Intelligent mine-oriented main data application system construction method

Abstract

The invention discloses a main data application system construction method for an intelligent mine, and the method comprises the following steps: S1, constructing a fact table system based on the core business of the mine according to a big data dimension model and a data granularity recognition theory; constructing a corresponding series dimension table through a fact table system, determining a hierarchical structure of the series dimension table, and finally constructing and forming a mine big data snowflake model library; s2, on the basis of a mine big data snowflake model library, according to a main data identification model, mine main data and total evaluation scores thereof are identified from data items corresponding to the dimension table, and a mine big data main data system is formed; and S3, developing a mine main data application system based on a mine big data main data system of the mine big data snowflake model library. The mine data covered by the method is more comprehensive and complete, the identification accuracy of the main data is higher, the main data is prevented from being missed, the integrity of the main data is ensured, and the key problems in the aspects of intelligent mine big data standard system construction, data quality improvement, data asset management optimization and the like are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of mine data processing, and in particular to a method for constructing a master data application system for intelligent mines. Background Art

[0002] Intelligent mining is an important part of the high-quality development of the national economy. The construction of intelligent mines faces the problems of inconsistent data standards and specifications and difficult system integration. As a result, mining companies have formed "data barriers" and "information islands" in production, safety, management and other business data, making it difficult to achieve an intelligent development pattern based on data fusion and sharing, and also restricting the high-quality development of mining companies.

[0003] For example, the Chinese invention patent application with application number 202410024533.7 discloses a data governance system based on mine big data. The above solution collects data from heterogeneous data sources corresponding to multiple business modules through multiple data collection interfaces, and processes and stores data based on unified data description standards and data storage standards to form data assets that can be shared. It makes up for the shortcomings of the current smart mining platform in data governance, data standards, etc., and through standardized ETL process processing, it realizes standardized management of data, realizes data value, provides assistance to coal mines, and realizes the innovation of coal mine business models. However, mine data is scattered and messy, and there is overlap between multiple departments, with a large amount of redundant data. It is difficult to retrieve the key data of the mine efficiently, and it is difficult to ensure that the key data of the mine remains consistent and accurate throughout the enterprise.

[0004] For example, the Chinese invention patent application with application number 202210729769.1 discloses a master data identification method for smart mines. By establishing a tree-like data graph feature data and constructing an indicator discrimination matrix, important hot spot data is marked and analyzed based on the importance of n evaluation indicators and the preliminary scoring of expert evaluation, thereby improving its recognition rate. The tree-like classification method reduces the redundancy of the data and makes a certain degree of distinction on the data, thereby improving its computing speed.

[0005] The above scheme only discloses the method for identifying the master data of the mine, but does not specifically disclose how to pre-process the scattered and messy mine data. It is difficult to ensure the integrity of the mine data, and often causes some master data to be missed. In addition, the above scheme only relies on the indicator discriminant matrix to identify and analyze the master data, and the recognition accuracy still has room for improvement. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for constructing a master data application system for intelligent mines, so as to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for constructing a master data application system for intelligent mines includes the following steps:

[0009] S1. Based on the big data dimension model and data granularity identification theory, the mining big data is dimensionally modeled and a fact table system based on the core business of the mine is constructed. The corresponding series dimension tables are constructed through the fact table system, and the hierarchical structure of the series dimension tables is determined based on the data granularity theory and the snowflake model. Finally, a multi-level structured mining big data snowflake model library is constructed.

[0010] S2. Based on the constructed mine big data snowflake model library and the pre-designed master data identification model, the mine master data and its total evaluation score are identified from the data items corresponding to the dimension table of one level to form a mine big data master data system.

[0011] S3. Develop a mine master data application system based on the mine big data snowflake model library to provide support for the query, application and promotion of mine master data.

[0012] As a further solution of the present invention, the specific steps of step S1 are:

[0013] S101. Before building the series dimension table, clarify the granularity of the fact table. Each fact table is based on business events and also contains multiple descriptive attributes related to the business events. Each descriptive attribute is used as a series foreign key of the fact table. The series foreign key is used to point to the corresponding series dimension table.

[0014] S102. Define a hierarchical structure for a series dimension table based on business requirements and data granularity. The series dimension table includes dimension tables at multiple levels. The series foreign key of the fact table serves as the primary key of the dimension table at the adjacent level. Each dimension table also includes multiple descriptive attributes related to its primary key, namely, the foreign keys of the dimension table. Each foreign key of the dimension table corresponds to related management data.

[0015] S103. According to the complexity of the dimension, the dimension table is refined, and the foreign key of the dimension table of the previous level is used as the primary key of the dimension table of the next level. This process is repeated to form a snowflake data model. By integrating various snowflake data models, a multi-level structured mining big data snowflake model library is constructed.

[0016] As a further solution of the present invention, the fact table system includes: a geological fact table, a production fact table, a safety fact table and / or an operation and management fact table;

[0017] The mine big data snowflake model library includes: a geological data snowflake model, a production data snowflake model, a safety data snowflake model and / or an operation and management data snowflake model.

