An intelligent digital park operation and maintenance data storage method and system

Abstract

The application relates to the technical field of big data storage, in particular to an intelligent digital park operation and maintenance data storage method and system, which comprises the following steps: acquiring data of various operation and maintenance parameters corresponding to each business category at each time in each time period in the digital park; determining redundancy dependence, noise interference and effective contribution of the various operation and maintenance parameters in each time period; extracting a score vector corresponding to each principal component; clustering elements of all time points in the score vector, calculating information ambiguity and information irrelevance of each principal component, determining information invalidity and discriminant coefficient of the various operation and maintenance parameters in each time period, and screening and storing all kinds of operation and maintenance parameters in each time period. The application improves the storage quality of operation and maintenance data of the intelligent digital park, can guarantee that key information contained in the operation and maintenance data is accurate and relatively complete, and can also save a large amount of storage resources.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of big data storage, in particular to an operation and maintenance data storage method and system for an intelligent digital park. BACKGROUND

[0002] An intelligent digital park realizes real-time collection, analysis and application of data through technologies such as the Internet of Things, big data and cloud computing. The storage of operation and maintenance data as underlying technology support provides strong basic guarantee for park equipment monitoring, resource allocation and security management, and the operation and maintenance data storage can also serve as a core basis for optimizing management decisions.

[0003] In a traditional digital park, there are a large amount of redundant and invalid data in different kinds of operation and maintenance data, and different kinds of operation and maintenance data are greatly affected by noise, resulting in a large amount of storage resources occupied when storing the operation and maintenance data of the digital park, low storage efficiency and low data quality of the stored operation and maintenance data, increased risk of decision-making, and thus affecting the reliability of decision-making. SUMMARY

[0004] In order to solve the above technical problems, an operation and maintenance data storage method and system for an intelligent digital park are provided to solve the existing problems.

[0005] The technical problem of the application is solved by providing an operation and maintenance data storage method and system for an intelligent digital park, comprising the following steps:

[0006] In the first aspect, the application provides an operation and maintenance data storage method for an intelligent digital park, which comprises the following steps:

[0007] The redundancy dependency of various operation and maintenance parameters in each time period is determined according to the discrete condition of all data of various operation and maintenance parameters in each business category in each time period and the correlation condition of data between different kinds of operation and maintenance parameters;

[0008] The distribution characteristics of the same data of various operation and maintenance parameters in each time period are analyzed, the noise interference degree of various operation and maintenance parameters in each time period is calculated, and the effective contribution degree of various operation and maintenance parameters in each time period is determined in combination with the redundancy dependency;

[0009] Based on the correlation condition of data between various operation and maintenance parameters and the remaining operation and maintenance parameters in each business category in each time period, the score vectors corresponding to each principal component are extracted through dimension reduction analysis;

[0010] clustering elements of all time points in the score vector, analyzing dispersion between and within all cluster clusters of each principal component, calculating information fuzziness of each principal component; calculating information irrelevance of each principal component through correlation between all cluster centers corresponding to time points of each principal component and other principal components, and difference of the score vector, and determining information invalidity of various operation and maintenance parameters in each period in combination with the information fuzziness;

[0011] Based on the effective contribution degree and the information invalidity, determine the discriminant coefficient of various operation and maintenance parameters in each period, filter and store all kinds of operation and maintenance parameters in each period.

[0012] Preferably, the determination of the redundancy dependence of various operation and maintenance parameters comprises:

[0013] Calculate the correlation degree of all data between the any operation and maintenance parameter and other various operation and maintenance parameters in each business category in each period;

[0014] The sum of the absolute values of the correlation degrees between the any operation and maintenance parameter and all other operation and maintenance parameters is denoted as similarity;

[0015] Calculate the dispersion degree of all data of the any operation and maintenance parameter in each period;

[0016] The redundancy dependence is the ratio of the similarity and the dispersion degree.

