Mine wellhead-oriented thermal load prediction method and system

Abstract

The invention relates to the technical field of intelligent industrial energy management and prediction control, in particular to a thermal load prediction method and system for a mine wellhead. Comprising the following steps: S1, acquiring operation parameter data and meteorological data of mining equipment; the operation parameters comprise operation time data of the fixing equipment and the transportation equipment; the meteorological data is temperature data of the position of the mine; s2, acquiring static aggregation points and dynamic aggregation points in the operation process of the mining equipment based on the operation parameter data and the meteorological data of the mining equipment; according to the method, through multi-source data fusion and dynamic feature analysis, mining area heat load distribution is accurately described, and limitation of a traditional model is broken through; thermodynamic constraint and a compensation mechanism are coupled to dynamically optimize an equipment strategy, so that the energy efficiency real-time performance is improved; edge intelligence and closed-loop feedback are combined to guarantee safe operation, a full-link collaborative management scheme is formed, and intelligent and refined upgrading of mining area energy management is promoted.

Description

Technical Field

[0001] The present invention relates to the field of intelligent industrial energy management and predictive control technology, and in particular to a heat load prediction method and system for a mine wellhead. Background Art

[0002] In the scenarios of safe production and energy efficiency management in the field of mineral mining, traditional environmental control solutions generally rely on static analysis models built based on a large range of standardized parameters. This type of method has significant limitations when dealing with complex geological environments and dynamic climate conditions: its data model usually only covers macro-regional characteristics, and it is difficult to adapt to the special environmental parameters and real-time change requirements of local operating scenarios, resulting in serious deficiencies in the solution's refined control and dynamic adaptation capabilities. At the same time, the traditional energy consumption evaluation logic does not fully correlate the dynamic coupling relationship between the actual operating status of the equipment and environmental fluctuations, especially when environmental conditions frequently cross preset thresholds. Static rules are still used to infer equipment behavior, resulting in significant deviations between the evaluation results and actual working conditions. In addition, the performance characteristics of key equipment are overly simplified into fixed modes, which fail to reflect their dynamic response laws in actual complex interactive scenarios, further exacerbating the risk of inaccurate system control.

[0003] The technical bottleneck of traditional methods and the existing technical system face multiple contradictions: first, the static model has weak adaptability to the environmental heterogeneity of local operating scenarios and cannot respond to real-time demand changes in dynamic scenarios; second, energy consumption assessment relies too much on simplified assumptions, which breaks the inherent correlation between equipment behavior and environmental fluctuations, resulting in decision-making basis being divorced from actual working conditions; third, the dynamic change law of equipment performance is statically processed, which seriously weakens the system's ability to analyze complex interactive effects. These problems are particularly prominent in the high-risk and highly dynamic environment of mineral mining, manifested in increased energy waste, accumulated safety hazards and inaccurate energy efficiency management. It is urgent to reconstruct the control logic through a new technical paradigm to achieve coordinated optimization of environment-equipment-energy efficiency. Summary of the invention

[0004] In order to overcome the disadvantage of the incompatibility between a static system and a dynamic complex environment, the present invention provides a heat load prediction method and system for a mine wellhead.

[0005] The technical implementation scheme of the present invention is: a heat load prediction method for a mine wellhead, comprising the following steps: S1: Acquire mining equipment operating parameter data and meteorological data; the operating parameters include operating time data of fixed equipment and transportation equipment; the meteorological data is temperature data of the location of the mine; S2: Based on the operation parameter data and meteorological data of the mining equipment, obtain the static aggregation points and dynamic aggregation points during the operation of the mining equipment; the static aggregation points are the heat load characteristics of the local high-temperature areas formed by the frequent start and stop of fixed equipment; the dynamic aggregation points are the heat load characteristics of the cold and hot alternating areas caused by the movement of transportation equipment; S3: Calculate the static aggregation points using the first adjustment formula and calculate the dynamic aggregation points using the second adjustment formula; based on the static aggregation points and dynamic aggregation points, calculate the optimal control parameters of the heat load using the optimization formula; S4: Based on the optimized data acquisition method, construct a heat load prediction model for the mine shaft and generate an energy efficiency optimization strategy; dynamically optimize the parameter configuration scheme of the static aggregation points and dynamic aggregation points based on the prediction results.

