Warehouse warehouse-in and warehouse-out management method and system serving coal mine directional drilling

Abstract

The invention relates to a warehouse in-out management method and system serving coal mine directional drilling, and belongs to the field of coal mine underground drilling. The method comprises the steps that equipment information in a warehouse is collected in real time; performing standardization processing on the number of each type of equipment in each warehouse to obtain a standardization value; determining the inventory condition of each type of equipment in each warehouse according to the standardized value; constructing an equipment residual life prediction module to obtain an equipment residual life prediction value; if the inventory of any type of equipment in any warehouse is insufficient, allocating the type of equipment in the warehouse with sufficient inventory to the warehouse with insufficient inventory, and generating a plurality of allocation schemes; for each deployment scheme, calculating a multi-target optimization value; and selecting a deployment scheme corresponding to the minimum multi-target optimization value as a final deployment scheme. The purpose of advanced overall management of warehouse equipment can be achieved, so that the equipment stopping time is shortened, the drilling efficiency is improved, and automatic and intelligent development of coal mine drilling is promoted.

Description

Technical Field

[0001] This application relates to the field of underground drilling in coal mines, specifically to a warehouse entry and exit management method and system for directional drilling in coal mines. Background Technology

[0002] Coal mine underground drilling is a drilling operation conducted within mine roadways or working faces, primarily used for geological exploration, gas extraction, water hazard prevention, rockburst control, and auxiliary engineering (such as cable laying, ventilation, and drainage). Its core objective is to ensure safe and efficient coal mining, while also serving resource exploration and disaster prevention functions. The unique production environment of underground coal mines dictates that the required equipment and materials cannot be transported as readily as those available for surface drilling. In recent years, many coal mines have seen rapid development in information systems for intelligent working faces; however, due to the frequent changes in underground drilling locations and the long distances between drilling rigs, the necessary equipment and spare parts are difficult to transport in a timely manner, hindering production.

[0003] Current warehouse management relies heavily on manual planning, replenishing the warehouse with spare parts according to fixed cycle plans. Often, a specific spare part is only called from the warehouse when it is lacking. Due to the special transportation system in mines, equipment downtime and idle personnel are common occurrences. If a scarce spare part is encountered, prolonged equipment downtime can significantly impact the construction progress. Summary of the Invention

[0004] In order to overcome at least one deficiency in the prior art, this application provides a warehouse entry and exit management method and system for directional drilling in coal mines.

[0005] Firstly, a warehouse inbound and outbound management method for directional drilling in coal mines is provided, including: Real-time collection of equipment information in the warehouse, including the warehouse to which the equipment belongs, equipment type, quantity of equipment, current total operating time of equipment, historical operating time of equipment, and number of times equipment has been repaired; The quantity of each type of equipment in each warehouse is standardized to obtain standardized values; the inventory status of each type of equipment in each warehouse is determined based on the standardized values. A module for predicting the remaining life of equipment is constructed. This module includes a time-series feature extraction module, a static feature extraction module, and a prediction module. The historical operating time of the equipment is input into the time-series feature extraction module to obtain time-series features. The current total operating time and the number of equipment maintenance operations are input into the static feature extraction module to obtain static features. The time-series features and static features are input into the prediction module to obtain the predicted value of the remaining life of the equipment. If the inventory of any type of equipment in any warehouse is in short supply, then the equipment of that type in the warehouse with sufficient inventory will be transferred to the warehouse with short supply, and multiple transfer plans will be generated. For each allocation plan, a multi-objective optimization value is calculated based on the predicted remaining lifespan of the equipment, the transportation time of the allocation plan, the transportation cost of the allocation plan, and the standardized value of the equipment. The allocation plan with the smallest multi-objective optimization value is selected as the final allocation plan.

[0006] In one embodiment, the quantity of each type of equipment in each warehouse is standardized to obtain a standardized value using the following formula:

[0007]

[0008] in, For the first Standardized values ​​for a certain type of equipment within a warehouse. For the first The quantity of a certain type of equipment in a warehouse This represents the average quantity of a certain type of equipment in each warehouse. n The total number of warehouses, This represents the standard deviation of a certain type of equipment within each warehouse.

