropy of the residual heat dissipation data after deleting the data level as a second calculation factor, and taking the difference value between the first calculation factor and the second calculation factor as the information entropy change amount of the data level.
Preferably, the method for acquiring the primary data and the secondary data according to the information entropy change amount comprises the following steps:
for any one acquisition time, taking a sequence formed by the information entropy change amounts corresponding to the elements in the heat dissipation data sequence at the acquisition time as an information amount sequence at each acquisition time according to the information entropy change amounts corresponding to each element in the heat dissipation data sequence at the acquisition time;
taking the average value of elements in the information quantity sequence as entropy change average value, taking the element corresponding data which is greater than or equal to the entropy change average value in the information quantity sequence as main data of the acquisition time, and taking the element corresponding data which is smaller than the entropy change average value in the information quantity sequence as secondary data of the acquisition time.
Preferably, the method for acquiring the data criticality according to the criticality of the heat dissipation data in the process of visualizing the performance of the heat sink comprises the following steps:
for any one acquisition time, acquiring the heat dissipation stability of the acquisition time according to the heat dissipation data sequences of all times in the neighbor sequence corresponding to the acquisition time and the heat dissipation data sequences of the abrupt change time;
for any type of heat dissipation data in the heat dissipation data sequence at the acquisition time, respectively acquiring main data and secondary data in the heat dissipation data sequence at the acquisition time, and acquiring the time sequence information criticality of each type of heat dissipation data at the acquisition time according to the difference between the entropy weight of each type of heat dissipation data at the acquisition time and the entropy weight of the main data and the entropy weight of the secondary data;
taking a natural constant as a base number, taking a calculation result taking the heat radiation stability at the acquisition moment as an index as a first product factor, taking the time sequence information criticality of each type of heat radiation data at the acquisition moment as a second product factor, and taking the product of the first product factor and the second product factor as the data criticality of each type of heat radiation data at the acquisition moment.
Preferably, the method for obtaining the heat dissipation stability of the acquisition time according to the heat dissipation data sequence of all the times in the neighbor sequence corresponding to the acquisition time and the heat dissipation data sequence of the abrupt change time comprises the following steps:
for any one acquisition time, acquiring all mutation times in all acquisition times by utilizing a BG sequence segmentation algorithm, and taking the accumulated sum of pearson correlation coefficients between the acquisition time heat dissipation data sequence and all mutation time heat dissipation data sequences as a molecule;
taking the starting time of a computer as the starting time, respectively acquiring the time interval between each acquisition time and the acquisition time from the starting time to the acquisition time, taking a sequence formed by a preset number of acquisition times in a descending sequence arrangement result of the time interval as a neighbor sequence of the acquisition time, and taking the accumulated sum of the pearson correlation coefficient between the acquisition time heat dissipation data sequence and each acquisition time heat dissipation data sequence in the neighbor sequence as a denominator;
and taking the ratio of the numerator to the denominator as the heat radiation stability at the acquisition time.
Preferably, the method for obtaining the time sequence information criticality of each type of heat dissipation data at the collection time according to the difference between the entropy weight of each type of heat dissipation data at the collection time and the entropy weight of the main data and the entropy weight of the secondary data comprises the following steps:
in the method, in the process of the invention,is the time sequence information criticality of the heat dissipation data of class i at time a,/->N is the data amount of the main data and the data amount of the secondary data at the time a respectively, p is the p-th secondary data in the heat dissipation data sequence at the time a, j is the j-th main data in the heat dissipation data sequence at the time a,/>、/>、/>The entropy weight values corresponding to the i-type heat dissipation data, the main data j and the secondary data p are respectively +.>、/>、/>I-type heat dissipation dataInformation entropy change of data level corresponding to primary data j and secondary data p, +.>Is a parameter adjusting factor.
Preferably, the method for acquiring the data distance between the acquisition moments according to the data criticality comprises the following steps:
in the method, in the process of the invention,is the fitting degree of the heat radiation state of the time a and the time b, < >>、/>The main data in the a time data sequence and the b time data sequence are respectively gathered, and the +.>、/>The data sequence at time a and the number of data in the main data set are respectively +.>Is the number of data in the intersection between the sets, +.>Is a data information threshold, < >>Is a similarity threshold;
is the data distance between time a and time b, < >>Is the total number of categories of heat dissipation data in the heat dissipation data sequence at each moment,/for each moment>、/>The data criticality of the heat dissipation data of the type i at the time a and the time b respectively.
