Underground coal mine intelligent follow-up method and underground coal mine transportation system

Abstract

The invention relates to the technical field of underground coal mine transportation, and discloses an underground coal mine intelligent follow-up method and an underground coal mine transportation system. The method comprises the following steps: acquiring running state data such as running speed, load weight and position coordinates of transportation equipment; dividing stable and fluctuating operation intervals according to the change trend of the operation speed, and screening target transportation equipment by combining the load weight difference in the intervals; calculating a dynamic following index of each target device in the stable interval by combining the position coordinate change track and the running speed fluctuation characteristics of the target device; determining a reference following weight of each target device in the stable interval according to a load weight change trend, a position coordinate offset and a dynamic following index in the fluctuation interval; and screening the main control equipment based on the reference following weight, constructing a follow-up collaborative network in combination with the same target equipment data in the fluctuation interval, and generating an equipment dynamic adjustment strategy according to the follow-up collaborative network, so as to realize accurate collaborative regulation and control of the underground transportation equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of coal mine underground transportation, in particular to a coal mine underground intelligent follow-up method and a coal mine underground transportation system. BACKGROUND

[0002] In the production operation of a coal mine, the stable operation of the transportation equipment is directly related to the continuity and efficiency of the entire mining process. At present, the coal mine underground transportation system mostly adopts a fixed parameter control mode, that is, a unified operation threshold and adjustment strategy are set for the transportation equipment, without fully considering the differences in the actual operation process of different equipment. For example, part of the transportation equipment may have frequent fluctuations in running speed due to factors such as changes in the slope of the roadway, uneven distribution of coal seams, etc., while another part of the equipment may maintain a relatively stable running state for a long period of time. The existing control mode cannot effectively distinguish between these equipment with different running states, and often uses the same regulation and control standard, resulting in increased energy consumption during equipment operation, and may also cause equipment failure risk due to untimely regulation and control. In the scenario of multiple transportation equipment working together, the existing technology lacks comprehensive analysis and utilization of key running state data such as equipment load weight and position coordinates. When there is a load mutation or position deviation of the transportation equipment in a certain area, it is difficult to quickly identify the target equipment that needs to be focused on and to construct an effective collaborative control mechanism according to the running correlation between the equipment. This situation not only causes low overall running efficiency of the transportation equipment, but also may cause congestion of the transportation line due to insufficient coordination between the equipment, affecting the production progress of the coal mine. At the same time, the traditional manual monitoring and adjustment method relies on the experience judgment of the operator, and has problems such as response lag and low control precision, which is difficult to adapt to the complex and changeable working environment of the coal mine, and cannot meet the needs of intelligent mining operation for efficient and stable operation of the transportation system. With the advancement of intelligent transformation of the coal mining industry, the requirement for the automation control level of the underground transportation system is continuously improving, and the shortcomings of the existing transportation equipment control technology in adaptability, collaboration and precision are increasingly prominent, and an intelligent follow-up method that can dynamically adjust according to the actual running state of the equipment is needed to solve many problems faced by the current transportation system. SUMMARY

[0003] The present application relates to the technical field of coal mine underground transportation, in particular to a coal mine underground intelligent follow-up method and a coal mine underground transportation system.

[0004] To achieve the above-mentioned purpose, the present application provides a coal mine underground intelligent follow-up method, which comprises: Collecting running state data of underground transportation equipment in a coal mine, the running state data including equipment running speed, equipment load weight and equipment position coordinates; Dividing a stable running interval and a fluctuation running interval according to a change trend of the equipment running speed, and screening target transportation equipment based on a difference in the equipment load weight in the stable running interval and the fluctuation running interval; Under the target transportation equipment, combining a change trajectory of the equipment position coordinates with fluctuation characteristics of the equipment running speed, calculating a dynamic following index of each target transportation equipment in the stable running interval; According to a change trend of the equipment load weight in the fluctuation running interval and an offset amount of the equipment position coordinates, combining the dynamic following index, determining a reference following weight of each target transportation equipment in the stable running interval; Based on the reference following weight, screening the target transportation equipment in the stable running interval as a master control equipment, combining running state data of the same target transportation equipment in the fluctuation running interval to construct a follow-up coordination network, and generating a dynamic adjustment strategy of the transportation equipment according to the follow-up coordination network.

[0005] Preferably, the method for dividing the stable running interval and the fluctuation running interval according to the change trend of the equipment running speed comprises: Extracting time series data of the equipment running speed, and calculating a speed change rate of adjacent time points; Screening a continuous time interval with a speed change rate lower than a preset threshold as the stable running interval, and the rest intervals as the fluctuation running interval.

[0006] Preferably, the method for obtaining the dynamic following index comprises: Under the target transportation equipment, selecting equipment with running state data in the stable running interval as a reference equipment, and selecting equipment with running state data in the fluctuation running interval as a following equipment; Calculating a trajectory similarity of the reference equipment and the following equipment on the equipment position coordinates, combining a matching degree of the equipment running speed, and obtaining a dynamic influence factor of the reference equipment on the following equipment; Based on the correlation between the dynamic influence factor and the equipment load weight, calculating a dynamic following index of the reference equipment.

[0007] Preferably, the method for determining the reference following weight comprises: Extracting a load weight change amplitude of the following equipment in the fluctuation running interval, combining an offset distance of the equipment position coordinates, and calculating a dynamic adjustment demand of the following equipment; Normalizing a product of the dynamic following index and the dynamic adjustment demand to obtain the reference following weight.

