Time series data storage system and method for intelligent mine based on industrial Internet of Things

Abstract

The invention relates to the technical field of time series data storage, and particularly discloses a time series data storage system for an intelligent mine based on industrial Internet of Things, which comprises a data acquisition module for respectively acquiring time series sensing data of various operation equipment efficiencies of each operation type in a white shift and a night shift on a workday, transmitting the data and the corresponding operator scheduling time sequence data to a central database for preprocessing so as to generate operation data; according to the invention, an embedding method based on deep learning is used, and an embedding layer is designed to map high-dimensional discrete data, i.e., scheduling of information of each person in different time periods, into a low-dimensional continuous vector space, so that the efficiency of the operation data is improved. And a deep learning model is used for learning, embedding and representing and establishing association with the basic efficiency to which the equipment operated by the personnel information belongs, so that the combination with different equipment under different scheduling information is greatly improved, and the overall working efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field related to time series data storage, and in particular to a time series data storage system and method based on an industrial Internet of Things smart mine. Background Art

[0002] Smart mine time-series data storage systems are primarily designed to efficiently manage and analyze the large amounts of time-series data generated during mining operations. The core purpose of these systems is to handle the massive amounts of data transmitted in real time by various sensors and monitoring equipment in smart mines. This data often includes timestamps and reflects time-varying information such as environmental monitoring, equipment status, and production processes. However, existing mine time-series data storage systems are unable to rationally schedule work and equipment operations based on equipment and personnel efficiency, resulting in low personnel and equipment efficiency.

[0003] Therefore, it is necessary to design a time series data storage system and method based on industrial Internet of Things smart mines to solve the technical problems of low work efficiency of personnel and equipment in the existing technology. Summary of the Invention

[0004] In view of this, the present invention proposes a time series data storage system and method based on the industrial Internet of Things smart mine, which aims to solve the technical problem that the existing mine time series data storage system in the prior art cannot reasonably arrange personnel scheduling and equipment operation status according to equipment work efficiency and personnel work efficiency during use, resulting in low personnel work efficiency and equipment work efficiency.

[0005] In one aspect, the present invention proposes a time series data storage system based on an industrial Internet of Things (IIoT) smart mine, comprising: a data acquisition module that collects time series perception data on the efficiency of various operating equipment during day and night shifts for each operation type on weekdays, as well as corresponding operator scheduling time series data, and transmits the data to a central database for preprocessing to generate operation data;

[0006] The data analysis module encodes and sorts the efficiency of each workday's operation data. In the database, it establishes corresponding indexes based on the time-series perception data of the efficiency of various operating equipment during the day and night shifts for each operation type. It also analyzes the basic efficiency data of the equipment during the day or night shift when different personnel are using it. Using a deep learning-based embedding method, an embedding layer is designed to map the high-dimensional discrete data, namely the schedule of each personnel information at different time periods, into a low-dimensional continuous vector space. The deep learning model is then used to learn the embedding representation and establish an association with the basic efficiency of the equipment operated by the personnel information.

[0007] A distributed storage module automatically partitions and stores the time-series-aware data on different nodes according to the coding order, and copies the basic efficiency data to a non-volatile memory and updates it regularly to ensure data integrity and high availability;

[0008] The indexing module uses structured query language to query and index the central database to obtain the corresponding personnel scheduling plans and equipment efficiency plans when the preset efficiency standards are met, and generate operation reports.

[0009] Specifically, in the data acquisition module, letter codes are set for the data of each of the operation types, including drilling ZO and blasting BO, loading AO and transportation YO, crushing PO and screening SO, grinding MO and beneficiation KO, tailings treatment WO and wastewater management FO, ventilation TO and drainage RO, and equipment maintenance records WH and repair records XL. Data acquisition is achieved through sensors, and a collection cycle is set for the data of each operation type according to a time series score, which quantifies multiple characteristics of each operation type.

[0010] Specifically, data is collected on the work efficiency of personnel on working days for each job type, including personnel efficiency data collection. By using the work time management system to calculate personnel working hours and rest time, marking day shifts and night shifts separately, the production volume and work task completion status of each shift of personnel are recorded. Through personnel efficiency analysis, the attendance rate, average working hours, and task completion rate of day shift and night shift personnel are compared to identify efficient and inefficient time periods.