Preferably, the method for obtaining the performance association index according to the data feature matrix corresponding to the different performance states of the radiator by the heat dissipation data sequence at each moment includes:
in the method, in the process of the invention,is the information difference between the time a and the kth cluster, < >>Is a heat radiation data sequence at the moment a, M is a characteristic matrix corresponding to the kth cluster +.>Line number of->Is a feature matrix->A sequence of f-th line elements in (c),is the sequence->、/>DTW distance between>Is a data effective index of heat dissipation data at time a, < >>Is the variance of the information difference between time a and K clusters, < >>Is a parameter adjusting factor;
is a performance related index of the heat dissipation data sequence at time a, < >>The heat dissipation stability at time a.
Preferably, the method for acquiring the screening data for visualization according to the performance association index comprises the following steps:
for any cluster, calculating the performance association degree of each moment in the cluster, respectively obtaining the minimum value of the performance association degree in each cluster, and taking the average value of the minimum values of the performance association degrees in all clusters as a screening threshold value;
and respectively acquiring the difference value between the performance association degree of each acquisition time in the cluster and the screening threshold value, and taking the heat dissipation data sequence of which the performance association degree is larger than zero and corresponds to the acquisition time as screening data.
Preferably, the method for obtaining the visual result of the heat dissipation data according to the screening data of each cluster includes:
and acquiring corresponding screening data according to the heat radiation data request of the analyst, acquiring analysis and rendering results corresponding to the screening data, acquiring a visual page according to HTML, CSS, javaScript by combining with the jQuery library, and acquiring a visual chart in the page by utilizing a visual plug-in.
The beneficial effects of the invention are as follows: according to the method, the data association degree is built according to the information quantity and the time sequence characteristics of different heat dissipation data, and the data criticality considers the primary and secondary relations of different heat dissipation data at each moment. And secondly, acquiring a feature matrix corresponding to each grade of performance through PCA principal component analysis, and constructing a performance association index based on the data characteristics of the data sequence and the feature matrix at each moment, wherein the performance association index considers the correlation degree of the data sequence at each moment and the feature matrix in different states.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a flow chart of a computer heat sink performance analysis system based on data visualization enhancement according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an organization structure of a computer heat sink performance analysis system based on data visualization enhancement according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flowchart of a computer heat sink performance analysis system based on data visualization enhancement according to an embodiment of the present invention is shown, the system includes: the system comprises a data acquisition module, a data classification module, a data screening module and a performance analysis module.
The invention uses a computer radiator on a desktop computer as a performance analysis object, and in order to accurately analyze the performance of the radiator, the data sensor or computer parameter monitoring software is used for collecting attribute data related to the performance of the radiator, wherein the attribute data comprise fan rotating speed, fan noise, fan power, fan caliber, CPU temperature and frequency, and the data sensor comprises an audio sensor, a temperature sensor and a rotating speed sensor.
At the starting time of computer start, 20 cycles of data are collected, the time interval between two adjacent times of data collection is 5s, the collected data amount of each related data in each cycle is N, and the size of N is checked to be 200. After data acquisition is completed, transmitting acquired data to a data processing module in a computer radiator performance analysis system, preprocessing the acquired data received by the data processing module by using a k-nearest neighbor filling method, wherein the k-nearest neighbor filling is a known technology, the specific process is not repeated, the preprocessed data is used as heat dissipation data of the computer radiator, a sequence formed by all acquired data at each moment is used as a heat dissipation data sequence at each moment, and the heat dissipation data sequence at the moment a is recorded as a heat dissipation data sequence at the moment a。
So far, the heat dissipation data sequence of each acquisition time is obtained.
For performance analysis of a computer radiator, the data classification module directly uploads all relevant heat dissipation data to a display system for visual analysis is generally limited by perception caused by high-dimensional multi-data characteristics, which leads to recognition deviation of the visual system. Data visualization while enhancing the performance analyst's knowledge of the radiator related data, the quality of interaction between the performance analyst and the radiator related data determines the quality of the visualization results. Therefore, the invention considers that the interaction quality between personnel and heat dissipation data is improved while the data complexity is reduced, and the enhancement of the visual effect of the heat dissipation data is realized.
