Advancing type working face propulsion control method and system

Abstract

The invention belongs to the field of coal mining, and discloses an advancing type working face propulsion control method and system, which dynamically optimizes the operation time sequence and action parameters of each device by deeply sensing the operation parameters and energy consumption characteristics of the device in combination with load trend prediction and risk assessment, thereby effectively stabilizing the overall instantaneous total power load of the working face, and improving the working efficiency. 'Peak clipping and valley filling' of energy consumption is realized, the overall instantaneous total power load of a working face can be effectively stabilized, the impact of a power supply system is reduced, the energy utilization efficiency is improved, and the service life of equipment is prolonged.

Description

Technical Field

[0001] This application relates to the field of coal mining, and more specifically, to a method and system for controlling the advancement of a forward working face. Background Technology

[0002] In automated longwall mining faces, continuous coal production relies on the close coordination of the mining machine, scraper conveyor, and hydraulic supports. The mining machine is responsible for cutting the coal seam, and the cut coal is transported out by the scraper conveyor, while the hydraulic supports provide roof support and propel the face equipment forward.

[0003] Currently, to achieve the goal of low-carbon and high-efficiency production, each core piece of equipment has achieved localized optimization of energy consumption at its own level. However, in the analysis of actual long-term operating data, the overall instantaneous total electrical energy consumption of the working face often exhibits significant and large-scale periodic fluctuations. These fluctuations are not entirely caused by the failure of a single piece of equipment or drastic changes in geological conditions, but are related to the coordinated operation modes between various pieces of equipment.

[0004] The fundamental reason for this peak in total energy consumption is the concentration or overlap of instantaneous load peaks in multiple high-power equipment during the production cycle. For example, when a coal mining machine is cutting hard coal seams or encountering interbedded rock, the load on its cutting motor will increase sharply, reaching an instantaneous power peak. When a scraper conveyor is transporting a large amount of coal or encountering increased local resistance, its drive motor may also reach full load or even overload. When hydraulic supports perform concentrated support movement, especially when a large number of supports simultaneously or in batches perform high-energy-consuming actions such as lowering, moving, and raising supports within a short period of time, the power demand of its hydraulic pump station will surge instantaneously.

[0005] Existing control systems typically perform sequential control based on pre-defined work processes and the geometrical relationships between equipment. For example, when a coal mining machine completes a cutting cycle and passes a set of hydraulic supports, the control system immediately issues a shift command to that set or batch of supports. The scraper conveyor continues to operate during coal cutting and continues transporting after the shift is completed. This rigid, event-triggered control logic lacks the ability to globally perceive and predict the instantaneous power load of the entire working face. This leads to situations where, during certain production periods, the coal mining machine is under high-load cutting, the scraper conveyor is running at full load, and the hydraulic support group happens to simultaneously initiate a concentrated shift. In this case, the instantaneous power demand of the coal mining machine, scraper conveyor, and hydraulic supports—the "three machines" system—simultaneously reaches or approaches its peak, forming a huge peak in total power demand.

[0006] The simultaneous or near-simultaneous peak load of multiple high-power devices significantly impacts the power supply system at the work site. First, instantaneous high power demand can cause voltage drops, affecting the stable operation of electrical equipment. Second, during power transmission, higher instantaneous peak loads typically result in greater line losses and lower power supply efficiency. This is because when cables and transformers transmit large currents, the heat loss generated by their internal resistance is proportional to the square of the current; peak currents significantly increase instantaneous losses. Furthermore, frequent instantaneous high-power surges accelerate the wear and tear on power transmission equipment, transformers, and internal electrical components at the work site (such as motors, cables, and switchgear), shortening their lifespan and increasing maintenance costs and the risk of failure.

[0007] Currently, although some systems possess real-time monitoring capabilities for the total power of the working face, these are primarily used for overload protection or historical data recording, rather than proactively smoothing the load. These systems cannot dynamically adjust the operating sequence and pace of each piece of equipment based on predictions of future load trends. Therefore, existing methods cannot proactively and proactively adjust the timing of support relocation, or, when necessary, fine-tune the traction speed of the coal mining machine or the operating speed of the scraper conveyor to stagger or smooth the load peaks of these devices. This control strategy, lacking global load prediction and coordinated scheduling capabilities, cannot effectively achieve "peak shaving and valley filling" of the overall energy consumption of the working face, limiting further improvements in low-carbon and high-efficiency production levels. The core issue is that existing methods treat the operation of each piece of equipment as independent events or events coordinated solely based on geometric positional relationships, neglecting the dynamic coupling relationship of the instantaneous power load of each piece of equipment, and the potential to smooth the total energy consumption peak by optimizing this coupling relationship.

[0008] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0009] To address the shortcomings of existing technologies, this application provides a forward-moving working face propulsion control method and system, which can effectively mitigate the overall instantaneous total power load of the working face, reduce the impact on the power supply system, improve energy utilization efficiency, and extend the service life of the equipment.

[0010] In a first aspect, this application provides a method for controlling the advancement of an automated longwall mining face, the method comprising the following steps:

[0011] A1. Obtain the operating parameters of the coal mining machine, scraper conveyor, and hydraulic support to identify the current production cycle of the working face;

[0012] A2. Obtain the energy consumption characteristic parameters of the coal mining machine, scraper conveyor and hydraulic support under different production cycles, as well as the safe upper limit of the total power of the working face;

[0013] A3. Based on the current production cycle, the operating parameters, geological conditions, work plan, and energy consumption characteristic parameters, predict the instantaneous power load change trend of the coal mining machine, scraper conveyor, and hydraulic support in the future period, and assess the superposition risk of the overall instantaneous total power load of the working face in conjunction with the total power safety limit of the working face.

[0014] A4. Based on the instantaneous power load change trend, the superimposed risks, and the safe upper limit of the total power of the working face, adjust the operating sequence and action parameters of the coal mining machine, scraper conveyor, and hydraulic support to smooth out the overall instantaneous total power load of the working face.

[0015] Secondly, this application provides a forward-moving face propulsion control system for controlling the propulsion process of an automated longwall coal mining face. The system includes:

[0016] The cycle time recognition module is used to acquire the operating parameters of the coal mining machine, scraper conveyor and hydraulic support, in order to identify the current production cycle time of the working face;

[0017] The feature acquisition module is used to acquire the energy consumption characteristic parameters of the coal mining machine, scraper conveyor and hydraulic support under different production cycles, as well as the safe upper limit of the total power of the working face.

[0018] The load trend prediction module is used to predict the instantaneous power load change trend of the coal mining machine, scraper conveyor and hydraulic support in the future period based on the current production cycle, the operating parameters, geological condition information, work plan and energy consumption characteristic parameters, and to assess the superposition risk of the overall instantaneous total power load of the working face in combination with the total power safety limit of the working face.

[0019] The collaborative scheduling module is used to adjust the operating sequence and action parameters of the coal mining machine, scraper conveyor and hydraulic support according to the instantaneous power load change trend, the superimposed risk and the safe upper limit of the total power of the working face, so as to smooth the overall instantaneous total power load of the working face.

[0020] In summary, the forward-moving working face propulsion control method and system provided in this application, through in-depth perception of equipment operating parameters and energy consumption characteristics, combined with load trend prediction and risk assessment, dynamically optimizes the operation sequence and action parameters of each piece of equipment, thereby effectively smoothing the overall instantaneous total power load of the working face, achieving "peak shaving and valley filling" of energy consumption, effectively smoothing the overall instantaneous total power load of the working face, reducing the impact on the power supply system, improving energy utilization efficiency, and extending the service life of equipment. Attached Figure Description

[0021] Figure 1 A flowchart of a forward working face propulsion control method provided in this application.

[0022] Figure 2 A schematic diagram of a forward-moving working face propulsion control system provided in this application.

[0023] In the diagram: 1. Beat recognition module; 2. Feature acquisition module; 3. Load trend prediction module; 4. Cooperative scheduling module. Detailed Implementation

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0025] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0026] refer to Figure 1 This application provides a method for controlling the advancement of an automated longwall mining face, which is used to control the advancement process of the face. The method includes the following steps:

[0027] A1. Obtain the operating parameters of the coal mining machine, scraper conveyor, and hydraulic support to identify the current production cycle of the working face;

[0028] A2. Obtain the energy consumption characteristic parameters of the coal mining machine, scraper conveyor and hydraulic support under different production cycles, as well as the safe upper limit of the total power of the working face;

[0029] A3. Based on the current production cycle, the operating parameters, geological conditions, work plan, and energy consumption characteristic parameters, predict the instantaneous power load change trend of the coal mining machine, scraper conveyor, and hydraulic support in the future period, and assess the superposition risk of the overall instantaneous total power load of the working face in conjunction with the total power safety limit of the working face.