[0011] Specifically, data is collected on each staff member's work efficiency during the working day, including calculating each employee's attendance rate, average working hours, task completion rate, and health indicators during the day and night shifts, analyzing the changing trends of each employee's work efficiency during different time periods during the day and night shifts, and identifying efficient and inefficient periods.

[0012] Specifically, for the above-mentioned work types, data collection is performed on equipment work efficiency information. Equipment work data is collected through sensors, and the changing trends of equipment operating efficiency during different time periods of day and night shifts are analyzed to identify high-efficiency and low-efficiency periods. Weights are set based on personnel efficiency data and equipment efficiency data to calculate comprehensive weighted data, including:

[0013] Time periods: Determine the weights of different time periods based on the specific working hours of day and night shifts and key periods, including peak and off-peak periods;

[0014] Efficiency difference: Analyze the efficiency differences between day and night shifts in different time periods based on historical data and set corresponding weights;

[0015] Task type: Set the weights of different tasks based on their impact on overall efficiency;

[0016] Environmental factors: Set corresponding weights based on the impact of environmental factors such as power supply stability, temperature, and light on efficiency;

[0017] Weight calculation formula:

[0018] Wtotal=Wt×We×Wtk×Wenv

[0019] Where Wt is the time period weight, and the weights of different time periods are determined based on the specific working hours of day and night shifts and key periods (such as peak and off-peak periods);

[0020] Where We is the efficiency difference weight. According to the efficiency data, the weights of different time periods of day shift and night shift are set;

[0021] Where Wtk is the task type weight, according to the weight of different task types;

[0022] Among them, Wenv is the weight of environmental factors, according to the weight of environmental factors.

[0023] Specifically, in order to sort the operating data of each working day by efficiency coding, we collected the operating data of each working day, day shift, and night shift, including equipment efficiency, personnel efficiency, and task completion status. We calculated the efficiency value of each working day according to the efficiency formula, generated a unique efficiency code for each working day, and sorted the working days according to the efficiency value.

[0024]

[0025] is the efficiency of the i-th device, is the weight of the i-th device, is the efficiency of the i-th person, is the weight of the i-th device.

[0026] Specifically, the temporal score calculation formula is:

[0027]

[0028] Among them, S i is the temporal score of the i-th type of homework, w j is the weight of the jth feature, x ij is the jth eigenvalue of the i-th job type, and n is the number of features

[0029] Collection cycle setting formula:

[0030]

[0031] Among them, P iis the data collection period for the i-th job type, T max The maximum allowed collection period.

[0032] Specifically, using a deep learning-based embedding method, an embedding layer is designed to map high-dimensional discrete data, namely the schedule of each person's information at different time periods, into a low-dimensional continuous vector space. A deep learning model is then used to learn the embedding representation and establish an association with the basic efficiency of the equipment operated by the person's information. Specifically, the following steps are involved:

[0033] Data preparation:

[0034] Let Xi represent the shift information of the i-th person, and Ei represent the basic efficiency of the equipment he operates;

[0035] Embedding layer design: Use the embedding matrix W to map high-dimensional discrete data Xi to low-dimensional continuous vector v i =W·X i

[0036] Deep learning model design: input layer: accepts embedded representation vi, hidden layer: several fully connected layers, each layer has activation function ReLU, output layer: predicts the basic efficiency Ei of the equipment;

[0037] Loss function:

[0038]

[0039] Optimization algorithm: Use Adam optimizer to update model parameters.

[0040] Specifically, a unique benefit code generated by a person’s workday is imported into the embedding layer and converted into high-dimensional discrete data Xi, which is mapped to a low-dimensional continuous vector. Each category corresponds to a low-dimensional vector. Similarly, for equipment information and equipment efficiency Embedding processing is performed so that different equipment types or characteristics can be mapped to points in a high-dimensional space. The distance between them reflects their similarity, and the basic efficiency of the equipment to which personnel information operations belong is established.