According to the collected heat radiation data, an entropy weight method is utilized to obtain the entropy weight of each type of heat radiation data, the larger the entropy weight is, the larger the information content contained in the corresponding heat radiation related data is, the larger the change degree of the heat radiation data is, and the entropy weight of the ith type of heat radiation data is recorded asThe entropy weighting method is a well-known technique, and the specific process is not repeated. For a computer radiator, the change of heat dissipation data reflects that the computer is in different working states, and correspondingly, the heat dissipation performance of the radiator is different. In addition, on the premise of different performances of the radiator, the distribution condition of each type of heat dissipation data is different, and the degree of correlation with the performance of the radiator is also different, for example, if the temperature of the CPU is increased, the heat dissipation performance is not necessarily lower, and the heat dissipation performance is not necessarily caused, and the heat dissipation performance is also possibly caused by factors such as internal dust, memory bank faults and the like.
And counting the value range of each type of heat dissipation data, taking each unequal collected data as a data level, calculating the difference value of the information entropy of the heat dissipation data before and after deleting the data level of the heat dissipation data of any one data level, and if the difference value of the information entropy is smaller, indicating that the information quantity of the data corresponding to the data level is smaller and the association degree of the state of the heat sink is weaker. Further, each of the following is acquired separatelyThe information entropy change amount corresponding to each data level of the heat dissipation data is recorded as the information entropy change amount corresponding to the c data level in the i-th heat dissipation data. Secondly, acquiring mutation points in each type of heat dissipation data at all acquisition moments by utilizing a BG sequence segmentation algorithm, wherein the moment corresponding to the mutation points in all types of heat dissipation data is taken as the mutation moment, and the BG sequence segmentation algorithm is a known technology and a specific process is not repeated.
For any one acquisition time, according to the value of each type of heat dissipation data in the corresponding heat dissipation data sequence, respectively acquiring the data level and the information entropy change amount corresponding to each type of heat dissipation data, and taking a sequence consisting of the information entropy change amounts corresponding to elements in the data sequence at each time as an information amount sequence H at each acquisition time.
Secondly, dividing the data in the acquired data sequence at each moment into main data and secondary data according to the strength of the association degree between the elements in the acquired data sequence at each moment and the radiating state of the radiator corresponding to each moment. The dividing process is as follows: for any acquisition time, acquiring an information quantity sequence H corresponding to the time, marking the average value of elements in the information quantity sequence as entropy change average value, taking element corresponding data which is larger than or equal to the entropy change average value in the information quantity sequence H as main data of the time, and taking element corresponding data which is smaller than the entropy change average value in the information quantity sequence H as secondary data of the time.
Based on the analysis, a data criticality V is constructed here for representing the criticality of different heat dissipation data in each moment data sequence in the performance process of the visual heat sink, and the data criticality of the heat dissipation data of class i at the moment a is calculated：
In the method, in the process of the invention,is the heat radiation stability at time a, +.>Is the number of mutation moments in all acquired data, t is the t-th mutation moment in all acquired data,/-in->Is the number of acquisition moments in a neighbor sequence, b is the b-th moment in a neighbor sequence of a moment, wherein the neighbor sequence refers to +.>Sequences of individual acquisition moments, < >>The magnitude of (2) takes the empirical value of 7./>、/>、/>The heat dissipation data sequences are respectively the a time, the t time and the b time, and the +.>、/>Respectively the sequences->、/>Between, sequence->、/>Pearson correlation coefficient therebetween. />The larger the value of (a), the more likely the time a is near the transition time of different performance of the heat sink.
Is the time sequence information criticality of the heat dissipation data of class i at time a,/->N is the data amount of the main data and the data amount of the secondary data at the time a respectively, p is the p-th secondary data in the heat dissipation data sequence at the time a, j is the j-th main data in the heat dissipation data sequence at the time a,/>、/>、/>The entropy weight values corresponding to the i-type heat dissipation data, the main data j and the secondary data p are respectively +.>、/>、/>Respectively i-type heat dissipation data, main data j and secondary data pInformation entropy change of data level, +.>Is a parameter regulating factor, and is a herb of Jatropha curcas>The function of (2) is to prevent the denominator from being 0, < >>The size of (2) is 0.001./>The larger the value of the (a) is, the larger the information amount of the heat dissipation data of the type i is, and the higher the key degree of the heat dissipation data of the type i is.