[0008] Preferably, the method for constructing the follow-up coordination network comprises: Select the target transport equipment with the reference following weight higher than the preset weight threshold as the master control equipment; Calculate the difference of the running state data of the following equipment and the master control equipment in the fluctuation running interval, and determine the cooperative response degree of the following equipment to the master control equipment in combination with the adjustment trend of the equipment load weight; According to the order from high to low of the cooperative response degree, a hierarchical following relationship is constructed to form a follow-up cooperative network.

[0009] Preferably, the generation method of the dynamic adjustment strategy comprises: In the follow-up cooperative network, the running state data of the master control equipment and the first-level following equipment is taken as the first adjustment basis; The running state data of the second-level and below following equipment is taken as the second adjustment basis; Based on the priority relationship of the first adjustment basis and the second adjustment basis, the dynamic adjustment strategy of the transport equipment is generated.

[0010] Preferably, the priority of the first adjustment basis is higher than that of the second adjustment basis, and the generation method of the dynamic adjustment strategy further comprises: If there is a conflict between the first adjustment basis and the second adjustment basis, the data of the first adjustment basis is preferentially used for dynamic adjustment.

[0011] Preferably, the execution method of the dynamic adjustment strategy comprises: According to the cooperative response degree of each level in the follow-up cooperative network, the running parameters of the master control equipment and the following equipment are dynamically adjusted to ensure the cooperative operation of the transport equipment.

[0012] Preferably, the adjustment of the running parameters comprises synchronous optimization of the equipment running speed and balanced allocation of the equipment load weight.

[0013] Preferably, the present application further comprises a coal mine underground transportation system, which comprises: A running state acquisition module is used to acquire the running state data of the coal mine underground transport equipment in real time, and the running state data comprises the equipment running speed, the equipment load weight and the equipment position coordinates; A running interval division module is used to divide the stable running interval and the fluctuation running interval according to the change trend of the equipment running speed, and to select the target transport equipment based on the difference of the equipment load weight in the stable running interval and the fluctuation running interval; A dynamic following calculation module is used to calculate the dynamic following index of each target transport equipment in the stable running interval in combination with the change trajectory of the equipment position coordinates and the fluctuation characteristics of the equipment running speed under the target transport equipment; The benchmark following weight determination module is used for determining a benchmark following weight of each target transportation device in the stable operation interval according to a change trend of the device load weight in the fluctuation operation interval and a deviation of the device position coordinate, in combination with a dynamic following index; The follow-up cooperative network construction module is used for constructing a follow-up cooperative network based on the operation state data of the same target transportation device in the fluctuation operation interval, in combination with the target transportation device in the stable operation interval as the master control device based on the benchmark following weight; The dynamic adjustment strategy generation module is used for generating a dynamic adjustment strategy of the transportation device according to the follow-up cooperative network.

[0014] Compared with the prior art, the beneficial effects of the present application are: Through comprehensive collection and deep analysis of the operation state data of the transportation device, accurate identification and classification of the device operation state are realized. By dividing the stable operation interval and the fluctuation operation interval according to the change trend of the device operation speed, and in combination with the load weight difference to screen the target transportation device, the device that needs to be focused on can be quickly locked, and the problem of insufficient regulation and control pertinence caused by the traditional control mode of using a unified standard for all devices is avoided, so that the device regulation and control is more in line with the actual operation demand. After the target transportation device is determined, the dynamic following index is calculated in combination with the change trajectory of the device position coordinate and the operation speed fluctuation characteristics, which provides a scientific basis for subsequent accurate regulation and control. This index calculation method based on multidimensional data can fully reflect the actual operation characteristics of the device, compared with the traditional regulation and control method relying on only a single parameter, the accuracy of the device operation state judgment is effectively improved, which is helpful to develop a regulation and control strategy more in line with the actual demand of the device. By determining the benchmark following weight in combination with the change trend of the device load weight in the fluctuation operation interval, the position coordinate deviation and the dynamic following index, the regulation and control priority of different target transportation devices in the stable operation interval can be determined. Based on the benchmark following weight, the master control device is screened, which can ensure that in the multi-device cooperative operation scene, the key device affecting the overall operation efficiency is preferentially regulated and controlled, and in combination with the operation data of the same target device in the fluctuation interval, a follow-up cooperative network is constructed, realizing the cooperative regulation and control among the devices. This cooperative regulation and control mechanism can break through the limitation of traditional independent regulation and control of the device, so that multiple transportation devices form an organic whole. When a device appears abnormal operation, other associated devices can quickly respond and adjust according to the cooperative network, avoiding the risk of overall transportation system paralysis caused by single device failure, and improving the stability and reliability of the overall operation of the transportation device. The follow-up cooperative network and dynamic adjustment strategy constructed by the method can realize dynamic self-adaptive regulation and control of the transportation equipment, and can adjust the regulation and control mode in real time according to the change of the equipment operation state without manual intervention, thereby reducing the error and hysteresis of manual operation and reducing the working strength of the operator. Meanwhile, through the accurate regulation and control of the equipment operation state and the efficient cooperation between the equipment, the equipment operation parameters can be optimized, unnecessary energy consumption can be reduced, the probability of equipment failure can be reduced, and the service life of the equipment can be prolonged.