[0017] Preferably, the calculation of the noise interference degree of various operation and maintenance parameters comprises:

[0018] Statistically count the number of continuous occurrence of the same data of various operation and maintenance parameters in each period, and obtain the maximum number;

[0019] Using the run test algorithm, calculate the test statistic of all data of various operation and maintenance parameters in each period;

[0020] The noise interference degree is the ratio of the maximum number and the absolute value of the test statistic.

[0021] Preferably, the effective contribution degree is the reciprocal of the product of the redundancy dependence and the noise interference degree.

[0022] Preferably, the extraction of the score vector corresponding to each principal component comprises:

[0023] Arrange the absolute values of the correlation degrees between the any operation and maintenance parameter and all other operation and maintenance parameters in each business category in each period in descending order, select a preset number of operation and maintenance parameters at the front of the arrangement as the associated operation and maintenance parameters corresponding to the any operation and maintenance parameter;

[0024] The dimensionality reduction analysis is performed on all data of all associated operation and maintenance parameters corresponding to the any one operation and maintenance parameter in each service category in each period, and a score vector corresponding to each principal component is extracted.

[0025] Preferably, the information ambiguity of each principal component is calculated, including:

[0026] The mean of the difference between all elements in each cluster and the cluster center thereof is calculated, denoted as an average difference, and the sum of the average difference of all cluster centers corresponding to each principal component is calculated, denoted as a relative deviation of each principal component;

[0027] The information entropy of the cluster center of each cluster corresponding to each principal component is calculated.

[0028] The information ambiguity is the product of the relative deviation and the information entropy.

[0029] Preferably, the information irrelevance of each principal component is calculated, including:

[0030] The sum of the difference between the score vector of each principal component and all other principal components is denoted as a relative difference;

[0031] The sum of the absolute values of the correlation coefficients corresponding to the moment of the cluster center of all clusters between each principal component and all other principal components is calculated, denoted as a correlation degree of each principal component.

[0032] The information irrelevance is the ratio of the relative difference and the correlation degree.

[0033] Preferably, the information invalidity is the sum of the product of the information ambiguity and the information irrelevance of all principal components corresponding to each operation and maintenance parameter.

[0034] Preferably, the determination of the discriminant coefficient of each operation and maintenance parameter in each period includes screening and storing all operation and maintenance parameters in each period, including:

[0035] The ratio of the information invalidity and the effective contribution degree is normalized as the discriminant coefficient of each operation and maintenance parameter, and all data of all operation and maintenance parameters in each period with a discriminant coefficient less than a preset threshold are stored.

[0036] In a second aspect, the embodiments of the present application further provide an intelligent digital park operation and maintenance data storage system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of the intelligent digital park operation and maintenance data storage method in any one of the above aspects when executing the computer program.

[0037] The present application has at least the following beneficial effects:

[0038] The application has the beneficial effects that the data correlation between the various operation and maintenance parameters is considered to reflect the independent storage value of the operation and maintenance parameter; the random characteristics and data repetition of the data change of the various operation and maintenance parameters are considered to reflect the influence degree of the data of the operation and maintenance parameter on the noise interference; the information effectiveness contained in the various operation and maintenance parameters is considered to reflect the data storage value of the operation and maintenance parameter, and the risk of occupying storage resources by a large amount of repeated operation and maintenance data is effectively reduced; the dimensionality reduction analysis is performed on the remaining operation and maintenance parameters with high correlation with the various operation and maintenance parameters, the score vector of each principal component is extracted, the elements of all time points in the score vector are clustered, the difference between the elements in each cluster and the cluster center and the discrete distribution of the cluster centers of all clusters of each principal component are analyzed, the information fuzziness of each principal component is calculated, the clustering structure and distribution discrete condition of the elements in the score vector of each principal component are considered to reflect the discrete degree and uncertainty degree of the information contained in the principal component, and the information representation of the principal component on the operation and maintenance parameter is explained; further, the difference change between the score vectors of each principal component and the remaining principal components and the correlation of the time points of the cluster centers of all clusters are calculated, the information irrelevance of each principal component is calculated, the distribution difference of the elements in the score vectors of each principal component and the remaining principal components and the correlation of the cluster centers in time are considered to reflect the degree of invalid information contained in the principal component; the information invalidity of the various operation and maintenance parameters in each time period is determined, the data pollution of the operation and maintenance parameter and the storage value thereof are reflected; the discriminant coefficients of the various operation and maintenance parameters in each time period are determined, all the operation and maintenance parameters in each time period are screened and stored, the operation and maintenance parameters with high storage value are selected for data storage, the redundant and invalid data information is eliminated, the storage quality of the operation and maintenance data of the intelligent digital park is improved, the storage efficiency is improved, the key information contained in the operation and maintenance data is ensured to be accurate and relatively complete, a large amount of storage resources is saved, and the reliability of decision-making is improved. BRIEF DESCRIPTION OF DRAWINGS