[0041] On the other hand, the present invention also provides a time series data storage system based on an industrial Internet of Things smart mine, comprising:

[0042] S1. Collect time-series sensing data on the efficiency of various operating equipment during the day and night shifts for each operation type, as well as the corresponding operator scheduling time-series data, and transmit them to the central database for pre-processing to generate operation data;

[0043] S2. Encode and sort the efficiency of each workday's operation data. In the database, establish corresponding indexes based on the time-series perception data of the efficiency of various operating equipment during the day and night shifts for each operation type. Analyze the basic efficiency of the equipment during the day or night shift when different personnel are using it. Use a deep learning-based embedding method to design an embedding layer to map the high-dimensional discrete data, namely the schedule of each personnel information at different time periods, into a low-dimensional continuous vector space. Use a deep learning model to learn the embedding representation and establish an association with the basic efficiency of the equipment operated by the personnel information.

[0044] S3. Use structured query language to query and index the central database to meet the personnel scheduling plan and equipment efficiency plan corresponding to the preset efficiency standards, and generate operation reports.

[0045] Compared with the prior art, the beneficial effect of the present invention lies in that this application uses an embedding method based on deep learning, designs an embedding layer to map high-dimensional discrete data, namely the schedule of each personnel information in different time periods, into a low-dimensional continuous vector space, and uses a deep learning model to learn the embedding representation and establish an association with the basic efficiency of the equipment operated by the personnel information, thereby greatly improving the combination with different equipment under different scheduling information and improving overall work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0047] Figure 1 A flowchart of a time series data storage system based on an industrial Internet of Things smart mine provided by an embodiment of the present invention;

[0048] Figure 2 This is a functional block diagram of the time series data storage method based on the Industrial Internet of Things smart mine provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0050] See Figure 1 As shown, this embodiment provides a time series data storage system based on an industrial Internet of Things smart mine, including: a data acquisition module, which collects time series perception data on the efficiency of various operating equipment in the day and night shifts of each operation type on weekdays, as well as the corresponding operator scheduling time series data, and transmits them to a central database for preprocessing to generate operation data;

[0051] Data collection is conducted on the work efficiency of personnel on working days for each job type, including personnel efficiency data collection. By using the work time management system to calculate personnel working hours and rest time, marking day shifts and night shifts separately, the production volume and work task completion status of each shift are recorded. Through personnel efficiency analysis, the attendance rate, average working hours, and task completion rate of day shift and night shift personnel are compared to identify efficient and inefficient time periods.

[0052] Specifically, data is collected on each staff member's work efficiency during the working day, including calculating each employee's attendance rate, average working hours, task completion rate, and health indicators during the day and night shifts, analyzing the changing trends of each employee's work efficiency during different time periods of the day and night shifts, and identifying efficient and inefficient periods.

[0053] Specifically, for the above-mentioned work types, data collection is conducted on equipment work efficiency information. Equipment work data is collected through sensors, and the changing trends of equipment operating efficiency during different time periods of day and night shifts are analyzed to identify high-efficiency and low-efficiency periods. Based on the weights set by personnel efficiency data and equipment efficiency data, comprehensive weighted data is calculated, including:

[0054] Time periods: Determine the weights of different time periods based on the specific working hours of day and night shifts and key periods, including peak and off-peak periods;

[0055] Efficiency difference: Analyze the efficiency differences between day and night shifts in different time periods based on historical data and set corresponding weights;

[0056] Task type: Set the weights of different tasks based on their impact on overall efficiency;

[0057] Environmental factors: Set corresponding weights based on the impact of environmental factors such as power supply stability, temperature, and light on efficiency;

[0058] Weight calculation formula:

[0059] Wtotal=Wt×We×Wtk×Wenv

[0060] Where Wt is the time period weight, and the weights of different time periods are determined based on the specific working hours of day and night shifts and key periods (such as peak and off-peak periods);

[0061] Where We is the efficiency difference weight. According to the efficiency data, the weights of different time periods of day shift and night shift are set;

[0062] Where Wtk is the task type weight, according to the weight of different task types;

[0063] Among them, Wenv is the weight of environmental factors.