[0009] In one embodiment, determining the inventory status of each type of equipment in each warehouse based on standardized values ​​includes: If the standardized value of a certain type of equipment in a warehouse satisfy If so, the inventory of this type of equipment in the warehouse is normal; If the standardized value of a certain type of equipment in a warehouse satisfy If so, then the warehouse has sufficient stock of this type of equipment; If the standardized value of a certain type of equipment in a warehouse satisfy If so, the warehouse will have a shortage of this type of equipment.

[0010] In one embodiment, the method further includes: Visualize inventory data: Use different colors to represent different inventory levels; inventory levels include normal, sufficient, and tight.

[0011] In one embodiment, the device's historical operating time is input into the time-series feature extraction module to obtain time-series features, including: The device's historical running time is input into a bidirectional LSTM layer to obtain the hidden state; The hidden state is input into the attention mechanism to obtain temporal features; the temporal features are calculated using the following formula:

[0012]

[0013] in, As a time series feature, For time steps, For time step, For time step Attention weights For time step The hidden state, For attention context vectors, It is the hyperbolic tangent function. This is the weight matrix. For bias parameters, This is the maintenance event attenuation coefficient. The time decay rate, This is the time step of the most recent maintenance.

[0014] In one embodiment, the current total operating time and the number of equipment maintenance visits are input into the static feature extraction module to obtain static features, including: The current total running time of the equipment is normalized using Min-Max to obtain the normalized total running time; The normalized total runtime is input into the fully connected layer to obtain the total runtime feature representation vector; Convert equipment repair counts into unique thermal codes; One-hot encoding is input into the embedding layer to obtain the repair count embedding vector; The static features are obtained by concatenating the total running time feature vector and the maintenance number embedding vector.

[0015] In one embodiment, time-series features and static features are input into the prediction module to obtain a predicted value for the remaining life of the device, including: Temporal features and static features are concatenated to obtain fused features; The fused features are input into the fully connected layer to obtain the predicted remaining lifetime of the device.

[0016] In one embodiment, for each allocation plan, a multi-objective optimization value is calculated based on the predicted remaining lifespan of the equipment, the transportation time of the allocation plan, the transportation cost of the allocation plan, and the standardized value of the equipment, using the following formula:

[0017] in, For the first The multi-objective optimization value of the allocation scheme. For the first The transportation time for each allocation plan For the maximum allowable transportation time, For the first The transportation cost of each allocation plan, To maximize the allowable transportation cost, For the first Standardized equipment values ​​for each allocation plan For the first The maximum service life of the equipment in each allocation scheme For the first The predicted remaining lifespan of the equipment in each allocation plan. As a weight for transportation time, As a weight for transportation costs, As equipment margin weight, Weighted by the remaining lifespan of the equipment.

[0018] In one embodiment, the warehouse to which the equipment belongs includes: a headquarters warehouse, a regional warehouse, and an underground warehouse.

[0019] Secondly, a warehouse inbound / outbound management system for directional drilling in coal mines is provided, including: The data acquisition module is used to collect equipment information in the warehouse in real time. The equipment information includes the warehouse to which the equipment belongs, the type of equipment, the quantity of equipment, the current total running time of the equipment, the historical running time of the equipment, and the number of times the equipment has been repaired. The inventory status determination module is used to standardize the quantity of each type of equipment in each warehouse to obtain a standardized value; and to determine the inventory status of each type of equipment in each warehouse based on the standardized value. The remaining life prediction module is used to construct the equipment remaining life prediction module. The equipment remaining life prediction module includes a time-series feature extraction module, a static feature extraction module, and a prediction module. The equipment's historical operating time is input into the time-series feature extraction module to obtain time-series features; the equipment's current total operating time and the number of equipment maintenance times are input into the static feature extraction module to obtain static features; the time-series features and static features are input into the prediction module to obtain the equipment's remaining life prediction value. The allocation scheme generation module is used to allocate the equipment of that type from warehouses with sufficient inventory to warehouses with insufficient inventory if the inventory of any type of equipment in any warehouse is tight, and to generate multiple allocation schemes. The inbound / outbound decision module is used to calculate a multi-objective optimization value for each allocation plan based on the predicted remaining life of the equipment, the transportation time of the allocation plan, the transportation cost of the allocation plan, and the standardized value of the equipment; and selects the allocation plan with the minimum multi-objective optimization value as the final allocation plan.