[0039] The intelligent digital park operation and maintenance data storage method of the application will be further described in detail below with reference to the accompanying drawings.

[0040] Figure 1 A step flowchart of the intelligent digital park operation and maintenance data storage method provided by the embodiment of the application is shown in the figure.

[0041] Figure 2 The step flow chart of the method for obtaining the effective contribution degrees of various operation and maintenance parameters in each period provided by the embodiment of the present application is shown in the following.

[0042] Figure 3 The step flow chart of the method for obtaining the effective contribution degrees of various operation and maintenance parameters in each period provided by the embodiment of the present application is shown in the following. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical scheme and advantages of the present application more clear, the intelligent digital park operation and maintenance data storage method and system provided by the present application is further described in detail below in combination with the drawings and implementation examples. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0045] Please refer to Figure 1 which shows the step flow chart of the intelligent digital park operation and maintenance data storage method provided by an embodiment of the present application. The method comprises the following steps:

[0046] Step 1, obtaining the data of each business category corresponding to various operation and maintenance parameters at each time in each period in the digital park.

[0047] The intelligent digital park builds a smart operation and maintenance management system with the management core of park safety, facilities, transportation, efficiency and environment through Internet of Things technology, Internet technology and intelligent methods. The smart operation and maintenance management system contains multiple business categories, and each business category corresponds to the data of multiple operation and maintenance parameters at different times in each period.

[0048] In the present embodiment, the data of each business category corresponding to multiple operation and maintenance parameters is obtained by calling the database API interface of different business modules in the smart operation and maintenance management system. For example, the data of multiple operation and maintenance parameters such as device running status, device fault code and device maintenance record is obtained through the database API interface of the device business module in the smart operation and maintenance management system; the data of multiple operation and maintenance parameters such as power consumption, water resource consumption, gas consumption and device energy efficiency is obtained through the database API interface of the energy business management module in the smart operation and maintenance management system; the data of multiple operation and maintenance parameters such as weather, water quality, dust and noise is obtained through the database API interface of the environment business management module in the smart operation and maintenance management system; the data of multiple operation and maintenance parameters such as building occupancy rate, tenant number and tenant to be expired is obtained through the database API interface of the building business management module in the smart operation and maintenance management system. The length of each period is 1 hour, and the implementer can set it according to the actual situation as other implementation manners.

[0049] All the obtained data are normalized and missing value filled to obtain data of various operation and maintenance parameters in each business category at different time;

[0050] In the embodiment, Z-Score standardization method is used for normalization, wherein the Z-Score standardization method is a known technology and will not be described here again, and as other embodiments, implementers can use other methods of prior art, for example, maximum and minimum normalization method, and the embodiment does not specially limit this; and median filling method is used for missing value filling, wherein the median filling method is a known technology and will not be described here again, and as other embodiments, implementers can use other methods of prior art, for example, mean filling method, and the embodiment does not specially limit this.

[0051] Thus, data of various operation and maintenance parameters in each business category at different time are obtained.