[0006] Preferably, the obtaining of the operation parameter data and meteorological data of the mining equipment includes: Obtain the operation time of the fixed equipment and transportation equipment, and extract the operation intersection time of the fixed equipment and transportation equipment; Based on the operation intersection time, extract the start and stop moments of the fixed equipment and transportation equipment.

[0007] Preferably, the obtaining of the static aggregation points and dynamic aggregation points during the operation of the mining equipment based on the operation parameter data and meteorological data of the mining equipment includes: Based on the operation intersection time and start and stop moments, calculate the time difference between the restart after the equipment stops and the generation of a new intersection with another equipment, and use it as the calculation moment of the static aggregation points and dynamic aggregation points.

[0008] Preferably, the calculating of the static aggregation points using the first adjustment formula includes: The first adjustment formula is as follows, where, is the heat load value of the static aggregation point, is the total number of start and stop frequencies, is the equipment energy efficiency coefficient related to temperature, is the operation time of the fixed equipment related to temperature, is the start and stop frequency coefficient, is the depth of the mine where the equipment is located, is the operation power of the fixed equipment, is the heat diffusion coefficient related to temperature.

[0009] Preferably, the calculating of the dynamic aggregation points using the second adjustment formula includes: The second adjustment formula is as follows, where, is the heat load value of the dynamic aggregation point, is the internal temperature of the mine is the target temperature at the mine shaft is the mine air velocity at the location of the temperature-related transportation equipment is the cross-sectional area of the mine shaft is the operating time of the transportation equipment is the reference operating time of the temperature-related transportation equipment is the effective length of the heat conduction path

[0010] Preferably, based on the static aggregation point and the dynamic aggregation point, calculating the optimal control parameters of the heat load using an optimization formula, including: Based on the static aggregation point and the dynamic aggregation point, obtaining the heat load data of the mine shaft If the heat load data of the shaft calculated by the static and dynamic aggregation points is higher than the preset upper threshold, the equipment is delayed from being turned off to enhance heat dissipation; if the heat load data of the shaft calculated by the static and dynamic aggregation points is lower than the preset lower threshold, the equipment is turned on in advance to maintain heat balance

[0011] Preferably, calculating the optimal control parameters of the heat load using the optimization formula, including: The optimization formula is as follows where is the power of the heating equipment is the static proportionality coefficient is the dynamic differential coefficient is the delay compensation coefficient is the time decay constant is the delay time from the start of equipment start / stop The optimization formula generates the optimal energy efficiency parameters that satisfy the mine heat balance equation through the linear superposition of the static aggregation point and the dynamic aggregation point and dynamic differential compensation, where the delay compensation term is used to correct the time lag effect of the prediction model

[0012] Preferably, based on the optimized data acquisition method, constructing a mine shaft heat load prediction model and generating an energy efficiency optimization strategy, including: Integrating the operating parameters of fixed equipment, dynamic transportation trajectories, and environmental meteorological data, constructing a spatio-temporal aligned feature matrix, and extracting the heat accumulation effect of fixed equipment start / stop and the transient fluctuations of the transportation path of transportation equipment as key factors Adopting a hierarchical hybrid model architecture, the basic layer uses random forest regression to fit the non-linear relationship of energy efficiency, the time series layer predicts the dynamic changes of the load through an LSTM network, and the physical constraint layer uses the thermodynamic equation as a loss function term to enforce energy conservation The model training introduces a joint loss function, couples the mean square error with the heat conduction delay penalty term, and optimizes the generalization performance based on spatio-temporal cross-validation; In the deployment stage, spatio-temporal prediction of heat load is realized through edge computing, an energy efficiency optimization strategy is dynamically generated and the warning threshold is triggered, driving the mine heat management system to execute according to the predicted optimal configuration.

[0013] Preferably, the parameter configuration scheme for dynamically optimizing the static aggregation point and the dynamic aggregation point based on the prediction result includes: Based on the measured heat load and the predicted value of the heat load, the predicted value with a deviation amount greater than the preset threshold is obtained and defined as the deviation predicted value; Based on the deviation predicted value, the corresponding equipment operating parameters are obtained; The corresponding equipment operating parameters are adjusted from small to large according to the magnitude of the deviation predicted value.