[0018] As a further solution of the present invention, the master data identification model in step S2 includes a master data identification index system and an evaluation model, and the evaluation model includes an AHP index weight calculation module, an entropy weight method index entropy weight calculation module, a combination index weight calculation module and a comprehensive evaluation module;

[0019] The master data identification indicator system includes primary indicators, secondary indicators, indicator definitions, and indicator scoring standards. The primary indicators include shareability, stability, accuracy, and importance. The secondary indicators corresponding to shareability are the number of shared business departments and system span. The secondary indicators corresponding to stability are data validity period and update frequency. The secondary indicators corresponding to accuracy are uniqueness and anomaly rate. The secondary indicators corresponding to importance are business priority and security classification. Each secondary indicator has a corresponding indicator definition and indicator scoring standard.

[0020] Based on the master data identification indicator system, the importance of each indicator corresponding to the data item in one of the hierarchical dimension tables in the mine big data snowflake model library is assigned to obtain the corresponding mine data indicator score data set.

[0021] As a further solution of the present invention, the calculation method of the AHP index weight calculation module is:

[0022] S211. Construct a comparison matrix: Perform pairwise comparisons on mining data indicators (such as the number of shared business departments, system span, data validity period, update frequency, uniqueness, anomaly rate, business priority, and security classification) to establish a comparison matrix A, which is expressed as follows:

[0023]

[0024] Among them, A represents the comparison matrix, a ij is the value compared between data index elements i and j, and n is the total number of elements.

[0025] S212. Calculate indicator weights:

[0026] Calculate the sum S of each column of the comparison matrix j :

[0027]

[0028] By adding each value a ij Divide by the sum of its corresponding column S j , get the normalized value Constructing a normalized comparison matrix Its expression is as follows:

[0029]

[0030] Calculate the weight w of each data indicator element i , we get the weight vector W, which is expressed as follows:

[0031]

[0032] W=[w1,w2,w3…w i …,w n ](i=1,2,…,n);

[0033] S213, consistency check:

[0034] Calculate the maximum eigenvalue λ max , whose expression is:

[0035]

[0036] Calculate the consistency index (CI):

[0037]

[0038] Calculate the consistency ratio (CR):

[0039]

[0040] Among them, RI is the random consistency index;

[0041] As a further solution of the present invention, the calculation method of the entropy weight calculation module of the entropy weight method indicator is:

[0042] S221. Data Standardization:

[0043] (Since the dimensions and numerical ranges of different indicators may be different, the first step is to standardize the mining data indicator score dataset. This application uses range standardization).

[0044] The range standardization method is used to standardize the mine data index score data set, and its expression is:

[0045]

[0046] Among them, x ef is the value of the fth indicator of the mine data item e, x' ef is the index value after standardization, min(x f ) is the minimum value of the f-th index, max(x f ) is the maximum value of the f-th index;

[0047] S222. Calculate the entropy value and entropy weight of each indicator:

[0048] According to each standardized indicator value, calculate its proportion p in the total of the indicator ef , whose expression is:

[0049]

[0050] Calculate the entropy value h of each indicator f , whose expression is:

[0051]

[0052] Where n is the number of samples;

[0053] Calculate the redundancy d f , whose expression is:

[0054] d f =1-h f #(10);

[0055] The higher the redundancy, the more important the indicator is and the more concentrated the information is;

[0056] Finally, the entropy weight w of each indicator is calculated by redundancy f , whose expression is:

[0057]

[0058] Among them, m is the number of indicators, and the entropy weight reflects the importance of each indicator in the evaluation;

[0059] As a further solution of the present invention, the calculation method of the combined index weight calculation module is:

[0060] Combining the AHP indicator weight and the entropy weight method indicator entropy weight, the combined weight is calculated and the subjective and objective combined weighting is performed. The expression is:

[0061]

[0062] Among them, W z is the comprehensive weight of the indicator, w i is the weight of the data indicator element calculated by AHP, w f is the entropy weight calculated by the entropy weight method.

[0063] As a further solution of the present invention, the calculation method of the comprehensive evaluation module is:

[0064] Calculate the comprehensive score of the data items corresponding to each dimension table. The expression is:

[0065]

[0066] Among them, s e is the comprehensive score of the e-th mine data item, w j is the combined weight, x ef is the value of the fth indicator of the eth mine data item;

[0067] Compare the comprehensive score with the pre-set judgment threshold. If the comprehensive score exceeds the judgment threshold, the corresponding data item will be included in the main data table;

[0068] By summarizing and establishing the master data table and its total evaluation score, a mine big data master data system is formed.

[0069] An electronic device, comprising: a memory storing a computer program for the aforementioned method for constructing a master data application system for intelligent mines;

[0070] A processor is used to retrieve and execute the computer program stored in the memory.

[0071] A computer-readable storage medium stores a computer program, which, when executed, implements the aforementioned method for constructing a master data application system for intelligent mines.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] 1. Based on the big data dimensional model and data granularity identification theory, the present invention conducts dimensional modeling on mining big data and constructs a fact table system based on the core business of the mine. The corresponding series dimension tables are constructed through the fact table system, and the hierarchical structure of the series dimension tables is determined based on the data granularity theory and the snowflake model. Ultimately, a multi-level structured mining big data snowflake model library is constructed. The mining data covered by the mining big data snowflake model library is more comprehensive and complete, providing a foundation for the subsequent identification of key data, and solving key issues in the construction of intelligent mining big data standard systems, data quality improvement, and data asset management optimization.