[0052] In step 2, redundancy dependency of various operation and maintenance parameters at each time period is determined through discrete conditions of all data of various operation and maintenance parameters in each business category at each time period in the digital park and correlation conditions of data between different operation and maintenance parameters; distribution characteristics of same data of various operation and maintenance parameters at each time period are analyzed, noise interference degree of various operation and maintenance parameters at each time period is calculated, and effective contribution degree of various operation and maintenance parameters at each time period is determined in combination with the redundancy dependency.

[0053] Further, the step flow chart of the method for obtaining the effective contribution degree of various operation and maintenance parameters at each time period provided in the embodiment is shown in Figure 2

[0054] In different business categories of the intelligent digital park, redundant or low information amount data not only wastes storage resources, but also threatens storage accuracy and consistency of operation and maintenance data and limits play of value of operation and maintenance data. Data of various operation and maintenance parameters under each business category in the intelligent digital park exist repetition, and fixed noise interference is serious in data collection, which produces a large amount of redundant invalid data, and data correlation in multiple dimensions of various operation and maintenance parameters in each business category can also effectively identify invalid information in data.

[0055] ​Secondly, when the higher the correlation between different kinds of operation and maintenance parameters in each business category and the smaller the data fluctuation, the higher the redundant dependency between different kinds of operation and maintenance parameters in each business category, that is, it can be obtained by deriving the data of the remaining kinds of operation and maintenance parameters, so that the value of independent storage is lower. At the same time, when the data collection of the digital park is seriously affected by noise, the data repetition characteristics of different kinds of operation and maintenance parameters are more obvious, reflecting that the contribution of this kind of operation and maintenance parameter to the information integration and storage of the digital park is lower, and the overall data quality is more affected.

[0056] Based on the above analysis, by analyzing the correlation change and data fluctuation between different kinds of operation and maintenance parameters in each business category, the redundant dependency is calculated, specifically:

[0057] The dispersion degree of all data of each kind of operation and maintenance parameter in each business category at each time period is calculated.

[0058] In this embodiment, the dispersion degree is measured by calculating the coefficient of variation of each kind of operation and maintenance parameter in each business category at each time period, wherein the calculation process of the coefficient of variation is a known technology, which will not be repeated here. As other embodiments, the implementer can use other methods of prior art, such as standard deviation, variance, information entropy, etc. This embodiment does not make special restrictions.

[0059] The correlation degree between all data of any kind of operation and maintenance parameter and the remaining kinds of operation and maintenance parameters in each business category at each time period is calculated.

[0060] In this embodiment, the correlation degree is measured by calculating the Pearson correlation coefficient between all data of each kind of operation and maintenance parameter and the remaining kinds of operation and maintenance parameters in each business category, wherein the calculation of the Pearson correlation coefficient is a known technology, which will not be repeated here. As other embodiments, the implementer can use other methods of prior art, such as cosine similarity, Spearman correlation coefficient, etc. This embodiment does not make special restrictions; for example, for water quality and dust, the Pearson correlation coefficient between all data of water quality and dust at each time period is calculated; for power consumption and gas consumption, the Pearson correlation coefficient between all data of power consumption and gas consumption at each time period is calculated.

[0061] The sum of the absolute values of the correlation degrees between the said any kind of operation and maintenance parameter and the remaining kinds of operation and maintenance parameters in each business category at each time period is denoted as similarity;

[0062] The ratio of the similarity to the dispersion degree is used as the redundant dependency of each kind of operation and maintenance parameter in each business category at each time period.

[0063] It should be noted that the smaller the dispersion degree, the smaller the data volatility of each service category in each period. The greater the similarity, the higher the data correlation between each service category in each period and the remaining service category. The greater the redundancy dependence, the lower the value of independent storage of each service category in each period.

[0064] Further, the periodic fluctuation characteristics and data repetition of each service category in each period are analyzed, and the noise interference degree is calculated, specifically:

[0065] The number of continuous occurrence of the same data of each service category in each period is counted, and the maximum number is obtained.

[0066] The run test algorithm is used to calculate the test statistics of all data of each service category in each period.

[0067] It should be noted that the run test algorithm is a known technology and will not be described here.