[0014] Preferably, a heat load prediction system for a mine shaft includes: Multi-source data prediction module: Collect equipment operation time, transportation equipment trajectory, and mine temperature data, extract the intersection time of equipment operation and start / stop moments, and synchronously construct a spatio-temporal database; Thermodynamic feature optimization module: Based on the intersection time difference of operation, calculate the static aggregation point of fixed equipment through the first adjustment formula, and calculate the dynamic aggregation point of transportation equipment through the second adjustment formula; Heat load prediction module: Integrate the static and dynamic aggregation point data, generate a prediction result using a hierarchical model architecture, and trigger a warning through edge computing; Prediction deviation compensation module: According to the prediction and measured deviation amount, use the joint loss function to reversely adjust the coefficient.

[0015] Beneficial effects: Through multi-source data deep fusion and dynamic feature extraction technology, the present invention accurately identifies the spatio-temporal characteristics of the collaborative operation of mine equipment, and constructs an optimization system covering the whole process of heat load generation, transfer, and regulation. Based on the collaborative analysis of static and dynamic aggregation points, it breaks through the limitation of the traditional static model in characterizing the complex heat field interaction effect, and realizes the high-precision characterization of the heat load distribution characteristics in the mining area; through the coupled design of thermodynamic constraints and dynamic compensation mechanisms, it dynamically optimizes the equipment start / stop strategy and the heat supply configuration scheme, significantly improving the real-time performance and environmental adaptability of energy efficiency regulation; combined with edge intelligent computing and closed-loop feedback mechanism, it synchronously ensures the anti-freezing safety of the shaft and the operation stability of the equipment, effectively suppressing the risk of local heat imbalance. Finally, a full-link collaborative management scheme from data collection, feature analysis, strategy generation to dynamic correction is formed, providing a systematic solution with energy efficiency optimization, safety control, and low-carbon operation and maintenance value for mines, and promoting the upgrading of energy management in mining areas towards intelligence and refinement. Description of the Drawings

[0016] Figure 1 This is a flowchart of the heat load prediction method for the mine shaft opening of the present invention; Figure 2 This is a schematic structural diagram of the heat load prediction system for the mine shaft opening of the present invention. Specific embodiments

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0018] Embodiment 1: A heat load prediction method for a mine shaft opening, as Figure 1 shown, includes the following steps: S1: Obtain the operation parameter data of mine equipment and meteorological data; the operation parameters include the operation time data of fixed equipment and transportation equipment; the meteorological data is the temperature data of the location where the mine is located; S2: Based on the operation parameter data of the mine equipment and the meteorological data, obtain the static aggregation points and dynamic aggregation points during the operation of the mine equipment; the static aggregation points are the heat load characteristics of the local high-temperature areas formed by the frequent start and stop of fixed equipment; the dynamic aggregation points are the heat load characteristics of the cold and hot alternating areas caused by the movement of transportation equipment; S3: Calculate the static aggregation points using the first adjustment formula and calculate the dynamic aggregation points using the second adjustment formula; based on the static aggregation points and dynamic aggregation points, calculate the optimal control parameters of the heat load using the optimization formula; S4: Based on the optimized data acquisition method, construct a heat load prediction model for the mine shaft opening, generate an energy efficiency optimization strategy; dynamically optimize the parameter configuration scheme of the static aggregation points and dynamic aggregation points based on the prediction results.

[0019] Obtaining the operation parameter data of mine equipment and meteorological data includes: Obtain the operation time of fixed equipment and transportation equipment, and extract the operation intersection time of fixed equipment and transportation equipment; Based on the operation intersection time, extract the start and stop moments of fixed equipment and transportation equipment.