[0020] Compared to existing technologies, this application offers the following advantages: It establishes a three-tiered warehouse system—headquarters, regional, and underground—covering all equipment, accessories, and consumables used in underground coal mine drilling. It standardizes equipment information at different scales and determines the inventory status of each type of equipment within each region based on standardized values. By constructing an equipment remaining life prediction module, it forecasts the remaining lifespan of equipment. Based on the remaining lifespan and standardized equipment values, it automatically generates optimal allocation plans, achieving the goal of proactive and comprehensive management of warehouse equipment. This reduces equipment downtime, improves drilling efficiency, and promotes the development of automated and intelligent coal mine drilling. Attached Figure Description

[0021] This application can be better understood by referring to the description given below in conjunction with the accompanying drawings, which, together with the detailed description below, are incorporated in and form part of this specification. In the drawings: Figure 1 A flowchart illustrating the warehouse inbound and outbound management method for directional drilling in coal mines is shown. Figure 2 A schematic diagram of a three-level warehouse for directional drilling in a coal mine is shown. Detailed Implementation

[0022] Exemplary embodiments of the present application will be described below with reference to the accompanying drawings. For clarity and brevity, not all features of the actual embodiments are described in the specification. However, it should be understood that many embodiment-specific decisions can be made in the development of any such actual embodiment to achieve the developer’s specific objectives, and these decisions may vary as the embodiments differ.

[0023] It should also be noted that, in order to avoid obscuring this application with unnecessary details, only the device structure closely related to the solution according to this application is shown in the accompanying drawings, while other details that are not closely related to this application are omitted.

[0024] It should be understood that this application is not limited to the described embodiments by virtue of the following description with reference to the accompanying drawings. In this document, embodiments may be combined with each other, features may be substituted or borrowed between different embodiments, and one or more features may be omitted in one embodiment, where feasible.

[0025] This application provides a warehouse inbound and outbound management method for directional drilling in coal mines. Figure 1 A flowchart illustrating the warehouse inbound and outbound management method for directional drilling in coal mines is shown. (See attached diagram.) Figure 1 The methods include: Step S1: Collect equipment information in the warehouse in real time. The equipment information includes the warehouse to which the equipment belongs, the type of equipment, the quantity of equipment, the current total running time of the equipment, the historical running time of the equipment, and the number of times the equipment has been repaired.

[0026] Coal mine drilling has various uses, including geological exploration, gas extraction, water hazard control, rock burst management, and auxiliary engineering. Each type of operation requires different equipment and drilling tools. Based on frequency of use, this embodiment categorizes drilling projects into three types: water hazard control projects, gas extraction projects, and other projects. Warehouse equipment is specifically classified into six categories: large equipment, major equipment accessories, miscellaneous equipment accessories, drilling tools, geological surveying instruments, and general consumables. Warehouse managers at all levels regularly upload inventory information according to these categories.

[0027] Based on the six equipment categories, the movement of some scattered spare parts after they leave the warehouse is difficult to track, posing a risk of loss and causing resource waste. The three drilling projects are numbered A, B, and C respectively; the warehouse equipment is classified into six categories, numbered (10~19), (20~29), (30~39), (40~49), (50~59), and (60~69), where each number from 10 to 19 represents a specific category of large equipment; for example, 10 represents a drilling rig within large equipment. The warehouses to which the equipment belongs include: headquarters, regional, and downhole, numbered (1a~1z), (2a~2z), and (3a~3z) respectively. Specific numbers are marked as 001. For example: a water hazard prevention project - large equipment drilling rig - a headquarters, numbered A-10-1a-001.

[0028] Here, the establishment of a three-tiered warehouse system—headquarters, regional, and underground—is based on the classification of equipment locations in the information uploaded by personnel. In the coal mine directional drilling project, when equipment is taken out of the warehouse, it must first be transferred to different coal mine regions, such as Shanxi, Inner Mongolia, and Anhui, and then transferred from the region to the specific coal mine. Similarly, when the project is completed, the equipment is also returned to the warehouse level by level, first from the specific underground project to the regional warehouse, and then back to the headquarters warehouse. Figure 2 A schematic diagram of a three-level warehouse for directional drilling in a coal mine is shown.

[0029] Step S2: Standardize the quantity of each type of equipment in each warehouse to obtain standardized values; determine the inventory status of each type of equipment in each warehouse based on the standardized values.