[0030] A4. Based on the instantaneous power load change trend, the superimposed risks, and the safe upper limit of the total power of the working face, adjust the operating sequence and action parameters of the coal mining machine, scraper conveyor, and hydraulic support to smooth out the overall instantaneous total power load of the working face.

[0031] Among them, operating parameters are data that indicate the current working status of the equipment, such as current, voltage, speed, and position. These can be collected in real time by sensors or obtained from the equipment controller to monitor the equipment's operating status in real time and provide basic data for subsequent analysis.

[0032] Among them, the production cycle refers to the periodic working mode formed by coal mining, coal transportation, support and other operations during the continuous production process of a coal mining face. It can be identified by analyzing the combination characteristics of equipment operating parameters, and is used to provide contextual information of the current working face to guide load forecasting and scheduling.

[0033] Among them, energy consumption characteristic parameters are quantitative indicators that describe the power consumption characteristics of equipment under a specific production cycle. For example, the average instantaneous power corresponding to different operating steps under a specific production cycle can be obtained through statistical analysis of historical data and is used to characterize the power demand pattern of equipment under different operating conditions.

[0034] The upper limit of the total power safety of the working face refers to the maximum instantaneous total power load that the power supply system of the coal mining face can withstand. It is a preset threshold used to ensure the stable operation of the power supply system and the safety of the equipment, and is used to set the boundary for overall load control.

[0035] Among them, the instantaneous power load change trend refers to the predicted curve or pattern of the instantaneous power demand of the coal mining machine, scraper conveyor and hydraulic support changing over time in the future. It can be obtained by model prediction by combining equipment operating parameters, geological conditions, work plans and energy consumption characteristic parameters, and is used to proactively identify potential load peaks.

[0036] Among them, superimposed risk refers to the degree and probability that the total instantaneous power load of the working face may exceed the safe limit in the future. It can be quantified by comparing the predicted load trend with the safe limit and is used to assess the potential power supply shock and equipment damage risk.

[0037] Among them, the operation sequence and action parameters refer to the execution order, start time, duration and specific operation command settings of various operation tasks of coal mining machine, scraper conveyor and hydraulic support, such as coal mining machine traction speed, scraper conveyor speed, hydraulic support shifting interval, etc., which can be adjusted by optimization algorithms or rules to change the power load curve of the equipment and achieve load smoothing.

[0038] The core innovation of this application lies in the fact that by acquiring equipment operating parameters and energy consumption characteristic parameters, it can proactively predict the instantaneous power load change trend of the coal mining machine, scraper conveyor and hydraulic support, and assess the superposition risk of the overall load in combination with the safety upper limit of the total power of the working face. In this way, the operating sequence and action parameters of each piece of equipment can be dynamically adjusted, which can effectively suppress the instantaneous total power load of the working face, achieve the effect of "peak shaving and valley filling", and improve energy utilization efficiency and equipment operation stability.

[0039] Specifically, this method addresses the problem of large fluctuations and potential peak overlap in the overall instantaneous total power load of the coal mining face by comprehensively sensing, intelligently predicting, and collaboratively scheduling the operating status of the coal mining machine, scraper conveyor, and hydraulic supports. First, by acquiring the operating parameters of the coal mining machine, scraper conveyor, and hydraulic supports, the system can monitor the current operating status of each piece of equipment in real time and identify the current production cycle of the working face, providing contextual information for subsequent load prediction and collaborative scheduling. Simultaneously, the system acquires the energy consumption characteristic parameters of each piece of equipment under different production cycles, as well as the safe upper limit of the total power of the working face. This data provides a basis for quantifying equipment power consumption patterns and setting overall load control boundaries. Based on this, the system, by integrating multi-source information including the current production cycle, real-time operating parameters, geological conditions, work plans, and energy consumption characteristic parameters, predicts the instantaneous power load change trends of the coal mining machine, scraper conveyor, and hydraulic supports over a future period. This forward-looking prediction can identify potential load peaks. Subsequently, the system compares the predicted overall instantaneous total power load trend with the safe upper limit of the total power at the working face, quantifies the potential risk of load superposition, and identifies the critical periods and extent that may lead to power supply shocks. Finally, based on the predicted instantaneous power load change trend, the assessed superposition risk, and the safe upper limit of the total power at the working face, the system dynamically adjusts the operating sequence and motion parameters of the coal mining machine, scraper conveyor, and hydraulic supports. This adjustment strategy is targeted; for example, it can stagger the operation of high-load equipment or fine-tune its operating parameters, thereby effectively smoothing the overall instantaneous total power load at the working face, achieving "peak shaving and valley filling," and avoiding the situation where multiple high-power equipment simultaneously reaches peak load. The entire process forms a closed-loop control, from sensing, prediction, assessment to adjustment, ensuring the energy utilization efficiency of the working face and the stability of equipment operation.

[0040] Through the above-described solution, this application effectively solves the problem of large fluctuations in the overall instantaneous total power load of the working face caused by the concentration or overlap of instantaneous load peaks of multiple high-power devices in automated longwall mining faces. This solution achieves smoothing of the overall instantaneous total power load of the working face, avoids power supply system impacts, reduces line losses, extends the service life of electrical equipment, and improves the energy utilization efficiency and production stability of the working face.

[0041] Preferably, the operating parameters of the coal mining machine include the cutting motor current, traction motor current, traction speed, and coal mining machine position information;

[0042] The operating parameters of the scraper conveyor include the drive motor current and the scraper conveyor speed;

[0043] The operating parameters of the hydraulic support include the pressure and flow rate of the pump station, the movement status of the hydraulic support, and the advancing position of each hydraulic support.

[0044] The production cycle includes the coal cutting stage, the reversing stage, the frame shifting stage, and the coal cleaning stage.

[0045] Specifically, the operating parameters of the coal mining machine are refined into cutting motor current, traction motor current, traction speed, and coal mining machine position information. These parameters comprehensively reflect the real-time load and motion status of the coal mining machine under different operating modes. For example, the cutting motor current and traction motor current directly indicate the power consumption of the coal mining machine during cutting and traction processes, the traction speed supplements its motion efficiency information, and the coal mining machine position information provides its spatial coordinates within the working face, which is crucial for determining its current production stage. Meanwhile, the operating parameters of the scraper conveyor are defined as drive motor current and scraper conveyor speed. The drive motor current directly reflects the load magnitude when conveying coal, while the speed indicates its conveying capacity and operating status. These data enable the system to monitor the load status of the scraper conveyor in real time. Furthermore, the operating parameters of the hydraulic supports are defined as the pump station pressure and flow rate, the hydraulic support shifting action status, and the advancement position of each hydraulic support. The pump station pressure and flow rate are directly related to the instantaneous power demand of the hydraulic support group, the shifting action status clearly indicates the occurrence of high-energy-consuming actions, and the advancement position provides the spatial distribution and overall advancement status of the support group. By acquiring these detailed and comprehensive operating parameters, the system can accurately sense and characterize the instantaneous power load of the coal mining machine, scraper conveyor, and hydraulic support.

[0046] Based on this, this application divides the working face production cycle into four stages: coal cutting, reversing, support shifting, and coal cleaning. This precise cycle division allows the system to correlate the acquired detailed operating parameters with specific production modes, thereby more accurately identifying the current production state of the working face. For example, when the coal mining machine's cutting motor current is high and the traction speed is stable, it can be identified as the coal cutting stage; when the coal mining machine's position information shows that it has reached the end and the traction motor current fluctuates, it can be identified as the reversing stage; when the pump station pressure and flow rate increase significantly and the support shifting action is active, it can be identified as the support shifting stage; when the scraper conveyor drive motor current is stable and the coal mining machine and hydraulic supports move less, it can be identified as the coal cleaning stage. This detailed parameter acquisition and precise cycle division provide a solid data foundation and accurate state context for subsequent steps. It enables the system to more accurately acquire energy consumption characteristic parameters under different production cycles, more accurately predict the instantaneous power load change trend of each piece of equipment in the future, and, combined with the working face's total power safety limit, to more reliably assess the superposition risk of the overall instantaneous total power load of the working face. Ultimately, based on these precise perceptions, features, and predictions, the system can make more effective and targeted scheduling adjustments, thereby significantly smoothing the overall instantaneous total power load of the working face and avoiding scheduling deviations caused by insufficient information or inaccurate identification.