[0020] For further explanation, by extracting the operating intersection time of the fixed equipment and the transportation equipment, the time window for their collaborative operation is locked, providing a spatio-temporal benchmark for subsequent analysis. Based on the intersection time, the start and stop moments are further extracted to identify the critical points of equipment interaction (such as the heat load superposition period or the cold and heat alternation nodes). The internal logic is to screen key events through time alignment to provide dynamic time series characteristics for thermodynamic modeling. Example of time accuracy: The fixed equipment collects data at the second level (millisecond-level time stamp), and the transportation equipment has data at 5-second intervals, which is interpolated and aligned to the second level, and NTP synchronization ensures an error < 1 second. Example of exception handling: For transportation data interruption time ≤ 3 minutes, linear interpolation is used to fill it; for transportation data interruption time > 3 minutes, the data is removed and marked as "invalid window". Example of start and stop determination: Fixed equipment: The power fluctuation ≥ 10% lasts for 30 seconds and the temperature rise ≥ 2 °C / minute; Transportation equipment: The speed returns to zero (unit: m / s) and the load is cleared (unit: ton) for 2 minutes.

[0021] Based on the operating parameter data and meteorological data of the mining equipment, static aggregation points and dynamic aggregation points during the operation of the mining equipment are obtained, including: Based on the operating intersection time and the start and stop moments, calculate the time difference between the restart of a device after it stops and the generation of a new intersection with another device, which is used as the calculation moment for the static aggregation point and the dynamic aggregation point. For further explanation, by calculating the time difference between the generation of a new intersection after the device stops and restarts, the thermodynamic characteristics of the static and dynamic aggregation points are distinguished: Static aggregation point: The time difference between the restart of the fixed equipment and the new intersection with the transportation equipment reflects the local heat accumulation effect of the periodic start and stop of the fixed heat source (such as the heat superposition during the overlapping period when the crusher restarts 2 hours after shutdown and the belt conveyor is running); Dynamic aggregation point: The time difference between the movement of the transportation equipment and the new intersection with the fixed equipment characterizes the cold and heat alternation cycle caused by the spatial displacement of the dynamic heat source (the dynamic aggregation point characterizes the periodic heat fluctuation intensity caused by the spatial displacement of the dynamic heat source, specifically the time difference between the movement of the transportation equipment and the new intersection with the fixed equipment, and its heat load value is related to the equipment spacing and heat conduction rate (Example: The ore truck moves from area A 50 meters away from the ventilation equipment to area B 100 meters away, and the heat exchange cycle is extended to 2.3 times the original cycle.)). During the modeling process, the time difference needs to be correlated with the physical spacing and thermal conductivity parameters of the equipment to avoid errors in the heat flux density distribution caused by relying solely on the time dimension calculation (such as the heat accumulation in the deep well area being underestimated by 30%).

[0022] Calculate the static aggregation point using the first adjustment formula, including: The first adjustment formula is as follows, where, is the heat load value of the static aggregation point, is the total number of start and stop frequencies, is the equipment energy efficiency coefficient related to temperature, is the operating time of the fixed equipment related to temperature, is the start-stop frequency coefficient, is the depth of the mine where the equipment is located, is the operating power of the fixed equipment, is the thermal diffusion coefficient related to temperature.

[0023] For further explanation, the formula constructs a thermal load intensity model of the static aggregation point by quantifying the start-stop heat accumulation effect of the fixed equipment at different ambient temperatures. Parameter functions: : The actual thermal output power of the equipment, is the equipment energy efficiency coefficient related to temperature (decreases at low temperatures and higher power is required to maintain thermal balance); : The operating duration corrected for temperature (e.g., the equipment needs to extend the operating time to compensate for heat loss at -5°C); : The start-stop frequency coefficient (when starting and stopping frequently >1, amplifying the thermal shock effect); ( ): The depth-related thermal attenuation factor, the larger (deep well) or (unit: W / (m·K)) the smaller (poor heat transfer), the exponential term approaches 1, and heat is more likely to accumulate. Combined effects: Power-depth coupling: High-power equipment in deep wells (such as downhole pumps) contribute significantly to the thermal load due to having a large value; Temperature-frequency compensation: In a low-temperature environment, decrease and extension together increase the thermal output demand, and frequent start-stop ( with a high value) further amplifies the cumulative effect. Example: A certain crusher at -10°C has its mine depth increased from 50 meters (shallow layer, =300 W / (m·K)) to 200 meters (deep layer, =150 W / (m·K)) and starts and stops 5 times a day ( =1.2), the value of the shallow-layer equipment is increased by about 3 times, and based on this, its continuous operating duration is dynamically limited to avoid local overheating.

