Coal mine enterprise production data preprocessing method and system

Abstract

The invention discloses a coal mine enterprise production data preprocessing method and system, and the method comprises the steps: obtaining first production data based on all data source management units in a coal mine enterprise; and performing data cleaning on the first production data to generate second production data. And data reduction is carried out on the second production data, data conversion is carried out after data reduction is completed, and third production data is generated. And performing data integration and storage on the third production data to obtain a production database of the coal mine enterprise. According to the coal mine enterprise production data preprocessing method and system provided by the invention, data reduction is performed on the data acquired by different enterprises, the data complexity is reduced by reducing the data dimension, the data scale is reduced, and the value density of the data is improved. And based on rule conversion or metadata conversion, performing data conversion on the data subjected to data reduction, unifying data types, and facilitating data collection. And scientific monitoring, prediction and work arrangement of coal mine enterprises are facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial production data processing, and in particular to a method and system for preprocessing production data of a coal mining enterprise. Background Art

[0002] With the popularization of data analysis methods based on big data, various industrial enterprises have also begun to use big data analysis methods to help enterprises analyze and predict various aspects of operations. In view of this, the value of data itself is increasingly recognized by various enterprises, and the data quality of various types of data obtained is also valued by various enterprises.

[0003] In order to improve the data quality of data obtained by various enterprises, pre-processing methods are usually used to improve the availability of data by screening and deleting abnormal data. However, for coal mining enterprises, due to the particularity of their production process, in the continuous advancement of the production process, the data timeliness is high, and the common pre-processing methods are still unable to handle the large differences in data between enterprises. The collected data itself has many defects and the data quality is uneven. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: for coal mining enterprises, due to the particularity of their production process, in the continuous advancement of the production process, the data timeliness is relatively high, and the common preprocessing method is still unable to handle the large data differences between enterprises. The collected data itself has various defects and the data quality is uneven.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: A method for preprocessing production data of a coal mining enterprise, comprising: obtaining first production data based on all data source management units in the coal mining enterprise.

[0007] The first production data is cleaned to generate second production data.

[0008] Data reduction is performed on the second production data, and after the data reduction is completed, data conversion is performed to generate third production data.

[0009] The third production data is integrated and stored to obtain a production database of the coal mining enterprise.

[0010] As a preferred solution of the coal mining enterprise production data preprocessing method described in the present invention, wherein: the first production data includes unit production data corresponding to each data source management unit.

[0011] As a preferred solution of the coal mining enterprise production data preprocessing method described in the present invention, wherein: the data cleaning of the first production data to generate the second production data includes inconsistency detection, noise identification, data filtering and correction of the first production data to complete data cleaning.

[0012] Inconsistent data is detected through rule matching, and metadata is eliminated through inconsistency detection.

[0013] For the first production data, a prediction algorithm is set, possible values ​​based on the prediction algorithm are determined, and the possible values ​​are used for filling.

[0014] The production data after data cleaning is used as the second production data.

[0015] As a preferred solution of the coal mining enterprise production data preprocessing method of the present invention, wherein: the data reduction is performed on the second production data, and after the data reduction is completed, the data conversion is performed to generate the third production data, including, for the second production data, determining a key features corresponding to the data. The key feature is an n-dimensional orthogonal vector.

[0016] Key features are sorted according to their importance and key features whose importance is lower than a preset threshold are deleted.

[0017] The deleted data is projected in a low-dimensional space, where the dimension of the low-dimensional space is n minus the deleted key features.

[0018] The reduced second production data is divided into multiple sets, and the multiple sets are clustered into b non-overlapping clusters, and the b non-overlapping clusters are sampled to obtain c clusters.

[0019] Based on rule conversion, the production data corresponding to the c clusters are converted into a unified data type to obtain third production data.

[0020] As a preferred solution of the coal mining enterprise production data preprocessing method described in the present invention, wherein: for the second production data, determining a key features corresponding to the data includes determining an initial key feature set based on the second production data, setting key feature judgment conditions, and calculating the relevance of the key features in the second production data to this type of data.