[0064] The data analysis module encodes and sorts the efficiency of each workday's operation data. In the database, it establishes corresponding indexes based on the time-series perception data of the efficiency of various operating equipment during the day and night shifts for each operation type. It also analyzes the basic efficiency data of the equipment during the day or night shift when different personnel are using it. Using a deep learning-based embedding method, an embedding layer is designed to map the high-dimensional discrete data, namely the schedule of each personnel information at different time periods, into a low-dimensional continuous vector space. The deep learning model is then used to learn the embedding representation and establish an association with the basic efficiency of the equipment operated by the personnel information.

[0065] In the data acquisition module, letter codes are set for the data of each operation type, including drilling ZO and blasting BO, loading AO and transportation YO, crushing PO and screening SO, grinding MO and beneficiation KO, tailings treatment WO and wastewater management FO, ventilation TO and drainage RO, and equipment maintenance records WH and repair records XL. Data acquisition is achieved through sensors, and a collection cycle is set for the data of each operation type according to a time series score, which quantifies multiple characteristics of each operation type.

[0066] Specifically, in order to sort the operating data of each working day by efficiency coding, we collected the operating data of each working day, day shift, and night shift, including equipment efficiency, personnel efficiency, and task completion status. We calculated the efficiency value of each working day according to the efficiency formula, generated a unique efficiency code for each working day, and sorted the working days according to the efficiency value.

[0067]

[0068] is the efficiency of the i-th device, is the weight of the i-th device, is the efficiency of the i-th person, is the weight of the i-th device.

[0069] Specifically, the temporal score calculation formula is:

[0070]

[0071] Among them, Si is the temporal score of the i-th type of homework, w j is the weight of the jth feature, x ij is the jth eigenvalue of the i-th job type, and n is the number of features

[0072] Collection cycle setting formula:

[0073]

[0074] Among them, P i is the data collection period for the i-th job type, T max The maximum allowed collection period.

[0075] Specifically, using a deep learning-based embedding method, an embedding layer is designed to map high-dimensional discrete data, namely the schedule of each person's information at different time periods, into a low-dimensional continuous vector space. A deep learning model is then used to learn the embedding representation and associate it with the basic efficiency of the equipment operated by the person's information. Specifically, the following steps are involved:

[0076] Data preparation:

[0077] Let Xi represent the schedule information of the i-th person, and Ei represent the basic efficiency of his operating equipment;

[0078] Embedding layer design: Use the embedding matrix W to map high-dimensional discrete data Xi to low-dimensional continuous vector v i =W·X i

[0079] Deep learning model design: input layer: accepts embedded representation vi, hidden layer: several fully connected layers, each layer has activation function ReLU, output layer: predicts the basic efficiency Ei of the equipment;

[0080] Loss function:

[0081]

[0082] Optimization algorithm: Use Adam optimizer to update model parameters.

[0083] A distributed storage module automatically partitions and stores the time-series-aware data on different nodes according to the coding order, and copies the basic efficiency data to a non-volatile memory and updates it regularly to ensure data integrity and high availability;

[0084] The indexing module uses structured query language to query and index the central database to obtain the corresponding personnel scheduling plans and equipment efficiency plans when the preset efficiency standards are met, and generate operation reports.

[0085] Specifically.

[0086] Specifically, a unique benefit code generated by a person’s workday is imported into the embedding layer to convert it into high-dimensional discrete data Xi and mapped to low-dimensional continuous vectors. Each category corresponds to a low-dimensional vector. Similarly, for equipment information and equipment efficiency Embedding processing is performed so that different equipment types or characteristics can be mapped to points in a high-dimensional space. The distance between them reflects their similarity, and the basic efficiency of the equipment to which personnel information operations belong is established.