[0024] Calculate the dynamic aggregation point using the second adjustment formula, including: The second adjustment formula is as follows, Among them, is the thermal load value of the dynamic aggregation point, is the internal temperature of the mine, is the target temperature of the mine shaft, is the mine wind speed at the location of the transportation equipment related to temperature, is the cross-sectional area of the mine shaft, is the operating time of the transportation equipment, is the reference operating time of the transportation equipment related to temperature, is the effective length of the heat conduction path.

[0025] For further explanation, the formula quantifies the dynamic heat exchange effect caused by the movement of the transportation equipment and characterizes the heat load fluctuation intensity in the cold and hot alternating area. Parameter functions: : Trigger the calculation only when the temperature in the well is higher than the target temperature to avoid ineffective heat dissipation (e.g., = 5 °C, = 8 °C contributes a temperature difference of 3 °C); The wind speed and the cross-sectional area of the wellhead jointly reflect the heat convection efficiency (the higher the wind speed, the faster the heat dissipation, but the high value expands the heat exchange area); The non-linear relationship between the operation time and the reference time reflects the marginal effect of heat accumulation (e.g., = 2 times contributes a coefficient of ln(3) ≈ 1.1 times); : The depth correction term of the transportation equipment, with the unit of meter. The greater the depth (e.g., = 300 meters), the attenuation of the unit heat load along with the extension of the conduction path. Combined effect: Temperature difference - wind speed game: A high temperature difference ( - ) requires a high wind speed ( rising) to suppress the value. Conversely, a low wind speed will amplify the heat accumulation; Time - depth coupling: A long operation time ( rising) is more likely to cause local overheating in shallow ( descending) mines (e.g., = 100 meters, the value is 2.7 times higher than that of = 300 meters). Example: A transport vehicle operates for 6 hours ( = 150 m) at a well depth of 150 meters, a wind speed of 0.5 m / s ( = 0.5), and a temperature difference of 4 °C ( - = 4 °C). Substituting into the formula for calculation gives , which is about 2.3 times higher than the reference working condition ( = 100 m, ), ) = 1.0 m / s, = 2 hours, )).

[0026] Based on the above-mentioned static aggregation point and dynamic aggregation point, use the optimization formula to calculate the optimal control parameters of the heat load, including: Based on the static aggregation point and dynamic aggregation point, obtain the heat load data of the mine wellhead; If the wellhead heat load data calculated by the static and dynamic aggregation points is higher than the preset upper threshold, the equipment shutdown is delayed to enhance heat dissipation; if the wellhead heat load data calculated by the static and dynamic aggregation points is lower than the preset lower threshold, the equipment is started in advance to maintain heat balance.

[0027] For further explanation, the wellhead heat load is calculated by superimposing the static aggregation point (fixed equipment heat accumulation) and the dynamic aggregation point (transport equipment heat fluctuation), and compared with the preset safety threshold to trigger equipment regulation: Internal logic: When the measured heat load is higher than the threshold (risk of overloading), the equipment shutdown is delayed to enhance heat dissipation; when the predicted heat load is lower than the threshold (underload energy efficiency waste), the equipment is started in advance to maintain heat balance, forming a closed-loop control. The threshold dynamic generation algorithm is determined by comprehensively weighting the historical heat load extreme value distribution, the calculation results of the thermodynamic simulation model, and the industry safety standards.