[0068] The ratio of the maximum number to the absolute value of the test statistics is taken as the noise interference degree of each service category in each period.

[0069] It should be noted that the smaller the absolute value of the test statistics, the more random the data of each service category in the intelligent digital park, and the more seriously affected by noise. The greater the maximum number, the more repeated data caused by noise, the greater the noise interference degree, and the more seriously affected by noise.

[0070] Further, based on the redundancy dependence and the noise interference degree, the effective contribution degree is determined, specifically:

[0071] The reciprocal of the product of the redundancy dependence and the noise interference degree is taken as the effective contribution degree of each service category in each period.

[0072] It should be noted that the effective contribution degree reflects the data redundancy and invalidity of the corresponding service category in the intelligent digital park. The greater the effective contribution degree, the higher the data storage value of the service category, and the more significant the effective information contained in the data.

[0073] Thus, the effective contribution degree of each service category in each period is obtained.

[0074] Step 3, based on the correlation between each service category in each time period and various operation and maintenance parameters and the rest of the operation and maintenance parameters, the score vector corresponding to each principal component is extracted through dimension reduction analysis; the elements of the score vector at all times are clustered, the dispersion of all clusters of each principal component is analyzed, and the information fuzziness of each principal component is calculated; the correlation between all cluster centers corresponding to each principal component and the rest of the principal components at the same time, and the difference of the score vector, the information irrelevance of each principal component is calculated, and the information invalidity of various operation and maintenance parameters in each time period is determined in combination with the information fuzziness.

[0075] Secondly, the data of various operation and maintenance parameters has high dimension characteristics, for example, the energy management module of the digital park contains various operation and maintenance parameters such as power consumption, water resource consumption, gas consumption and equipment energy efficiency, and various operation and maintenance parameters may be affected by the rest of the multiple operation and maintenance parameters. Only by judging the redundancy of various operation and maintenance parameters through effective contribution degree, the redundant or highly correlated operation and maintenance parameters may be stored, which not only wastes a large amount of storage resources, but also reduces the data quality in the digital park.

[0076] Therefore, when the data dispersion of the rest of the operation and maintenance parameters affecting various operation and maintenance parameters in each service category of the intelligent digital park is higher and the clustering structure is more obvious, it means that the business characteristics of the operation and maintenance parameters are worse, the information lacks effective clustering boundary, and the operation and maintenance parameters are meaningless. At the same time, when the correlation between various operation and maintenance parameters and the rest of the operation and maintenance parameters is stronger, it means that the redundancy of the operation and maintenance parameters is higher.

[0077] Based on the above analysis, the rest of the operation and maintenance parameters affecting various operation and maintenance parameters is selected from each service category, and dimension reduction analysis is performed on it to further analyze the redundancy of various operation and maintenance parameters, specifically:

[0078] The absolute value of the correlation degree between each service category in each time period and all other operation and maintenance parameters is arranged in descending order, and the top pre-set number of operation and maintenance parameters is selected, which is recorded as the associated operation and maintenance parameters corresponding to the operation and maintenance parameters.

[0079] In this embodiment, the top 10 operation and maintenance parameters are selected. As an alternative, the implementer can set it according to the actual situation.

[0080] The dimension reduction analysis is performed on all data of all associated operation and maintenance parameters corresponding to the operation and maintenance parameters in each service category in each time period, and the score vector corresponding to each principal component is extracted;

[0081] In the embodiment, a principal component analysis algorithm (PCA) is used for dimensionality reduction analysis, and the PCA algorithm is a known technique and will not be described herein.

[0082] The K-Mediods clustering algorithm is used to cluster all elements of the score vector of each principal component at all time points to obtain a plurality of clustering clusters.

[0083] In the embodiment, the K-Mediods clustering algorithm is used for clustering, and the K-Mediods clustering algorithm is a known technique and will not be described herein.

[0084] The mean of the difference between all elements in each clustering cluster corresponding to each principal component and the clustering center thereof is calculated, denoted as the average difference, and the sum of the average differences of all clustering clusters corresponding to each principal component is calculated, denoted as the relative deviation of each principal component.