[0030] Step S3: Construct a remaining life prediction module for equipment. The remaining life prediction module includes a time-series feature extraction module, a static feature extraction module, and a prediction module. The historical operating time of the equipment is input into the time-series feature extraction module to obtain time-series features. The current total operating time of the equipment and the number of equipment maintenance are input into the static feature extraction module to obtain static features. The time-series features and static features are input into the prediction module to obtain the predicted value of the remaining life of the equipment.

[0031] Step S4: If the inventory of any type of equipment in any warehouse is tight, then the equipment of that type in the warehouse with sufficient inventory will be transferred to the warehouse with tight inventory, and multiple transfer plans will be generated.

[0032] For example, if the drilling rig inventory in warehouse 2a is tight, while the drilling rig inventory in warehouse 2b is sufficient, then the drilling rig numbered 0001 in warehouse 2b will be transferred to warehouse 2a, thus generating a transfer plan.

[0033] Step S5: For each allocation plan, calculate the multi-objective optimization value based on the predicted remaining lifespan of the equipment, the transportation time of the allocation plan, the transportation cost of the allocation plan, and the standardized value of the equipment; select the allocation plan corresponding to the minimum multi-objective optimization value as the final allocation plan.

[0034] For example, regarding the allocation plan: Drilling rig numbered 0001 in warehouse 2b is allocated to warehouse 2a. The predicted remaining life of the equipment refers to the predicted remaining life of drilling rig numbered 0001. The allocation plan transportation time refers to the transportation time from warehouse 2b to warehouse 2a. The allocation plan transportation cost refers to the transportation cost from warehouse 2b to warehouse 2a. The standardized equipment value refers to the standardized value corresponding to the drilling rig category in warehouse 2b.

[0035] This embodiment establishes a three-tiered warehouse system: headquarters, regional, and underground, covering all equipment, accessories, and consumables used in underground coal mine drilling. Equipment information of varying quantities is processed in a standardized manner, and the inventory status of each type of equipment within each region is determined based on standardized values. An equipment remaining life prediction module is constructed to forecast the remaining lifespan of equipment. Based on the remaining lifespan and standardized equipment values, an optimal allocation plan is automatically generated to achieve proactive and comprehensive management of warehouse equipment, thereby reducing equipment downtime, improving drilling efficiency, and promoting the development of automated and intelligent coal mine drilling.

[0036] In one embodiment, after collecting equipment information from the warehouse, due to the large variety of equipment categories and the significant difference in quantity between large equipment and spare parts, the warehouse equipment information is retrieved. Then, for example, if the equipment is a drilling rig, the Z-Score normalization function is used to standardize the quantity of a specific type of equipment in a warehouse, and the standardized value is obtained. Specifically, the following formula can be used:

[0037]

[0038] in, For the first Standardized values ​​for a certain type of equipment within a warehouse. For the first The quantity of a certain type of equipment in a warehouse This represents the average quantity of a certain type of equipment in each warehouse. n The total number of warehouses, This represents the standard deviation of a certain type of equipment within each warehouse.

[0039] In one embodiment, determining the inventory status of each type of equipment in each warehouse based on standardized values ​​includes: If the standardized value of a certain type of equipment in a warehouse satisfy If so, the inventory of this type of equipment in the warehouse is normal; If the standardized value of a certain type of equipment in a warehouse satisfy If so, then the warehouse has sufficient stock of this type of equipment; If the standardized value of a certain type of equipment in a warehouse satisfy If so, the warehouse will have a shortage of this type of equipment.

[0040] Furthermore, the method also includes: visualizing inventory data: using different colors to represent different inventory levels; inventory levels include normal, sufficient, and tight. For example, a normal inventory level is indicated by yellow, meaning no short-term allocation is needed; a sufficient inventory level is indicated by green, meaning it can be transferred to other warehouses; and a tight inventory level is indicated by red, meaning replenishment is needed.

[0041] In this embodiment, the differences in data scales between different categories of devices are eliminated through standardization, and the results are visualized through color classification.

[0042] In one embodiment, step S3 involves inputting the device's historical operating time into the time-series feature extraction module to obtain time-series features, including: First, the device's historical runtime is input into a bidirectional LSTM layer to obtain the hidden state. For example, the equipment's historical operating time is [500, 600, 700, ...], which represents the operating time recorded for each month within the last 12 months, in hours. Here, This represents the hidden state corresponding to the current time step t.