[0028] The optimal control parameters of the heat load are calculated using the optimization formula, including: The optimization formula is as follows, where, is the power of the heating equipment, is the static proportionality coefficient, is the dynamic differential coefficient, is the delay compensation coefficient, is the time decay constant, is the delay time from the start of equipment start / stop; The optimization formula generates the optimal energy efficiency parameters that satisfy the mine heat balance equation through the linear superposition of the static and dynamic aggregation points and dynamic differential compensation, where the delay compensation term is used to correct the time lag effect of the prediction model. For further explanation, the formula generates the optimal heating power configuration by fusing the static / dynamic heat load prediction values and the dynamic change trend, combined with the delay compensation mechanism, to achieve real-time control of heat balance. Parameter functions: The static proportional term directly responds to the current heat load demand (e.g., when and , = 0.8 contributes 120 kW of benchmark power); The dynamic differential term anticipates the change trend of the heat load (e.g., when increases by 20 kW per hour, = 0.5 adds 10 kW of power in advance to suppress the temperature rise rate); The delay compensation term exponentially decays to correct the historical heat load lag effect (e.g., when = 0.1 and = 10 minutes, the compensation amount decays to 37% of the initial value, avoiding overshoot). Combined effect: Trend anticipation and real-time response coordination: The static term ensures basic heating, and the dynamic differential term captures transient fluctuations (such as sudden shutdown of transport equipment resulting in sudden drop, the derivative term immediately reduces the heating power); hysteresis effect suppression: the compensation term decays over time to prevent interference from overestimating the early static heat load on current control (e.g., after 1 hour of shutdown, the compensation amount drops to a negligible level). Example: When a sudden ventilation failure occurs in the mine ( the growth rate reaches 30 kW / min, = 0.6), an additional 18 kW of power is buffered in advance through the derivative term to warm up; meanwhile, after 30 minutes of shutdown ( = 0.2, = 30), the static compensation amount decays to 0.2% of the initial value to avoid redundant heating.

[0029] Based on the optimized data acquisition method, construct a heat load prediction model for the mine shaft, and generate an energy efficiency optimization strategy, including: Integrate the operating parameters of fixed equipment, dynamic transportation trajectories, and environmental meteorological data to construct a spatio-temporally aligned feature matrix, and extract the start-stop heat accumulation effect of fixed equipment and the transient fluctuations of the transportation paths of transportation equipment as key factors; Adopt a hierarchical hybrid model architecture. The basic layer uses random forest regression to fit the non-linear relationship of energy efficiency, the time series layer predicts the dynamic changes of the load through the LSTM network, and the physical constraint layer uses the thermodynamic equation as a loss function term to enforce energy conservation; The model training introduces a joint loss function, couples the mean square error and the heat conduction delay penalty term, and optimizes the generalization performance based on spatio-temporal cross-validation; During the deployment phase, spatio-temporal prediction of heat load is achieved through edge computing, dynamically generating energy efficiency optimization strategies and triggering warning thresholds, driving the mine heat management system to execute according to the predicted optimal configuration. Further explanation is that the process constructs a full-link closed loop of data-model-deployment: Data integration logic: Eliminate the dimensional and temporal deviations of multi-source data (equipment, transportation, meteorology) through spatio-temporal alignment, extract key thermodynamic factors (start-stop heat accumulation, path transient fluctuations), and provide high-purity feature inputs for modeling; Hierarchical model logic: The basic layer (machine learning) fits static energy efficiency non-linearity, the temporal layer (deep learning) captures dynamic evolution laws, and the physical constraint layer (thermodynamic equations) enforces energy conservation to achieve "data-driven + physically credible" hybrid prediction; Training optimization logic: In the joint loss function, the mean square error optimizes prediction accuracy, the heat conduction delay penalty term constrains the model to conform to the physical conduction rate, and spatio-temporal cross-validation avoids local overfitting; Deployment control logic: Edge computing realizes low-latency prediction, dynamically generates strategies and triggers warnings, forming a real-time control closed loop. Example, data alignment: Align the transportation trajectory GPS data (1Hz) and device sensor data (10Hz) to 10Hz through timestamp interpolation; Physical constraint: Add an energy conservation term to the loss function (apply a penalty when the predicted heat output - environmental heat dissipation ≠ 0); Edge deployment: Replace the LSTM in the temporal layer with TinyLSTM to adapt to the memory limitations of edge devices, ensuring that the prediction delay < 200ms.

[0030] Based on the predicted results, dynamically optimize the parameter configuration schemes of static aggregation points and dynamic aggregation points, including: Based on the measured heat load and the predicted value of the heat load, obtain the predicted value with a deviation greater than the preset threshold and define it as the deviation predicted value; Based on the deviation predicted value, obtain the corresponding device operating parameters; Adjust the corresponding device operating parameters from small to large according to the magnitude of the deviation predicted value.