[0087] See Figure 2 As shown, the present invention also provides a time series data storage method based on industrial Internet of Things smart mines, including:

[0088] S1. Collect time-series sensing data on the efficiency of various operating equipment during the day and night shifts for each operation type, as well as the corresponding operator scheduling time-series data, and transmit them to the central database for pre-processing to generate operation data;

[0089] S2. Encode and sort the efficiency of each workday's operation data. In the database, establish corresponding indexes based on the time-series perception data of the efficiency of various operating equipment during the day and night shifts for each operation type. Analyze the basic efficiency of the equipment during the day or night shift when different personnel are using it. Use a deep learning-based embedding method to design an embedding layer to map the high-dimensional discrete data, namely the schedule of each personnel information at different time periods, into a low-dimensional continuous vector space. Use a deep learning model to learn the embedding representation and establish an association with the basic efficiency of the equipment operated by the personnel information.

[0090] S3. Use structured query language to query and index the central database to meet the personnel scheduling plan and equipment efficiency plan corresponding to the preset efficiency standards, and generate operation reports.

[0091] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0092] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0093] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0094] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A time series data storage system based on industrial Internet of Things smart mine, characterized in that: include: The data collection module collects the time-series perception data of the efficiency of various operating equipment in the day shift and night shift of each operation type on weekdays, as well as the corresponding operator scheduling time-series data, and transmits them to the central database for pre-processing to generate operation data; The data analysis module encodes and sorts the benefits of the operation data of each working day, and in the database, establishes corresponding indexes based on the time-series perception data of the efficiency of various operating equipment in the day shift and night shift on working days for each operation type, and analyzes the basic efficiency data of the equipment in the day shift or night shift when different personnel are using it. Using the embedding method based on deep learning, the embedding layer is designed to map the high-dimensional discrete data, that is, the schedule of each personnel information in different time periods, into a low-dimensional continuous vector space, and the deep learning model is used to learn the embedding representation to establish an association with the basic efficiency of the equipment operated by the personnel information; A distributed storage module that automatically partitions and stores the time-series-aware data on different nodes according to the coding order, and copies the basic efficiency data to a non-volatile memory and updates it regularly to ensure data integrity and high availability; The indexing module uses structured query language to query and index the personnel scheduling plan and equipment efficiency plan corresponding to the preset benefit standards in the central database, and generates a work report.

2. The time series data storage system based on the industrial Internet of Things smart mine according to claim 1 is characterized in that: In the data acquisition module, letter codes are set for the data of each of the operation types, including drilling ZO and blasting BO, loading AO and transportation YO, crushing PO and screening SO, grinding MO and beneficiation KO, tailings treatment WO and wastewater management FO, ventilation TO and drainage RO, and equipment maintenance records WH and repair records XL, wherein data acquisition is achieved through sensors, and a collection cycle is set for the data of each of the operation types according to a time series score, and the time series score is to quantify multiple characteristics of each operation type.

3. The time series data storage system based on the industrial Internet of Things smart mine according to claim 2 is characterized in that: Data collection is conducted on the work efficiency of personnel on working days for each type of operation, including personnel efficiency data collection. The working hours and rest time of personnel are calculated by using the work time management system, and the day shift and night shift are marked separately. The production volume and work task completion status of each shift are recorded. Through personnel efficiency analysis, the attendance rate, average working hours and task completion rate of day shift and night shift personnel are compared to identify efficient and inefficient time periods.

4. The time series data storage system based on the industrial Internet of Things smart mine according to claim 3 is characterized in that: Data is collected on each staff member’s work efficiency during the working day, including calculating each employee’s attendance rate, average working hours, task completion rate, and health indicators during the day and night shifts, analyzing each employee’s work efficiency change trends during different time periods during the day and night shifts, and identifying efficient and inefficient time periods.

5. The time series data storage system based on the industrial Internet of Things smart mine according to claim 4 is characterized in that: For the above work types, data collection is performed on equipment work efficiency information. Equipment work data is collected through sensors, and the change trend of equipment operation efficiency in different time periods of day shift and night shift is analyzed to identify high-efficiency and low-efficiency time periods. Weights are set according to personnel efficiency data and equipment efficiency data to calculate comprehensive weight data, including: Time periods: Determine the weights of different time periods based on the specific working hours of day and night shifts and key periods, including peak and off-peak periods; Efficiency difference: Analyze the efficiency difference between day shift and night shift in different time periods based on historical data and set corresponding weights; Task type: Set the weights of different tasks based on their impact on overall efficiency; Environmental factors: Set corresponding weights based on the impact of environmental factors such as power supply stability, temperature, and light on efficiency; Weight calculation formula: Wtotal=Wt×We×Wtk×Wenv Where Wt is the time period weight. The weights of different time periods are determined according to the specific working hours of the day shift and night shift and key time periods (such as peak and non-peak periods); Where We is the efficiency difference weight. According to the efficiency data, the weights of different time periods of day shift and night shift are set; Where Wtk is the task type weight, according to the weight of different task types; Among them, Wenv is the weight of environmental factors, according to the weight of environmental factors.