[0074] 2. The present invention first assigns importance to each indicator of the data item corresponding to one of the hierarchical dimension tables in the mine big data snowflake model library based on the master data identification index system, and obtains the corresponding mine data indicator score data set. Then, the hierarchical analysis method is used to calculate the weight of the mine data indicator score data set. Then, the entropy value and entropy weight of each data item indicator are calculated using the entropy weight method. Then, the combined weight is calculated by combining the AHP indicator weight and the entropy weight of the entropy weight method indicator. A comprehensive evaluation is then performed, and the comprehensive score is compared with a pre-set judgment threshold. If the comprehensive score exceeds the judgment threshold, the corresponding data item is included in the master data table. By summarizing and establishing the master data table and its total evaluation score, a mine big data master data system is constructed. Through the above method, the recognition accuracy of the master data is higher, the master data is prevented from being missed, and the integrity of the master data is guaranteed.

[0075] 3. The present invention incorporates mine master data into the master data table. On the one hand, it can ensure the consistency and accuracy of key mine data throughout the enterprise, promote collaborative communication, reduce information silos and information asymmetry, and effectively avoid data inconsistency or errors that may lead to misunderstandings or wrong decisions;

[0076] 4. Through master data management, the present invention can better analyze the current situation and predict trends in the decision-making process, manage equipment and production plans more effectively, optimize production processes, reduce downtime, improve equipment utilization and production efficiency, optimize resource utilization and production scheduling, and reduce production costs and management costs;

[0077] 5. On the other hand, the present invention also helps enterprises to respond to market changes more quickly and enhance their market competitiveness. At the same time, it helps enterprises comply with laws and regulations, reduce the risk of safety accidents, improve employees' safety awareness, reduce resource waste and environmental load, and promote the sustainable development of mines. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 It is a schematic diagram of the process of the present invention;

[0079] Figure 2 Schematic diagram of the geological data snowflake model in the present invention;

[0080] Figure 3 A schematic diagram of the production data snowflake model in the present invention;

[0081] Figure 4 Schematic diagram of the security data snowflake model in the present invention;

[0082] Figure 5 A schematic diagram of the snowflake model of business management data in the present invention;

[0083] Figure 6 Schematic diagram of the specific process of step S2 in the present invention. DETAILED DESCRIPTION

[0084] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0085] See also Figure 1 A method for constructing a master data application system for intelligent mines includes the following steps:

[0086] S1. Based on the big data dimension model and data granularity identification theory, the mining big data is dimensionally modeled and a fact table system based on the core business of the mine is constructed. The corresponding series dimension tables are constructed through the fact table system, and the hierarchical structure of the series dimension tables is determined based on the data granularity theory and the snowflake model. Finally, a multi-level structured mining big data snowflake model library is constructed.

[0087] S2. Based on the constructed mine big data snowflake model library and the pre-designed master data identification model, the mine master data and its total evaluation score are identified from the data items corresponding to the dimension table of one level to form a mine big data master data system.

[0088] S3. Develop a mine master data application system based on the mine big data snowflake model library to provide support for the query, application and promotion of mine master data.

[0089] In the present invention, the specific steps of step S1 are:

[0090] S101. Before building a series dimension table, clarify the granularity of the fact table (as the basic unit). Each fact table is based on a business event (such as production, geology, safety, and operations) and also contains multiple descriptive attributes related to the business event. Each descriptive attribute is used as a series foreign key in the fact table. The series foreign key is used to point to the corresponding series dimension table.

[0091] S102. Define a hierarchical structure for a series dimension table based on business requirements and data granularity. The series dimension table includes dimension tables at multiple levels. The series foreign key of the fact table serves as the primary key of the dimension table at the adjacent level. The primary key of the dimension table is used for data retrieval. Each dimension table also includes multiple descriptive attributes related to its primary key, namely, the foreign keys of the dimension table. Each foreign key of the dimension table corresponds to related management data.

[0092] S103. According to the complexity of the dimension, the dimension table is refined, and the foreign key of the dimension table of the previous level is used as the primary key of the dimension table of the next level. This process is repeated to form a snowflake data model. By integrating various snowflake data models, a multi-level structured mining big data snowflake model library is constructed.

[0093] In the present invention, the fact table is associated with the primary key of the dimension table through its own series of foreign keys, and the foreign key of the dimension table of the previous level is used as the primary key of the dimension table of the next level. This is repeated to construct a snowflake data model. The fact table and each dimension table form a total of four hierarchical structures. For example, the "basic geology" of the geological fact table points to the basic geology dimension table, the "coal quality classification" of the basic geology dimension table points to the coal quality classification dimension table, and the "bituminous coal" of the coal quality classification dimension table points to the bituminous coal dimension table. The bituminous coal dimension table has descriptive attributes such as fixed carbon content, adhesiveness, calorific value, volatile matter yield, colloid layer thickness, hardness, ignition point and density. Each dimension level contains different granularity data.

[0094] In the present invention, the specific process of dimensional modeling of mining big data is as follows:

[0095] 1) Based on intelligent mining big data, extract the main fact data to form geological fact tables, production fact tables, safety fact tables, and business management fact tables. These fact tables can basically cover the basic information and business areas of coal industry mining big data.

[0096] The series of foreign keys of the geological fact table are also the primary keys of its dimension tables, which specifically include: basic geology, mining conditions, disaster conditions, etc.

[0097] The series of foreign keys of the production fact table are also the primary keys of its dimension table, which specifically include: coal mining, tunneling, power supply and distribution, hoisting, main transportation, auxiliary transportation, ventilation, compressed air, water supply, drainage, cooling and refrigeration, washing, scheduling management, production technology management, roof management, electromechanical management, blasting management, etc.