[0021] The key features with the highest correlation in the initial key feature set are added to the candidate key feature set, and the information increment of the candidate key feature set is determined.

[0022] According to the similarity of each key feature in the candidate key feature set and the preset judgment condition, it is judged whether the information increment of the candidate key feature set meets the requirement. If it does not meet the requirement, it is judged that the information increment of the candidate key feature set meets the requirement, and the candidate key feature set is determined as a key feature corresponding to the second production data. If it meets the requirement, it is judged that the information increment of the candidate key feature set does not meet the requirement, and the initial key feature set is re-determined.

[0023] As a preferred solution of the coal mining enterprise production data preprocessing method of the present invention, wherein: judging whether the information increment of the candidate key feature set meets the requirements according to the similarity of each key feature in the candidate key feature set and the preset judgment condition includes: the judgment condition for whether the information increment of the key feature set meets the requirements is expressed as:

[0024]

[0025] Among them, S b represents the similarity of the candidate key feature set before deleting the current feature vector, S a It represents the similarity of the candidate key features after deleting the current feature vector, S t Indicates the preset similarity threshold.

[0026] As a preferred solution of the coal mine enterprise production data preprocessing method of the present invention, wherein: the data integration and storage of the third production data to obtain the production database of the coal mine enterprise includes generating the production database of the coal mine enterprise based on the third production data. The production database is a data cube.

[0027] A coal mining enterprise production data preprocessing system, characterized by: comprising:

[0028] The collection module obtains the first production data based on all data source management units in the coal mining enterprise.

[0029] The cleaning module cleans the first production data to generate second production data.

[0030] The reduction and conversion module performs data reduction on the second production data, and performs data conversion after completing the data reduction to generate third production data.

[0031] The database generation module integrates and stores the third production data to obtain the production database of the coal mining enterprise.

[0032] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0033] A computer-readable storage medium stores a computer program, which implements the steps of the method described above when executed by a processor.

[0034] Beneficial effects of the present invention: The present invention improves data quality, improves data reliability, and also improves data consistency, accuracy, and availability. Data reduction is performed on data obtained from different enterprises, and the complexity of the data is reduced by reducing the data dimension, which reduces the data size and increases the value density of the data. Based on rule conversion or metadata conversion, the data that has completed data reduction is converted to unify data types for easy data collection. A unified database facilitates subsequent analysis based on big data technology, which helps coal mining enterprises to conduct scientific monitoring, prediction, and work arrangements. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0036] Figure 1 An overall flow chart of a method and system for preprocessing production data of a coal mining enterprise provided in the first embodiment of the present invention.

[0037] Figure 2 A flowchart of a method for preprocessing production data of a coal mining enterprise and a data reduction and data conversion method of the system is provided for the first embodiment of the present invention.

[0038] Figure 3 A flowchart of a method for preprocessing production data of a coal mining enterprise and a method for determining key features based on information increment in a system is provided as a first embodiment of the present invention.

[0039] Figure 4 A structural block diagram of a coal mine enterprise production data preprocessing system of a coal mine enterprise production data preprocessing method and system provided in a third embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0041] Example 1, reference Figure 1 to Figure 3 , is an embodiment of the present invention, and provides a method for preprocessing production data of a coal mining enterprise, comprising:

[0042] S101: Acquire first production data based on all data source management units in the coal mining enterprise.

[0043] Specifically, the first production data at least includes unit production data corresponding to each data source management unit.

[0044] Generally, coal enterprises are large in size, have complex businesses, and have a larger corporate structure than ordinary enterprises. Therefore, they will set up corresponding data source management units in different business directions to process the data obtained in a certain business direction.

[0045] S102: Cleaning the first production data to generate second production data.

[0046] Specifically, the first production data is subjected to inconsistency detection, noise identification, data filtering and correction to complete data cleaning. Inconsistent data is detected by rule matching, and metadata is eliminated by inconsistency detection. For the first production data, a prediction algorithm is set to determine possible values ​​based on the prediction algorithm, and the possible values ​​are used for filling.