The data criticality reflects the degree of association of different heat dissipation data and the state of the heat sink in each time data sequence. The more likely the time a is near the conversion time of different performances of the radiator, the more approximate the sizes of various heat dissipation data corresponding to the abrupt change time and various heat dissipation data in the time a data sequence,the larger the value of (a) is, the closer the time a is to the switching time of different performances of the radiator, the larger the fluctuation degree of heat dissipation data in the adjacent time data sequences is, and the +.>The smaller the value of +.>The greater the value of (2); the larger the information quantity of the i-type heat dissipation data at the time a, the larger the entropy weight of the i-type heat dissipation data, the larger the information entropy change quantity, and the more likely the information entropy change quantity is the main data in the time a data sequence, and the larger the difference between the i-type heat dissipation data at the time a and the secondary data in the time a data sequence is, the more the difference is, the more the information entropy change quantity is the main data in the time a data sequence, and the information entropy change quantity is the main data in the time a data sequence>The larger the value of (a) is, the smaller the difference between the heat dissipation data of the type i at the time a and the main data in the data sequence at the time a is, the +.>The smaller the value of +.>The greater the value of (2); i.e. < ->The larger the value of the (i) type heat radiation data is, the larger the relevance between the i type heat radiation data and the performance of the radiator at the time a is, and the more likely the i type heat radiation data is the key data for the subsequent visual analysis of the performance of the radiator. The data criticality performance association degree considers the primary and secondary relations of different heat dissipation data at each moment, and has the advantages of improving the calculation accuracy of the measurement distance in the subsequent data clustering process, and being beneficial to improving the interaction quality between personnel and heat dissipation data.
Further, according to the data distance in the clustering process obtained by the data criticality, the data distance between the time a and the time b is calculated：
In the method, in the process of the invention,is the fitting degree of the heat radiation state of the time a and the time b, < >>、/>The main data in the a time data sequence and the b time data sequence are respectively gathered, and the +.>、/>The data sequence at time a and the number of data in the main data set are respectively +.>Is the number of data in the intersection between the sets. />Is a data information threshold, < >>Is a similar threshold value->、/>The empirical values of 0.6 and 0.3 are respectively adopted. />The larger the value of a, the more similar the radiator states corresponding to the time a and the time b are, the more likely the radiator data belong to the same performance.
Is the data distance between time a and time b, < >>Is the total number of categories of heat dissipation data in the heat dissipation data sequence at each moment,/for each moment>、/>The data criticality of the heat dissipation data of the type i at the time a and the time b respectively.
Furthermore, according to the working state of the computer in the using process, the performance of the radiator is divided into four different performance grades, namely low, medium, good and high, and the performance is gradually improved from low to high. And clustering the acquired data at all moments by using a K-means clustering algorithm, wherein the magnitude of K takes an empirical value of 4, the data distance is used as a measurement distance in the clustering process, and the heat dissipation data in each cluster is used as heat dissipation data under a performance level. The K-means clustering algorithm is a well-known technique, and the specific process is not described in detail.
So far, the clustering result of the heat dissipation data is obtained.
And the data screening module is used for further analyzing the mutual influence relation among different types of heat dissipation data after the heat dissipation data are clustered, so that if the heat dissipation performance is displayed only through the visual result of the data size, the influence among different types of data of the heat dissipation is ignored. For example, when other types of heat dissipation data are unchanged, the faster the rotation speed of the fan of the radiator is, the larger the generated noise is, and the size of the diameter of the radiator also affects the ventilation amount, so that the noise is affected, that is, the rotation speed and the noise of the fan of the radiator with the same performance may be different due to different diameters of the radiator.
And for any type of heat dissipation data in each cluster, a principal component analysis PCA algorithm is utilized to obtain a principal component analysis result of each type of heat dissipation data, and principal components with the accumulated contribution rate of each type of heat dissipation data reaching 90% are reserved. Taking i-type heat dissipation data as an example, respectively marking the 1 st main component reserved by the i-type heat dissipation data in the 1 st cluster asY-type heat dissipation data coexist in the 1 st cluster, the first main component is often the component with the most information of the original data reserved, the PCA main component analysis algorithm is a known technology, and the specific process is not repeated. Secondly, respectively acquiring main components of the rest heat dissipation data in the 1 st cluster, taking a matrix formed by first main components of all heat dissipation data in the 1 st cluster as a data characteristic matrix of the radiator under the corresponding performance of the 1 st cluster, wherein if the first main components with unequal lengths appear, the first main component with the longest length is taken as the column number of the data characteristic matrix, and the rest first main components ensure that the quantity of elements in each row in the matrix is equal in a mode of 0 terminal supplement. Further, respectivelyObtaining data characteristic matrixes corresponding to the K clusters, and judging the association degree of each moment in each cluster and different performance heat dissipation data by calculating the difference between the acquired data at each moment in each cluster and the radiator data characteristic matrixes under each performance.