[0098] The series of foreign keys in the safety fact table are also the primary keys of its dimension tables. They specifically include: roof management, rock burst prevention and control, water hazard prevention and control, fire prevention and control, gas prevention and control, dust prevention and control, heat hazard control, safety monitoring system, underground worker management, video surveillance, communication, risk classification management, accident hazard investigation and control, superior safety inspection, unsafe behavior management, accident management, safety training, occupational health, emergency management, environmental protection, etc.

[0099] The series of foreign keys of the business management fact table are also the primary keys of its dimension tables, which specifically include: human resources management, financial management, audit management, material management, equipment management, transportation and marketing management, energy conservation and emission reduction management, scientific and technological management, project management, legal management, comprehensive management, information management, etc.

[0100] 2) The fact table is linked to the primary keys of the dimension tables through its own set of foreign keys. The foreign keys of the dimension tables in the previous level are used as the primary keys of the dimension tables in the next level. This process is repeated to form a snowflake data model. The fact table and each dimension table form a four-level hierarchy, of which the fact table is the first level, and the dimension tables form the remaining three levels. Obviously, the dimension tables can also form four, five, or more levels as needed.

[0101] Geological Data Snowflake Model

[0102] like Figure 2 As shown in the figure, the series of foreign keys of the geological fact table are also the primary keys of its dimension tables, which specifically include: basic geology, mining conditions, disaster conditions, etc.

[0103] The dimensional tables under the first layer of basic geology include: basic geological data, media classification, field exploration, reserves, geological maps, etc.; the dimensional tables under the first layer of disaster conditions include: gas disasters, water disasters, surrounding rock disasters, rock burst, heat damage, mine floods, dust disasters, etc.

[0104] The dimensional tables under the second-layer geological basic data include: coal seams, coal seam roofs and floors, geological structures, etc.; the dimensional tables under the second-layer coal quality classification include: bituminous coal, anthracite, lignite, etc.; the dimensional tables under the second-layer reserve data include: second-layer resource reserve basic information, loss report information, reserve dynamic information, reserve summary calculation, reserve increase and decrease transfer, mining area loss information, etc.

[0105] The third-layer dimension table for bituminous coal includes: coalification degree, fixed carbon content, volatile matter yield, colloidal layer thickness, hardness, flash point, density, etc.; the third-layer dimension table for anthracite includes: coalification degree, fixed carbon content, volatile matter yield, density, hardness, flash point, calorific value, etc.; the third-layer dimension table for lignite includes: moisture, volatile components, flash point, fixed carbon content, density, hardness, ash content, etc. Due to space limitations, we will not elaborate on this further, and the accompanying figures do not show all of them.

[0106] Production data snowflake model

[0107] like Figure 3 As shown, the series of foreign keys of the production fact table are also the primary keys of its dimension table, which specifically include: coal mining, tunneling, power supply and distribution, hoisting, main transportation, auxiliary transportation, ventilation, compressed air, water supply, drainage, cooling and refrigeration, washing, scheduling management, production technology management, production planning management, electromechanical management, blasting management, etc.

[0108] The maintenance items under the first layer of coal mining include: coal mining machine, support system, fluid supply system, power supply system, transportation system, centralized control center, etc.; the maintenance items under the first layer of tunneling include: tunneling equipment, transportation equipment, etc.; the maintenance items under the first layer of main transportation include: belt conveyor, coal blending equipment, etc.

[0109] The maintenance tables under the second layer of tunneling equipment include: tunneling machine, integrated miner, continuous miner, anchor drilling vehicle, TBM fast tunneling, etc.; the maintenance tables under the second layer of transportation equipment include: shuttle car, crawler transfer crusher, belt conveyor protection status, belt conveyor belt tensioning device, belt conveyor CST start, etc.

[0110] The third-level dimension tables for roadheaders include: roadheader footage position, roadheader start / stop status, communication heartbeat, control mode, current operation mode, etc.; the third-level dimension tables for integrated miners include: integrated miner footage position, integrated miner start / stop status, communication heartbeat, etc.; the third-level dimension tables for continuous miners include: continuous miner position, continuous miner status, control mode, etc.; the third-level dimension tables for anchor drills include: anchor drill position, anchor drill status, anchor drill operation mode, etc.; the third-level dimension tables for TBM fast excavation include: TBM fast excavation status, cutterhead speed, cutterhead torque, etc. Due to space limitations, these tables will not be detailed here, and the accompanying figures do not fully illustrate them.

[0111] Security Data Snowflake Model

[0112] like Figure 4 As shown in the figure, the series of foreign keys of the safety fact table are also the primary keys of its dimension table, which specifically include: roof management, rock burst prevention and control, water hazard prevention and control, fire prevention and control, gas prevention and control, dust prevention and control, heat hazard control, safety monitoring system, underground workers management, video monitoring, communication, risk classification management, accident hazard investigation and control, superior safety inspection, unsafe behavior management, accident management, safety training, occupational health, emergency management, environmental protection, etc.

[0113] The maintenance tables under the first level of fire prevention and control include: fire prevention and extinguishing management, basic fire monitoring information, bundle pipe monitoring system, etc.; the maintenance tables under the first level of gas prevention and control include: gas extraction monitoring system, gas extraction system, outburst prevention management, gas inspector inspection system, coal and gas outburst monitoring system, etc.; the maintenance tables under the first level of dust prevention and control include: online monitoring system, dust prevention and control management, etc.; the maintenance tables under the first level of heat damage control include: heat damage control implementation, heat damage monitoring, etc.