[0015] In the coal mine underground scene of multiple equipment cooperative operation, the method can effectively improve the overall operation efficiency of the transportation equipment, avoid the problem of transportation line blockage caused by insufficient cooperation between the equipment, and ensure the continuity of the coal mine underground production operation. Meanwhile, the application of the method can promote the development of the coal mine underground transportation system in the direction of intelligence and automation, meet the needs of the intelligent transformation of the coal mine industry, provide strong guarantee for the orderly development of the intelligent mining operation in the underground, and further improve the overall benefit of the coal mine underground production. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A working principle diagram of the coal mine underground intelligent follow-up method is described. Figure 2 A flowchart for stable and fluctuating operation interval division is described. Figure 3 A flowchart for follow-up cooperative network construction is described. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0018] Please refer to Figure 1The application provides an intelligent follow-up method for underground coal mines, which comprises the following steps: dividing an operation process into a stable operation interval and a fluctuation operation interval by monitoring the change trend of the operation speed of equipment in real time, and the division is based on the calculation of the speed change rate to ensure the accuracy of the interval identification. The target transportation equipment is screened out by using the difference in the load weight of the equipment in the stable operation interval and the fluctuation operation interval. After the target transportation equipment is determined, the dynamic following index of each target transportation equipment in the stable operation interval is calculated by combining the change trajectory of the equipment position coordinates and the fluctuation characteristics of the equipment operation speed, and the index reflects the collaborative potential of the equipment in the stable state. According to the change trend of the load weight of the equipment in the fluctuation operation interval and the offset of the equipment position coordinates, the reference following weight of each target transportation equipment in the stable operation interval is determined by combining the calculated dynamic following index, and the weight value is used to quantify the influence of the equipment in the collaborative network. The master control equipment is screened out based on the reference following weight, and the follow-up collaborative network is constructed by combining the operation state data of the same target transportation equipment in the fluctuation operation interval, the network integrates the interaction between the equipment through the hierarchical relationship, thereby generating a dynamic adjustment strategy to realize the intelligent follow-up control of the transportation equipment. The whole scheme emphasizes the data-driven interval division and weight calculation to ensure the improvement of the transportation efficiency and safety performance in the complex environment of underground coal mines.

[0019] Embodiment 1: refer to Figure 2 The method for dividing the stable operation interval and the fluctuation operation interval according to the change trend of the equipment operation speed relies on the detailed processing and pattern recognition of the continuous time sequence data, and the whole process starts from the continuous monitoring of the equipment operation speed by the operation state acquisition module. The speed data is collected by the sensors installed on the transportation equipment, which record the instantaneous speed of the equipment at fixed time intervals to form a speed curve changing with time, and the integrity and accuracy of the data collection are the basis for all subsequent analysis. The original speed data needs to be preprocessed to eliminate interference, and the preprocessing includes filtering and smoothing processing, and the moving average method or low-pass filter is used to suppress random noise and short-time mutations, so that the overall trend of the speed change is more clear and identifiable. The speed time sequence after preprocessing is sent to the calculation unit for change rate analysis, and the calculation of the speed change rate of adjacent time points is the core step, and the speed change rate sequence at each time is obtained by calculating the difference between the speed value at the current time point and the speed value at the previous time point, and then dividing the time interval between the two time points. The change rate sequence directly reflects the smoothness of the equipment operation and is the direct basis for distinguishing different operation intervals.

[0020] The setting of the preset threshold is a process that needs to consider historical operation data and actual working conditions. The threshold is usually not a fixed value but a numerical range that is dynamically adjusted according to the type of equipment, the condition of the roadway, and the different transportation tasks. For example, in the transportation section with frequent load changes, the threshold may be set relatively loose to accommodate larger normal fluctuations, while on the fixed line that requires smooth operation, the threshold will be set more strictly. After determining the threshold, the system starts to scan the speed change rate sequence and identify those continuous time segments whose change rate is continuously below the preset threshold. These segments are marked as potential stable running intervals. The identification process uses a sliding window algorithm, and the window size is set according to the sampling frequency to ensure that statistically significant stable periods can be captured. For a continuous interval, not only is the change rate of each point within the interval required to be below the threshold, but the duration of the interval is also required to reach the minimum length threshold to avoid misjudging occasional short-term stability as a stable interval. The fluctuation running interval is defined as all time segments outside the stable interval, especially those with a speed change rate significantly higher than the threshold. These intervals usually correspond to the start, acceleration, deceleration, stop or external disturbance of the equipment.

[0021] The results of interval division need to be further verified and corrected. The system will cross-verify the divided stable intervals with the load data and location information of the equipment at that time. If an interval meets the stable characteristics in terms of speed change rate, but the load weight fluctuates sharply or the location coordinates jump abnormally at the same period, the interval may need to be re-evaluated and may be classified as a fluctuation interval or treated as a special working condition. This multi-dimensional verification mechanism improves the reliability of interval division and avoids misjudgment that may be caused by a single speed indicator. The divided interval information, including the start time, end time, interval type (stable or fluctuation), and statistical characteristics such as average speed, speed variance within the interval, is completely recorded and stored in the time series database. These interval records with rich metadata provide a structured time frame and feature labels for subsequent equipment screening, index calculation, and weight determination.