[0085] In the embodiment, the mean of the absolute value of the difference between all elements in each clustering cluster corresponding to each principal component and the clustering center thereof is calculated, denoted as the average difference.

[0086] The information entropy of the clustering center of all clustering clusters corresponding to each principal component is calculated, and the product of the relative deviation and the information entropy is used as the information fuzziness of each principal component.

[0087] It should be noted that the calculation of the information entropy is a known technique and will not be described herein. In addition, the greater the relative deviation, the more dispersed the elements corresponding to the principal component relative to the clustering center, and it is difficult to form a clear clustering structure. The greater the information entropy, the more uniform the distribution between all clustering centers corresponding to the principal component, and there is no obvious clustering boundary. The greater the information fuzziness, the higher the dispersion and uncertainty of the score vector of the principal component in the clustering analysis, and the information of the principal component is difficult to clearly reflect the information characteristics of the corresponding operation and maintenance parameters.

[0088] The sum of the differences between the score vectors of each principal component and all other principal components is denoted as the relative difference.

[0089] In the embodiment, the sum of the Kullback Leibler (KL) divergences between each principal component and all other principal components is calculated, denoted as the relative difference, and the calculation of the KL divergence is a known technique and will not be described herein.

[0090] The sum of the absolute values of the correlation coefficients corresponding to the time points of the clustering centers of all clustering clusters between each principal component and all other principal components is calculated, denoted as the correlation degree of each principal component.

[0091] In the embodiment, the correlation coefficient is measured by calculating the Pearson correlation coefficient of the clustering centers of all clusters at the corresponding time of each principal component and all other principal components. Alternatively, other methods known in the art can be used, such as the Spearman correlation coefficient, and the embodiment does not make special limitations.

[0092] The ratio of the relative difference and the correlation degree is used as the information irrelevance of each principal component.

[0093] In the embodiment, the calculation formula of the information irrelevance of each principal component is as follows:

[0094]

[0095] wherein, C k is the information irrelevance of the kth principal component, G k,h is the KL divergence of the score vector between the kth principal component and the hth principal component, r k,h is the Pearson correlation coefficient of the clustering centers of all clusters at the corresponding time between the kth principal component and the hth principal component, H is the number of all principal components corresponding to the any operation and maintenance parameter, and ε is a preset value greater than 0, which avoids the denominator being 0. In the embodiment, ε is 0.01, and alternatively, the implementer can set it according to the actual situation.

[0096] It should be noted that the greater the relative difference and the smaller the correlation degree, the greater the information difference between each principal component and the other principal components, but the similarity between them in time is low, which reflects that the similarity between the principal component and the other principal components in the clustering structure is low, and the more unnecessary information is contained in the principal component, and the greater the information irrelevance.

[0097] The sum of the product of the information ambiguity and the information irrelevance of all principal components corresponding to the any operation and maintenance parameter is used as the information invalidity of the any operation and maintenance parameter.

[0098] It should be noted that the greater the information invalidity, the more serious the data pollution of the operation and maintenance parameter, and the less storage value it has.

[0099] Thus, the information invalidity of the any operation and maintenance parameter is obtained.

[0100] Step 4, based on the effective contribution degree and the information invalidity, the discriminant coefficient of each operation and maintenance parameter in each period is determined, all operation and maintenance parameters in each period are screened and stored.

[0101] Further, when the effective contribution degree of each business category of various operation and maintenance parameters of the intelligent digital park is greater and the information invalidity is smaller, it indicates that the data storage value of the operation and maintenance parameter is higher, the redundant invalid data is less, and it is more beneficial to improve the operation and maintenance data integration storage quality of the intelligent digital park and play the operation and maintenance data value.