[0114] The dimensional tables under the second-layer gas extraction monitoring system include: basic parameters of basic analog quantities of the second-layer coal mining face roof, basic parameters of switch quantities, basic parameters of cumulative quantities, environmental monitoring parameters, extraction pump monitoring parameters, gas storage tank monitoring parameters, etc.; the dimensional tables under the second-layer gas extraction system include gas extraction inspection radius, gas extraction rate, gas extraction management system, three-defense devices, monitoring reports, certificates / identification materials, drawings and technical data, etc.

[0115] The dimension table under the third layer gas extraction investigation radius includes: coal seam permeability coefficient, borehole gas flow attenuation coefficient, effective extraction radius, etc. The dimension table under the third layer gas extraction rate includes: mine absolute gas emission volume Q (m 3 / min), mine gas extraction rate (%), absolute gas emission volume Q (m 3 The third-level gas extraction management system includes: detailed criteria for evaluating mine gas extraction compliance, gas extraction management and assessment and reward and punishment systems, extraction project inspection and acceptance systems, regular meetings prior to extraction, and technical file management systems. Due to space limitations, these are not detailed here, and the accompanying figures do not fully illustrate them.

[0116] Snowflake Model for Business Management Data

[0117] like Figure 5 As shown, the series of foreign keys of the business management fact table are also the primary keys of its dimension tables, which specifically include: human resources management, financial management, audit management, material management, equipment management, transportation and marketing management, energy conservation and emission reduction management, scientific and technological management, project management, legal management, comprehensive management, information management, etc.

[0118] The dimension tables under the first level of human resource management include: personnel management, human resource planning, recruitment management, training management, performance management, salary management, etc.; the dimension tables under the first level of financial management include: accounting, fund management, cost management, comprehensive budget management, financial analysis, etc.; the dimension tables under the first level of audit management include: audit plan, audit system, audit content, audit results, audit report preparation, internal and external audit coordination management, etc.

[0119] The dimension tables under the second-level capital management include: capital planning - plan preparation, capital planning - plan approval, capital allocation - fund transfer, capital allocation - fund allocation, capital settlement - payment management, capital settlement - collection management, capital deposit and loan - deposit management, capital deposit and loan - loan management, etc.; the dimension tables under the second-level cost management include: cost planning - basic information, cost planning - preparation basis, cost accounting - accounting method, cost assessment, cost control, etc.; the dimension tables under the second-level comprehensive budget management include: budget preparation, budget control, budget adjustment, budget analysis, budget assessment, etc.

[0120] The dimension tables for budget preparation at the third level include: preparation time, preparation cycle, and preparation method; the dimension tables for budget control at the third level include: budget item name, control standard, and implementation progress; the dimension tables for budget adjustment at the third level include: adjustment item name, adjustment reason, and adjustment range; the dimension tables for budget analysis at the third level include: budget item name, variance reason, and variance rate; and the dimension tables for budget assessment at the third level include: assessment indicators, completion rate, and responsibility center. Due to space limitations, this will not be discussed in detail, and the accompanying figures do not fully reflect this information.

[0121] In the present invention, the master data identification model in step S2 includes a master data identification index system and an evaluation model, and the evaluation model includes an AHP index weight calculation module, an entropy weight method index entropy weight calculation module, a combination index weight calculation module and a comprehensive evaluation module;

[0122] The master data identification indicator system includes primary indicators, secondary indicators, indicator definitions, and indicator scoring standards. The primary indicators include shareability, stability, accuracy, and importance. The secondary indicators corresponding to shareability are the number of shared business departments and system span. The secondary indicators corresponding to stability are data validity period and update frequency. The secondary indicators corresponding to accuracy are uniqueness and anomaly rate. The secondary indicators corresponding to importance are business priority and security classification. Each secondary indicator has a corresponding indicator definition and indicator scoring standard.

[0123] Among them, the master data identification indicator system is shown in Table 1.

[0124] Table 1: Description of identification indicators and scores in the master data identification indicator system

[0125]

[0126] With reference to the identification indicators and score descriptions in the master data identification indicator system, the importance of each indicator of the corresponding data item in one of the hierarchical dimension tables in the mine big data snowflake model library is assigned using a 1-9 evaluation scale to obtain the corresponding mine data indicator score data set; among them, the indicators of the data items include the number of shared business departments, system span, data validity period, update frequency, uniqueness, anomaly rate, business priority and / or security classification; among them, the 1-9 evaluation scale is obtained by Santy using the Delphi method, and the specific content is shown in Table 2 below.

[0127] Table 2: Evaluation scale from 1 to 9

[0128] scale meaning 1 Indicates that two elements have the same importance. 3 Indicates that compared with two elements, the former is slightly more important than the latter 5 Indicates that compared with two elements, the former is obviously more important than the latter 7 Indicates that compared with two elements, the former is extremely important than the latter 9 Indicates that compared with two elements, the former is more important than the latter 2，4，6，8 Indicates the middle value of the above adjacent judgment Countdown from 1 to 9 Indicates the importance of comparing the order of the corresponding two elements

[0129] In the present invention, the calculation method of the AHP (Analytical Hierarchy Process) index weight calculation module is:

[0130] S211. Construct a comparison matrix: Perform pairwise comparisons on mining data indicators (such as the number of shared business departments, system span, data validity period, update frequency, uniqueness, anomaly rate, business priority, and security classification) to establish a comparison matrix A, which is expressed as follows:

[0131]

[0132] Among them, A represents the comparison matrix, a ij is the value compared between data index elements i and j, and n is the total number of elements.