[0022] The entire interval division process is designed as a real-time running stream processing module that can continuously receive new speed data points and update the status of the current running interval in real time. This real-time capability enables the system to respond promptly to changes in device operating status, providing near real-time decision-making basis for subsequent follow-up control. The processing module also includes a feedback mechanism that can automatically or with operator intervention adjust the threshold parameters when the system finds that the interval divided based on the current threshold deviates significantly from the actual situation, achieving adaptive optimization of the threshold. In addition, to deal with data loss or transient interference caused by the complex electromagnetic environment underground, the module also designs data completion and abnormal recovery logic to ensure that the interval can still be accurately divided as much as possible in the case of missing individual data points. Through the above series of careful data processing and analysis steps, the change trend of the device running speed is effectively converted into the explicit identification of stable and fluctuating running intervals, laying a solid time reference and state classification foundation for the subsequent processes of the entire intelligent follow-up method.

[0023] The application effect of this division method depends largely on the rationality of the initial parameter settings, so the system needs a parameter tuning phase at the initial deployment. During this period, the operation and maintenance personnel can calibrate key parameters such as preset threshold, sliding window size, and minimum stable duration, combined with historical operation logs and actual operation experience. The system also records the division results under different parameter settings to assist personnel in comparison and selection. Once the parameters are determined, the entire division process can run automatically without the need for continuous human intervention. For multiple transport devices that may be running simultaneously in a large coal mine, this method supports parallel processing, dividing intervals for multiple speed data streams simultaneously through distributed computing resources to ensure that the system processing capability can keep up with the speed of data generation. The division results are not only used for the subsequent steps of this invention, but can also be independently output as a report for device running status evaluation, helping management personnel understand the running stability of the device at different times, and providing valuable data insights for device maintenance and scheduling.

[0024] Embodiment 2: The core of the acquisition method for the dynamic following index is to quantify the degree of influence of one device on another device in a fluctuating state, which is calculated through multi-dimensional data fusion. The implementation process starts with the clear definition of the target device, which is the result of the screening completed in the previous step. The system will mark the devices that exhibit complete and coherent running state data in the stable running interval as reference devices, and their behavior in the specified time period is considered to have reference value and stability. At the same time, devices with running state data in the fluctuating running interval are defined as following devices, whose running trajectories and parameters exhibit varying degrees of uncertainty or change. The correspondence between reference devices and following devices is not always fixed, and the system will dynamically establish and dissolve this association according to real-time running intervals and data availability. A device may serve as a reference device in one time period and become a following device for other devices in another time period.

[0025] Calculating the similarity of the trajectories of two devices in position coordinates is the first step in obtaining the dynamic influence factor. Position data comes from positioning systems deployed in the roadway or sensors such as odometers and inertial measurement units provided by the devices themselves. The comparison of trajectory similarity is not a simple comparison of position points, but rather an examination of the consistency of the shape, direction, and change pattern of the motion path of both in the entire running interval. Since devices may not pass through the same position at the same time, the algorithm needs to have the ability to nonlinearly align and compare time series. One commonly used method is to examine the similarity of the shapes of the two trajectories, ignoring their absolute positions and strict correspondence in time, which can eliminate the effects of different start times or initial position differences. Another auxiliary judgment is to analyze the local features of the trajectory, such as turning radius, straight line segment length, and distribution of stopping points, which are geometric and topological properties that can more deeply reveal the internal connection of the running mode. The evaluation of speed matching degree focuses on analyzing the coordination of speed change rhythm and amplitude of two devices in comparable time periods. Even if the absolute speed values are different, if the acceleration and deceleration behaviors occur in a correlated manner in time, it may indicate a certain following relationship. The matching analysis of the speed sequence needs to normalize the two time series to eliminate dimension and reference differences, and then calculate statistical quantities such as dynamic time warping distance or correlation coefficient.

[0026] The calculation of the dynamic influence factor is a comprehensive embodiment of trajectory similarity and speed matching degree. Different weight coefficients are usually assigned to these two components for fusion. The setting of the weight needs to consider the priority of the specific application scenario. In the underground transportation environment, the similarity of the trajectory may be more reflective of the spatial dependence between devices than the instantaneous matching of the speed. Therefore, the weight of the trajectory similarity may be set higher. The fused dynamic influence factor is a dimensionless value, and its size directly reflects the strength of the traction or influence that the behavior pattern of the reference device may have on the following device. The higher the factor value, the better the coordination of the two devices. The correlation between this dynamic influence factor and the load weight of the device needs to be investigated next. The load data is obtained through weighing sensors or motor current estimation. The analysis of the law of the change of the dynamic influence factor with the change of the load can reveal the role played by the load in the interaction between devices. If it is found that the change of the load weight of the reference device significantly affects the dynamic influence factor between it and the following device, it means that the load is an important correlation parameter and must be considered in the calculation of the final index.

[0027] The generation of the dynamic following index is a concentrated embodiment of all the above analysis results. It integrates the dynamic influence factor and the load correlation into a single index through a suitable mathematical model. This model can be linear or nonlinear. The calculation of the index needs to ensure the unity of the dimension and the reasonableness of the range. The result is usually normalized to a certain interval between zero and one to facilitate the comparison between different devices and the subsequent weight calculation. The entire calculation process is encapsulated as a software module that can be called repeatedly. This module receives the historical running data of the reference device and the following device from the database, and outputs the dynamic following index value after a series of internal operations. The module contains data validity checking logic inside, which can automatically identify and process common data problems such as data missing, timestamp misplacement, and sensor abnormality, to ensure the reliability of the output result.