Based on the analysis, a performance association index L is constructed, used for representing the association degree of each moment and different performance states of the radiator, and the performance association index of the heat radiation data sequence at the moment a is calculated：
In the method, in the process of the invention,is the information difference between the time a and the kth cluster, < >>Is a heat radiation data sequence at the moment a, M is a characteristic matrix corresponding to the kth cluster +.>Line number of->Is a feature matrix->A sequence of f-th line elements in (c),is a sequence of/>、/>The DTW distance between the two is a known technology, and the specific process is not described again. />The larger the value of the (a) is, the larger the information difference between the acquired data at the time a and the feature matrix corresponding to the kth cluster is.
Is a data effective index of heat dissipation data at time a, < >>Is the variance of the information difference between time a and K clusters, < >>Is a parameter regulating factor, and is a herb of Jatropha curcas>The function of (2) is to prevent the denominator from being 0, < >>The size of (2) is 0.001./>The greater the value of (a), the greater the effectiveness of collecting data at time a, and the greater the accuracy of analyzing the performance of the radiator.
Is a performance related index of the heat dissipation data sequence at time a, < >>The heat dissipation stability at time a.
The performance-related index reflects the absence of the radiator at each timeThe degree of association of performance states. The larger the difference between the acquired data at the time a and the feature matrix corresponding to the kth cluster is, the larger the information difference between the data contained in the corresponding performance of the kth cluster at the time a is,the greater the value of +.>The greater the value of (2); the higher the effective degree of the data acquisition at the moment a is, the greater the similarity between the moment a and the data corresponding to 1 performance in the 4 performance grades is, the +.>The greater the value of (2), i.e +.>The greater the value of (a), the stronger the degree of correlation between the collected data at the time a and the performance of a certain level of the radiator, and the higher the accuracy of analyzing the performance of the radiator by the collected data at the time a.
Further, calculating the performance association degree of each moment in each cluster, respectively obtaining the minimum value of the performance association degrees of the moment in K clusters, taking the average value of the minimum value of the K performance association degrees as a screening threshold, and deleting the data sequence of the moment, corresponding to the moment, of which the performance association degree in each cluster is smaller than the screening threshold. The basis of the screening is that if the correlation degree of the heat dissipation data sequence at a certain moment and the feature matrix in different states is relatively close, the heat dissipation data at the moment is abnormal data.
The performance association index considers the correlation degree of the data sequence at each moment and the feature matrix under different states, and has the beneficial effects that abnormal data can be screened through the difference of the feature matrix corresponding to the performance of the radiator at each moment and different levels and the stability degree of heat dissipation, so that the influence of the abnormal data on the visual precision is avoided, and the error caused by the evaluation analysis of the performance of the radiator is prevented.
To this end, screening data for visualization is obtained.
And the performance analysis module is used for obtaining screening data for visualization according to the steps, and obtaining a visualization result of the heat dissipation data based on the screening data in each cluster. The computer radiator performance analysis system based on data visualization enhancement in the invention is based on a B/S architecture, and in order to accelerate performance analysis speed, the whole analysis system is divided into a performance analysis module, a data screening module, a data classification module and a data acquisition module, as shown in figure 2.
Furthermore, a performance analysis module in the system directly faces to an analyst, the analyst completes various interaction actions with the performance analysis system through a browser, in the testing process of each performance index, the performance analysis module sends a heat dissipation data request of the analyst, a data screening module receives the heat dissipation data request and sends the heat dissipation data analysis request to a data classification module, the data classification module returns analysis response to the data screening module, after receiving analysis response, the data screening module acquires corresponding screening order data according to the steps, and returns the screening data to the performance analysis module, and the analysis and rendering are completed and presented to a user. The pages in the performance analysis module are mainly realized by combining HTML, CSS, javaScript with the jQuery library, and the visual view of the chart in the pages is realized by means of the Echarts visual plug-in, wherein the use of the jQuery library and the Echarts visual plug-in is a known technology, and the specific process is not repeated.
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 (10)
1. The computer radiator performance analysis system based on data visualization enhancement is characterized by comprising the following modules:
the data acquisition module acquires a heat dissipation data sequence at each moment in the test process;
the data classification module is used for acquiring information entropy change according to the statistical result of the value range of each type of heat dissipation data; acquiring main data and secondary data according to the information entropy change amount; acquiring data criticality according to the criticality of heat dissipation data in the performance process of the visual radiator; acquiring data distances between acquisition moments according to the data criticality, and acquiring a clustering result of heat dissipation data by using a K-means algorithm by taking the data distances as measurement distances between the acquisition moments during clustering;
the data screening module is used for respectively acquiring first principal components of each type of heat dissipation data in each cluster by using a principal component analysis PCA algorithm, and taking a matrix formed by the first principal components of all types of heat dissipation data in each cluster as a data characteristic matrix of each cluster; acquiring a performance association index according to a data feature matrix corresponding to different performance states of the radiator and a heat radiation data sequence at each moment; acquiring screening data for visualization according to the performance association index;
and the performance analysis module is used for acquiring a visual result of the heat dissipation data according to the screening data of each cluster, and acquiring a performance analysis result of the radiator according to the visual result.