[0133] S212. Calculate indicator weights:

[0134] Calculate the sum S of each column of the comparison matrix j :

[0135]

[0136] By adding each value a ij Divide by the sum of its corresponding column S j , get the normalized value Constructing a normalized comparison matrix Its expression is as follows:

[0137]

[0138] Calculate the weight w of each data indicator element i , we get the weight vector W, which is expressed as follows:

[0139]

[0140] W=[w1,w2,w3…w i …,w n ](i=1,2,…,n);

[0141] S213, consistency check:

[0142] Calculate the maximum eigenvalue λ max , whose expression is:

[0143]

[0144] Calculate the consistency index (CI):

[0145]

[0146] Calculate the consistency ratio (CR):

[0147]

[0148] Among them, RI is the random consistency index; the random consistency index is usually the value shown in the following table (determined according to the order of the comparison matrix).

[0149] Table 3: Random consistency indicators

[0150] n RI 1 0 2 0 3 0.58 4 0.90 5 1.12 6 1.24 7 1.32 8 1.41 9 1.45

[0151] In the present invention, the calculation method of the entropy weight calculation module of the entropy weight method indicator is:

[0152] S221. Data Standardization:

[0153] The dimensions and numerical ranges of different indicators may be different, so it is necessary to standardize the mining data indicator score dataset. This application specifically uses the range normalization method to standardize the mining data indicator score dataset, and its expression is:

[0154]

[0155] Among them, x ef is the value of the fth indicator of the mine data item e, x' ef is the index value after standardization, min(x f ) is the minimum value of the f-th index, max(x f ) is the maximum value of the f-th index;

[0156] S222. Calculate the entropy value and entropy weight of each indicator:

[0157] According to each standardized indicator value, calculate its proportion p in the total of the indicator ef , whose expression is:

[0158]

[0159] Calculate the entropy value h of each indicator f , whose expression is:

[0160]

[0161] Where n is the number of samples;

[0162] Calculate the redundancy d f , whose expression is:

[0163] d f =1-h f #(10);

[0164] The higher the redundancy, the more important the indicator is and the more concentrated the information is;

[0165] Finally, the entropy weight w of each indicator is calculated by redundancy f, whose expression is:

[0166]

[0167] Among them, m is the number of indicators, and the weight value reflects the importance of each indicator in the evaluation;

[0168] In the present invention, the calculation method of the combined index weight calculation module is:

[0169] Combining the AHP indicator weight and the entropy weight method indicator entropy weight, the combined weight is calculated and the subjective and objective combined weighting is performed. The expression is:

[0170]

[0171] Among them, W z is the comprehensive weight of the indicator, w i is the weight of the data indicator element calculated by AHP, w f is the entropy weight calculated by the entropy weight method.

[0172] In the present invention, the calculation method of the comprehensive evaluation module is:

[0173] Calculate the comprehensive score of the data items corresponding to each dimension table. The expression is:

[0174]

[0175] Among them, s e is the comprehensive score of the e-th mine data item, w j is the combined weight, x ef is the value of the fth indicator of the eth mine data item;

[0176] Compare the comprehensive score with the pre-set judgment threshold. If the comprehensive score exceeds the judgment threshold, the corresponding data item will be included in the main data table;

[0177] By summarizing and establishing master data tables and their total evaluation scores, a mine big data master data system is constructed.

[0178] As a specific embodiment, step S2 specifically includes:

[0179] 1) Referring to the identification indicators and score descriptions in the master data identification indicator system, the importance of each indicator corresponding to the data items of the 922 third-level dimension tables in the mine big data snowflake model library was assigned using an evaluation scale of 1 to 9, and the third-level mine data indicator score data set was obtained, as shown in Table 4 below.

[0180] Table 4: 922 third-tier mine data indicator score datasets (partial)

[0181]

[0182]

[0183] 2) AHP method to calculate weights:

[0184] The comparison matrix A of the four first-level indicators is established as follows:

[0185]

[0186] Similarly, the comparison matrices A1, A2, A3, and A4 of the secondary indicators under the four primary indicators are established as follows:

[0187]

[0188] According to the comparison matrix A constructed above, Python is used to calculate the eigenvalues and eigenvectors of the matrix. Then, the eigenvectors are normalized to obtain the weights of the four first-level indicators. Then, a consistency test is performed according to the formula. From the test results, all the first-level indicators meet the requirements.

[0189] The weights of shareability, stability, accuracy, and importance are calculated to be W = [0.289723190.102980520.541788630.06550766]. There are four identification indicators in total, so the RI is 0.90. The calculated CR = 0.04930 < 0.90, indicating acceptable consistency.

[0190] Using the same steps and calculation methods for all secondary indicators, we obtain the following data, as shown in Table 5. A11, A12, A21, A22, A31, A32, A41, and A42 represent the number of business departments, system span, data validity period, update frequency, uniqueness, anomaly rate, business level, and security level, respectively.