[0031] Further explanation is that the process realizes the progressive adjustment of device parameters through prediction-measurement deviation feedback: Internal logic: When the prediction deviation exceeds the threshold, it is determined that the model is locally inaccurate, and the relevant device parameters are inversely located; Adjust from small to large according to the deviation magnitude to avoid large parameter mutations causing system oscillations and achieve stable optimization. Example, threshold dynamic setting: The threshold is set to ±10kW in winter (higher tolerance during cold snaps), and adjusted to ±5kW in summer; Adjustment strategy: Adjust the device power by 2% when the deviation is 20kW, and adjust by 5% when the deviation is 50kW to avoid step changes; Conflict arbitration: If both the ventilator and the heater are associated with deviations, give priority to adjusting the heat source device (heater weight coefficient 0.7).

[0032] Embodiment 2: On the basis of Embodiment 1, a heat load prediction system for a mine shaft, such as Figure 2As shown, it includes: Multi-source data prediction module: Collect equipment operation time, transportation equipment trajectory, and mine temperature data, extract the intersection time of equipment operation and start / stop moments, and synchronously build a spatio-temporal database; Thermodynamic feature optimization module: Based on the difference in operation intersection time, calculate the static aggregation points of fixed equipment through the first adjustment formula, and calculate the dynamic aggregation points of transportation equipment through the second adjustment formula; Heat load prediction module: Integrate static and dynamic aggregation point data, generate prediction results using a hierarchical model architecture, and trigger warnings through edge computing; Prediction deviation compensation module: According to the prediction and measured deviation amount, use the joint loss function to reverse-adjust the coefficient.

[0033] The above has introduced this application in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A method for predicting the heat load facing the mine shaft, characterized in that, It includes the following steps: S1: Obtain the operation parameter data and meteorological data of the mining equipment; the operation parameters include the operation time data of the fixed equipment and the transportation equipment; the meteorological data is the temperature data of the location where the mine is located; S2: Based on the operation parameter data and meteorological data of the mining equipment, obtain the static aggregation points and dynamic aggregation points during the operation of the mining equipment; the static aggregation points are the heat load characteristics of the local high-temperature areas formed by the frequent start and stop of the fixed equipment; The dynamic aggregation points are the heat load characteristics of the cold and hot alternating areas caused by the movement of the transportation equipment; S3: Calculate the static aggregation points using the first adjustment formula and calculate the dynamic aggregation points using the second adjustment formula; Based on the static aggregation points and dynamic aggregation points, calculate the optimal control parameters of the heat load using the optimization formula; S4: Based on the optimized data acquisition method, construct a heat load prediction model for the mine shaft and generate an energy efficiency optimization strategy; Dynamically optimize the parameter configuration scheme of the static aggregation points and dynamic aggregation points based on the prediction results.

2. The thermal load prediction method for a mine shaft opening according to claim 1, characterized in that, The obtaining of the operation parameter data and meteorological data of the mining equipment includes: Obtain the operation time of the fixed equipment and the transportation equipment, and extract the operation intersection time of the fixed equipment and the transportation equipment; Based on the operation intersection time, extract the start and stop moments of the fixed equipment and the transportation equipment.

3. A thermal load prediction method and system for a mine shaft entrance according to claim 1, characterized in that, The obtaining of the static aggregation points and dynamic aggregation points during the operation of the mining equipment based on the operation parameter data and meteorological data of the mining equipment includes: Based on the operation intersection time and the start and stop moments, calculate the time difference between the restart after the equipment stops and the generation of a new intersection with another equipment, and use it as the calculation moment of the static aggregation points and dynamic aggregation points.

4. A method for predicting the heat load for a mine shaft entrance according to claim 1, characterized in that, The calculation of the static aggregation points using the first adjustment formula includes: The first adjustment formula is as follows, Among them, is the heat load value of the static aggregation point, is the total number of start-stop frequencies, is the equipment energy efficiency coefficient related to temperature, is the operating time of the fixed equipment related to temperature, is the start-stop frequency coefficient, is the depth of the mine where the equipment is located, is the operating power of the fixed equipment, is the heat diffusion coefficient related to temperature.