2. The computer radiator performance analysis system based on data visualization enhancement according to claim 1, wherein the method for obtaining the information entropy change according to the statistical result of the value range of each type of heat dissipation data is as follows:
for each type of heat dissipation data, counting the value ranges of each type of heat dissipation data at all acquisition moments, and taking each unequal acquisition data as one data level in each type of heat dissipation data;
for any one data level, taking the information entropy of heat dissipation data of the data level as a first calculation factor, taking the information entropy of the residual heat dissipation data after deleting the data level as a second calculation factor, and taking the difference value between the first calculation factor and the second calculation factor as the information entropy change amount of the data level.
3. The computer heat sink performance analysis system based on data visualization enhancement according to claim 1, wherein the method for obtaining the primary data and the secondary data according to the information entropy change amount is as follows:
for any one acquisition time, taking a sequence formed by the information entropy change amounts corresponding to the elements in the heat dissipation data sequence at the acquisition time as an information amount sequence at each acquisition time according to the information entropy change amounts corresponding to each element in the heat dissipation data sequence at the acquisition time;
taking the average value of elements in the information quantity sequence as entropy change average value, taking the element corresponding data which is greater than or equal to the entropy change average value in the information quantity sequence as main data of the acquisition time, and taking the element corresponding data which is smaller than the entropy change average value in the information quantity sequence as secondary data of the acquisition time.
4. The computer heat sink performance analysis system based on data visualization enhancement according to claim 1, wherein the method for obtaining the data criticality according to the criticality of heat dissipation data in the process of visualizing heat sink performance is as follows:
for any one acquisition time, acquiring the heat dissipation stability of the acquisition time according to the heat dissipation data sequences of all times in the neighbor sequence corresponding to the acquisition time and the heat dissipation data sequences of the abrupt change time;
for any type of heat dissipation data in the heat dissipation data sequence at the acquisition time, respectively acquiring main data and secondary data in the heat dissipation data sequence at the acquisition time, and acquiring the time sequence information criticality of each type of heat dissipation data at the acquisition time according to the difference between the entropy weight of each type of heat dissipation data at the acquisition time and the entropy weight of the main data and the entropy weight of the secondary data;
taking a natural constant as a base number, taking a calculation result taking the heat radiation stability at the acquisition moment as an index as a first product factor, taking the time sequence information criticality of each type of heat radiation data at the acquisition moment as a second product factor, and taking the product of the first product factor and the second product factor as the data criticality of each type of heat radiation data at the acquisition moment.
5. The computer radiator performance analysis system based on data visualization enhancement according to claim 4, wherein the method for obtaining the heat dissipation stability of the acquisition time according to the heat dissipation data sequence of all times in the neighbor sequence corresponding to the acquisition time and the heat dissipation data sequence of the abrupt change time is as follows:
for any one acquisition time, acquiring all mutation times in all acquisition times by utilizing a BG sequence segmentation algorithm, and taking the accumulated sum of pearson correlation coefficients between the acquisition time heat dissipation data sequence and all mutation time heat dissipation data sequences as a molecule;
taking the starting time of a computer as the starting time, respectively acquiring the time interval between each acquisition time and the acquisition time from the starting time to the acquisition time, taking a sequence formed by a preset number of acquisition times in a descending sequence arrangement result of the time interval as a neighbor sequence of the acquisition time, and taking the accumulated sum of the pearson correlation coefficient between the acquisition time heat dissipation data sequence and each acquisition time heat dissipation data sequence in the neighbor sequence as a denominator;
and taking the ratio of the numerator to the denominator as the heat radiation stability at the acquisition time.
6. The computer radiator performance analysis system based on data visualization enhancement according to claim 4, wherein the method for obtaining the time sequence information criticality of each type of heat dissipation data at the collection time according to the difference between the entropy weight of each type of heat dissipation data at the collection time and the entropy weight of the main data and the entropy weight of the secondary data is as follows:
in the method, in the process of the invention,is the time sequence information criticality of the heat dissipation data of class i at time a,/->N is the data amount of the main data and the data amount of the secondary data at the time a, p is the p secondary data in the heat dissipation data sequence at the time a, and j is the j primary data in the heat dissipation data sequence at the time aData on demand->、/>、/>The entropy weight values corresponding to the i-type heat dissipation data, the main data j and the secondary data p are respectively +.>、/>、/>The information entropy change amounts of the i-type heat dissipation data, the main data j and the secondary data p corresponding to the data level are respectively +.>Is a parameter adjusting factor.