[0047] In some implementations, step A1, identifying the current production cycle time of the workface, includes:

[0048] A101. Analyze the instantaneous rate of change of each operating parameter, the duration of the parameter state, and the correlation between parameters to obtain the parameter combination characteristics;

[0049] A102. Compare the parameter combination features with the preset historical parameter patterns corresponding to each production cycle to obtain the preliminary identification result of the current production cycle;

[0050] A103. Based on the preliminary identification results, and in conjunction with the fluctuation range and degree of abnormality of the parameter combination features, determine the confidence level of the preliminary identification results;

[0051] A104. When the confidence level of the preliminary identification result is not lower than a preset threshold, the preliminary identification result shall be taken as the final identification result of the current production cycle.

[0052] A105. When the confidence level of the preliminary identification result is lower than a preset threshold, execute:

[0053] Based on the fluctuation range and abnormality degree of the parameter combination features, abnormal data processing is performed on the parameter combination features to obtain the processed parameter combination features;

[0054] The processed parameter combination features are compared again with the preset historical parameter patterns corresponding to each production cycle to obtain the final identification result of the current production cycle.

[0055] Among these, the instantaneous rate of change reflects the speed and direction of parameter change over a short period, revealing the dynamic characteristics of equipment operation; the duration of a parameter state indicates how long a parameter or a group of parameters maintains a specific state, helping to distinguish between instantaneous fluctuations and stable operating modes; and the correlation between parameters reveals the synergistic or restrictive relationships between different devices or different parameters of the same device, reflecting the overall operational coordination of the system. By integrating the above multi-dimensional information, parameter combination characteristics can be formed, which can comprehensively characterize the abstract data set of the working face production cycle.

[0056] In assessing the reliability of preliminary identification results, a quantitative indicator called confidence level is introduced. Confidence level comprehensively considers the fluctuation range and degree of anomaly of the current parameter combination features. When the confidence level of the identification result is insufficient, anomaly processing can be performed on the parameter combination features. This is a data preprocessing technique aimed at identifying, correcting, or removing outliers, noise, or inconsistent data in the parameters to improve data quality and the accuracy of pattern recognition.

[0057] This application improves the accuracy and robustness of production cycle identification by introducing multi-dimensional feature analysis, confidence assessment, and abnormal data processing and re-comparison mechanisms. It solves the problem of low confidence or inaccuracy that may occur with traditional identification methods in complex and variable operating environments, providing reliable input for subsequent power load prediction and coordinated scheduling. Specifically, firstly, by analyzing the instantaneous change rate of various operating parameters, the duration of parameter states, and the correlation between parameters, parameter combination features can be obtained. This multi-dimensional and dynamic feature extraction method can comprehensively and accurately capture the essential characteristics of equipment operation under different production cycles, overcoming the limitations of single-parameter or static parameter pattern recognition, and providing rich and discriminative information for subsequent cycle identification. Subsequently, the obtained parameter combination features are compared with the preset historical parameter patterns corresponding to each production cycle to obtain the preliminary identification result of the current production cycle. This is the basic step in extracting meaningful cycle information from complex operating data. Based on this, the confidence level of the preliminary identification result can be determined by combining the fluctuation range and abnormality degree of the parameter combination features. This mechanism introduces a quantitative assessment of the reliability of preliminary identification results, helping to identify those with high uncertainty due to data fluctuations, noise, or pattern ambiguity, and avoiding the direct adoption of potentially erroneous judgments under high uncertainty. When the confidence level of the preliminary identification result is not lower than a preset threshold, it indicates that the current data pattern highly matches a certain historical cycle pattern and the data quality is good. At this time, the preliminary identification result can be directly used as the final identification result for the current production cycle, ensuring real-time performance. However, when the confidence level of the preliminary identification result is lower than the preset threshold, the system does not simply abandon or report an error. Instead, it actively performs abnormal data processing on the parameter combination features based on the fluctuation range and abnormality degree of the parameter combination features, obtaining processed parameter combination features. This processing can identify and eliminate or correct abnormal or noisy data that leads to a decrease in confidence, thereby purifying the data quality and making it closer to the true cycle pattern. Subsequently, the processed and purified parameter combination features are compared again with the preset historical parameter patterns corresponding to each production cycle to obtain an accurate and reliable final identification result.

[0058] Through this adaptive confidence assessment and anomaly data processing mechanism, this application can identify the current working face production cycle with high confidence, ensuring accuracy and stability even in complex, variable, or abnormal operating environments. This high-confidence cycle identification result, as the output of step A1, provides accurate and reliable input for subsequent power load prediction, enabling the prediction results to accurately reflect actual working conditions. Furthermore, based on accurate predictions, adjustments to the operating sequence and motion parameters of the coal mining machine, scraper conveyor, and hydraulic supports can effectively mitigate the overall instantaneous total power load of the working face, reduce power supply system impact and equipment wear, thereby improving the robustness and optimization effect of the entire working face propulsion control method.

[0059] As a preferred implementation, the identification of the current production cycle time at the working face can be specifically achieved as follows:

[0060] In step A101, when analyzing the instantaneous rate of change of various operating parameters, differential calculation or linear regression slope within a sliding window can be used. For example, the instantaneous rate of change of the current of the coal mining machine's cutting motor can be obtained by calculating the ratio of the difference in current values ​​between adjacent sampling points to the time interval. The duration of a parameter state can be determined using a state machine model; for example, if the traction speed of the coal mining machine remains within a preset range for a continuous period, it is considered to be in a stable traction state. The correlation between parameters can be calculated using statistical methods such as Pearson correlation coefficient or mutual information; for example, analyzing the correlation between the traction speed of the coal mining machine and the cutting motor current. Through these analyses, a multidimensional vector can be constructed as a parameter combination feature.

[0061] In step A102, when comparing the parameter combination features with preset historical parameter patterns, a machine learning classifier, such as a support vector machine (SVM), random forest, or deep neural network, can be used. These classifiers can be pre-trained with a large amount of historical data to learn typical parameter combination feature patterns corresponding to different production cycles (such as coal cutting, reversing, frame shifting, and coal cleaning). When real-time parameter combination features are input, the classifier can output the production cycle category to which they most likely belong as a preliminary identification result.

[0062] In step A103, when determining the confidence level of the preliminary identification result, the numerical set of each feature parameter in the parameter combination feature within a preset time window can be obtained. Based on this numerical set, the fluctuation index of each feature parameter, such as standard deviation or coefficient of variation, and the deviation index of each feature parameter from the historical parameter pattern corresponding to the preliminary identification result, such as Euclidean distance or Mahalanobis distance, can be calculated. Subsequently, these fluctuation indexes and deviation indexes can be combined, and the confidence level of the preliminary identification result can be determined by methods such as weighted averaging or fuzzy logic reasoning. This confidence level can be a value between 0 and 1.

[0063] In step A104, when the confidence level of the preliminary identification result is not lower than a preset threshold, for example, when the confidence level is greater than 0.8, the system can directly use the preliminary identification result as the final identification result of the current production cycle and pass it to the subsequent power load prediction module.

[0064] In step A105, when the confidence level of the preliminary identification result is lower than a preset threshold, for example, when the confidence level is less than 0.8, the system can perform abnormal data processing on the parameter combination features based on the fluctuation range and degree of abnormality. This processing may include outlier detection based on statistical methods, such as Z-score or box plot methods, to identify and remove or correct abnormal data points. For missing or abnormal data, interpolation (such as linear interpolation, spline interpolation) or smoothing (such as moving average, Gaussian filtering) methods can be used to obtain processed parameter combination features. Subsequently, these processed parameter combination features are input again into the classifier used in step A102 for comparison to obtain the final identification result of the current production cycle.

[0065] In fact, the historical parameter combination features corresponding to the historical parameter patterns of each production cycle can be statistically analyzed in advance to obtain the reference parameter combination features corresponding to each production cycle. When comparing the parameter combination features with the preset historical parameter patterns corresponding to each production cycle, the similarity between the real-time parameter combination features and the reference parameter combination features of each production cycle can be calculated, and the production cycle corresponding to the reference parameter combination features with the highest similarity can be used as the recognition result.

[0066] Preferably, step A103 may include:

[0067] Obtain the set of values ​​of each feature parameter in the parameter combination feature within a preset time window;

[0068] Based on the numerical set, calculate the fluctuation index (representing the fluctuation range) of each feature parameter, and the deviation index (representing the degree of abnormality) of each feature parameter from the historical parameter pattern corresponding to the preliminary identification result.

[0069] The confidence level of the preliminary identification result is determined by combining the fluctuation index and the deviation index.

[0070] The preset time window refers to a continuous time range used to collect and examine changes in parameter values ​​during data analysis. It can be set according to the duration of the actual production cycle, the data sampling frequency, or the system response requirements.

[0071] The numerical set refers to a series of discrete or continuous numerical data collected within a preset time window for each feature parameter in the parameter combination feature.