[0043] Then, the hidden state is input into the attention mechanism to obtain the temporal features; the temporal features are represented by the following formula:

[0044]

[0045] in, As a time series feature, For time steps, For time step, For time step Attention weights For time step The hidden state; It is the hyperbolic tangent function; The attention context vector is a trainable vector whose function is to map the hidden state of the LSTM to a scalar energy value, thereby calculating the attention weight at each time step. Its value is related to the attention dimension, which is usually set to d / 2, where d is the dimension of the LSTM hidden layer. This is the weight matrix, which is related to the attention dimension and the LSTM hidden layer dimension, and is a trainable weight matrix. For bias parameters, This is the maintenance event attenuation coefficient. This is the time decay rate, which can be set to 0.1. This is the time step of the most recent maintenance. For example, if the current time step t is the 8th month, If it is the 6th month, then .

[0046] In this embodiment, a time decay rate is introduced into the attention mechanism to give higher weight to recent repairs.

[0047] In one embodiment, step S3 involves inputting the current total operating time and the number of equipment maintenance visits into the static feature extraction module to obtain static features, including: First, the current total running time of the equipment is normalized using Min-Max to obtain the normalized total running time, which can be obtained using the following formula:

[0048] in, The normalized total runtime. This represents the current total uptime of the equipment. This is the minimum service life of the equipment. This refers to the maximum service life of the equipment.

[0049] Then, the normalized total runtime is input into the fully connected layer to obtain the total runtime feature representation vector; Then, the number of equipment repairs is converted into unique hot codes. Here, the number of repairs is divided into 4 intervals: 0 times for no repair record, 1-2 times for basic maintenance, 3-5 times for frequent repairs, and more than 5 times for excessive repairs. Unique hot codes are assigned to these four intervals as follows: 0 times: [1,0,0,0], 1-2 times: [0,1,0,0], 3-5 times: [0,0,1,0], 5 times: [0,0,0,1].

[0050] Then, the one-hot encoding is input into the embedding layer to obtain the repair count embedding vector; Then, the total running time feature vector and the maintenance number embedding vector are concatenated to obtain the static features.

[0051] In one embodiment, step S3, inputting time-series features and static features into the prediction module to obtain the predicted remaining lifespan of the device, includes: First, temporal features and static features are concatenated to obtain fused features; Then, the fused features are input into the fully connected layer to obtain the predicted remaining lifetime of the device.

[0052] Furthermore, the equipment remaining life prediction module is a trained module, and the loss function used in the specific training process is as follows:

[0053]

[0054]

[0055] in, For loss function, For mean square error loss, For attention regularization, This is the regularization coefficient, usually set to 0.1. Indicates the first Predicted remaining lifespan for each sample Indicates the first The true remaining lifespan of each sample, where N is the number of samples. For time step Attention weights For time step.

[0056] In this embodiment, the mean squared error loss term ensures the accuracy of the model's predictions, making the predicted remaining lifetime as close as possible to the true value. The attention regularization term ensures the rationality of attention, preventing excessive concentration or dispersion of attention weights.

[0057] In one embodiment, for each allocation plan, a multi-objective optimization value is calculated based on the predicted remaining lifespan of the equipment, the transportation time of the allocation plan, the transportation cost of the allocation plan, and the standardized value of the equipment, using the following formula:

[0058] in, For the first The multi-objective optimization value of the allocation scheme. For the first The transportation time for each allocation plan For the maximum allowable transportation time, For the first The transportation cost of each allocation plan, To maximize the allowable transportation cost, For the first Standardized equipment values ​​for each allocation plan For the first The maximum service life of the equipment in each allocation scheme For the first The predicted remaining lifespan of the equipment in each allocation plan. As a weight for transportation time, As a weight for transportation costs, As equipment margin weight, Weighted by the remaining lifespan of the equipment.

[0059] In this embodiment, For the transportation time item, longer transportation times are penalized, meaning shorter times are better. This represents the weight of transportation time; it can be set higher in urgent situations. For transportation costs, high costs are penalized, meaning lower costs are better. This represents the weight of transportation costs; it can be set higher when costs are tight. For warehouse equipment surplus items, A negative value indicates a higher standardized value, suggesting a greater inventory surplus and encouraging outbound shipments. This is the equipment margin weight. The larger the standardized value, the higher this value should be set, so as to prioritize the consumption of warehouses with high inventory. For emergency items related to remaining lifespan, the closer to... The greater the punishment.