7. The computer radiator performance analysis system based on data visualization enhancement according to claim 1, wherein the method for acquiring the data distance between the acquisition moments according to the data criticality is as follows:
in the method, in the process of the invention,is the fitting degree of the heat radiation state of the time a and the time b, < >>、/>The main data in the a time data sequence and the b time data sequence are respectively gathered, and the +.>、/>The number of data in the time data sequence and the main data set are respectively a,is the number of data in the intersection between the sets, +.>Is a data information threshold, < >>Is a similarity threshold;
is the data distance between time a and time b, < >>Is the total number of categories of heat dissipation data in the heat dissipation data sequence at each moment,/for each moment>、/>The data criticality of the heat dissipation data of the type i at the time a and the time b respectively.
8. The computer radiator performance analysis system based on data visualization enhancement according to claim 1, wherein the method for obtaining the performance association index according to the data feature matrix corresponding to the different performance states of the radiator according to the heat dissipation data sequence at each moment is as follows:
in the method, in the process of the invention,is the information difference between the time a and the kth cluster, < >>Is a heat radiation data sequence at the moment a, M is a characteristic matrix corresponding to the kth cluster +.>Line number of->Is a feature matrix->A sequence of f-th line elements in (c),is the sequence->、/>DTW distance between>Is a data effective index of heat dissipation data at time a, < >>Is the variance of the information difference between time a and K clusters, < >>Is a parameter adjusting factor;
is a performance related index of the heat dissipation data sequence at time a, < >>The heat dissipation stability at time a.
9. The enhanced computer heat sink performance analysis system based on data visualization of claim 1, wherein the method for obtaining screening data for visualization according to the performance association index is as follows:
for any cluster, calculating the performance association degree of each moment in the cluster, respectively obtaining the minimum value of the performance association degree in each cluster, and taking the average value of the minimum values of the performance association degrees in all clusters as a screening threshold value;
and respectively acquiring the difference value between the performance association degree of each acquisition time in the cluster and the screening threshold value, and taking the heat dissipation data sequence of which the performance association degree is larger than zero and corresponds to the acquisition time as screening data.
10. The computer radiator performance analysis system based on data visualization enhancement according to claim 1, wherein the method for obtaining the visualization result of the heat dissipation data according to the screening data of each cluster is as follows:
and acquiring corresponding screening data according to the heat radiation data request of the analyst, acquiring analysis and rendering results corresponding to the screening data, acquiring a visual page according to HTML, CSS, javaScript by combining with the jQuery library, and acquiring a visual chart in the page by utilizing a visual plug-in.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311197159.2A CN116955117B (en) | 2023-09-18 | 2023-09-18 | Computer radiator performance analysis system based on data visualization enhancement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311197159.2A CN116955117B (en) | 2023-09-18 | 2023-09-18 | Computer radiator performance analysis system based on data visualization enhancement |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116955117A true CN116955117A (en) | 2023-10-27 |
| CN116955117B CN116955117B (en) | 2023-12-22 |
Family
ID=88456789
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311197159.2A Active CN116955117B (en) | 2023-09-18 | 2023-09-18 | Computer radiator performance analysis system based on data visualization enhancement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116955117B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050141782A1 (en) * | 2003-12-31 | 2005-06-30 | Honeywell International Inc. | Principal component analysis based fault classification |
| US20170371757A1 (en) * | 2016-06-28 | 2017-12-28 | Beijing Baidu Netcom Science And Technology, Ltd. | System monitoring method and apparatus |
| CN110110000A (en) * | 2018-01-24 | 2019-08-09 | 阿里巴巴集团控股有限公司 | Methods of exhibiting and device |
| CN110390358A (en) * | 2019-07-23 | 2019-10-29 | 杨勇 | A kind of deep learning method based on feature clustering |