[0072] Among them, the volatility index refers to the statistical quantity used to quantify the degree of change or stability of the value of a characteristic parameter within a preset time window. It can be calculated using statistical methods such as standard deviation, variance, coefficient of variation, range, or mean absolute deviation.

[0073] The deviation index refers to a statistical measure used to quantify the degree of deviation between the current set of feature parameter values ​​and the historical parameter patterns corresponding to the preliminary identification results. It can be calculated using methods such as Euclidean distance, Mahalanobis distance, percentage deviation, mean squared error, or correlation coefficient. For example, the mean squared error between the average value of a feature parameter within a preset time window and the average value of that feature parameter in the historical patterns corresponding to the preliminary identification results can be used as the deviation index for that feature parameter. Alternatively, the reciprocal of the correlation coefficient between the parameter curve of a feature parameter within a preset time window and the historical pattern curve of that feature parameter can be used as the deviation index.

[0074] Confidence level refers to a quantitative assessment of the reliability or accuracy of the preliminary identification results. It can be a probability value between 0 and 1, or a level representing the degree of credibility. When determining the confidence level of the preliminary identification results by combining fluctuation and deviation indicators, a weighted summation model can be set up, assigning different weights to the fluctuation and deviation indicators, and then adding them together to obtain a comprehensive score. The higher this comprehensive score, the higher the confidence level of the preliminary identification results. Alternatively, a rule-based expert system or fuzzy logic system can be used to classify the confidence level into different levels such as "high," "medium," and "low" based on different combinations of fluctuation and deviation indicators.

[0075] This application provides a quantitative and systematic method for refining the confidence assessment of preliminary identification results. Specifically, firstly, by acquiring the numerical set of each feature parameter in the parameter combination feature within a preset time window, a data foundation is laid for subsequent dynamic analysis. This data collection method allows the evaluation of parameter status to move beyond a single instantaneous value and capture its dynamic behavior over a period of time, thus more comprehensively reflecting the true state of the parameters. Based on this, according to the acquired numerical set, fluctuation indices and deviation indices from the historical parameter patterns corresponding to the preliminary identification results are calculated for each feature parameter. The fluctuation index measures the inherent stability or degree of change of the parameter within the time window, while the deviation index quantifies the degree of conformity between the current parameter behavior and the expected historical pattern. This dual-indicator calculation ensures that the evaluation of the parameter combination feature considers both its own dynamic characteristics and its matching degree with known patterns, providing a multi-dimensional quantitative basis for confidence assessment. Finally, by comprehensively considering the fluctuation index and the deviation index, the confidence level of the preliminary identification results is determined. This comprehensive evaluation method avoids the one-sidedness that may result from a single indicator, and ensures the comprehensiveness and objectivity of the confidence level judgment.

[0076] By employing the methods described above, this application can more accurately determine the reliability of the preliminary identification results. In the entire working face propulsion control method, accurate production cycle identification is the foundation for subsequent power load prediction and coordinated scheduling. Once the confidence level of the preliminary identification results is precisely quantified, the system can decide, based on its level, whether to directly adopt the preliminary results or to process abnormal data of the parameter combination characteristics and compare them again, thereby ensuring that the finally identified production cycle is stable and meets expectations. This refined processing of confidence level judgment improves the accuracy and reliability of production cycle identification, thus providing a more reliable input for subsequent power load prediction. This makes the assessment of power load superposition risk more accurate, ultimately achieving effective adjustment of the operating sequence and action parameters of the coal mining machine, scraper conveyor, and hydraulic support, thereby more effectively mitigating the overall instantaneous total power load of the working face and achieving the expected energy-saving and consumption-reducing goals.

[0077] In some implementations, step A2, which involves obtaining energy consumption characteristic parameters of the coal mining machine, scraper conveyor, and hydraulic support under different production cycles, includes:

[0078] A201. Collect instantaneous power data of coal mining machines, scraper conveyors and hydraulic supports at different production cycles during historical operation, and record them as historical instantaneous power data;

[0079] A202. Based on the historical instantaneous power data, determine the average instantaneous power of the coal mining machine, scraper conveyor, and hydraulic support corresponding to different operating steps under different production cycles, and use it as the energy consumption characteristic parameter.

[0080] This solution collects instantaneous power data from the coal mining machine, scraper conveyor, and hydraulic supports during historical operation at different production cycles, using this data as historical instantaneous power data to lay the foundation for subsequent extraction of energy consumption characteristic parameters. Collecting instantaneous power data ensures the capture of subtle and rapidly changing power fluctuations during equipment operation, which is crucial for identifying and analyzing instantaneous peak loads, as total power peaks are often caused by the superposition of these instantaneous fluctuations. Data collection during historical operation means accumulating a large amount of real and comprehensive data, covering various operating conditions and anomalies, thus making the extracted characteristic parameters more representative and robust. Data collection at different production cycles directly addresses the cyclical characteristics of face production, ensuring that the collected data reflects the unique energy consumption patterns of the equipment at different stages such as coal cutting, reversing, support shifting, and coal cleaning. This is indispensable for subsequent accurate prediction and scheduling based on the current production cycle.

[0081] Based on this, and using historical instantaneous power data, the average instantaneous power of the coal mining machine, scraper conveyor, and hydraulic support at different production cycles corresponding to different operating steps is determined as an energy consumption characteristic parameter. Each piece of equipment performs at least one operating step in each production cycle, and the corresponding average instantaneous power is calculated for each operating step. This process transforms the raw instantaneous power data into characteristic parameters with practical guiding significance. Determining these parameters based on historical instantaneous power data ensures that they are based on empirical summaries of actual operating conditions, rather than theoretical derivations, thus improving their accuracy. Determining the average instantaneous power provides the typical energy consumption level of each operating step of the equipment at a specific production cycle. In this way, this scheme can comprehensively and accurately characterize the energy consumption characteristics of the coal mining machine, scraper conveyor, and hydraulic support at different production cycles corresponding to different operating steps. This provides reliable and detailed data support for subsequent steps to accurately predict instantaneous power load change trends and assess superimposed risks, thereby enabling collaborative scheduling to more effectively smooth the overall instantaneous total power load of the working face. This refined acquisition of energy consumption characteristics enables subsequent load forecasting and risk assessment to more accurately identify potential power peak superposition risks, thus providing a solid data foundation for adjusting work sequence and action parameters, and effectively improving the smoothing effect of the overall instantaneous total power load of the working face.

[0082] Specifically, steps A201 and A202 can be executed for historical operation processes under different geological conditions to generate energy consumption characteristic parameters under different geological conditions.

[0083] In some implementations, step A3 includes:

[0084] A301. Based on the current production cycle, the operating parameters, the geological conditions information, and the work plan, determine the expected work sequence and expected action timing of the coal mining machine, scraper conveyor, and hydraulic support in the future period;

[0085] A302. Based on the expected work sequence and expected action timing, and combined with the energy consumption characteristic parameters, predict the instantaneous power load change trends of the coal mining machine, scraper conveyor and hydraulic support in the future period of time;

[0086] A303. Summarize the instantaneous power load variation trends of the coal mining machine, scraper conveyor, and hydraulic support to obtain the predicted trend of the overall instantaneous total power load of the working face;

[0087] A304. Compare the predicted trend of the overall instantaneous total power load of the working face with the safe upper limit of the total power of the working face, and identify the time window and magnitude of the predicted trend that exceeds the safe upper limit of the total power of the working face, so as to quantify the superposition risk of the overall instantaneous total power load of the working face.

[0088] The work plan includes the expected work tasks of the coal mining machine, scraper conveyor and hydraulic support in the next coal cutting cycle, which can be obtained from the working face management system.

[0089] The geological conditions information may include at least one of the following: coal seam hardness, rock inclusion distribution, faults, etc.

[0090] The expected work sequence refers to a series of specific operational steps that the coal mining machine, scraper conveyor, and hydraulic support will each perform within a future period. This can be achieved through pre-defined work processes, task decomposition based on production plans, or task lists generated by intelligent planning algorithms. The expected action sequence refers to the specific start and end times of each operational step within these expected work sequences, which can be determined using timestamps, relative time intervals, or synchronization triggering conditions. For example, based on inputs such as the current production cycle time, operating parameters, geological conditions, and work plans, a work planning algorithm based on rule-based reasoning or machine learning models can be run. This algorithm can analyze historical work data and current operating conditions to dynamically generate the expected work sequence (and the expected action sequence accurate to the second) for each piece of equipment within a future period.