[0047] The production data after data cleaning is used as the second production data.

[0048] It should be noted that inconsistent data is detected through rule matching, and matching rules are constructed by setting at least one rule condition, such as the order of different features in the data content, the corresponding units of the data, the characteristic symbols carried by different data features, etc. The required accuracy level can also be set for different rule conditions, such as selecting the required accuracy level from low (30%), medium (60%), high (80%) and precise (100%).

[0049] Furthermore, clustering algorithms such as DBSCAN (Density-Based Spatial Clustering of Applications with Noise) can be used to detect data inconsistencies and identify noise points. Abnormal data can also be detected based on the Isolation Forest algorithm.

[0050] By setting the prediction algorithm, determining the possible values ​​based on the prediction algorithm, and using the possible values ​​for filling, the knn algorithm can be used for nearest neighbor filling, that is, the knn algorithm is fitted with non-missing data, and the continuous features are weighted averaged, and the discrete features are weighted voted to predict the data at the target location. It is also possible to use the data fitting model based on the random forest algorithm to predict the missing data to complete the data filling. It is also possible to perform multiple interpolation of data based on repeated simulations, that is, generate multiple filling values ​​for the missing target location, and form a corresponding complete filling data set based on different filling values. Statistical analysis is performed on each complete data set, and each filled data set is integrated to generate the final filled data set to complete the data filling.

[0051] S103: Perform data reduction on the second production data, and perform data conversion after the data reduction is completed to generate third production data.

[0052] It should be noted that the verification data (Valid Data) is determined by the scenario user, or the verification data is obtained by sampling from a data set.

[0053] S104: Integrate and store the third production data to obtain a production database of the coal mining enterprise.

[0054] Specifically, the production database of the coal mining enterprise is generated based on the third production data. The production database is a conventional database or a data cube.

[0055] Among them, data cubes are more suitable for parallel computing architectures and distributed file systems. Users can choose an appropriate database model based on the type of analysis system for subsequent big data analysis, and there is no limitation here.

[0056] Figure 2 A schematic flow chart of a data reduction and data conversion method according to an embodiment of the present invention is shown.

[0057] S201: For the second production data, determine a key features corresponding to the data.

[0058] Specifically, the key feature is an n-dimensional orthogonal vector.

[0059] S202: Sort the key features according to their importance, and delete the key features whose importance is lower than a preset threshold.

[0060] It should be noted that the heuristic algorithm for compressing the search space, i.e. the greedy algorithm, is used to make the best selection locally. By establishing a reduction set of empty key features, the best key features are selected and added to the reduction set using the above algorithm, and then iterated. In each iteration, the best key features among the remaining key features are added to the reduction set.

[0061] After completing the feature collection of the specification set, gradually remove the worst key features currently still in the specification set.

[0062] in,

[0063] By using consistency measurement, the minimum key features can be determined to describe the second production data, and the key features that will not cause contradictions can be found. After that, the number of inconsistencies generated by each dimension based on the target key features is calculated, that is, the second production data is described by each key feature, and compared with the description of the second production data by the target key feature, and the number of inconsistencies in each group of comparisons is checked. The key features with fewer inconsistencies are more important.

[0064] For example, for the key feature of electricity consumption, among the four key features of the number of people going down the mine, time period, duration, and number of mine closures, if the number of inconsistencies generated after comparing the number of people going down the mine with the electricity consumption is the least, then the key feature of the number of people going down the mine is the most important; if the number of inconsistencies generated after comparing the number of mine closures with the electricity consumption is the largest, then the key feature of the number of mine closures is the least important. Based on this, the key features to be deleted are determined by using the number of inconsistencies as a preset threshold. Alternatively, through correlation measurement, for the key features of the target, the correlation between other key features and the key features of the target is calculated to determine the strength of the correlation between each key feature and the key features of the target, and the correlation is used as the preset threshold to determine the key features to be deleted.

[0065] S203: Projecting the deleted data in a low-dimensional space.