[0028] In order to adapt to the dynamic changes of the downhole environment, the dynamic following indicators are not calculated once, but as a periodic updating process, the updating frequency is consistent with or an integer multiple of the acquisition frequency of the running state data. The system will maintain a dynamic following indicator list, which records and updates the indicator values of all potential device pairs in the network in real time. This list constitutes the data basis for the construction of the follow-up cooperative network. In terms of specific technical implementation, trajectory similarity calculation may involve a large number of spatial geometric operations, which has certain requirements for the computing power of the processor. In actual deployment, optimization algorithms or special hardware need to be considered to meet the real-time requirements. Speed matching analysis may use related technologies in the field of signal processing to extract effective matching features from speed data that may have noise. The entire process of obtaining dynamic following indicators embodies the idea of multi-source information fusion, organically combining different types of data such as position, speed, and load to form a quantitative indicator that can comprehensively evaluate the coordination potential between devices. The correctness and stability of this indicator directly affect the accuracy of the subsequent benchmark following weight and the final performance of the entire follow-up control system.

[0029] Robustness of the algorithm also needs to be considered during implementation. For example, when the trajectory data of a following device is interrupted due to loss of positioning signals, the system should automatically switch to an estimation mode based on the remaining valid data segments, or directly mark the data in this time period as unavailable to avoid misleading indicators. For load weight correlation analysis, attention needs to be paid to the hysteresis effect and nonlinear characteristics of load changes. Sometimes, simple linear correlation analysis may not be sufficient to characterize the true relationship, and more advanced statistical analysis tools need to be introduced. The dynamic following indicator module usually provides a debugging interface, allowing developers or operators to view intermediate calculation results such as trajectory similarity detail comparison charts and speed sequence matching conditions, which is very helpful for algorithm parameter tuning and problem troubleshooting. Through the above detailed design and implementation, the acquisition of dynamic following indicators not only completes the calculation tasks specified by the method, but also provides an effective means for the intelligent follow-up system to understand the dynamic relationship between devices.

[0030] Example 3: The core of the determination method for the benchmark following weight lies in the combination of the dynamic adjustment needs of the device in the fluctuation operation interval and the dynamic following index calculated in the stable operation interval to form a comprehensive quantitative evaluation value. The implementation process begins with the extraction and analysis of the following device data in the fluctuation operation interval. The system first retrieves all the running state data records of the target following device in the identified fluctuation interval from the time series database. The calculation of the load weight variation amplitude is the first step. The amplitude is not simply the difference between the maximum and minimum values, but needs to consider the degree of change and the change pattern of the load in the entire fluctuation interval. The system calculates the statistical dispersion of the load weight sequence, such as the standard deviation or the mean absolute deviation, and also analyzes the trend slope of the load change, whether it is a rapid rise, a slow decline, or a sharp oscillation. The calculation of the device position coordinate offset is more complex. The offset is not the displacement relative to a fixed origin, but specifically refers to the deviation of the device's movement trajectory in the fluctuation operation interval from the "normal" movement pattern exhibited in the previous stable operation interval. This deviation can manifest as path bending, detouring, stagnation, or reverse movement.

[0031] To quantify this deviation, the system needs to first establish the benchmark movement pattern of the device in the stable operation interval, which can be an ideal path or a movement vector field. The calculation of the offset can be represented as: where the integral symbol represents the cumulative calculation over the entire time period, and represent the start time and end time of the fluctuation operation interval, respectively. The double vertical bar symbol represents the Euclidean norm of the calculated vector, i.e., the length of the vector. The vector function represents the actual movement speed vector of the device at time , which contains both the speed and direction information derived from the device's running speed and position coordinates. The vector function represents the expected movement speed vector at the same time , based on the benchmark movement pattern established in the stable interval. The Greek letter (Lambda) represents the final offset result calculated. This formula calculates the cumulative amount of the difference between the actual movement and the expected movement in the speed vector over the entire fluctuation interval, which can comprehensively reflect the comprehensive offset of the device in terms of direction and speed.

[0032] After obtaining the two basic quantities, the load weight variation range and the position coordinate offset, they need to be synthesized into a value representing the urgency of adjusting the current state of the device, that is, the dynamic adjustment demand. The synthesis process usually involves weighting, and the allocation of weights reflects the different emphasis on load stability and path tracking accuracy in different application scenarios. For example, in the area of heavy load transportation and narrow roadway space, the weight of position offset may be set higher, because a small path deviation may cause collision risk; while in the scene of light load but requiring throughput, the weight of load variation may be higher, because it directly affects the transportation efficiency. The larger the value of the dynamic adjustment demand, the farther the current running state of the device deviates from its ideal or expected mode, and the more it needs to be brought back on track through external intervention or adjustment.

[0033] The system will obtain the dynamic following index of the device from the dynamic following calculation module, which quantifies the coordination potential between the device as a follower and the reference device. The combination of the dynamic following index and the dynamic adjustment demand adopts a multiplication model, and its internal logic is: a device with high coordination potential (high dynamic following index) should be given higher attention and adjustment priority if its current state deviates from the expected degree is also large (high dynamic adjustment demand), so the reference following weight is also correspondingly higher. The multiplication model can amplify the effect of the combination of "high potential, high deviation", so that it stands out in the weight ranking. The original weight value obtained by direct multiplication may have a large difference in orders of magnitude, and it is difficult to compare directly between different devices, so it must be normalized. Normalization aims to map the original weight values of all devices to a unified and comparable range, and the most commonly used is the zero to one interval. Normalization processing needs to be calculated based on the original weight value set of all devices to be evaluated in the current system, and the commonly used method includes the minimum-maximum value normalization, that is, finding the minimum and maximum values in the set, and then linearly scaling each original weight to the target interval. The final result after normalization is the reference following weight, which comprehensively reflects the inherent coordination properties and real-time state deviation of the device, and provides a scientific basis for subsequent master device selection.