6. The time series data storage system based on the industrial Internet of Things smart mine according to claim 5 is characterized in that: In order to sort the operation data of each working day by benefit coding, the operation data of each working day, day shift and night shift are collected, including equipment efficiency, personnel efficiency, and task completion status. The benefit value of each working day is calculated according to the benefit formula, a unique benefit code is generated for each working day, and the working days are sorted according to the benefit value; is the efficiency of the ith device, is the weight of the ith device, is the efficiency of the ith person, is the weight of the ith device.

7. The time series data storage system based on the industrial Internet of Things smart mine according to claim 6 is characterized by: The calculation formula of time series score is: Among them, S i is the temporal score of the i-th job type, w j is the weight of the jth feature, x ij is the jth characteristic value of the i-th job type, and n is the number of characteristics Collection cycle setting formula: Among them, P i is the data collection period for the i-th job type, T max is the maximum allowed collection period.

8. The time series data storage system based on the industrial Internet of Things smart mine according to claim 7 is characterized in that: Using the embedding method based on deep learning, the embedding layer is designed to map the high-dimensional discrete data, i.e. the schedule of each personnel information at different time periods, into a low-dimensional continuous vector space, and the deep learning model is used to learn the embedding representation and establish an association with the basic efficiency of the equipment operated by the personnel information, including: Data preparation: Let Xi represent the shift information of the ith person, and Ei represent the basic efficiency of the equipment he operates; Embedding layer design: Use the embedding matrix W to map high-dimensional discrete data Xi to low-dimensional continuous vector v i =W·X i Deep learning model design: input layer: accepts embedded representation vi, hidden layer: several fully connected layers, each layer has activation function ReLU, output layer: predicts the basic efficiency Ei of the equipment; Loss function: Optimization algorithm: Use Adam optimizer to update model parameters.

9. The time series data storage system based on the industrial Internet of Things smart mine according to claim 8 is characterized in that: The unique benefit code generated by the personnel working day is imported into the embedding layer to convert the high-dimensional discrete data Xi into a low-dimensional continuous vector. Each category corresponds to a low-dimensional vector. Similarly, the equipment information and equipment efficiency Embedding processing is performed so that different equipment types or characteristics can be mapped to points in a high-dimensional space. The distance between them reflects their similarity, and the basic efficiency of the equipment to which the personnel information operation belongs is established.

10. A time series data storage method based on an industrial Internet of Things smart mine, applied to the time series data storage system based on an industrial Internet of Things smart mine as claimed in any one of claims 1 to 9, characterized in that: include: S1, respectively collect the time series perception data of the efficiency of various operating equipment in the day shift and night shift of each operation type on weekdays, as well as the corresponding operation personnel scheduling time series data, and transmit them to the central database for preprocessing to generate operation data; S2. Encode and sort the benefits of the operation data of each working day, and in the database, establish corresponding indexes based on the time series perception data of the efficiency of various operating equipment in the day shift and night shift of each operation type on working days, and analyze the basic efficiency of the equipment in the day shift or night shift when different personnel use it. Use the embedding method based on deep learning to design an embedding layer to map the high-dimensional discrete data, that is, the schedule of each personnel information in different time periods, into a low-dimensional continuous vector space, and use the deep learning model to learn the embedding representation and establish an association with the basic efficiency of the equipment operated by the personnel information; S3. Use structured query language to query and index the central database to obtain the corresponding personnel scheduling plan and equipment efficiency plan when the preset efficiency standards are met, and generate work reports.

Images