[0060] This embodiment can automatically balance the multi-dimensional conflicts of transportation time, transportation cost, warehouse equipment surplus, and remaining lifespan urgency to achieve the globally optimal outbound decision.

[0061] Based on the same inventive concept as the warehouse inbound / outbound management method for directional drilling in coal mines, this embodiment also provides a corresponding warehouse inbound / outbound management system for directional drilling in coal mines, including: The data acquisition module is used to collect equipment information in the warehouse in real time. The equipment information includes the warehouse to which the equipment belongs, the type of equipment, the quantity of equipment, the current total running time of the equipment, the historical running time of the equipment, and the number of times the equipment has been repaired. The inventory status determination module is used to standardize the quantity of each type of equipment in each warehouse to obtain a standardized value; and to determine the inventory status of each type of equipment in each warehouse based on the standardized value. The remaining life prediction module is used to construct the equipment remaining life prediction module. The equipment remaining life prediction module includes a time-series feature extraction module, a static feature extraction module, and a prediction module. The equipment's historical operating time is input into the time-series feature extraction module to obtain time-series features; the equipment's current total operating time and the number of equipment maintenance times are input into the static feature extraction module to obtain static features; the time-series features and static features are input into the prediction module to obtain the equipment's remaining life prediction value. The allocation scheme generation module is used to allocate the equipment of that type from warehouses with sufficient inventory to warehouses with insufficient inventory if the inventory of any type of equipment in any warehouse is tight, and to generate multiple allocation schemes. The inbound / outbound decision module is used to calculate a multi-objective optimization value for each allocation plan based on the predicted remaining life of the equipment, the transportation time of the allocation plan, the transportation cost of the allocation plan, and the standardized value of the equipment; and selects the allocation plan with the minimum multi-objective optimization value as the final allocation plan.

[0062] The warehouse entry and exit management system for directional drilling in this embodiment has the same inventive concept as the warehouse entry and exit management method for directional drilling in coal mine described above. Therefore, the specific implementation of this system can be found in the embodiment section of the warehouse entry and exit management method for directional drilling in coal mine described above, and its technical effects correspond to the technical effects of the above method, so it will not be repeated here.

[0063] The above descriptions are merely various embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A warehouse entry and exit management method for directional drilling in coal mines, characterized in that, include: Real-time collection of equipment information within the warehouse, including the warehouse to which the equipment belongs, equipment type, equipment quantity, current total operating time, historical operating time, and number of maintenance visits; The quantity of each type of equipment in each warehouse is standardized to obtain a standardized value; the inventory status of each type of equipment in each warehouse is determined based on the standardized value. A device remaining life prediction module is constructed, which includes a time-series feature extraction module, a static feature extraction module, and a prediction module; the historical operating time of the device is input into the time-series feature extraction module to obtain time-series features; The current total operating time of the equipment and the number of times the equipment has been repaired are input into the static feature extraction module to obtain static features; The time-series features and the static features are input into the prediction module to obtain the predicted value of the remaining lifespan of the equipment. If the inventory of any type of equipment in any warehouse is in short supply, then the equipment of that type in the warehouse with sufficient inventory will be transferred to the warehouse with short supply, and multiple transfer plans will be generated. For each allocation plan, a multi-objective optimization value is calculated based on the predicted remaining lifespan of the equipment, the transportation time of the allocation plan, the transportation cost of the allocation plan, and the standardized value of the equipment. The allocation scheme corresponding to the minimum multi-objective optimization value is selected as the final allocation scheme.

2. The method as described in claim 1, characterized in that, in, The quantity of each type of equipment in each warehouse is standardized to obtain a standardized value using the following formula: in, For the first Standardized values ​​for a certain type of equipment within a warehouse. For the first The quantity of a certain type of equipment in a warehouse This represents the average quantity of a certain type of equipment in each warehouse. n The total number of warehouses. This represents the standard deviation of a certain type of equipment within each warehouse.

