Keywords

1 Introduction

With the advent of the big data era, massive data information is explosive growth. How can we extract effective knowledge from these massive data information and express and store it, so as to better understand the use for humans and machines is a hot research topic at present. However, the content multi-source heterogeneity and loose structure on the Internet brings great challenges to the extraction of knowledge, data fusion, and storage [1, 2]. Knowledge map is the subject that produces in this background. In recent years, knowledge map has received extensive attention, and the application of knowledge map has been greatly developed through data collation, data mining, machine learning, and expert system. But at present, the application research of knowledge map is still in its infancy, and there are many limitations, low efficiency and poor expansibility.

In 2019, the “two sessions” of the national network put forward the requirements for building world-class energy Internet enterprises around “three types and two networks,” speeding up the construction of the universal electric power Internet of Things, fully applying modern information technology, advanced communication technology such as “big cloud mobile intelligence chain,” vigorously improving the ability of automatic data collection, automatic acquisition, flexible application, and realizing “one source of data, one map of power grid, one line of business” to promote the unified management and application of company data [3, 4].

Based on the knowledge map, data fusion technology research is around the power grid company business, relying on the whole business unified data center construction results, using the knowledge map “node, edge, color” construction technology to achieve the network company business, data and grid structure relationship series, so as to establish a new business and big data innovative application of the big data fusion model and to explore the formation of a data fusion system applicable to the company.

2 Status of Data Use and Problems

Since the 13th Five-Year Plan, the national network company has carried out the whole business unified data center, the whole area comprehensive demonstration project construction, and has initially realized the data fusion convergence. With the expansion of business, data application has been upgraded from regular monitoring analysis and professional decision analysis to “big data analysis” and “big data mining” of “cross-business.” However, there are still many problems in the use of data. First, the data relationship of each professional system is more complex, and non-professional personnel cannot quickly understand the data of other specialties. Second, at the operational level, each profession, each system is still relatively independent state. Third, the cross-professional data use. We need business personnel, technical personnel, data management personnel, and other multi-party support, autonomous, self-help application ability is weak.

In order to effectively support the company’s business expansion and application, further promote the ability of data application, in order to better solve the problem of cross-professional data fusion application, so as to carry out knowledge map-based data fusion technology research and application.

3 Key Approaches to Knowledge Map-Based Data Fusion

As a unified whole composed of hair, transmission, transformation, distribution and use of each link, the grid needs to coordinate and dispatch each other from the operation characteristics. With the increasing input of power generation and storage equipment, we have more and more the improvement of protection devices for transmission network security and stability, the increasing demand response load, the increasing access of electric vehicles and charging piles, the increasing coverage of data acquisition devices such as smart meters and remote measurement and control terminals, the increasing scale and complexity of power grid operation in the region. Therefore, the data fusion technology based on knowledge map is needed to realize the business integration of transmission and distribution network (Fig. 1).

Fig. 1
figure 1

Key approaches to knowledge map-based data fusion

Full size image

3.1 Build Business Process Maps

The business process map construction is an effective way to realize the integration of business and data among various specialties, which provides a convenient channel for the application of stock business and the combing of data requirements. However, due to the lack of effective integration mechanism between business processes, business management is still decentralized. In order to realize the business map fusion, the business topology model is constructed from the point of view of business management and application. At the same time, the business topology fusion degree is analyzed by using graph calculation and big data analysis technology, the business data fusion situation is judged, the business topology map is continuously improved, and finally, the business process map is formed [5, 6].

3.2 Construction of Equipment Asset Profiles

In view of the business process data, we can sort out the existing business system and business process, construct the business process map for the core and curable business, gradually realize the full coverage of power grid business, establish the intelligent business flow management application map of integrated business management, process visualization, dynamic update and expand, form “business one chart,” connect the process links with business data effectively, make the data and business mutually confirm, and realize the purpose of resource unification, result interaction, up-down linkage, application chart calculation, big data analysis, and so on existing data analysis. Thus, the sub-maps of each business are associated and integrated, and the full coverage of the business process maps is realized step by step (Fig. 2).

Fig. 2
figure 2

Build business flowchart

Full size image

3.3 Build Data Asset Profiles

Based on the company’s existing business data, using data integration to obtain technical research results, relying on the company’s enterprise information model standards, to build the whole business data asset map, form a “data one map,” achieve business data cross-professional, cross-system unified, standardized storage and management, dynamically update the data asset map according to the source end business adjustment, and achieve unified data standards, source traceability, dynamic update data asset management and application.

Using knowledge map and big data analysis technology, this paper analyzes the data association relationship between various specialties and systems, constructs the equipment asset map, business process map and data asset map, respectively, solidifies and precipitates the existing business through the knowledge map, realizes the association between the three maps, and constructs the “grid one map,” so as to achieve the purpose of quickly responding to the data requirements of various specialties and units and efficiently constructing the application of big data analysis.

3.4 Building Business Data Fusion Maps

The “one map of the grid” establishes the data standard for the cross-professional data fusion; at the same time, it also preliminarily realizes the fusion and penetration of each professional data, lays the foundation for the subsequent data fusion model construction, and provides the efficient and high-quality model iterative updating ability.

By using the techniques of graph calculation and big data analysis, the linkage analysis mechanism of equipment asset grid map, business process map, and data asset map is established, and the equipment asset operation management process is related to equipment asset grid map, business map, and data map, and the business flow and data flow are effectively combined to realize the mutual verification of equipment asset operation, business process and business data and further improve the level of enterprise lean management.

Using graph calculation, machine learning and other technologies to realize the horizontal association of equipment asset map, business process map and data asset map, to support multidimensional management of power grid business, to establish equipment asset view business and data, to view related equipment and data from business process, to view corresponding process and equipment from data, to meet business management and application requirements in all directions, to realize rapid response of data demand, effectively support business innovation and adaptability adjustment of operation monitoring, and to improve data application efficiency [7].

3.5 Data Fusion Support

Based on domestic and foreign mature theories and advanced technologies such as data standard governance, master data identification, data fusion degree analysis, and so on, combined with the company’s business system, establish data standard discrimination and governance mechanism around business data fusion, perfect master data identification method, apply data integration to obtain advanced technology, support business data integration, improve data fusion depth through data fusion degree analysis, optimize data fusion model, and support demand response and construction of cross-professional and cross-domain big data application.

By combing the present situation of the company’s master data, using the way of dividing according to the business domain or according to the application level, using the ORC identification technology and the comprehensive weighting method, the complete master data identification method is constructed, and the operation flow of master data identification is established.

4 Conclusion

Through the research of data fusion technology based on knowledge map, it promotes the new business application and management mode of the combination of company data, business, and grid truss and has achieved remarkable results in supporting various business departments, prefectural and municipal companies to carry out data application analysis and bring into play the value of data. First, through the open application of data fusion technology, we can quickly find out the problems of non-uniform standards, non-standard, and poor data quality in company business data, so as to promote the formation of corporate data standard governance system, establish a network of data fusion relationship, and provide reference for the company to formulate data management model. Second, through the research and application of data fusion technology based on knowledge map, aiming at the characteristics of dispersion, diversity and complexity of current company business data, we deeply analyze the results of company business data fusion and carry out typical case studies to extract a number of businesses.