[0102] Based on the above analysis, based on the effective contribution degree and the information invalidity, a discriminant coefficient is determined, specifically:

[0103] The normalized result of the ratio of the information invalidity and the effective contribution degree is taken as the discriminant coefficient of the operation and maintenance parameter;

[0104] In the embodiment, the sigmoid function is used for normalization processing, wherein the sigmoid function is a known technology and will not be described here. As other implementation manners, the implementer can use other methods of prior art, for example, tanh function, and the embodiment does not specially limit this.

[0105] It should be noted that the discriminant coefficient reflects the information contribution effective value degree and data storage quality condition of various operation and maintenance parameters in each business category in integrated storage; the greater the discriminant coefficient is, the lower the independent storage value of the operation and maintenance parameter is, the more significant the redundant invalid information is, the more seriously the noise is affected, and the storage of the data of the operation and maintenance parameter will affect the overall storage quality of the operation and maintenance data in the intelligent digital park. The step flow chart of the discriminant coefficient provided in the embodiment is as shown in Figure 3

[0106] Further, based on the discriminant coefficient, the storage value of various operation and maintenance parameters is evaluated, and the data is stored, specifically:

[0107] All data of all kinds of operation and maintenance parameters whose discriminant coefficients are less than a preset threshold value in each period are stored, and are stored in a database;

[0108] In the embodiment, the preset threshold value is 0.6, and as other implementation manners, the implementer can set it according to actual conditions.

[0109] It should be noted that when the discriminant coefficient is greater than or equal to the preset threshold value, it indicates that the operation and maintenance parameter contains more redundant invalid information, has lower storage value, has greater influence on data storage quality, and has greater storage resource waste degree; otherwise, the operation and maintenance parameter contains less redundant invalid information, has higher storage value, has slight influence on data storage quality, and by selecting all kinds of operation and maintenance parameters whose discriminant coefficients are less than the preset threshold value for data storage, the data integration quality is improved and the storage resource is released.

[0110] ​By deploying an ETL tool in the operation and maintenance data storage database of the intelligent digital park, and using JSON-LD to realize semantic annotation, structured labels are added to each piece of data, including various operation and maintenance parameters, time information, and discriminant coefficients; a MySQL relational database is used for data storage, and the data can also be stored by encryption algorithm to ensure data security. The data storage process and encryption algorithm are both well-known technologies, and will not be described here.

[0111] Based on the same inventive concept as the above method, the embodiments of the present application also provide an operation and maintenance data storage system for an intelligent digital park, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any one of the above methods for storing operation and maintenance data of an intelligent digital park.

[0112] It should be understood that, although Figure 1 The steps in the flowchart are displayed in sequence according to the direction of the arrows, but these steps are not necessarily executed in sequence according to the direction of the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, Figure 1 At least part of the steps in the above embodiments can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these sub-steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or sub-steps or stages of other steps.

[0113] The technical features of the above embodiments can be combined in any way. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0114] The above embodiments only express several embodiments of the present application, and the description is more specific and detailed, but it should not be understood as a limitation of the present application. It should be noted that for ordinary skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, therefore, any simple modification, equivalent change and modification of the above embodiments according to the technical essence of the present application, which does not deviate from the technical solution of the present application, all belong to the protection scope of the technical solution of the present application.