[0066] It should be noted that the dimension of the low-dimensional space is n minus the number of deleted key features. For example, if the number of key features deleted in step S202 is d, the dimension of the low-dimensional space is nd.

[0067] S204: Divide the second production data after dimension reduction into multiple sets, and cluster the multiple sets into b non-overlapping clusters.

[0068] S205: Sampling b disjoint clusters to obtain c clusters.

[0069] It should be noted that the sampling method can be stratified sampling, systematic sampling, cluster sampling or random sampling.

[0070] S206: Based on rule conversion or metadata conversion, the production data corresponding to the c clusters are converted into a unified data type to obtain third production data.

[0071] Furthermore, according to the above method, by determining the key features in the production data, deleting the parts whose importance is lower than the threshold, and using the projection of the low-dimensional space to generate the corresponding new set, dimensionality reduction can be completed, reducing the dimension, complexity and data scale of the data. At the same time, the value density of the data is improved, the difficulty of further data analysis is reduced, and the analysis efficiency is improved. In addition, by sampling the clusters generated by clustering, the data scale is further streamlined, while fully retaining the core content of the data, the data volume is effectively reduced, and it is more convenient to save data without affecting the application of data analysis.

[0072] Figure 3 A schematic flow chart of a method for determining key features based on information increment according to an embodiment of the present invention is shown.

[0073] S301: Determine an initial key feature set according to the second production data, and set key feature determination conditions.

[0074] Specifically, the judgment condition for whether the information increment of the key feature set meets the requirements is

[0075]

[0076] Among them, S b is the similarity of the candidate key feature set before deleting the current feature vector. a is the similarity of the candidate key features after deleting the current feature vector. S t is the preset similarity threshold.

[0077] Furthermore, for example, Table 1 shows the data corresponding to the key characteristics of three coal mining companies in the past three months.

[0078] Table 1 Data corresponding to the key characteristics of three coal mining companies in the past three months

[0079]

[0080] Among them, the working status of coal mining enterprises is taken as the key feature of the target, and then the other key features are divided into three levels based on historical experience. The monthly power consumption is divided into 0-500000 (kw·h), 500001-1000000 (kw·h), 1000001-1500000 (kw·h). The daily average power consumption of a single mine is divided into 7000-12000 (kw·h), 12001-17000 (kw·h), 17001-22000 (kw·h). The number of people going down the mine per day is divided into 0-20 people, 21-40 people, and 41-60 people. The number of working mines in the month is divided into 3, 2, and 1. The daily average number of people going down the mine in a single mine is divided into 11-15 people. 16-20 people. 21-25 people. If the daily average power consumption of a single mine is used as the key feature to determine whether a coal mining enterprise has violated the regulations and engaged in overproduction, then in the data for each month, the number of times each key feature is in the same level as the daily average power consumption of a single mine is calculated. If the monthly power consumption is 0, the number of people going down the mine per day is 0, the number of working mines in that month is 1, and the average number of people going down the mine per day is 3.

[0081] The product of the above ratio and 100% is S b , S a Then it is 100%, S t is 100%.

[0082] Based on the above judgment conditions, it can be known that the monthly power consumption 0 = S b ≤S a =1, number of people working underground per day 0 = S b ≤S a =1, the number of working mines in that month 1 = S b ≤S a =1, average number of people going down to a single mine per day 3 = S b >S a = 1. That is, the average number of people going down the mine in a single mine per day meets the judgment conditions of key characteristics, the information increment meets the requirements, and the rest do not meet the requirements.

[0083] S302: Calculate the relevance of the key feature in the second production data to the data of this type.

[0084] S303: adding the key features with the highest correlation in the initial key feature set to the candidate key feature set, and determining the information increment of the candidate key feature set.

[0085] S304: judging whether the information increment of the candidate key feature set meets the requirement according to the similarity of each key feature in the candidate key feature set and the preset judgment condition.

[0086] If the requirement is not met, then step S305 is executed. If the requirement is met, then step S306 is executed.