| CN113191891A (en) * | 2021-05-27 | 2021-07-30 | 中国工商银行股份有限公司 | Data processing method, device and system |
| CN113420691A (en) * | 2021-06-30 | 2021-09-21 | 昆明理工大学 | Mixed domain characteristic bearing fault diagnosis method based on Pearson correlation coefficient |
| CN113778831A (en) * | 2021-09-27 | 2021-12-10 | 广州文远知行科技有限公司 | Data application performance analysis method, device, equipment and medium |
| CN114519086A (en) * | 2022-02-17 | 2022-05-20 | 北京大学深圳研究生院 | Incremental interactive clustering visualization method and system for credible cloud data sharing |
| CN115752582A (en) * | 2022-11-21 | 2023-03-07 | 国网四川省电力公司巴中供电公司 | Visual monitored control system of rack |
-
2023
- 2023-09-18 CN CN202311197159.2A patent/CN116955117B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050141782A1 (en) * | 2003-12-31 | 2005-06-30 | Honeywell International Inc. | Principal component analysis based fault classification |
| US20170371757A1 (en) * | 2016-06-28 | 2017-12-28 | Beijing Baidu Netcom Science And Technology, Ltd. | System monitoring method and apparatus |
| CN110110000A (en) * | 2018-01-24 | 2019-08-09 | 阿里巴巴集团控股有限公司 | Methods of exhibiting and device |
| CN110390358A (en) * | 2019-07-23 | 2019-10-29 | 杨勇 | A kind of deep learning method based on feature clustering |
| CN113191891A (en) * | 2021-05-27 | 2021-07-30 | 中国工商银行股份有限公司 | Data processing method, device and system |
| CN113420691A (en) * | 2021-06-30 | 2021-09-21 | 昆明理工大学 | Mixed domain characteristic bearing fault diagnosis method based on Pearson correlation coefficient |
| CN113778831A (en) * | 2021-09-27 | 2021-12-10 | 广州文远知行科技有限公司 | Data application performance analysis method, device, equipment and medium |
| CN114519086A (en) * | 2022-02-17 | 2022-05-20 | 北京大学深圳研究生院 | Incremental interactive clustering visualization method and system for credible cloud data sharing |
| CN115752582A (en) * | 2022-11-21 | 2023-03-07 | 国网四川省电力公司巴中供电公司 | Visual monitored control system of rack |
Non-Patent Citations (1)
| Title |
|---|
| 尹成群等: "基于小波变换和混合神经网络的短期负荷预测", 《电力自动化设备》, pages 40 - 44 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116955117B (en) | 2023-12-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106021362B (en) | Generation, image searching method and the device that the picture feature of query formulation represents | |
| CN103258000B (en) | Method and device for clustering high-frequency keywords in webpages | |
| CN109783879B (en) | Radar radiation source signal identification efficiency evaluation method and system | |
| CN110826618A (en) | Personal credit risk assessment method based on random forest | |
| WO2022052548A1 (en) | Data analysis method and apparatus, electronic device, and storage medium | |
| CN111861781A (en) | Feature optimization method and system in residential electricity consumption behavior clustering | |
| CN114676749A (en) | Power distribution network operation data abnormity judgment method based on data mining | |
| CN112016618A (en) | Measurement method for generalization capability of image semantic segmentation model | |
| CN115617784A (en) | Data processing system and processing method for informationized power distribution | |
| CN117725437B (en) | Machine learning-based data accurate matching analysis method | |
| CN115081515A (en) | Energy efficiency evaluation model construction method and device, terminal and storage medium | |
| CN111445280A (en) | Model generation method, restaurant ranking method, system, device and medium | |
| CN110019763B (en) | Text filtering method, system, equipment and computer readable storage medium | |
| CN113591400A (en) | Power dispatching monitoring data anomaly detection method based on feature correlation partition regression | |
| CN116955117B (en) | Computer radiator performance analysis system based on data visualization enhancement | |
| CN113537278A (en) | Wine detection method and system based on 1D-CNN electronic tongue | |
| CN116668151A (en) | Network intrusion detection method and device based on improved CSA optimization SVM | |
| CN112860531B (en) | Block chain wide consensus performance evaluation method based on deep heterogeneous graph neural network | |
| CN116304295A (en) | User energy consumption portrait analysis method based on multivariate data driving | |
| CN118277723B (en) | Data center-based learning data statistical analysis method and related device | |
| CN117609529B (en) | Substitution search method and system for electronic components | |
| CN113723835B (en) | Water consumption evaluation method and terminal equipment for thermal power plant | |
| CN114647386B (en) | Big data distributed storage method based on artificial intelligence | |
| CN113837632A (en) | Engineering machinery product competitiveness quantitative analysis method and system and storage medium | |
| Huiskes | Aspect-based relevance learning for image retrieval |
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 |