Claims

1. A method for storing operation and maintenance data in an intelligent digital park, characterized in that, The method includes the following steps: By analyzing the discreteness of all data for various operation and maintenance parameters in each business category during different time periods within the digital park, as well as the correlation between different types of operation and maintenance parameters, the redundancy dependency of various operation and maintenance parameters during different time periods can be determined. Analyze the distribution characteristics of the same data for various operation and maintenance parameters in different time periods, calculate the noise interference degree of various operation and maintenance parameters in different time periods, and combine the redundancy dependency degree to determine the effective contribution of various operation and maintenance parameters in different time periods. Based on the correlation between various operation and maintenance parameters and other operation and maintenance parameters in each business category under each time period, the score vector corresponding to each principal component is extracted through dimensionality reduction analysis. Cluster the elements at all times within the score vector, analyze the inter-cluster and intra-cluster discreteness of all clusters of each principal component, and calculate the information ambiguity of each principal component; calculate the information irrelevance of each principal component by considering the correlation between each principal component and the other principal components at all times corresponding to the cluster centers, as well as the differences in the score vector; and determine the information invalidity of various operation and maintenance parameters at each time period by combining the information ambiguity. Based on effective contribution and information invalidity, the discrimination coefficients of various operation and maintenance parameters under each time period are determined, and all operation and maintenance parameters under each time period are screened and stored. The determination of the redundancy dependency of various operation and maintenance parameters includes: Calculate the correlation between any one type of operation and maintenance parameter and all other operation and maintenance parameters in each business category under each time period; The sum of the absolute values ​​of the correlation between any one type of operation and maintenance parameter and all other types of operation and maintenance parameters is denoted as the similarity. Calculate the dispersion of all data for any one type of operation and maintenance parameter in each time period; The redundancy dependency is the ratio of the similarity to the degree of dispersion.

2. The operation and maintenance data storage method for an intelligent digital park as described in claim 1, characterized in that, The calculation of noise interference levels for various operation and maintenance parameters includes: Count the number of consecutive occurrences of the same data for various operation and maintenance parameters in different time periods, and obtain the maximum number. The run-length test algorithm is used to calculate the test statistics of all data for various operation and maintenance parameters in each time period. The noise interference level is the ratio of the maximum quantity to the absolute value of the test statistic.

3. The operation and maintenance data storage method for an intelligent digital park as described in claim 1, characterized in that, The effective contribution is the reciprocal of the product of the redundancy dependency and the noise interference.

4. The operation and maintenance data storage method for an intelligent digital park as described in claim 1, characterized in that, The extraction of the score vector corresponding to each principal component includes: Arrange the absolute values ​​of the correlation between any one type of operation and maintenance parameter and all other operation and maintenance parameters in descending order for each business category in each time period, and select a preset number of operation and maintenance parameters that are ranked first, and record them as the associated operation and maintenance parameters corresponding to the one type of operation and maintenance parameter. Dimensionality reduction analysis is performed on all data of all associated operation and maintenance parameters corresponding to any operation and maintenance parameter in each business category under each time period, and the score vector corresponding to each principal component is extracted.

5. The operation and maintenance data storage method for an intelligent digital park as described in claim 1, characterized in that, The calculation of the information ambiguity of each principal component includes: Calculate the mean difference between all elements in each cluster and their cluster center, denoted as the average difference. Calculate the sum of the average differences of all clusters corresponding to each principal component, denoted as the relative deviation of each principal component. Calculate the information entropy of the cluster centers of all clusters corresponding to each principal component; The information ambiguity is the product of the relative deviation and the information entropy.

6. The operation and maintenance data storage method for an intelligent digital park as described in claim 1, characterized in that, The calculation of the information irrelevance of each principal component includes: The sum of the differences between each principal component and the score vectors of all other principal components is denoted as the relative difference; The sum of the absolute values ​​of the correlation coefficients between each principal component and the cluster centers of all clusters at the corresponding time points is calculated and denoted as the correlation degree of each principal component. The information irrelevance is the ratio of the relative difference to the relevance.

7. The operation and maintenance data storage method for an intelligent digital park as described in claim 1, characterized in that, The information invalidity is the sum of the products of the information ambiguity and the information irrelevance of all principal components corresponding to various operation and maintenance parameters.

8. The operation and maintenance data storage method for an intelligent digital park as described in claim 1, characterized in that, The process of determining the discrimination coefficients for various operation and maintenance parameters in each time period, and filtering and storing all operation and maintenance parameters in each time period includes: The normalized result of the ratio of the invalidity of the information to the effective contribution is used as the discrimination coefficient of various operation and maintenance parameters; all data of all operation and maintenance parameters whose discrimination coefficient is less than the preset threshold in each time period are stored.

9. A data storage system for operation and maintenance of an intelligent digital park, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the operation and maintenance data storage method for an intelligent digital park as described in any one of claims 1-8.

Images