[0091] The instantaneous power load change trend refers to the dynamic curve of the equipment's instantaneous power changing over time on a future time axis. It can be represented using time series data, a power-time function, or a set of discrete power points. After determining the expected operation sequence and action timing of each piece of equipment, this timing information can be matched with energy consumption characteristic parameters under corresponding geological conditions. The matched energy consumption characteristic parameters are then used to predict the instantaneous power load change trend. For example, if a coal mining machine is expected to perform a cutting operation within a certain time period, and geological information indicates that the coal seam in the area has high hardness, the instantaneous power load change trend within that time period can be predicted based on the energy consumption characteristic parameters of the coal mining machine under the "high-hardness coal seam cutting" condition in historical data. This prediction can be achieved using methods such as lookup tables, regression models, or machine learning models, mapping the expected actions to specific power curves.

[0092] Among them, the predicted trend of the overall instantaneous total power load of the working face refers to the dynamic curve of the total power demand of the entire working face in the future period after superimposing the instantaneous power load change trends of all equipment in the working face.

[0093] The time window here refers to a specific period during which the total power exceeds the safe limit in the predicted trend, which can be defined by the start and end times. The magnitude refers to the specific value or percentage by which the total power exceeds the safe limit within this time window, which can be quantified by the maximum exceedance value, the average exceedance value, or the integral exceedance value.

[0094] Among them, superimposed risks can be represented by risk level, risk index or risk probability.

[0095] This method, through the synergistic effect of the above steps, achieves accurate prediction and risk quantification assessment of the instantaneous power load of the working face. First, in step A301, the system integrates multi-dimensional data such as current production cycle time, operating parameters, geological conditions, and work plans to dynamically and accurately infer the specific operations and timing of each piece of equipment in the future, thereby determining the expected operation sequence and expected action timing of the coal mining machine, scraper conveyor, and hydraulic support. This step is the foundation for accurate load prediction, avoiding the limitations of simple presets or experience-based judgments, and providing highly relevant input for subsequent power load prediction. Based on this, step A302 uses the expected operation sequence and expected action timing determined in A301, combined with pre-acquired energy consumption characteristic parameters, to transform the expected operating behavior of each piece of equipment into specific power load predictions. By combining the expected actions of the equipment with known energy consumption characteristics, the system can generate detailed instantaneous power load curves for each piece of equipment over a future time period, providing necessary data support for identifying potential load superposition. Next, in step A303, the system superimposes the individual instantaneous power load trends of each device predicted in A302 to generate a total power load prediction curve for the entire working face over a future period. This is a crucial step in identifying total power peaks, visually demonstrating the overall situation after the load of each device is superimposed. This allows the system to examine the power demand of the working face from a global perspective, laying the foundation for subsequent risk assessment. Finally, in step A304, the system precisely compares the total power load prediction trend obtained in A303 with the preset safe upper limit of total power for the working face, clearly identifying which time periods will exceed the safe threshold, as well as the specific values ​​and durations of the exceedance. This identification of time windows and magnitude transforms the abstract superimposed risk into specific, quantifiable indicators, making risk assessment no longer a vague judgment but a precise quantitative result. Through the above detailed prediction and assessment process, this scheme enables the perception of the overall instantaneous total power load of the working face to move from qualitative to quantitative, and from coarse to fine.

[0096] This precise prediction and quantitative risk assessment provides a solid data foundation for subsequent adjustments to the timing and motion parameters of operations. For example, when adjusting the timing and motion parameters of the coal mining machine, scraper conveyor, and hydraulic supports, the system can formulate targeted "peak shaving and valley filling" strategies based on the identified specific over-limit time windows and magnitudes. This avoids blind or excessive adjustments, effectively smoothing the overall instantaneous total power load of the working face, ensuring the stable operation of the power supply system, and improving energy efficiency. This refined prediction and assessment capability is a key element in achieving efficient operation of the entire forward-moving working face propulsion control method.

[0097] Preferably, step A304 may include:

[0098] Based on the predicted trend of the overall instantaneous total power load of the working face and the safe upper limit of the total power of the working face, identify all independent over-limit windows in the predicted trend that exceed the safe upper limit of the total power of the working face;

[0099] For each identified independent over-limit window, calculate the maximum instantaneous power over-limit value within that independent over-limit window and the cumulative power integral exceeding the total power safety limit of the working surface;

[0100] Based on the duration, maximum instantaneous power exceedance value, and cumulative power integral of each independent exceedance window, and in conjunction with preset evaluation criteria, the risk weight of each independent exceedance window is calculated.

[0101] By combining the risk weights of all independent over-limit windows, the superimposed risk index of the overall instantaneous total power load of the working face is obtained.

[0102] The independent overload window refers to a continuous period of time in the predicted trend where the power load continuously exceeds the safe limit. It can be achieved by using time series analysis to identify the intersection of the power curve and the safe limit to determine the start and end points, thereby discretizing the continuous overload events into independent events.

[0103] The maximum instantaneous power exceedance value refers to the maximum difference between the predicted power load and the safe upper limit of the total power of the working face within any independent exceedance window. It can be obtained by traversing all instantaneous power data points within the identified exceedance window and selecting the maximum difference that exceeds the safe upper limit of the total power of the working face.

[0104] The cumulative power integral refers to the area within any independent over-limit window where the predicted power load exceeds the safe upper limit of the total power of the working face. It can be calculated by integrating the power and time of the over-limit portion, for example, through numerical integration methods (such as the trapezoidal rule or the rectangular rule).

[0105] Among them, the preset evaluation criteria refer to the set of rules or calculation models used to quantify the degree of risk of exceeding the limit event. They can be implemented by using weight coefficients set based on expert experience, risk assessment models trained based on historical data, or weighted summation formulas containing multiple risk factors (such as duration, exceedance range, and cumulative energy).

[0106] Risk weight refers to a numerical value that quantifies the severity of risk for a single independent over-limit window. It can be calculated by comprehensively considering multiple dimensions such as duration, maximum instantaneous power over-limit value, and cumulative power integral through a preset evaluation criterion, resulting in a single numerical value representing the risk level of that window. For example, the preset evaluation criterion can be a weighted formula: Risk Weight = (Duration Coefficient × Duration Normalized Value) + (Maximum Instantaneous Power Over-Limit Coefficient × Maximum Instantaneous Power Over-Limit Normalized Value) + (Cumulative Power Integral Coefficient × Cumulative Power Integral Normalized Value), where the normalized values ​​are the values ​​obtained after normalizing the corresponding parameters, and the coefficients can be set based on historical data analysis or expert experience.

[0107] The superimposed risk index is a quantitative indicator representing the total risk of the overall instantaneous total power load of the working face over a future period, obtained by comprehensively considering the risk weights of all independent over-limit windows. It can be obtained by weighted summation of the risk weights of all independent over-limit windows, taking the maximum value, or through other aggregation algorithms.

[0108] This scheme, based on a comparison of the predicted trend of the overall instantaneous total power load of the working face with the safe upper limit of total power, further refines the quantification method of superimposed risks. First, the system identifies all independent overload windows based on the predicted power load trend and the safe upper limit. This process decomposes continuous overload periods into discrete, independently analyzable events, because different overload events may have different characteristics and impacts, requiring separate analysis and evaluation. This lays the foundation for subsequent refined risk quantification, avoiding the simplistic treatment of all overloads as a single entity and improving the granularity of risk identification.

[0109] Next, for each identified independent overload window, the system calculates its maximum instantaneous power overload value and the cumulative power integral exceeding the safety limit. The maximum instantaneous power overload value directly reflects the instantaneous impact intensity of the overload event on the power supply system and equipment, while the cumulative power integral quantifies the cumulative effect of prolonged or high-intensity overload on energy loss and equipment wear. By simultaneously considering both instantaneous peak values ​​and cumulative effects, this scheme can more comprehensively capture the harmfulness of overload events, overcoming the limitations of assessment based solely on a single magnitude.

[0110] Based on this, the system calculates the risk weight for each independent overload window according to its duration, maximum instantaneous power overload value, and cumulative power integral, combined with preset evaluation criteria. This step integrates three key dimensions of the overload event: duration, instantaneous impact intensity, and cumulative energy impact, and assigns different weights according to preset evaluation criteria, thereby calculating the comprehensive risk weight for each independent overload window. This multi-dimensional comprehensive evaluation enables the system to distinguish the severity of different types of overload events. For example, a short-duration but high-intensity overload may have different risk weights than a long-duration but smaller-amplitude overload, thus providing a more refined basis for subsequent decision-making.

[0111] Finally, the system integrates the risk weights of all independent over-limit windows to obtain the superimposed risk index of the overall instantaneous total power load of the working face. By integrating the risk weights of all independent over-limit events, this scheme can derive a single, quantitative indicator representing the superimposed risk of the total power load of the entire working face over a future period. This superimposed risk index not only reflects the risk of a single over-limit event but also embodies the overall risk that multiple over-limit events may bring, providing a comprehensive and operable risk assessment result for the subsequent coordinated scheduling module.