[0191] Table 5: Secondary indicator weights

[0192] index Weight index Weight index Weight index Weight <![CDATA[A 11 ]]> 0.67 <![CDATA[A 21 ]]> 0.5 <![CDATA[A 31 ]]> 0.8 <![CDATA[A 41 ]]> 0.25 <![CDATA[A 12 ]]> 0.33 <![CDATA[A 22 ]]> 0.5 <![CDATA[A 32 ]]> 0.2 <![CDATA[A 42 ]]> 0.75

[0193] Summarize the weights of indicators at all levels and calculate the absolute weight of each secondary indicator. The absolute weight of each secondary indicator is the relative weight of the indicator to its primary indicator multiplied by the weight of the primary indicator. See Table 6 below for details.

[0194] Table 6: Summary of indicator weights

[0195]

[0196] 3) Entropy weight method to calculate entropy value and entropy weight:

[0197] Import the data in Table 4 and use Python to calculate the entropy value h of each indicator f And the entropy weight w of each indicator f , the results are shown in Table 7 below.

[0198] Table 7: Entropy value and entropy weight of each indicator

[0199]

[0200]

[0201] 4) AHP combined entropy weight method to determine the combined weight

[0202] Calculate the combined weights, assign subjective and objective weights, and use the weights obtained by the AHP method as follows:

[0203] w i =[0.194114537 0.095608653 0.05149026 0.05149026 0.4334309040.108357726 0.016376915 0.049130745],

[0204] Entropy weight obtained by entropy weight method:

[0205] w f =[0.108260403 0.092708769 0.131908635 0.116242261 0.2225932480.113156704 0.118421095 0.096708886],

[0206] Substituting into formula (12) we get the combined weight:

[0207] W z =[0.15534182 0.10088661 0.08831281 0.08290278 0.3328438 0.118657350.04719056 0.07386427].

[0208] 5) Calculate the comprehensive score of the data

[0209] Calculate the comprehensive score s of all third-level dimension table data items through the comprehensive evaluation module e(Formula 13) compares the comprehensive score with the pre-set judgment threshold to achieve accurate classification of mine master data. In a specific example, the judgment threshold is set to "7", and the third-level dimension table data items with a comprehensive score exceeding the judgment threshold "7" are included in the master data table. For example, taking the mine data "reserve increase, decrease and transfer" as an example, its number of shared business departments, system span, data validity period, update frequency, uniqueness, anomaly rate, business priority, and security classification index value is [866610266]. Its comprehensive score is calculated through the mine master data identification model as follows:

[0210] s e =8*0.15534182+6*0.10088661+6*0.08831281+6*0.08290278+10*0.3328438+2*0.11865735+6*0.04719056+6*0.07386427=7.16742944>7;

[0211] Its comprehensive score is greater than the judgment threshold of "7", so the mine data "reserve increase and decrease conversion" is included in the main data table to better utilize the data, optimize the production process, improve production efficiency, reduce costs, and support decision-making.

[0212] 6) Master data table and its evaluation score summary

[0213] Through the master data identification model, the master data of the data items in the third-level dimension table in the mine big data snowflake model library are identified, and finally a mine master data application system is built. The intelligent mine master data table and its total score are obtained as shown in the table below.

[0214] Table 8: Master data table and its total evaluation score

[0215]

[0216] Constructing a mine master data application system through the method of the present invention is helpful to comply with regulations, improve safety, and improve environmental sustainability; the present invention incorporates mine master data into the master data table, on the one hand, can ensure that the key data of the mine remains consistent and accurate throughout the enterprise, promote collaborative communication, reduce information islands and information asymmetry, and effectively avoid data inconsistency or errors leading to misunderstandings or decision-making errors; in addition, through master data management, the current situation can be better analyzed and trends can be predicted in the decision-making process, equipment, production plans, etc. can be more effectively managed, production processes can be optimized, downtime can be reduced, equipment utilization and production efficiency can be improved, resource utilization and production scheduling can be optimized, and production costs and management costs can be reduced; on the other hand, it helps enterprises to respond to market changes more quickly and enhance market competitiveness. At the same time, it helps enterprises comply with regulations, reduce the risk of safety accidents, improve employees' safety awareness, reduce resource waste and environmental load, and promote the sustainable development of mines.

[0217] An electronic device, comprising: a memory storing a computer program for the aforementioned method for constructing a master data application system for intelligent mines;

[0218] A processor is used to retrieve and execute the computer program stored in the memory.

[0219] A computer-readable storage medium stores a computer program, which, when executed, implements the aforementioned method for constructing a master data application system for intelligent mines.

[0220] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A method for constructing a master data application system for intelligent mines, characterized in that: The steps include: S1. Based on the big data dimension model and data granularity identification theory, the mining big data is dimensionally modeled and a fact table system based on the core business of the mine is constructed. The corresponding series dimension tables are constructed through the fact table system, and the hierarchical structure of the series dimension tables is determined based on the data granularity theory and the snowflake model. Finally, a multi-level structured mining big data snowflake model library is constructed. S2. Based on the constructed mine big data snowflake model library and the pre-designed master data identification model, the mine master data and its total evaluation score are identified from the data items corresponding to the dimension table of one level to form a mine big data master data system. S3. Develop a mine master data application system based on the mine big data snowflake model library to provide support for the query, application and promotion of mine master data.