5. A method for predicting the heat load for a mine shaft entrance, characterized in that, according to claim 1 The calculation of the dynamic aggregation points using the second adjustment formula includes: The second adjustment formula is as follows, Among them, is the heat load value of the dynamic aggregation point, is the internal temperature of the mine, is the target temperature at the mine shaft opening, is the mine air velocity at the location of the temperature-related transportation equipment, is the cross-sectional area of the mine shaft opening, is the operation time of the transportation equipment, is the reference operation time of the temperature-related transportation equipment, is the effective length of the heat conduction path.

6. A thermal load prediction method for a mine shaft entrance according to claim 1, characterized in that, The calculation of the optimal control parameters of the heat load using the optimization formula based on the static aggregation points and dynamic aggregation points includes: Based on the static aggregation points and dynamic aggregation points, obtain the heat load data of the mine shaft; If the heat load data of the mine shaft calculated by the static and dynamic aggregation points is higher than the preset upper limit threshold, delay turning off the equipment to enhance heat dissipation; if the heat load data of the mine shaft calculated by the static and dynamic aggregation points is lower than the preset lower limit threshold, turn on the equipment in advance to maintain heat balance.

7. A heat load prediction method for a mine shaft opening according to claim 1, characterized in that The calculation of the optimal control parameters of the heat load using the optimization formula includes: The optimization formula is as follows, Among them, is the power of the heating equipment, is the static proportionality coefficient, is the dynamic differential coefficient, is the delay compensation coefficient, is the time decay constant, is the delay time starting from the start and stop of the slave device; The optimization formula generates the optimal energy efficiency parameters that satisfy the mine heat balance equation through the linear superposition of the static aggregation points and dynamic aggregation points and dynamic differential compensation, where the delay compensation term is used to correct the time lag effect of the prediction model.

8. A thermal load prediction method for a mine shaft opening according to claim 1, characterized in that, The construction of the heat load prediction model for the mine shaft and the generation of the energy efficiency optimization strategy based on the optimized data acquisition method includes: Integrate the operation parameters of the fixed equipment, the dynamic transportation trajectory and the environmental meteorological data, construct a spatio-temporally aligned feature matrix, and extract the heat accumulation effect of the start and stop of the fixed equipment and the transient fluctuation of the transportation path of the transportation equipment as the key factors; Adopt a hierarchical hybrid model architecture. The basic layer uses random forest regression to fit the non-linear relationship of energy efficiency. The time series layer predicts the dynamic change of load through the LSTM network. The physical constraint layer takes the thermodynamic equation as a loss function term to enforce energy conservation; The model training introduces a joint loss function, couples the mean square error and the heat conduction delay penalty term, and optimizes the generalization performance based on spatio-temporal cross-validation; In the deployment stage, spatio-temporal prediction of the heat load is realized through edge computing, an energy efficiency optimization strategy is dynamically generated and the warning threshold is triggered, driving the mine heat management system to execute according to the predicted optimal configuration.

9. A thermal load prediction method for a mine shaft entrance according to claim 1, characterized in that, The parameter configuration scheme of the static aggregation point and the dynamic aggregation point is dynamically optimized based on the prediction result, including: Based on the measured heat load and the predicted value of the heat load, obtain the predicted value with the deviation amount greater than the preset threshold and define it as the deviation predicted value; Based on the deviation predicted value, obtain the corresponding device working parameters; Adjust the corresponding device working parameters from small to large according to the magnitude of the deviation predicted value.

10. A heat load prediction system for a mine shaft entrance, which is used to implement the heat load prediction method for a mine shaft entrance according to any one of claims 1-9, and is characterized in that, Include: Multi-source data prediction module: Collect the equipment operation time, transportation equipment trajectory, and mine temperature data, extract the equipment operation intersection time and start / stop time, and synchronously construct a spatio-temporal database; Thermodynamic feature optimization module: Based on the operation intersection time difference, calculate the static aggregation point of the fixed equipment through the first adjustment formula, and calculate the dynamic aggregation point of the transportation equipment through the second adjustment formula; Heat load prediction module: Integrate the static and dynamic aggregation point data, generate a prediction result using a hierarchical model architecture, and trigger a warning through edge computing; Prediction deviation compensation module: According to the prediction and measured deviation amount, use the joint loss function to inversely adjust the coefficient.

Images