[0087] S305: Determine whether the information increment of the candidate key feature set meets the requirement, and determine the candidate key feature set as a key features corresponding to the second production data.

[0088] S306: Determine whether the information increment of the candidate key feature set does not meet the requirement.

[0089] Re-execute step S301.

[0090] It should be noted that according to the above method, the initial key feature set can be determined from the data that has completed data cleaning, and its relevance to the data of this type can be calculated to determine the information increment of the feature set with the highest relevance, and whether the information increment meets the requirements based on the judgment conditions, so as to determine the key feature set, and further complete the dimensionality reduction of the data, more accurately select the key feature set that can better reflect the data characteristics and is more in line with the meaning of the data, and make the dimensionality reduction based on this better retain the core dimensions of the data, better and more scientifically retain the data value, and reduce the complexity of the data. At the same time, in the feature extraction process, by using the data increment for screening, information redundancy is avoided, and the space cost and time cost of data storage are effectively controlled.

[0091] The computer device may be a server. The computer device includes a processor, a memory, an input / output interface (I / O for short) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data cluster data of a power monitoring system. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method is implemented.

[0092] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0093] Example 2 is an embodiment of the present invention, which provides a method and system for preprocessing production data of a coal mining enterprise. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0094] First, multiple data sources of a coal mining enterprise, including mining equipment, transportation systems, and environmental monitoring equipment, were selected to collect the first production data. These data cover information such as production time, output, equipment status, etc., with a total of about 10,000 records. Existing technologies use traditional data cleaning methods, including manual filtering and simple statistical analysis, which often leads to data loss and information inconsistency.

[0095] When cleaning the first production data, the method of the present invention is used to generate the second production data through inconsistency detection, noise identification and data filtering. The specific steps include: 1) using a rule matching algorithm to detect inconsistencies in the data. 2) performing metadata elimination to ensure data integrity. 3) applying a prediction algorithm to fill in missing values.

[0096] Data reduction is performed on the second production data, a key features are selected, and low-importance features are deleted. After feature selection and sorting using n-dimensional orthogonal vectors, the data is divided into multiple non-overlapping clusters for sampling, and finally the third production data is generated.

[0097] Finally, the third production data is integrated to construct a production database of the coal mining enterprise. The comparison of the experimental results of the present invention and the prior art is shown in Table 2.

[0098] Table 2 Experimental results

[0099]

[0100] It can be seen from the experimental data table that the present invention has shown significant advantages in many aspects. First, from the perspective of record cleaning effect, the prior art only retains about 70%-78% of the data records after processing, while the present invention can effectively clean up to about 92%-95%. This difference shows that the method of the present invention significantly improves the data retention rate and ensures higher quality data for subsequent analysis.

[0101] In terms of feature reduction, the number of key features in the prior art is kept between 19 and 22, while the present invention retains only 12 to 15 key features through effective feature selection. This feature simplification process not only reduces the consumption of computing resources, but also improves the efficiency of model training and avoids the overfitting problem caused by redundant features.

[0102] In terms of data integration time, the existing technology takes between 4 and 6 hours, while the processing time of the present invention is shortened to 1.5 to 2 hours. This time difference significantly improves the efficiency of data processing, allowing coal mining companies to respond to production needs more quickly and optimize the decision-making process.

[0103] In summary, through comparative data analysis, it can be seen that the coal mining enterprise production data preprocessing method of the present invention has obvious innovation and advantages in improving data cleaning efficiency, reducing key feature redundancy and shortening data integration time. These effects will play a positive role in promoting the production management and intelligent transformation of coal mining enterprises, showing the practicality and market potential of the present invention.

[0104] Embodiment 3, as Figure 4, is an embodiment of the present invention, and provides a coal mining enterprise production data preprocessing system, the system comprising: a collection module 401, a cleaning module 402, a reduction conversion module 403 and a database generation module 404. Among them, the collection module 401 is used to obtain the first production data based on all data source management units in the coal mining enterprise.