[0034] The whole weight determination process is designed as a configurable computing pipeline. The operation and maintenance personnel can fine-tune the system behavior by adjusting the weight parameters when the synthetic dynamic adjustment requirements, selecting different normalization methods, etc. to adapt to the specific characteristics of different mines and different transport lines. The computing module has fault tolerance capability. When a certain follower device lacks sufficient stable interval data to establish the baseline motion pattern, the system can use the global average pattern or the typical pattern of similar devices as a substitute to estimate the offset, ensuring the continuity of weight calculation. The update cycle of the reference follower weight is synchronized with the division of the running interval and the calculation of the dynamic following index, ensuring the consistency of the data in the whole decision chain. The determination of the weight value converts the abstract relationship between devices and unstable running state into a specific and sortable numerical value, enabling the system to objectively identify which devices need to be paid more attention to in the current fluctuating environment and which devices have the potential to become stable pivots in the coordination network, thus taking a key step towards building an efficient and reliable follow-up coordination network.

[0035] Embodiment 4: refer to Figure 3 The construction of the follow-up coordination network starts from the selection from the device group whose baseline following weight has been calculated. The system sets a preset weight threshold, which is usually determined according to the weight distribution of devices with higher coordination efficiency in historical running data. All devices whose baseline following weight exceeds the threshold are officially determined as master devices. These devices have shown higher dynamic following index in the previous stable running interval and have significant adjustment requirements in the current fluctuating running interval, so they are given the responsibility of leading the coordination network. Next, the running state data difference between each non-master device and each master device in the fluctuating running interval needs to be calculated. The difference calculation covers three core parameters: device running speed, device load weight, and device position coordinates, refer to Table 1.

[0036] Table 1: Device running state difference and coordination response calculation

[0037] The adjustment trend of the load weight of the device is determined by analyzing the slope or the tendency of the curve fitting of the load of the follower device within the fluctuation interval over time. The trend coefficient is quantified as a value between -1 and 1. A positive value indicates that the load tends to increase, and a negative value indicates that the load tends to decrease. The absolute value size indicates the degree of obviousness of the trend. The determination of the cooperative response degree is a comprehensive calculation process. It normalizes and weights the speed difference, load weight difference, and position coordinate offset to obtain a preliminary difference comprehensive value. Then, the comprehensive value is combined with the load adjustment trend coefficient. The combination needs to consider the consistency of the direction of the trend and the expected direction of the command of the master device. For example, if the master device needs to increase the load and the follower device also presents a load increase trend, the cooperative response degree will be positively enhanced.

[0038] According to the order of the cooperative response degree from high to low, the system establishes a queue of follower devices for each master device. The follower devices with the highest cooperative response degree are determined as the first-level follower devices of the master device. They are most synchronized with the running state of the master device, and the expected effect of the response command is the best. The devices with the second highest response degree are listed as the second-level follower devices, and so on, thereby forming a tree or hierarchical network structure with the master device as the root node, i.e., the follow-up cooperative network. In this network, the connection relationship is not absolutely unique. A follower device may have calculated cooperative response degrees with multiple master devices at the same time, but finally it will only be assigned to the level of the master device with the highest cooperative response degree. This ensures the clarity of the network structure and the uniqueness of the command transmission. After the follow-up cooperative network is constructed, the generation of the dynamic adjustment strategy immediately expands on the basis of this network structure. The strategy generation clearly distinguishes the priority of the data basis. The running state data of the master device itself and the data of the first-level follower devices directly controlled by it are jointly used as the first adjustment basis. The data of the first adjustment basis has the highest real-time and importance because they directly reflect the running condition of the core layer of the network. The data of the second-level follower devices, third-level follower devices, and other lower-level devices are collectively referred to as the second adjustment basis. These data are used to evaluate the coordination of the whole network and discover potential systematic deviations.

[0039] The generation process of the dynamic adjustment strategy is a priority-based synthetic logic. The strategy engine first analyzes the data in the first adjustment basis, such as the speed and load information of the master device T-01 and its first-level follower device F-03, to generate core adjustment instructions for the sub-network, such as fine-tuning the operating speed of T-01 to achieve better synchronization with F-03. Subsequently, the system scans the data in the second adjustment basis to check the status of the second-level follower device F-05 and other devices. If it is found that the status of F-05 deviates greatly from the expected effect of the instructions generated based on the first basis, auxiliary calibration instructions will be generated. However, when the second adjustment basis analysis results conflict with the core instructions generated by the first adjustment basis, the decision logic clearly stipulates that the conclusion of the first adjustment basis should be given priority. For example, if it is judged according to the data of the master device T-01 and the first-level follower device F-03 that the operating speed needs to be slightly increased, but the data of the second-level follower device F-05 shows that it is currently under heavy load and cannot be accelerated, the system will still give priority to the speed-up instruction, and generate an independent and relatively mild speed-up suggestion or warning for F-05, without negating the core speed-up decision. This priority setting ensures that the stable core of the collaborative network can be adjusted first, avoiding the impact of the overall decision-making efficiency due to the abnormality of the edge devices.