2. The method for constructing a master data application system for intelligent mines according to claim 1, characterized in that: The specific steps of step S1 are: S101. Before building the series dimension table, clarify the granularity of the fact table. Each fact table is based on business events and also contains multiple descriptive attributes related to the business events. Each descriptive attribute is used as a series foreign key of the fact table. The series foreign key is used to point to the corresponding series dimension table. S102. Define a hierarchical structure for a series dimension table based on business requirements and data granularity. The series dimension table includes dimension tables at multiple levels. The series foreign key of the fact table serves as the primary key of the dimension table at the adjacent level. Each dimension table also includes multiple descriptive attributes related to its primary key, namely, the foreign keys of the dimension table. Each foreign key of the dimension table corresponds to related management data. S103. According to the complexity of the dimension, the dimension table is refined, and the foreign key of the dimension table of the previous level is used as the primary key of the dimension table of the next level. This process is repeated to form a snowflake data model. By integrating various snowflake data models, a multi-level structured mining big data snowflake model library is constructed.

3. The method for constructing a master data application system for intelligent mines according to claim 2, characterized in that: The fact table system includes: a geological fact table, a production fact table, a safety fact table and / or an operation and management fact table; The mine big data snowflake model library includes: a geological data snowflake model, a production data snowflake model, a safety data snowflake model and / or an operation and management data snowflake model.

4. The method for constructing a master data application system for intelligent mines according to claim 1, characterized in that: The master data identification model in step S2 includes a master data identification index system and an evaluation model, wherein the evaluation model includes an AHP index weight calculation module, an entropy weight method index entropy weight calculation module, a combination index weight calculation module, and a comprehensive evaluation module; The master data identification indicator system includes primary indicators, secondary indicators, indicator definitions, and indicator scoring standards. The primary indicators include shareability, stability, accuracy, and importance. The secondary indicators corresponding to shareability are the number of shared business departments and system span. The secondary indicators corresponding to stability are data validity period and update frequency. The secondary indicators corresponding to accuracy are uniqueness and anomaly rate. The secondary indicators corresponding to importance are business priority and security classification. Each secondary indicator has a corresponding indicator definition and indicator scoring standard. Based on the master data identification indicator system, the importance of each indicator corresponding to the data item in one of the hierarchical dimension tables in the mine big data snowflake model library is assigned to obtain the corresponding mine data indicator score data set.

5. The method for constructing a master data application system for intelligent mines according to claim 4, characterized in that: The calculation method of the AHP indicator weight calculation module is: S211. Construct a comparison matrix: perform pairwise comparisons on the mining data indicators and establish a comparison matrix A, which is expressed as: Among them, A represents the comparison matrix, a ij is the value compared between data indicator elements i and j, and n is the total number of data indicator elements. S212. Calculate indicator weights: Calculate the sum S of each column of the comparison matrix j : By adding each value a ij Divide by the sum of its corresponding column S j , get the normalized value Constructing a normalized comparison matrix Its expression is as follows: Calculate the weight w of each data indicator element i , we get the weight vector W, which is expressed as follows: W=[w1,w2,w3…w i …,w n ](i=1,2,…,n); S213, consistency check: Calculate the maximum eigenvalue λ max , whose expression is: Calculate the consistency index (CI): Calculate the consistency ratio (CR): Among them, RI is the random consistency index.

6. The method for constructing a master data application system for intelligent mines according to claim 5, characterized in that: The calculation method of the entropy weight calculation module of the entropy weight method indicator is: S221. Data Standardization: The range standardization method is used to standardize the mine data index score data set, and its expression is: Among them, x ef is the value of the fth indicator of the mine data item e, x' ef is the index value after standardization, min(x f ) is the minimum value of the f-th index, max(x f ) is the maximum value of the f-th index; S222. Calculate the entropy value and entropy weight of each indicator: According to each standardized indicator value, calculate its proportion p in the total of the indicator ef , whose expression is: Calculate the entropy value h of each indicator f , whose expression is: Where n is the number of mine data samples; Calculate the redundancy d f , whose expression is: d f =1-h f #(10); The higher the redundancy, the more important the indicator is and the more concentrated the information is; Finally, the entropy weight w of each indicator is calculated by redundancy f , whose expression is: Among them, m is the number of indicators, and the entropy weight reflects the importance of each indicator in the evaluation.

7. The method for constructing a master data application system for intelligent mines according to claim 6, characterized in that: The calculation method of the combined indicator weight calculation module is: Combining the AHP indicator weight and the entropy weight method indicator entropy weight, the combined weight is calculated and the subjective and objective combined weighting is performed. The expression is: Among them, W z is the comprehensive weight of the indicator, w i is the weight of the data indicator element calculated by AHP, w f is the entropy weight calculated by the entropy weight method.

8. The method for constructing a master data application system for intelligent mines according to claim 7, characterized in that: The calculation method of the comprehensive evaluation module is: Calculate the comprehensive score of the data items corresponding to each dimension table. The expression is: Among them, s e is the comprehensive score of the e-th mine data item, w j is the combined weight, x ef is the value of the fth indicator of the eth mine data item; Compare the comprehensive score with the pre-set judgment threshold. If the comprehensive score exceeds the judgment threshold, the corresponding data item will be included in the main data table; By summarizing and establishing the master data table and its total evaluation score, a mine big data master data system is formed.

9. An electronic device, characterized in that: include: A memory storing a computer program of the method for constructing a master data application system for an intelligent mine according to any one of claims 1 to 8; A processor is used to retrieve and execute the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the method for constructing a master data application system for intelligent mines as described in any one of claims 1 to 8 is implemented.

Images