[0105] The cleaning module 402 is used to construct a sample value function and a feature value function, and initialize the sample value function and the feature value function.

[0106] The reduction and conversion module 403 is used to perform data reduction on the second production data, and perform data conversion after completing the data reduction to generate third production data.

[0107] The database generation module 404 is used to integrate and store the third production data to obtain a production database of the coal mining enterprise.

[0108] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for preprocessing production data of a coal mining enterprise, characterized in that: include: Based on all data source management units in the coal mining enterprise, first production data is obtained; Performing data cleaning on the first production data to generate second production data; Performing data reduction on the second production data, and performing data conversion after completing the data reduction to generate third production data; The third production data is integrated and stored to obtain a production database of the coal mining enterprise.

2. The method for preprocessing production data of a coal mining enterprise according to claim 1, characterized in that: The first production data includes unit production data corresponding to each data source management unit.

3. The method for preprocessing production data of a coal mining enterprise according to claim 2, characterized in that: The performing data cleaning on the first production data to generate the second production data includes performing inconsistency detection, noise identification, and data filtering and correction on the first production data to complete data cleaning; Detect inconsistent data through rule matching and eliminate metadata through inconsistency detection; For the first production data, a prediction algorithm is set, possible values ​​based on the prediction algorithm are determined, and the possible values ​​are filled in; The production data after data cleaning is used as the second production data.

4. The method for preprocessing production data of a coal mining enterprise according to claim 3, characterized in that: The performing data reduction on the second production data and performing data conversion after completing the data reduction to generate the third production data includes, for the second production data, determining a key features corresponding to the data; The key feature is the n-dimensional orthogonal vector; Sort key features by their importance and delete key features whose importance is lower than a preset threshold; Project the deleted data into a low-dimensional space whose dimension is n minus the deleted key features; The reduced second production data is divided into multiple sets, and the multiple sets are clustered into b non-overlapping clusters, and the b non-overlapping clusters are sampled to obtain c clusters; Based on rule conversion, the production data corresponding to the c clusters are converted into a unified data type to obtain third production data.

5. The method for preprocessing production data of a coal mining enterprise according to claim 4, characterized in that: Determining the a key features corresponding to the second production data includes determining an initial key feature set according to the second production data, setting key feature determination conditions, and calculating the relevance of the key features in the second production data to the data of this type; The key features with the highest correlation in the initial key feature set are added to the candidate key feature set, and the information increment of the candidate key feature set is determined; According to the similarity of each key feature in the candidate key feature set and the preset judgment condition, it is judged whether the information increment of the candidate key feature set meets the requirements; If the requirement is not met, it is determined that the information increment of the candidate key feature set meets the requirement, and the candidate key feature set is determined to be a key features corresponding to the second production data; If the requirements are met, it is determined that the information increment of the candidate key feature set does not meet the requirements, and the initial key feature set is re-determined.

6. The method for preprocessing production data of a coal mining enterprise according to claim 5, characterized in that: The step of judging whether the information increment of the candidate key feature set meets the requirement according to the similarity of each key feature in the candidate key feature set and the preset judgment condition includes: the judgment condition for whether the information increment of the key feature set meets the requirement is expressed as: Among them, S b represents the similarity of the candidate key feature set before deleting the current feature vector, S a It represents the similarity of the candidate key features after deleting the current feature vector, S t Indicates the preset similarity threshold.

7. The method for preprocessing production data of a coal mining enterprise according to claim 6, characterized in that: The step of integrating and storing the third production data to obtain a production database of the coal mining enterprise includes generating the production database of the coal mining enterprise based on the third production data; The production database is a data cube.

8. A coal mining enterprise production data preprocessing system using the method according to any one of claims 1 to 7, characterized in that: The acquisition module acquires the first production data based on all data source management units in the coal mining enterprise; A cleaning module, cleaning the first production data to generate second production data; a reduction conversion module, performing data reduction on the second production data, and performing data conversion after completing the data reduction to generate third production data; The database generation module integrates and stores the third production data to obtain the production database of the coal mining enterprise.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images