[0040] The execution of the dynamic adjustment strategy is based on the fact that the follow-up collaborative network has been built, and the hierarchical relationship and corresponding collaborative response degree between the master device, the first-level follow-up device, and the second-level and below follow-up devices in the system have been determined. The core principle of strategy execution is strict adherence to priority, that is, the priority of the first adjustment basis is absolutely higher than the second adjustment basis. After the strategy generation module outputs a preliminary adjustment scheme, the conflict detection mechanism will start immediately. The mechanism compares the core instructions calculated based on the first adjustment basis in the preliminary scheme with the global state reflected by the second adjustment basis. The typical scenario of conflict may be the mismatch of speed instructions. For example, according to the real-time data of the master device T-01 and its first-level follow-up device F-03, the system calculates that adjusting the running speed to 1.5 meters per second can optimize the transportation efficiency of the current section, which is the conclusion of the first adjustment basis. However, the second adjustment basis data shows that the second-level follow-up device F-05 is abnormally heavy in load and its motor performance is temporarily limited, and its current maximum stable running speed can only reach 1.3 meters per second. If F-05 is forced to follow the speed of 1.5 meters per second, it may cause the motor to overload or the control system to be unstable. At this time, the conflict is obvious, and the conflict resolution logic will unconditionally adopt the instruction of the first adjustment basis, that is, maintain the core speed target of 1.5 meters per second. For the conflict of F-05 with the core instruction, the system will not modify the core instruction, but will generate an independent, secondary compensatory instruction for F-05, such as instructing F-05 to make its best effort to rise to its safe upper limit of 1.3 meters per second per second, while sending an alarm information to the dispatch center, prompting that there is a device performance bottleneck that may need manual intervention. This processing method ensures that the collaborative efficiency of the network core layer is not dragged down by the problems of the edge devices, while giving necessary attention to abnormal situations.

[0041] The system begins to dynamically adjust the operating parameters of the master device and the follower device according to the cooperative response degree of each level in the follow-up cooperative network. The cooperative response degree plays a key role in adjusting the amplitude and timing here. For devices with high cooperative response degree, the adjustment instruction will be more direct and rapid, and it is expected to closely follow the pace of the master device. For devices with low cooperative response degree, the adjustment will be more gradual and cautious, allowing longer response time and larger tolerance range. The synchronization optimization of device operating speed is the primary task of parameter adjustment. The system does not simply set all devices to the same target speed, but constructs a speed distribution based on the master device speed, considering the level and response degree. The speed of the master device is set by the system according to the global working condition or adjusted by itself according to its own state priority. The speed instruction of the first-level follower device is attached with a dynamic adjustment item based on the real-time speed difference between the master device and the first-level follower device and its cooperative response degree. The role of this adjustment item is to quickly narrow the gap and achieve close following. The speed of the second-level follower device may be synchronized with the first-level follower device as the reference benchmark, rather than directly targeting the master device, thus forming a level transmission chain of speed instructions, avoiding system oscillation caused by all devices chasing a single target at the same time.

[0042] The balanced allocation of device load weight is another key operating parameter adjustment, which is particularly important when multiple devices cooperate to transport long distances or there are load transfer points. Load balancing does not pursue absolute average, but dynamically allocates according to the real-time carrying capacity, position and cooperative response degree of the device. The system will monitor the load rate of each device in the network in real time. If it finds that the load rate of a device is significantly higher than that of other devices of the same level, and there is an opportunity to distribute the load on the path, the system will generate a load balancing suggestion instruction. For example, the master device T-01 has a light load, while the first-level follower device F-03 has a high cooperative response degree and a heavy load, and a filling point that can adjust the feed amount by controlling the distributor is approaching. The system may instruct to appropriately increase the loading amount of T-01 and slightly reduce the loading amount of F-03 in the next filling period, so that the load rates of the two devices tend to be close. For devices with low cooperative response degree, if their load is abnormal, the system may prefer to take the form of warning or suggesting path adjustment rather than directly performing complex real-time load scheduling, because this may introduce uncertainty.

[0043] The whole execution process is a closed-loop control. The adjustment instructions are issued to the local controllers of each device through the downhole communication network for execution. The effects after execution, such as new running speed and actual load data, are fed back to the system through the running state acquisition module. The system compares the difference between the expected effect of the instruction and the actual feedback data to evaluate the effectiveness of the adjustment strategy and fine-tune the subsequent instructions. This continuous feedback-evaluation-adjustment mechanism enables the dynamic follow-up control to adapt to the complex and variable environment underground, gradually reduces the differences in running states between devices, and improves the smoothness and efficiency of the whole transportation system. The execution module also records all the issued instructions and their results, providing data support for analyzing system performance and optimizing control algorithms. Through this execution method based on priority conflict resolution and hierarchical parameter adjustment, the intelligent follow-up method can be transformed from a theoretical model into a practical and usable control logic, driving the underground transportation devices to work collaboratively like a well-trained team.