[0112] This multi-dimensional and refined quantification of overload risk enables this solution to more accurately assess the overall power load risk of the working face, thus providing a more precise basis for formulating subsequent load mitigation strategies. Compared to methods that only identify the time window and magnitude of overload, this solution can distinguish the severity of different overload scenarios, such as short-term high-intensity peak loads versus long-term moderate overloads, or the cumulative energy consumption during overload periods. This allows subsequent adjustment strategies to more effectively mitigate the risks brought by various power peaks, avoiding poor load mitigation effects, continuous equipment wear, or low energy efficiency.

[0113] In some implementations, step A4 includes:

[0114] A401. Based on the instantaneous power load change trend, the superimposed risks, and the total power safety limit of the working face, determine multiple candidate adjustment schemes; each candidate adjustment scheme includes a combination adjustment of the operation sequence and action parameters of the coal mining machine, scraper conveyor, and hydraulic support;

[0115] A402. For each candidate adjustment scheme, evaluate its effect on the overall instantaneous total power load of the working face, its impact on production efficiency, and its impact on operational safety to determine the comprehensive impact assessment result of each candidate adjustment scheme;

[0116] A403. Based on the comprehensive impact assessment results, select an optimal adjustment plan;

[0117] A404. Based on the optimal adjustment scheme, adjust the operating sequence and action parameters of the coal mining machine, scraper conveyor, and hydraulic support.

[0118] Among them, multiple candidate adjustment schemes refer to a set of different equipment operation timing and action parameters that can be selected and generated by algorithms or rules based on the instantaneous power load change trend, superimposed risks and total power safety limit of the current working face. These can be implemented by heuristic search, optimization algorithms (such as genetic algorithms and particle swarm optimization) or rule-based generation methods.

[0119] Among them, combined adjustment refers to the coordinated, non-independent adjustment of the operation sequence (e.g., start-up, stop, and support relocation timing) and action parameters (e.g., traction speed, support relocation speed, and pump station pressure) of the three core equipment types: coal mining machine, scraper conveyor, and hydraulic support, in order to achieve overall load stabilization. It can be achieved by using predefined combined strategies, linkage adjustment based on model prediction, or dynamic coordination under real-time feedback control.

[0120] Among them, evaluating the effect of candidate adjustment schemes on the overall instantaneous total power load of the working face, the impact on production efficiency, and the impact on operational safety refers to conducting multi-dimensional and multi-indicator quantitative analysis on each candidate adjustment scheme to predict its comprehensive benefits and potential risks in practical applications. This can be achieved through simulation, regression analysis based on historical data, expert system evaluation, or multi-objective optimization models.

[0121] The comprehensive impact assessment result refers to the quantitative or ranking result obtained by weighting, normalizing or using multi-criteria decision analysis methods (such as the analytic hierarchy process or the TOPSIS method) of the above multi-dimensional assessment results. It is used to measure the overall merits of each candidate adjustment scheme and can be implemented by a scoring mechanism, ranking list or multi-dimensional vector representation.

[0122] Among them, the optimal adjustment scheme refers to the scheme that achieves the best balance between power smoothing, production efficiency and operational safety among all candidate adjustment schemes based on the comprehensive impact assessment results. It can be achieved by maximizing the comprehensive score, minimizing the risk-weighted cost, or satisfying the best performance index under all constraints.

[0123] This method introduces a systematic, multi-objective decision-making adjustment mechanism to finely control the operating sequence and motion parameters of the working face equipment, thereby effectively mitigating the instantaneous total power load while balancing production efficiency and operational safety. Specifically, after receiving information on the instantaneous power load change trend, superimposed risks, and the safe upper limit of the total power of the working face provided by previous steps, the system first enters the scheme generation stage. In this stage, based on this input information, the system intelligently determines multiple candidate adjustment schemes. These schemes are not single or fixed, but rather involve combined adjustments to the operating sequence and motion parameters of the coal mining machine, scraper conveyor, and hydraulic support. This means that the system recognizes the dynamic coupling relationship between the three machines and seeks overall optimization through coordinated rather than independent adjustments. This multi-scheme generation mechanism allows the system to explore a broader solution space to cope with complex and ever-changing working face conditions. Subsequently, for each generated candidate adjustment scheme, the system conducts a multi-dimensional evaluation. This evaluation not only focuses on its effect on mitigating the overall instantaneous total power load of the working face, but more importantly, it also considers the impact on production efficiency and operational safety. The evaluation of the load mitigation effect ensures that the proposed solution effectively addresses power peaking issues; the evaluation of the impact on production efficiency avoids sacrificing production speed or output for load mitigation, which is crucial for maintaining economic benefits; and the evaluation of the impact on operational safety ensures that any adjustments will not introduce new safety hazards or lower existing safety levels—a non-negotiable bottom line in mining operations. By comprehensively considering these key indicators, the system obtains a comprehensive impact assessment result, providing a reliable basis for subsequent decision-making. After obtaining the comprehensive impact assessment results of all candidate solutions, the system enters the decision-making stage, selecting the optimal adjustment solution based on these assessment results. This decision-making process based on quantitative assessment results avoids the limitations of human experience or single indicators, ensuring the scientific and rational nature of the selected solution, thereby achieving the best balance between power mitigation, production efficiency, and operational safety. Finally, once the optimal adjustment solution is determined, the system adjusts the operating sequence and action parameters of the coal mining machine, scraper conveyor, and hydraulic supports according to the specific instructions of that solution. This creates a closed-loop control method, from predicting future loads and assessing risks, to generating multiple solutions, evaluating multiple objectives, selecting the optimal solution, and finally executing the plan. This effectively mitigates the overall instantaneous total power load on the workface while ensuring production efficiency and operational safety. This systematic decision-making and execution process enables the workface to operate more intelligently and efficiently, significantly improving the overall low-carbon and high-efficiency production level.

[0124] In one specific embodiment, this method can be implemented as follows:

[0125] In step A401, when the system predicts that the total power load of the working face will exceed the safety limit within a certain period of time and identifies the specific time window and magnitude of the over-limit, it can generate multiple candidate adjustment schemes based on a preset adjustment rule base and optimization algorithm. For example, for a predicted power peak, the system can generate the following schemes: Scheme 1: Reduce the traction speed of the coal mining machine by 10% and delay the movement of some hydraulic supports by 5 seconds; Scheme 2: Keep the traction speed of the coal mining machine unchanged, but stagger and disperse the movement of all hydraulic supports, and fine-tune the operating speed of the scraper conveyor; Scheme 3: Dynamically adjust the traction speed of the coal mining machine while ensuring the minimum production speed, and optimize it in conjunction with the timing of the batch movement of hydraulic supports. These schemes can be generated using a genetic algorithm, with power mitigation as the main objective, and production efficiency and operational safety as constraints, to generate a series of initial schemes that meet the basic requirements.

[0126] In step A402, for these generated candidate adjustment schemes, a simulation model can be used to simulate the operation of each scheme under the actual working face environment. For example, for Scheme 1, the simulation model can calculate its peak power reduction (smoothing effect), coal output change per unit time (production efficiency impact), and equipment load fluctuation (operational safety impact) within the over-limit time window. The smoothing effect can be quantified as the percentage reduction in peak power or the reduction in the over-limit power integral; the production efficiency impact can be quantified as the change in output per shift or advance speed; the operational safety impact can be quantified as the equipment overload risk index or the stress level of key components. These quantitative indicators can then be input into a weighted evaluation model, for example, by determining the weights of power smoothing, production efficiency, and operational safety through the analytic hierarchy process (AHP), and then calculating the comprehensive score for each scheme to obtain the comprehensive impact evaluation result of each candidate adjustment scheme.

[0127] In step A403, the system automatically selects an optimal adjustment scheme based on these comprehensive impact assessment results. For example, if a comprehensive scoring system is used, the system will select the scheme with the highest score. If multiple schemes have similar scores, the system can make a final selection based on preset priority rules (e.g., prioritizing production efficiency while ensuring safety, followed by power mitigation).

[0128] Finally, in step A404, once the optimal adjustment scheme is determined, adjustment instructions are sent to the controllers of the coal mining machine, scraper conveyor, and hydraulic supports according to the specific instructions of the scheme. For example, if the optimal scheme is to "reduce the traction speed of the coal mining machine by 10% and delay the movement of some hydraulic supports by 5 seconds," then a new traction speed setpoint is sent to the coal mining machine controller, and a delayed movement instruction is sent to the hydraulic support group controller. This ensures that these adjustments are executed accurately and in a timely manner, thereby effectively mitigating the instantaneous total power load of the working face while maintaining production efficiency and operational safety.