3. The method as described in claim 1, characterized in that, in, Based on the standardized values, determine the inventory status of each type of equipment in each warehouse, including: If the standardized value of a certain type of equipment in a warehouse satisfy If so, the inventory of this type of equipment in the warehouse is normal; If the standardized value of a certain type of equipment in a warehouse satisfy If so, then the warehouse has sufficient stock of this type of equipment; If the standardized value of a certain type of equipment in a warehouse satisfy If so, the warehouse will have a shortage of this type of equipment.

4. The method as described in claim 1, characterized in that, The method further includes: Visualize inventory data: Use different colors to represent different inventory levels; these levels include normal, sufficient, and tight.

5. The method as described in claim 1, characterized in that, in, The historical operating time of the device is input into the time-series feature extraction module to obtain time-series features, including: The historical running time of the device is input into a bidirectional LSTM layer to obtain the hidden state; The hidden state is input into the attention mechanism to obtain temporal features; the temporal features are calculated using the following formula: in, As a time series feature, For time steps, For time step, For time step Attention weights For time step The hidden state, For attention context vectors, It is the hyperbolic tangent function. This is the weight matrix. For bias parameters, This is the maintenance event attenuation coefficient. The time decay rate, This is the time step of the most recent maintenance.

6. The method as described in claim 1, characterized in that, in, The current total operating time of the equipment and the number of times the equipment has been repaired are input into the static feature extraction module to obtain static features, including: The current total running time of the device is normalized using Min-Max to obtain the normalized total running time; The normalized total runtime is input into the fully connected layer to obtain the total runtime feature representation vector. Convert the number of equipment repairs into unique thermal codes; The one-hot encoding is input into the embedding layer to obtain the maintenance count embedding vector; The total running time feature vector and the maintenance number embedding vector are concatenated to obtain static features.

7. The method as described in claim 1, characterized in that, in, The time-series features and the static features are input into the prediction module to obtain the predicted remaining lifespan of the equipment, including: The temporal features and the static features are concatenated to obtain the fused features; The fused features are input into the fully connected layer to obtain the predicted remaining lifetime of the device.

8. The method as described in claim 1, characterized in that, in, For each allocation plan, a multi-objective optimization value is calculated based on the predicted remaining lifespan of the equipment, the transportation time of the allocation plan, the transportation cost of the allocation plan, and the standardized value of the equipment, using the following formula: in, For the first The multi-objective optimization value of the allocation scheme For the first The transportation time for each allocation plan For the maximum allowable transportation time, For the first The transportation cost of each allocation plan, To maximize the allowable transportation cost, For the first Standardized equipment values ​​for each allocation plan For the first The maximum service life of the equipment in each allocation scheme For the first The predicted remaining lifespan of the equipment in each allocation plan. As a weight for transportation time, As a weight for transportation costs, As equipment margin weight, Weighted by the remaining lifespan of the equipment.

9. The method as described in claim 1, characterized in that, The warehouses to which the equipment belongs include: headquarters warehouse, regional warehouse and underground warehouse.

10. A warehouse inbound and outbound management system for directional drilling in coal mines, characterized in that, include: The data acquisition module is used to collect equipment information in the warehouse in real time. The equipment information includes the warehouse to which the equipment belongs, the type of equipment, the quantity of equipment, the current total running time of the equipment, the historical running time of the equipment, and the number of times the equipment has been repaired. The inventory status determination module is used to standardize the quantity of each type of equipment in each warehouse to obtain a standardized value; and to determine the inventory status of each type of equipment in each warehouse based on the standardized value. The remaining life prediction module is used to construct the equipment remaining life prediction module. The equipment remaining life prediction module includes a time-series feature extraction module, a static feature extraction module, and a prediction module. The historical operating time of the equipment is input into the time-series feature extraction module to obtain time-series features. The current total operating time of the equipment and the number of times the equipment has been repaired are input into the static feature extraction module to obtain static features; The time-series features and the static features are input into the prediction module to obtain the predicted value of the remaining lifespan of the equipment. The allocation scheme generation module is used to allocate the equipment of that type from warehouses with sufficient inventory to warehouses with insufficient inventory if the inventory of any type of equipment in any warehouse is tight, and to generate multiple allocation schemes. The inbound / outbound decision module is used to calculate multi-objective optimization values ​​for each allocation plan based on the predicted remaining life of the equipment, the transportation time of the allocation plan, the transportation cost of the allocation plan, and the standardized value of the equipment. The allocation scheme corresponding to the minimum multi-objective optimization value is selected as the final allocation scheme.

Images