[0044] It should be noted that the relational terms herein such as first and second and the like are used solely to distinguish one from another entity or action, without necessarily requiring or implying that there is any such relationship or order between such entities or actions. Also, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0045] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements and variations can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. An intelligent follow-up method in a coal mine underground, characterized by, The method comprises the following steps: Collecting operation state data of underground coal mine transportation equipment, the operation state data comprising equipment operation speed, equipment load weight and equipment position coordinates; Dividing stable operation intervals and fluctuation operation intervals according to the change trend of the equipment operation speed, and screening target transportation equipment based on the difference in the equipment load weight in the stable operation intervals and the fluctuation operation intervals; Under the target transportation equipment, combining the change trajectory of the equipment position coordinates with the fluctuation characteristics of the equipment operation speed, calculating the dynamic following index of each target transportation equipment in the stable operation intervals; According to the change trend of the equipment load weight in the fluctuation operation intervals and the offset of the equipment position coordinates, combining the dynamic following index, determining the reference following weight of each target transportation equipment in the stable operation intervals; Based on the reference following weight, screening the target transportation equipment in the stable operation intervals as the master control equipment, combining the operation state data of the same target transportation equipment in the fluctuation operation intervals to construct a follow-up coordination network, and generating a dynamic adjustment strategy of the transportation equipment according to the follow-up coordination network.

2. The intelligent following method for underground coal mine of claim 1, wherein, The method for dividing the stable operation intervals and the fluctuation operation intervals according to the change trend of the equipment operation speed comprises: Extracting time series data of the equipment operation speed, and calculating the speed change rate of adjacent time points; Screening a continuous time interval with a speed change rate lower than a preset threshold as a stable operation interval, and screening the remaining intervals as fluctuation operation intervals.

3. The intelligent following method for underground coal mine of claim 2, wherein, The method for obtaining the dynamic following index comprises: Under the target transportation equipment, selecting equipment with operation state data in the stable operation intervals as reference equipment, and selecting equipment with operation state data in the fluctuation operation intervals as following equipment; Calculating the trajectory similarity of the reference equipment and the following equipment on the equipment position coordinates, combining the matching degree of the equipment operation speed, and obtaining a dynamic influence factor of the reference equipment on the following equipment; Based on the correlation between the dynamic influence factor and the equipment load weight, calculating the dynamic following index of the reference equipment.

4. The intelligent following method for underground coal mine of claim 3, wherein, The method for determining the reference following weight comprises: Extracting the load weight change amplitude of the following equipment in the fluctuation operation intervals, combining the offset distance of the equipment position coordinates, and calculating the dynamic adjustment demand of the following equipment; Normalizing the product of the dynamic following index and the dynamic adjustment demand to obtain the reference following weight.

5. The intelligent following method for underground coal mine of claim 4, wherein, The method for constructing the follow-up coordination network comprises: Selecting target transportation equipment with a reference following weight higher than a preset weight threshold as master control equipment; Calculating the operation state data difference of the following equipment and the master control equipment in the fluctuation operation intervals, combining the adjustment trend of the equipment load weight, and determining the cooperative response degree of the following equipment to the master control equipment; Constructing a hierarchical following relationship in the order from high to low according to the cooperative response degree to form a follow-up coordination network.

6. The intelligent following method for underground coal mine of claim 5, wherein, The method for generating the dynamic adjustment strategy comprises: In the follow-up coordination network, taking the operation state data of the master control equipment and the first following equipment as the first adjustment basis; Taking the operation state data of the second and subsequent following equipment as the second adjustment basis; Based on the priority relationship of the first adjustment basis and the second adjustment basis, generating a dynamic adjustment strategy of the transportation equipment.

7. The intelligent following method for underground coal mine of claim 6, wherein, The priority of the first adjustment basis is higher than that of the second adjustment basis, and the method for generating the dynamic adjustment strategy further comprises: If the first adjustment basis conflicts with the second adjustment basis, the data of the first adjustment basis is preferentially used for dynamic adjustment.

8. The intelligent following method for underground coal mine of claim 7, wherein, The execution method of the dynamic adjustment strategy comprises: According to the cooperative response degree of each level in the follow-up cooperative network, the operating parameters of the master device and the follower device are dynamically adjusted to ensure the cooperative operation of the transportation device.

9. The intelligent following method for underground coal mine of claim 8, wherein, The adjustment of the operating parameters includes the synchronization optimization of the device operating speed and the balanced allocation of the device load weight.

10. An underground coal mine haulage system characterised in that, It comprises: An operating state acquisition module is configured to acquire operating state data of the underground coal mine transportation device in real time, wherein the operating state data comprises device operating speed, device load weight and device position coordinates; An operating interval division module is configured to divide stable operating intervals and fluctuating operating intervals according to the change trend of the device operating speed, and to select target transportation devices based on the difference in the device load weight in the stable operating intervals and the fluctuating operating intervals; A dynamic following calculation module is configured to calculate the dynamic following index of each target transportation device in the stable operating intervals in combination with the change trajectory of the device position coordinates and the fluctuation characteristics of the device operating speed under the target transportation device; A reference following weight determination module is configured to determine the reference following weight of each target transportation device in the stable operating intervals in combination with the dynamic following index according to the change trend of the device load weight in the fluctuating operating intervals and the offset of the device position coordinates; A follow-up cooperative network construction module is configured to select the target transportation devices in the stable operating intervals as master devices based on the reference following weight, and to construct a follow-up cooperative network in combination with the operating state data of the same target transportation devices in the fluctuating operating intervals; A dynamic adjustment strategy generation module is configured to generate a dynamic adjustment strategy of the transportation device according to the follow-up cooperative network.

Images