[0129] This method, by introducing a systematic multi-objective decision-making mechanism, overcomes the problems of suboptimal adjustment schemes and even negative impacts on production efficiency or operational safety in existing technologies. By considering not only the effect on mitigating instantaneous total power load when determining adjustment schemes, but also simultaneously evaluating their impact on production efficiency and operational safety, and selecting the optimal scheme based on these assessments, it ensures that the implemented adjustment scheme achieves a balance among multiple objectives. This enables the working face to maintain or optimize coal production while mitigating power peaks, reducing power supply system impacts and equipment losses, and avoiding the introduction of new safety hazards, thereby significantly improving the overall low-carbon and high-efficiency production level of the working face.

[0130] Preferably, step A401 may include:

[0131] Based on the instantaneous power load change trend, the superimposed risk, and the total power safety limit of the working face, identify the key time window for power load easing and determine the target easing magnitude to be achieved within the key time window.

[0132] Based on the key time window and the corresponding target smoothing amplitude, the feasible operation sequence adjustment range and action parameter adjustment range of the coal mining machine, scraper conveyor and hydraulic support are determined from the preset adjustment rules.

[0133] Based on the operation timing adjustment range and the motion parameter adjustment range, multiple combination adjustment schemes for the operation timing and motion parameters of the coal mining machine, scraper conveyor, and hydraulic support are generated as candidate adjustment schemes;

[0134] A rapid preliminary effect evaluation is performed on all the candidate adjustment schemes, and schemes that cannot meet the target smoothing range or may cause security risks are eliminated, resulting in a final number of candidate adjustment schemes.

[0135] The critical time window refers to the period in the predicted trend of the overall instantaneous total power load change of the working face that may have exceeded the power limit or has already occurred. It can be identified by time period marking, event trigger points, or interval division based on threshold detection. For example, when the predicted total power exceeds a preset percentage of the safe upper limit of the total power of the working face, this time period is marked as the critical time window.

[0136] The target mitigation magnitude refers to the amount of power reduction required to lower the overall instantaneous total power load of the working face to below the safe upper limit or a preset target value within the identified critical time window. It can be determined using absolute power values, relative percentages, or power integrals. For example, the difference between the predicted peak value within the critical time window and the safe upper limit of the working face's total power can be used as the baseline target mitigation magnitude. Then, a correction coefficient is obtained based on the superimposed risk index (e.g., by looking up a preset superimposed risk index-correction coefficient mapping table). This correction coefficient is then used to correct the baseline target mitigation magnitude to obtain the final target mitigation magnitude.

[0137] Among them, the preset adjustment rules refer to the pre-stored constraints, operating procedures and optimization strategies for adjusting the operation sequence and action parameters of coal mining machines, scraper conveyors and hydraulic supports. These rules can be stored and called using rule bases, expert system knowledge bases or strategy sets based on historical optimization experience.

[0138] The operation sequence adjustment range refers to the interval on the time axis where the operation actions such as starting, stopping, reversing or moving of the coal mining machine, scraper conveyor and hydraulic support can be adjusted without affecting the continuity of production and safety. It can be limited by time offset, time window or relative timing relationship.

[0139] Among them, the adjustment range of action parameters refers to the variable range of equipment operating parameters such as the traction speed of the coal mining machine, the cutting speed, the running speed of the scraper conveyor, and the moving speed of the hydraulic support within the allowable range. It can be defined by numerical range, discrete value set, or percentage adjustment amount.

[0140] Generating multiple combined adjustment schemes can employ various algorithms, such as exhaustive search or grid search, to combine the operating timing and motion parameters of the coal mining machine, scraper conveyor, and hydraulic support within a defined adjustment range, using a certain step size or discrete points. Heuristic algorithms, such as genetic algorithms and particle swarm optimization, can also be used to iteratively optimize the search space and generate promising combined schemes. Rule-based generation can also be employed, generating a series of common and effective combination patterns based on preset priorities or empirical rules.

[0141] The preliminary effect assessment refers to a rapid and simplified performance and safety check of the generated candidate adjustment schemes to preliminarily determine whether they meet the basic requirements or pose significant risks. This can be achieved through rule-based rapid verification, simplified model simulation, or heuristic algorithms for quick screening. For example, a rapid preliminary effect assessment can use a simplified power load prediction model to simulate the total power curve of each candidate scheme after adjustment, determining whether it can reduce the peak power below the target smoothing level. Schemes that fail to meet the target smoothing level are eliminated if simulation results show that the adjusted total power peak power is still above the safety limit or fails to reach the set smoothing target. Schemes that may pose safety risks are eliminated by checking whether they violate safety constraints such as minimum safe distances, minimum support shift intervals, and maximum load durations. For example, if a scheme results in the coal mining machine being too close to the support, or the scraper conveyor being overloaded for an extended period, it should be eliminated.

[0142] This solution refines the steps for identifying multiple candidate adjustment schemes, aiming to generate candidate adjustment schemes in a structured, targeted, and pre-screened manner, thereby improving the efficiency and quality of scheme generation and laying the foundation for subsequent scheme evaluation and selection. This solution achieves this goal by first identifying the critical time window for power load mitigation based on instantaneous power load change trends, superimposed risks, and the safe upper limit of the total power at the working face, and then determining the target mitigation magnitude to be achieved within that critical time window. This step allows subsequent adjustment schemes to precisely focus on the periods where actual power over-limit risks exist and clearly defines the mitigation targets. Identifying critical time windows avoids blind adjustments to the entire production process, improving the targeting of adjustments; determining the target mitigation magnitude provides a quantitative standard for evaluating the effectiveness of subsequent schemes. Based on this, and using the critical time window and the corresponding target mitigation magnitude, the feasible adjustment ranges for the operating sequence and motion parameters of the coal mining machine, scraper conveyor, and hydraulic support are determined from preset adjustment rules. This step, after clarifying the adjustment targets, further limits the feasible adjustment space for each piece of equipment. By utilizing pre-defined adjustment rules, it was ensured that the determined adjustment range met equipment performance, operating procedures, and safety requirements, avoiding the generation of infeasible solutions or those that might damage equipment or cause safety accidents. This provided a safe and executable boundary for the generation of subsequent combined solutions. Subsequently, based on the adjustment range of the operation sequence and motion parameters, multiple combined adjustment schemes for the operation sequence and motion parameters of the coal mining machine, scraper conveyor, and hydraulic support were generated as candidate adjustment schemes. After clarifying the adjustment objectives and the adjustment range of each piece of equipment, this step systematically generated multiple potential solutions by combining these executable adjustment options. Because these combined schemes were generated within a defined and executable range, they inherently possess a certain degree of rationality and feasibility, avoiding the generation of a large number of invalid or random schemes. Finally, a rapid preliminary effect evaluation was conducted on all candidate adjustment schemes, eliminating schemes that could not meet the target smoothing range or might cause safety risks, resulting in a final number of candidate adjustment schemes. This crucial pre-screening step quickly eliminated schemes that were clearly unacceptable or posed safety risks before submitting them for more detailed evaluation. Preliminary evaluation can significantly reduce the number of options that require detailed evaluation, thereby significantly improving the efficiency of the entire optimization process and ensuring that the best option selected can effectively mitigate power load while ensuring production safety.

[0143] Through the combined effect of the above steps, this scheme provides a high-quality and efficient set of candidate schemes for subsequent scheme evaluation and selection. Compared with directly generating a large number of unscreened schemes, this scheme significantly reduces the number of invalid schemes through forward-looking identification, constrained generation, and rapid preliminary evaluation. This reduces the computational burden and time cost of subsequent detailed evaluation, making it easier to identify the best adjustment scheme from a smaller, better set of schemes. In turn, it improves the smoothing effect and efficiency of the overall instantaneous total power load of the working face.

[0144] refer to Figure 2 This application provides a forward-moving face propulsion control system for controlling the propulsion process of an automated longwall coal mining face. The system includes:

[0145] The cycle time identification module 1 is used to obtain the operating parameters of the coal mining machine, scraper conveyor and hydraulic support, in order to identify the current production cycle time of the working face (for details, please refer to step A1 above).

[0146] Feature acquisition module 2 is used to acquire the energy consumption characteristic parameters of the coal mining machine, scraper conveyor and hydraulic support under different production cycles, as well as the safe upper limit of the total power of the working face (for details, please refer to step A2 above).

[0147] The load trend prediction module 3 is used to predict the instantaneous power load change trend of the coal mining machine, scraper conveyor and hydraulic support in the future period based on the current production cycle, the operating parameters, geological conditions information, work plan and energy consumption characteristic parameters, and to assess the superposition risk of the overall instantaneous total power load of the working face in combination with the total power safety limit of the working face (the specific process can be referred to step A3 above).

[0148] The collaborative scheduling module 4 is used to adjust the operating sequence and action parameters of the coal mining machine, scraper conveyor and hydraulic support according to the instantaneous power load change trend, the superimposed risk and the safe upper limit of the total power of the working face, so as to smooth the overall instantaneous total power load of the working face (the specific process can be referred to step A4 above).

[0149] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A forward-moving face advance control method for controlling the advance process of an automated longwall mining face, characterized in that, The steps of this method include: A1. Obtain the operating parameters of the coal mining machine, scraper conveyor, and hydraulic support to identify the current production cycle of the working face; A2. Obtain the energy consumption characteristic parameters of the coal mining machine, scraper conveyor and hydraulic support under different production cycles, as well as the safe upper limit of the total power of the working face; A3. Based on the current production cycle, the operating parameters, geological conditions, work plan, and energy consumption characteristic parameters, predict the instantaneous power load change trend of the coal mining machine, scraper conveyor, and hydraulic support in the future period, and assess the superposition risk of the overall instantaneous total power load of the working face in conjunction with the total power safety limit of the working face. A4. Based on the instantaneous power load change trend, the superimposed risks, and the safe upper limit of the total power of the working face, adjust the operating sequence and action parameters of the coal mining machine, scraper conveyor, and hydraulic support to smooth out the overall instantaneous total power load of the working face.

2. The method for controlling the propulsion of a forward-moving working face according to claim 1, characterized in that, The operating parameters of the coal mining machine include the cutting motor current, traction motor current, traction speed, and coal mining machine position information. The operating parameters of the scraper conveyor include the drive motor current and the scraper conveyor speed; The operating parameters of the hydraulic support include the pressure and flow rate of the pump station, the movement status of the hydraulic support, and the advancing position of each hydraulic support. The production cycle includes the coal cutting stage, the reversing stage, the frame shifting stage, and the coal cleaning stage.

3. The method for controlling the propulsion of a forward-moving working face according to claim 2, characterized in that, Step A1, the steps for identifying the current production cycle time of the workface, include: A101. Analyze the instantaneous rate of change of each operating parameter, the duration of the parameter state, and the correlation between parameters to obtain the parameter combination characteristics; A102. Compare the parameter combination features with the preset historical parameter patterns corresponding to each production cycle to obtain the preliminary identification result of the current production cycle; A103. Based on the preliminary identification results, and in conjunction with the fluctuation range and degree of abnormality of the parameter combination features, determine the confidence level of the preliminary identification results; A104. When the confidence level of the preliminary identification result is not lower than a preset threshold, the preliminary identification result shall be taken as the final identification result of the current production cycle. A105. When the confidence level of the preliminary identification result is lower than a preset threshold, execute: Based on the fluctuation range and abnormality degree of the parameter combination features, abnormal data processing is performed on the parameter combination features to obtain the processed parameter combination features; The processed parameter combination features are compared again with the preset historical parameter patterns corresponding to each production cycle to obtain the final identification result of the current production cycle.

4. The method for controlling the propulsion of a forward-moving working face according to claim 3, characterized in that, Step A103 includes: Obtain the set of values ​​of each feature parameter in the parameter combination feature within a preset time window; Based on the numerical set, calculate the fluctuation index of each feature parameter and the deviation index of each feature parameter from the historical parameter pattern corresponding to the preliminary identification result; The confidence level of the preliminary identification result is determined by combining the fluctuation index and the deviation index.

5. The method for controlling the propulsion of a forward-moving working face according to claim 2, characterized in that, Step A2, which involves obtaining the energy consumption characteristic parameters of the coal mining machine, scraper conveyor, and hydraulic support under different production cycles, includes: A201. Collect instantaneous power data of coal mining machines, scraper conveyors and hydraulic supports at different production cycles during historical operation, and record them as historical instantaneous power data; A202. Based on the historical instantaneous power data, determine the average instantaneous power of the coal mining machine, scraper conveyor, and hydraulic support corresponding to different operating steps under different production cycles, and use it as the energy consumption characteristic parameter.

6. The method for controlling the propulsion of a forward-moving working face according to claim 1, characterized in that, Step A3 includes: A301. Based on the current production cycle, the operating parameters, the geological conditions information, and the work plan, determine the expected work sequence and expected action timing of the coal mining machine, scraper conveyor, and hydraulic support in the future period; A302. Based on the expected work sequence and expected action timing, and combined with the energy consumption characteristic parameters, predict the instantaneous power load change trends of the coal mining machine, scraper conveyor and hydraulic support in the future period of time; A303. Summarize the instantaneous power load variation trends of the coal mining machine, scraper conveyor, and hydraulic support to obtain the predicted trend of the overall instantaneous total power load of the working face; A304. Compare the predicted trend of the overall instantaneous total power load of the working face with the safe upper limit of the total power of the working face, and identify the time window and magnitude of the predicted trend that exceeds the safe upper limit of the total power of the working face, so as to quantify the superposition risk of the overall instantaneous total power load of the working face.

7. The method for controlling the propulsion of a forward-moving working face according to claim 6, characterized in that, Step A304 includes: Based on the predicted trend of the overall instantaneous total power load of the working face and the safe upper limit of the total power of the working face, identify all independent over-limit windows in the predicted trend that exceed the safe upper limit of the total power of the working face; For each identified independent over-limit window, calculate the maximum instantaneous power over-limit value within that independent over-limit window and the cumulative power integral exceeding the total power safety limit of the working surface; Based on the duration, maximum instantaneous power exceedance value, and cumulative power integral of each independent exceedance window, and in conjunction with preset evaluation criteria, the risk weight of each independent exceedance window is calculated. By combining the risk weights of all independent over-limit windows, the superimposed risk index of the overall instantaneous total power load of the working face is obtained.

8. The method for controlling the propulsion of a forward-moving working face according to claim 1, characterized in that, Step A4 includes: A401. Based on the instantaneous power load change trend, the superimposed risks, and the total power safety limit of the working face, determine multiple candidate adjustment schemes; each candidate adjustment scheme includes a combination adjustment of the operation sequence and action parameters of the coal mining machine, scraper conveyor, and hydraulic support; A402. For each candidate adjustment scheme, evaluate its effect on the overall instantaneous total power load of the working face, its impact on production efficiency, and its impact on operational safety to determine the comprehensive impact assessment result of each candidate adjustment scheme; A403. Based on the comprehensive impact assessment results, select an optimal adjustment plan; A404. Based on the optimal adjustment scheme, adjust the operating sequence and action parameters of the coal mining machine, scraper conveyor, and hydraulic support.

9. The method for controlling the propulsion of a forward-moving working face according to claim 8, characterized in that, Step A401 includes: Based on the instantaneous power load change trend, the superimposed risk, and the total power safety limit of the working face, identify the key time window for power load easing and determine the target easing magnitude to be achieved within the key time window. Based on the key time window and the corresponding target smoothing amplitude, the feasible operation sequence adjustment range and action parameter adjustment range of the coal mining machine, scraper conveyor and hydraulic support are determined from the preset adjustment rules. Based on the operation timing adjustment range and the motion parameter adjustment range, multiple combination adjustment schemes for the operation timing and motion parameters of the coal mining machine, scraper conveyor, and hydraulic support are generated as candidate adjustment schemes; A rapid preliminary effect evaluation is performed on all the candidate adjustment schemes, and schemes that cannot meet the target smoothing range or may cause security risks are eliminated, resulting in a final number of candidate adjustment schemes.

10. A forward-moving face propulsion control system for controlling the propulsion process of an automated longwall coal mining face, characterized in that, The system includes: The cycle time recognition module is used to acquire the operating parameters of the coal mining machine, scraper conveyor and hydraulic support, in order to identify the current production cycle time of the working face; The feature acquisition module is used to acquire the energy consumption characteristic parameters of the coal mining machine, scraper conveyor and hydraulic support under different production cycles, as well as the safe upper limit of the total power of the working face. The load trend prediction module is used to predict the instantaneous power load change trend of the coal mining machine, scraper conveyor and hydraulic support in the future period based on the current production cycle, the operating parameters, geological condition information, work plan and energy consumption characteristic parameters, and to assess the superposition risk of the overall instantaneous total power load of the working face in combination with the total power safety limit of the working face. The collaborative scheduling module is used to adjust the operating sequence and action parameters of the coal mining machine, scraper conveyor and hydraulic support according to the instantaneous power load change trend, the superimposed risk and the safe upper limit of the total power of the working face, so as to smooth the overall instantaneous total power load of the working face.

Images